Interactive Storytelling
Interactive Storytelling
Interactive Storytelling
Interactive
Storytelling
14th International Conference
on Interactive Digital Storytelling, ICIDS 2021
Tallinn, Estonia, December 7–10, 2021, Proceedings
Lecture Notes in Computer Science 13138
Founding Editors
Gerhard Goos
Karlsruhe Institute of Technology, Karlsruhe, Germany
Juris Hartmanis
Cornell University, Ithaca, NY, USA
Interactive
Storytelling
14th International Conference
on Interactive Digital Storytelling, ICIDS 2021
Tallinn, Estonia, December 7–10, 2021
Proceedings
123
Editors
Alex Mitchell Mirjam Vosmeer
National University of Singapore Amsterdam University of Applied Sciences
Singapore, Singapore Amsterdam, The Netherlands
This Springer imprint is published by the registered company Springer Nature Switzerland AG
The registered company address is: Gewerbestrasse 11, 6330 Cham, Switzerland
Preface
As part of our ongoing efforts to improve the review process, some changes were
made this year. As in the past, the review process was strictly double-blind, used a
structured and detailed review form, and included an extended discussion phase
between the reviewers, steered by our area chairs, to attempt to build a consensus
opinion. A minimum of three reviews per paper were requested before the decision,
with additional reviews solicited on the recommendations of reviewers, or in the light
of their discussions. In addition, we included a rebuttal phase, and final decisions were
made at a virtual program chairs meeting, which included all the area chairs. We
believe that including a virtual program chair meeting as part of the process helps to
improve the transparency of the review process. We welcome feedback from both the
authors and reviewers to help us continue to refine and strengthen the way that we run
the conference. We want to thank our area chairs for their hard work and participation
in the meta-reviews process: Ben Samuel, Boyang “Albert” Li, Frank Nack, Ruth
Aylett, Vincenzo Lombardo, Cristina Sylla, Christian Roth, and Lissa
Holloway-Attaway.
Finally we want to thank the members of the ICIDS community who have served as
reviewers this year, particularly given the ongoing difficult and changing circumstances
around the world. The commitment of our reviewers to provide high-quality reviews
and constructive and insightful discussions is a credit to our community, and helps to
maintain the rigour and integrity of our ongoing development of this exciting and
growing field.
Committees
within our community and the discipline at large. Projects to tackle will be determined
by a needs assessment of the community members, but one initiative already underway
focuses on trans-inclusive publishing practices in terms of facilitating author name
changes. Please reach out to Tess directly if you are interested in working together
(tess.tanen at gmail.com). Come join us and let’s make a difference together!
Graduate Research committee – led by Sarah Brown. This task force puts a focus on
research by graduate students (Masters and PhD) to further exchange, provide support,
and offer a forum for early career researchers. Contact Sarah Brown to join the
committee (sarah.brown at ufl.edu).
Promotion and Advancement committee – led by Hartmut Koenitz and Josh Fisher and
including Luis Bruni and Colette Daiute. The aim of this committee is to create a tenure
equivalency document and recruit a team of expert reviewers for tenure and
examination. Those interested should reach out to Hartmut Koenitz (hkoenitz at
gmail.com).
IDN in Education committee – led by Jonathon Barbara. This committee will be
looking into how IDN can become a part of school (K-12) curricula and will be
producing a white paper with recommendations. Students are also welcome to join as
task force members! Those interested should reach out to Jonathon Barbara (barbaraj at
tcd.ie).
Task Forces
Task Force on Inclusive Pricing Structure – led by Agnes Bakk. This task force will be
looking into how to adjust registration for membership and conference registration
according to GDP. Those interested should reach out to Agnes Bakk (bakk at mome.
hu).
Task Force on ARDIN Outreach – led by Maria Cecilia Reyes. Aims of this task
force are to create awareness about Interactive Digital Narratives and around ARDIN,
and to build partnerships with industry, art, and educational institutions, among other
key stakeholders. Contact Maria Cecilia Reyes (mariaceciliareyesr at gmail.com) for
further information.
Organization
Organization Committee
General Chairs
Michael Mateas UC Santa Cruz, USA
David Lamas Tallinn University, Estonia
Workshop Chair
Mirjam Vosmeer Amsterdam University of Applied Sciences,
The Netherlands
Virtual Chairs
María Cecilia Reyes Akademie Schloss Solitude, Germany
Joshua Fisher Columbia College Chicago, USA
Logistics Chairs
Annegret Kiivit Tallinn University, Estonia
Sirli Peda Tallinn University, Estonia
General Board
Ágnes Bakk Moholy-Nagy University of Art and Design, Hungaria
Luis Bruni Aalborg University, Denmark
Clara Fernandez-Vara New York University, USA
Josh Fisher Columbia College Chicago, USA
Andrew Gordon University of Southern California, USA
Mads Haahr Trinity College Dublin, Ireland
Michael Mateas University of California, Santa Cruz, USA
Valentina Nisi University of Madeira, Portugal, and Carnegie Mellon
University, USA
Mirjam Palosaari Eladhari Södertörn University, Sweden
Tess Tanenbaum University of California Irvine, USA
David Thue Carleton University, Canada, and Reykjavik
University, Iceland
Program Committee
Panos Amelidis Bournemouth University, UK
Sasha Azad North Carolina State University, USA
Julio Bahamon University of North Carolina at Charlotte, USA
Ágnes Karolina Bakk Moholy-Nagy University of Art and Design Budapest,
Hungary
Paulo Bala Universidade Nova de Lisboa and ITI/LARSyS,
Portugal
Jonathan Barbara Saint Martin’s Institute of Higher Education, UK
Marguerite Barry University College Dublin, Ireland
Organization xi
Narrative Systems
Monster Power. Rebel Heart. Gay Sword. Queer Structures and Narrative
Possibility in PbtA Tabletop Roleplaying Games. . . . . . . . . . . . . . . . . . . . . 179
PS Berge
Interviews Towards Designing Support Tools for TTRPG Game Masters . . . . 283
Devi Acharya, Michael Mateas, and Noah Wardrip-Fruin
Interacting with Climate Change: A Survey of HCI and Design Projects and
Their Use of Transmedia Storytelling . . . . . . . . . . . . . . . . . . . . . . . . . . . . 338
Marta Ferreira, Nuno Nunes, and Valentina Nisi
haahrm@tcd.ie
1 Introduction
While visual discrimination between non-diegetic user interfaces and diegetic elements
of a virtual world seen through a Virtual Reality (VR) headset is aided by depth per-
ception, the aural dimension presents a challenge for the player to assess whether a
sound is diegetic or not. Building on Bernstein’s [1] analysis of audio in terms of what
information sounds provide the player, Ekman classifies sounds in relationship to the
diegesis of their referent as ‘the thing being told by the sound’ [2]. She presents two
approaches to such an assessment: (i) whether its apparent source is itself diegetic; or
(ii) whether the non-player inhabitants of the virtual world react to the sound. Both
approaches are dependent on a non-guaranteed relationship between the sound and its
potential source’s visual representation or lack thereof. The ambiguity is compounded
by the layers of information carried by audio signals: frequency, timbre, and semantics
[3] and its exaggerated use to compensate for visual shortcomings [4] such as the use of
the ‘menacing zombie drone for brains’ [5] in Zombies, Run! [6].
In his essay on Narrative Discourse, Genette [14] uses Vendryes’s definition of voice as
used in his Traité d’accentuation grecque (Treaty of Greek Accentuation): “the mode of
action of the verb considered for its relation to the subject” ([15] as cited in [14]). Thus,
voice determines “who speaks?” where the subject of the action includes the narrator and
the narratee(s). In the context of fiction, the narrator is a fictional character invented by
the author, thereby separating the act of narration from the author’s writing and indeed
from the author being in the narrative [14]. This frees the author from narrating in the
diegetic first-person (homodiegetic narration), and gives the option to narrate in the non-
diegetic third-person (heterodiegetic narration) [14], which in turn results in new options
for narrating the protagonist’s experience of the plot: as the protagonist (autodiegetic
narrator), as a non-protagonist character diegetic to the story (homodiegetic narrator),
or as a non-diegetic, unrepresented, character (heterodiegetic narrator).
Aare [16] clearly distinguishes between the voice (who speaks) and the perspective
(who sees). Genette categorises perspective into three forms: (i) in Internal focalization,
knowledge is bound to one character, usually the protagonist-narrator; (ii) in External
focalization, knowledge about the protagonist is limited to a third-person’s perspective;
and (iii) in Non-focalization, the knowledge would not be limited to the perspective of a
single character. This could suggest the existence of a non-diegetic implied author [14]
Who Am I that Acts? 5
or an impersonal voice of the narrative, linked to no one character in particular [17] and
comparable to a third-person heterodiegetic narration.
Narration has been an important counterpart to mimesis in theatre [18] when, in
prologues and epilogues, the presence of narrators on stage present a prolepsis (what is
yet to be mimetically shown) and, in some non-Western drama, such as Japanese plays,
the importance of the narrator is as much as that of the main characters in terms of lines
and time on stage. In games, prologues find their equivalent in tutorial missions. Due to
their interactive nature, games most often use tutorials to guide the player-narratees in
their participation of the narration through their gameplay. Such players are addressed
using voices whose diegesis depends on the game’s genre. Non-diegetic narration serves
to direct the mimetic gameplay, giving motivation and justification for the player’s in-
game actions, similar to ‘generative narrators,’ whose narration instantiates action [18].
Diegetic voice, situated inside the virtual environment via spatialisation effects and the
Barthesian concept of ‘voice grain’ to assist in matching the aural dimension with the
visual representation of the source, needs to be loaded with emotional semantics in order
to help deliver the narrative and cohere it with the gameplay [4].
VR marries theatrical mimetic performance with gameplay as, rather than pushing keys
or buttons, the player virtually embodies the action-triggering performance. The ques-
tion “Who speaks?” is pertinent to the VR player who, sensorially immersed in the
virtual world, will attempt to seek out a diegetic owner of the voice and understand the
perspective and focalisation of this voice. A non-focalised owner would have access to
information beyond one’s knowledge which may confuse the player and lessen realism.
Non-diegesis, or even trans-diegesis, would lessen the VR player’s immersion, because
they are reminded of their physical self’s role as an external audience. Diegetic voices
presenting an external focalisation, on the other hand, would represent a realistic char-
acter’s perspective and thus heighten immersion [19]. Once the characters behind the
voices around the player and their nature are identified and, especially, when addressed
directly by these characters, the next question for the player is self-identification: ‘who
am I that speaks?’ or rather, due to the affordable strong literal interaction [20] of VR,
‘who am I that acts?’ What is the player’s avatar’s relationship to those around them in
the virtual world? How are they expected to behave?
When the audience is addressed in the second person ‘you’ in a narrative, their presence
in the virtual space is acknowledged and reaffirmed [21], whether they have no visible
impact upon the narrative or its space, or they are given a role as co-creators [22],
giving them a share in the responsibility of decision-making [23]. In digital interactive
experiences such as first-person perspective games and interactive documentaries (i-
docs), this responsibility demands knowledge of who they are relative to the storyworld
and the characters within it. Following Bell and Ensslin [21], answering the question ‘who
am I that acts?’ in VR demands sensitivity to media-specific affordances, specifically
6 J. Barbara and M. Haahr
having the player entering the virtual world through the aural and visual senses of a
virtual character diegetic to the storyworld and demanding an interactive role in the
virtual space. So, who could this character be?
In traditional narratives (e.g., novels or films) the narratee may identify as: (1) the
narrator, who may or may not be the protagonist of the story; (2) the protagonist of a
story, told by a third-party narrator; or (3) an onlooker to a protagonist’s story, narrated
by the protagonist or a third party [24]. These identities are now considered in a VR
context:
The first identity has the player identify as the narrator by assigning a voice to the
player’s avatar that mismatches the player’s own voice breaks immersion [25], and thus
using the player’s own voice for commands or reading out choices may be considered
instead, as used successfully in the game In Verbis Virtus [26]. Using the player’s own
voice would allow us to address the original question, ‘who am I that speaks?’ and would
also help further immerse the player through ‘ludonarrative consonance’ by matching
voice with action [4]. Higher presence can be achieved if the voice is augmented with
real-time echoes to model the spatialisation effects of the virtual space being inhabited.
Such effects suggest that the source of the sound is diegetically situated in the virtual
space, and since the source of that sound is the player him/herself, then this supports the
experience of telepresence [27].
The second identity has the player identify as the protagonist [28], which is akin to
the use of second-person voice in Choose Your Own Adventure (CYOA) books, where
players have a say in the narration of the story through the interaction afforded by the
narrative device. The reader is addressed with a “you” as the main protagonist within
the narrative (intradiegetic narration) but also as the decision-maker in charge of the
non-diegetic interaction (extradiegetic narration), causing the narratee to shift alternately
between the two [24]. Interactive Fiction, offline text adventure games, Multiuser Dun-
geons (MUDs), and their object-oriented variations (MOOs), as digital adaptations of
CYOA books, also employ second-person voice as they address the player in the role
of a character – the same player who writes textual commands to guide the narrative in
their role as co-author [25]. In the case of the digital game The Stanley Parable [29], the
player’s actions cause Stanley to behave differently from the narrator’s description, as
the role of the player from narratee to protagonist slowly reveals itself to the narrator.
In the context of VR interactive narratives, however, having the player inhabit the
avatar’s visual and aural senses may minimise the separation between intradiegetic and
extradiegetic narration by embedding their interaction into their diegetic agency and
may thus augment their feeling of presence in the interactive narrative’s storyworld.
This may cause tension between the authorial control of the narrative and the agency
provided to the player, known as ‘ludonarrative dissonance’ [30]. For the third identity
of the player as an onlooker, Larsen [31] suggests a solution in the use of second-person
point-of-view that allows viewers to participate as sidekicks (diegetic observers) to the
protagonist without the ability to modify the narrative structure. This solution lets the
audience tackle side quests alongside the main narrative, while still allowing the main
narrative to progress resolutely, irrespective of player action. Larsen’s approach also
suggests a preference for non-protagonist roles of the VR player in order to support
‘ludonarrative consonance’. Thus, the character may be a bystander who perceives the
Who Am I that Acts? 7
4 Case Studies
We now consider the use of voice in a close reading of two VR documentaries. Both are
situated during World War II, but our interest is in their use of voice, rather than their
setting or subject matter. Both case studies project the viewer as a non-protagonist non-
narrating character, but they use second-person voice differently and provoke different
levels of self-identification through the provided agency.
immigrants. Randall serves as the embodied narrator who takes the VR player along
an imagined journey of his grandfather Yonezo in the VR short documentary1 entitled
The Book of Distance [34]. In the experience, Yonezo is presented through original pho-
tographs, digitized and made virtually available for picking up for close-up inspection,
during which Randall’s voice explains who is who in the photos. A game of horse-
shoes – which serves as a connection between Yonezo and his nephew as evidenced by
one of the photos – is presented to the player as an ice-breaker into the interactive narra-
tive, empowering the player with agency and bringing out Randall’s character. Playing
on the lack of ambient light, Randall often moves off into the shadows, shifting the
player’s attention to the unfolding story that his off-screen voice narrates. Members of
Yonezo’s family, such as Randall’s father, are presented through their Japanese cartoon
representation as well as the playback of recorded vocal interactions. Yonezo’s family
in Hiroshima, including his younger sister, are presented through animated coloured
silhouettes but are not given a voice, reflecting the narrator’s lack of familiarity with
them.
Throughout the first part of the experience, the player acts as a sidekick to Randall’s
grandfather, helping him pack his clothes and photographic camera into his luggage for
his voyage to Canada, responding to his family and future wife’s waving whilst on the
ship, having his passport stamped at the Canadian customs, taking photos of his house-
building activities, giving a hand in clearing up the land, building the house, sowing
the strawberries and serving them at the dinner table. All this ends abruptly as World
War II starts and Yonezo’s family are taken away by Canadian military to an internment
camp for the Japanese. All their possessions are taken away: the camera, the house, the
strawberry business, and likewise the player’s narrative agency is greatly reduced.
Yonezo eventually returned to freedom, but it was hard earned and he never spoke
about it to his nephew. Randall’s lack of familiarity with this part of the story is reflected
in the narrative’s shift of focus away from the grandfather onto the father, and in the
player’s lack of agency, distancing them from the storyworld. The photographic camera
is back in post-war Yonezo’s hands as he takes photos of his growing children, photos
that the player had seen at the beginning of the story. These are now brought back to
the player’s scrutiny with the addition of more recent, colourful photos of the grown
characters. The story ends with the player looking through photos of Randall’s father’s
childhood while, in the background, Randall discusses them with his father, seeking to
further understand the last years of his grandfather’s life in Canada.
5 Discussion
The two case studies provide contrasting examples of the use of second-person voice and
its effect on self-identification from the viewer’s behalf. With respect to the Holocaust
VR film The Last Goodbye, Zalewska [33] reports that the subject of the VR film is the
player’s experience of the Majdanek camp rather than that of Gutter’s. Doubts are also
cast on the perceived fidelity of the experience due to the scripted narration making the
experience more of a documentary than a testimony. While Gutter imparts his painful
1 https://www.youtube.com/watch?v=b9DDoeeQq6g.
Who Am I that Acts? 9
experience of the concentration camp and his turmoil in remembering nothing about his
twin sister except her golden braid, there is no requirement or incentive for the player
to identify themselves with Gutter or with any other relevant character. The player is
offered no meaningful agency except linear narrative progression. Thus while it is clear
that Gutter is the one ‘who speaks’, there is no self-identification for the player.
The Book of Distance tackles the VR player’s self-identity very differently. The
player is immediately given agency through a book presented in front of them with
instructions to turn the page – reminiscent of CYOA books. From the beginning, the
player is addressed with the second-person voice: “To You, the Time Traveller” and,
as soon as Randall (the narrator) makes an entrance, he addresses the player directly,
teaching them how to throw the horseshoe. This second-person address of the player
continues throughout the journey of discovery, with the narrator’s voice (who speaks?)
embodied by Randall’s character explicitly represented in the scene or implied to be
hiding in the shadows around the player.
The Book of Distance makes use of mechanics and structures frequently used in
games and traditional films, even though the experience itself is neither. The empower-
ment of the player’s agency and its subsequent reduction is a common technique used
to challenge the player by limiting his/her skills. In The Book of Distance, however, the
effect is not a feeling of increased challenge but a sense of loss of freedom, of identity. It
starts with a prologue where modern-day Randall presents his grandfather, whose story
we explore together throughout the rest of the experience, until we reunite with his father
for the story’s epilogue. We are then told exactly how Randall knows of his grandfa-
ther’s life: through the photos, his father’s testimony, and the letters received from the
Canadian government, to name a few. The experience uses touched-up heterodiegetic
narration, attempting an internal focalisation as the nephew tries his best to understand
his grandfather’s experience of his life as a Japanese immigrant in war-time Canada.
Self-identification suffers, however, as the actual role of the VR player is indetermi-
nate. Observation of Randall’s exploration of his grandfather’s story through his photo
album evolves into active participation as he starts recounting his grandfather’s story. As
players, we become active sidekicks to the grandfather, helping him paint, write letters,
prepare the luggage for his voyage to Canada, and making him aware of his family
waving him off. But as the experience progresses, it is not clear whose sidekicks we
are: Randall’s or his grandfather’s. As we help the latter build his house, sow his fields,
serve the strawberries on the table, the question begs to be asked: who am I that acts? A
friend of the family perhaps? This becomes even less clear when, as Yonezo is separated
from his family and taken away to a field, it is the player who gets to raise the lever
that traps the Japanese farmer inside the internment camp. Are we now sidekicks of
the Canadian government? It appears to be impossible to reconcile these three different
roles: chronologically separate sidekicks to the nephew and to the grandpa and morally
separate sidekicks to the grandpa and to the Canadian military, and it raises the ques-
tion: Whose side is the sidekick on? As a result, the player identifies with none of these
diegetic roles, but instead the experience constructs a non-diegetic player role whose
responsibility is to push the narrative forward. By not sticking to a specific persona, the
virtual character that is embodied becomes transient across time and actions, such that
the question, ‘who am I that acts’, does not resolve to a specific characterisation, leaving
10 J. Barbara and M. Haahr
the player with themselves as enactors of the experience. This was a design choice with
the player only identifying themselves as ‘a part of the story’ [34].
Thus, neither of the two experiences manages to successfully assist in self-
identification. The Last Goodbye, through its limited agency and non-address of the
player only serves to inform of the terrible loss of Pinchas Gutter at the hands of the Nazi
regime. The Book of Distance makes it a point to continuously address the player and to
provide agency that serves LudoNarrative Coherence as the gameplay contributes to the
progression of the narrative while the empowerment and reduction of agency progresses
with the narrative. However, as the experience follows a prologue/interlude/epilogue
structure pertinent to theatre and film, the diegetic identity of the player transcends time,
and the fruitful agency of the player’s character places them on the same diegetic space
as the younger grandfather and his family to which the nephew seems to have no access,
except as an observer. The already fragile link between the two trans-chronological per-
sonas is further disconnected with the entrapment action that results in Yonezo Okita’s
confinement in the internment camp, the character’s body language showing as much
disbelief at the player’s traitorous act as his introvert self could afford.
6 Conclusion
References
1. Bernstein, D.: Creating an interactive audio environment. Gamasutra 1, 2013 (1997)
2. Ekman, I.: Meaningful noise: understanding sound effects in computer games. Proc. Digit.
Arts Cult. 17 (2005)
3. Wallmark, Z., Kendall, R.A.: Describing sound: the cognitive linguistics of timbre. In: The
Oxford Handbook of Timbre, p. 14. Oxford University Press, New York (2018). https://doi.
org/10.1093/oxfordhb/9780190637224.013
4. Ward, M.: Voice, videogames, and the technologies of immersion, p. 15 (2010)
5. Gardner, K.: Braaiinnsss!: zombie-technology, play and sound. In: Technologies of the Gothic
in Literature and Culture, pp. 81–93. Routledge, New York (2015).
6. Guerilla Bandit: Zombies, Run! (2012)
Who Am I that Acts? 11
7. Barker, J.M.: The Tactile Eye: Touch and the Cinematic Experience. University of California
Press (2009)
8. Busselle, R., Bilandzic, H.: Measuring narrative engagement. Media Psychol. 12, 321–347
(2009). https://doi.org/10.1080/15213260903287259
9. Mildorf, J.: Reconsidering second-person narration and involvement. Lang. Lit. 25, 145–158
(2016)
10. McMahan, A.: Immersion, engagement, and presence: a method for analyzing 3-D video
games. In: The Video Game Theory Reader, pp. 89–108. Routledge (2013)
11. Brown, E., Cairns, P.: A grounded investigation of game immersion. In: CHI 2004 Extended
Abstracts on Human Factors in Computing Systems, pp. 1297–1300. ACM, Vienna (2004)
12. Lombard, M., Ditton, T.: At the heart of it all: The concept of presence. J. Comput.-Mediat.
Commun. 3, JCMC321 (1997)
13. Nordahl, R., Nilsson, N.C.: The sound of being there: presence and interactive audio in
immersive virtual reality. In: The Oxford Handbook of Interactive Audio. Oxford University
Press, United Kingdom (2014). https://doi.org/10.1093/oxfordhb/9780199797226.013.013
14. Genette, G.: Narrative Discourse: An Essay in Method. Cornell University Press, Ithaca (1983)
15. Vendryes, J.: Traité d’accentuation grecque. C. Klincksieck (1904)
16. Aare, C.: A narratological approach to literary journalism: how an interplay between voice
and point of view create empathy with the other, p. 34 (2016)
17. Nielsen, H.S.: The impersonal voice in first-person narrative fiction. Narrative 12, 133–150
(2004)
18. Richardson, B.: Point of view in drama: diegetic monologue, unreliable narrators, and the
author’s voice on stage. Comp. Drama 22, 193–214 (1988). https://doi.org/10.1353/cdr.1988.
0017
19. Barreda-Ángeles, M., Aleix-Guillaume, S., Pereda-Baños, A.: An “empathy machine” or a
“just-for-the-fun-of-it” machine? Effects of immersion in nonfiction 360-video stories on
empathy and enjoyment. Cyberpsychol. Behav. Soc. Netw. 23, 683–688 (2020)
20. Ryan, M.-L.: Immersion vs. interactivity: virtual reality and literary theory. SubStance 28,
110–137 (1999)
21. Bell, A., Ensslin, A.: “I know what it was. You know what it was”: second-person narration
in hypertext fiction. Narrative 19, 311–329 (2011)
22. Walmsley, B.: Co-creating theatre: authentic engagement or inter-legitimation? Cult. Trends.
22, 108–118 (2013)
23. Ciancio, G.: Active spectatorship, changes and novelties in the performing arts sector. In:
Bonet, L. and Négrier, E. (eds.) Breaking the Fourth Wall: Proactive Audiences in the
Performing Arts, pp. 90–96 (2018)
24. DelConte, M.: Why you can’t speak: second-person narration, voice, and a new model for
understanding narrative. Style 37, 204–219 (2003)
25. Vosmeer, M., Roth, C., Koenitz, H.: Who are you? Voice-over perspective in surround video.
In: Nunes, N., Oakley, I., Nisi, V. (eds.) ICIDS 2017. LNCS, vol. 10690, pp. 221–232. Springer,
Cham (2017). https://doi.org/10.1007/978-3-319-71027-3_18
26. Ferrari, M.: In verbis virtus. Presented at the (2015). https://doi.org/10.1007/978-3-319-
05326-4_18
27. Johansson, M.: VR for your ears: dynamic 3D audio is key to the immersive experience by
mathias johansson· illustration by Eddie guy. IEEE Spectr. 56, 24–29 (2019)
28. Fludernik, M.: Second person fiction: Narrative “you” as addressee and/or protagonist. AAA
- Arb. Aus Angl. Am. 18(2), 217–247 (1993)
29. Wreden, D., Pugh, W.: Stanley parable (2011)
30. Hocking, C.: Ludonarrative Dissonance in Bioshock. http://clicknothing.typepad.com/click_
nothing/2007/10/ludonarrative-d.html. Accessed 24 Jan 2016
12 J. Barbara and M. Haahr
31. Larsen, M.: Virtual sidekick: second-person POV in narrative VR. J. Screenwrit. 9, 73–83
(2018). https://doi.org/10.1386/josc.9.1.73_1
32. Traum, D., et al.: New dimensions in testimony: digitally preserving a holocaust survivor’s
interactive storytelling. In: Schoenau-Fog, H., Bruni, L., Louchart, S., Baceviciute, S. (eds.)
ICIDS 2015. LNCS, vol. 9445, pp. 269–281. Springer, Cham (2015). https://doi.org/10.1007/
978-3-319-27036-4_26
33. Zalewska, M.: The last goodbye (2017): virtualizing witness testimonies of the holocaust,
p. 8 (2017)
34. Oppenheim, D., Okita, R.L.: The book of distance: personal storytelling in VR. In: ACM
SIGGRAPH 2020 Immersive Pavilion, pp. 1–2. ACM, Virtual Event USA (2020). https://doi.
org/10.1145/3388536.3407896
A Synset-Based Recommender Method
for Mixed-Initiative Narrative World
Creation
1 Introduction
Narrative Worlds (NW) are environments designed to support enacting a given
story [2]. In the fast-paced industry of games and films, NWs are becoming
more realistic and complex in nature. NW designers need new techniques and
workflows to rapidly explore their creative ideas while preserving story consis-
tency and integrity. Tools such as the game engines Unity or Unreal Engine have
become increasingly popular for the creation of virtual worlds in form of game
levels or virtual film production. Yet, these tools lack any narrative understand-
ing and leave designers with full creative responsibility.
c Springer Nature Switzerland AG 2021
A. Mitchell and M. Vosmeer (Eds.): ICIDS 2021, LNCS 13138, pp. 13–28, 2021.
https://doi.org/10.1007/978-3-030-92300-6_2
14 M. R. B. Perez et al.
2 Related Work
3 Mixed-Initiative Approach
In this section, we describe the basis of our mixed-initiative approach, as well as
its interactive workflow, both illustrated in Fig. 1.
First, assume that an NW designer has a working space (e.g., game engine
or level editor) that allows to create virtual locations for an NW; we refer to
this as NW canvas. We improve the designer’s workflow with a recommender
method, which also supports exploration. Our method suggests plausible enti-
ties for an NW based on common and novel associations to narrative locations,
actions and other entities. Similar to a screenplay, the designer first selects a
16 M. R. B. Perez et al.
particular scene of the story and an NW canvas is shown with the location’s
layout of the scene; the recommender then suggests entities associated only to
the location. The designer iteratively adds, removes and arranges entities by
exploring recommendations or searching for specific content. The designer walks
through each plot point or selects entities to further guide the recommender. At
this point, the recommendations are based on the location, action and selected
entities. The designer can select additional features to guide the exploration pro-
cess. In the following subsections, we describe in detail the main components of
our approach.
We represent an action and its involved entities by means of a plot point (pp);
a semantically-coherent structure which describes an important event of a story.
The argument structure of a plot point is determined by the verb of an action
(e.g., eat, give) and involved entities, each slot denoted by a semantic role (SR).
SR names can be customized to provide clarity about the role an entity has in
the plot point. A general structure of the plot point representation with N slots
is as follows:
The number of slots and their corresponding SRs depend on the common argu-
ment structure of a verb. In our approach, we used the argument structure of
verbs and SRs already provided by the hand-crafted lexical-semantic resource
VerbAtlas2 [8]. To illustrate our representation, consider a story with a scene
happening in a living room in the evening, with this sequence of two plot points:
S1 . [INT.] [LIVING ROOM] - [EVENING]
Space Location Time
pp1 . [Bob]Agent [Give] [Pizza]Theme [Sally]Recipient
Subject Verb Direct Object Indirect Object
pp2 . [Sally]Agent [Eat] [Pizza]Patient
Subject Verb Direct Object
The first line is a scene description S1 , followed by the two plot points, pp1
and pp2 . In pp1 , give is used as a transitive verb of three argument slots: the
subject Bob is the Agent who initiates the action, the direct object Sally is
the Recipient, and the indirect object pizza is the thing being transferred as
Theme. In pp2 , eat is used as a transitive verb of two argument slots; the subject
Sally is the Agent who performs the action, and the direct object pizza is the
thing affected as Patient.
An NW can have several locations, each of them decorated with physical entities
that are coherent with associations suggested by a story [2]. We identify two
types of NW content: explicit content and plausible content. The explicit content
consists of every entity specified in a plot point. For example, in “Bob drinks
coffee”, coffee must exist in the location to support the action drink. In contrast,
plausible content is every entity that potentially fits in the location, but their
implicit nature is more of an open question. These entities might be required
or only serve as decoration for the location. Traditionally, designers determine
plausible content based on their own creative experience. Our method assists
designers with suggestions based on the following learned associations:
2
VerbAtlas provides semantically coherent structures of verbs, see verbatlas.org.
18 M. R. B. Perez et al.
V
J= f (Xij )(wiT wj + bi + bj − log Xij )2 (1)
i,j=1
Then, we sort the result according to the average angular distances ϕ̄dists and
select the k entities with the smallest distances.
results, analogous to a boolean NOT query (e.g. animal NOT bird ). Thus, we
perform vector negation as Q NOT Q− to extract a subspace Q+ of vectors that
has no features in common with Q− [32]. We obtain qi+ ∈ Q+ by subtracting the
vector projections of all synset vectors qj− ∈ Q− on each qi ∈ Q:
qi · qj−
Q+ = {qi+ ∈ Q+ : qi+ = qi − − q ∈ Q}
||
q − ||2 qj , ∀i (5)
j− ∈Q−
j
q
We replace Q for Q+ in Eq. (4) as ϕ̄dists (Q+ , W, V ) and retrieve the top k
entities with shortest distance. Interestingly, this feature can be used to neutral-
ize common human subjective biases (e.g., gender and cultural biases) prone to
be found in models learned from text corpora [4].
We chose this criterion due to the tendency to form smaller, compact clusters.
The optimal clusters are determined via the silhouette coefficient [23]: a value
from −1 (bad fit) to 1 (good fit) which measures the quality of the clusters in
terms of how good each entity fits in its own cluster. Given the mean intra-cluster
distance ai and the mean nearest-cluster distance bi , we compute the coefficient
si for each entity and the average coefficient s̄ as follows:
|R|
b¯i − a¯i 1
si = s̄ = si (7)
max(a¯i , b¯i ) |R| i=1
We further constrain clusters by only considering the steps where their size
is below the maximum threshold Tmax = 3 to ensure compact clusters. For
convenience, we sort clusters in ascending order by the average of the score
obtained with Eq. (4) for entities cj in each cluster Ck ∈ C:
1
ϕ¯c dists (Q, W, C) = {ϕc ∈ R : ϕc = ϕ̄dists (Q, W, Ck ), ∀Ck ∈ C} (8)
|Ck |
cj ∈Ck
5
For hierarchical clustering visit nlp.stanford.edu/IR-book/completelink.html
22 M. R. B. Perez et al.
4 Interactive Prototype
Fig. 3. The story “Once upon a time, a pirate...”: with 3 plot point per scene/location.
Mixed-Initiative Narrative World Creation 23
In Fig. 4, we show three main panels of TaleForge: the entities panel, scenes
panel and canvas panel. In the entities panel, the designer searches for entities,
obtains recommendations and configures the recommender by setting categories,
weights, number of results, clustering and an unrelatedness bin for NOT queries.
An entity appears with a name, icon6 , category color and description when
hovering over it. The novelty of a recommended entity is visualized as a lighter
(more novel) to darker (least novel) background. In the scenes panel, the designer
sees every scene of the story and selects one for editing the location’s layout. In
the canvas panel, the designer creates a location by placing entities into cells of
the canvas. The designer can select individual plot points and entities to guide
6
Entities’ icons are populated automatically from thenounproject.com
24 M. R. B. Perez et al.
Fig. 5. Designer’s workflow while creating an NW for “Once upon a time, a pirate...”.
the recommender. The front end of TaleForge was implemented in Unity, and
the recommender method is served by a Python’s flask API.
In Fig. 5, we show the recommender’s output at various steps of the designer’s
workflow, together with the final layouts of the locations for our showcase story.
To explain the designer’s workflow, consider the first scene S1 as an example.
Here, the designer creates a layout for a forest. When selecting this scene, a plot
points menu (top) and an empty NW canvas (bottom) is shown. The designer
adds entities by searching for them and dragging/dropping them. Further, it is
possible to paint on the canvas, e.g., to place several trees, by selecting the object
and moving the mouse over the canvas, while holding the mouse button pressed.
The recommender is prompted when interacting with any element. At first, it
Mixed-Initiative Narrative World Creation 25
only considers entities related to the location (the forest). The designer selects
the category plant and adds recommended entities like pine trees and bushes.
When entities are added to the canvas, the novelty scores are updated, thus,
recommended entities such as bark, trunk and lichen appear lighter (more novel).
The designer can enforce a different output of the recommender by changing the
narrative query weights or using the unrelatedness bin. The designer prompts
the recommender by selecting a pine tree and lowers the weight of the location to
put more emphasis on the selected object, causing pinecone to appear in the top
results. Next, the designer might be interested in entities that are not related to
tree, so the designer drags and drops tree into the unrelatedness bin and activates
clustering. Consequently, the recommender suggests and groups different types
of mushrooms.
The designer selects a plot point to consider the influence of an action. In
pp1 , the designer selects the category animal to add animals in the location
for finding food. The output shows different birds which might not be interesting
for the designer. Thus, the designer drags and drops bird to the unrelatedness
bin and activates clustering to obtain groups of animals unrelated to bird, like
chipmunk, squirrel, deer, and so on. In pp2 , the designer selects the category
artifact and substance to find entities related to making a fire, such as
wood, firewood, axe, and others. In pp3 , the recommender outputs entities for
preparing food, however, these are common cooking utensils and might not be
interesting enough for a story, where a pirate is involved. Thus, the designer
selects pirate from the plot point and discovers caldron in the recommendations.
Figure 5 presents similar designer workflows for Scenes S2 and S3 .
5 Conclusion
References
1. Amato, A.: Procedural content generation in the game industry. In: Korn, O., Lee,
N. (eds.) Game Dynamics, pp. 15–25. Springer, Cham (2017). https://doi.org/10.
1007/978-3-319-53088-8 2
2. Balint, J.T., Bidarra, R.: Procedural generation of narrative worlds (2021). Sub-
mitted for publication
3. Banerjee, S., Pedersen, T.: An adapted Lesk algorithm for word sense disambigua-
tion using WordNet. In: Gelbukh, A. (ed.) CICLing 2002. LNCS, vol. 2276, pp.
136–145. Springer, Heidelberg (2002). https://doi.org/10.1007/3-540-45715-1 11
4. Caliskan, A., Bryson, J.J., Narayanan, A.: Semantics derived automatically from
language corpora contain human-like biases. Science 356(6334), 183–186 (2017).
https://arxiv.org/abs/1608.07187
5. Castaño, O., Kybartas, B., Bidarra, R.: Talebox-a mobile game for mixed-initiative
story creation. In: Proceedings of DiGRA-FDG (2016). https://graphics.tudelft.nl/
Publications-new/2016/CKB16/
6. Day, W.H., Edelsbrunner, H.: Efficient algorithms for agglomerative hierarchi-
cal clustering methods. J. Classif. 1(1), 7–24 (1984). https://doi.org/10.1007/
BF01890115
7. Deterding, S., et al.: Mixed-initiative creative interfaces. In: Proceedings of the 2017
CHI Conference Extended Abstracts on Human Factors in Computing Systems,
pp. 628–635 (2017). https://doi.org/10.1145/3027063.3027072
8. Di Fabio, A., Conia, S., Navigli, R.: VerbAtlas: a novel large-scale verbal seman-
tic resource and its application to semantic role labeling. In: Proceedings of the
2019 Conference on Empirical Methods in Natural Language Processing and the
9th International Joint Conference on Natural Language Processing (EMNLP-
IJCNLP), pp. 627–637 (2019). https://doi.org/10.18653/v1/D19-1058
9. Gupta, T., Schwing, A., Hoiem, D.: ViCo: word embeddings from visual co-
occurrences. In: Proceedings of the IEEE/CVF International Conference on Com-
puter Vision, pp. 7425–7434 (2019). https://arxiv.org/abs/1908.08527
10. Hartsook, K., Zook, A., Das, S., Riedl, M.O.: Toward supporting stories with pro-
cedurally generated game worlds. In: 2011 IEEE Conference on Computational
Intelligence and Games (CIG 2011), pp. 297–304. IEEE (2011). https://doi.org/
10.1109/CIG.2011.6032020
11. Kottur, S., Vedantam, R., Moura, J.M., Parikh, D.: Visual word2vec (vis-w2v):
learning visually grounded word embeddings using abstract scenes. In: Proceedings
of the IEEE Conference on Computer Vision and Pattern Recognition, pp. 4985–
4994 (2016). https://arxiv.org/abs/1511.07067
12. Kybartas, B., Bidarra, R.: A semantic foundation for mixed-initiative computa-
tional storytelling. In: Schoenau-Fog, H., Bruni, L.E., Louchart, S., Baceviciute, S.
(eds.) ICIDS 2015. LNCS, vol. 9445, pp. 162–169. Springer, Cham (2015). https://
doi.org/10.1007/978-3-319-27036-4 15
13. Kybartas, B., Bidarra, R.: A survey on story generation techniques for authoring
computational narratives. IEEE Trans. Comput. Intell. AI Games 9(3), 239–253
(2016). https://doi.org/10.1109/TCIAIG.2016.2546063
14. Liapis, A., Yannakakis, G.N., Nelson, M.J., Preuss, M., Bidarra, R.: Orchestrating
game generation. IEEE Trans. Games 11(1), 48–68 (2018). https://doi.org/10.
1109/TG.2018.2870876
15. Liapis, A., Yannakakis, G.N., Togelius, J.: Sentient sketchbook: computer-aided
game level authoring. In: FDG, pp. 213–220 (2013). http://julian.togelius.com/
Liapis2013Sentient.pdf
Mixed-Initiative Narrative World Creation 27
16. Linden, R.V.D., Lopes, R., Bidarra, R.: Designing procedurally generated levels.
In: Proceedings of the second AIIDE workshop on Artificial Intelligence in the
Game Design Process, November 2013
17. Liu, Z.Q., Leung, K.M.: Script visualization (ScriptViz): a smart system that makes
writing fun. Soft. Comput. 10(1), 34–40 (2006). https://doi.org/10.1007/s00500-
005-0461-4
18. Miller, G.A.: WordNet: An Electronic Lexical Database. MIT Press, Cambridge
(1998). https://doi.org/10.7551/mitpress/7287.001.0001
19. Navigli, R., Ponzetto, S.P.: BabelNet: the automatic construction, evaluation and
application of a wide-coverage multilingual semantic network. Artif. Intell. 193,
217–250 (2012). https://doi.org/10.1016/j.artint.2012.07.001
20. Pennington, J., Socher, R., Manning, C.D.: GloVe: global vectors for word repre-
sentation. In: Proceedings of the 2014 Conference on Empirical Methods in Natural
Language Processing (EMNLP), pp. 1532–1543 (2014). https://doi.org/10.3115/
v1/D14-1162
21. Porteous, J., Ferreira, J.F., Lindsay, A., Cavazza, M.: Extending narrative plan-
ning domains with linguistic resources. In: Proceedings of the 19th International
Conference on Autonomous Agents and Multiagent Systems (AAMAS 2020), pp.
1081–1089. International Foundation for Autonomous Agents and Multiagent Sys-
tems (IFAAMAS) (2020). https://dl.acm.org/doi/abs/10.5555/3398761.3398887
22. Porteous, J., Lindsay, A., Read, J., Truran, M., Cavazza, M.: Automated extension
of narrative planning domains with antonymic operators. In: Proceedings of the
14th International Conference on Autonomous Agents and Multiagent Systems
(AAMAS 2015), May 2015. https://doi.org/10.5555/2772879.2773349
23. Rousseeuw, P.J.: Silhouettes: a graphical aid to the interpretation and validation
of cluster analysis. J. Comput. Appl. Math. 20, 53–65 (1987). https://doi.org/10.
1016/0377-0427(87)90125-7
24. Samuel, B., Mateas, M., Wardrip-Fruin, N.: The design of writing buddy: a mixed-
initiative approach towards computational story collaboration. In: Nack, F., Gor-
don, A.S. (eds.) ICIDS 2016. LNCS, vol. 10045, pp. 388–396. Springer, Cham
(2016). https://doi.org/10.1007/978-3-319-48279-8 34
25. Schlechtweg, D., Oguz, C., Walde, S.S.I.: Second-order co-occurrence sensitivity
of skip-gram with negative sampling. arXiv preprint arXiv:1906.02479 (2019).
https://arxiv.org/abs/1906.02479
26. Shaker, N., Shaker, M., Togelius, J.: Ropossum: an authoring tool for designing,
optimizing and solving cut the rope levels. In: Ninth Artificial Intelligence and
Interactive Digital Entertainment Conference (2013). https://ojs.aaai.org/index.
php/AIIDE/article/view/12611
27. Shaker, N., Togelius, J., Nelson, M.J.: Procedural Content Generation in Games.
Springer, Heidelberg (2016). https://doi.org/10.1007/978-3-319-42716-4
28. Skorupski, J., Jayapalan, L., Marquez, S., Mateas, M.: Wide ruled: a friendly inter-
face to author-goal based story generation. In: Cavazza, M., Donikian, S. (eds.)
ICVS 2007. LNCS, vol. 4871, pp. 26–37. Springer, Heidelberg (2007). https://doi.
org/10.1007/978-3-540-77039-8 3
29. Smelik, R.M., Tutenel, T., Bidarra, R., Benes, B.: A survey on procedural modelling
for virtual worlds. Comput. Graph. Forum 33(6), 31–50 (2014). https://doi.org/
10.1111/cgf.12276
30. Smelik, R.M., Tutenel, T., de Kraker, K.J., Bidarra, R.: Interactive creation of
virtual worlds using procedural sketching. In: Eurographics (Short papers), pp.
29–32 (2010). https://doi.org/10.2312/egsh.20101040
28 M. R. B. Perez et al.
31. Smith, G., Whitehead, J., Mateas, M.: Tanagra: a mixed-initiative level design tool.
In: Proceedings of the Fifth International Conference on the Foundations of Digital
Games, pp. 209–216. ACM (2010). https://doi.org/10.1145/1822348.1822376
32. Widdows, D.: Orthogonal negation in vector spaces for modelling word-meanings
and document retrieval. In: Proceedings of the 41st Annual Meeting of the Associ-
ation for Computational Linguistics, pp. 136–143 (2003). https://doi.org/10.3115/
1075096.1075114
33. Xu, C., Chen, Z., Li, C.: Obj-GloVe: scene-based contextual object embedding.
arXiv preprint arXiv:1907.01478 (2019). https://arxiv.org/abs/1907.01478
34. Yannakakis, G.N., Liapis, A., Alexopoulos, C.: Mixed-initiative co-creativity. In:
FDG (2014). https://www.um.edu.mt/library/oar/handle/123456789/29459
Ressaca and Dispersão: Experiments
in Non-linear Cinema
1 Introduction
This demo paper describes two interactive cinema experiences created by the
author, namely Ressaca (Hangover) and Dispersão (Dispersion). Both are meant
for movie theaters, have feature film length (between 75 and 120 min) and por-
tray fictional stories based in local historical events. They also share a collective
device of agency and a non-branching narrative based on audiovisual lexia [8]
which are the movie sequences. We will describe narrative, interactive and tech-
nical aspects and situate them in the context of interactive visual storytelling.
2 Ressaca
The movie follows the puberty and teenage years of a middle-class Brazilian
during the turbulent times of re-democratization in the 80’s and 90’s. During that
period, the country faced economic crisis, hyperinflation, dead and impeached
presidents.
The project’s interaction device is a 1-m round touch screen that stands
between the main screen and the audience. The device (Fig. 1), baptized as
c Springer Nature Switzerland AG 2021
A. Mitchell and M. Vosmeer (Eds.): ICIDS 2021, LNCS 13138, pp. 29–33, 2021.
https://doi.org/10.1007/978-3-030-92300-6_3
30 B. C. Vianna
Engrenagem (Sprocket) was developed and programmed by Maı́ra Sala [2], and
is inspired by the musical interface named Reactable [5]. It is manipulated any
one person with sufficient knowledge of the material and interface. It works
effectively as an editing tool, allowing the user to create an order for the pre-
edited sequences and even the individual shots of the film in real time. The usual
session starts with 5–10 sequences overhead, and subsequent lexia were added
impromptu, according to the will of the editor, which could be influenced by the
reaction of the audience, his or her mood, or a specific goal like focusing on one
of the stories. The movie has 128 sequences in total [6], and a session could use
anything between fifty to eighty from them.
Fig. 1. The director between the main theater screen and the Engrenagem interface
Ressaca was released in 2008 and exhibited in several festivals and venues
throughout the world until 2011. It won four awards, including Best Film in
Cinesquemanovo 2009 [7], in Porto Alegre, Brazil. This was the only festival
in which it took part of the competition. In others, the project was shown in
non-competitive screenings.
The goal of the project was to offer a non-linear experience that didn’t disrupt
the immersive cinema experience by asking the audience to make choices. The
“editor” – the director himself, most of the times – assumes a role of representing
the audience [9]. At the same time, the imposing presence of the interface and the
Ressaca and Dispersão: Experiments in Non-linear Cinema 31
3 Dispersão
In 2018, we started a new project continuing our research on cinema and non-
linear narratives. Dispersão (Dispersion) [3] is also a feature-length experience
created for movie theaters, and offers a collective narrative agency. But instead
of relying on a representative to decide on the storytelling route, this project
puts the structuring of the sequences on the hands of viewers.
The audience of the film receives information on the story and characters
through two different streams: the first and more visible is the big theater screen,
where live action scenes take place. And the other is the mobile phone of each
viewer. A social network app was created specially for the project, again by
creative developer Maı́ra Sala. This software imitates the mechanism of a micro-
blogging platform, except that its participants are not the moviegoers, but the
characters of the movie. Thus, events on the main story told on the theater
screen might trigger some posts and comments inside the app.
Even though the spectators cannot make posts, they can interact with the app
by reacting to them. And that is exactly the interactive strategy for storytelling:
the system will choose the next sequences depending on the engagement with
each post. The rationale here follows the logic of existing commercial social
networks. If a posts provokes more engagement it means that the audience is
interested in the character, and/or the plot around it. In the case of a platform
like Twitter, this would make the post be shown to a greater number of users (as
it proves to be more popular) and would also bring up more posts related to it to
the particular user who “liked” it. And in the case of the collective audience of
Dispersão in the movie theater, this would have the effect of guiding the stories
towards that plot and/or character, by selecting specific sequences.
32 B. C. Vianna
4 Conclusion
When discussing the future of the Interactive Digital Narratives field, Murray
proposes this image of a kaleidoscope of multiple taxonomies and artifacts [4].
The projects described here are to be found in a very specific branch of this
looking glass, the one dedicated to collective interactive experiences that take
place inside the movie theater. But even within this small field, they belong to
a more exotic leaf which aggregates the ones produced in the periphery of the
western world - in developing countries.
It is hard to claim novelty on such projects when there isn’t even a reference
frame to discuss them in the same language. A quick lookup on an academic
search engine reveals three pages of articles dedicated to the project Ressaca
written in Portuguese or Spanish, and not one single document in English, even
after more than a decade from its inception.
Practitioners from the global south are used to being subject to “discoveries”,
specially if their practice is performed in non hegemonic languages. This demo
proposes to fill a gap in the documentation of such practice. By opening a small
crack in this wall, I hope that new projects - like “Dispersão” - can find a voice
of their own, free from a colonialist perspective.
Ressaca and Dispersão: Experiments in Non-linear Cinema 33
References
1. Bordwell, D.: Narration in the Fiction Film. Routledge, Abingdon (2013)
2. Coutinho, E.L., Pinto, I.: A participação do espectador no filme Ressaca (2011)
3. Dispersão website. https://web.archive.org/web/20190710073406/, http://
dispersao.net/. Accessed 11 Oct 2021
4. Murray, J.H.: Research into interactive digital narrative: a kaleidoscopic view. In:
Rouse, R., Koenitz, H., Haahr, M. (eds.) ICIDS 2018. LNCS, vol. 11318, pp. 3–17.
Springer, Cham (2018). https://doi.org/10.1007/978-3-030-04028-4 1
5. Reactable. https://en.wikipedia.org/wiki/Reactable. Accessed 11 Oct 2021
6. Ressaca edited scenes. https://archive.org/details/ressaca. Accessed 11 Oct 2021
7. Ressaca wins Cinemaesquemanovo Festival. https://web.archive.org/web/
20091029191702/https://g1.globo.com/Noticias/Cinema/0,,MUL1354133-7086,
00.html. Accessed 11 Oct 2021
8. Ross, C.: Hypertext 2.0: The Convergence of Contemporary Critical Theory (1999)
9. Rodrigues, A.C.: Live Cinema: Autoria Coletiva em Narrativas Objetivas e Subje-
tivas (2015)
10. van Dyke Parunak, H.: Hypermedia topologies and user navigation. In: Proceedings
of the Second Annual ACM Conference on Hypertext, pp. 43–50 (1989). https://
doi.org/10.1145/74224.74228
Pedagogical Challenges in Social Physics
Authoring
1 Introduction
Emergent narratives are narrative structures which are not predefined but rather
generated by users through interactions and choices [2,13]. These narratives have
exponentially larger spaces of possible playthroughs as compared to more tra-
ditional narrative experiences, including modern video games using branching
narrative logic [26]. Social physics engines like Comme il Faut (CiF) [20] and
its spiritual successor Ensemble [27] offer one approach to the creation of emer-
gent narratives by leveraging concepts relevant to the broader domain of social
physics, including network theory and predicate logic. Social physics games can
c Springer Nature Switzerland AG 2021
A. Mitchell and M. Vosmeer (Eds.): ICIDS 2021, LNCS 13138, pp. 34–47, 2021.
https://doi.org/10.1007/978-3-030-92300-6_4
Pedagogical Challenges in Social Physics Authoring 35
take the form of social puzzles [17], where players must try to anticipate the
potential social effects of their actions [18], considering, for example, the range
of likely emotional responses by non-player characters [12]. Social physics sys-
tems in the CiF family have been implemented in mods to AAA games such as
Skyrim and Conan Exiles, with user studies suggesting that players prefer and
gravitate toward NPCs guided by such engines [10,11,23].
Unlike pre-written narratives, however, the authoring of content for the CiF
and Ensemble engines requires the composition of predicate logic defining the
logical structures and rules of the interactive world, including the realm of pos-
sible truths and states–known as a schema–as well as the individual rules which
govern social behaviors and the range of actions that players and NPCs can take.
Previous research has identified general challenges in the field of interactive dig-
ital narrative storytelling [14], surveyed issues presented by existing interactive
narrative authoring tools [9], and examined pain points in interactive narrative
authoring caused by authors’ lack of knowledge of underlying computational
models [31]. However, the existing research has not yet examined the unique
challenges of authoring for systems like Ensemble, wherein content is authored
solely as first-order predicate logic rules rather than narrative fragments.
The authoring process may differ greatly depending on the goals of the
authors. For example, in early applications of the CiF and Ensemble engines,
content authors generally took an informal approach; for example, the authors
of the game Prom Week –consisting primarily of computer scientists–strove to
approximate the sociological and psychological situations portrayed by films in
the “teen movies” genre [28], using examples from films of the genre for ref-
erence [19]. Subsequent projects like VESPACE and Vox Populi: The Ustra-
dian Games, however, have required more stringent authoring processes. Both
of these projects, being multidisciplinary collaborations between researchers in
artificial intelligence, the humanities, and social sciences, required the training
of researchers from outside the field of computer science in social physics author-
ing. In this paper, we will introduce these two projects and outline the various
difficulties in author training which they have revealed, as well as the numer-
ous methods which we will employ in our tutorial game in order to address and
circumvent them.
Based on experience and feedback from VESPACE and Vox Populi, several
pedagogical issues appear to occur with some regularity in collaborative social
physics authoring efforts, and our hope is that documenting these challenges will
aid future teachers and authors in their authoring endeavors. In order to address
these difficulties head on, the training hurdles enumerated in our analysis have
inspired our current tutorial game project, which seeks to guide users in effec-
tive social physics authoring by leveraging the unique pedagogical opportunities
afforded by open-ended gameplay; much of our tutorial game design has come
directly from written responses that VESPACE workshop participants took part
in and which we will draw from in this paper.
To address the challenges identified in our analysis, we have designed and
prototyped a tutorial game to train users by reinforcing learning through explo-
ration, so that learners can immediately experience the expressiveness of the
36 D. DeKerlegand et al.
Ensemble engine through real-time feedback, all while being guided by various
challenges and motivational goals. In the game, players will guide a household
of pets through social interactions, exploring the narrative and social effects of
authorial choices as they complete challenges and unlock new authoring capabili-
ties. The term “SimCity effect” has been used to describe video games possessing
this ability to develop player understanding of internal structures and mecha-
nisms through gameplay [21,34,35]. We have sought to design our tutorial game
with the SimCity effect in mind, utilizing open-ended gameplay and exploration,
along with motivating techniques such as skill trees, in order to instill a famil-
iarity with the functionality and potentiality of the Ensemble engine.
2 Authoring Projects
In this paper, we focus on the authoring challenges of two major collaborative
efforts, the VESPACE and Vox Populi: The Ustradian Games projects; we chose
these two projects based on their complexity and the fact that both projects were
interdisciplinary in nature, consisting of collaboration between computer science
and artificial intelligence researchers on one hand and humanities and social sci-
ence scholars on the other. In contrast with other social physics-based interactive
narrative games, the authoring for these projects was largely performed by par-
ticipants with no previous training in social physics or computer science. The
VESPACE project included an authoring collaboration with French literary his-
torians, which was conducted through numerous authoring workshops that ran
in parallel with the development of a collaborative web-based authoring tool.
The Vox Populi: The Ustradian Games project saw collaboration with educa-
tion and assessment specialists in order to develop a game which challenges users
to think critically as they explore an unfamiliar culture.
Authoring for the VESPACE project has required the construction of a large
corpus of rules and a schema tailored to the social norms of 18th century France;
importantly, this content must be written by literary historians, with citations
supporting each authoring decision. Before the authoring process could begin, we
needed to conduct significant training workshops for prospective authors, as well
as to develop a new authoring tool facilitating collaborative authoring. We first
conducted a series of short workshops as we iterated through the development
sprints of the tool, ultimately working up to the deployment of the tool and its
use in a week-long remote workshop [6]. The first round of VESPACE author
training included four half-day workshops, attended by a group of six and guided
by three of the participants. Two of the three workshop leaders were computer
science researchers, while the other leader was a professor of literary history
who had undergone some one-on-one training in social physics concepts and
terminologies. The other three participants were graduate students in literary
history. These short workshops held a three-fold value: as training sessions for
authors, as a means of identifying significant pedagogical challenges in the use
of Ensemble, and as beta-testing scenarios for the authoring tool.
The second round of VESPACE author training consisted of one week of
day-long workshops and was attended by eight participants, including the previ-
ous trio of leaders and five graduate and PhD students from various humanities
disciplines. The end result of this workshop was a fully-realized schema for VES-
PACE consisting of 10 categories and a total of 113 types, as well as a rich rule
set of 230 rules. Following this week-long workshop, we tasked all eight partici-
pants with composing an 8–12 page formal reflection outlining their experience,
asking them to describe any difficulties they faced. We then conducted an infor-
mal analysis of these reflections, maintaining a list of challenges described by the
authors. Following the informal analysis, we performed a more thorough anal-
ysis, keeping track of the frequency with which each challenge was mentioned,
combining closely related challenges into categories, looking for keywords indi-
cating a particular challenge’s severity, and collecting illustrative quotes for the
paper. The results of this analysis are an important source of experiential data
concerning the authorial challenges which we will codify and analyze in this
paper.
eral and abstract. Some VESPACE authors found this particularly challenging,
with one workshop participant writing that “Complex and less concrete elements
in the schema were much more difficult to agree upon and incorporating them
revealed both the constraints and the possibilities of thinking within the Ensem-
ble authoring system. For example, my group was very concerned with the issue
of theater etiquette, but baffled by the question of how the intricate rules of
eighteenth-century public conduct could be summed up by the kind of one-word
terms that comprised the schema.”
currently provide a means of determining how much of that potential rule space
has been covered by authored content. We have developed a general heuristic
for approaching this problem in the form of a set authoring order; during the
authoring process, we first consider all relationships, then networks, then sta-
tuses, and so on. However, this heuristic approach should ultimately be replaced
by more exact methods, for example author assistance visualization techniques
[8] informing authors on the current authored rule space, which could be filtered
by category or type.
Actions in Ensemble define the range of all possible decisions that a player or
NPC can make in a game. Because actions have preconditions and effects (how
the action changes the social state), authors must make deliberate choices when
composing both rules and actions, so that rules will have a real effect on the
range of possible and probable actions, making for rich social experiences.
learn by doing, which means diminishing the time to feedback or feedback loop
becomes an important pedagogical tool. If authors can visualize, experience,
and play with the effects of their authorial choices, they can begin to develop an
intuitive sense of how authoring decisions relate to the creation of social worlds.
Previous research has identified expressive range as a measure of genera-
tive space, whereby a set of quantitative metrics and models can be defined for
graphing and analyzing an expressive range with an ultimate goal of qualita-
tive interpretation and analysis [16,29,30]. Others have discussed methods for
expanding expressive range [15,32], but there may also be uses for decreasing or
modulating expressive range depending on the context. Our tutorial game seeks
to intentionally constrain the expressive range of the system as players begin,
making the range manageable for learning, then gradually releasing that expres-
sive range as players accomplish in-game goals in order to guide them naturally
through the learning experience.
see the effects of their choices gradually play out in the social world, experienc-
ing immediate feedback in the form of character actions and responses that will
illustrate the expressive range of the Ensemble engine.
users to expand social rule coverage in order to make certain behaviors and
responses possible, with the anticipation that experiencing these effects in real
time will give authors a more intuitive sense of the correlation between rule
authoring and the range of possible social behaviors. We will begin by allowing
players to only activate rules from a predetermined set, requiring them to select
rules wisely in order to achieve some social responses in the game. As illustrated
in Fig. 2(a), not every rule is initially active; only volition rules managed by the
player will be leveraged by the game. As a core aspect of gameplay, players can
activate rules and experience immediate feedback as they change the range of
narrative potential. Figure 2(b) shows that once a player has activated the rule
“Social animals want to become playmates” and brought new characters into the
game world having the “social” trait, those characters can now be affected by
the rule, leading them to potentially become playmates. This is a pedagogically
valuable interaction whereby players learn the effects of authoring by experienc-
ing them directly. Gradually, players can unlock the capacity to modify existing
rules and to forge new rules, but only after they have witnessed the narrative
effects of changes in rule space.
Eventually, as players level up their rule authoring skills, they will gain the
option of creating entirely new rules in order to accomplish some goal. As players
modify the list of active rules, we will provide visualizations of social rule cov-
erage and existing rule space, in order to illustrate the effects of rule authoring
in a more theoretical sense. Due to the importance of weight balancing in the
Ensemble authoring process, yet another branch of rule authoring on the skill
tree will involve weight tuning, wherein players will gain the ability to mod-
ify rule weights so that the impacts of particular rules will change, with quests
challenging the player to modify rule weights in order to achieve some goal.
5 Conclusions/Future Work
One of the strengths of social physics is its ability to foster collaboration and
creative problem solving across disciplines; due to its underlying reliance on
first-order logic predicates as opposed to statistical models, researchers and cre-
ators of various backgrounds can leverage this authoring paradigm to develop
immersive, rich interactive experiences. However, there is a learning curve to
the authoring process, which means it will be important to produce effective
authoring tools and learning materials that can address the particular difficul-
ties involved in social physics authoring. In this paper, we have compiled data
from previous authoring collaborations in order to codify and analyze the most
persistent and troublesome challenges in social physics authoring pedagogy. We
have also introduced our tutorial game design, which will minimize the feedback
loop and gradually introduce player-authors to Ensemble paradigms in intuitive
ways, directly addressing the challenges laid out in the paper through the design
choices enumerated. With our analysis of pedagogical challenges complete and
the tutorial game designed and prototyped, our next step will be to complete
development of the game and submit it to user testing and user studies, adjusting
game elements as necessary based on user feedback and success studies.
46 D. DeKerlegand et al.
References
1. Antoun, C., Antoun, M., Ryan, J.O., Samuel, B., Swanson, R., Walker, M.A.:
Generating natural language retellings from prom week play traces. In: Proceedings
of the PCG in Games (2015)
2. Aylett, R.: Emergent narrative, social immersion and “storification”. In: Proceed-
ings of the 1st International Workshop on Narrative and Interactive Learning Envi-
ronments, pp. 35–44 (2000)
3. Carmichael, G., Mould, D.: A framework for coherent emergent stories. In: FDG
(2014)
4. Chen, S., Nelson, M., Mateas, M.: Evaluating the authorial leverage of drama
management. In: Proceedings of the AAAI Conference on Artificial Intelligence
and Interactive Digital Entertainment, vol. 4 (2009)
5. Cole, A.: Connecting player and character agency in videogames. Text Special (49)
(2018)
6. DeKerlegand, D., Samuel, B., Leichman, J.: Encoding socio-historical exegesis as
social physics predicates. In: International Conference on the Foundations of Digital
Games, pp. 1–9 (2020)
7. François, P., Laroche, F., Leichman, J., Rubellin, F.: Mettre en place les fondations
d’un outil de restitution à large audience: Vespace (2018)
8. Garbe, J., Reed, A.A., Dickinson, M., Wardrip-Fruin, N., Mateas, M.: Author
assistance visualizations for ice-bound, a combinatorial narrative. In: FDG (2014)
9. Green, D., Hargood, C., Charles, F.: Contemporary issues in interactive storytelling
authoring systems. In: Rouse, R., Koenitz, H., Haahr, M. (eds.) ICIDS 2018. LNCS,
vol. 11318, pp. 501–513. Springer, Cham (2018). https://doi.org/10.1007/978-3-
030-04028-4 59
10. Guimaraes, M., Santos, P., Jhala, A.: CiF-CK: an architecture for social NPCs in
commercial games. In: 2017 IEEE Conference on Computational Intelligence and
Games (CIG), pp. 126–133. IEEE (2017)
11. Guimarães, M., Santos, P., Jhala, A.: Prom week meets skyrim (2017)
12. Jenkins, H.: Game design as narrative architecture. Computer 44(53), 118–130
(2004)
13. Koenitz, H.: Narrative in video games (2019)
14. Koenitz, H., Eladhari, M.P.: Challenges of IDN research and teaching. In: Cardona-
Rivera, R.E., Sullivan, A., Young, R.M. (eds.) ICIDS 2019. LNCS, vol. 11869, pp.
26–39. Springer, Cham (2019). https://doi.org/10.1007/978-3-030-33894-7 4
15. Thue, D., Schiffel, S., Guðmundsson, T.Þ., Kristjánsson, G.F., Eirı́ksson, K.,
Björnsson, M.V.: Open world story generation for increased expressive range. In:
Nunes, N., Oakley, I., Nisi, V. (eds.) ICIDS 2017. LNCS, vol. 10690, pp. 313–316.
Springer, Cham (2017). https://doi.org/10.1007/978-3-319-71027-3 33
16. Kybartas, B., Verbrugge, C., Lessard, J.: Expressive range analysis of a possible
worlds driven emergent narrative system. In: Rouse, R., Koenitz, H., Haahr, M.
(eds.) ICIDS 2018. LNCS, vol. 11318, pp. 473–477. Springer, Cham (2018). https://
doi.org/10.1007/978-3-030-04028-4 54
17. McCoy, J., Treanor, M., Samuel, B., Mateas, M., Wardrip-Fruin, N.: Prom week:
social physics as gameplay. In: Proceedings of the 6th International Conference on
Foundations of Digital Games, pp. 319–321 (2011)
18. McCoy, J., Treanor, M., Samuel, B., Reed, A.A., Wardrip-Fruin, N., Mateas, M.:
Prom week. In: Proceedings of the International Conference on the Foundations of
Digital Games, pp. 235–237 (2012)
Pedagogical Challenges in Social Physics Authoring 47
19. McCoy, J., Treanor, M., Samuel, B., Tearse, B., Mateas, M., Wardrip-Fruin, N.:
Authoring game-based interactive narrative using social games and Comme il Faut.
In: Proceedings of the 4th International Conference & Festival of the Electronic
Literature Organization: Archive & Innovate, vol. 50. Citeseer (2010)
20. McCoy, J., Treanor, M., Samuel, B., Reed, A.A., Mateas, M., Wardrip-Fruin, N.:
Social story worlds with Comme il Faut. IEEE Trans. Comput. Intell. AI Games
6(2), 97–112 (2014)
21. Mitchell, A.: Reflective rereading and the SimCity effect in interactive stories. In:
Schoenau-Fog, H., Bruni, L.E., Louchart, S., Baceviciute, S. (eds.) ICIDS 2015.
LNCS, vol. 9445, pp. 27–39. Springer, Cham (2015). https://doi.org/10.1007/978-
3-319-27036-4 3
22. Moallem, J.D., Raffe, W.L.: A review of agency architectures in interactive drama
systems. In: 2020 IEEE Conference on Games (CoG), pp. 305–311. IEEE (2020)
23. Morais, L., Dias, J., Santos, P.A.: From caveman to gentleman: a CiF-based social
interaction model applied to Conan exiles. In: Proceedings of the 14th International
Conference on the Foundations of Digital Games, pp. 1–11 (2019)
24. Rubellin, F., François, P.: Le théâtre du xviiie siècle, plus vivant que jamais (2018)
25. Ryan, J., Walker, M.A., Wardrip-Fruin, N.: Toward recombinant dialogue in inter-
active narrative. In: Seventh Intelligent Narrative Technologies Workshop (2014)
26. Samuel, B., et al.: Playing the worlds of prom week. In: Narrative Theory, Litera-
ture, and New Media, pp. 87–105. Routledge (2015)
27. Samuel, B., Reed, A.A., Maddaloni, P., Mateas, M., Wardrip-Fruin, N.: The ensem-
ble engine: next-generation social physics. In: Proceedings of the Tenth Interna-
tional Conference on the Foundations of Digital Games (FDG 2015), pp. 22–25
(2015)
28. Samuel, B.: Crafting stories through play. Ph.D. thesis, UC Santa Cruz (2016)
29. Smith, G., Whitehead, J., Mateas, M.: Tanagra: a mixed-initiative level design
tool. In: Proceedings of the Fifth International Conference on the Foundations of
Digital Games, pp. 209–216 (2010)
30. Smith, G., Whitehead, J., Mateas, M.: Tanagra: reactive planning and constraint
solving for mixed-initiative level design. IEEE Trans. Comput. Intell. AI Games
3(3), 201–215 (2011)
31. Spierling, U., Szilas, N.: Authoring issues beyond tools. In: Iurgel, I.A., Zagalo,
N., Petta, P. (eds.) ICIDS 2009. LNCS, vol. 5915, pp. 50–61. Springer, Heidelberg
(2009). https://doi.org/10.1007/978-3-642-10643-9 9
32. Summerville, A.: Expanding expressive range: evaluation methodologies for pro-
cedural content generation. In: Fourteenth Artificial Intelligence and Interactive
Digital Entertainment Conference (2018)
33. Treanor, M., McCoy, J., Sullivan, A.: A framework for playable social dialogue. In:
Proceedings of the AI and Interactive Digital Entertainment Conference (AIIDE
2016) (2016)
34. Wardrip-Fruin, N.: Three play effects-eliza, tale-spin, and sim city. Digital Human-
ities, pp. 1–2 (2007)
35. Wardrip-Fruin, N.: Expressive Processing: Digital Fictions, Computer Games, and
Software Studies. MIT Press, Cambridge (2009)
An Approach to Multiplayer Interactive
Fiction
1 Introduction
For a long time, people have been using storytelling as a way of entertainment.
In recent years, Tabletop Role-playing Games (TTRPG)1 have felt a resurgence
in popularity [5,6]. It is not far-fetched to assume that, under the right orien-
tation, TTRPG players would also enjoy partaking in Interactive Fiction (IF)2 .
Some TTRPG feature the possibility of separating private information and pub-
lic information, i.e., a player has access to certain information based on their
1
Tabletop Role-playing Games: a form of role-playing game in which the partic-
ipants describe their characters’ actions through speech.
2
Interactive Fiction: a software simulating environments in which players use text
commands to control characters and influence the environment.
c Springer Nature Switzerland AG 2021
A. Mitchell and M. Vosmeer (Eds.): ICIDS 2021, LNCS 13138, pp. 48–60, 2021.
https://doi.org/10.1007/978-3-030-92300-6_5
An Approach to Multiplayer Interactive Fiction 49
character and the information that makes sense their character to know while
keeping it private from other player characters. This happens while simultane-
ously having the players share the same table and talk with each other, both
in character and out of character, giving them the chance to share their private
information. TTRPG also have a very strong social component. Being able to
play with your friends enhances the experience in a very significant way. Being
able to play these kinds of games with a heavy focus on story creation where
the player takes an active part but is at the same time interacting with other
players as they wish to is something not many digital games have been able to
emulate. This act of communication and companionship helps build new rela-
tionships and create greater investment in the characters and story. For these
reasons, we propose the creation of a new interaction paradigm that fulfills these
requirements and offers a fully fleshed-out two-player experience in the form of
interactive fiction.
3 Interaction Paradigm
Our main objective is to develop a short narrative experience based on the fol-
lowing private vs public information paradigm (see diagram in Fig. 1). We will
use a ‘couch co-op’ approach, where players have access to a shared computer
screen and an individual phone screen that also acts as a controller. We will
focus on the interaction between two players, each controlling a different char-
acter. There is a public space that is public to both characters and players. It
consists of a shared screen. Whatever is displayed on the screen is known by both
3
A Way Out, https://www.ea.com/games/a-way-out. Last accessed May 19, 2021.
4
Monster Prom, http://monsterprom.pizza. Last accessed May 19, 2021.
50 M. Farias and C. Martinho
characters and can be read by both players. There is a semi-private space that
is public for the players, but private to the characters. It consists of the phys-
ical space the players share (e.g. sharing a room or communicating via a voice
call). Players use this space to discuss strategies or the story as it progresses,
and influence their character’s behavior, but this is private information to their
characters. Finally, there is a private space that is private to the players and
characters. This is represented by each phone screen. Whatever is displayed on
the phone is private information that the respective player and character know
but is private to the rest of the players and their characters.
4 Related Work
the actions taken by the currently playing character, without the possibility of
interference.
the second: the second player will be starved of agency over the story. A way
to prevent starvation is to employ gating, i.e. the system stays on hold until
both players have finished their sections and choose to progress. However, the
employment of gating creates deadlocks: when one player finishes reading their
section before the other player, then the first player will experience a deadlock,
i.e. they won’t be able to proceed until the second player finishes reading their
own section, leaving them stuck without anything else to do.
5 Approach
We decided to create an interactive fiction experience that runs for approxi-
mately 40 min where two players play simultaneously, each having a different
character. We constructed a short story that follows three housemates that do
not take each other too seriously. Players control two of these friends while trying
to coordinate a surprise birthday party for the third friend, choosing the per-
fect gift for the occasion. The story is divided into three parts: a shared section
where players meet and discuss together the upcoming party, a split section
where each character goes to a different store to buy a birthday present and
possibly a birthday cake, moving separately, and a final scene where the players
are back together with their friend for the party and gift exchange.
The story encourages players to cooperate. We separated public from private
information: the players are presented with information privately that directly
results from their interactions with the virtual world of the story - this can
come from dialogue with Non-Player Characters (NPC), character’s emotions,
thoughts, etc. They can choose to share this information with the other player
either in its entirety, partially or not at all. This, of course, will influence the
possible choices the players have in future interactions based on their knowledge.
The story progresses and depending on the choices and coordination of the play-
ers, different endings are possible. If the players play as intended, coordinating
their efforts for the birthday party, they might unlock an ending where they
throw the best birthday party of all time. If the players fail this cooperation or
one player tries to misguide their partner, they might end up picking a bad gift
or there might not be a birthday cake at the party.
The players both share a screen but all their inputs are collected through an
application that runs on the player’s phone and collects the player’s choices in
the story. When they are sharing a scene, the public outputs can be displayed
simultaneously for both of them but their decisions are kept private by the
system. This allows for verbal communication so that the players can strategize
their choices or turn the experience into a competition.
6 Prototype
6.1 Story
The story follows a typical branch-out format applying the foldback strategy [2].
Players can diverge through the story and meet up in key moments. There are
An Approach to Multiplayer Interactive Fiction 53
6.2 Application
We developed our application using Unity 8 (see Fig. 2A). To set up a server and
establish connections between the user controllers and the application running
the experience, we set up a Photon Unity Networking (PUN) application with
the Photon Engine 9 . Then, we launch two instances of the application - one
7
Ink, https://www.inklestudios.com/ink/, Last accessed May 19, 2021.
8
Unity, https://unity.com/, Last accessed May 19, 2021.
9
Photon engine, https://www.photonengine.com/en-US/Photon, Last accessed on
May 19, 2021.
54 M. Farias and C. Martinho
for the shared information (main application) and one to receive the private
information and collect user inputs (worker application).
The worker application, starts by asking the players to select their character
(see Fig. 2B) and then connects to the server. The application displays private
information and choices to the player, if available. If there are no choices, the
application still displays the private text but only shows a button with the
instruction continue (see Fig. 2C). When the player makes a choice or decides
to continue, the worker application notifies the main application. The worker
application is built to run on Android platforms, and optimized to run on smart-
phones.
Regarding the main application, it is responsible for starting the experience
and the PUN server. On the first scene (shared) the application loads the ‘ink’
scripts and, once loaded, the application sends the current passage of the story
and the possible choices to the players. At the same time, it must update the
shared interface to display the sprites of the characters as well as the current
text on the public script. Afterward, it awaits feedback from the players. An
example of the shared interface for a shared scene can be viewed in Fig. 3A.
This is followed by a split scene, where the characters are separated in the
story space. The structure of this scene is somewhat similar to a shared scene,
the difference being the presence of two distinct public information sources. Since
the characters are separated, the application only has to wait for the answer of a
player to update the story state of that player. Once a player finishes their story,
the main application must make that player wait until the other player finishes
their story. Once both players are done, the application must ensure the loading
of the next scene. An example of the split interface can be viewed in Fig. 3B.
The final scene is identical to the first, a shared scene with the exception
that previous choices carry over to it. Everything else executes exactly like the
first scene. Once the ‘Ink’ scripts are finished, the story ends.
7 Evaluation
7.1 Experimental Procedure
Our prototype was designed to be tested in a ‘couch co-op’ setting, however, due
to the pandemic situation that coincided with the period of development of our
work, we adapted the procedure. We set up the experiment to be played during
a peer-to-peer call (over Skype, Discord, or equivalent). The main application
is running on the test supervisor environment and its screen is shared with
the players. Each player has the worker application running on their phone.
Communication between players is possible at all times - just like it would be if
players were in the ‘couch co-op’ condition.
Before starting the test, the participants fill the demographics section of the
questionnaire: gender, age, reading habits and opinions, and experiences with
interactive fiction in digital games. Players are then given a brief description of
the experience and instructed on how to start. During the experience, the super-
visor listens and takes note of the following metrics: the interactions between
players - if players were communicative with each other or not and how this
An Approach to Multiplayer Interactive Fiction 55
Fig. 3. A (top) - shared scene on the public interface. B (bottom) - split scene on the
public interface.
varied with the different scenes, as well as the questions asked to the supervisor
regarding the application. Other metrics are being automatically recorded by the
application, namely the average and maximum time a player waits for the other
player’s response for the story to progress. After the experience is complete, the
players are asked to complete the second part of the questionnaire. This section
includes the modules sensory and imaginative immersion, negative affect and
positive affect from the core module of the Game Experience Questionnaire [4].
We also included the entirety of the social presence module to assess how the
players felt about the joint experience. This module of the questionnaire encom-
passes psychological involvement, both empathy and negative feelings as well as
behavioural engagement. We also included some simple questions to determine
how well the paradigm was perceived by the players. After the questionnaire is
complete, we conducted a short semi-structured interview with the participant.
56 M. Farias and C. Martinho
The majority of players appreciated the metaphor and praised the concept:
11 out of the 12 players complimented the novelty of the experience. However,
many expressed concern over the overwhelming amount of information presented
to them, especially during the separated scenes. We asked them if they tried to
follow along with the other player. The majority of players said they did not.
Out of the 6 groups, only one tried to choose adequate gifts with each other.
In situations where one player finished the scene before the other, they started
reading the other’s scene. We asked if it was hard to keep track of both interfaces
at the same time. Players uniformly agreed that it needed a lot of effort and even
then it was not clear where they should be looking. Many players gave suggestions
such as adding a sound effect or vibration effect that would help guide the players’
attention or some visual indication to both players when feedback is expected.
Aligned with the answers to the previous questions, players had no idea of the
actions taken by the other player’s character in the separated scene. Apart from
that, players considered they had a good understanding of the other player’s
character’s actions.
While playing the story, we had a varying range of interaction behaviors.
Some participants spent the entire experience talking about the story while
other groups said nothing at all. Even among the groups that communicated,
not all tried to achieve a good outcome by discussing their choices. We asked the
players what interfered with this communication and what could have been done
An Approach to Multiplayer Interactive Fiction 57
differently to promote communication. Some players answered they did not feel
it was necessary to communicate, they enjoyed playing the interactive fiction as
it was. We postulate that, since we are exploring a new concept of interactive
fiction that most players had not tried before, these groups that did not display
much interaction were focused on how their in-story interactions influenced the
story’s outcome and for that reason tended to ignore the possibility of verbal
interactions. These results, although positive, prompted some changes on our
approach.
A common complaint users had at first was about the difficulty they felt when
shifting their attention between interfaces while playing through the story. To
make the new information more noticeable on the shared screen, we implemented
a flashing effect when new lines of text appear. The new lines are also presented
in a different color (yellow) from the older lines. To notify the player of new
information on the phone, we implemented vibration. This feature is present
in both the shared scenes as well as the split scenes since it is private to each
player, i.e. only the player holding the phone is notified by its vibration. We also
implemented a feature that consisted of visual thought bubbles in the shared
scenes: whenever a user receives private information on their private interface,
a thought bubble pops up above the corresponding character and it stays there
until the player gives feedback. This way, the players have an idea of where
to look and know why the story isn’t progressing in the case that one of the
characters is “still thinking”. This does not apply to the split scene. We also
included a new passage in each script during the split scene to encourage players
to verbally communicate with each other. At a certain point, characters will think
something along the lines of “I wonder what the other character is buying...”.
All playable characters will have this thought at one point or another.
We tested our new version with 6 new pairs of participants. The duration of the
experiment was aligned with the values from the first experiment. Results for
the GEQ were again positive and we registered a noticeable increase in empa-
thy, which was supported by a Mann-Whitney U-Test (U = 38.00, p = 0.049)
(Table 2). This suggests the new features we implemented in our prototype could
help promote empathy between players. Players grasped the concept behind the
dual interface, however, some participants felt the private interface was actively
trying to push towards certain choices - quoting one participant, when trying
to explain what was being presented to him on his phone interface: “The infor-
mation on my phone was directed more for my character and to influence my
choices.”. A possible explanation is that because we are now guiding where the
players should look for novel information, this might have lead players to feel
being manipulated in some capacity.
58 M. Farias and C. Martinho
We asked players what interface they paid more attention to. Responses were
mixed. Some players paid more attention to their own private interface while
others paid attention to the shared interface. In this second experiment, we got
8 out of the 12 players answering that they paid attention to both interfaces
in equal parts – which supports the adequacy of vibration and thought bubbles
- although they admit it is still a bit tiring looking from one interface to the
other. Many players confessed it was hard to keep track of both interfaces since
the shared information moved so fast and a bit taxing looking from one screen
to the other. Other players said the fast movement of information was not that
disruptive since the screen still kept the info for some time. Players pointed out
that, at some points, the experience showed prompts simultaneously on both
screens and they considered this behavior to be confusing.
Many more groups confessed to being attentive to the other player’s behavior
during this round of testing. Players that answered they paid attention to both
characters and recalled specific details about these scenes were also the players
that reported paying attention to both interfaces at the same time. Players that
weren’t paying attention to the other admitted that it is taxing trying to pay
attention to both shared screens, especially when their counterpart is not doing
the same. When asked if they understood all actions taken by the other player’s
character, answers followed the same pattern from the first experiment. People
understood the actions taken by the other character, but were apprehensive
about the origin of other player’s actions.
We also inquired players about how they felt regarding their verbal inter-
actions and what could be done to improve these interactions. Many players
expressed that they did not know they were allowed to talk outside of their char-
acters. Players felt like communication was not necessary to play the story. Other
players still had this communication happen, as before, which helps support our
thesis that players can still communicate and this factor is more dependent on
the player’s personality.
We asked if they felt the vibration and thought bubbles were useful in guid-
ing the players’ attention. Regarding the vibration, the players’ reactions were
mixed. Many found it too disruptive to the experience since the vibration was
a little bit stronger in some devices than we intended it to be. Other play-
ers thought it was helpful to notify them of new information available on the
An Approach to Multiplayer Interactive Fiction 59
8 Closing Guidelines
In this paper, we explored a new paradigm for multiplayer interactive fiction. We
developed a prototype to test our hypothesis and conducted two rounds of tests
with players, implementing mechanisms to help guide the players’ attention. We
concluded that these changes harbored improvements in the overall experience.
As a final conclusion of our work, we defined some simple guidelines a project
similar to the one we tried to develop should follow to be successful, according
to our experience:
– Separate public and private information;
– On split scenes, public information should be described as if it was being seen
from a ‘faraway’ perspective;
– Create a story in a foldback strategy - players should meet on the same shared
scene from time to time;
– Pay attention to starvation and deadlocks, while trying to minimize the input
needed from the player - our passages prompted the player to spam the “con-
tinue” button without much regard to what was written on one of the two
interfaces; personalize feedback instructions - players got tired of pressing
“continue” over and over.
– Keep short story segments grouped by space, i.e., public or private, in longer
story segments - i.e., when instructions are showing up on one screen, the
other should not be getting new instructions simultaneously. Do not show
new information on both screens at the same time, players will not know
where to look.
– Try to keep the story segments the same size for both players, i.e. each passage
takes roughly the same time to read. This will help bridge the impact of
starvation and deadlocks. Make sure the size of the story segments is not
dependent on the players’ choices during split scenes, otherwise, players may
get very different story lengths, creating long waiting times for one of the
players.
– When displaying information on a different screen from the previous screen
information was displayed on, warn the player. In our work, we used vibra-
tion and thought bubbles to help guide this attention. These changes greatly
improved our players’ comprehension of the story.
For more implementation and evaluation details, please refer to [3].
References
1. Bernstein, M.: Card shark and thespis: exotic tools for hypertext narrative. In:
Proceedings of the 12th ACM Conference on Hypertext and Hypermedia, HYPER-
TEXT ’01, pp. 41–50. Association for Computing Machinery, New York (2001).
https://doi.org/10.1145/504216.504233
2. Crawford, C.: Chris Crawford on Interactive Storytelling (New Riders Games), p.
366. New Riders Games (2004)
3. Farias, M.: An approach to multiplayer interactive storytelling. Master’s thesis,
Instituto Superior Técnico, University of Lisbon (2021)
4. IJsselsteijn, W., de Kort, Y., Poels, K.: The game experience questionnaire. Tech-
nische Universiteit Eindhoven (2013)
5. Kosa, M., Spronck, P.: Towards a tabletop gaming motivations inventory (TGMI).
In: Zagalo, N., Veloso, A.I., Costa, L., Mealha, Ó. (eds.) VJ 2019. CCIS, vol. 1164,
pp. 59–71. Springer, Cham (2019). https://doi.org/10.1007/978-3-030-37983-4 5
6. Sousa, M., Bernardo, E.: Back in the game. In: Zagalo, N., Veloso, A.I., Costa, L.,
Mealha, Ó. (eds.) VJ 2019. CCIS, vol. 1164, pp. 72–85. Springer, Cham (2019).
https://doi.org/10.1007/978-3-030-37983-4 6
7. Spawforth, C., Gibbins, N., Millard, D.E.: StoryMINE: a system for multiplayer
interactive narrative experiences. In: Rouse, R., Koenitz, H., Haahr, M. (eds.) ICIDS
2018. LNCS, vol. 11318, pp. 534–543. Springer, Cham (2018). https://doi.org/10.
1007/978-3-030-04028-4 62
8. Spawforth, C., Millard, D.: Multiplayer games as a template for multiplayer narra-
tives: a case study with dark souls. In: HT (2017)
9. Spawforth, C., Millard, D.E.: A framework for multi-participant narratives based
on multiplayer game interactions. In: Nunes, N., Oakley, I., Nisi, V. (eds.) ICIDS
2017. LNCS, vol. 10690, pp. 150–162. Springer, Cham (2017). https://doi.org/10.
1007/978-3-319-71027-3 13
Highlight the Path Not Taken to Add
Replay Value to Digital Storytelling
Games
1 Introduction
did not take to create more desire to replay the game again and explore differ-
ent outcomes. The feedback was composed of storytelling elements to generate
feelings of regret in the players towards the actions they did not take.
1
Twine Homepage, https://twinery.org/.
64 S. Gamito and C. Martinho
things that could have happened but are no longer available for story advance-
ment (e.g. having the maid bump into a table and break a bottle filled with
venom that could have been used to poison a particular character in the story).
The idea is to bring to the attention of the players that things have changed and
are no longer available (e.g. they could have picked up the bottle earlier in the
scene but did not, now it is too late). Additionally, through initial playtesting,
we insured that regret was used to make players wonder about their decisions
but not feel bad about the overall play experience, in a manner that would make
them leave the game annoyed or frustrated. Players would play one or the other
version of the game and fill a questionnaire that, in conjunction with logged
information from the game, would help us put our assumptions to the test.
Implementation-wise, the system follows a graph architecture where each
node represents a decision moment in the storyline leading to new decisions and
nodes in the graph (see Fig. 1). We used a world state approach, i.e. by updating
variables based on choice and outputting different text based on the value of these
variables, we kept the number of nodes manageable. The game is structured in
chapters, outlining where branches start and when they join together, while also
serving as starting points for the players’ future replays. Full details regarding
both storyline and implementation can be found in [8].
Fig. 2. Adding inaction feedback to decision nodes: the dashed line demonstrates the
traditional approach and the solid line demonstrates the feedback system approach.
Highlight the Path Not Taken to Add Replay Value 65
When (A) is chosen, the consequences of the action are directly described to
the player and the story moves forward (e.g. “you grab the bottle and quickly
hide it from sight before anyone notices.”). However, if node (A) is not cho-
sen, and the system realizes (through graph traversal and appropriate tagging)
that node (A) is no longer available to be chosen by the player during this
playthrough, then, before moving to node (I), an extra node (F1) is triggered
that provides feedback regarding the inaction (e.g. “while leaving the tavern,
you see the maid bump unto the table next to you, the bottle that seconds ago
was standing on the table just breaks into tiny pieces, leaving a poisonous odor
in the air”). The intent is to show the player that something important (in terms
of story flow) was (voluntarily or involuntarily) missed by the player, and that
node (A) is now off-limits. Additionally, a node (F2) is created that reinforces
the fact that this option is no longer available to the player (e.g. a new option
available to the player in node (I) is to “examine the broken bottle” that would
lead to (F2) “the bottle is now broken and its foul content lost forever, who
would have left such a bottle unattended?”). Overall, it should be clear by then
that another option was available to be chosen.
After all, (F1) and (F2) nodes are created in the story graph, the story
designer must fill them with adequate content so the system can provide appro-
priate inaction feedback when required. This feedback needs to appeal to the
player’s senses, such as visual cues, sounds, smells, and vibrations.
5 Evaluation
We invited a set of participants to play one of two versions of “The Ballad of the
Wizard and Sacrifice” and answer a few questions. The inaction feedback was
activated in one version and disabled in the other. First, the participants would
answer a set of demographic questions and report their familiarity with inter-
active storytelling. Then, they would play one of the two versions of the game,
as much as they wanted, while the game would be logging their choices and
progression. Finally, they would fill a second questionnaire in which they would
rate their experience with the game and report on emotions related to regret. We
measured the game experience with the Game Experience Questionnaire [9] and
five complementary items of Choice Perception, Narrative Perception, Agency,
Action, and Inaction. The evaluation of the emotions of regret and disappoint-
ment was accomplished using the RDS [14]. Finally, we added questions directly
related to the game story understanding for control. The participants were not
made aware of the manipulation until the end of the experiment. Due to the
constraints related to the worldwide pandemic at the time of the experiment,
this process was conducted remotely. A link was given to each participant, that
would direct them to a website that would guide them, step by step, through the
experimental procedure. The average playing time was 20 min but varied based
on the number of playthroughs.
66 S. Gamito and C. Martinho
A pilot was conducted with 4 participants playing the version with inaction
feedback. This preliminary evaluation assessed the overall experience of going
through the narrative with the inaction feedback and gathered feedback regard-
ing both the game and the evaluation procedure.
Participants expressed a positive feeling while going through the game. The
regret index had a much higher score than the disappointment index, which
means they attributed the consequences of bad outcomes to their own actions.
The experience was overall enjoyable, all participants played the game multiple
times and explored different choices when going through the same decision point.
A potential issue was, however, identified in the procedure. All participants
reported a positive emotion after playing (regarding the overall experience).
Because a participant could have felt regret during a playthrough but being able
to cope with this feeling by addressing the issue in the following playthrough
(maybe motivated to replay because of this same feeling of regret), asking the
question at the end of the experience would have prevented us to detect this
effect. To ensure this was not the case, we decided to ask participants to fill the
RDS items right after the first playthrough and capture the (eventually induced)
emotion of regret after going through the game once.
The final evaluation was conducted with a total of 64 participants, equally dis-
tributed between the two versions: V1 implemented the “standard” approach
and only provided feedback to the actions of the player; V2 provided all that
V1 provided with the addition of inaction feedback based on our model.
Regarding the results from the RDS questionnaire, we found a significant
difference2 in affective reaction between V1 and V2 (t(62) = 2.361, p = 0.021,
x̄(V 1) = 1.840, x̄(V 2) = 0.910). Participants felt more positive emotions towards
V2 and more negative emotions (e.g. sadness, sorrow) towards V1. This could be
a result of the players in V1 having less feedback that would provide them with an
explanation for the outcome of their actions and ‘inactions’. Such feedback would
provide them with immediate hints to courses of actions that could change or
prevent that outcome. In opposition, the feedback provided in V2 helped players
better understand their choice, and could have provided them with a stronger
feeling of closure. We found moderate similar feelings of both regret (x̄(V 1) =
1.953, x̄(V 2) = 1.750) and disappointment (x̄(V 1) = 1.844, x̄(V 2) = 1.734) in
both versions.
Results gathered from the game logs (see Fig. 3) showed more participants
replayed the game in V2 but fewer endings were reached. In opposition, in V1, a
greater diversity of endings was explored, but with fewer participants achieving
2
Independent t-tests were used when comparing distributions that both passed the
Shapiro-Wilk normality test and a Mann-Whitney U test were used when one of the
distribution did not. The test were performed with IBM SPSS 26.
Highlight the Path Not Taken to Add Replay Value 67
6 Discussion
In this section, we discuss some results related to the participants’ familiarity
with interactive fiction and their inherent preferences regarding this medium.
Players who enjoy and are familiar with branching narratives in video games
and replay them multiple times to get different stories, felt more challenged when
playing V1 than V2 (U = 91, p = 0.022, x̄(V 1) = 1.750, x̄(V 2) = 0.950). This
might be a result of the richer feedback provided by V2 when compared to V1.
In V1, it is harder to understand the branches the story could take.
Players who do not enjoy interactive fiction or are not familiar with it,
reported feeling more tired in V1 than V2 (U = 44, p = 0.009, x̄(V 1) = 0.920,
x̄(V 2) = 0.290). The feedback given in V2 provided with a more enjoyable ride
for these participants, who are not fully aware of what characterises the genre.
Players who reported replaying this type of games more than once, got to
play at least thrice and felt V2 better conveyed the consequences of their actions
than V1 (t(33) = 2.371, p = 0.024, x(V 1) = 2.160, x̄(V 2) = 2.940). However,
they also reported feeling more irritated toward V2 than V1 (U = 61, p = 0.029,
x̄(V 1) = 0.330, x̄(V 2) = 0.790). This could be the consequence of V2 regularly
68 S. Gamito and C. Martinho
pointing that something could have been done differently and they would have
to go through the game once more to experience it, which moves the 100%
completion goal post further away at each game iteration. This feeling could
have been amplified by the repetitive nature of the text-based interface of the
game although players were not forced to go through the game from the start,
and could resume at the start of any previously visited chapter.
Players who do not feel the need to replay these games after a first
playthrough reported a lesser sense of agency in V2 than V1 (t(27) = 2.705,
p = 0.014, x̄(V 1) = 3.038, x̄(V 2) = 2.094). This could be related to the fact
that pointing out what could have happened differently to a player that has no
intentions of replaying the game can have a negative impact on this dimension
of the experience. Additionally, when considering these same players, we found
that more played the game just once in V1 than V2 (U = 46, p = 0.007, x̄(V 1)
= 0.670, x̄(V 2) = 0.170). As such, providing inaction feedback could motivate
such players to replay the game at least a second time. Furthermore, we found
that, although they had the option to restart from any chapter, these partici-
pants actually replayed the game from Chap. 1 more often in V2 than in V1 (U
= 184, p = 0.050, x̄(V 1) = 0.280, x̄(V 2) = 0.050).
Finally, participants who played through the game only once found V2 more
impressive than V1 (t(22) = 2.305, p = 0.031, x̄(V 1) = 1.210, x̄(V 2) = 2.140).
As such, inaction feedback may help with first impressions.
References
1. Albarracin, D., Hepler, J., Tannenbaum, M.: General action and inaction
goals. Curr. Direct. Psychol. Sci. 20, 119–123 (2011). https://doi.org/10.1177/
0963721411402666
2. Baron, J., Ritov, I.: Omission bias, individual differences, and normality. Organ.
Behav. Human Decis. Process. 94, 74–85 (2004). https://doi.org/10.1016/j.obhdp.
2004.03.003
3. Bohner, G., Dickel, N.: Attitudes and attitude change. Ann. Rev. Psychol. 62,
391–417 (2011). https://doi.org/10.1146/annurev.psych.121208.131609
4. Connolly, T., Reb, J.: Regret in health-related decisions. Health Psychol.: Off. J.
Divis. Health Psychol. Am. Psychol. Assoc. 24, S29–S34 (2005). https://doi.org/
10.1037/0278-6133.24.4.S29
5. Connolly, T., Zeelenberg, M.: Regret in decision making. Curr. Dir. Psychol. Sci.
11(6), 212–216 (2002). https://doi.org/10.1111/1467-8721.00203
6. David, B.: Regret in decision making under uncertainty. Oper. Res. 30, 961–981
(1982). https://doi.org/10.1287/opre.30.5.961
7. Diefendorff, J., Hall, R., Lord, R., Strean, M.: Action-state orientation: construct
validity of a revised measure and its relationship to work-related variables. J. Appl.
Psychol. 85, 250–63 (2000). https://doi.org/10.1037/0021-9010.85.2.250
8. Gamito, S.: Highlight the path not taken to add replay value to a storytelling video
game. Master’s thesis, Instituto Superior Técnico, University of Lisbon (2021)
9. IJsselsteijn, W., de Kort, Y., Poels, K.: The game experience questionnaire. Tech-
nische Universiteit Eindhoven (2013)
10. Inman, J., Dyer, J., Jia, J.: A generalized utility model of disappointment and
regret effects on post-choice valuation. Mark. Sci. 16, 97–111 (2000). https://doi.
org/10.1287/mksc.16.2.97
11. Kahneman, D., Tversky, A.: The psychology of preferences. Sci. Am. 246, 160–173
(1982). https://doi.org/10.1038/scientificamerican0182-160
12. Kutscher, L., Feldman, G.: The impact of past behavior normality on regret: repli-
cation and extension of three experiments of the exceptionality effect. Cogn. Emot.
33, 901–914 (2019). https://doi.org/10.1080/02699931.2018.1504747
13. Loomes, G., Sugden, R.: Regret theory: an alternative theory of rational choice
under uncertainty. Econ. J. 92, 805–824 (1982). https://doi.org/10.1287/opre.30.
5.961
14. Marcatto, F., Ferrante, D.: The regret/disappointment scale: an instrument for
assessing regret and disappointment in decision making. Judgm. Decis. Mak. 3,
87–99 (2008)
15. Pieters, R., Zeelenberg, M.: A theory of regret regulation 1.1. J. Consum. Psychol.
17, 29–35 (2007). https://doi.org/10.1207/s15327663jcp1701 6
70 S. Gamito and C. Martinho
16. Roese, N.J.: Counterfactual thinking. Psychol. Bull. 121, 133–148 (1997). https://
doi.org/10.1037/0033-2909.121.1.133
17. Spranca, M., Minsk, E., Baron, J.: Omission and commission in judgment and
choice. J. Exp. Soc. Psychol. 27, 76–105 (1991). https://doi.org/10.1016/0022-
1031(91)90011-T
18. Zeelenberg, M.: Anticipated regret, expected feedback and behavioral decision-
making. Other publications tisem, Tilburg University, School of Economics and
Management (1999). https://EconPapers.repec.org/RePEc:tiu:tiutis:38371d1b-
31fd-45b0-860f-b83a4e416fbf
19. Zeelenberg, M., van den Bos, K., Dijk, E., Pieters, R.: The inaction effect in the
psychology of regret. J. Pers. Soc. Psychol. 82, 314 (2002). https://doi.org/10.
1037/0022-3514.82.3.314
Narrative Text Generation
from Abductive Interpretations Using
Axiom-Specific Templates
1 Introduction
A popular approach in research on narrative text generation is to first represent
story content formally as symbolic structures, which are then converted into nat-
ural language text using a variety of approaches. Elson [2] proposed the Story
Intention Graph as a formalism for encoding the interpretation of stories as
symbolic causal structures, reminiscent of the Causal Network Model of psychol-
ogists Trabasso and van den Broek [18]. Using a software tool for hand-authoring
these representations [3], different research teams have succeeded in authoring
sizable corpora of story representations, and devising novel algorithms for con-
verting these representations into fluent natural-language texts [2,12]. Although
these text-generation systems are typically quite sophisticated in their use of
numerous grammatical subsystems and lexical resources, the overall lesson from
The project or effort depicted was or is sponsored by the U.S. Army Research Labo-
ratory (ARL) under contract number W911NF-14-D-0005, and that the content of the
information does not necessarily reflect the position or the policy of the Government,
and no official endorsement should be inferred.
c Springer Nature Switzerland AG 2021
A. Mitchell and M. Vosmeer (Eds.): ICIDS 2021, LNCS 13138, pp. 71–79, 2021.
https://doi.org/10.1007/978-3-030-92300-6_7
72 A. S. Gordon and T. S. Wang
tual templates directly with the axioms used to assemble the story graph in the
first place. Using axiom-specific templates, we show that a trivial template sys-
tem with simple manipulations for noun phrases is sufficient to generate text from
formal story graphs that is as fluent as those produced by previous approaches.
2 Axiom-Specific Templates
The basic idea in our narrative text generation approach is to utilize sentence-
length templates that are specific to individual knowledge base axioms, rather
than trying to assemble grammatical sentences from groups of connected nodes
in the structured interpretation graph. The rationale is that the knowledge base
axioms used to construct the interpretation graph already identify a coherent
set of interrelated nodes (logical literals) during the search process. When an
axiom participates in building the interpretation, its constituent literals (and
their variable bindings) provide all the necessary information to express the
inference as a fluent natural-language sentence.
To illustrate this idea, consider the following knowledge base axiom, used by
Gordon [4] to correctly answer question 83 of the TriangleCOPA benchmark.
This axiom captures the commonsense idea that if somebody attacks some-
one that you like, then you are likely to be angry at the attacker. During the
interpretation process, this axiom would be used to replace an assumption that
unifies with the consequent with the three assumptions in the antecedent, along
with the necessary variable substitutions. In a subsequent narration process, a
text template can be used to express this inference as a single sentence.
When the pronoun class of an entity has been provided, our implementa-
tion will favor referencing it using a pronoun rather than a proper or common
noun, guided by specific directives provided by the template author. Our system
supports pronoun substitution for subjects (he), objects (her), dependent pos-
sessives (their), independent possessives (hers), and reflexive pronouns (herself),
as in the following example:
4 Evaluation
We evaluate our approach by directly comparing it to the previous work of Ahn
et al. [1], where over-generating and ranking is used to assemble content from
connected nodes in the story graph into fluent sentences. As in their previous
work, we apply our approach to 100 formal interpretations of problems in the
TriangleCOPA benchmark.
Modelled after the Choice of Plausible Alternatives (COPA) benchmark [17]
that is widely used in computational linguistics research, TriangleCOPA was
conceived as an end-to-end evaluation for systems that jointly perform the tasks
perception, interpretation, and narration. Each of its 100 questions consist of a
sentence describing a situation involving three characters and a common setting,
a question about the commonsense interpretation of the situation, and two plau-
sible answers, where one was uniformly preferred by human raters. Unlike the
original COPA evaluation, each TriangleCOPA question includes an animated
video clip of the situation to support computer vision research on action recog-
nition, a formal representation of the question and each alternative to support
1
https://github.com/asgordon/EtcAbductionCS.
76 A. S. Gordon and T. S. Wang
a. Q 5. Insults and yelling flew back and forth as the circle and triangle
argued loudly in the house. Finally, the triangle had had enough
and walked out, slamming the door behind it. As angry as the
circle was at the triangle, it was very sad and knew that this may
be the end of their relationship.
Q 83. The circle is trying to get away from the cops and pushes the
small triangle to get out of its way. The big triangle feels attacked
that the circle pushed its friend and chases after the circle too.
b. Q 5. Big Triangle was inside. Circle was inside. Big Triangle argued
with Circle because Big Triangle was angry at Circle. Big Trian-
gle exited. He closed a door. Circle moved to the corner because
Circle was feeling sad.
Q 83. Circle approached Little Triangle in order to attack Little Tri-
angle. Circle pushed on Little Triangle to attack Little Triangle.
Big Triangle chased Circle because Big Triangle was angry at
Circle.
c. Q 5. The big triangle argues with the circle. He is inside the box be-
cause he is asleep. She is inside it because she is asleep. He exits
it and closes the door. She goes to the corner because he argues
with her.
Q 83. The circle approaches the little triangle and pushes him in order
to attack him. The big triangle chases her because the big triangle
likes the little triangle and she attacks the little triangle.
see several improvements that could be made to our approach, which may not
be easily assessed using only the metric of perplexity. Instead, we are most
encouraged by the finding that our approach generates text that is at least as
good as Ahn et al., using a much simpler method.
5 Conclusions
The use of structured graphs to represent story content has aided progress in
automated interpretation and narration by allowing researchers to focus their
efforts on either of the two different parts of the problem, namely graph con-
struction and natural language generation. However, a downside of this separa-
tion is that certain opportunities to exploit synergies across these two processes
are not immediately evident. The problem addressed in this paper is one such
example, where the assembly of sentences from connected nodes in the graph is
greatly simplified by attaching templates directly to the axioms used to make
these connections during the interpretation process. Here we exploit a particular
feature of interpretations constructed using etc.etera abduction, namely that the
etc.etera literals present in a solution indicate exactly which axioms were used in
its construction, along with the variable bindings for each universally quantified
variable. Using text templates and straightforward methods for including proper
nouns, common nouns, and pronouns, the difficult grammatical problems of sen-
tence construction can be largely avoided. While the resulting narrative text is
similar in quality to that of more sophisticated approaches, our hope is that
the simplicity of our method encourages researchers to shift their development
efforts toward more interesting aspects of the narrative text generation problem,
such as content selection and discourse planning.
References
1. Ahn, E., Morbini, F., Gordon, A.S.: Improving fluency in narrative text generation
with grammatical transformations and probabilistic parsing. In: Proceedings of the
9th International Natural Language Generation Conference, pp. 70–73. Association
for Computational Linguistics, Stroudsburg (2016)
2. Elson, D.K.: Modeling narrative discourse. Columbia University (2012)
3. Elson, D.K., McKeown, K.R.: A platform for symbolically encoding human nar-
ratives. In: AAAI Fall Symposium: Intelligent Narrative Technologies, pp. 29–36
(2007)
4. Gordon, A.S.: Commonsense interpretation of triangle behavior. In: Thirtieth
AAAI Conference on Artificial Intelligence, pp. 3719–3725. AAAI Press, Palo Alto
(2016)
5. Gordon, A.S., USC EDU: Interpretation of the Heider-Simmel film using incre-
mental etcetera abduction. Adv. Cogn. Syst. 6, 1–16 (2018)
6. Gordon, A.S., Spierling, U.: Playing story creation games with logical abduction.
In: Rouse, R., Koenitz, H., Haahr, M. (eds.) ICIDS 2018. LNCS, vol. 11318, pp.
478–482. Springer, Cham (2018). https://doi.org/10.1007/978-3-030-04028-4 55
Narrative Text Generation from Abductive Interpretations 79
7. Heider, F., Simmel, M.: An experimental study of apparent behavior. Am. J. Psy-
chol. 57(2), 243–259 (1944)
8. Hobbs, J.R.: Resolving pronoun references. Lingua 44(4), 311–338 (1978)
9. Hobbs, J.R.: Ontological promiscuity. In: Proceedings of the 23rd Annual Meeting
on Association for Computational Linguistics, pp. 60–69. Association for Compu-
tational Linguistics (1985)
10. Hobbs, J.R., Stickel, M.E., Appelt, D.E., Martin, P.: Interpretation as abduction.
Artif. Intell. 63(1–2), 69–142 (1993)
11. Inoue, N., Inui, K.: ILP-based inference for cost-based abduction on first-order
predicate logic. J. Nat. Lang. Process. 20(5), 629–656 (2013)
12. Lukin, S.M., Walker, M.A.: Narrative variations in a virtual storyteller. In:
Brinkman, W.-P., Broekens, J., Heylen, D. (eds.) IVA 2015. LNCS (LNAI), vol.
9238, pp. 320–331. Springer, Cham (2015). https://doi.org/10.1007/978-3-319-
21996-7 34
13. Maslan, N., Roemmele, M., Gordon, A.S.: One hundred challenge problems for
logical formalizations of commonsense psychology. In: Proceedings of the Twelfth
International Symposium on Logical Formalizations of Commonsense Reasoning,
pp. 107–113. AAAI Press, Palo Alto (2015)
14. Meadows, B.L., Langley, P., Emery, M.J.: Seeing beyond shadows: incremental
abductive reasoning for plan understanding. In: Plan, Activity. and Intent Recog-
nition: Papers from the AAAI 2013 Workshop, pp. 24–31. AAAI Press, Palo Alto
(2013)
15. Mir, R., Felbo, B., Obradovich, N., Rahwan, I.: Evaluating style transfer for text.
In: Proceedings of the 2019 Conference of the North American Chapter of the
Association for Computational Linguistics: Human Language Technologies, vol. 1
(Long and Short Papers), pp. 495–504. Association for Computational Linguistics,
Minneapolis (2019)
16. Radford, A., Wu, J., Child, R., Luan, D., Amodei, D., Sutskever, I.: Language
models are unsupervised multitask learners (2019)
17. Roemmele, M., Bejan, C., Gordon, A.: Choice of plausible alternatives: an evalua-
tion of commonsense causal reasoning. In: Proceedings of the AAAI Spring Sympo-
sium on Logical Formalizations of Commonsense Reasoning, Stanford University
(2011)
18. Trabasso, T., Van Den Broek, P.: Causal thinking and the representation of nar-
rative events. J. Mem. Lang. 24(5), 612–630 (1985)
19. Yeh, A.: More accurate tests for the statistical significance of result differences. In:
COLING 2000 Volume 2: The 18th International Conference on Computational
Linguistics (2000)
Using Wearable Devices to Participate in 3D
Interactive Storytelling
Abstract. There exist various ways for a user to participate in a story through
interactive narratives. Most previous work uses traditional user interfaces for inter-
actions with limited modifications to story elements. It is a challenge to allow a
user to participate in an interactive story through the first-person view in a 3D
virtual environment. In this work, we propose to use wearable motion capture
(mocap) devices to enable a user to play as a character in a 3D virtual scene and
interact with the environment and other virtual characters in real time. The interac-
tions will affect how the story develops as well as the result. In such an interactive
storytelling system, we have designed methods to interpret user actions as well
as to generate parameterizable animations according to the interactions. We have
conducted a user study to evaluate our system by comparing a traditional con-
troller with a wearable device. The experimental results reveal that the interaction
methods we have designed are more intuitive and easier to use, compared to the
controller. In addition, the users are willing to try to play with the system multiple
times, which confirms the replay value of our interactive storytelling system.
1 Introduction
As computer technologies advance, there exist more and more opportunities for the
audience to change their roles as observers and actively participate in a story in a 3D
virtual environment. Through interactions with the virtual environment as well as other
characters in a story, one can change the plot of the story or the contents of story elements
on the fly. The audience will be able to break the fourth wall of narratives in various
ways. In a typical 3D digital interactive storytelling setting, one can use traditional user
interface devices such as a keyboard and a mouse to interact with the narrative system by
selecting a story branch from a menu or navigating to a Non-Character Player (NPC) to
retrieve information and make a conversation. The interaction may not be intuitive and
the responses from the system may also be limited. On the other hand, the development
of new sensing and VR technologies is opening up new directions for novel applications.
For example, somatosensory devices such as Leap Motion, Kinect, and other wearable
motion capture devices are becoming more affordable, people are starting to use these
devices to develop novel applications for interactive storytelling.
As the technologies develop, motion capture (mocap) has become a popular and
affordable way to generate character animations in 3D applications. If we use the mocap
devices to control the motions of a 3D avatar for the user, how to interpret these motions
and generate responsive interactions becomes an important research issue. Besides, cap-
tured motions are mostly used as canned motions for NPCs when certain events are
triggered or certain conditions for the story or game are satisfied. If the behaviors of
NPCs cannot adapt to user interactions, the plausibility of the virtual scene, as well
as the replay value of the application, will be greatly reduced. Thus, in this work, we
aim to design a 3D interactive storytelling system allowing intuitive 3D avatar control
and generation of responsive and appropriate NPC behaviors. We will describe how the
system is designed and report the experimental results from a user study.
In the following sections, we will first describe the related work about our research.
In Sect. 3, we will describe how we have designed and implemented the interactive
storytelling system. Then, we will report the result of the experiment that we have
designed to evaluate the system. Lastly, we conclude the paper with some remarks and
future research directions.
2 Related Work
Interactive storytelling is a form of entertainment allowing the authors and audience to co-
create a unique experience of the story through user interactions. 3D interactive narrative
is a special form emphasizing using a 3D virtual environment to deliver a story. To realize
such an application, many techniques are involved such as drama generation, authoring
tools, character animation generation, scripting languages, user interface design, etc. The
Oz project [1] is one of the earliest interactive storytelling systems utilizing the concept
of agents to provide impromptu interactions with users. Spierling et al. [13] introduced
the concept of Narrative Formalism into interactive storytelling and use a layered model
to create interactive stories and contents.
Unlike text-based interactive drama creation, the challenges for 3D interactive sto-
rytelling are more on the dynamic generation of multimedia contents and the design of
multi-modal user interfaces. Kistler et al. [7] and Yang and Li [15] all have proposed
to use gestures to interact with the objects in the virtual environment or to select a
branch in a story graph. Cavazza et al. [3] have designed a multimodal user interface to
allow a user to use voice input to influence the behaviors of virtual characters as well as
the plot in an immersive 3D scene. In [4], a gesture recognition module also has been
implemented for interactive storytelling. Brown et al. [2] reported a study attempting to
find the features of iconic gestures for retelling a story. Piplica et al. [10] presented a
system for combining improvisational acting with full-body motions detected through
Kinect to support the co-creation of interactive narratives. Mousas et al. [8] proposed
to detect user gestures with Kinect to trigger actions of virtual characters. Rhodin et al.
[11] proposed to use Kinect and Leap Motion and predefined settings of gestures and
velocities to present the corresponding animation of virtual characters. Eubanks et al.
[5] recently presented two studies about the investigation of how body tracking fidelity
is related to avatar embodiment.
There has been much research on the automatic generation or editing of charac-
ter animations. For example, Tonneau et al. [14] proposed a technique to edit existing
82 T.-Y. Li and W.-H. Wang
motion clips to respond to large environmental changes on the ground. Shoulson et al.
[12] proposed an animation testbed allowing a user to leverage a character’s animation
and navigation capabilities when authoring both individual decision-making and com-
plex interactions. In [6], the authors used precomputed semantic information about the
environment to choose appropriate animation clips for the character to reach a given
goal. Mousas et al. [9] also have proposed a system called CHASE allowing a novice
user to design scripts with parameters to produce the desired animation. Besides, Yang
and Li [15] also proposed to use a scripting language in XML to generate the animations
for different story nodes in 3D interactive storytelling.
a wearable device mentioned above and inputted into the system in a streaming format
(BVH). Since the capture motions have many degrees of freedom, to reduce the control
and computation complexity, we have only used a few parts of the body, such as hands
and legs, for interpretation. Motion interpretation is needed because some of the player
motions are designed to trigger certain events with interaction with the environment or
NPCs. In addition, some motions, such as sitting down on a chair, cannot happen in the
physical space as in the virtual space. Therefore, we have divided the inputs into two
modes: direct input and command input.
Direct Input Mode. In this mode, the motions captured for the player are mirrored
directly into the motions of the avatar. Therefore, the player can move his body as if
he was the avatar. When certain conditions described in interaction scripts are satisfied,
corresponding actions from the environment or the NPC will be triggered to respond
to the player’s intention for interaction. For example, in Table 1, we have listed the
types of motions that will be recognized as meaningful actions and trigger events with
corresponding motions by the affected objects or the NPC. For example, when the hand
of a user avatar approaches an object, such as a glass or a hand of an NPC, and a grasp
motion is performed, the object will be snapped into and moves with the hand (Fig. 2).
Similarly, an object is released when the player ungrasps. The push motion is similar
and can be used to push objects, such as doors, before entering or exiting a room. The
motion of a short walk reflects a short-distance walk or position adjustment when the
player moves around in the physical space.
Fig. 2. The graspable object is highlighted (right) when the hand is close enough
Command Input Mode. Unlike the direct input mode where all motions are mapped to
the avatar, there could be actions that the player would like to perform in the virtual world
84 T.-Y. Li and W.-H. Wang
but is not physically feasible in the real world. For example, the physical space could
be limited and does not allow the player to freely walk for a long distance as one may
expect in the virtual world. Similarly, certain motions are not feasible simply because
there are no corresponding objects in the real world. For example, when one would like
to sit down on a chair in the virtual world, he/she cannot do so because there may not
be a chair ready for this in the real world. As a result, we have to design special motions
allowing a player to specify his intention for doing this type of motion, which we call the
Command Input Mode. In the current implementation, we have designed three motions
for command inputs as shown in Table 2. For the long-distance walk, since the player is
not transporting his body, the avatar will move along his facing direction. To perform the
sit-down motion, the player has to touch a chair first and then bend his knees to tell the
system his intention for sitting down. Once the action is triggered, the player can stand
up and allow the avatar to display the sit-down animation. Similarly, when the avatar
is in a sit-down situation, the player can bend his knees to specify the intention for the
avatar to stand up.
Animation Enforcing Mode. There are also some situations where we would like to
enforce the display of certain animations to ensure the progress of a story or to simplify
the interaction with the environment or NPC. In this mode, the player is still allowed
to move his head to watch the display of enforced animations. In our demonstrative
example, two situations may trigger this mode. The first case occurs when the story
develops to a point where the player is shot by the suspect. Then a lying-down animation
will be enforced even though the player does not perform this action. The second case
is when the player (playing as a policeman) takes the suspect to the police car after
arresting him, both need to enter the car. We choose to enforce the animations for both
characters to avoid complex interactions and motion coordination which are not crucial
for experiencing the story.
and story script. The interaction script is used to define how the environment or the
NPCs should react to user interaction while the story script defines the story graph with
branches for the interaction narratives.
Story Script. Unlike an interaction script which describes the responses of the envi-
ronment or NPCs, the story script is used to describe the story graph consisting of story
nodes connected with branches. It also includes nodes that are time-triggered, which
means that some animations may start voluntarily without interactions. Thus, the story-
teller can take an initiative to drive the story instead of waiting for the player to interact
and trigger events. For example, the script in Fig. 4 describes that the suspect will start
to move to a chair by walking for 15 s at an absolute or relative time.
86 T.-Y. Li and W.-H. Wang
In addition to determining the priorities of the animations, the diversity and quality
of the animations will also affect a player’s immersion experience. By the diversity
of animations, we mean that the responses that the player gets from NPC should vary
according to the development of the story or the emotion of the NPC. According to the
history of interactions with the player, we maintain three emotional states: bad, good,
and great for the NPC. When a responsive action needs to be taken by the NPC, the
animation manager will select an appropriate animation according to the emotional state
of the NPC. Therefore, the responses a player sees may vary at different times of play.
Using Wearable Devices to Participate in 3D Interactive Storytelling 87
The quality of the animation for an NPC can be considered from several aspects
including the factors of cost and complexity. Since most of the animations for the NPCs
are prepared and stored in the motion database, the diversity of animations will affect the
cost of preparing such an interactive storytelling system. In our animation management
system, we have attempted to minimize the need for creating a large animation database
by reusing existing motions as much as possible. For example, many motions only focus
on a certain part of the body. As such, one can decompose a motion into different parts
that can be recomposed to form new animations. For example, as shown in Fig. 6,
the hand gesture of the NPC does not depend on the lower body state. Thus, we can
decompose the animation for the upper-body gesture and use it when the NPC is sitting
on the chair.
In addition to selecting different animation clips for different NPC emotions, the
location of the user avatar, as controlled by the player, cannot be determined in advance.
Thus, the animation management module may need to make a minor adjustment on the
facing direction of the NPC toward the player to conduct a conversation. This kind of
minor adjustment also happens when the player needs to have physical interactions with
the NPC such as in the situation of arresting the suspect by grasping his hand. As shown
in Fig. 7, the configuration of the hand being grasped is adjusted with inverse kinematics
according to the player’s hand location. We have used the built-in IK controller module
in Unity3D for the implementation of this function.
Fig. 8. Snapshots of story development during the interactive storytelling (a) mission setup, (b)
clothes changing, (c) suspect in a bar, (d) inviting for a drink, (e) releasing crime information, (f)
arresting the suspect, (g) taking him to the car, and (h) putting him in prison.
The procedure for the experiment was a within-subject design and worked as follows.
Each subject would experience the story twice with the VIVE controller and the wearable
device, respectively. The order of the two experiences was random such that the learning
effect could be canceled. Before the formal session started, we first explained to the
subjects how the system worked and allowed the subjects to practice how to interact with
the objects in a tutorial session through the wearable device and the VIVE controller,
respectively. For the VIVE controller, we have adopted the most common ways of
interacting with objects. For example, we use a touchpad to trigger teleport, a hair-
trigger to grasp, and a grip to switch between sit-down and stand-up. At the end of each
formal session, the subjects were asked to fill in a questionnaire about the operations,
story, system functions, and immersion of their experiences.
After each formal session, the subjects were asked to fill in three questionnaires for the
evaluation of our system. The questions in the survey are answered on a 5-point Likert
scale with 1 to 5 meanings strongly disagree to strongly agree. Three questionnaires
are about the assessments on user interface operations, system and story, and immersion
experience, respectively. The results (means and standard deviation) for both formal runs
(controller and wearable device) are shown in Tables 3, 4 and 5. A paired two-tailed t-test
has also been conducted for each question.
From the results in Table 3, we can find that the average scores for the wearable
device are all higher than the controller. The variation for the wearable device is also
lower than the controller. However, only the questions of intuitiveness on the grasping
90 T.-Y. Li and W.-H. Wang
and sit-down and stand-up operations have achieved significant differences between the
two types of interfaces. In the interview after the experiment, we also found that the
scores for the controller quite depend on the prior experience of the subjects while all
subjects consider the wearable devices are more intuitive to use.
The results in Table 4 reveal that the user feedback about the story and system
functions are all very positive for either interaction interface but no significant difference
has been found. The subjects have found that being able to interact with the NPC and
influence the development of the story is interesting.
Table 5 shows the result of assessing the immersion of the player. The subjects are
more immersed in the environment with the wearable device but all enjoy playing in the
scene with both input devices. The variation of the scores is higher for the controller
session probably because their familiarity with the controller varies. On the other hand,
from questions 28 and 30, we can find significant differences between the two interfaces,
and the wearable mocap device provides a more intuitive interface allowing a player to
immerse into the virtual environment more easily.
Using Wearable Devices to Participate in 3D Interactive Storytelling 91
In the current system, the dialogs between the player and the NPC are all pre-
determined and pre-recorded. Therefore, although the story has multiple plotlines and
endings, the system still cannot change the story on the fly with customized contents
or dynamic dialogs. To allow this to happen, more research on dynamic drama creation
and scene generation will be necessary. Besides, although the cost for wearable mocap
devices is becoming affordable, the quality for stability may be compromised. We will
be looking forward to the development of this type of input device for it to become more
affordable and precise.
Acknowledgment. The authors would like to thank the sponsor of this research: Ministry of
Science and Technology in Taiwan under contract MOST108-2221-E-004–007-MY3.
References
1. Bates, J.: Virtual reality, art, and entertainment. Presence: J. Teleoper. Virtual Environ. 1(1),
133–138 (1992)
2. Brown, S.A., et al.: Towards a gesture-based story authoring system: design implications
from feature analysis of iconic gestures during storytelling. In: Cardona-Rivera, R., Sullivan,
A., Young, R. (eds.) ICIDS 2019. LNCS, vol. 11869, pp. 364–373. Springer, Cham (2019).
https://doi.org/10.1007/978-3-030-33894-7_38
3. Cavazza, M., Lugrin, J.L., Pizzi, D., Charles, F.: Madame bovary on the HolodeckL immer-
sive interactive storytelling. In: Proceedings of the 15th ACM International Conference on
Multimedia, pp. 651–660 (2007)
4. Cavazza, M., Charles, F., Mead, S.J., Martin, O., Marichal, X., Nandi, A.: Multimodal acting
in mixed reality interactive storytelling. IEEE Multimed. 11(3), 30–39 (2004)
5. Eubanks, J.C., Moore, A.G., Fishwick, P.A., McMahan, R.P.: The effects of body tracking
fidelity on embodiment of an inverse-kinematic avatar for male participants. In: Proceedings
of 2020 IEEE International Symposium on Mixed and Augmented Reality (2020)
6. Kapadia, M., et al.: PRECISION: precomputing environment semantics for contact-rich char-
acter animation. In Proceedings of the 20th ACM SIGGRAPH Symposium on Interactive 3D
Graphics and Games, pp. 29–37 (2016)
7. Kistler, F., Sollfrank, D., Bee, N., André, E.: Full body gestures enhancing a game book
for interactive story telling. In: Si, M., Thue, D., André, E., Lester, J.C., Tanenbaum, T.J.,
Zammitto, V. (eds.) ICIDS 2011. LNCS, vol. 7069, pp. 207–218. Springer, Heidelberg (2011).
https://doi.org/10.1007/978-3-642-25289-1_23
8. Mousas, C.: Towards developing an easy-to-use scripting environment for animating virtual
characters. arXiv preprint arXiv:1702.03246 (2017)
9. Mousas, C., Anagnostopoulos, C.-N.: Performance-driven hybrid full-body character control
for navigation and interaction in virtual environments. 3D Res. 8(2), Article No. 124 (2017)
10. Piplica, A., DeLeon, C., Magerko, B.: Full-body gesture interaction with improvisational
narrative agents. In: Nakano, Y., Neff, M., Paiva, A., Walker, M. (eds.) IVA 2012. LNCS, vol.
7502, pp. 514–516. Springer, Heidelberg (2012). https://doi.org/10.1007/978-3-642-33197-
8_63
11. Rhodin, H., et al.: Generalizing wave gestures from sparse examples for real-time character
control. In: Proceedings of ACM SIGGRAPH Asia 2015, vol. 34, no. 6, Article No. 181
(2015)
12. Shoulson, A., Marshak, N., Kapadia, M., Badler, N.I.: ADAPT: the agent development and
prototyping testbed. IEEE Trans. Vis. Comput. Graph. 20(7), 1035–1047 (2014)
Using Wearable Devices to Participate in 3D Interactive Storytelling 93
13. Spierling, U., Grasbon, D., Braun, N., Iurgel, I.: Setting the scene: playing digital director in
interactive storytelling and creation. Comput. Graph. 26(1), 31–44 (2002)
14. Tonneau, S., Al-Ashqar, R.A., Pettré, J., Komura, T., Mansard, N.: Character contact re-
positioning under large environment deformation. In: Proceedings of the 37th Annual
Conference of the European Association for Computer Graphics, pp. 127–138 (2016)
15. Yang, C.C., Li, T.-Y.: Participating in narratives with motion-sensing technologies in a 3D
interactive storytelling system. In: Proceedings of Computer Graphics Workshop, Taipei
(2015)
Inbox Games: Poetics and Authoring
Support
1 Introduction
Interactive narratives tell stories with some form of reader interaction, but those
interactions can take many forms and serve a number of diegetic and thematic
purposes. For instance, the player-as-protagonist model (arguably the default)
gives the player the point of view of a specific character and asks them to make
in-story decisions for that character. Alternatively, a player-as-coauthor model
asks the player to draft and/or edit parts of the story (e.g.: 18 Cadence [23]); a
player-as-detective model asks players to examine and iteratively discover frag-
mented pieces of a story’s aftermath to determine a past event sequence (e.g.:
Her Story [4]); a player-as-director model asks them to manipulate aspects of
characters’ environment or the actions of multiple characters from an omniscient
point of view (e.g.: Prom Week [21]). Each of these diegetic framings serves a
different type of storytelling.
In this paper, our diegetic framing of interest is what we call the inbox game: a
work that employs an overlapping set of formal techniques to deliver a narrative
by asking the player to send and receive asynchronous messages with non-player
characters. We take an expansive view of this idea by including not only games
that look like literal email inboxes, but also those that model player interac-
tion on other forms of text-based messaging, such as traditional letter-writing,
SMS text-messaging, and social media posting. We are interested in this class
of games because we think they can evoke a compelling set of narrative themes,
including the conference themes of disconnection and reconnection. The para-
doxical feelings of simultaneous connection and disconnection created by online
communication are natural to explore within an interface mimicking one players
already use to keep in touch with friends and family online.
c Springer Nature Switzerland AG 2021
A. Mitchell and M. Vosmeer (Eds.): ICIDS 2021, LNCS 13138, pp. 94–106, 2021.
https://doi.org/10.1007/978-3-030-92300-6_9
Inbox Games: Poetics and Authoring Support 95
Our interest in this form has led us to develop an authoring language and
game engine designed specifically for this mode of storytelling. We describe this
language and its implementation status in Sect. 4.
This paper’s contribution is twofold, paralleling the two disciplinary
approaches to research methodology. First, we present a theoretical descrip-
tion of the poetic affordances of inbox games, through definition, example, and
description of the relationships between formal properties and narrative effects.
Second, we present a technical design effort in the form of our software tool for
authoring inbox games.
2 Related Work
Mawhorter et al.’s work on choice poetics [20] provides a model for this paper by
integrating the theoretical foundation of poetics (relations between form and con-
tent) [7] with theories of game design and interactive media, such as procedural
rhetoric [5] and link pragmatics in hypertext games [29]. As in the choice poet-
ics paper, one of our goals is to provide a preliminary vocabulary for discussing
the relationship between formal elements (interface mimesis and epistolary sto-
rytelling) and their aesthetic and narrative effects on players. Previous work
surveying the email novel as a literary form performs similar poetic analysis of
a related genre [24,25], though email novels are not interactive.
Kreminski and Wardrip-Fruin’s work defining “storylets” as a model of inter-
active narrative authorship is structurally similar to ours in that they also cata-
logue several existing narrative works and outline a design space based on salient
common features [15]. The idea of storylets itself is also relevant to this work in
that an individual message (e.g. email) could be codified as a storylet, and an
engine for inbox games could be adapted to interoperate with storylets as a data
structure.
Grayscale is an inbox game implemented on top of the Chimeria project [10],
an engine for modeling group affinities for virtual characters in stories. Chime-
ria:Grayscale represents a central example of the form we are attempting to
define and support. Further examples are discussed in the next section.
3.1 Examples
Inbox games allow the player to send and receive messages to and from non-
player characters, but within those constraints, they vary heavily. We first
present a collection of examples that we consider within or adjacent to the space
96 C. Martens and R. J. Simmons
Fig. 3. Screenshots of the drafting interfaces in First Draft of the Revolution and 13 min
of Light.
Inbox Games: Poetics and Authoring Support 97
of inbox games and identify some of their key characteristics. Table 1 summarizes
the games and characteristics discussed.
There are two existing games we are aware of that do this with an interface
that simulates an email inbox: Digital: A Love Story [16] and Grayscale [10].
Screenshots of their interfaces can be seen in Figs. 1 and 2. In both cases, an
opening frame story transitions quickly to the interface of an email client, in
which the player progresses by reading emails and sending replies. In Digital, a
reply is sent by clicking on the “reply” button, but the contents of the player’s
message are left implicit. In Grayscale, the player usually has several choices for
how to respond, and the choice affects the outcome of the game. It is implied
that the text in the choice links for the player is the literal text of the email that
is sent.
There are a few other examples of games that emphasize the “epistolary”
nature of inbox games—sending and receipt of messages—without emphasis on
digital interface mimesis, including First Draft of the Revolution [27] and 13 min
of Light [14], which simulate writing handwritten letters (Fig. 3). (Notably,
13 min of Light also has a portion of gameplay that mimics Reddit’s forum
interface.) These games also allow the player to edit the message before sending
it, which adds an element of coauthorship to the player’s role, rather than simply
selecting pre-authored responses.
There are several other games that emphasize digital interface mimesis but
deviate from the strictly asynchronous epistolary nature of email. We Should
Talk [30], Lifeline [1], and A Normal Lost Phone [2] all imitate text messaging
interfaces for in-game dialogue. Don’t Take it Personally, Babe [17] (by the same
author as Digital: A Love Story) mimics the social network dynamics of Facebook
walls, and HoloVista [3] specifically evokes Instagram, including posting photos
with filters and receiving likes and comments.
The above examples form a Wittgensteinian family resemblance [22] in that they
share a meaningful set of overlapping formal features with an empty intersection.
These formal features include:
Game RC RD M IM
Digital: A Love Story BBS/Email
Grayscale Email
Holovista Instagram
A Normal Lost Phone Email and text messages
First Draft of the Revolution Handwritten letters
13 min of Light Handwritten letters/Reddit
Don’t Take it Personally, Babe Facebook
Lifeline Text messaging
We Should Talk Text messaging
Talk’s text messaging interface where individual words and subphrases of a single
message can be selected and modified independently. This mechanic extends the
metaphor of the inbox beyond reply selection and into reply composition, which
is why Emily Short classifies them as “games of coauthorship” [26].
Fig. 4. An inbox game can build a response choice mechanic on top of an interface for
selecting suggestions that Google has already taught to a large population of users.
interfaces. Grayscale had to invent its own choice interface for email replies,
but in the system we describe in Sect. 4, we can simply reappropriate Google’s
interface to provide the user with a choice-based interface they are likely to have
already encountered (Fig. 4).
“I’ve learned that real-world human-to-human tactile contact will kill you,
and that all human interaction, whether it be social, political, spiritual,
sexual, or interpersonal should be contained in the much more safe, much
more real interior digital space. That the outside world, the non-digital
world, is merely a theatrical space in which one stages and records content
for the much more real, much more vital digital space.
One should only engage with the outside world as one engages with a coal
mine. Suit up, gather what is needed, and return to the surface.”
Narrative themes supported by inbox games include both the good and bad of
mediated communication: a sense of being separated from physical intimacy and
connection, but also learning new languages of intimacy that can be expressed
in unfamiliar ways. Many LGBTQ+ millennials and zoomers grew up with the
internet as the main space that felt safe to express and explore their identity
during adolescence [18], and this experience is reflected in games like Digital: A
Love Story and A Normal Lost Phone.
At the same time, especially during the COVID-19 pandemic, lots of people
experienced feelings of dissociation and being “trapped” behind digital inter-
faces, struggling to express intimacy and nuance through them. danah boyd
Inbox Games: Poetics and Authoring Support 101
coined the term “context collapse” [8,19] to explain one facet of this phe-
nomenon: internet spaces like email inboxes often collide social worlds from the
workplace, family, and extracurricular spaces. Grayscale emphasizes that chal-
lenging emotional and political labor can take place in the email inbox [9], despite
the medium’s apparent lack of conversational nuances associated with embodied
communication (e.g.: sexual harassment can take place via the strictly verbal
medium of email, not only via physical touch). HoloVista traps its protagonist
in a space generated by their own internal turmoil, employing themes of cog-
nitive distance from reality and difficulty distinguishing one’s internet presence
from one’s reality.
4 Authoring Support
Inspired by this emerging genre, we designed an authoring tool called Inbox,
consisting of a custom markup language and a rendering engine, to allow an
author to create an inbox game. We take inspiration from Twine [13] and Ink [12]
in terms of language minimality and prioritizing the writing of legible narrative
text over control flow code. Our design of Inbox served as an investigation into
the minimal set of authoring affordances that could support multithreading and
response choice.
It will feel less believable if a player receives a reply to their message immediately
after sending it. There are a few different possible ways to solve this: for example,
Grayscale simply adds a timed delay between sending an email and receiving new
emails. Our design choice for Inbox is to modify the inbox only when an email is
sent, returning the player to an inbox that has already changed rather than ever
showing an email’s arrival. To model the “time” it would take for a recipient to
respond to their sent messages, we essentially add a one-step delay: the effects
of a player’s action won’t happen until they take a subsequent action. In other
words, the player can take an action on a thread, but any responses to that action
won’t appear until immediately after the player takes an action on a different
thread.
This interaction style won’t be the only one that authors want, but it is one
that promotes an experience of context collapse (as described in Sect. 3.3): taking
an action on a given thread instantaneously redirects the player’s attention back
to an unrelated thread. This model also allows us to greatly simplify the engine
due to fact that the most recently received email can be thought of as the
“current scene,” which entirely defines the set of available actions. On the other
hand, it limits the expressive affordances for authoring. There is no way, for
example, to permit a player to send multiple emails on a thread in a row: sending
an email is an action, and no other action will be available until the engine adds
a new scene to the thread, which won’t happen until the player takes action on
a different thread. Additionally, since it is more straightforward to implement
responses that always react to the last email received, Inbox does not naturally
facilitate authoring a conversation where the set of available responses shifts as
the conversation evolves.
The Inbox authoring language gives authors three main expressive affordances,
threads, scenes and (player) actions. A thread is a list of scenes associated with
an email subject. A scene corresponds to a single email received by the player.
In addition to the usual data and metadata associated with an email (text,
attachments, a list of senders and recipients), a scene defines a set of actions,
which include response choices and their effects.
In Fig. 6, a complete Inbox script is provided, written in a markup language
called Camperdown designed for flexible interactive authoring.1 (Both Fig. 4 and
Fig. 5 are taken from this script.) The threads are marked by lines starting with
single # marks (lines 11 and 20), and additional scenes in the second thread
are marked by lines starting with a pair of # marks (lines 49 and 56, which
are additional scenes in the thread that starts on line 20).
The most common player action is responding to the thread by sending an
email. The author lists the player’s available responses in the script by specifying
1
https://package.elm-lang.org/packages/brilliantorg/backpacker-below/latest/.
Inbox Games: Poetics and Authoring Support 103
Fig. 6. The same inbox story written in Inbox (left) and in Ink (right).
contents of each response, allowing for response choice (but not response draft-
ing). The structure of the !respond command detailing a potential response
is shown in Fig. 7.
Because the archive action is disabled by default, the author must explicitly
indicate where the player may archive the thread. In Fig. 6, the !archive
command appears on lines 18, 54, and 61.
Fig. 7. Authoring an email response in the markup language used for Inbox.
for interface mimesis—lines 1–9 of the Inbox script, for example, serve only to
connect characters with the identifying information that an email client would
normally expose.
An inbox game in Ink seems to require at least one bit of state per thread
to track the current “scene position” of that thread—though because the first
thread has no scenes, this particular script only requires the variable on line 4
to track the state of the second thread.
The most notable difference between the two presentations, though, is not
state but circularity. An inbox game fundamentally proceeds forward as new
messages received and sent. The possible evolutions of an inbox game form a
directed graph, and the Inbox script is a straightforward linearization of this
directed graph. An Ink representation of an inbox game, on the other hand, has
little choice but to loop back on itself in order to implement response choice
and multithreading. Every [Reconsider] action must point back to its parent
email, and every sent email must return to the inbox. For an inbox game, this is
inessential complexity, as the simpler structure of the Inbox script demonstrates.
5 Future Work
The current implementation of Inbox was intended to include the minimum set
of functionality needed to tell stories with a threaded inbox. We avoided adding
a relatively important feature to our first prototype: at present, scene selection
is the only way of influencing the story’s state. Text, actions, and scenes can-
not depend on choices the player made more than one choice in the past, and
message threads cannot influence one another. An earlier prototype included
such state-manipulation primitives, and we removed them in the interest of cre-
ating a “minimum viable product” in which a context collapse experience was
emergently generated by multithreaded conversation. However, we came to the
conclusion that this restriction is probably too limiting in the long run.
A natural extension of our language and implementation is to allow choices
to be enabled or disabled, and allow message delivery to be blocked, based on
whether users have received or viewed certain scenes. This simple mechanism
adds an enormous amount of expressive power, as demonstrated by Ink’s similar
functionality of automatically supporting conditional tests on whether the player
has seen any named piece of content. For example, this feature would grant
Inbox Games: Poetics and Authoring Support 105
enough expressive power to gate scenes on the user having reached a certain
point in multiple conversations across multiple threads. It would not be difficult
to support arbitrarily complex state tracking, giving threads and scenes the
same expressive affordances as storylets, but our current approach is to grow
the language conservatively, driven by authorial need.
In addition, the consequences of any particular action are always delayed
by exactly one player action, as described in Sect. 4.2. We plan to implement a
more general set of message delay controls, including delaying for any number
of player actions, delaying until some condition is set in the global state, and
delaying for specified wall-clock time (as in Grayscale’s approach).
6 Conclusion
In this paper, we defined a class of games called inbox games, citing several
examples and identifying poetic relations between form and function in these
games. We describe an implementation of a novel authoring language for inbox
games and demonstrate how it supports key formal elements of the genre.
Our analytical process identifies that the epistolary diagetic framing and use
of familiar digital interfaces lend themselves particularly well to ICIDS 2021’s
conference theme of [Re|Dis]Connection. Specifically, these games explore nar-
rative themes of digitally mediated intimacy, isolation, surveillance, and context
collapse. We posit that these narrative themes are especially relevant today in a
world of global (though inequitably distributed) internet connectivity and global
pandemic, and our work contributes an effort to support technical forms that
assist storytellers with adopting these themes more effectively.
References
1. 3 Minute Games: Lifeline (2015). https://www.3minute.games/
2. Accidental Queens: A normal lost phone (2017). https://anormallostphone.com/
3. Aconite Studios: HoloVista (2020). http://www.aconite.co/holovista/
4. Barlow, S.: Her story (2015). http://www.herstorygame.com/
5. Bogost, I.: Persuasive Games: The Expressive Power of Videogames. MIT Press,
Cambridge (2010)
6. Burnham, B.: Inside (2021)
7. Butcher, S.H., et al.: The Poetics of Aristotle. Macmillan, London (1907)
8. Davis, J.L., Jurgenson, N.: Context collapse: theorizing context collusions and col-
lisions. Inf. Commun. Soc. 17(4), 476–485 (2014)
106 C. Martens and R. J. Simmons
9. Hardesty, L.: 3Q: D. Fox Harrell on his video game for the #MeToo
era. MIT News (2018). https://news.mit.edu/2018/3q-d-fox-harrell-video-game-
metoo-sexual-misconduct-0119
10. Harrell, D.F., Ortiz, P., Downs, P., Wagoner, M., Carré, E., Wang, A.: Chime-
ria: Grayscale: an interactive narrative for provoking critical reflection on gender
discrimination. MATLIT: Materialidades da Literatura 6(2), 217–221 (2018)
11. Iacovides, I., Cox, A., Kennedy, R., Cairns, P., Jennett, C.: Removing the HUD:
the impact of non-diegetic game elements and expertise on player involvement. In:
Proceedings of the 2015 Annual Symposium on Computer-Human Interaction in
Play, pp. 13–22 (2015)
12. The ink narrative scripting language. Inkle Studios (2021). https://www.
inklestudios.com/ink/, version 1.0.0
13. Jane, F.: Untangling Twine: a platform study. In: DiGRA 3913 - Proceedings of the
2013 DiGRA International Conference: DeFragging Game Studies, August 2014
14. Jod: 13 minutes of light (2016). https://jod.itch.io/13-minutes-of-light
15. Kreminski, M., Wardrip-Fruin, N.: Sketching a map of the storylets design space.
In: Rouse, R., Koenitz, H., Haahr, M. (eds.) ICIDS 2018. LNCS, vol. 11318, pp.
160–164. Springer, Cham (2018). https://doi.org/10.1007/978-3-030-04028-4 14
16. Love, C.: Digital: A love story. Love Conquers All Games (2010). https://
scoutshonour.com/digital/
17. Love, C.: don’t take it personally, babe, it just ain’t your story. Love Conquers
All Games (2011). https://scoutshonour.com/donttakeitpersonallybabeitjustainty
ourstory/
18. Lucero, L.: Safe spaces in online places: social media and LGBTQ youth. Multicult.
Educ. Rev. 9(2), 117–128 (2017)
19. Marwick, A.E., Boyd, D.: I tweet honestly, i tweet passionately: Twitter users, con-
text collapse, and the imagined audience. New Media Soc. 13(1), 114–133 (2011)
20. Mawhorter, P., Mateas, M., Wardrip-Fruin, N., Jhala, A.: Towards a theory of
choice poetics. In: Proceedings of the 9th International Conference on Foundations
of Digital Games (2014)
21. McCoy, J., Treanor, M., Samuel, B., Mateas, M., Wardrip-Fruin, N.: Prom week:
social physics as gameplay. In: Proceedings of the 6th International Conference on
Foundations of Digital Games, pp. 319–321 (2011)
22. Medin, D.L., Wattenmaker, W.D., Hampson, S.E.: Family resemblance, conceptual
cohesiveness, and category construction. Cogn. Psychol. 19(2), 242–279 (1987)
23. Reed, A.: 18 Cadence (2014). http://aaronareed.net/18-cadence/
24. Rettberg, J.W.: Email novel. The Johns Hopkins Guide to Digital Media, pp. 178–
79 (2014)
25. Rotunno, L.: User IDs: email novels and the search for identity. A/B: Auto/Biogr.
Stud. 21(1), 70–82 (2006)
26. Short, E.: Games of coauthorship. Emily Short’s blog (online) (2015). https://
emshort.blog/2015/01/14/games-of-co-authorship/
27. Short, E., Daly, L.: First draft of the revolution (2012). https://www.inklestudios.
com/firstdraft/
28. Squinkifer, D.: Tentacles growing everywhere (2015). https://squinky.itch.io/
tentacles
29. Tosca, S.P.: A pragmatics of links. In: Proceedings of the eleventh ACM on Hyper-
text and hypermedia, pp. 77–84 (2000)
30. Whitethorn Games: We should talk. Whitethorn (2020). https://whitethorndigital.
com/we-should-talk
A Plan-Based Formal Model of Character
Regret
1 Introduction
The feeling of regret is central to many different types of stories. For example,
remorse and regret of past choices or mistakes provides the background and
motivation to change for many characters in redemptive arcs, like Darth Vader
in the Star Wars [15] saga. Regret can also prompt non-redemptive character
growth, like Kevin McCallister in Home Alone [7] who comes to appreciate his
family after regretting a wish that they would disappear. Sometimes characters
feel a sense of loss and regret about an outcome even when they are happy with
and confident about their past choices, like Mia and Sebastian in La La Land [5].
In addition to empathizing with characters in linear stories, regret is amplified in
interactive narrative contexts where participants can regret their own actions and
influence on story events and outcomes [17]. A formal model of regret would allow
linear and interactive narrative systems to identify, reason about, and generate
situations that produce regret for narrative characters and human participants.
In this paper, we present a formal model of commissive character regret in the
context of explicit, discrete choices in plan-based narratives.
Our model is based on narrative AI planning and identifies regret using char-
acter goals and choice actions. The plan-based choices our model uses have
explicit framings and outcomes. They also have a discrete number of choice
options. Psychological work has identified regret can arise through either action
or inaction that results in an unfavorable outcome [10]. Regret caused by an
action is called commissive and regret caused by inaction is called omissive. Our
model focuses on commissive regret, which is caused by the character’s past
actions. Regret arises from a counterfactual reasoning process that compares
the current situation to an alternate possible world where the reasoner took a
different past action and received a more favorable outcome. For example, a
commonly reported regret is not finishing college or graduate school [10]. To
identify this regret, the person reflects on the time in their life they stopped
pursuing education and imagines their life if they had instead continued. They
imagine the fulfillment or opportunities they would have and feel regret about
their decision to stop. Similarly, our model works by identifying when character
actions make a goal they have in the story world unachievable along with an
alternate possible world where the goal is achieved or still possible.
We begin by discussing related work in narrative planning, choice poetics,
and the study of regret. We then introduce our formalism and model mechanics.
We give an example of the mechanics using a dilemma based on the film La La
Land. Finally, we discuss situations in which characters may feel regret that are
not currently supported by our formalization.
2 Related Work
This section provides background on narrative planning, choice poetics, and non-
computational approaches to studying regret. We begin with narrative planning,
which is the computational paradigm our model is built in.
2.3 Regret
In psychology literature, regret is viewed as a “counterfactual emotion”. This
means regret arises from a comparison between a real-world outcome and a
counterfactual possible world where a more favorable outcome was reached [32].
Additionally, regret can be felt over the short or long-term and stems from acts
of omission or commission [11]. Omissive regret stems from actions never taken,
like never learning to speak a second language. Commissive regret from a taken
110 M. Martinelli and J. Robertson
action, like ordering a bad dish at a restaurant. In this paper, we establish a plan-
based model of commissive narrative character regret. Our model establishes
regret over a narrative interval by identifying a choice action, outcome, character
goal, and counterfactual outcome that produce commissive regret. We give a La
La Land example to show this process in the context of a particular type of
choice called a dilemma. A dilemma is a type of choice structure where each
choice option leads to an undesirable outcome for the character [16]. In the next
section, we introduce our plan-based model of regret.
Finally, characters can act in the story world and change the state by taking
actions described by operators. Each operator o = lo , po , eo , co ∈ O consists
of a unique name or label l, a conjunctive set of first-order literal preconditions
p, a conjunctive set of first-order literal effects e, and a set of characters that
must consent to the action being taken. The set of free variables that occur in
an operator’s preconditions and effects are called the operator’s parameters.
An inciting action is the decision made that directly makes a desired outcome
inaccessible in the future. For this action to be part of a choice, there must also
be an alternate action where the goal remains accessible.
This alternate action produces an alternate possible world where the charac-
ter’s goal can still be achieved or is true.
This type of choice will create commission-based regret for character c with
respect to their goal g in every state after si , because their goal will never become
true as a result of the action ai . Additionally, there is at least one alternative
action c could have taken in state si to keep their goal g accessible in the future.
In the next section we illustrate these mechanics with an example dilemma based
on the film La La Land.
4 Example
To illustrate our model we use an example dilemma from the film La La Land [5].
The film is in Los Angeles and is about an unlikely relationship between an
aspiring actress, Mia, and jazz pianist, Sebastian. At the start of the film, Mia
is struggling to begin her career while Sebastian dreams of owning his own jazz
club. Near the end of the story, Mia is offered a breakout acting role in Paris.
Sebastian encourages her to follow her dream while he works to build his jazz
club. The two pursue and ultimately attain their dream careers, but lose touch
with one another in the process. For our purposes, we will treat this situation
as an explicit dilemma for Mia, with a choice between pursuing a career on her
own or staying with Sebastian. A dilemma is a type of choice structure where
each option leads to an undesirable outcome for the character [16]. No matter
what Mia chooses, she has one positive and one negative outcome. Pursuing her
career has the explicit outcome of not ending up with Sebastian and staying
together has the outcome of not attaining her dream career. Either Mia finishes
the story with her relationship with Sebastian or her career, but not both.
4.1 Setup
4.2 Result
No matter what Mia decides, our model predicts she will experience regret with
respect to one of her character goals. If Mia chooses to stay in the relationship she
will not be a successful actress, so staying in the relationship is an inciting action.
Additionally, there is an alternate action where she pursues her career which
produces an alternate possible world where she becomes a successful actress.
Our model identifies that Mia will regret not becoming a successful actress with
respect to her decision to stay in the relationship. Conversely, if Mia chooses to
pursue her career she will not be with Sebastian, so pursuing her career is an
inciting action. Additionally, there is an alternate action where she stays in the
relationship which produces the outcome the she and Sebastian are still together.
Our model identifies that Mia will regret not being with Sebastian with respect
to her decision to pursue her career. This outcome is illustrated in Fig. 2.
Fig. 2. The outcome of Mia’s choice to pursue her career. In this case, Pursue Career
becomes the inciting action with regard to Mia and Sebastian being together and Stay
Together is an alternate action where the goal remains accessible. Fulfilled goals are
indicated with a check-mark while unsatisfied goals are indicated with an empty box.
5 Conclusion
References
1. Bae, B.-C., Young, R.M.: A use of flashback and foreshadowing for surprise arousal
in narrative using a plan-based approach. In: Spierling, U., Szilas, N. (eds.) ICIDS
2008. LNCS, vol. 5334, pp. 156–167. Springer, Heidelberg (2008). https://doi.org/
10.1007/978-3-540-89454-4 22
2. Bahamón, J.C., Barot, C., Young, R.M.: A goal-based model of personality for
planning-based narrative generation. In: Twenty-Ninth AAAI Conference on Arti-
ficial Intelligence (2015)
3. Barot, C., et al.: Bardic: generating multimedia narrative reports for game logs. In:
Proceedings of the 10th Workshop on Intelligent Narrative Technologies at the 13th
AAAI Conference on Artificial Intelligence and Interactive Digital Entertainment
(2017)
4. Cardona-Rivera, R.E., Robertson, J., Ware, S.G., Harrison, B., Roberts, D.L.,
Young, R.M.: Foreseeing meaningful choices. In: AAAI Conference on Artificial
Intelligence and Interactive Digital Entertainment, vol. 10, pp. 9–15 (2014)
5. Chazelle, D.: La La Land (2016)
6. Christensen, M., Nelson, J., Cardona-Rivera, R.: Using domain compilation to add
belief to narrative planners. In: AAAI Conference on Artificial Intelligence and
Interactive Digital Entertainment, vol. 16, pp. 38–44 (2020)
7. Columbus, C.: Home alone (1990)
8. Fendt, M.W., Harrison, B., Ware, S.G., Cardona-Rivera, R.E., Roberts, D.L.:
Achieving the illusion of agency. In: Oyarzun, D., Peinado, F., Young, R.M.,
Elizalde, A., Méndez, G. (eds.) ICIDS 2012. LNCS, vol. 7648, pp. 114–125.
Springer, Heidelberg (2012). https://doi.org/10.1007/978-3-642-34851-8 11
9. Fendt, M.W., Young, R.M.: Leveraging intention revision in narrative planning to
create suspenseful stories. IEEE Trans. Comput. Intell. AI Games 9(4), 381–392
(2016)
10. Gilovich, T., Medvec, V.H.: The experience of regret: what, when, and why. Psy-
chol. Rev. 102(2), 379 (1995)
11. Gilovich, T., Medvec, V.H., Kahneman, D.: Varieties of regret: a debate and partial
resolution. Psychol. Rev. 105(3), 602 (1998)
12. Haslum, P.: Narrative panning: compilations to classical planning. J. Artif. Intell.
Res. 44, 383–395 (2012)
13. Jhala, A., Young, R.M.: Cinematic cisual discourse: representation, generation, and
evaluation. IEEE Trans. Comput. Intell. AI Games 2(2), 69–81 (2010)
14. Kokkinakis, A.V., et al.: DAX: data-driven audience experiences in esports. In:
ACM International Conference on Interactive Media Experiences, IMX 2020, pp.
94–105. Association for Computing Machinery, New York (2020)
15. Lucas, G.: Star Wars: Episode IV - A New Hope (1977)
16. Mawhorter, P., Mateas, M., Wardrip-Fruin, N.: Generating relaxed, obvious, and
dilemma choices with Dunyazad. In: AAAI Conference on Artificial Intelligence
and Interactive Digital Entertainment, pp. 58–64 (2015)
17. Mawhorter, P., Mateas, M., Wardrip-Fruin, N., Jhala, A.: Towards a theory of
choice poetics. In: International Conference on the Foundations of Digital Games.
ACM (2014)
18. Mawhorter, P., Zegura, C., Gray, A., Jhala, A., Mateas, M., Wardrip-Fruin, N.:
Choice poetics by example. In: Arts, vol. 7, p. 47. Multidisciplinary Digital Pub-
lishing Institute (2018)
116 M. Martinelli and J. Robertson
19. McDermott, D., et al.: PDDL - the planning domain definition language. Technical
report CVC TR98003/DCSTR1165, Yale Center for Computational Vision and
Control (1998)
20. Miller, C., Dighe, M., Martens, C., Jhala, A.: Crafting interactive narrative games
with adversarial planning agents from simulations. In: Bosser, A.-G., Millard, D.E.,
Hargood, C. (eds.) ICIDS 2020. LNCS, vol. 12497, pp. 44–57. Springer, Cham
(2020). https://doi.org/10.1007/978-3-030-62516-0 4
21. Ramirez, A., Bulitko, V.: Automated planning and player modeling for interactive
storytelling. IEEE Trans. Comput. Intell. AI Games 7(4), 375–386 (2014)
22. Riedl, M.O., Young, R.M.: Narrative planning: balancing plot and character. J.
Artif. Intell. Res. 39, 217–268 (2010)
23. Riedl, M.O., Bulitko, V.: Interactive narrative: an intelligent systems approach. AI
Mag. 34(1), 67 (2013)
24. Robertson, J., Cardona-Rivera, R.E., Young, R.M.: Invisible dynamic mechanic
adjustment in virtual reality games. In: IEEE International Conference on Artificial
Intelligence and Virtual Reality, pp. 282–289. IEEE (2020)
25. Robertson, J., Jhala, A., Young, R.M.: Efficient choice enumeration for narrative
world design. In: International Conference on the Foundations of Digital Games,
pp. 1–10 (2019)
26. Robertson, J., et al.: Wait, but why?: assessing behavior explanation strategies
for real-time strategy games. In: 26th International Conference on Intelligent User
Interfaces, IUI 2021, pp. 32–42. Association for Computing Machinery, New York
(2021)
27. Robertson, J., Young, R.M.: Finding Schrödinger’s gun. In: Proceedings of the
AAAI Conference on Artificial Intelligence and Interactive Digital Entertainment,
pp. 153–159 (2014)
28. Robertson, J., Young, R.M.: Gameplay as on-line mediation search. In: Proceedings
of the AAAI Artificial Intelligence and Interactive Digital Entertainment Confer-
ence, pp. 42–48 (2014)
29. Robertson, J., Young, R.M.: Automated gameplay generation from declarative
world representations. In: Proceedings of the AAAI Artificial Intelligence and Inter-
active Digital Entertainment Conference, pp. 72–78 (2015)
30. Robertson, J., Young, R.M.: Interactive narrative intervention alibis through
domain revision. In: Intelligent Narrative Technologies Workshop at the AAAI
Artificial Intelligence and Interactive Digital Entertainment Conference, pp. 49–52
(2015)
31. Robertson, J., Young, R.M.: Perceptual experience management. IEEE Trans.
Games 11(1), 15–24 (2019)
32. Roese, N.J., Morrison, M.: The psychology of counterfactual thinking. Hist. Soc.
Res./Historische Sozialforschung 16–26 (2009)
33. Teutenberg, J., Porteous, J.: Efficient intent-based narrative generation using mul-
tiple planning agents. In: International Conference on Autonomous Agents and
Multi-Agent Systems, pp. 603–610 (2013)
34. Ware, S., Young, R.M.: CPOCL: a narrative planner supporting conflict. In: Pro-
ceedings of the AAAI Conference on Artificial Intelligence and Interactive Digital
Entertainment (2011)
35. Ware, S.G., Garcia, E.T., Shirvani, A., Farrell, R.: Multi-agent narrative experience
management as story graph pruning. In: AAAI Conference on Artificial Intelligence
and Interactive Digital Entertainment, vol. 15, pp. 87–93 (2019)
A Plan-Based Formal Model of Character Regret 117
36. Ware, S.G., Young, R.M.: Glaive: a state-space narrative planner supporting inten-
tionality and conflict. In: Proceedings of the 10th AAAI international conference
on Artificial Intelligence and Interactive Digital Entertainment, pp. 80–86 (2014).
(Awarded Best Student Paper)
37. Ware, S.G., Young, R.M.: Intentionality and conflict in the best laid plans interac-
tive narrative virtual environment. IEEE Trans. Comput. Intell. AI Games 8(4),
402–411 (2015)
38. Young, R.M.: Notes on the use of plan structures in the creation of interactive
plot. In: AAAI Fall Symposium on Narrative Intelligence, pp. 164–167 (1999)
39. Young, R.M., Ware, S.G., Cassell, B.A., Robertson, J.: Plans and planning in
narrative generation: a review of plan-based approaches to the generation of story,
discourse and interactivity in narratives. Sprache Datenverarbeitung Spec. Issue
Formal Comput. Models Narrative 37(1–2), 41–64 (2013)
Narraport: Narrative-Based Interactions
and Report Generation with Large Datasets
1 Introduction
Large amounts of data are currently being produced and stored relating to all aspects
of our daily lives. By some estimates, we generate around 2.5 quintillion bytes of data
daily spread over social media platforms. This wealth of data is often of great interest
to analysts and researchers working on digital forensics in cases where there is data
associated with legal cases. The size and scope of data analysis makes this a challenging
tasks for analysts working within our legal system. This challenge is intensified further
due to the variety of tools that need to be incorporated into the analysis process and their
outputs need to be presented in interpretable and verifiable reports for non-experts. For
this task analysts need to effectively filter and triage data in order to ensure that retrieve
relevant data and are able to succinctly present their analysis results.
In legal cases pertaining to corporate malfeasance, public attorneys and investiga-
tors need to work with analysts to triage through large numbers of communications
such as email, chat messages, and social media messages to look for evidence. For this
task, analysts and researchers develop workflows that incorporate available computa-
tional tools to effectively sort through this data. Further, they need to then document
and create summary reports for further discussion, integration of segments assigned to
multiple analysts, and generation of reports for final outcomes of the investigation [17].
c Springer Nature Switzerland AG 2021
A. Mitchell and M. Vosmeer (Eds.): ICIDS 2021, LNCS 13138, pp. 118–127, 2021.
https://doi.org/10.1007/978-3-030-92300-6_11
Interactive Generation of Narrative Reports 119
As part of evaluating and building this system we worked extensively with a team
comprised of two analysts and two scientists specializing in research supporting ana-
lysts. We worked with them for over a year in an iterative development process. Anal-
ysis of large datasets where there is not a clear target is an iterative process [22]. This
is usually characterized by transitions between high-level exploration followed by a
deeper look into elements of interest. This process iteratively leads to more specific
filters along different dimensions. For instance, an analyst may want to look at the
frequency of emails about a particular topic over time as a temporal chart. Then they
identify peaks or valleys in the chart and focus on patterns of email within the month in
the neighborhood of these features. Even after this step, they end up with several hun-
dred emails to analyze. From here they could filter by emails that only include senders
or receivers at the upper-management level or based on other meta-data such as emails
that include attachments. Eventually, the process leads to one or a chain of emails of
interest that leads them through the process of discovery of relevant content to their
investigation. Finally, they generate a multi-modal (text+charts) report that summarizes
their exploration and presents items or patterns of interest with justifications for them.
Analysis Report
Stage Stage
Topic Detection
Report Generation
Email Parsing & Annotation Filtering Narrative Templates
and Visualization
Source Social Network
Analysis
Fig. 1. A pipeline for the storytelling with data paradigm from emails sources.
This paper describes two contributions with respect to AI support for analysts. First,
we describe the integration of a pipeline (see Fig. 1) of established AI tools and algo-
rithms in the workflow for analysts. Second, we present a novel idea for automatically
filtering data and generating reports that are constructed with narrative frames that are
instantiated based on constraints satisfied by trends and relationships in the underlying
data. These narrative frames encode a number of recognizable data patterns that can be
presented as stories from the dataset.
We tested our system on both the Enron [12] dataset (about 500K emails) and the
Avocado Research Email Collection (closer to 1M emails) [16]. The Enron scandal
[18] was a well-known corporate collapse with corresponding investigation. Avocado
was an information technology software and services firm developing products for the
mobile internet market, operating from the late 1990s to the middle of the first decade
of the 21st century. We used this collection extensively in our work with analysts but
the license prohibits most sharing of information about the dataset.
Our system uses an assortment of existing NLP and social network analysis tech-
niques to analyze our email datasets. The key dynamic we are seeking to capture is the
relationship between users over time. This will allow us to identify patterns of change
that correspond to narrative patterns that we can exploit later. We parse the dataset and
store as a directed hyper-graph where nodes are email addresses and edges are mes-
sages. There are other representations that afford better performance depending on the
120 C. M. Potts and A. Jhala
desired metric [7], but in terms of application to narrative patterns we leave this to fur-
ther research. There are many topics of discussion in a large organization. An important
filter for emails in this dataset is based on relevance of messages and message threads
by topics. For topic modeling we use Latent Dirichlet Analysis (LDA) [4]. We further
extend the LDA model used in this paper, following the work of TIARA [13], to com-
pute topic strength over time. This allows analysts to filter emails based on topic across
different time frames through an interactive interface. For social network analysis, we
use graph based measures over the graph of all email address units in the dataset. These
include in/out degree and proximity prestige, where these different measures offer dif-
ferent views of the social network [21].
2 Related Work
This work builds on a basis of natural language processing and social network analysis
algorithms. These are rich areas of research. We present a system for using currently
popular techniques and algorithms to generate automated reports. The field of auto-
mated email analysis includes interactive systems like TIARA [13]. TIARA develops
a few novel measures for topic strength and ranking, but also allows a user to inter-
act with several interactive visualizations. It does not however, contain any automated
reporting capabilities. Bardic [2] is an interactive tool for analyzing video game relays.
Similar to TIARA, it provides a number of interactive visualizations, but most relevant
to this work it attempts to find and present narrative patterns from the replay data. The
approach is similar in that it uses a decomposition planner to match and apply narrative
templates. Unlike Bardic, we externally solve for story patterns then use the planner to
exclusively solve the corresponding presentation plan. Sultan et al. [11] use a hierar-
chical deep-learning approach to automatically generate narrative captions for images.
Battad and Si [3] also do co-generate of visual plots and narrative captions. Our work is
more expansive than either because we do co-generation of complete narrative reports
with multiple visual artefacts (as see in Fig. 3).
Vesanto and Hollmén [19] consider the processes of data mining in general, and
present an automated system for the generation of preliminary reports. This phase is
meant to be followed by interactive analysis. It takes a similar approach in that it begins
by generating a number of standard measures over a target data source and presents that
information alongside visualizations. The intent of this report is to enable a data miner
to further explore and thus is about getting a general understanding of the data source.
Erete et al. [8] examine the use of storytelling by non-profit organizations. NPOs
use narratives to help engage stakeholders promote causes. McKenna et al. [14] did
a user study on popular data-driven narratives on the web to determine the effect of
different visual encodings. Traditional timelines used for data may not fit ideally with
storytelling approaches. Brehmer et al. [6] consider this question and explore the addi-
tional dimensions that can be explore in narrative. We find this relevant to our work
because we present a timeline in our narrative templates based upon detected inflec-
tions points in data that we use an story inflection points. An interesting contemporary
system is DataShot [20]. Given a tabular data source, the system automatically gener-
ates a fact sheet using a number of visual techniques. While we use email data, both
Interactive Generation of Narrative Reports 121
systems attempt to create engaging visual artefacts. DataShot does not attempt to create
reports however, or use storytelling techniques.
Now we present a selection of hybrid narrative/data patterns for use in our report gen-
eration. These are not meant to be exhaustive but ones that are easily identifiable and
relevant to the patterns in our dataset. The representation of stories is in terms of tem-
plates with rules that are based on statistical properties of the underlying data. These
templates can be nested. Our representation is inspired from early storytelling systems
122 C. M. Potts and A. Jhala
such as James Meehan’s Tale-Spin [15] that incorporated a set of heuristic rules over a
knowledge base of predicates about the world to create sequences of actions for char-
acters based on logical inference over STRIPS-style operators [9].
For instance, the rags to riches pattern discussed later can be used to frame a entire
report as one person gaining influence, or to introduce a communicative partner. The
system reasons over these templates to ensure that all parties discussed are properly
introduced. The sample patterns are named for their narrative counterparts. Five of the
patters are described below with the corresponding rules presented in Table 1.
Table 1. Details of narrative templates. Here “Person” refers to any email addresses with email
volume above a given threshold. A “relative set” is one or more other email addresses above
the same threshold. A measure f is either message volume (total or by topic) or an influence
proxy. “Introduce”, “Visualize”, “Highlight”, and “Contrast” correspond to presentation plans.
The constraints correspond directly to visuals that will be inserted into the final reports and the
presentation plans are partial plans to be solved by the discourse planner.
Rags to Riches. We identify a user who along any of several metrics (topic influ-
ence, email volume, etc.) shows a dramatic increase. This corresponds to the differ-
ence between the user’s least and most active periods. We also need to identify at least
one inflection point where their influence reaches a relative maximum, followed by a
relative minimum, followed by another increase. Next we foreshadow their eventual
success in text, followed by a visual representation showing increase from their initial
low point. Finally we show their final state making sure to contrast both with the orig-
inal comparison but also with their initial state. Visually the report keeps the overall
graph, showing the entire trajectory, in view. Through use of text templates, the reader
is guided to focus on interesting parts of the visual.
Fall from Grace. This is the reverse of rags to riches and involves the same data con-
ditions in reverse. The presentation largely follows the same pattern both in visual rep-
resentation in terms of choice of graphs and in textual template descriptions.
Pairs: Best Buddies. Here we look for two people in our graph that develop a closer
relationship over time. This could also be between an individual and a group. Charac-
terized on mutual email volume both in terms of sent and received emails. We look for
patterns where the interaction begins with low volume then builds. Several interesting
visualizations are generated to provide details of the relationship such as distribution of
topic terms over time, length, frequency, sentiment of emails.
Pairs: Falling Out. Opposite of best buddies where the relationship appears to dimin-
ish over time. This one is a little more challenging to characterize because someone
leaving a company often shows this pattern where their emails abruptly stop. As such
the presentation accounts for the uncertain nature of the decrease.
Juxtaposition of Rise and Fall. This is another common narrative trope where there is
a juxtaposition of rise and fall of influence or relationships. This is a meta pattern in that
we either recognize a best buddies and a falling out or a rags to riches and a fall from
grace. This meta pattern then realizes the overall report by correlating the trajectory of
both sub-patterns. A single combined visual for both patterns is used as the primary
focal point. This includes both data trajectories visual coded to contrast, enabling the
reader to draw their inference about the significance of the correlation. This pattern is
also visible during internal personnel changes in significant roles.
Discourse Plan
Visualize G.N.'s E.V. Visualize S.S.'s E.V. Clause Intro G.N. w/ Name & Company Clause Introduce Email Vol. Clause Intro w/ Name & Company
Figure Figure
Highlight G.N.'s low starting E.V. Plot G.N.'s E.V. over Time Plot S.S.'s E.V. over Time
Foreshadow G.N's rise Contrast S.S.'s high starting E.V. Indicate G.N.'s low starting E.V. on Plot Indicate S.S.'s high starting E.V. on Plot
Contrast
Highlight G.N.'s Jan 2000 E.V. high Foreshadow S.S.'s fall Clause S.S.'s high starting E.V. Clause G.N.'s low starting E.V.
Highlight G.N.'s Feb 2000 E.V. dip Contrast S.S.'s Jan 2000 E.V. low Clause Foreshadow G.N/S.S. E.V. Switch
Contrast
Highlight G.N.'s November 2000 E.V. high Contrast S.S.'s March 2000 rise Clause S.S.'s Jan 2000 E.V. low Clause G.N.'s Jan 2000 E.V. high
Contrast
Contrast S.S.'s Jun 2001 low Clause S.S.'s March 2000 rise Clause G.N.'s G.N.'s Feb 2000 E.V. dip
Contrast
Clause S.S.'s Jun 2001 low Clause G.N.'s November 2000 E.V. high
Fig. 2. Partial discourse and realization plan for Fig. 3. The story-layer is omitted because we
use a modified version of the tripartite model from Barot et al. [1, 2] where we select a pattern
externally from the planner.
Interactive Generation of Narrative Reports 125
works could be potentially set up in our proposed analysis workflow and could lead to
interesting biographies that could be generated for authors, tools, systems, or ideas.
References
1. Barot, C., Potts, C.M., Young, R.M.: A tripartite plan-based model of narrative for narra-
tive discourse generation. In: Proceedings of the Joint Workshop on Intelligent Narrative
Technologies and Social Believability in Games at the 11th AAAI Conference on Artificial
Intelligence and Interactive Digital Entertainment, pp. 2–8 (2015)
2. Barot, C., et al.: Bardic: generating multimedia narrative reports for game logs. In: Working
Notes of the AIIDE Workshop on Intelligent Narrative Technologies (2017)
3. Battad, Z., Si, M.: Apply storytelling techniques for describing time-series data. In: Rouse,
R., Koenitz, H., Haahr, M. (eds.) ICIDS 2018. LNCS, vol. 11318, pp. 483–488. Springer,
Cham (2018). https://doi.org/10.1007/978-3-030-04028-4 56
4. Blei, D.M., Ng, A.Y., Jordan, M.I.: Latent Dirichlet allocation. J. Mach. Learn. Res. 3(Jan),
993–1022 (2003)
5. Booker, C.: The Seven Basic Plots: Why We Tell Stories. A&C Black (2004)
6. Brehmer, M., Lee, B., Bach, B., Riche, N.H., Munzner, T.: Timelines revisited: a design
space and considerations for expressive storytelling. IEEE Trans. Vis. Comput. Graph. 23(9),
2151–2164 (2017)
7. Engel, O.: Clusters, recipients and reciprocity: extracting more value from email communi-
cation networks. Proc. Soc. Behav. Sci. 10, 172–182 (2011)
8. Erete, S., Ryou, E., Smith, G., Fassett, K.M., Duda, S.: Storytelling with data: examining
the use of data by non-profit organizations. In: Proceedings of the 19th ACM Conference on
Computer-Supported Cooperative Work & Social Computing, CSCW 2016, pp. 1273–1283.
Association for Computing Machinery, New York (2016)
9. Fikes, R.E., Nilsson, N.J.: STRIPS: a new approach to the application of theorem proving to
problem solving. Artif. Intell. 2(3–4), 189–208 (1971)
10. Freeman, H.: Shape description via the use of critical points. Pattern Recogn. 10(3), 159–
166 (1978). https://doi.org/10.1016/0031-3203(78)90024-9. https://www.sciencedirect.com/
science/article/pii/0031320378900249. The Proceedings of the IEEE Computer Society
Conference
11. Nahian, M.S.A., Tasrin, T., Gandhi, S., Gaines, R., Harrison, B.: A hierarchical approach for
visual storytelling using image description. In: Cardona-Rivera, R.E., Sullivan, A., Young,
R.M. (eds.) ICIDS 2019. LNCS, vol. 11869, pp. 304–317. Springer, Cham (2019). https://
doi.org/10.1007/978-3-030-33894-7 30
12. Klimt, B., Yang, Y.: The Enron corpus: a new dataset for email classification research. In:
Boulicaut, J.-F., Esposito, F., Giannotti, F., Pedreschi, D. (eds.) ECML 2004. LNCS (LNAI),
vol. 3201, pp. 217–226. Springer, Heidelberg (2004). https://doi.org/10.1007/978-3-540-
30115-8 22
13. Liu, S., Zhou, M.X., Pan, S., Song, Y., Qian, W., Cai, W., Lian, X.: TIARA: interactive, topic-
based visual text summarization and analysis. ACM Trans. Intell. Syst. Technol. (TIST) 3(2),
1–28 (2012)
14. McKenna, S., Henry Riche, N., Lee, B., Boy, J., Meyer, M.: Visual narrative flow: exploring
factors shaping data visualization story reading experiences. Comput. Graph. Forum 36(3),
377–387 (2017)
15. Meehan, J.R.: TALE-SPIN, an interactive program that writes stories. In: IJCAI, vol. 77, pp.
91–98 (1977)
Interactive Generation of Narrative Reports 127
16. Oard, D., Webber, W., Kirsch, D.A., Golitsynskiy, S.: Avocado research email collection
LDC2015T03. Linguistic Data Consortium, Philadelphia (2015). https://doi.org/10.35111/
wqt6-jg60
17. Robertson, J., Harrison, B., Jhala, A.: Interactive summarization for data filtering and triage.
In: The Thirty-Third International Flairs Conference (2020)
18. Staff, E.: Timeline: a chronology of Enron corp. New York Times (2006). https://www.
nytimes.com/2006/01/18/business/worldbusiness/timeline-a-chronology-of-enron-corp.
html
19. Vesanto, J., Hollmén, J.: An automated report generation tool for the data understanding
phase. In: Abraham, A., Jain, L., van der Zwaag, B.J. (eds.) Innovations in Intelligent Sys-
tems. Studies in Fuzziness and Soft Computing, vol. 140, pp. 203–219. Springer, Heidelberg
(2004). https://doi.org/10.1007/978-3-540-39615-4 8
20. Wang, Y., Sun, Z., Zhang, H., Cui, W., Xu, K., Ma, X., Zhang, D.: DataShot: automatic
generation of fact sheets from tabular data. IEEE Trans. Vis. Comput. Graph. 26(1), 895–
905 (2020)
21. Wilson, G., Banzhaf, W.: Discovery of email communication networks from the Enron cor-
pus with a genetic algorithm using social network analysis. In: 2009 IEEE Congress on
Evolutionary Computation, pp. 3256–3263 (2009)
22. Wongsuphasawat, K., et al.: Voyager 2: augmenting visual analysis with partial view spec-
ifications. In: Proceedings of the 2017 CHI Conference on Human Factors in Computing
Systems, pp. 2648–2659 (2017)
23. Young, R.M., Pollack, M.E., Moore, J.D.: Decomposition and causality in partial-order plan-
ning. In: AIPS, pp. 188–194 (1994)
Lean-Back Machina: Attention-Based
Skippable Segments in Interactive Cinema
1 Introduction
The fruition of interactive cinema provides new opportunities for telling stories
within large entertainment industries. Interactive cinema offers both new venues
c Springer Nature Switzerland AG 2021
A. Mitchell and M. Vosmeer (Eds.): ICIDS 2021, LNCS 13138, pp. 128–141, 2021.
https://doi.org/10.1007/978-3-030-92300-6_12
Attention-Based Skippable Segments in Interactive Cinema 129
for artists to explore and new types of experiences for the audience. Because the
opportunities appear so vast, understanding how to design the experiences that
maintain cinema’s appeal is essential.
The term lean-back serves as a metaphor that describes traditional media
experiences such as cinema [16,26] – We lean back, and behold the spectacle
unfolding unlike lean-forward mediums such as games [11]. For the narrative
absorption, it has been argued that active choice becomes a disruption [5,27] -
including even just the mediation of choice through an interface [25].
Passive Brain-Computer Interfaces [33] (BCIs) have therefore become of
interest in recent projects of brain-controlled film [23,27] as a mean to tap into
the mind of the viewer, and to apply passive control for changes in a film for a
more personalized experience. Previous studies have sought to artistically aug-
ment and personalize the film experience, such as using BCI measurements of
attention and eyelid blinks to control sound, shots, and timelines [22]. It appears
however that conscious control, though passive, can still disrupt the narrative
immersion [22,26]. Proposed lean-back interactions seek to increase narrative
immersion without posing demands to the user, or them necessarily being con-
scious of the control [26].
This project seeks to leverage lean-back interactions in a practical manner
for the viewer. Cinema as a product for the masses arguably suffers, as smaller
tangents of a plot or embellishments of its story world are discarded to provide
a lean product with mass appeal—sometimes seen with director’s cut opposed
to the studio’s cinematic version. Here, interactive cinema can earn a functional
quality as a way to indulge both the creator’s and viewers’ interests at a more
individual and subjective level.
Therefore, this project investigates how individual interests can be catered
to with the use of a framework for skipping past segments of a film found bor-
ing by the viewer, while maintaining a coherent narrative with authorial con-
trol. Our own short film, which was an adaptation of scenes from the movie Ex
Machina (2014)1 , included 4 zones with content that could be skipped based on
the viewer’s attention. With a between-subject research design, we aimed to find
a difference in narrative engagement. Though no significant difference was found
between the two versions of the experiences in self-reported narrative engage-
ment, there is still knowledge to inform future endeavors with brain-controlled
film.
This paper will first review related work, then present the background for
brain-controlled cinema and areas researched for the presented framework. Then
we will describe the experiment and the evaluation methods, followed by a dis-
cussion of the main findings of the study, and a conclusion.
2 Related Work
Interactive cinema has been experimented with for many decades [27], though the
first to apply passive control by measuring physiological responses was Tikka’s
1
https://www.imdb.com/title/tt0470752/.
130 N. E. Raursø et al.
installation ’The Obsession’ [31]. Since then, the term of ’neurocinematics’ has
been coined by Hasson et al. [13] who investigated films’ effect on our brains, and
now, in turn, how this can affect the film itself, with the field of brain-controlled
films. This section will look into some examples of these.
In a study by Pike et al. [22] with the film called The Disadvantages of
Time Travel [2], the occurrence of viewers’ eyelid blinks, as well as attention
and meditation levels, are used to control the view of the narrative as well
as blending between video layers. In continuation of this research, Ramchurn
et al. [26] conducted a study of the brain-controlled film The MOMENT [3].
The film blends narrative threads and soundtracks using an attention-based
algorithm with data from the previous scene to determine the combination. The
algorithm was designed to maintain continuity while also allowing for variation
and subconscious control. A drop in attention triggers a cut between threads,
the thread continues if attention is maintained or increased. Pike et al. reflected
on their work with brain-controlled films [24], and proposed themes for research
questions yet to be answered, among them, how individuals’ subjectivity can be
catered to and whether viewers should be conscious of control.
Kierkels and Pun [17] investigated if the viewer’s interest in movie scenes
could be discerned using various physiological measures, not including EEG.
They found no correlations between interest and movie genre, arousal and
valence, but that changes in interest level across a scene and between view-
ers were detectable. Kirke et al. [19] used physiological measurements, among
them EEG, evaluated at specific points in order to determine which ending of
the film a viewer would find most exciting. They found that their system was
possibly able to succeed in this task, but that it was inconclusive to what degree
the system did so by appropriate selection of measurements.
This study attempts to explore how consumer-grade BCIs can be leveraged
further in designing experiences with practical benefits for the audience and
does not require the creators to completely rethink their art in the creation of
interactive films.
3 Background
2
http://neurosky.com.
132 N. E. Raursø et al.
Fig. 1. The conceptual framework for Attention Zones and its segments.
A zone contains at least two segments, as first, the viewer needs to be intro-
duced to the content before it can be determined whether there is an interest in
the topic. The first segment serves as a hook, to assess if the succeeding segments
are of interest. Therefore, only succeeding segments are skippable. Attention is
compared to a pre-zone baseline, with no major plot points that could spike
attention. If there is more than two segments, then baseline for the third seg-
ment is the second segment. If a particular segment is not of interest to the
viewer and the segment is thereby skipped, we also need to consider what comes
after the segment. After all, if the viewer was disinterested in a particular aspect
or part of the story, continuing with something boring would risk losing the
interest altogether. So, we want to re-engage them with a new beat of the story
- here, elements of suspense, surprise and curiosity should be considered as a
way to amend lowered interest.
To apply this framework, an interactive film needed to be produced. Due
to resources and scope of the project, it was decided to base our short film
production on an existing movie. The movie Ex Machina (See footnote 1) is
about the programmer Caleb who performs a Turing test on tech mogul Nathan’s
newest creation, the humanoid AI Ava. It was chosen as it covers different topics
that could be elaborated upon or condensed and featured several plot points
that could be used in a short film of its own right.
134 N. E. Raursø et al.
The short film created that applies this framework contains four zones with
content that can be skipped if the viewer’s attention drops. The content of the
zones was considered for how they elaborate on topics of the film, topics that
might be of different interest to viewers. Zone 1, as illustrated in Fig. 2, features
entirely new content in the film, while the remaining zones and film are adapted
from several scenes of the original Ex Machina script.
The themes of the remaining 3 zones concern respectively; Caleb’s work and
his personal life, the philosophical Knowledge Argument, and Ava testing Caleb.
The exciting plot developments for each of these zones are respectively; Ava
asking if Caleb likes Nathan, Caleb asking Ava if she knows he is brought there
to test her, and Ava asking what will happen if she fails his test.
The short film used in the experimental setup can be seen here3 , and lasts
13 min, 3 min of which are skippable. The link features annotations for when the
different zones and segments starts. Our short film production was made using
virtual production techniques, more specifically with the use of live LED wall
in-camera virtual production. This is a state of the art technique where output
from the real-time game engine Unity to a live LED video wall. to a live LED
wall is used together with real-time DMX lighting and camera tracking to create
final-pixel imagery in camera, and has been employed in the recent and popular
series The Mandalorian 4 [14,15].
5 Methods
To evaluate if the presented framework can be used to increase narrative engage-
ment, a lean-back interactive short film was produced as described in the previous
section.
With a between-subject experimental design, two independent groups were
tested; The experimental group viewed an interactive version, and the control
group viewed a non-interactive version. Sampling of participants was done by
convenience sampling. To ensure equal distribution of skips in both groups, par-
ticipants in the non-interactive group watched a previous participant’s version
from the experimental group. The setup featured a projector to give a cinema-like
3
https://youtu.be/9qROrN4eVkU.
4
https://www.imdb.com/title/tt8111088/.
Attention-Based Skippable Segments in Interactive Cinema 135
5
https://store.neurosky.com/pages/mindwave.
136 N. E. Raursø et al.
6 Results
The experiment was conducted on 24 students, 9 females and 15 males in age
range of 20–27 years old. Two participants had only seen some of the original
movie, 11 had seen it, and 11 had not. In the control group six participants had
not seen the original movie, two had seen some if it. In the experimental group,
five had not seen it. The following sections will first present findings between
the experimental and control conditions, then qualitative data gathered from
interviews and observations, and finally results of assessment of the skipping
algorithm.
6.1 Experiment
A normal distribution was observed in the data with Shapiro-Wilk test (p > .05)
homogeneity of variance with Levene’s test (p > .05). A two-tailed independent
t-test was used to detect significance in the result data. No significant difference
was found in the total self-reported narrative engagement scores between the
group viewing the interactive version (M = 41.08, SE = 6.39) and the group
viewing the non-interactive version (M = 41.92, SE = 6.53), t(22) = −.32, p =
.76, r = .07.
Upon deeper inspection of the data, the narrative engagement sub-scales
showed no indication of difference, nor did the continuation desire: Both condi-
tions had almost the same scores, with the control condition having a slightly
higher but non-significant score in both scales. However, the attention levels
between the groups (see Fig. 3 (A)) differed, as the group viewing the interac-
tive version (M = 46.77, SE = 5.14) had higher attention levels than the group
viewing the non-interactive version (M = 43.11, SE = 4.44), t(22) = 1.87, p =
.04, r = .39.
Fig. 3. A) Box-plots depicting the measured attention between experimental and con-
trol group. B) Depicts the attention levels across the film with smoothed by a rolling
average. C) Depicts the self-reported interest scores for each of the zones.
The confusion matrix, which was developed to assess the performance of the
skipping algorithm compared to expected skips based on participants’ interest
ratings, showed the following results: true positive (correctly identified skip) =
12, true negative (correctly identified no skip) = 49, false positive (incorrectly
identified skip) = 25, false negative (incorrectly identified no skip) = 10. Deriva-
tions from the confusion matrix are presented in Table 1.
7 Discussion
mechanism was ambiguous and covert it its working, some users still attempted
to exert control. In our design, could viewers be able exert control over their
attention to elicit skips by disengaging with the content? How this would work
at a conscious or subconscious level, and how this relates to the viewing experi-
ence, and if it is even desirable, require further research to be answered.
8 Conclusion
In recent years, the usage of passive control in interactive cinema has opened
unexplored territory with the field of brain-controlled cinema. Previous projects
had used a ‘loose’ mapping in assigning functionality to the metric of ’attention’
from consumer-grade Brain-controlled Interfaces.
With this project it has been explored how individual film viewers’ subjective
interests can be catered to, by attempting to use the attention measurement as an
indication of interest in specific parts of a film. A short prototype film adaptation
of scenes from the movie Ex Machina was made, that contained 4 zones with
content that could be skipped according to the viewer’s level of attention.
The results from the between-subject test showed no significant difference
in narrative engagement between the experimental and control group. There
was however an indication of a significant difference found in their measured
attention-levels, with the experimental group having a higher average atten-
tion than the control group. Furthermore, there appeared to be indications that
gender differences in attention-levels affected the performance of the skipping
algorithm, as well as whether participants had seen the original film. Addition-
ally, it was found that skips often did not seem to correlate with participants’
self-reported interests in the presented topics to any reliable degree. Uncertain-
ties about how the attention data can be interpreted and how affected it was by
external factors make it difficult to form any definitive conclusions. However, due
to the average attention being significantly higher for the experimental group, it
suggests that the interactive film with skippable segments did have some impact
that remains to be fully accounted for.
References
1. Imdb - black mirror: Bandersnatch. https://www.imdb.com/title/tt9495224/.
Accessed 15 Oct 2021
2. Imdb - the disadvantages of time travel. https://www.imdb.com/title/tt8072006/.
Accessed 15 Oct 2021
3. Imdb - the moment. https://www.imdb.com/title/tt7853742/. Accessed 15 Oct
2021
4. Barto, A., Mirolli, M., Baldassarre, G.: Novelty or surprise? Front. Psychol. 4, 907
(2013)
140 N. E. Raursø et al.
5. Ben-Shaul, N.: Can narrative films go interactive? New Cinem.: J. Contemp. Film
2(3), 149–162 (2004)
6. Benford, S., et al.: Contesting control: journeys through surrender, self-awareness
and looseness of control in embodied interaction. Human-Comput. Interact. 36,
1–29 (2020)
7. Bitner, R., Le, N.-T., Pinkwart, N.: A concurrent validity approach for EEG-based
feature classification algorithms in learning analytics. In: Nguyen, N.T., Hoang,
B.H., Huynh, C.P., Hwang, D., Trawiński, B., Vossen, G. (eds.) ICCCI 2020. LNCS
(LNAI), vol. 12496, pp. 568–580. Springer, Cham (2020). https://doi.org/10.1007/
978-3-030-63007-2 44
8. Brewer, W.F., Lichtenstein, E.H.: Stories are to entertain: a structural-affect theory
of stories. J. Pragmat. 6(5–6), 473–486 (1982)
9. Busselle, R., Bilandzic, H.: Measuring narrative engagement. Media Psychol. 12(4),
321–347 (2009)
10. Csikszentmihalyi, M.: Flow: The Psychology of Optimal Experience. Harper Peren-
nial (1991)
11. Dewdney, A., Ride, P.: The New Media Handbook. Routledge, London (2006)
12. Gottlieb, J., Lopes, M., Oudeyer, P.Y.: Motivated cognition: neural and compu-
tational mechanisms of curiosity, attention, and intrinsic motivation. In: Recent
Developments in Neuroscience Research on Human Motivation. Emerald Group
Publishing Limited (2016)
13. Hasson, U., Landesman, O., Knappmeyer, B., Vallines, I., Rubin, N., Heeger, D.J.:
Neurocinematics: the neuroscience of film. Projections 2(1), 1–26 (2008)
14. Kadner, N.: The Virtual Production Field Guide volume 1, vol. 1. Epic Games
(2019)
15. Kadner, N.: The Virtual Production Field Guide Volume 2, vol. 2. Epic Games
(2021)
16. Katz, H.E.: The media handbook : a complete guide to advertising media selection,
planning, research, and buying/Helen Katz. LEA’s Communication Series, 3rd edn.
L. Erlbaum Associates, Mahwah (2006)
17. Kierkels, J.J.M., Pun, T.: Towards detection of interest during movie scenes. In:
First International Workshop on Multimodal Interactions Analysis of Users in a
Controlled Environment (MIAUCE2008) (2008)
18. Kim, Y., Moon, J., Lee, H.J., Bae, C.S., Sohn, S.: Integration of electroencephalog-
raphy based services into consumer electronics. In: 2012 IEEE 16th international
symposium on consumer electronics, pp. 1–2. IEEE (2012)
19. Kirke, A., et al.: Unconsciously interactive films in a cinema environment-a demon-
strative case study. Digit. Creat. 29(2–3), 165–181 (2018)
20. Lim, C.G., et al.: A brain-computer interface based attention training program for
treating attention deficit hyperactivity disorder. PLoS ONE 7(10), e46692 (2012)
21. Pari-Larico, S., Llerena-Urday, B., del Carpio, Á.F., Rosas-Paredes, K., Esquicha-
Tejada, J.: Evaluation of brain attention levels using Arduino and Neurosky Mind-
wave EEG according to age and sex (2020)
22. Pike, M., Ramchurn, R., Benford, S., Wilson, M.L.: # scanners: exploring the
control of adaptive films using brain-computer interaction. In: Proceedings of the
2016 CHI Conference on Human Factors in Computing Systems, pp. 5385–5396
(2016)
23. Pike, M., Ramchurn, R., Wilson, M.L.: # scanners: integrating physiology into
cinematic experiences. In: Proceedings of the 2015 ACM SIGCHI Conference on
Creativity and Cognition, pp. 151–152 (2015)
Attention-Based Skippable Segments in Interactive Cinema 141
24. Pike, M., Ramchurn, R., Wilson, M.L.: Two-way affect loops in multimedia expe-
riences. In: Proceedings of the 2015 British HCI Conference, pp. 117–118 (2015)
25. Polaine, A.: The flow principle in interactivity. In: Proceedings of the Second Aus-
tralasian coNference on Interactive Entertainment, pp. 151–158 (2005)
26. Ramchurn, R., Martindale, S., Wilson, M.L., Benford, S.: From director’s cut to
user’s cut: to watch a brain-controlled film is to edit it. In: Proceedings of the 2019
CHI Conference on Human Factors in Computing Systems, pp. 1–14 (2019)
27. Ramchurn, R., Martindale, S., Wilson, M.L., Benford, S., Chamberlain, A.: Brain-
controlled cinema. In: Nijholt, A. (ed.) Brain Art, pp. 377–408. Springer, Cham
(2019). https://doi.org/10.1007/978-3-030-14323-7 14
28. Schoenau-Fog, H., Louchart, S., Lim, T., Soto-Sanfiel, M.: Narrative engagement
in games - a continuation desire perspective. In: Proceedings of the 8th Interna-
tional Conference on the Foundations of Digital Games (FDG 2013). Foundations
of Digital Games Conference Proceedings, vol. 8, pp. 384–387. Society for the
Advancement of the Science of Digital Games (2013). sNO/CNO: 156495 Foun-
dations of Digital Games Conference Proceedings; Foundations of Digital Games:
The 8th International Conference on the Foundations of Digital Games, FDG 2013;
Conference date: 14–05-2013 Through 17–05-2013
29. Siler, T.L.: Neurotranslations: interpreting the human brain’s attention system.
AIMS Med. Sci. 3(2), 179–202 (2016)
30. Smith, T.J.: The attentional theory of cinematic continuity. Projections 6(1), 1–27
(2012)
31. Tikka, P., Vuori, R., Kaipainen, M.: Narrative logic of enactive cinema: obsession.
Digit. Creat. 17(4), 205–212 (2006)
32. Vorderer, P., Klimmt, C., Ritterfeld, U.: Enjoyment: at the heart of media enter-
tainment. Commun. Theory 14(4), 388–408 (2006). https://doi.org/10.1111/j.
1468-2885.2004.tb00321.x
33. Zander, T.O., Kothe, C.: Towards passive brain-computer interfaces: applying
brain-computer interface technology to human-machine systems in general. J. Neu-
ral Eng. 8(2), 025005 (2011)
A Quantified Analysis of Bad News
for Story Sifting Interfaces
1 Introduction
Social simulation has demonstrated itself to be a powerful tool for creating works
of interactive digital storytelling [18,20,22]. Social relationships and dynamics
are often a core part of resonant stories, and therefore finding ways to computa-
tionally represent them remains an important area of research for the interactive
narrative community. Thus far, many narratives driven by social simulation have
relied on the properties of emergent narrative [32], in which no formal represen-
tation of narrative structure (e.g., an Aristotelian dramatic arc [2]) is encoded,
and instead narrative patterns naturally emerge through the actions of the simu-
lated agents. These patterns are then surfaced to readers/players through mech-
anisms appropriate to the experience in question, in a process that has come to
be known as story sifting [15,29]: somehow searching the simulated space to find
narratively salient gold amidst the dross of other simulated output.
Story sifting technology is still nascent, however, with ample room for
exploration. A notable example of this can be seen in the award-winning
c Springer Nature Switzerland AG 2021
A. Mitchell and M. Vosmeer (Eds.): ICIDS 2021, LNCS 13138, pp. 142–156, 2021.
https://doi.org/10.1007/978-3-030-92300-6_13
A Quantified Analysis of Bad News for Story Sifting Interfaces 143
game/theatrical performance piece Bad News [34]. Powered by the social simu-
lation system Talk of the Town [30], Bad News asks players to explore a unique
procedurally generated small American town with one hundred and forty years
of simulated history. The player is tasked with three objectives: discovering the
identity of a resident that has recently passed away, searching for the deceased’s
next of kin, and delivering the eponymous bad news of their passing. Interaction
in this piece is real-time spoken communication between the player and a live
human performing as the game’s Actor, who may be asked to assume the role of
any of the hundreds of the town’s residents. Behind the scenes, another human
deemed the Wizard (so termed from Wizard-of-Oz interaction techniques [31])
is performing the role of story sifter; scouring the simulated history for material
that is narratively salient. In truth, the Wizard bears the responsibility of both
story sifter and drama manager [23], as the material they find must not only
be narratively salient, but also in service of the ultimate quality of the player’s
experience, whose needs are dynamically changing based on a number of factors.
A simple example of one such factor is the experience duration; the designers of
Bad News determined that the ideal length for a playthrogh is between forty-five
and sixty minutes. Thus, the Wizard had the task of searching the simulated his-
tory for interesting narrative patterns (e.g., starcrossed lovers, rival businesses,
disgruntled family dynamics, and other stressors of smalltown life), but revealing
them in such a way as to not make the game too easy or too hard (e.g., laying
breadcrumbs that would bring the player to the next of kin immediately upon
the start of play—even if narratively salient—would likely produce a short and
ultimately narratively unsatisfying experience). This is all made more complex
by virtue of the fact that the Wizard does not present their findings directly,
but rather conveys this information to the Actor, who bears ultimately respon-
sibility over how and when any of this information is revealed to the player.
Moreover, the simulated history has no GUI interface; it can only be explored
via Python code. This led to the Bad News experience resulting in the Wizard
livecoding in Python, the Actor performing with the player, and the Wizard and
Actor simultaneously communicating with each other via text messages while
performing their respective roles.
Though perhaps the spectacle of this is unique to Bad News—and indeed
likely contributed to its positive reception—it renders the game frustratingly
difficult to distribute. The work presented in this paper is an attempt to enable
all of the above in a manner which consolidates the responsibilities of Wizard
and Actor into a single role. At present, the cognitive load of simultaneously
role playing and story sifting are too much for a single person to perform. Thus,
we present the initial development of an interface intended to enable the consol-
idation of the Actor and Wizard roles, tentatively titled the WizActor inter-
face. Its design draws directly from quantitative and qualitative data drawn
from dozens of Bad News playthroughs. However, though Bad News informs the
design of this interface, it is intended to be generalizable enough to be applicable
in any situation in which an interactive narrative driven by social simulation and
storysifting leverages a human storyteller. This has a wide range of application
144 B. Samuel et al.
2 Related Work
This work finds itself at a cross section of many different branches of research and
practice (akin to other cross-disciplinary techniques such as Research Creation
[33]). One such branch is the theatre, both traditional and experimental. Bad
News is a piece of interactive theatre such as Coffee! A Misunderstanding, Sleep
No More, and the performances of Improbotics [13,19,41]. Of these three it is
most closely aligned with Improbotics, as they are both fully improvised. How-
ever, Bad News dialogue is informed by the simulation but is generated by the
human Actor, whereas improbotics uses natural language generation techniques
to produce dialogue for both human and robotic performers. Bad News is also
a piece of procedural content generation. Though there are many approaches to
procedural content generation [39,40], Bad News takes much inspiration from
the world generation and simulation techniques of Dwarf Fortress [1]. It is also
a piece of table top role-playing akin to Dungeons and Dragons [9], a domain
which has seen its own share of of PCG research ranging from the generated
music of Bardo [25], the characters and relationships of Fiascomatic [11], and
the settings and story beats of Dear Leader [12].
Gamalyzer [24] analyzed playthroughs of Prom Week in a quantitative man-
ner, but was concerned with similarity/dissimilarity between playthroughs as
opposed to any specific patterns found in the playthroughs. Rameshkumar and
Bailey [27] performed high-level textual analysis of role playing sessions from the
Critical Role series, but as with Gamalyzer, there was no assessment of patterns
in the playthroughs. Leece and Jhala [16] used sequence mining similar to our
approach to find patterns of actions in StarCraft, but to our knowledge no one
has investigated logs of actions of simulation sifting/role playing before.
A Quantified Analysis of Bad News for Story Sifting Interfaces 145
Fig. 2. Cumulative distributions for print related commands (2a for Subject, 2b for
Information)
Fig. 3. Cumulative distributions for assignment related commands (3a for Label, 3b
for Value)
Turning to the Assignment commands, we first note that they were much
less common than the Print commands – 1,048 Assignment vs 11,260 Print
commands. Another note is that the Assignment commands are not as extreme
as Print commands – the most common label (‘bars’) only accounts for 7.5%
and the most common value (d.mother, d.father) only accounts for 6.8% of
the Assignment commands respectively. We note that the discrepency in that
the most common label does not match the most common information – this
comes from the fact that ‘bars’ was often assigned to multiple times in the
face of missing information (e.g., First g.city.businesses of type(’Bar’),
then g.city.businesses of type(’Hotel’), and finally g.city.former
businesses of type(’Bar’)). The five most common labels were (1) bars at 79
commands, (2 & 3) mom & dad at 66, (4) ‘sibs’ (siblings) at 58, and (5) hotel
at 32. Figure 3a shows the CDF for assignment labels. The five most common
values were (1) d.mother, d.father at 66 commands, (2) d.siblings at 58
, (3) g.city.businesses of type(’Bar’) at 44, (4) d.love interest at 37,
and (5) nok[0] at 24. Figure 3b shows the CDF for assignment information.
148 B. Samuel et al.
Both Print and Assignment commands take the form of a Pareto distri-
bution [26] – sometimes known as the 80–20 distribution (i.e., 80% of outcomes
are due to 20% of causes and 20% of outcomes are due to 80% of causes). In the
case of Bad News Print commands – the top 5 most common subjects account
for over 50% of all use-cases, and the top 20 account for two-thirds. This means
that being able to prioritize those 20 subjects will cover the bulk of interactions,
easing the requirements for a WizActor.
dad.occupations[-1]
dad.occupations[-1].terminus.date
A Quantified Analysis of Bad News for Story Sifting Interfaces 149
future moves simple and accessible. Though ultimately intended for use in many
simulation-driven performance experiences, in the parlance of Bad News, it is
intended to facilitate the Wizard ’s most common needs (outlined in Sect. 3),
such that the roles of Actor and Wizard could conceivably be collapsed and
performed by a single individual with minimal cognitive overhead.
The prototype is being developed as a web application using Data Driven
Documents (the D3.js library) [3]. Figure 7 depicts an early version of the tool,
demonstrating its initial functionality. Though the inspiration for the design
of this tool comes in part from the aforementioned analysis of Bad News
playthroughs, it is intended to be usable for a variety of domains and game
experiences that depend on human performers (e.g., the Bad News “actor”, a
Tabletop Role Playing Game “game master”, etc.) parsing through, discovering,
and delivering simulation-generated content. To showcase a different narrative
domain, the underlying data used in the following examples was generated via
a Kismet simulation that represents a collection of thirty regency era socialites
attending a ball. Each character in the simulation is composed of individual
character traits (such as “snoopy”, “extrovert”, and “drunkard”), statuses (such
as “sad”, “drunk”, and “embarrassed”) and is depicted as a node in a D3 force
directed graph. The edges of the graph are directed and represent the source
character’s affinity towards the target; unbounded integers that could be nega-
tive (indicating disdain) or positive (indicating affection). In the following figures
the solid green edges represent positive affinities and the dashed red edges rep-
resent negative ones.
Figure 5 presents a view of the entire interface. The top portion depicts the
complete cast of characters, who can be added or removed from the graph indi-
vidually via checkboxes. Beneath that there is a slider that adjusts the “relation-
ship threshold” value which impacts how many edges between nodes are drawn.
Beneath that is trait and status information about a selected character (here
Penelope Leosfel), and to the right of that are buttons that highlight narratively
salient patterns (such as “Find a Rival” or “Find all Friends”). Beneath all that
is, finally, the graph itself.
Being able to add characters to the graph individually is helpful; in Bad
News, for example, players were often most concerned with learning more infor-
mation about characters they had already met. This allows for the WizActor
to gradually add characters as the player encounters them, focusing the infor-
mation offered by the WizActor interface to be most pertinent to the current
game. However, as discussed in Sect. 3, the Wizard also commonly needs to query
the simulation for information regarding characters that the player has yet to
encounter – either at the start or in media res. However, these queries nearly
always were in relationship to characters the player had already encountered or
learned about – e.g., finding the family tree of an encountered character. The
Kismet world used for this didn’t have familial relations, but finding friends, love
interests, and rivals are spiritually similar tasks. There are a set of buttons (e.g.,
Find a Friend and Find a Rival ) that adds a new character to the graph who
fulfills the desired pattern when pressed. This automated search for patterns is
A Quantified Analysis of Bad News for Story Sifting Interfaces 151
Fig. 5. An early prototype of the WizActor interface using generated Kismet data
simulating a regency era ball. Solid green lines signify positive affinities, dashed red
lines represent signify negative affinities. (Color figure online)
a simple form of story-sifting, and enables the WizActor to quickly discover new
characters to weave into the story. The patterns presented here (“friend”, “rival”,
etc.) were authoring decisions specific to this particular simulated world. Kismet
permits authors to create their own patterns of arbitrary complexity, and the
WizActor interface dynamically populates its search functionality based on the
current world.
Though its primary intended use case is to facilitate live story-sifting while
in-game, the WizActor interface is also a helpful mechanism for better under-
standing the generative space of a Kismet simulation, and can serve as a tool to
verify that generated content is desirable. Figures 6, 7a, and 7b highlight some
of this sanity checking.
Figure 6 depicts the results of clicking the “Find all Friends” button; it intro-
duces into the graph every character who appears in the “Friends” pattern. Note
how the resulting graph is clearly partitioned; the simulation has naturally cre-
ated a group of people who all love each other (through the mutually connected
solid green lines towards the top of the graph) and another group that all despise
each other (connected through the dashed red lines towards the bottom). This
is an interesting phenomenon in and of itself—it enables the author of the sim-
ulation to reflect on whether this result is desirable and to modify the Kismet
file accordingly—but it also reveals potentially interesting characters for narra-
tive moves. Here, there are a few characters, such as Peggy Ameham, Geoffrey
Edstock, and Dulcima Kirkford who are generally liked but who also have con-
nections to the “angry” group. Those characters serve as narratively interesting
152 B. Samuel et al.
candidates if the WizActor needs to find a way to ferry the player from one
group to another.
Figures 7a and 7b highlight the significance of the aforementioned “relation-
ship threshold” value. In this Kismet world, every character has an affinity value
for every other character; depicting them all would result in a graph that is dif-
ficult to decipher. The threshold value determines the minimum magnitude a
relationship must have in order to be drawn (e.g., if the value is twelve, then
dashed red edges will only be drawn at negative twelve or less, and solid green
edges will only be drawn at twelve and above). Figure 7a has the threshold at
twenty, and consequently shows significantly fewer links, while Fig. 7b relaxes
the threshold to eight (and begins veering dangerously close to losing decipher-
ability). Higher numbers can be used to discover particularly strong connections
between characters, which can form the basis of narrative content. For example,
Fig. 7a tells us that Alfred and Lionel have a stronger affinity for one another
than any other pair in the simulation, and Camille and Lambert are linked via
a very interesting chain of people who strongly hate each other. Both of these
phenomena were naturally occurring—but buried—in the simulation, but the
WizActor tool unearthed them easily, offering potential narrative fuel to in turn
to be surfaced to the player. Similarly, lowering the threshold consequently shows
more links, further confirming the two ’factions’ here and revealing additional
characters that could be considered “on the border” between the two groups.
Fig. 6. A depiction of the interface after the “Find all Friends” button is pressed
This paper presents initial findings from a quantified analysis of Wizard logs
from the game Bad News. These findings revealed the commands most often
entered by the Wizard while playing Bad News, which in turn speaks to the
simulated information and emergent narrative most crucial to be surfaced for
A Quantified Analysis of Bad News for Story Sifting Interfaces 153
Fig. 7. Two WizActor graphs depicting character affinities. Figure 7a includes all thirty
characters with the “relationship threshold” value set to twenty. 7b is the same but
with the value at eight.
References
1. Adams, T., Adams, Z.: Dwarf fortress. Game [Windows, Mac, Linux], Bay 12
(2006)
2. Aristotle: Poetics (Penguin Classics). Penguin Classics (1997). http://www.
amazon.com/Poetics-Penguin-Classics-Aristotle/dp/0140446362
3. Bostock, M., Ogievetsky, V., Heer, J.: D3 data-driven documents. IEEE Trans.
Vis. Comput. Graph. 17(12), 2301–2309 (2011)
4. Compton, K., Kybartas, B., Mateas, M.: Tracery: an author-focused generative
text tool. In: Schoenau-Fog, H., Bruni, L.E., Louchart, S., Baceviciute, S. (eds.)
ICIDS 2015. LNCS, vol. 9445, pp. 154–161. Springer, Cham (2015). https://doi.
org/10.1007/978-3-319-27036-4 14
5. Compton, K., Mateas, M.: Casual creators. In: Proceedings of the Sixth Interna-
tional Conference on Computational Creativity, p. 228 (2015)
6. DeKerlegand, D., Samuel, B., Leichman, J.: Encoding socio-historical exegesis as
social physics predicates. In: International Conference on the Foundations of Digital
Games, pp. 1–9 (2020)
7. Gage, P.: A new algorithm for data compression. C Users J. 12(2), 23–38 (1994)
8. Garbe, J., Reed, A.A., Dickinson, M., Wardrip-Fruin, N., Mateas, M.: Author
assistance visualizations for ice-bound, a combinatorial narrative. In: FDG (2014)
9. Gygax, G., Cook, D.: The Dungeon Master Guide, No. 2100, 2nd edn. (Advanced
Dungeons and Dragons). TSR, Inc., Lake Geneva (1989)
10. Horswill, I.: Imaginarium: a tool for casual constraint-based PCG. In: Proceedings
of the AIIDE Workshop on Experimental AI and Games (EXAG) (2019)
11. Horswill, I.D.: Fiascomatic: a framework for automated fiasco playsets. In: Eleventh
Artificial Intelligence and Interactive Digital Entertainment Conference (2015)
12. Horswill, I.: Dear leader’s happy story time: a party game based on automated
story generation. In: WS-16-21. pp. 39–45. AAAI Workshop - Technical Report, AI
Access Foundation, 12th AAAI Conference on Artificial Intelligence and Interactive
Digital Entertainment, AIIDE 2016; Conference date: 08–10-2016 Through 09–10-
2016 (January 2016)
13. Kiai, D.: Coffee! a misunderstanding (2014)
14. Kreminski, M., Dickinson, M., Mateas, M.: Winnow: a domain-specific language
for incremental story sifting. In: Proceedings of the AAAI Conference on Artificial
Intelligence and Interactive Digital Entertainment, vol. 17, pp. 156–163 (2021)
15. Kreminski, M., Dickinson, M., Wardrip-Fruin, N.: Felt: a simple story sifter. In:
Cardona-Rivera, R.E., Sullivan, A., Young, R.M. (eds.) ICIDS 2019. LNCS, vol.
11869, pp. 267–281. Springer, Cham (2019). https://doi.org/10.1007/978-3-030-
33894-7 27
16. Leece, M.A., Jhala, A.: Sequential pattern mining in starcraft: brood war for short
and long-term goals. In: Tenth Artificial Intelligence and Interactive Digital Enter-
tainment Conference (2014)
17. Marsella, S.C., Pynadath, D.V., Read, S.J.: Psychsim: agent-based modeling of
social interactions and influence. In: Proceedings of the International Conference
on Cognitive Modeling, vol. 36, pp. 243–248 (2004)
18. Mateas, M., Stern, A.: Façade: an experiment in building a fully-realized interactive
drama. In: Game Developers Conference, vol. 2, pp. 4–8 (2003)
19. Mathewson, K.W.: Humour-in-the-loop: improvised theatre with interactive
machine learning systems (2019)
A Quantified Analysis of Bad News for Story Sifting Interfaces 155
20. McCoy, J., Treanor, M., Samuel, B., Reed, A.A., Wardrip-Fruin, N., Mateas, M.:
Prom week. In: Proceedings of the International Conference on the Foundations of
Digital Games, pp. 235–237 (2012)
21. McCoy, J., Treanor, M., Samuel, B., Reed, A.A., Mateas, M., Wardrip-Fruin, N.:
Social story worlds with comme il faut. IEEE Trans. Comput. Intell. AI Games
6(2), 97–112 (2014)
22. Meehan, J.R.: Tale-spin, an interactive program that writes stories. In: Proceedings
of the 5th International Joint Conference on Articial Intelligence, Aug. 1977. vol.
1, pp. 91–98 (1977)
23. Nelson, M.J., Mateas, M., Roberts, D.L., Isbell, C.L., Jr.: Declarative optimization-
based drama management (dodm) in the interactive fiction anchorhead. Comput.
Graph. Appl. 26(3), 32–41 (2006)
24. Osborn, J.C., Samuel, B., McCoy, J.A., Mateas, M.: Evaluating play trace (dis)
similarity metrics. In: Tenth Artificial Intelligence and Interactive Digital Enter-
tainment Conference (2014)
25. Padovani, R.R., Ferreira, L.N., Lelis, L.H.: Bardo: emotion-based music recommen-
dation for tabletop role-playing games. In: Thirteenth Artificial Intelligence and
Interactive Digital Entertainment Conference (2017)
26. Pareto, V.: Cours d’économie politique: professé à l’Université de Lausanne, vol.
1. F. Rouge (1896)
27. Rameshkumar, R., Bailey, P.: Storytelling with dialogue: a critical role dungeons
and dragons dataset. In: Proceedings of the 58th Annual Meeting of the Association
for Computational Linguistics, pp. 5121–5134 (2020)
28. Roberts, D.L., Isbell, C.L.: A survey and qualitative analysis of recent advances
in drama management. Int. Trans. Syst. Sci. Appl. Spec. Issue Agent Based Syst.
Hum. Learn. 4(2), 61–75 (2008)
29. Ryan, J.: Curating simulated storyworlds. Ph.D. thesis, UC Santa Cruz (2018)
30. Ryan, J.O., Summerville, A., Mateas, M., Wardrip-Fruin, N.: Toward characters
who observe, tell, misremember, and lie. In: Eleventh Artificial Intelligence and
Interactive Digital Entertainment Conference (2015)
31. Ryan, J.O., Summerville, A.J., Samuel, B.: Bad news: a game of death and com-
munication. In: Proceedings of the 2016 CHI Conference Extended Abstracts on
Human Factors in Computing Systems, pp. 160–163 (2016)
32. Ryan, M.L.: From narrative games to playable stories: toward a poetics of interac-
tive narrative. Storyworlds A J. Narrative Stud. 1, 43–59 (2009)
33. Sadati, S.H., Mitchell, C.: Serious game design as research-creation to address
sexual and gender-based violence. Int J Qual Methods 20, 16094069211046130
(2021)
34. Samuel, B., Ryan, J., Summerville, A.J., Mateas, M., Wardrip-Fruin, N.: Bad news:
an experiment in computationally assisted performance. In: Nack, F., Gordon, A.S.
(eds.) ICIDS 2016. LNCS, vol. 10045, pp. 108–120. Springer, Cham (2016). https://
doi.org/10.1007/978-3-319-48279-8 10
35. Shibolet, Y., Knoller, N., Koenitz, H.: A framework for classifying and describing
authoring tools for interactive digital narrative. In: Rouse, R., Koenitz, H., Haahr,
M. (eds.) ICIDS 2018. LNCS, vol. 11318, pp. 523–533. Springer, Cham (2018).
https://doi.org/10.1007/978-3-030-04028-4 61
36. Smith, A.M., Lewis, C., Hullett, K., Smith, G., Sullivan, A.: An inclusive taxonomy
of player modeling. University of California, Santa Cruz, Technical report. UCSC-
SOE-11-13 (2011)
156 B. Samuel et al.
37. Smith, G., Whitehead, J.: Analyzing the expressive range of a level generator. In:
Proceedings of the 2010 Workshop on Procedural Content Generation in Games,
pp. 1–7 (2010)
38. Summerville, A., Samuel, B.: Kismet: a small social simulation language. In: Sum-
merville, A., Samuel, B. (2020, September). Kismet: a Small Social Simulation
Language. In: 2020 International Conference on Computational Creativity (ICCC).
(Casual Creator Workshop). ACC (2020)
39. Summerville, A., et al.: Procedural content generation via machine learning
(pcgml). IEEE Trans. Games 10(3), 257–270 (2018)
40. Togelius, J., Yannakakis, G.N., Stanley, K.O., Browne, C.: Search-based procedural
content generation: a taxonomy and survey. IEEE Trans. Comput. Intell. AI Games
3(3), 172–186 (2011)
41. Worthen, W.B.: The written troubles of the brain: sleep no more and the space of
character. Theatre J. 79–97 (2012)
Digital Storytelling and Social Interaction
in Cultural Heritage - An Approach for Sites
with Reduced Connectivity
Abstract. As mobile devices gradually became more pervasive, there has been
an evolution in their use in the museum and cultural heritage context. Tradition-
ally, audio guides and mobile mediated narratives in general, have been a single
user experience. However, as a museum visit is, in most cases, social in nature,
different experimental applications attempt to implement sociality in a variety of
experience designs. In this work, we seek to support digitally-mediated social
interaction for collocated users experiencing digital storytelling in a cultural her-
itage context, addressing the issue of decentralized communication across devices
that are collocated, without a single point of coordination. This requirement stems
from the fact that most cultural sites lack the reliable infrastructure to support
network connectivity of any kind. To support social interaction, we extend an IDN
authoring framework with support for collaborative experiences and evaluate it
through a study with twelve participants.
1 Introduction
Audio guides have played a significant role in museums and heritage sites [33]. As
mobile devices gradually became more pervasive, there has been an evolution in the use
of this technology, yet its full potential still has not been fully realized. The technology is
there, however the issue has shifted to that of an effective experience design that would
convey successfully the objective to promote visitor engagement. Digital storytelling is
one of the recognized and widely used approaches to that effect [21, 27] and an increasing
number of cultural institutions employ narratives of different forms to enhance the visitor
experience, “overextending” in some cases the use of the term “digital storytelling” to
even include purely informational content [13].
Traditionally, mobile mediated narratives in this context have been a single user
experience [34, 36, 37]. Visitors are given a device or asked to download an app, and are
left to wander alone, isolated by the other visitors by the use of headphones. However, a
museum visit is in most cases social in nature [3, 7, 18, 19] as people rarely come alone
[2]. Cultural sites have been characterized as “social places where people in groups
gather to collectively share their experiences and collectively make meaning” [6]. And
mobile devices have indeed been accused of isolating the visitor from their surroundings
and each other [30]. Past research has confirmed that the use of personal mobile devices
by collocated users often isolates the group’s members, absorbing their attention on their
device – creating thus an isolating bubble or “cocoon” [10, 25]. Recent work has been
attempting to reconcile sociality in heritage sites with mobile digital experiences and to
reverse this effect, turning the mobile phones into the incentive or focus for meaningful
communication and interaction between the visitors [1, 12, 14, 20].
In this work we focus on digitally-mediated social interaction for collocated users
experiencing mobile-mediated digital storytelling in a cultural heritage context. In this
experience type, the users are allowed to proceed in parts of the experience at their own
pace. At specific points, they are asked by the system to interact with each other, either
by engaging in conversation in relation to the narrative themes or in another type of
shared activity. We build upon previous work, that focused on the conceptual aspects
of this approach [14], to explore an implementation of system-based synchronization
of the collocated devices, beyond user-initiated coordination. To that end, we propose
an implementation framework, building upon the Story Maker tool authoring tool for
interactive digital narratives [28, 29]. Our objective is (a) to explore the effect of this
type of system-driven synchronization on the visitor experience and (b) to address the
authoring aspects of the approach, taking into account that the content creators in this
context are not experienced programmers but rather digital heritage practitioners. As
an additional challenge, we address the issue of decentralized communication across
devices that are collocated, without a single point to orchestrate the synchronization.
In Sect. 2 we present in more detail the background research that has been the
motivation for this work, while in Sect. 3 we identify specific requirements for our
implementation, offer a brief description of the Story Maker framework and present our
implementation approach which extends this framework. Section 4 focuses on the user
evaluation, followed by the Conclusions in Sect. 5.
Social interaction has been examined in a cultural context as an aspect of the visit e.g.
[5, 22, 35] and from the perspective of the study of group behavioral patterns [4, 24].
Existing research has sought to promote specific forms of social interaction between
groups, including situations where visitors are not in close proximity [16, 17]. Different
applications implement sociality in a variety of experience designs. Gamified approaches
are typically based on a treasure hunting theme [2, 15], whereas other approaches employ
individual user trajectories that combine isolated exploration and reflection with shared
activities and may include collocated interactions [8, 20]. An example of this latter type
combines mobile drama with coordinated narrative variations to induce conversations
in small groups [1, 9].
Our work focuses on the use of digital storytelling in a cultural heritage context as
an individual mobile user experience, combined with points where the users are guided,
Digital Storytelling and Social Interaction in Cultural Heritage 159
by the device, to interact with each other, usually to engage in some type of common
exploration activity or, most often, dialogue. One example is the facilitated dialogue
experience for a group of 6–8 users for the York Minster Cathedral described in [11].
In this case one of the users assumes an enhanced role, to facilitate the dialogue that is
prescribed as a social activity in between the storytelling parts of the experience.
The work in [14, 23] describes a mobile-based digital storytelling experience for two
users, “The story of Building 52”, developed for the archaeological site of Çatalhöyük,
near Konya in Turkey. The experience follows a story-centric, drama-based digital story-
telling approach where two fictional characters, Abla, a Neolithic inhabitant and Archie,
an archeologist, narrate events from their past on the site. The evaluation of the expe-
rience has established with statistical significance the positive outcomes of designing
explicit social interaction points between the users into digital storytelling. Conversation
between group members has positive effects on visitors’ engagement with the overall
experience and significantly fosters historical empathy and learning. The same study also
confirmed that integrating social interaction points in a digital storytelling, experienced
individually by each user, does not break the user’s engagement with the story flow.
In this case a technologically minimalistic approach was adopted. The interaction
points were implemented as individual prompts to each user to initiate the interaction.
There was no underlying synchronization mechanism to ensure that the digital story-
telling experiences proceeded in parallel. In other words, there was no means to control
whether users reached the same interaction point at the same time, so as to start the
activity together. The users themselves controlled their own experience pace, coordi-
nating (or not) their actions. The study observed a variety of approaches by the users,
concluding that “the decision not to enforce or even technically support synchronization
did not seem to hamper the experience for users. Participants were able to drift apart if
they wanted to, and pairs who decided to align their experiences were able to socially
coordinate their digital stories, while at the same time maintaining individual control of
their personal device” [14], finding this result consistent with the work of [8].
An important point, however, was that the ease with which the participants’ self-
coordinated is most likely due also to other factors and cannot be considered a gen-
eralizable argument against the use of system-supported synchronization, at least not
without more targeted experiments to explore participant self-coordination. The authors
consider the specific story design, including “structure, branching factor and depth over
the script level, temporal placement and frequency of interaction points, etc.” [14] as
important factors to that effect, becoming even more crucial if significant differences
are introduced between the narrative variations offered to the participants.
Furthermore, the fact that the experience type is an interactive narrative, even if it is
designed for the users to proceed in a roughly parallel trajectory, the fact that users are
“able to proceed at their own pace, exploring in-depth some parts of the story, or moving
along quickly, or even skipping other parts”, does “not guarantee that they will arrive at
any given [social] interaction point at the same time” [14].
In this sense, as the authors also conclude, although Suh et al. point out that social
negotiation is more or less a prerequisite for shared experiences [26], it cannot technically
support synchronization. To this end, in this work we propose and assess an approach
160 E. Vrettakis et al.
3 System Overview
In this section we present our approach, starting from the definition of requirements.
Content Variations. The users are simultaneously shown content, which is different
for every user, and then prompted to discuss with their companion(s) about it. This
activity aims to support the information gap technique where users are presented each
with different information and asked to exchange their perspectives on the topic. As an
example, the user following the character of Abla in the story example of the Çatalhöyük
“Building 52” will be presented with a representation of the Neolithic home whereas
Archie, with a view of the home as it was while excavated.
“Gift Giving” Activity. One user may offer a virtual object to another user. The object
may appear on the other user’s screen. This object is selected from a collection, motivat-
ing users to explore more deeply this collection so as to decide which is more appropriate.
An example in the context of the Çatalhöyük digital storytelling experience could be
“Select the most appropriate funerary object for your companion. Explain to them the
reasons for this choice.”
Facilitator Control. One of the participants can be assigned the role of facilitator. This
role has advanced rights in comparison with the others, allowing the user to control the
pace of the experience and make decisions for the whole group in branching points.
Polling Branching Point. Participants are asked to select an option in a menu, rep-
resenting a branching point of the story. Then, the system evaluates their choices to
make an informed decision for the path of the story they will all, collectively follow.
The author can choose between different activity variations, in which users can see the
polling results and, optionally, a pie-chart with the vote distribution. This activity aims to
engage users in a decision-making process based on the current state of the experience.
In case of a tie, the final decision can be either facilitator-driven or system-driven.
Lastly, our implementation aims to address the issue of reduced connectivity by
exploring solutions for decentralized communication across devices that are collocated,
without a single point to orchestrate the synchronization.
In the remainder of this section we focus on our implementation approach for
these requirements, aiming for the right balance between promoting visitor engagement
through social interaction and an authoring approach appropriate for non-experts.
For this work we build upon the Story Maker [28] authoring tool. The Story Maker offers
a framework for creating IDN experiences, from start to end. It consists of a) the Story
Design Editor (SDE) used to define the structure and concept of the branching narrative,
and b) the Storyboard Editor (SBE) [29] used to transform the author’s concept into an
interactive experience.
SDE’s objective is to support the design of the story concept (i.e. the story plot) and
structure through the use of a WYSIWYG editor, extended with additional features for
branching narrative creation. The narrative text can be organized in Parts, defined as
coherent parts of the narrative and Branches that interconnect them. Parts are defined as
abstract building blocks that can be used to define the different pieces of the narrative.
They can be freely linked together using Branches, leading from one Part to another.
Branches help define the possible paths of the experience. For the authors to control
the availability of a path at any given point of the experience, we introduced the use
of Conditions and Tags. Tags can be associated with Parts or Branches, and be used in
a Branch condition to show or hide the specific Branch, giving the authors additional
control over the flow of the experience, depending on the paths selected by the users.
The conditions are determined by enforcing boolean constraints on the tags a user has
encountered, defining whether a tag has been met during an experience, or not.
162 E. Vrettakis et al.
The Story Maker Storyboard Editor (SBE) covers the production of the final IDN
experience. Parts of the script can be implemented in multimedia format through sets of
Screens and Branches through Menus or Jumps. Jumps are implicit transitions, selected
automatically according to conditions that are set while the user proceeds.
When complete, the produced experience can be made available through a web-based
player as well as a native mobile application for Android devices. In this work, we extend
the latter by adding support for collaborative and facilitated features, as described in the
following section.
3.3 Implementation
To address the need for the framework to be effective in a context of reduced connectivity,
we opted to use Google’s Nearby Connections [31] for the P2P functionality. Google’s
Nearby Connections is a P2P networking framework that enables an application to
discover, connect to, and exchange data with other nearby devices, regardless of network
connectivity status. It leverages a variety of wireless interfaces available on modern
mobile devices (i.e. Wi-Fi, Bluetooth and BLE) to offer high-bandwidth, low-latency,
encrypted P2P networking. Nearby Connections aims to eliminate the vagaries posed
by different android versions and wireless interface hardware vendors.
Our framework provides three types of requests: a) Base Requests, b) Screen
Requests and c) Menu Requests.
Base Requests: The Base Requests class contains requests related to framework-
specific actions such as handling the player component’s status, moving between screens,
bookkeeping, exception handling, etc.
All requests not directly handled by the Story Player (i.e. Base Requests) are del-
egated to the respective Screen’s MVC to handle them. Programming hooks (i.e. han-
dleRequest) are exposed to enable developers to easily extend any Screen type (and their
subtypes) with collaborative features.
Screen Requests: Screen requests are specializations of Base Requests, extending them
with an ID attribute. The ID corresponds to the ID of the Screen the request refers to;
if the request ID attribute matches the ID of the Screen currently shown, the handler
associated with the request gets called. An example would be toggling between the text
and image views of a Screen.
Types of Menu Requests. We have implemented three types of Menu Requests: a) Role
Requests, b) Poll Requests and c) Remote Choice Requests.
Role Requests: Role requests are tightly coupled with role assigning menus. A role can
be used to restrict the user’s navigation to certain paths of the story, thus introducing role-
specific subpaths that serve alternative content between users of the same experience.
Currently, the cardinality of the relationship between users and roles is 1–1 but one
could consider a future extension to also handle role-groups. A collaborative experience
Digital Storytelling and Social Interaction in Cultural Heritage 163
can have at most one instance of this type of menu. This limitation is imposed by the
way we chose to implement the dynamic nature of roles. The role essentially acts as a
prefix for Tags collected while traversing the experience graph. Using the example of the
“Building 52” experience, the two variations of the story, implementing the perspective
of Abla and Archie can be realized by the corresponding two roles. By making a choice
on a Role Menu, the user acquires a role Tag, that can also be used for conditionals. So in
the story we may denote that the user with the role Abla, while viewing the Screen about
Bucrania (the decorative use of bull’s horns), “collects” the tag “PerspectiveBucrania”
with the expression Abla:PerspectiveBucrania).
Poll Requests: Poll requests are tightly coupled with voting menus. These menus give
the ability for a group of users to collectively choose which path of the story they want
to follow. Users can change their votes as many times as they want while the voting
procedure is still open. The voting procedure stops when all users have successfully
voted at least once. The result is either the Branch that gathered the majority of votes or
a random selection between the finalist Branches in case of a tie.
Remote Choice Requests: Remote choice requests enable a user to change the course
of the plot for another user. Conceptually, this functionality can be thought of as the
ability of a user to remotely make a Branch selection for the other. This request type
may support the implementation of the gift giving activity, described in Sect. 4.
Currently, the Story Maker tool framework provides four distinct templates for menus
(i.e. Simple, QR-Code, Interactive Image and NFC). Every Screen or Menu can be easily
converted to its collaborative counterpart; i.e. we can add polling capabilities to any menu
type just by implementing the related API hooks. These new collaborative capabilities,
effectively provide us with twelve new types of menus.
Storytelling Engine in a valid state (i.e. there has to be at least one user-visible Screen or
an eligible Branch that leads to a user-visible Screen). Thus, to identify terminal states,
it is enough to explore all of the outgoing Branches of a Part; if there is no transition
leading to a Part bearing some user-visible content, then the state is considered terminal.
To identify whether the user can reach a valid state, the Storytelling Engine traverses the
story graph in a Depth-First Search manner until an eligible path gets discovered.
Evaluating the Global Experience State. Tags are essentially equivalent to globally
scoped Boolean variables; the existence of a tag corresponds to logical True while the
absence of it, to a logical False. Tags can be combined to represent Boolean conditions
except for OR. This poses some limitations on the conditionals that an author can express.
The user acquires tags by visiting parts or following Branches. Tags can also be set as
prerequisites for certain graph paths; if the user’s tag inventory matches the criteria
imposed for following a certain path, then the user can follow that path.
We have extended the Branch condition evaluation mechanism to support tags in
the form Abla:House. The previous expression states that the user with the role Abla
needs to acquire the tag House to follow a specific Branch. Having a global view of each
user’s state, lets us express and evaluate conditionals that range beyond the scope of a
single device. This way tags can be used to implement synchronization primitives. We
can distinguish several patterns of usage for this mechanism, some of which are:
The author may want to establish explicit synchronization points in which all the
users wait for the “slowest” user to reach them (i.e. barriers). To this end, they need to
follow these steps:
Digital Storytelling and Social Interaction in Cultural Heritage 165
• Create a branch li for transitioning ri to pt , ri ∈ Roles. Let pi be the part li is in. The
li ’s may belong to different parts.
• Let also pi assign the tag ti when visited.
• Add the condition ri on each li .
• Add the condition rj :tj on each li , ∀ rj ∈ Roles \ { ri }.
For example, suppose we have three users, bearing the roles of Abla, Archie and
Felix that are currently in parts pAbla , pArchie and pFelix respectively. We also have the
branches lAbla , lArchie , lFelix that are in pAbla , pArchie and pFelix respectively and point to
ptarget . Lastly, pAbla , pArchie and pFelix assign the tags tAbla , tArchie and tFelix respectively
when visited. Then the conditional expressions for the respective branches will be:
If the Storytelling Engine, taking into account the global story state, evaluates the
conditions of li to True, then li can be selected by the user with the role ri , ri ∈ {Abla,
Archie, Felix}. Following the same strategy, the author can impose synchronization
points on a subset of users, just by including only the roles in question.
Authoring Synchronization Points. In this section we present the definitions of uni-
lateral and bilateral synchronization points. These can be mixed, thus creating complex
synchronization primitives between users.
Unilateral Synchronization Points: There exists p, p ∈ Parts such that the users with r1 ,
r2 ∈ Roles paths cross on p. Let’s suppose that we want r2 to transition to p before r1 ; in
other words, we want to block r1 from reaching p as long as r2 has not. To achieve this
behavior, we have to do the following steps:
The condition r1 ensures that the branch will be followed only by the user bearing
the role r1 and r2 :t renders l1 unavailable while r2 hasn’t acquired t.
Bilateral Synchronization Points. There exists p, p ∈ Parts such that the users with r1 ,
r2 ∈ Roles paths cross on p. Let’s suppose that we want r1 and r2 to be able to transition
to p together. To achieve this behavior, we have to do the following steps:
• Add the conditions r2 and r1 :t1 to l2 . The condition r2 ensures that the branch will be
followed only by the user holding role r2 and r1 : t1 renders l2 unavailable while r1
hasn’t acquired t1 . Let also the part containing l2 assign the tag t2 to r2 .
For example, suppose we want the user with the role Archie to have viewed the
Screen “Bucrania” thus having collected the tag “PerspectiveBucrania”, as well as the
user with the role Abla to have collected the tag House to let them both move forward.
In Fig. 1, on the left device, the user gets informed that they should wait for their
partner, while in the middle, we can see that the next button is no longer available and
therefore the user is forced to wait for their partner. Once the user of the device on the
right reaches the synchronization point (right), the next-screen button becomes available
again on both devices, allowing the users to proceed.
4 Evaluation
To assess the users’ outlook on the enforced synchronization points, we have organized
and conducted six trial sessions with twelve participants in six groups of two members
each. Our objective was to examine general user experience aspects of the approach as
well as how the users negotiated synchronization with user control over the experience.
After a brief introduction to the research objectives and the signing of consent forms,
the users were handed one mobile device each. For the purposes of the evaluation we
used the digital storytelling experience “The story of building 52” [14, 23]. The users
experienced the story off-site, in a room dedicated to the evaluation and after a brief
video presentation about the site. They were instructed to proceed within the experience
as they saw fit, even to end it if they wished to. While engaged in the experience, an
evaluator was discreetly observing and took notes. At the end of the session, the users
filled in a brief questionnaire. Each session took approximately 45 min to complete.
The participants’ age ranged between 19 and 45 years old (3 were in the range 19–
25, 5 in 26–35 and 4 in 36–45). 75% identified themselves as “female” while the rest
25% as “male”. Most participants stated moderate to high interest in History (91.6%)
and Archaeology (83.3%). The participants were selected so as to be familiar with each
Digital Storytelling and Social Interaction in Cultural Heritage 167
other (group-wise), either as friends or colleagues. This was a study design decision
since in real conditions people that visit a cultural site together know each other.
The questionnaire to be completed by all users at the end of the evaluation [32]
was based on the one in [14] and included visitor demographic characteristics, quality of
experience, social presence and questions related to the user agency and sense of control.
This last part focused on how the users perceive the interaction points in terms of their
effect on their sense of control over the experience. The questionnaire combined state-
ments, rated by the respondents on a five-point Likert scale, from “Strongly disagree”
to “Strongly agree”, with open-ended questions designed to record qualitative feedback.
The original questionnaire was presented in Google Forms format.
4.1 Results
Our study showed that most participants were positive about sharing an IDN experience
with another person (μ = 4.58, σ = 0.51). The participants enjoyed the time spent
on conversation activities stimulated by the narrative (μ = 4.67, σ = 0.49). However,
further research is needed to validate this finding (e.g. would a group of non-acquainted
participants also share a positive view on shared experiences?). The majority of the
participants (μ = 4.58, σ = 0.51) found the mobile experiencing app easy to understand
and use.
All participants were in favor of synchronized experiences, also arguing that the
kind of experience we delivered would not be possible without it. The majority (75%)
mentioned that the interactive nature of the experience helped them keep their focus on
the plot while also creating a sense of responsibility towards their partners. Several of
the participants (66.6%) stated that the topics raised by the experience for discussion
motivated them to work in tandem, provoked interesting discussions and information
sharing, something they would otherwise be reluctant to do. Moreover, we observed
that most pairs of participants had a, give-or-take, steady pace relative to their partner’s.
We believe that this is due to the careful design of the experience and the appropriate
placement of the synchronization points. All participants stated that at some point they
had to wait for their companion. The majority used the waiting time as an opportunity to
review the information offered (either on the screen they were in or by moving to previous
screens) and to reflect in order to prepare for the discussion with their companions
regarding the questions posed by the narrative. We also noticed that some of the users
indulged in looking over their companions’ screens.
The most prominent negative aspect was that some participants felt insecure while
waiting for their companions. This feeling of insecurity stemmed from not knowing
in what state their companion was in (i.e. has she made a selection or is she waiting
for me?). Lastly, two participants stated they would prefer to receive additional status
updates for the actions performed by their partners (i.e. which role they chose, etc.).
In this work we extend the Story Maker framework to support social interaction in dig-
ital storytelling for cultural heritage. This meant that we had to adapt the principles of a
168 E. Vrettakis et al.
framework initially designed for single user interactive, to work in a multi-user setting.
To this end, we had to enforce guidelines for authoring the story structure and imple-
ment synchronization principles. An additional aim was to implement the collaborative
experience in a decentralized fashion, to address the fact that most cultural sites lack the
reliable infrastructure to support stable network connectivity.
The evaluation results provided useful insights regarding the usability and effective-
ness of our approach as most of our users were quite positive about sharing an IDN
experience with another and argued in favor of system based synchronization, feeling
that it would not be possible without it. Taking into account that the implementation of
synchronization and the social interaction points is based on the use of tags and con-
ditions and does not require programming skills, we are confident that with brief train-
ing authors would be able to use the tool to create group digital storytelling activities.
However, further evaluation is needed to validate this approach with story creators.
Additionally, our design, by exposing programming hooks, enables developers to
easily extend Screens (and any subtype of it) with collaborative features and is also open
to further extensions. As an example, taking advantage of the byte stream payload type
provided by the Nearby Connections API to support features such as streaming audio or
video between devices (i.e. a tour guide acting as a facilitator, could use her device to
show every participant a specific exhibit through live video feed). File-sharing can also
be used to share in-app recorded streams and commemorative photos.
Furthermore, we aim to introduce physical user collaboration aspects. For example,
many devices come with an integrated NFC tag reader/writer; one could think of an
extension that will utilize the underlying hardware by exploiting Android BEAM API.
This could introduce playfulness to some activities by requiring the users, for example,
to touch their devices together for a specific interaction to happen (e.g. “Put your devices
together to unite the house’s floor plan and explore all its objects”). Android’s BEAM
can also be used to provide functionality similar to Remote Choice Requests.
The development of such collaborative functionality in the Story Maker tool implies
enhancements for the authoring tool itself. Specifically, conditional branches are a topic
of Story Maker tool that is not thoroughly tested and its API may completely change in
the future (i.e. there are thoughts about developing a simple domain-specific language
that will add support for actual variables, a full set of Boolean algebra operators, and
more). Moreover, the tool could be extended to provide semi-automated functionality
aiding the author in specifying roles of the user and other similar functionalities. Finally,
the authoring tool could identify problems of the plot (for example unreachable paths)
at authoring time and not at runtime.
Acknowledgements. This research has been co-financed by the European Regional Development
Fund of the European Union and Greek national funds through the Operational Program Compet-
itiveness, Entrepreneurship and Innovation, under the call RESEARCH –CREATE –INNOVATE
(project “ARIA - Augmenting the Reception of music through Innovative solutions and Archives”,
code:T2EDK-02084)”.
Digital Storytelling and Social Interaction in Cultural Heritage 169
References
1. Callaway, C.B., Stock, O., Dekoven, E.: Experiments with mobile drama in an instrumented
museum for inducing conversation in small groups. ACM Trans. Interact. Intell. Syst. 4(1),
1–39 (2014)
2. Cabrera, J.S., et al.: Mystery in the museum: collaborative learning activities using handheld
devices. In: Proceedings of the 7th International Conference on Human Computer Interaction
with Mobile Devices and Services (MobileHCI 2005). Association for Computing Machinery,
New York, NY, USA, pp. 315–318 (2005)
3. Dierking, L.D., Falk, J.H.: Family behavior and learning in informal science settings: a review
of the research. Sci. Educ. 78, 57–72 (1994)
4. Dim, E., Kuflik, T.: Group situational awareness: being together. In: Workshop on Personal-
ization in Mobile and Pervasive Computing, in Conjunction with UMAP 2009. Trento, Italy
(2009)
5. Emmanouilidis, C., Koutsiamanis, R.A., Tasidou, A.: Mobile guides: taxonomy of archi-
tectures, context awareness, technologies and applications. J. Netw. Comput. Appl. 36(1),
103–125 (2012)
6. Falk, J.H., Dierking, L.D.: Learning from Museums, 2nd edn. Rowman & Littlefield, Lanham,
Maryland (2018)
7. Falk, J.H.: Identity and the Museum Visitor Experience. Left Coast Press, Walnut Creek, CA
(2009)
8. Fosh, L., Benford, S., Reeves, S., Koleva, B., Brundell, P.: See me, feel me, touch me, hear
me: trajectories and interpretation in a sculpture garden. In: Proceedings of the SIGCHI
Conference on Human Factors in Computing Systems. ACM, New York, pp. 149–158 (2013)
9. Huws, S., John, A., Kid, J.: Evaluating the affective dimensions of Traces-Olion: a subtle mob
at St Fagans National Museum of History, Wales. In: IEEE Proceedings of the 3rd International
Congress and Expo of Digital Heritage, San Francisco, USA, pp. 26–30, October 2018. IEEE,
San Francisco (2018). (in press)
10. Jarusriboonchai, P., et al.: Personal or social? designing mobile interactions for co-located
interaction. In: Proceedings of the 8th Nordic Conference on Human-Computer Interaction:
Fun, Fast, Foundational (NordiCHI 2014). Association for Computing Machinery, New York,
NY, USA, pp. 829–832 (2014)
11. Gargett, K.: Re-thinking the Guided Tour: a Critical Exploration of the use of Co-creation,
Dialogue and Facilitation Practices for Democratic Social Engagement at York Minster. MA in
Cultural Heritage Management, Department of Archaeology. University of York, UK (2018)
12. Katifori, A., Kourtis, V., Perry, S., Pujol, L., Vayanou, M., Chrysanthi, A.: Cultivating mobile-
mediated social interaction in the museum: towards group-based digital storytelling experi-
ences. In: MW2016: Museums and the Web 2016. Published 15 January 2016. Consulted 24
July 2019. https://mw2016.museumsandtheweb.com/paper/cultivating-mobile-mediated-soc
ial-interaction-in-the-museum-towards-group-based-digital-storytelling-experiences/
13. Katifori, A., Karvounis, M., Kourtis, V., Perry, S., Roussou, M., Ioanidis, Y.: Applying interac-
tive storytelling in cultural heritage: opportunities, challenges and lessons learned. In: Rouse,
R., Koenitz, H., Haahr, M. (eds.) ICIDS 2018. LNCS, vol. 11318, pp. 603–612. Springer,
Cham (2018). https://doi.org/10.1007/978-3-030-04028-4_70
14. Katifori, A., et al.: Let them talk!”: exploring guided group interaction in digital storytelling
experiences. ACM J. Comput. Cult. Heritage (2020). https://doi.org/10.1145/3382773
15. Klopfer, E., Perry, J., Squire, K., Jan, M.F., Steinkuehler, C.: Mystery at the museum: a
collaborative game for museum education. In: Proceedings of the 2005 Conference on Com-
puter Support for Collaborative Learning: Learning 2005: The Next 10 Years! (CSCL 2005).
International Society of the Learning Sciences, pp. 316–320 (2005)
170 E. Vrettakis et al.
16. Kuflik, T., Sheidin, J., Jbara, S., Goren-Bar, D., Soffer, P., Stock, O., Zancanaro, M.: Support-
ing small groups in the museum by context-aware communication services. In: Proceedings
of the 12th International Conference on Intelligent User Interfaces (IUI 2007). ACM, New
York, pp. 305–308 (2007)
17. Kuflik, T., Stock, O., Zancanaro, M., Gorfinkel, A., Jbara, S., Kats, S., Sheidin, J., Kashtan,
N.: A visitor’s guide in an active museum: Presentations, communications, and reflection. J.
Comput. Cult. Heritage (JOCCH), 3–11 (2011)
18. Leinhardt, G., Knutson, K.: Listening in on Museum Conversations. Altamira Press, Walnut
Creek, CA, USA (2004)
19. McManus, P.M.: Good companions: more on the social determination of learning-related
behaviour in a science museum. Int. J. Mus. Manage. Curatorship 7, 37–44 (1988)
20. Perry, S., Roussou, M., Mirashrafi, S., Katifori, A., McKinney, S.: Collaborative, shared dig-
ital experiences supporting sites, meaning-making at heritage sites. In: Lewi, H., Smith, W.,
Cooke, S., vom Lehn, D. (Eds.) The Routledge International Handbook of New Digital Prac-
tices in Galleries, Libraries, Archives, Museums and Heritage Sites. Routledge International
Handbook, Routledge (2019)
21. Pujol, L., Roussou, M., Poulou, S., Balet, O., Vayanou, M., Ioannidis, Y.: Personalizing
interactive digital storytelling in archaeological museums: the CHESS project. In: Earl, G.,
Sly, T., Chrysanthi, A., Murrieta-Flores, P., Papadopoulos, C., Ro-manowska, I., Wheatley, D.
(Eds.) Archaeology in the Digital Era. Papers from the 40th Annual Conference of Computer
Applications and Quantitative Methods in Archaeology (CAA) (2013). Southampton, UK,
pp. 26–29, March 2012. Amsterdam University Press (2012). https://arno.uva.nl/cgi/arno/
show.cgi?fid=545855
22. Raptis, D., Tselios, N., Avouris, N.: Context-based design of mobile applications for museums:
a survey of existing practices. In: Proceedings of the 7th Conference on Human-Computer
Interaction with Mobile Devices and Services, Mobile HCI, Salzburg, Austria, pp. 153–160
(2005)
23. Roussou, M., Pujol, L., Katifori, A., Chrysanthi, A., Perry, S., Vayanou, M.: The museum as
digital storyteller: collaborative participatory creation of interactive digital experiences. In:
MW2015: Museums and the Web 2015. Published 31 January 2015. Consulted 28 Septem-
ber 2015. http://mw2015.museumsandtheweb.com/paper/the-museum-as-digital-storyteller-
collaborative-participatory-creation-of-interactive-digital-experiences/
24. Tolmie, P., Benford, S., Greenhalgh, C., Rodden, T., Reeves, S.: Supporting group interac-
tions in museum visiting. In: Proceedings of the ACM Conference on Computer Supported
Cooperative Work (CSCW 2014), pp. 1049–1059 (2014)
25. Turkle, S.: Alone Together: Why We Expect More from Technology and Less from Each
Other. Basic Books, New York (2011)
26. Suh, Y., Shin, C., Woo, W., Dow, S., MacIntyre, B.: Enhancing and evaluating users’ social
experience with a mobile phone guide applied to cultural heritage. Pers. Ubiquit. Comput.
15(6), 649–665 (2011). https://doi.org/10.1007/s00779-010-0344-2
27. Twiss-Garrity, B., Fisher, M., Sastre, A.: The art of storytelling: enriching art museum exhibits
and education through visitor narratives. In: Trant, J., Bearman, D. (Eds.) Museums and the
Web 2008. Montreal, Quebec, Canada: Archives & Museum Informatics (2008)
28. Vrettakis, E., Kourtis, V., Katifori, A., Karvounis, M., Lougiakis, C., Ioannidis, Y.: Nar-
ralive – creating and experiencing mobile digital storytelling in cultural heritage. Digit. Appl.
Archaeol. Cult. Heritage 15, e00114 (2019)
29. Vrettakis, E., et al.: The story maker - an authoring tool for multimedia-rich interactive
narratives. In: Bosser, A.-G., Millard, D.E., Hargood, C. (eds.) ICIDS 2020. LNCS, vol.
12497, pp. 349–352. Springer, Cham (2020). https://doi.org/10.1007/978-3-030-62516-0_33
30. Wessel, D., Mayr, E.: Potentials and challenges of mobile media in museums. Int. J. Interact.
Mobile Technol. 1(1), 32–39 (2007)
Digital Storytelling and Social Interaction in Cultural Heritage 171
Abstract. There are no recipes or rules to develop games, any more than there
are to develop stories. Our aim was to develop a serious game with the goal of pro-
moting discussion and awareness among children around nutrition: FlavourGame.
For this purpose, we needed not only to design game mechanics, but also to create
a narrative that provided meaning to the game experience, in order to ensure sub-
strate to the context of nutrition. In this paper, we present a framework that serves
as a narrative design tool used in the development of this game and which can be
employed in the creation of serious story-based games.
1 Introduction
Narrative design has emerged in the world of game design only in the last decade [1, 2],
as a way of uniting story writing and game design [3–5]. The complexity of this union
arises from the fact that they are both central systems for generating human experience,
and as such, when united they collide by the force with which they draw to themselves
the control of experiences [6]. Therefore, we are facing a process still fluid in ways of
doing [7].
In this paper, we present the beats&units story-game integration framework, which
consists of a layered information management system that supports the ideation stage of
narrative design of serious games. It is organized in two levels: 1) the operational level,
composed by a single layer materialized into a set of cards that can be used to build
de narrative with the game; and 2) the informational level, constituted the several base
layers with the information useful to operate the operational level.
with the game, and the informational level, in which features the information about the
story and the game. These are organized in four groups: the conceptual layers, in which
layer 1 is focused on framing of possibilities of meaning, layer 2 on the mechanics, and
layer 3 on narrative structure; the design layers, in which layer 4 inspects the gameplay,
and layer 5 the game plot; the units layer, which expands the units of narrative; and the
surface layer, which is the tool that results from all the other layers.
To connect story and game, layer 4 resorts to the Forest Paths method by Alexan-
der Swords [11], modeling the relationship between the character’s story, and player’s
actions. It establishes the player as a starting point and the goal as the finishing line,
whereas in the middle is where the game process is developed that guarantees the player’s
involvement with the storyworld.
Up until layer 4, we have the story that we want people to retain in their heads, a
fable, but that is different from the artifact that stimulates the fable itself. In this sense,
174 N. Zagalo et al.
layer 5 takes Todorov’s [10] structure and adapts it to Freytag’s [12], abandoning the
chronology of the story, and following the plot as an organization of information capable
of generating interest and surprise.
Each card represents a Beat (Fig. 3) in a specific Scene. The cards are building blocks
of the relationship between narrative and gameplay, able to be combined into linear or
multilinear sequences of events with no fixed numbers of Scenes or Beats per Scene.
Beats and Units Framework: A Story-Game Integration Framework 175
Fig. 3. Each card represents a Beat. By interconnecting Beats, one can work on flow, pacing and
balancing in linear and non-linear sequences of events.
The current version of the beats&units framework allowed us to speed up the understand-
ing of the game and story experience, making the information visual and hierarchical
in time. By resorting to it we were able to understand what the player knows and does
not know about the game and the story, at each moment. It thus becomes possible to
measure the evolution of each character, in the eyes of the player, to evaluate the narrative
progression, to understand its moments of tension and breathing, and to adjust the game
progress with the narrative progress.
In a first moment, we used the tool to define the fundamental information to offer just
before the game starts, as well as the information to introduce in the dialogues. The goal
of any game lays in its interaction, so we cannot continuously stop the game to tell the
story in non-interactive modes, we need to work on the different modes of exposition,
from the dialogues to the scene objects – e.g. character presentation, space, setting, etc. –
in order to introduce the information that allows the evolution of the characters and the
progression of the narrative. So, the tool was very useful in determining these elements,
because as the information is introduced in the game, and appears in the beats&units
framework, we can better understand the flow of the experience, and understand if we
need to speed up the exposure or slow it down.
We consider that the tool has delivered as expected, in terms of supporting the creation
and management of narrative design. Naturally, for larger projects, it might be necessary
to take the structure and apply it in a programmable environment, which could offer us
links between cells, transportation of variables, as well as incorporating dialogues and
actions directly into the final implementation of the game.
The next steps are to finalize the implementation of the game, and to perform
sets of tests with users, namely working pre-tests and post-tests, to understand how
much the game experience may or may not contribute to future behavioral changes, and
consequently find about the level of truth infused in the game intent.
Acknowledgments. The authors would like to acknowledge POCI-FEDER and FCT for funding
this Project, under the Grant Agreement No. POCI-01–0145-FEDER-031024.
176 N. Zagalo et al.
References
1. Brunette, L.: What’s a Narrative Designer? https://web.archive.org/web/20171110210044/
http://www.bigfishgames.com/blog/guide-to-narrative-design-and-writing-video-game-sto
ries/. Accessed 15 July 2021
2. Jubert, T.: The Narrative Designer. http://litshowcase.org/content/the-narrative-designer/.
Accessed 15 July 2021
3. Batchelor, J.: Four questions to help improve your narrative design [Chris Bate-
men]. https://www.gamesindustry.biz/articles/2019-07-16-four-questions-to-help-improve-
your-narrative-design. Accessed 15 July 2021
4. Heussner, T., Finley, T.K., Hepler, J.B., Lemay, A.: The Game Narrative Toolbox. Routledge,
Burlington (2015). https://doi.org/10.4324/9781315766836
5. McRae, E.: Narrative Design for Indies: Getting Started. Fiction Engine Limited (2017)
6. Zagalo, N.: Engagement Design: Designing for Interaction Motivations. Springer, Cham
(2020). https://doi.org/10.1007/978-3-030-37085-5
7. Heussner, T.: The Advanced Game Narrative Toolbox (2019).https://doi.org/10.1201/978135
1014397
8. Sloan, M.C.: Aristotle’s nicomachean ethics as the original locus for the septem circumstan-
tiae. Class. Philol. 105, 236–251 (2010). https://doi.org/10.1086/656196
9. Schell, J.: The Art of Game Design: A Book of Lenses. Morgan Kaufmann (2008)
10. Todorov, T., Weinstein, A.: Structural analysis of Narrative. Nov. A Forum Fict. (1969). https://
doi.org/10.2307/1345003
11. Swords, A.: Forest Paths Method For Narrative Design. Alexander Swords, Melbourne (2020).
12. Freytag, G.: Technique of the Drama: An Exposition of Dramatic Composition and Art.
University Press of the Pacific, Forest Grove (1864)
13. Zagalo, N.: Design de Narrativa, Desenho de significado na experiência interativa. Convoc. -
Rev. Ciências Da Arte. 7, 57–76 (2019)
14. McKee, R.: Story: Substance, Structure, Style and the Principles of Screen Writing. Harper
Collins, New York (1997)
Interactive Narrative Theory
Monster Power. Rebel Heart. Gay Sword.
Queer Structures and Narrative Possibility in PbtA Tabletop
Roleplaying Games
PS Berge(B)
1 Introduction
1.1 Queer TRPGs: A Troubled Legacy
The history of tabletop roleplaying games (TRPGs, often stylized TTRPGs) and queer-
ness is contentious. On the one hand, mainstream TRPGs (like Dungeons and Dragons)
have been constricted by the conservative roots of the medium. As Stenros & Sihvonen
note in their history of queer representation in TRPG sourcebooks, “queer sexualities
started to figure in the role-playing game books towards the end of the 1980s. How-
ever, in these early depictions, male homosexuality is presented as especially villainous,
traitorous, and deceitful” [5]. Early TRPGs featured limitations on character gender [6]
and even labeled queer identities as “sexual disorders” [5, 7]. Broadly speaking, the
evolution of queerness in TRPGs, like queer video games, has followed a two-pronged
path: 1) industry-supported games have made limited effort towards basic representation
while 2) independent TRPGs push the boundaries of queer game possibilities.
In TRPG spaces, like other game spaces, queerness can be understood in two modes:
1) representing the experiences and identities of queer folks and 2) “as a way of being,
doing, and desiring differently” [8]. As queer game scholar Bo Ruberg notes [8], this
latter model builds on decades of queer theory [9–11] that examines queerness as a force
that disrupts hegemonic structures. Queer game scholarship that addresses TRPGs has
focused predominantly on queer representation, usually by examining sourcebooks and
social dynamics. Stenros & Sihvonen argue for increased queer representation when
they write “even the most fleeting mention, a veiled remark, would signal that the game
world does indeed feature queer people” [5]. They call such signals ‘cues’ for queer
players—noting that “cues for queer play remain a controversial issue… The publisher
and designer, the creators of the urtext, signal if queer play is tolerated, encouraged, or
expected” [1]. While signals for queer representation in TRPGs are important, overfo-
cusing on cues neglects possibilities for queer play that goes “beyond representation into
the mechanics, aesthetics, interfaces, and development practices of games” [3]. While
games like D&D have recently added invitations for queer players and characters [6],
mainstream TRPGs have struggled to develop meaningful queer structures.
While game scholars have begun exploring independent video games and the “Queer
Avant-Garde” [3], less notice has been paid to indie TRPGs built around queer struc-
tures. Designer Naomi Clark noted in an interview with Ruberg that she didn’t want
her roleplaying card game, Consentacle, to be “about queerness at the level of charac-
ters or narrative. Instead, the queerness comes out in the interactions between people”
[12]. Likewise, Avery Alder said that “games are made queer when they have structural
queerness. Structural queerness is fundamentally about challenging the frameworks of
how stories get told. It’s about subverting systems through queer mechanics and creating
new ways of seeing desire” [4]. This idea of ‘structural queerness’ is fundamental to
TRPG design that reshapes the narrative possibilities of play through queer interactions.
This study examines such structures across three Powered by the Apocalypse TRPGs—
Voidheart Symphony (VS) by Minerva McJanda, Thirsty Sword Lesbians (TSL) by April
Kit Walsh, and Apocalypse Keys (AK) by Jamila R. Nedjadi—to explicate a toolkit for
storygame design that opens possibilities for queer play.
sometimes that is all that is needed. There are metal bolts exposed on the side of the slide,
and the kids have invented their own game with those bolts…” [2]. TRPG mechanics
intended to work one way by the designer(s) can be ignored or reinterpreted by players.
At the same time, play might take place in what Dragon calls the “wide open field”
where “structure is unneeded” [2]. The playground, as a metaphorical lens, is useful for
examining TRPG systems not as singular, fixed mechanics (defined rules with concrete
outcomes), but as structures (acted upon by players) within a flexible system of inter-
pretation. Unlike most videogame narratives, where players only engage the surface
of the system through “abstracted interface” [13, 14] players interact with the struc-
tures of TRPGs directly [4]. In the same way that the physical structures of a playspace
anticipate, shape, and encourage the possibilities of play—so too the structures of a role-
playing game shape emergent narrative possibility. In this sense, my analysis of these
games focuses on 1) how their mechanics comprise larger structures of play, and 2) how
those structures support queer narratives.
In looking at queer TRPG structures, we must note how such structures are distinct
from more superficial elements. Llaura McGee describes a problem in video games
that interpret queerness through an oversimplified “shallow coating,” and notes “the
metaphor is in the mechanics, but the mechanics are so simplified that the metaphor is
meaningless” [15]. Likewise, Alder criticizes games that approach queerness through
characters and narrative but neglect queer mechanics—noting that ‘desire’ mechanics in
her game Monsterhearts stemmed out of a need to “make those mechanics queer” [4]. To
this end, I’m defining game elements that provide superficial signals and representational
flavor (what Anna Anthropy has called a “gay button” [16]) as queer cues [1], separate
from queer structures that shape and anticipate narrative play.
To help distinguish these, here’s an example: many TRPGs provide a space for
pronouns on character sheets. This is a mechanical cue to players that knowing and
respecting the pronouns of other players and characters is important, that they should
not be assumed, and is a “signal that the game world does indeed feature queer people”
[5]. Compare this, however, to Jay Dragon’s game Sleepaway, which gives the following
prompt, “For each camper, choose 2 genders (1 from each column):
• Each game relies on “playbooks” that designate the different roles the characters take
in the game (i.e. “The Beast” [TSL], “The Watcher” [VS], “The Fallen” [AK]).
• Gameplay is oriented around “moves.” Moves pause and shape the action of the fiction
and generally involve dice-rolls and/or choices. Moves are activated by a narrative
‘trigger’ that happens during collective storytelling (i.e. in AK, if a character confides
in another, they will trigger the ‘Reveal Your Heart’ move).
• Play and the fiction are structured as a ‘conversation’ between one MC (“Keeper of the
Doors” in AK, “Gaymaster” in TSL, and “Architect” in VS) and the players: sharing
what their characters (or NPCs and the world) are doing, saying, and thinking.
These games also draw from two influential PbtA titles: Masks: A New Generation
by Brendan Conway and Monsterhearts (MH) by Avery Alder (who was part of devel-
opment for both TSL and AK). Masks is a superhero game focused on teen drama and
cartoon-violence—with a robust system for relationships and emotional defeat. Both VS
and TSL note Masks as an inspiration, and Nedjadi has talked about his experiences run-
ning Masks as motivation for designing AK. Similarly, numerous scholars and designers
point to Alder’s work (MH and MH 2 in particular) as a turning point in the legacy
of both PbtA games and queer TRPGs [1, 7, 19, 21, 22]. MH, a messy drama about
monsters in high school, brought the queer potential of PbtA games to the fore. In this
sense, VS, TSL, and AK are products of a complex ecology within the independent TRPG
scene. While this study accounts for only a sample of PbtA games, I hope it illuminates
Monster Power. Rebel Heart. Gay Sword. 183
some of the ways that independent TRPG designers have developed a shared toolkit for
supporting queer narrative play.
figures—and by infiltrating them and defeating their Vassals, the rebels change the state
of the real world. Rebels must also manage a host of relationships (called Covenants)
represented by tarot cards. Rebel playbooks (The Authority, The Icon, The Heretic) note
characters’ civilian strengths and otherworldly powers.
Currently in development by Jamila R. Nedjadi (they/he), Apocalypse Keys was
originally distributed through itch.io during its beta releases, but is now being produced
by Evil Hat (also the publisher of TSL). This study used official playtest materials for
the game; content quoted here is subject to change. In AK, players are monstrous agents
working for a secretive, occult government branch called the DIVISION. Inspired by
storyworlds such as Hellboy, the players are monsters investigating monsters: stopping
Harbingers of the apocalypse from opening Doom’s Door. Yet each monstrous agent’s
playbook (The Last, The Surge, etc.) foretells a doomed future: as characters embrace
their Powers of Darkness, they risk becoming harbingers themselves.
2.1 Overview
Despite their disparate genres (lesbians, heist-action, mystery), the three sourcebooks
adopt shared structures that encourage queer narrative possibility: 1) safety tools 2)
Conditions 3) personal doom 4) Bonds and 5) drawing from extant media. These struc-
tures, though uniquely framed, all anticipate queer play and interactions. Notably, these
structures are not exclusively queer, but support both queer narrative possibility and
marginalized play across identities. Below, I characterize these structures and examine
how they are presented across each title:
• The X-Card (by John Stavropoulos), which gives players a card (digital or material)
that signals a need to pause the game in response to player discomfort or trauma.
• Script Change Tools (by Beau Jágr Sheldon) that allow players to signal (via cards,
gesture, or chat) a request to rewind, fast-forward, or pause the narrative.
• The CATS sheet (by Tomer Gurantz, adapted from Patrick O’Leary) that helps the
MC describe the Concept, Aim, Tone, and Subject Matter of the game.
• A system for sharing Lines (topics players don’t want to exist in the story) and Veils
(things they don’t want to see ‘onscreen’), developed by Ron Edwards.
[27]. TSL includes re-flavored versions of many of these tools (a check-in card, lines
and veils, and the X-card) and also a discussion about “The Palette” of the game, where
players share what story elements they wish to focus on [24]. AK refers players to the
toolkit, includes a digital CATS sheet, and a Greed/Red/Yellow system for describing
content players want to see and avoid (an adaptation of Lines and Veils) [28]. In every
case, these safety tools are frontloaded as the first and most crucial part of setup.
As community game designer Elizabeth Sampat wrote about her game Deadbolt,
“If Deadbolt is queer, it’s because it’s a safe space to be queer” [29]. Safety tools allow
players to establish boundaries for play and fiction—defining the narrative playground
through shared, consent-based rulebuilding. As Ashwell notes, these mechanics are not
framed as an attempt to limit content, rather “the stress is… that it makes a greater range
of content possible” [22]. Notably, safety tools also serve as a queer cue in that they clarify
audience and signal to marginalized players that safety is valued. TSL even includes a
clear statement “No Fascists or Bigots Allowed” (a strategy developed by iHunt designer
Olivia Hill [30]) along with a detailed list of the game’s expectations for players’ social
awareness and political values. Clark has noted the importance of designing space for
consent in tabletop games [12]. While the power dynamics of TRPGs can fall into toxic
tropes (relying on, say, antagonism between the facilitator and players), these safety tools
are rooted in trust. Rather than asking players to agree to ‘terms’ or relying on implied
consent, these tools are based around shared boundary-setting (through the “palette,”
lines, and veils), active consent (revoked at any time via the X-card or “Pause” tools),
and shared satisfaction.
each Playbook has its own four Conditions, and players can earn Darkness Tokens (a
crucial resource spent on moves) by roleplaying a Condition that affects them. Because
Conditions are unique to each Playbook, they build on that Playbook’s theme: The Last
(a grief-stricken survivor) might become Distant, Merciless, or Despairing. The proud
Fallen, on the other hand, turns Lustful, Raging, Forlorn, or Obsessed.
By choosing not to focus on physical harm, the Conditions system lingers on the
emotional aftermath of conflict. Unlike physical wounds, where characters might sleep
or drink a healing potion to ‘reset’ bodily harm, Conditions can’t be cleared by time.
Instead, they haunt the characters, and can only be resolved through care and support by
companions (at best) or by lashing out (at worst). Rather than framing Conditions as a
consequence (emphasizing the infliction as punishment), the focus is on the character’s
response—an opening of possibility for hurt and healing in the fiction.
Conditions are a queer reimagining of damage and pain. Importantly, these systems
do not demand that players or characters be messy or vulnerable, but instead encour-
age experimentation with what Ruberg has called for in video games: a “rich array of
emotions… that can in fact shape a game’s message as much as (if not more than) its
content and mechanics” [32]. Characters get vicious and fearful—rejecting an expecta-
tion of happiness, victory, and heroism. Conditions drive players to consider context: a
character who reacts to something by becoming Lustful (AK) must now think about how
that desire might shape the fiction. By confronting context through consensual pain [33],
these games invite players to “feel what we aren’t supposed to feel” [32] and engage
with emotional volatility and messy possibility.
In TSL, the structure for balancing personal doom is unique to each playbook. For
example, The Beast playbook has a “Feral” meter and a warning: “You may walk in
civilized circles, but sooner or later your feral truth will come to the fore” [24]. Depending
on the action The Beast takes in the fiction, they will increase or decrease their Feral
score: if their meter reaches zero, they lose access to all their Beast playbook moves.
If their meter reaches four, they Transform into their monstrous self and gain powerful
moves but expose themselves and allies to danger. Other playbooks in the game feature
a similar balancing act. The Chosen has a ‘Destiny’ that they must embrace or reject,
while The Trickster must balance their ‘Feelings’ meter. Each of TSL’s Playbooks feature
self-conflict and a precarious balance between the characters’ want for power/freedom
and cultural expectation. This also emerges in a special move, “Call on a Toxic Power”
which triggers when characters parlay with an evil entity: “You can approach them and
you may even find them helpful at times, but only those with strong Spirit can engage
with them and emerge unscathed” [24].
VS takes a similar approach through its use of the modified tarot. As characters
make choices, they build attunement with two cards: The World (“World is ‘you matter,’
a hand reaching down to pull others back up to their feet.”) and The Void (“Void is ‘I
matter,’ fists raised up in defiance against a hostile universe.”) [27]. Like the other games,
characters balance these two sources of power hand-in-hand: community obligations,
relationships, and the everyday vs. an internal, isolating, devastating power and “taking
your enemy’s strengths for yourself” [27]. When characters go too far and reach three
ranks in either attunement, they trigger one of two moves:
This structure reshapes narrative with possibilities for queer—and especially trans-
gender—tension: between helping your community and protecting yourself, between
hiding from and standing up to the world. As game designer Kara Stone has said, “hard-
ness goes with softness… the decision between whether you are going to heal or destroy.
Sometimes it’s not the right answer to put your healing energy towards something that
doesn’t love you back” [34]. As tensions escalate, players must make choices about
whether their characters are willing to grow closer to ruin, transform into beasts, or
channel dark powers from the Void to survive.
Bonds can also be nurtured when characters trigger the “Reveal Your Heart” move and
can be destroyed by activating some Ruin abilities. Bonds represent opportunities to
put mechanical teamwork into the fiction, but AK clarifies that Bonds aren’t always
positive: rivalries and tense relationships can also grow as Bonds. While Bonds suggests
a two-way relationship, TSL’s Strings (as in ‘heartstrings’) allow players to “Influence
With a String” and tempt a character or alter a roll. Strings give characters subtle (but
incomplete) power over one another, allowing players to boost their own experience
and grow closer in a short scene. In TSL, players also share Strings when a character
becomes Smitten with someone, foregrounding romantic possibility. Finally, VS adopts
this structure through its use of Minor Covenants and Major Covenants. Covenants are
defined by their role in the tarot deck and a brief description that states how a Covenant
will help a character and what help they require themselves. Characters may struggle to
protect and nurture their Covenants, or even betray them—making VS’s social system a
complex cycle of managing character’s needs and the needs of their Covenants.
These games continue the project that Alder has described with MH—which
famously includes a move to ‘turn on’ other monsters [35], prompting players “to con-
tend with what that desire means for their characters” [4]. Like structures for personal
doom, these mechanics support a queer world of social bonds filled with mutual empow-
erment, grief, needs, and desire [36]. Bonds reshape characters’ influences over one
another—opening possibility for queer play amidst rivalry, love, and friendship.
These games draw from established media and in doing so, reimagine the queer potential
of extant genres—building on a long legacy of queer readings and queer remixing of
media [8, 10, 37, 38]. For example, VS pulls extensively from the Persona franchise—
a videogame series about adolescents who can transform from ordinary civilians into
powerful, costumed heroes who wield inner-demons. Yet, as Jordan Youngblood has
pointed out, this franchise is riddled with homophobic and transphobic messaging and
reinforces heteronormative values [39]. VS rebuilds Persona’s ludic premise (and queer
potential), opening narrative possibility for trans-empowerment narratives of transfor-
mation and queer stories about navigating identities under crisis. Similarly, while the
Hellboy franchise (an inspiration for AK) rarely explores queerness directly, the poten-
tial of the themes of ‘dual-monstrosity’ [4] demands queer remix: as AK characters
encounter alienation, transformation, and reject self-narratives. TSL is overtly queer in
its messaging, and its sample adventures reference established queer media (like “Mon-
ster Queers of Castle Gayskull,” which parodies She-Ra and the Princesses of Power).
Yet TSL also references media with queer baggage. One adventure, “Sword Lesbians of
the Three Houses,” reimagines Nintendo’s Fire Emblem: Three Houses—a game popular
in queer shipper fandoms but which included queerbaiting [40]. By opening space for
reinterpretations of established media, these games prompt queer remix and subversive
play against normative genres.
Monster Power. Rebel Heart. Gay Sword. 189
This is loudest in VS: turning to the self (Void) gives characters destructive power that
can only be calmed with community support (World). At the same time, characters who
overcommit to the needs of their Covenants struggle to maintain their edge against the
Vassal. These are struggles of queer energy: between supporting the community around
you and fighting against larger, external systems of oppression.
Reframing violence: Conditions shape how these games frame physical violence.
In combat-focused TRPGs, a character’s ability to physically destroy and dominate
an enemy is paramount. Yet Conditions open possibilities for modes of conflict and
recognize that being defeated is different from being killed. In these games, violence
is only one avenue to defeating a potential enemy; players are prompted to rethink the
impact of their actions on an opponent. TSL especially drives this point: “Bear in mind
that inflicting Conditions is emotionally violent, and sometimes physically violent… that
said, some things are worth fighting for and conflict is often necessary before oppression
and toxic behavior can be halted” [24]. This is reinforced by the way these games
approach power: characters who take the ‘strongest’ moves are also prone to experience
Ruin (AK), struggle against the Void (VS), or expose themselves to danger (TSL). In this
way, they resist optimization and elevate character story over domination.
Building support between players: These games meaningfully underscore support
with other characters and the need to blow off steam. VS’s “Check In” move, AK’s
“Reveal Your Heart” move, and TSL’s “Emotional Support” move are triggered when
characters open up to others, clearing Conditions and gaining Bonds. These games
encourage scenes where characters support one another—driving roleplay that focuses
on teamwork and tension, healing and community, relationships and reciprocity.
References
1. Sihvonen, T., Stenros, J.: Cues for queer play: carving a possibility space for LGBTQ role-
play. In: Harper, T., Adams, M.B., Taylor, N. (eds.) Queerness in Play. PGC, pp. 167–184.
Springer, Cham (2018). https://doi.org/10.1007/978-3-319-90542-6_10
2. Dragon, J.: A dozen fragments on playground theory. Medium, 24 May 2021
3. Ruberg, B.: The Queer Games Avant-Garde: How LGBTQ Game Makers Are Reimagining
the Medium of Video Games. Duke University Press, Durham (2020)
4. Avery, A., Ruberg, B.: Avery alder: queer storytelling and the mechanics of desire. In: The
Queer Games Avant-Garde: How LGBTQ Game Makers Are Reimagining the Medium of
Video Games, pp. 183–191. Duke University Press (2020)
5. Stenros, J., Sihvonen, T.: Out of the dungeons: representations of queer sexuality in RPG
Source books. In: Evan, T. (eds.) Analog Game Studies. vol. 2, pp. 71–92. Carnegie Mellon
University (2017). https://doi.org/10.1184/R1/6686720.v1
6. Stokes, M.: Access to the page: queer and disabled characters in Dungeons and Dragons.
In: Aaron Trammell, Evan Torner, Emma Leigh Waldron, and Shelly Jones Analog Game
Studies, Volume 4, ed., 105–123. Carnegie Mellon University (2020). https://doi.org/10.1184/
R1/11929782.v1
7. Brown, A.M.L., Stenros, J.: Sexuality and the erotic in role-play. Routledge, In Role-Playing
Game Studies (2018)
8. Ruberg, B.: Video Games Have Always Been Queer. NYU Press, New York (2019)
9. Butler, J.: Undoing Gender. Routledge, New York, London (2004)
10. Sedgwick, E.K.: Novel Gazing: Queer Readings in Fiction. Duke University Press, Durham
(1997)
11. Ahmed, S.: Queer Phenomenology: Orientations, Objects. Duke University Press, Others
(2006)
12. Clark, N., Ruberg, B.: Naomi clark: disrupting norms and critiquing systems through “Good,
Nice Sex with a Tentacle Monster.” In: The Queer Games Avant-Garde: How LGBTQ Game
Makers Are Reimagining the Medium of Video Games, pp. 102–112. Duke University Press
(2020)
13. Tan, K., Mitchell, A.: Dramatic narrative logics: integrating drama into storygames with
operational logics. In: Bosser, A.-G., Millard, D.E., Hargood, C. (eds.) ICIDS 2020. LNCS,
vol. 12497, pp. 190–202. Springer, Cham (2020). https://doi.org/10.1007/978-3-030-62516-
0_17
14. Wardrip-Fruin, N.: Expressive Processing: Digital Fictions, Computer Games, and Software
Studies. MIT Press, Cambridge (2009)
15. McGee, L., Ruberg, B.: Llaura McGee: leaving space for messiness, complexity, and chance.
In: The Queer Games Avant-Garde: How LGBTQ Game Makers Are Reimagining the
Medium of Video Games (2020)
16. Anthropy, A.: Rise of the Videogame Zinesters: How Freaks, Normals, Amateurs, Artists,
Dreamers, Drop-outs, Queers, Housewives, and People Like You Are Taking Back an Art
Form. Seven Stories Press, New York (2012)
17. Dragon, J.: Sleepaway. Possum Creek Games (2020)
18. Halberstam, J.: Queer gaming: gaming, hacking, and going turbo. In: Ruberg, B., Shaw, A.
(eds.) Queer Game Studies. University of Minnesota Press, Minnesota (2017)
19. Reed, A.A.: Changeful Tales: Design-Driven Approaches Toward More Expressive Sto-
rygames. University of California, Santa Cruz (2017)
20. Powered by the Apocalypse (2021). http://apocalypse-world.com/pbta/policy. Accessed 8
June 2021
192 PS Berge
21. Torner, E.: Lyric games: geneaology of an online “Physical Games” Scene. DiGRA 2020 –
Proceedings of the 2020 DiGRA International Conference: Play Everywhere: 4 (2020)
22. Ashwell, S.K.: What Is A Storygame? These Heterogenous Tasks (2016)
23. Trammell, A.: Analog Games and the Digital Economy. Analog Game Studies (2019)
24. Walsh, A.K.: Gay Spaceship Games. Thirsty Sword Lesbians. Evil Hat Productions (2021)
25. McJanda, M., Mota, D.S.: Legacy Life Among the Ruins, 2nd edn. Modiphius Entertainment
Limited, UFO Press, Postapocalyptic RPG Hardback (2019)
26. Shaw, K., Lauren, B.-M.: TTRPG Safety Toolkit Guide v2.4 (2021)
27. McJanda, M.: Voidheart Symphony. 1.0. UFO Press (2021)
28. Jamila, N.: Sword Queen Games. Apocalypse Keys. Playtest Edition. Evil Hat Productions
(2021)
29. Sampat, E., Ruberg, B.: Elizabeth Sampat: safe spaces for queerness and games against
suffering. In: The Queer Games Avant-Garde: How LGBTQ Game Makers Are Reimagining
the Medium of Video Games (2020)
30. No fascists. RPG Museum (2021). https://rpgmuseum.fandom.com/wiki/No_fascists.
Accessed 17 June 2021
31. Conway, B.: Masks: A New Generation. Magpie Games (2016)
32. Ruberg, B.: No fun: the queer potential of video games that annoy, anger, disappoint, sadden,
and hurt. QED: J. GLBTQ Worldmaking 2, 108–124 (2015). https://doi.org/10.14321/qed.2.
2.0108. (Michigan State University Press)
33. Brice, M., Ruberg, B.: Mattie brice: radical play through vulnerability. In: The Queer Games
Avant-Garde: How LGBTQ Game Makers Are Reimagining the Medium of Video Games
The Queer Games Avant-Garde: How LGBTQ Game Makers Are Reimagining the Medium
of Video Games, Bo Ruberg. Duke University Press (2020)
34. Stone, K., Ruberg, B.: Softness, strength, and danger in games about mental health and
healing. In: The Queer Games Avant-Garde: How LGBTQ Game Makers Are Reimagining
the Medium of Video Games (2020)
35. Alder, A.: Monsterhearts 2. HT Publisher (2019)
36. Shaw, A.: The trouble with communities. In: Studies, I.Q.G. (ed.) Adrienne Shaw and Bo
Ruberg, pp. 153–162. University of Minnesota Press (2017)
37. Pow, W.: A trans historiography of glitches and errors. Feminist Media Histories 7, 197–230
(2021). https://doi.org/10.1525/fmh.2021.7.1.197. (University of California Press)
38. Macklin, C.: Finding the queerness in games. In: Studies, I.Q.G. (ed.) Bo Ruberg and Adrienne
Shaw, pp. 249–258. University of Minnesota Press (2017)
39. Youngblood, Jordan. “C’mon! Make me a man!”: Persona 4, Digital Bodies, and Queer
Potentiality. Ada: A Journal of Gender, New Media, and Technology (2013). https://doi.org/
10.7264/N3QC01D2
40. Berge, P.S., Britt, R.K.: Dance With Me, Claude: Creators, Catalyzers, and Canonizers in
the Fire Emblem: Three Houses Slash-Ship Fandom. Game Studies (Forthcoming, December
2021)
41. cavegirl. And like, you can’t design safety tools to account for this sort of thing. You just
can’t. Any tools you present will be weaponised by bad actors. (4/6). Tweet. @DyingStylishly
(2021)
Emergent Gameplay, Emergent Essaying
Kirsty Dunlop(B)
1 In Motion
In our post-internet1 landscape of Web 2.0, we have become used to the continual
motion of language and thought performed by us and before us on screen, enacted in
the constant updating of pages and the ability to repeatedly edit posts on social media
with the expectation of instant feedback. We are now intermedial beings, consumers
and readers, with the virtual and the actual no longer easily distinguishable. As Russell
1 ‘The postinternet is kind of to say, we don’t even log on anymore; this is just being. […] The
postinternet is kind of to say, what would still constitute an online experience of the sublime?
Is there a resistant potential in pursuing this, or staying with the sheer sense of the internet’s
dailiness?’ SPAM zine & Press, ‘>What is post-internet?’ [1].
argues in Glitch Feminism [2], it is now more accurate to state you are AFK (away-from-
keyboard), rather than existing IRL (in-real-life) (p. 5). Paralleling this blurring of the
environment, the framework of game design is beginning to spill over into creative and
experimental writing fields. The combination of Emergent Gameplay and experimental
essaying is a hybrid mergence, particularly apt and arguably vital to the rapidly shifting
rhythms of our everyday digital lives in the contemporary moment.
Emergence invokes the shifting motion of an ongoing creative process; there is a
tension and multiplicity in the word emergence, between quick action and slow unrav-
elling, of what is sudden and urgent (an “emergency”2 ) and what is unfurling. Whilst
this undercurrent is felt in the process of writing page-based forms, we are assured that
the writing is complete through our interaction with a finite physical manuscript. A
digital-born work, on the other hand, carries emergence, at the forefront of its thinking
and creation as well as its interaction with the reader, with both text and reader perfor-
matively enacting a sense of ongoing-ness. Here, we must embrace the difficulties and
pacing of the continual evolution emanating from the screen, whether we feel able to
explore a digital environment freely or are entrapped in a narrative.
Emergent Gameplay is a recognized aspect of game design where interactivity and
narrative come together: generally understood as ‘a game design term that refers to video
game mechanics that change according to the player’s actions […] Emergent gameplay
can also be created by adding multiple players to the same game environment and having
their individual actions impact the overall game narrative.’ (Techopedia) [4]. Horowitz
and Loony [5] add to this definition that ‘the term refers to complex outcomes that can
result from the interaction of simple rules […] There are two types of emergence com-
monly referred to by scholars, intentional and unintentional’ (p. 11). The combination of
simple actions and complex outcomes, intentionality and unintentionality, emphasizes
the nuance and openness of Emergent Gameplay’s narrative potential for communicating
complex affective responses and plot tension, as well as blurring the boundaries between
creator and player.
In writing an interactive work or game, the writer(s) must be hyper-aware of the
effects of this Emergent Gameplay, both on the over-arching structure and with regard
to the key micro moments. The writer must shift between experiencing the game as
player and creator, in order to understand how decision-making will impact the expe-
rience of playing the game, thinking beyond a single frame of text or play. Emergent
Gameplay is rich with possibilities. The uncertainty this multiplicity and reactiveness
embeds becomes intricately connected to the notion of the glitch and the verb glitching
(considered further below). Its complexity and potential is also related to notions of
essaying, hybridity, and experimentation on the page, and adds a specific kind of inter-
active quality and effect, which shall be seen within the hybrid framework of Emergent
Essaying.
2 ‘And the state of emergency is also always a state of emergence.’ writes Bhabha in The Location
of Culture (p. 59) [3].
Emergent Gameplay, Emergent Essaying 195
3 ‘Big Tech’ refers to the hegemony major technical companies have over society. For more on
the links between this control and the attention economy in our era, see Tanner in The Circle
of the Snake: Nostalgia and Utopia in the Age of Big Tech [7].
196 K. Dunlop
versions), and that it is this departure from the familiar that can act as a catalyst for
joy, potential, and unease, to engender a myriad of complex affects that revel in sur-
prise. I look towards Stewart’s [11] theoretical work Ordinary Affects to conceptualize
these responses: ‘Ordinary affects are the varied, surging capacities to affect and be
affected that give everyday life the quality of a continual motion of relations, scenes,
contingencies, and emergence.’ (p. 2).
How then does this subversion and consequent generation of affective responses
relate to Emergent Essaying, connecting back to ideas of wandering and losing one’s
way? Glitch must first be understood as it stands within these many fields and contexts,
as an active digital-born mode that creates entries into new beginnings or alternative con-
tinuations. The verb ‘glitching’ here also brings another mode into play: game scholars,
such as Meades and Consalvo, theorize glitching as a collective act of players unlocking
faults within a game to carve their own playing experience. Meades [12] promotes the
ideas of thinking through glitching as more complex than a negative interference whilst
Consalvo [13] connects glitching more overtly to acts of cheating, promoting the idea
that it is not a wholly negative intervention, but rather an important collaboration between
player, game, and writer, offered up by the specific emergent techniques enabled by the
technologies of the form. Glitching functions as an interventional act of emergence,
handing over control of the narrative structure to the reader and the potentialities of the
text itself. Page-based text can be open to varied interpretations and viewpoints, but in
the digital environment this occurs performatively, with the text and game never wholly
stable. The underwiring of the code is revealed, reflecting the fragility of binary (closed)
thought as further collaborative choices are opened within the context of the work.
Additional tensions also exist within the effects and purposes of glitching. After
a recent talk, I was asked whether the act of speed-running in a video game can be
considered a radical emergent glitching within the framework I am proposing. Speed-
running as framed by Scully-Blaker [14] is indeed an Emergent Gameplay practice:
‘the process of completing a game as quickly as possible without the use of cheats or
cheat devices’. Whilst this act of racing through a game is a radical intervention by the
player, it does not fall into the expansive wandering that I am positing with Emergent
Essaying. If speed-running exists as ‘post(human) performance art’ as posited by Hay
[15], in which speculation, competitiveness, and instant gratification are at the forefront,
then slow meandering through choice, and a re-framing of glitching with essaying as
a means to travel in different directions, inserts the human mind with all its errors and
capabilities of discovery back into the digital sphere.
The mode of glitching I envision within Emergent Essaying is subverting the opti-
mization rhythms of quick movement and success, so often geared towards solutionism
rather than the chaotic and difficult slower presence of our thoughts, which are capable
of curiosity, change, and alternative directions. This builds upon the definition of exper-
imental essaying posited by Lopate, ‘The essay is a notoriously flexible and adaptable
form. It possesses the freedom to move anywhere, in all directions.’ (p. xxxvii) [16]. This
takes on new relevance in Emergent Essaying, where digital writing tools such as Twine
turn readers into players, literally wandering between ideas. By embracing performative
glitches along with the potential for genuine errors, we can enact a playful spatiality
to wander and become lost in language. This thinking is in line with Robertson’s [17]
Emergent Gameplay, Emergent Essaying 197
words from ‘Time in the Codex’: ‘It is the most commodious sensation I can imagine,
this being lost.’ (p. 13). Commodious implies a comfortable space in which to roam
freely and explore. What roominess might gameplay offer us?
With that question in mind, there have been an increasing number of commercial
games that have reveled in a slower, more observant explorational narrative movement.
One example is the action-adventure game Subnautica [18], an open world survival
game that encourages the slow exploration of the ocean of an alien planet.
Alternative commodious modes of play can interact with the commodious sensation
of thinking in relation to reading, writing, and time, which is the experience and feeling
Robertson is describing. A rejection of the flattened idea of quick success and point
scoring as the primary mode of glitching makes room for more of the wonder that can be
found in the space of non-linear thinking and variables, discussed further below. Here,
in this liminality of the hybrid, is the possibility for a more exuberant “slippery space”.
3 To Try, To Play
Think of essay as a verb, as a becoming. To essay is to try within an experimental,
creative context. I parallel this with Carson’s statement [19]: ‘Consider incompleteness
as a verb’ (p. 29), from the hybrid collection Plainwater: Essays and Poetry. Emergent
Essaying seeks to expand this notion of ongoing-ness into the sphere of digital-born
play. What does it mean to play into an idea, to write into the glitch?
Hybridizing essaying with gameplay feels ever more important in our increasingly
digitalized world, where concepts of digital play can easily slide into a framework of
optimization. Instead, we can re-conceptualize play in line with emergence as a more
nuanced unravelling of thought and creation: a nexus of multiple ideas and genres.
Play invokes desire. Experimental essayist Blau DuPlessis writes: ‘The essay is restless
[…] always a little too thirsty.’ (p. 38) [20]; the merging of gameplay and playfulness
in language can enhance this sensation of continuous reaching, with the addition of
obstacles supplanting neat, conclusive arguments.
Emergent Essaying intertwines the acts of trying and playing. Here, the thinking
is innately interdisciplinary, combining elements of interactive fiction, gameplay tech-
niques, and hybrid essaying. Academic digital theory has traditionally focused on hyper-
text—the joining of fragmented pieces of text, through links, a form which Coover [21]
argues ‘offers the patient reader […] just such an experience of losing oneself to a
text’. However, digital writers and researchers, such as Montfort [22] have been increas-
ingly promoting the in-depth study and value of interactive fiction. Emergent Essaying
exists at the intersection of literary hypertext, the gameplay of interactive fiction, and
hybrid essaying, with the multi-potential layering of these seemingly disparate forms
and techniques.
A multi-media writer whose work embraces an essayistic approach, is Carpenter,
whose electronic writing/art often concerns the collision of historical documents with
the contemporary digital framework. In The Pleasure of the Coast [23], the user accesses
the work through a non-linear format, scrolling horizontally rather than vertically, to
enact an exploration of landscape. The work recalls the aesthetics of chapbooks, albeit
with the sense of space afforded by the digital realm. Pencil line drawings are joined
with kinetic text, playfully layering historical found text and fictional narration.
198 K. Dunlop
Carpenter uses the layering of aesthetics and user experience, forming a palimpsest
of image and text alongside the scrolling mechanic, to create an alternative mapping
effect that is imperfect and in motion, encouraging the user/reader to slow down and
examine how they engage with the work. The simplicity of the infinite coast, and the
performative glitch effect of forever scrolling sideways, asserts a lack of closure.
Carpenter also encourages a mode of error-making in writing, stating that ‘imper-
fections are deliberate’, including errors of translation. Here, the digital techniques and
artistic practice inform one another, leading to new slants in seeing and comprehending,
in the real-time pace of the on-screen reading experience.
When examining Carpenter’s expansion of essaying, apt connections can be found
in the Web 1.0 hypertext works of Jackson, whose landmark text Patchwork Girl [24],
brings together Mary Shelley’s Frankenstein [25], with contemporary understandings
of the post-human as posited in Haraway’s A Manifesto for Cyborgs [26], resulting in
a sprawling, monstrous conception of the female body through the rhizomatic form of
the hypertext reading structure. The movement of following a hyperlink to another page
of text induces an awareness of multi-layering: what is not accessed immediately is still
knowingly an integral part of the narrative.
Modes of critique and essaying which borrow from hypertextual rhythms are crop-
ping up in page-based creative-critical writing such as ‘hypercriticism’, highlighted by
Manifold Press as ‘an exploitation of hypertextual possibilities’ [27]. Emergent Essaying
adds layers of performance and interactivity, with the reader/player existing as a vital
part of the writing, performing this more ‘blissful’ mode of reading, that is ‘writerly,’
as posited by Barthes [28]. Emergent Essaying engages with gameplay techniques, such
as non-linearity, variables (collecting objects or scores that affect later parts of the nar-
rative), external data, randomization (diverging points emerging through chance for
different players), and intentional glitches alongside the potential for genuine malfunc-
tions, to influence how readers/players understand and engage with the thinking and
poetic effects of the work. The tension created by Ludonarrative Dissonance (described
by Hocking as ‘a powerful dissonance between what it is about as a game, and what
it is about as a story’, in relation to the game Bioshock [29]) must be kept in mind
during the writing and playing of Emergent Essaying. Techniques should not simply be
implemented because they are available in the digital realm but because they create new
experiences of understanding through play that are linked to the themes and connections
the writer is exploring. This ethos must be embraced when working with game writing
tools such as Twine, in which the author must also inhabit the role of player, creating an
environment in which gameplay influences essaying and vice versa.
Game designer Sampat subverts the idea that games are simply a mode of eliciting
empathy, instead arguing that they can function as empathetic machines, if empathy
is directly built into their structure by the designer(s) [30]. Here, not only content but
also structural systematic understanding can be subverted, demonstrated in Am I Part of
the Problem? [31], in which the player must address their own biases and approaches,
offering direct critical insight, tailored to the individual, as they answer questions in
order to understand their role in a conflict situation. The game recognizes that it holds
no definitive solutions. Instead, through play, it offers a look into the self that does not
feel prescriptive or reductive in its assumptions about the player. In combining elements
Emergent Gameplay, Emergent Essaying 199
from the examples above, Emergent Essaying opens up a new cross-form conversation,
reconfiguring and expanding the expectations of the essay in a multimedia landscape by
inviting the term ‘essay’ to co-mingle performatively with ‘play’.
Within this context, play must be understood in its many facets, beyond gameplay;
Huizinga [32] regards the function of play as just as important as work in society, a “free”
mode; the motivation of play being the experience it offers rather than the concrete
goal. We can connect this to Halberstam’s [33] conceptualization of low theory, as
a mode of play and child-like pleasure in process, crafted through error. The digital
gameplay of Emergent Essaying promotes a more playful approach to understanding,
with language play intertwining with digital interactive gameplay, each informing and
driving the other to capture new ways of seeing, doing, asking, and inviting. This layering
of play encourages us to think more receptively, questioning our knowledge and opening
us up to new experiences and information. Through play, we move beyond the individual
to a collective and collaborative space of thought. Thinking merges with play, emerges
from within it, sometimes consciously, often unconsciously.
If our sense of time is thwarted by the short attention spans and immediate responses
of Web 2.0, Emergent Essaying has the capability to expand how we experience and
consider time in this virtual world. Writers of these experimental digital forms can
continually re-frame their own considerations of narrative time, through the reflective,
slow thinking required to create complex linkages and gameplay techniques. Returning
to this idea of wandering, digital-born works have the capability to deliberately slow
down a player/reader’s mind through Emergent techniques, as well as providing them
with a sense of freedom to individually navigate the text.
The ‘glitch’ is an integral aspect of time in the practice of writing and experiencing
Emergent Essaying, as it exists as a moment of stasis and potential. I often think of the
glitch in relation to Derrida’s notion of ‘hauntology’ [36], a reminder of the work’s past
state and flux, its ghostly underpinnings, the mark the writer has left on the machine and
that the reader has yet to leave. There is also the sense of temporality and fragility, in
the awareness that the essay can ‘break’ or be meddled with by the reader, which opens
a more ambivalent mode of thinking.
The term Emergent Essaying creates immediate connections between what are often
thought of as disparate fields: bringing together elements of game design and experimen-
tal essaying. It challenges the ideas around what a game entails and an essay involves,
centering itself more around narrative exploration of ideas than the concrete pursuit of
goals of the former or the neat conclusive arguments of the latter. If essaying is a mode of
trying, Emergent Essaying inserts a more playful practice, understanding and enacting
of questioning and intervention, through a continually shifting dynamic between player,
writer, and game.
In order to demonstrate the practice of Emergent Essaying, and give the reader
an experience of one iteration it might take, a digital-born, alternative and exper-
imental version of this paper, written in Twine, can be accessed here: http://kirsty
dunlop.com/EmergentGameplayEmergentEssaying/digital.
References
1. SPAM zine & Press, “>What is post-internet?” https://www.spamzine.co.uk/what-is-post-int
ernet. Accessed 09 July 2021
2. Russell, L.: Glitch Feminism: A Manifesto. Verso Books, London (2020)
3. Bhabha, H.K.: The Location of Culture. Routledge, London (1994)
4. Techopedia def. Emergent Gameplay. https://www.techopedia.com/definition/27043/eme
rgent-gameplay. Accessed 09 July 2021
5. Horowitz, S., Looney, S.: The Essential Guide to Game Audio: The Theory and Practice of
Sound for Games. Focal Press, Burlington (2014)
6. Solnit, R.: A Field Guide to Getting Lost. Viking Penguin, New York (2005)
7. Tanner, G.: The Circle of the Snake: Nostalgia and Utopia in the Age of Big Tech. Zero Books,
London (2020)
8. Merriam-Webster def. Glitch. https://www.merriam-webster.com/word-of-the-day/glitch-
2019-05-15. Accessed 09 July 2021
9. Menkman, R.: The Glitch Moment(um). Institute of Network Cultures, Amsterdam (2011).
https://networkcultures.org/_uploads/NN%234_RosaMenkman.pdf. Accessed 09 July 2021
10. Myers, R.: glitcherature (2014). https://robmyers.org/glitcherature/. Accessed 09 July 2021
Emergent Gameplay, Emergent Essaying 201
11. Stewart, K.: Ordinary Affects. Duke University Press, Durham & London (2007)
12. Meades, A.: Why we glitch: process, meaning and pleasure in the discovery and documenta-
tion, sharing and use of videogame exploits. In: Well Played: a Journal on Video Games, Value
and Meaning, pp. 79–98, Carnegie Mellon University: ETC Press, Pittsburgh, PA (2013)
13. Consalvo, M.: Cheating: Gaining Advantage in Video Games. MIT Press, Cambridge (2009)
14. Scully-Blaker, R.: A practiced practice: speedrunning through space with de certeau and
virilio. Game Stud. Int. J. Comput. Game Res. 14(1) (2014). ISSN: 1604–7982. http://gam
estudies.org/1401/articles/scullyblaker. Accessed 09 July 2021
15. Hay, J.: Fully optimized: the (Post) human art of speedrunning. J. Posthuman Stud. 4(1),
5–24. The Pennsylvania State University, University Park, PA (2020). https://www.jstor.org/
stable/10.5325/jpoststud.4.1.0005
16. Lopate, P.: The Art of the Personal Essay. First Anchor Books, USA (1995)
17. Robertson, L.: Time in the Codex. In: Nilling: Prose Essays on Noise, Pornography, The
Codex, Melancholy, Lucretius, Folds, Cities and Related Aporias, pp. 9–18, 2nd edn.
Book*hug Press, Toronto (2012)
18. Cleveland, C.: Subnautica, 16 December 2014. https://store.steampowered.com/app/264710/
Subnautica/. Accessed 09 July 2021
19. Carson, A.: Plainwater: Essays and Poetry, Reprint Vintage Books, New York (2000)
20. DuPlessis, R.B.: Blue Studios. The University of Alabama Press, Alabama (2006)
21. Coover, R.: Praise for Patchwork Girl by Shelley Jackson (Eastgate) (1995). https://www.eas
tgate.com/catalog/PatchworkGirl.html. Accessed 7 July 2021
22. Montfort, N.: Twisty Little Passages: An Approach to Interactive Fiction. MIT Press,
Cambridge (2005)
23. Carpenter, J.R..: The Pleasure of the Coast (2019). http://luckysoap.com/pleasurecoast/en/
index.html. Accessed 09 July 2021
24. Jackson, S.: Patchwork Girl. Eastgate, Watertown, MA (1995). CD-ROM
25. Shelley, M.: Frankenstein, New Alma Classics, London (2014)
26. Haraway, D.J..: A Manifesto for Cyborgs: Science, Technology and Socialist Feminism in the
1980s (1985). https://monoskop.org/images/4/4c/Haraway_Donna_1985_A_Manifesto_for_
Cyborgs_Science_Technology_and_Socialist_Feminism_in_the_1980s.pdf. Accessed 09
July 2021
27. Manifold Press Submissions. https://www.manifoldcriticism.com/submissions. Accessed 09
July 2021
28. Barthes, R.: The Pleasure of the Text. Trans. Richard Miller. Reissue edn. Farrar, Straus &
Giroux Inc, New York (1975)
29. Hocking, C.: Ludonarrative dissonance in Bioshock. In: Click Nothing: Design from a Long
Time Ago, 07 October 2007. https://clicknothing.typepad.com/click_nothing/2007/10/ludona
rrative-d.html. Accessed 09 July 2021
30. Sampat, E.: Empathy Engines: Design Games that are Personal. Political and Profound.
CreateSpace Independent Publishing Platform, Scotts Valley, California (2017)
31. Sampat, E.: Am I Part of the Problem? (2017). https://elizabethsampat.itch.io/am-i-part-of-
the-problem. Accessed 09 July 2021
32. Huizinga, J.: Homo Ludens: A Study of the Play-Element in Culture. Maurice Temple Smith
Ltd, London (1970)
33. Halberstam, J.: The Queer Art of Failure. Duke University Press, Durham, North Carolina
(2011)
34. Jayemanne, D.: Chronotypology: a comparative method for analysing game time. In: Games
and Culture, 15(7) (2019, 2020). https://journals.sagepub.com/doi/10.1177/155541201984
5593. Accessed 09 July 2021
202 K. Dunlop
35. Quinn, Z., Patrick, L., Isaac, S.: Depression Quest, 14 February 2013. http://www.depressio
nquest.com/. Accessed 09 July 2021
36. Derrida, J.: Specters of Marx: the State of the Debt, the Work of Mourning and the New
International. Routledge, New York (1994)
Tale: Defamiliarizing Ludonarrative
Puzzles
1 Introduction
Tale1 is a 2.5D puzzle-platformer that seeks to explore and demonstrate ludonar-
rative design through an approach based on Mechanics as Metaphor and defa-
miliarization. Developed in Unity 3D Tale follows the relationship between two
nameless protagonists through a ruined fantasy world across 3 levels and 9 puz-
zles. The puzzles include a variety of platformer mechanics designed to explore
the emotional state of the protagonists and the nature of their relationship.
There is no text or dialogue, the play alone delivers a tale of communication and
experimentation, loneliness and anxiety, and collaboration and growth.
Ludonarrative narrative design seeks a blend of play and storytelling [1] where
by the mechanics of the game themselves may reinforce, and even tell, the story
through play. A common approach to this is the notion of mechanics as metaphor
where by the form and interaction of a gameplay mechanic becomes a metaphor
within the narrative. The earliest mention of this concept was likely Jason Rohrer
in a personal essay on his work on Passage [6] where the progression from left
to right and changing visuals are a metaphor for the passage of life, though
numerous other designers and scholars have discussed the concept. This can be
effectively paired with defamiliarization within design, a notion with its origins
in formalism [8] but more recently explored within ludonarrative [5]. A design
may be defamiliarized by subverting player expectations, breaking the games
1
Tale can be played here: https://ninofrazzitta.itch.io/tale as of 08/10/2021.
c Springer Nature Switzerland AG 2021
A. Mitchell and M. Vosmeer (Eds.): ICIDS 2021, LNCS 13138, pp. 203–207, 2021.
https://doi.org/10.1007/978-3-030-92300-6_18
204 A. Frazzitta and C. Hargood
own established rules, or surprising the player with new rules and systems. This
draws the players attention and can serve to highlight an element of ludonarra-
tive design, a mechanical metaphor, or even serve as a metaphor itself. While
ludonarrative is a well established concept in our field, particularly in analysis,
in this demonstration we provide an example of its use in practical design and
our own approach to its use in movement, puzzles, and defamiliarization.
2 Tale
Tale narrates a small adventure of a young boy entering a mysterious world
where he finds and frees a young girl belonging to an ancient civilization. The
story revolves around the theme of companionship, and the player will explore
three fundamental stages of this relationship as they progress with the game.
Firstly the player starts with the discovery of this new relationship focusing
on the sense of communication and experimenting with new things. The sec-
ond part has the two characters separated by an uncontrollable force, exploring
loneliness and oppressive anxiety. And finally, the characters are reunited and
we explore developing collaboration and growth. Through the duration of the
game the player as “the boy” will work with “the girl” as they take their journey
through the ancient ruins. The interactions between the two underline that they
need each other in order to progress on their journey. This is achieved by having
the boy pushing boxes and interacting with leavers, etc. while the girl uses magic
powers such as teleporting to progress.
The two characters come from different worlds and have different languages
so they cannot communicate properly with each other. The story explores their
struggles communicating and how they grow as time passes, showing how their
bond becomes stronger. While superficially the story may at first seem to have
a “save the princess” premise, this is quickly dispelled with a shift in focus to
the collaboration of the two characters. The story finishes the girl completing
her journey and taking her place at the ruins summit.
Our design makes use of Jesse Schell’s lenses of unification [7] helping the design
to bring together a cohesive experience that supports a dominant emotion of col-
laboration. There are three levels in the game, the forest, the underground dun-
geon, and the sky temple forming a three-act arc in the game, each is designed
to support the core emotional themes (communication and experimentation, fol-
lowed by loneliness and anxiety, followed by collaboration and growth). The level
design of those areas is meant to build the right context mirroring the character
state of mind - following Zammitto’s colour principles [9] that associate colours
to certain emotions, and Bura’s principles of emotion engineering [2] modifying
gameplay using framing devices such as music and visuals to build the appropri-
ate context allowing us to elicit the right emotions in time. In addition to that,
we have designed the player’s elevation based on the Lopez blockbuster intensity
graph [4] allowing us to further regulate the game emotional intensity (Fig. 1).
Tale: Defamiliarizing Ludonarrative Puzzles 205
The game presents classic platformer movements and interactions. The player
controls “the boy” with standard controls while “the girl” moves either in
response to leading and teleportation mechanics, or later moves in response to
puzzle context. The game is divided into small areas that contain environmental
puzzles. To complete a puzzle both characters must reach the end (which most
of the time is signified by a stone portal). For the basic mechanics, the game
has “classic” interactions such as pushing boxes, climbing ladders, pulling levers,
navigating lifts and moving platforms, and collecting keys. The core mechanic
used is the teleportation of “the girl” who can be positioned on platforms or
switches to complete the puzzle using initially static teleportation stones, and
later moveable teleporter orbs (as depicted in Fig. 2. Puzzles are solved through
careful positioning of both characters in order to collect keys and reach the exit,
While the player has fine control over the position of one character movement of
the other is more challenging and limited by teleportation and leading mechanics,
but becomes more flexible as the game progresses.
Fig. 2. Movement is key to Tale and teleportation stones and orbs are part of that
206 A. Frazzitta and C. Hargood
Fig. 3. Movement mechanics: leading, climbing, and bringing the portal together.
References
1. Aarseth, E.: A narrative theory of games. In: Proceedings of the International Con-
ference on the Foundations of Digital Games, FDG 2012, pp. 129–133. Associa-
tion for Computing Machinery, New York (2012). https://doi.org/10.1145/2282338.
2282365
2. Bura, S.: Emotion engineering in videogames: toward a scientific approach to under-
standing the appeal of videogames (2008). http://www.stephanebura.com/emotion/
3. Campbell, J.: The Hero with a Thousand Faces, vol. 17. New World Library (2008)
4. Lopez, M.: Gameplay fundamentals revisited: harnessed pacing & intensity. Gama-
sutra 12, 2008 (2008)
5. Mitchell, A., Kway, L., Neo, T., Sim, Y.T.: A preliminary categorization of tech-
niques for creating poetic gameplay. Game stud. 20(2) (2020)
6. Rohrer, J.: What i was trying to do with passage. Jason Rohrer’s webpage, Novem-
ber 2007. http://hcsoftware.sourceforge.net/passage/statement.html. Accessed 20
July 2021
7. Schell, J.: The Art of Game Design: A Book of Lenses. CRC Press, Boca Raton
(2008)
8. Shklovsky, V.: Art as Technique. University of Nebraska Press, Lincoln (1965)
9. Zammitto, V.L.: The expressions of colours. In: DiGRA Proceedings of the 2005
DiGRA International Conference: Changing Views: Worlds in Play (2005)
Emergent Narrative and Reparative Play
1 Introduction
Recent work on ludonarrative hermeneutics [10,12] has attempted to address
the question of how players make sense of narrative meaning in interactive digi-
tal narrative (IDN) systems, including narrative games. To date, this work has
largely focused on the analysis of games in which a strong protostory is deliber-
ately embedded by the designers: in other words, games that attempt to commu-
nicate certain preauthored narrative events to the player on every playthrough,
regardless of variations that might arise from one playthrough to the next. In
attempting to apply a hermeneutic approach to the analysis of narrative sand-
box games such as The Sims, where no particular narrative events are fixed
in place by the game’s creators, we are left with an open question: how do
players go about interpreting narrative meaning when the units of narrativity
embedded in a game by its designers are much smaller, much more abstract, and
much more freely recombined than in the archetypal cases of heavily protostory-
laden narrative games? To address this question, we propose that the narrative
C. Manning—Independent.
c Springer Nature Switzerland AG 2021
A. Mitchell and M. Vosmeer (Eds.): ICIDS 2021, LNCS 13138, pp. 208–216, 2021.
https://doi.org/10.1007/978-3-030-92300-6_19
Emergent Narrative and Reparative Play 209
2 Background
Key to the argument of how narrative sandbox games invite reparative reading,
and from there reparative play, is the idea that a work of interactive digital
210 J. Grinblat et al.
narrative that’s spun out of the upper circle. Of course the object will contain
ludonarrative devices that assist this process and anticipate aspects of the emerg-
ing narratives (things like mechanics, tone-setting art styles and sound design,
etc.). In this way we might say narrative sandbox games’ story volumes are open;
they sketch their boundaries to suggest shape but are less concerned with strict
in/out delineation. Sedgwick positions reparative as a weak theory in contrast
to the strong, totalizing impulse of the paranoid. (This is not disparaging; Sedg-
wick points to reparative reading’s acceptance of its limitations as a strength of
the theory.) Here we see the weakness in action; because the emerging narrative
does not benefit from an author’s prefiguration, because it may wander into a
thorny corner of its open story volume, the scope of its coherence may be local
instead of global, but it has the potential to be nourishing nonetheless.
It’s tempting to think of the narratives in narrative discovery games as also
arriving in disrepair. But the tools of repair, in the Sedgwickian formulation,
are furnished from “one’s own resources” and cannot be shipped with the game
object as part of its suite of mechanics. It’s more accurate to say these narratives
arrive intact but buried (think of the digging metaphor we used earlier) and that
you’re given tools of revelation: a compass to find the relevant sites and a hammer
and chisel to excavate them. How intact these buried narratives are can vary,
and this gives a bit of reparative flex to these broadly paranoid systems. Her
Story famously doesn’t communicate when you’ve satisfactorily unearthed what
it has hidden [16]. Return of the Obra Dinn, on the other hand, explicitly reifies
the revelation of its truths; it’s difficult (but humorous) to imagine the Obra
Dinn player who correctly matches three crewmates to their fates and triggers
the validation sequence but who persists in their doubt of the results.
We’ve formulated reparative reading in the context of IDN, but the repar-
ative practice is coextensive with the move from the hermeneutic circle to the
hermeneutic strip. The reparative work of interpretation happening in the lower
circle flows back to the upper; the repairer moves from interpreting events that
are occurring to reifying their interpretation as game actions that trigger the
next iteration of events. Reparative reading becomes reparative play; interpreta-
tion begets actions, and the cycle is repeated. Here the repairer acts as co-author
to the narrative object, using game affordances1 to mine narrative material [14]
and sculpting this material into a narrative work.
may go into aspiration failure, but that failure then becomes another part-object
to be assembled.
Because The Sims doesn’t prefigure its narratives, narrative must cohere
as the player moves through the hermeneutic strip, taking actions, witnessing
their effects, and interpreting the results. The player must interpret why a spe-
cific action is happening at any given moment. A spontaneous flirt might be
interpreted as a result of a sim’s promotion putting them in a good mood and
boosting their self-esteem, or it might be seen as a sign of true love. The cho-
sen reading then offers shape to the player’s next choices: if the autonomous
flirt is interpreted as a sign of the sim’s interest, then the player may choose to
send their sim on a date with the NPC of interest. The game actions taken as
a result of this interpretation then engender their own consequences that are in
turn subject to interpretation, and the cycle continues. These narrativizations
are influenced by a player’s “own resources”, the stories they’re most interested
in telling through this sort of imaginative play. The process by which this hap-
pens is one of extrapolative narrativization [7]: in making narrative sense of Sims
play, players do not simply transcribe the series of game events as it unfolds.
Instead, they confer additional layers of interpretation on these events, adding
extra details to the narrative-as-perceived—which then influences what actions
the player is inclined to take next.
Sims’ speech bubbles are a common intervention point for this extrapolative
narrativization. Player-authors of Sims retellings [3] often attempt to attribute
a meaning to the game’s abstract dialogue icons in order to shore up an inter-
pretation. In roBurky’s Alice and Kev [11]—a notable Sims 3 retelling centered
on a homeless father and daughter—the author looks at Alice’s first real adult
conversation with an NPC as a site of potential meaning, suggesting that a lake
might represent her sleeping rough in parks and a Yeti figure might be her ogre
of a father. Because Alice’s life in the story has been so shaped by roBurky’s
roleplaying of her as homeless, the author confers meaning on this conversation
by attributing referents to the otherwise ambiguous dialogue icons that would
cast the interaction as a meaningful opening-up. Another conversation between
two different sims involving a lake and a Yeti might suggest an entirely different
reading, such as a camping trip gone wrong. The same speech bubbles inter-
preted differently might lead players to take two very separate sets of actions; in
roBurky’s reading, a player might be moved to deepen the relationship with the
NPC, where in the camping trip scenario, a player might decide to send their
sim on another, hopefully more successful camping trip. Extrapolative narra-
tivization is the key by which players become co-authors, as their interpretive
frameworks overlay the game’s mechanics, guide them to actions that fulfill the
narratives suggested by their frameworks, and ultimately allow them to assemble
a cohesive, satisfying narrative through reparative play.
Even The Sims’s pre-structured scenarios, which could be considered pre-
assembled narrative pieces for discovery, are malleable to players’ intentions. The
Sims 2 shipped with several scripted events in its base neighborhoods; events
were set to trigger, but players could ignore those scenarios and instead focus on
214 J. Grinblat et al.
playing their own created characters. If the households with queued narrative
were opened after players had already changed the world state, the designer-
prefigured events would sometimes not be able to trigger, or their conditions
would cause them to play out differently as a result of the player’s previous
decisions elsewhere in the neighborhood. Players who wanted Cassandra Goth’s
wedding to Don Lothario to succeed rather than fail, as it was scripted to do
under default conditions, could go to his house and invite her over, raising his
relationship with her past the threshold at which he would not leave her at the
altar. That completed, they could reopen the Goth household to the wedding
scenario, still set up exactly as scripted, and complete the wedding as they
desired it to go. Mary-Sue Pleasant was scripted to always fail the chance card
that came up when her household was loaded, regardless of the action the player
took, causing her to come home and potentially catch her husband cheating on
her with their maid. Players discovered that ignoring the chance card entirely
avoided the trap; others simply had her husband send the maid home after she
was finished cleaning. When a player had decided how they wanted the scripted
event to play out, regardless of the game’s structure, their next actions could
subvert the game’s suggestions and instead offer a reparative reading in which the
player’s own preferred meaning would take precedence over the story suggested
by the game.
This sort of reassembly could not happen without the part-objects that Sedg-
wick discusses and which characterize narrative sandbox games. All Sims nar-
ratives arrive in disrepair, made up of small actions with specific game verbs
that have specific effects on the world state. These are most often very limited
in scope, affecting one sim or one household at a time; it is only through the
assemblage of many of these actions that major changes, like marriages and pro-
motions, occur. Game verbs like “study a skill” are part-objects in the larger
narrative of “Cassandra Goth got promoted”; without them, the latter cannot
occur mechanically, but also cannot be meaningful narratively. Any verb can
be a part-object, depending on what is meaningful to the player; that is what
makes The Sims such a clear instance of an open story volume. Reparative play,
and the hermeneutic strip, enable one to cut through a dizzying plethora of
possible meanings. As the player reifies an ever-evolving interpretation of events
and takes further actions that stem from that emerging interpretation, a dis-
parate mass of narrative parts, player-provided resources, and player-conferred
meanings coalesce into a satisfying narrative: “something like a whole”.
4 Conclusion
idea of extrapolative narrativization, through which the player brings their own
resources to bear on what the game offers. By interpreting game events, which
then spur further actions and interpretations, players engage in the hermeneutic
strip of meaning-making—but rather than narrowing in on a canonical, designer-
intended narrative meaning, the player instead constructs an assemblage of nar-
rative part-objects whose meaning is derived partly from the resources that this
player in particular has brought to bear. As a counterpart to a ludonarrative
hermeneutics of suspicion, Sedgwick’s framework thus points the way to a paral-
lel ludonarrative hermeneutics of repair, and from there, toward an investigation
into broader paradigms of interpretive orientation.
References
1. Aarseth, E.J.: Cybertext: Perspectives on Ergodic Literature. Johns Hopkins Uni-
versity Press, Baltimore (1997)
2. Eco, U.: The Open Work. Harvard University Press, Cambridge (1989)
3. Eladhari, M.P.: Re-tellings: the fourth layer of narrative as an instrument for cri-
tique. In: Rouse, R., Koenitz, H., Haahr, M. (eds.) ICIDS 2018. LNCS, vol. 11318,
pp. 65–78. Springer, Cham (2018). https://doi.org/10.1007/978-3-030-04028-4 5
4. Grinblat, J.: Emergent narratives and story volumes. In: Short, T., Adams, T.
(eds.) Procedural Generation in Game Design. CRC Press (2017)
5. Hawthorne, S.M.: Reparative reading as queer pedagogy. J. Fem. Stud. Relig.
34(1), 155–160 (2018)
6. Koenitz, H.: Towards a specific theory of interactive digital narrative. In: Koenitz,
H., Ferri, G., Haahr, M., Sezen, D., Sezen, T.I. (eds.) Interactive Digital Narrative:
History, Theory and Practice, pp. 91–105. Routledge (2015)
7. Kreminski, M., Samuel, B., Melcer, E., Wardrip-Fruin, N.: Evaluating AI-based
games through retellings. In: Proceedings of the AAAI Conference on Artificial
Intelligence and Interactive Digital Entertainment, vol. 15, pp. 45–51 (2019)
8. Love, H.: Truth and consequences: on paranoid reading and reparative reading.
Criticism 52(2), 235–241 (2010)
9. Ohito, E.O.: Refusing curriculum as a space of death for Black female subjects:
a black feminist reparative reading of Jamaica Kincaid’s girl. Curric. Inq. 46(5),
436–454 (2016)
10. Rezk, A.M., Haahr, M.: The case for invisibility: understanding and improving
agency in black mirror’s Bandersnatch and other interactive digital narrative
works. In: Bosser, A.-G., Millard, D.E., Hargood, C. (eds.) ICIDS 2020. LNCS,
vol. 12497, pp. 178–189. Springer, Cham (2020). https://doi.org/10.1007/978-3-
030-62516-0 16
11. roBurky: Alice and Kev (2009). https://aliceandkev.wordpress.com/
12. Roth, C., van Nuenen, T., Koenitz, H.: Ludonarrative hermeneutics: a way out and
the narrative paradox. In: Rouse, R., Koenitz, H., Haahr, M. (eds.) ICIDS 2018.
LNCS, vol. 11318, pp. 93–106. Springer, Cham (2018). https://doi.org/10.1007/
978-3-030-04028-4 7
13. Ruberg, B., Shaw, A.: Queer Game Studies. U of Minnesota Press, Minneapolis
(2017)
14. Ryan, J.: Curating Simulated Storyworlds. Ph.D. thesis, University of California,
Santa Cruz (2018)
216 J. Grinblat et al.
15. Sedgwick, E.K.: Paranoid reading and reparative reading, or, you’re so paranoid,
you probably think this essay is about you. In: Touching Feeling, pp. 123–152.
Duke University Press, Durham (2003)
16. Steam forum users: What to do? : Her Story General Discussions (2015). https://
steamcommunity.com/app/368370/discussions/0/530645446308194754/
17. Stone, K.: Time and reparative game design: queerness, disability, and affect. Game
Stud. 18(3) (2018)
Text, Retelling, and the Digital:
Reimagining the Mahabharata Through
Interactive Games
1 Introduction
Mahabharata, one of the great Indian epics, unravels the complexities of human
thought and condition. This paper focuses on the stories of marginalized char-
acters which remain unnoticed in the reader’s pursuit to uncover the popularly
known stories. With a contemporary lens of caste hierarchy and patriarchal
marginalization, “Hidimba”1 stands out as one of the characters who has the
strength to influence the plot but lacks the power to exercise her agency. As we
1
Hidimba is also called “Hidimba”, “Hidimbi” and “Hidimbaki” at different places.
We refer to her as Hidimba in this paper. Hidimba is referred to as“Asuri” and
“Rakshasi” (which roughly translates to a demoness). For the sake of this paper, we
call her a Rakshasi.
c Springer Nature Switzerland AG 2021
A. Mitchell and M. Vosmeer (Eds.): ICIDS 2021, LNCS 13138, pp. 217–221, 2021.
https://doi.org/10.1007/978-3-030-92300-6_20
218 R. Kar et al.
focus on the Odia retelling by Sarala Das, a 15th century Shudra2 poet, we bring
to light one such episode from the epic which depicts the strength of Hidimba’
character [2,7]. The narration of this story as a digital game aims to sensitize
the users about the existing biases and prejudices prevalent in our society and
to familiarise them with the ideas of agency, identity and perception. The game
unfolds differently for each user based on their choices at every juncture in the
episode, thus shifting the control of the story from the narrator to the user [5].
by the characters given their situation. In this manner, it becomes easier and
more fulfilling for the user to experience and practice empathy [3].
The pilot edition of the game5 was conducted through a Google Form. In this
pilot study, we narrate the confrontation between Hidimba and Draupadi drawn
from an episode in the Sarala Mahabharata. At the moment, the game allows
the user to make decisions on behalf of Hidimba. At every decision point, the
player gets to choose between two options6 . The story moves ahead on the basis
of these choices and consequently has multiple endings. One of the branches
leads to the original end whereas all others are created from the researchers’
poetic liberty. We attempted to create alternatives that would agree with the
characters’ personality yet spin the story in a different direction. The players are
also required to give reasoning for their choice of action. The set of choices are
unique to each user and indirectly represent their personality traits.
The pilot study was conducted with 39 participant- 20 females and 19 males.
Majority of our respondents belonged to the age group of 18–25 and 36–50 years.
It was observed that Hidimba, the protagonist of our pilot study, was one of the
most recognised marginal characters from the epic. The feedback suggested that
the dialogues which served as the background for each decision point needed
more clarity to encourage better decision making. We observed that when pro-
vided with options which do not affect a person in their real life, they always
tend to choose the more righteous option. It was difficult for the respondents to
realise the actual limitations of the situation and were inclined to choose the ideal
action. However, on further reflection and on being provided a similar dilemma
drawn from their personal experiences, they became more critical of their choices
and the rationale behind choosing them. The most prominent limitation of the
first iteration of this game was its textual character introductions which roman-
ticised Hidimba. It did not allow the users to form an impartial opinion of the
characters as one of our respondents said, “the introductions themselves seemed
to idolize certain characters”.
New media and technology have gathered a vast audience to share a wide
spectrum of content, including mythological stories. These stories have come a
long way from language specific literature to more experiential formats. The
visual representation plays a key role in driving the viewer’s imagination and
thought process. A “malicious” demon will be represented with features that
viewers attribute as scary and harmful. While a docile character would probably
be given softer features. It is observed that viewers are often shown what is more
desirable and conventionally acceptable, which often builds upon the pre-existing
biases even at the cost of straying away from authenticity [4].
5
Link to Pilot Study on Mahabharata.
6
Link to Flow of the Game.
220 R. Kar et al.
The introduction of the characters of the episode are, now, in the form of short
poems8 . Poems are a great form of abstraction and allow user interpretation to a
large extent. The same lines can be interpreted by a reader in their own subjective
way based upon their life experiences, mindset and thought process thus reducing
any bias introduced by the creators of the game. The characters as shown in Fig. 1
have been designed keeping in mind the perception of viewers in combination
with the literature. For example, Draupadi is represented as a lean built fair
woman in mainstream media, which aligns with the socially accepted beauty
standards in the modern world. This is different from how she is described in
the texts where she has a darker complexion. Similarly, Hidimba, in mainstream
media, is portrayed as a vicious rakshasi with “dehumanising” features, like
fangs, bone ornaments, deformities. The motivation behind redesigning these
characters is to challenge and try to eliminate social biases and stereotypes
based on physical appearances.
Using this interactive game as a tool, we break down Hidimba and all other
characters into entities everyone can relate to. We challenge the public idea of
a Rakshasi, and question the basis of their perception. When forced to think
about why the stories they read or the plays they’ve watched as children have a
particular representation of the character- both physical and behavioral - people
begin to realise the flexibility of these characters. They don’t necessarily fit into
a given mold. The game gives the users a chance to create their own opinion
about the marginalized characters from the Mahabharata, while experiencing
empathy and compassion for them in stories which deserve to be heard.
7
Link to the study of Visual Perception.
8
Link to the poetic character introductions.
Mahabharata Through Interactive Games 221
References
1. Patnaik, B.N.: Sarala Mahabharat: when Hidimbaki and Draupadi met...,
May 2008. https://saralamahabharat.blogspot.com/2008/05/when-hidimbaki-and-
draupadi-met.html
2. Dash, K.C.: Discourse of literary tradition and its socio-historical perspec-
tive. Lokaratna 4 (2011). http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.
1.1.737.4099&rep=rep1&type=pdf
3. Flottemesch, K.: Learning through narratives: the impact of digital story-
telling on intergenerational relationships. Acad. Educ. Leadersh. J. 17(3), 53–60
(2013). https://www.proquest.com/openview/3f3d0c08463329dd46d6f8fd4e9ce761/
1?pq-origsite=gscholar&cbl=38741
4. Goldschmidt, G., Smolkov, M.: Variances in the impact of visual stimuli on design
problem solving performance. Design Stud. 27(5), 549–569 (2006). https://doi.
org/10.1016/j.destud.2006.01.002. https://www.sciencedirect.com/science/article/
pii/S0142694X06000172
5. Robin, B.: The educational uses of digital storytelling, pp. 709–716. Association for
the Advancement of Computing in Education (AACE), March 2006. https://www.
learntechlib.org/primary/p/22129/
6. Sahu, D.U.N.: The great Indian epic: Mahabharat in Orissa, Assam and Ben-
gal, pp. 48–50, November 2008. http://magazines.odisha.gov.in/Orissareview/2008/
November-2008/engpdf/48-50.pdf
7. Satpathy, S., Nayak, J.K.: Sarala Mahabharata: introduction. Indian Lit. 58(3
(281)), 7–14 (2014). https://www.jstor.org/stable/44753707. Publisher: Sahitya
Akademi
A Coauthorship-Centric History
of Interactive Emergent Narrative
1 Introduction
Interactive emergent narrative (IEN) is an approach to the construction of inter-
active digital narrative experiences that aims to create computational systems
from which narrative naturally emerges, bottom-up, through simulation and user
interaction. Since the 1995 introduction [7] and 1999 popularization [2] of the
term “emergent narrative”, IEN has predominantly been framed as a solution
to the problem of creating narrative play experiences in which the player may
meaningfully participate as a character in an authored storyworld. For almost as
long, however, there has also existed an alternative perspective on the purpose
of IEN. This alternative perspective frames IEN as an approach to the creation
of play experiences in which the player takes on the role of the author of the
“emergent” narrative, rather than a participant. From this perspective, the goal
of IEN can be viewed as the provision of the user with creativity support [32]:
IEN games and systems must give the user the tools and materials they need to
construct a story of their own, even in the presence of barriers to player creativity
that might obstruct or inhibit the construction of a successful story [14].
These two contrasting perspectives on IEN, despite the substantial differences
between them, have remained rhetorically entangled due to the lack of a clear
distinction between the play-pleasures of authorship and the play-pleasures of
participation. Additionally, when IEN is discussed in a modern context, it is
c Springer Nature Switzerland AG 2021
A. Mitchell and M. Vosmeer (Eds.): ICIDS 2021, LNCS 13138, pp. 222–235, 2021.
https://doi.org/10.1007/978-3-030-92300-6_21
A Coauthorship-Centric History of Interactive Emergent Narrative 223
often taken to stand specifically for the form of IEN that targets participatory
user experiences.
We believe that several recent developments in the study of interactive dig-
ital narrative justify a reexamination of the player-authorship perspective on
IEN. The study of retellings [6], or the stories that players tell about their
play experiences (often in IEN games like The Sims and Dwarf Fortress), has
called attention to the cultural significance of these stories—and to the exten-
sive work that players do in constructing them, for instance by embellishing or
extrapolating beyond the bare events of play to craft a better story [13,22] and
constructing stories that ironically comment on or critique the IEN systems with
which they were produced [33]. James Ryan [26] has drawn a clear distinction
between the raw material of simulation and a particular telling or narrativization
of this material, proposing a new curationist perspective on emergent narrative
that highlights the work done by the human interactor in crafting a coherent
story from the disorderly and overwhelming outputs of a simulation engine. In
tabletop roleplaying games, recent years have seen the emergence of a clearer
distinction between games that focus on enabling player participation in a story
(like the traditional Dungeons and Dragons style of tabletop roleplaying) and
games that intend to enable player coauthorship of a story (like the “GM-less”
games Microscope and The Quiet Year ) [8]. And several recent efforts have been
made to construct digital IEN games in which the intended player experience is
one of coauthorship, rather than one of participation [10–12,30].
In this paper, we attempt to trace the history of emergent narrative (EN)
as both a term and a concept, with a particular focus on interactive emer-
gent narrative (IEN): the use of emergent narrative as an approach or solution
to the problems presented by the combination of interactivity and narrativity.
Walsh [34], Ryan [26] and Larsen et al. [16] have all made significant attempts to
disentangle the history of emergent narrative, and we draw extensively on their
efforts here. These existing histories, however, stop short of drawing a clear dis-
tinction between two frequently conflated uses of IEN: the use of IEN to enable
participation play and the use of IEN to enable authorship play. We therefore
focus especially on drawing out and clarifying this distinction, with an eye to
how this distinction can inform the design of IEN systems intended to facilitate
each kind of play.
We argue that one central play-pleasure of IEN lies in the use of IEN systems
by players to actively compose narratives. In this context, the computer functions
as a storytelling partner that supports the player’s storytelling practice, often
by keeping track of complicated storyworld state; elaborating on the player’s
actions in unexpected ways, or otherwise suggesting new directions in which the
narrative could be taken; and providing curatorial affordances [26] that assist
the player in extracting particularly resonant details of the play experience into
narrative form.
This use of IEN to enable player authorship of narrative is distinct from the
use of IEN to support the player’s participation in a storyworld through the
embodiment of a particular character. Although these uses are sometimes com-
224 M. Kreminski and M. Mateas
patible within a single IEN play experience, they are also frequently at odds with
one another. Artificially limiting the player’s viewpoint and agency to align with
that of a single character, for instance, may help to strengthen player identifica-
tion with the character in question (and thus the player’s sense of participation
in the world), but may simultaneously inhibit the player’s ability to tell stories
about storyworld events to which their point-of-view character did not directly
bear witness, or to “nudge” the storyworld in certain desirable ways in order to
promote the development of a particular narrative direction or theme. Therefore,
it is useful to consider the play-pleasures of participation and the play-pleasures
of authorship as related but distinct phenomena, and to maintain a consistent
awareness of which play-pleasures you intend to prioritize during IEN design.
One reason for confusion around the concept of emergent narrative is that
people have discussed the concept without using the term, and have also used the
term to talk about several different related concepts. In this paper, we will first
discuss early usage of the concept of emergent narrative prior to the appearance
of the term. Then we will discuss the birth, popularization, and development
of the term itself, including two distinct strands of thought that view emergent
narrative primarily through the lens of player participation and player author-
ship respectively, as well as an additional expansive perspective that attempts
to situate both interactive and non-interactive EN systems within a common
framework.
For the purposes of this paper, we adopt a definition of “authorship” that fol-
lows earlier scholarship on emergent narrative—particularly the definition given
by Louchart and Aylett (2004) [18], which holds that an author is someone who
“seeks control over the direction of a narrative in order to give it a satisfying
structure”. Though we recognize that this definition sidesteps the debate over
the boundary between reader and author that has taken place in hypertext com-
munities (e.g., in Landow’s work [15]), not to mention the extensive debate over
the broader concept of authorship that has unfolded in modern and postmodern
literary theory, we hope that this definition can nevertheless serve as a use-
ful jumping-off point from which to survey existing literature in the interactive
digital narrative tradition. Reconciliation of how authorship is discussed in an
IDN context with how it has been conceived of in literary theory more broadly
remains a potentially fruitful direction for future research.
2 Pre-interactive EN
Publication S P A Notes
Early worldsim story generators, Predate coinage of EN as a term, but
e.g. Tale-Spin (1975) [21] retroactively included by Ryan (2018)
Galyean (1995) [7] First academic use of EN as a term
Aarseth (1997) [1] Discusses EN without using term
directly; treats MUD players mostly as
participants but sometimes uses
coauthorship language
Murray (1997) [23] Discusses EN without using term directly
Aylett (1999, 2004) [2, 18] Introduces definition of EN that remains
most widely used today
Mateas (2002) [20] Cites Aylett but interprets EN
differently, without comment
Jenkins (2004) [9] Cites and disagrees with Murray; doesn’t
cite Galyean/Aylett/Mateas
Ryan, Marie-Laure (2006) [29] First occurrence of expansive definition
Walsh (2011) [34] Cites and disagrees with Ryan (2006)
Ryan, James (2015) [27] Closely follows Aylett’s definition
Ryan, James (2018) [26] Pivots from Ryan (2015) to an expansive
definition
Dehn [4] critiqued Tale-Spin for its limitations—particularly its need to begin
a simulation run with all eventually-necessary characters, props, locations, and
so on already invented in advance—and instead proposed an alternative app-
roach to story generation, founded not in Tale-Spin’s world simulation (simula-
tion of agents with various needs, desires, and so on) but in author simulation:
the simulation of the process by which a human author assembles a story. In
Dehn’s view, human authors take a variety of actions beyond placing characters
in a storyworld and allowing them to produce a story through their interactions.
For instance, authors may invent characters, props, and locations on the fly as
needed; plan out desired plot beats far in advance; deliberately develop charac-
ters and plotlines to engage with or comment on specific themes; or rearrange
the order in which events are presented to achieve a desired effect on the reader.
Dehn’s approach represented a break from emergent narrative, and helped to
found an alternative tradition of story generation founded on author simulation,
which is continued in part today by the robust field of planner-based story gen-
eration [24,36]. The analog story generation framework Plotto (1928) [3], which
226 M. Kreminski and M. Mateas
was later operationalized as a digital story generator [5], can also be viewed as
an early manifestation of story generation via author simulation.
In the early context of story generation, then, the emergent narrative app-
roach can largely be identified with bottom-up world simulation as opposed to
top-down author simulation. However, neither the early world simulation-driven
story generators nor their author simulation-driven counterparts present them-
selves as live interlocutors that can take input from a user midway through the
story generation process and adapt the story in response. Instead, these systems
aim to produce fully formed static narratives without a human in the loop. Due
to our focus on interactive emergent narrative specifically, we thus consider these
story generators to be largely outside the scope of our interest here.
Louchart and Aylett (2004) [18] develop this distinction and make it more
explicit. Here, Louchart and Aylett present three contrasting perspectives on the
role of the user in interactive storytelling (user-as-spectator, user-as-participant,
and user-as-author), and frame emergent narrative as an attempt to solve the
problems associated with the user-as-participant perspective specifically:
The role of the user is a key issue in interactive storytelling, with whether
the user is considered as an author or a participant within the story hav-
ing a major impact on theoretical approaches. The contradiction between
authorship and participation is an important element of the narrative para-
dox previously mentioned. On the one hand an author seeks control over
the direction of a narrative in order to give it a satisfying structure. On
the other hand a participating user demands the autonomy to act and
react without explicit authorial constraints. Casting the user either as a
spectator, with no ability to act, or as the author him or herself avoids
this problem, however it does not offer a solution for a participating user
in real-time interaction within a narrative display. It also limits the sto-
ryline to a single entity. We argue that a serious consideration of the user
as participant can actually present a solution to the narrative paradox,
in the sense that it would encourage the emergence of several storylines
while still leaving the user with the responsibility of conducting real-time
meaningful actions within the unfolding narrative.
This distinction between user roles in IEN is particularly useful for under-
standing how different authors understand the scope of emergent narrative as
a term or concept. In the remainder of this paper, we will make frequent refer-
ence to Aylett’s proposed user roles in order to contextualize the design goals of
different EN systems and approaches to EN system design.
Aylett’s conceptualization of emergent narrative as fundamentally based in
participation play is further supported (if only implicitly) by two early key works
of interactive narrative scholarship: Espen Aarseth’s Cybertext [1] and the first
edition of Janet Murray’s Hamlet on the Holodeck [23]. Both texts were published
in 1997, shortly after the first appearance of the term “emergent narrative”, but
prior to Aylett’s arrival on the scene in 1999. At the time, the term “emergent
narrative” was not yet in wide use, but both texts engage with emergence and its
relationship to interactive narrative. Moreover, although neither text explicitly
draws a clear distinction between participant and author as user roles, both deal
primarily with play experiences in which the user directly controls a particular
embodied character in the storyworld, and both tend to come down on the side
of narrative as either “merely” emergent from interaction or primarily produced
by the system rather than by the participating human user.
Aarseth’s interest in narrative emergence centers largely on the phenomenon
of collaborative improvisatory storytelling in multi-user dungeons, or MUDs. He
describes this collaborative storytelling as analogous to a “jazz jam session” [1,
p. 158] and frames players in MUDs as “literary cyborgs” [1, p. 160], collabo-
rating both with one another and with autonomous bots to construct textual or
228 M. Kreminski and M. Mateas
literary “happenings” that may lack the “grand structural schemes” of “prose
narrative”, but that are worthy of treatment as texts. This characterization
seems to position MUD players somewhere between participants and authors
from a narrative perspective, while largely avoiding the question of narrativity
one way or the other. Simultaneously, for Aarseth, “to be an ‘author’ means to
have configurative power over not merely content but also over a work’s genre
and form” [1, p. 164]. Through this lens, authorship may or may not be available
to players of MUDs and other highly malleable IEN systems (that allow users,
for instance, to define new commands or interaction mechanics), and remains
clearly out of reach for users of more closed IEN systems (such as The Sims,
Dwarf Fortress, and other IEN games that are widely known today).
The tension between Aarseth’s desire to characterize MUD play as a form of
collaborative storytelling and his hesitance to assign users of computational sys-
tems the status of “author” may be attributable to Aarseth’s view of early IEN
systems (like the MUD) as texts in and of themselves, rather than as tools for
producing texts. This conflation—the same one that Ryan [26] argues against,
with his admonition that the simulated storyworld is not itself an emergent
story—leads Aarseth to treat the unedited transcripts of MUD play sessions
as the artifacts of narrative interest. Without the distinction between narra-
tive material and retelling advanced in more recent work [6,13,14,16,22,26,33],
Aarseth’s view on the role of the user in IEN ends up suspended ambiguously
between the user-as-participant and user-as-author perspectives.
Murray’s position, meanwhile, is less ambiguous in its strong association of
narrativity with participation. Her discussion of MUDs positions the MUD as
a “collective creation” and a “digital narrative environment” in which stories
can take place [23, p. 103], but does not treat the MUD itself as a narrative
per se. MUDs for Murray are essentially a form of “participatory theater” [23,
p. 152], and the central design issue that they pose is not one of co-creativity
but of “discovering the conventions of participation” [23, p. 153] that will pre-
serve the participant’s sense of immersion. Similarly, in her discussion of different
player attitudes toward SimCity, Murray states that “for the wife, [the game]
was a narrative” [23, p. 105], but places the wife in the role of receiving the
story rather than creating it. This attitude is made more explicit by Murray’s
assertion that the “narrative quality” of SimCity is “expand[ed]” by changes
in later versions that “[allow] the player to live inside a more-detailed three-
dimensional city rather than only manipulate it from on high” [23, p. 106]—or,
essentially, that a simulation game becomes more of a narrative experience the
more closely identified a player becomes with a particular character in the story-
world. Ultimately, Murray’s goal is not to produce play experiences that center
around player authorship, but to extend into computational media the pleasures
of experiencing a story that someone else has authored. To Murray, “once we
understand simulations as interpretations of the world, the hand behind the mul-
tiform plot will feel as firmly present as the hand of the traditional author” [23,
p. 347]—and it is this firm sensation of authoredness that she believes interactive
narrative systems should ultimately aspire to provide.
A Coauthorship-Centric History of Interactive Emergent Narrative 229
Perhaps the first concrete evidence of this alternative view can be found in
Mateas’s 2002 dissertation [20]. Like Galyean before him, Mateas does not spend
much time on emergent narrative, listing it as one of a wide variety of potential
alternatives to the interactive drama approach to interactive narrative on which
he prefers to focus. But Mateas’s definition of emergent narrative is focused
on player authorship: for him, “emergent narrative is concerned with provid-
ing a rich framework within which individual players can construct their own
narratives, or groups of players can engage in the shared social construction of
narratives”, and emergent narrative can be explicitly contrasted with the tradi-
tional view of narrative as “highly structured experiences created by an author
for consumption by an audience” [20, p. 20]:
6 Conclusion
References
1. Aarseth, E.J.: Cybertext: Perspectives on Ergodic Literature. Johns Hopkins Uni-
versity Press, Baltimore (1997)
2. Aylett, R.: Narrative in virtual environments - towards emergent narrative. In:
Proceedings of the AAAI Fall Symposium on Narrative Intelligence, pp. 83–86
(1999)
3. Cook, W.W.: Plotto: The Master Book of All Plots. Tin House Books, Portland
(2011)
4. Dehn, N.: Story generation after TALE-SPIN. In: International Joint Conference
on Artificial Intelligence, pp. 16–18 (1981)
5. Eger, M., Potts, C., Barot, C., Young, R.M.: Plotter: operationalizing the master
book of all plots. In: Proceedings of the AAAI Conference on Artificial Intelligence
and Interactive Digital Entertainment, vol. 11 (2015)
6. Eladhari, M.P.: Re-tellings: the fourth layer of narrative as an instrument for cri-
tique. In: Rouse, R., Koenitz, H., Haahr, M. (eds.) ICIDS 2018. LNCS, vol. 11318,
pp. 65–78. Springer, Cham (2018). https://doi.org/10.1007/978-3-030-04028-4 5
7. Galyean, T.A.: Narrative Guidance of Interactivity. Ph.D. thesis, Massachusetts
Institute of Technology (1995)
8. Guzdial, M., et al.: Tabletop roleplaying games as procedural content generators.
In: International Conference on the Foundations of Digital Games (2020)
9. Jenkins, H.: Game design as narrative architecture. Computer 44(3), 118–130
(2004)
10. Kreminski, M., et al.: Cozy mystery construction kit: prototyping toward an AI-
assisted collaborative storytelling mystery game. In: Proceedings of the 14th Inter-
national Conference on the Foundations of Digital Games (2019)
11. Kreminski, M., Dickinson, M., Mateas, M., Wardrip-Fruin, N.: Why are we like
this?: Exploring writing mechanics for an AI-augmented storytelling game. In:
Electronic Literature Organization Conference (2020)
234 M. Kreminski and M. Mateas
12. Kreminski, M., Dickinson, M., Mateas, M., Wardrip-Fruin, N.: Why are we like
this?: The AI architecture of a co-creative storytelling game. In: International
Conference on the Foundations of Digital Games (2020)
13. Kreminski, M., Samuel, B., Melcer, E., Wardrip-Fruin, N.: Evaluating AI-based
games through retellings. In: Proceedings of the AAAI Conference on Artificial
Intelligence and Interactive Digital Entertainment, vol. 15, pp. 45–51 (2019)
14. Kreminski, M., Wardrip-Fruin, N.: Generative games as storytelling partners. In:
Proceedings of the 14th International Conference on the Foundations of Digital
Games (2019)
15. Landow, G.P.: Hypertext: The Convergence of Contemporary Critical Theory and
Technology. Johns Hopkins University Press, Baltimore (1991)
16. Larsen, B.A., Bruni, L.E., Schoenau-Fog, H.: The story we cannot see: on how a
retelling relates to its afterstory. In: Cardona-Rivera, R.E., Sullivan, A., Young,
R.M. (eds.) ICIDS 2019. LNCS, vol. 11869, pp. 190–203. Springer, Cham (2019).
https://doi.org/10.1007/978-3-030-33894-7 21
17. Laurel, B.K.: Toward the Design of a Computer-Based Interactive Fantasy System.
Ph.D. thesis, The Ohio State University (1986)
18. Louchart, S., Aylett, R.: The emergent narrative theoretical investigation. In: Nar-
rative and Interactive Learning Environments Conference, pp. 21–28 (2004)
19. Louchart, S., Truesdale, J., Suttie, N., Aylett, R.: Emergent narrative: past, present
and future of an interactive storytelling approach. In: Interactive Digital Narrative,
pp. 185–199. Routledge (2015)
20. Mateas, M.: Interactive Drama, Art and Artificial Intelligence. Ph.D. thesis,
Carnegie Mellon University (2002)
21. Meehan, J.R.: TALE-SPIN, an interactive program that writes stories. In: Inter-
national Joint Conference on Artificial Intelligence (1977)
22. Murnane, E.: Emergent Narrative: Stories of Play, Playing with Stories. Ph.D.
thesis, University of Central Florida (2018)
23. Murray, J.H.: Hamlet on the Holodeck: The Future of Narrative in Cyberspace.
MIT Press, Cambridge (1997)
24. Porteous, J.: Planning technologies for interactive storytelling. In: Nakatsu, R.,
Rauterberg, M., Ciancarini, P. (eds.) Handbook of Digital Games and Entertain-
ment Technologies. Springer, Singapore (2016). https://doi.org/10.1007/978-981-
4560-52-8 71-1
25. Roberts, D.L., Isbell, C.L.: A survey and qualitative analysis of recent advances
in drama management. Int. Trans. Syst. Sci. Appl. Spec. Issue Agent Based Syst.
Hum. Learn. 4(2), 61–75 (2008)
26. Ryan, J.: Curating Simulated Storyworlds. Ph.D. thesis, University of California,
Santa Cruz (2018)
27. Ryan, J.O., Mateas, M., Wardrip-Fruin, N.: Open design challenges for interactive
emergent narrative. In: Schoenau-Fog, H., Bruni, L.E., Louchart, S., Baceviciute,
S. (eds.) ICIDS 2015. LNCS, vol. 9445, pp. 14–26. Springer, Cham (2015). https://
doi.org/10.1007/978-3-319-27036-4 2
28. Ryan, M.L.: Narrative and the split condition of digital textuality. In: The Aes-
thetics of Net Literature: Writing, Reading and Playing in Programmable Media,
pp. 257–281. Columbia University Press (2005)
29. Ryan, M.L.: Avatars of Story. University of Minnesota Press, Minneapolis (2006)
30. Samuel, B., Mateas, M., Wardrip-Fruin, N.: The design of writing buddy: a mixed-
initiative approach towards computational story collaboration. In: Nack, F., Gor-
don, A.S. (eds.) ICIDS 2016. LNCS, vol. 10045, pp. 388–396. Springer, Cham
(2016). https://doi.org/10.1007/978-3-319-48279-8 34
A Coauthorship-Centric History of Interactive Emergent Narrative 235
31. Samuel, B., Ryan, J., Summerville, A.J., Mateas, M., Wardrip-Fruin, N.: Bad news:
an experiment in computationally assisted performance. In: Nack, F., Gordon, A.S.
(eds.) ICIDS 2016. LNCS, vol. 10045, pp. 108–120. Springer, Cham (2016). https://
doi.org/10.1007/978-3-319-48279-8 10
32. Shneiderman, B.: Creativity support tools: accelerating discovery and innovation.
Commun. ACM 50(12), 20–32 (2007)
33. Sych, S.: When the fourth layer meets the fourth wall: the case for critical game
retellings. In: Bosser, A.-G., Millard, D.E., Hargood, C. (eds.) ICIDS 2020. LNCS,
vol. 12497, pp. 203–211. Springer, Cham (2020). https://doi.org/10.1007/978-3-
030-62516-0 18
34. Walsh, R.: Emergent narrative in interactive media. Narrative 19(1), 72–85 (2011)
35. Weallans, A., Louchart, S., Aylett, R.: Distributed drama management: beyond
double appraisal in emergent narrative. In: Oyarzun, D., Peinado, F., Young,
R.M., Elizalde, A., Méndez, G. (eds.) ICIDS 2012. LNCS, vol. 7648, pp. 132–143.
Springer, Heidelberg (2012). https://doi.org/10.1007/978-3-642-34851-8 13
36. Young, R.M., Ware, S.G., Cassell, K.B., Robertson, J.: Plans and planning in
narrative generation: a review of plan-based approaches to the generation of story,
discourse and interactivity in narratives. Sprache und Datenverarbeitung Spec.
Issue Form. Comput. Models Narrative 37(1–2), 41–64 (2013)
Wrestling with Destiny: Storytelling
in Perennial Games
1 Introduction
“Games-as-a-service” [28,88] are the most popular form of games on the current
market1 . Games like Minecraft [67], League of Legends [81], Fortnite [41], and
1
Exact player counts are difficult to ascertain because they are often company
secrets unless announced, as e.g. Fortnite did [97], yet there are sources such as
the Steam Charts [98] (which only shows Steam games) or the updated Wikipedia
entry on most-played video games by player count https://en.wikipedia.org/wiki/
List of most-played video games by player count. Purveying these two lists give a
clear overall picture of the popularity of games-as-a-service.
c Springer Nature Switzerland AG 2021
A. Mitchell and M. Vosmeer (Eds.): ICIDS 2021, LNCS 13138, pp. 236–254, 2021.
https://doi.org/10.1007/978-3-030-92300-6_22
Wrestling with Destiny: Storytelling in Perennial Games 237
Destiny [25] have changed not only how games are made and distributed [39]
but also how they are played [111]. They are all “on-going”, with consistent
updates, patches and refreshments of their content, keeping players interested
and wanting to come back for more. It is proving to be a very successful business
model, and part of the undescribed success of these games lie in their storytelling,
their ability to tell on-going stories for and with the players. We argue that
it should be considered an important part of the experience of playing them.
Initially, one might assume their persistent, on-going nature renders them unfit
for plotted storytelling, as they cannot accomodate individual player actions as
meaningful change. In this work we instead focus on describing and analyzing
these experiences as storytelling experiences for whole audiences.
MMOs (Massive Multiplayer Online) can be considered a precursor to this
format. However, existing research provides few answers on how MMOs function
as storytelling experiences. At the height of research into Everquest and World
of Warcraft [5,53,75], many described MMOs as unable to tell stories in the
“traditional” sense: No three act structure, no plot [6,59]. The authored stories
of MMOs have not been in focus, despite plenty of plots one can get from World
of Warcraft (Sect. 4.1). Beyond games, there are many precedents for on-going
storytelling. Professional wrestling [73,105] has been telling on-going, live sto-
ries since the 80s [49], along with sports, reality TV, soap operas, long-running
book and comics series, that all share common traits with these on-going games.
When viewed as storytelling experiences, there are clear similarities. It can even
be associated to how mythological storytelling operates more broadly. League of
Legends has consistently added to its narrative world through websites, cinemat-
ics, and in-game changes [43,82], and traditional MMOs like World of Warcraft
are no different. We argue that these games tell traditionally plotted stories, yet
on an entirely different temporal scale than, say, Mass Effect [8].
We define these experiences as “perennial experiences”: Experiences that are
perpetual, temporally continuous, and has a universal chronicle. All games which
are perennial experiences can be called “perennial games”. The word perennial
is chosen because it encapsulates the perpetuity and recurrence of these expe-
riences: They repetitiously add new content (a new TV episode is released, a
new update to a game, a new sports match is played), but it is never fully
repetitious—it is always a new match, always a new patch, always a new episode.
The second aspect is that it takes time for these experiences to develop and
change. One update might not drastically change the status quo, but over many,
the fabric of their universes slowly becomes something new. Perennial games
are not a new genre of video games, in the popular understanding of genre as
defined by the primary interactions (such as “Shooter” or “RPG” (role playing
game)) [32,110], nor a genre of storytelling as defined in film studies, by stylistic
similarities [110]. The perennial nature of a game is not defined by its gameplay,
as a perennial experience is not defined by its storytelling content. Rather, it is
a form of storytelling many genres and media can perform, and it is in fact often
transmedic [50,85] in nature (although it does not have to be).
While perennial games have been discussed in storytelling terms [45,59,80,
94], this is the first time perennial experiences have been defined as a form
238 B. A. Larsen and E. Carstensdottir
of storytelling experience across games and not-games. This work aligns these
otherwise disparate experiences, and allows new ways to describe them. Using
the case study of Destiny [25,26], we highlight how this manifests in an existing
perennial game, how these stories naturally have a muddled diegesis and how
they create myths. These qualities have always been present in MMOs, yet it
takes significant time (as in real-world waiting time) to see it unfold. Perennial
games are among the most watched and played media in the world in 2021, and
is therefore of interest to the interactive storytelling community. While clearly
effective for storytelling, this aspect has been woefully understudied.
– Perpetuity. These experiences are perpetual. They do not have stated end
or explicit final goal: League of Legends has no obvious ending in sight. New
content, events, episodes, or updates can always add to the experience, and
the audience can always expect more for one reason or another. They might
end, as City of Heroes did in 2012 [34], but the ending is arbitrary rather
than planned from the outset. (This excludes most narratives.)
– Temporal Continuity. These experiences are tied to real-world time [113]
rather than experiential time [70]. The fictional space persists even when the
audience is not actively engaging with it: Wrestling matches occur without
you watching them, the “World” of Warcraft is there despite you being in
it, Doctor Who episodes aired before you watched them. The story does not
retract or repeat, it always appears in sequence and does not go backwards3 .
The fictional time might not flow 1:1 with the real world (though it can, in
e.g. Destiny), but it is operating in a similar fashion to the real world in the
sense that it does not revolve around a single person’s perception—even the
author’s. It is always progressing forward and changing, intermingled with
2
Referencing the earliest release date for these is inherently misleading, as these games
have changed from their release state. Therefore, whenever we reference a specific
example, we reference relevant version number or date-identifying information, along
with supplemental material showing recorded cinematics, dialogue transcripts etc.
as there is often no way of experiencing this inside the game experience today.
3
Spin-offs, flashbacks, reboots, or alterations of the timeline muddle this significantly.
The existence of such narrative devices could signify a lack of temporal continuity,
but we can still view the audience experience as linear and sequential regardless.
Wrestling with Destiny: Storytelling in Perennial Games 239
the time of the real world. A common signifier is how these experiences have
live, missable moments [83,109,114] where you “had to be there” [100]. (This
excludes static spaces and stories that only progress when engaged with.)
Fig. 1. This triangle shows the trifecta of control as shared between the author, the
game or system, and the audience, as they negotiate the diegesis and universal chronicle.
These then evolve concurrently over time, as shown by the figure’s third dimension,
which shapes the experience through the control all three exert.
The author, the game, and the audience are all progressing forward through
time, during the experience. In a traditional narrative process, the author grows
and changes during development, then publishes a fixed story to the audience,
who then are influenced by this story moving forward. But in a perennial expe-
rience, all three continuously move forward in time, change and mold the expe-
rience as they do. Each aspect of the experience is consistently reflected by the
other two. Perennial experiences embody the old saying “no one steps into the
same river twice”, through their very design.
Wrestling with Destiny: Storytelling in Perennial Games 241
6
This doesn’t discount players: Play is necessary for the perennial game to function.
7
This split should not be understood as a person can only engage with one or the other
exclusively, but rather as different lenses to understand the experience of interacting
with a piece of media.
242 B. A. Larsen and E. Carstensdottir
Fig. 2. This two-dimensional spectrum shows the various degrees of control the author
has (horizontal) and the audience has (vertical). While an inverse correlation that less
author control means more audience control exists in many cases, it is not always
the case, as highlighted by the diagonal line in the middle, which focuses on experi-
ences where neither have complete control. Note that the specific placements of the
experiences are exhibitory and should not be considered exact.
While there are experiences where clear lack of authorial control leads to
a high degree of audience control, there are also experiences where neither is
in full control; sports and sports-like experiences like Blaseball and wrestling
being the most obvious examples. The authors and audience each have a degree
of control, yet there is an element with much more to say: The emergence of
the system itself. No individual player, coach, audience member or television
promoter of a football match is in total control over what happens. The game
is decided by many players and systems coalescing into an emergent experience
[3] that is in no-one’s entire control. In Blaseball, the randomness of the game
gives a high degree of uncertainty for both the audience and the creators [61].
Similarly, in wrestling, neither the wrestlers or the writers or the audience are
in full control of the experience [35,63]. This is an example of how less authorial
control does not necessarily lead to more audience control, and while it is most
Wrestling with Destiny: Storytelling in Perennial Games 243
Through the creation of the universal chronicle over time by the three aspects, a
sense of diegesis8 is formed about what is part of the narrative world and what
is not. However, that is often a very blurry, shifting line. Wrestling is a great
example of this. It is common knowledge that wrestling is a “fake” sport, that is,
its matches are prescripted [4,63,91]. However, the real, unscripted outside world
frequently impacts the storytelling of wrestling, causing its pre-written narrative
to change in real-time to react to real-world events, and each participant will
frequently negotiate what is a part of the storyworld and what isn’t [63,73].
Simultaneously, as Sect. 4.2 will show, the fact that the audience experiences the
world in real-time causes the “virtual” events that happen in perennial stories to
become real, because they happen to real people in real situations. The universal
chronicle of perennial experiences runs parallel to the real world, and this causes
the distinction between what is part of the storyworld and what isn’t to blur
and weave, at all times. The trifecta of a perennial experience will always be in
a negotiation of what is part of its world, and what is not.
The case study of the video game Destiny will help to illustrate these effects
in a game context and how the properties of perennial experiences lead to this
kind of storytelling experience.
Destiny9 is a widely known game that has been studied with a view on identity
[30,48], player profiling and modelling [37,86,92,99], social media [29,38,107]
player networks and communication [9,76,77,87,103,106,108], economics [65],
competitive multiplayer [78,93], skill development [51,55,96], character creation
[64,89], archeology [40,79], and visuals [57,66,101], yet, on its story has only been
said it was poorly received at launch [14]. Since then, Destiny has turned into
a quintessential perennial game. The former Game Director for Destiny, Luke
Smith, defined Destiny as an “Action MMO in a single, evolving world” [20],
and while the genre itself doesn’t matter to its perennial status, the notion of a
“single, evolving world” is telling. It is perpetual and on-going, and its temporal
state is 1:1 with the real world: 1 day in our world is 1 day in the Destiny fiction;
events that happened a year ago, happened a year ago in both10 . To understand
8
Diegesis is here understood in the terms from Kleinman et al. [54].
9
We discuss Destiny [25] and Destiny 2 [26] as a single franchise and universe. The
story is a direct continuation, and they exist in the same narrative world, so there
is little reason to distinguish between them, other than distribution.
10
This is evident through a range of dialogue and interactions [36, 47], for example a
character referring to the Traveller coming alive “a year ago”, in 2018, referencing
an event that happened at the end of the Red War, in 2017 [47].
244 B. A. Larsen and E. Carstensdottir
how the universal chronicle in Destiny is formed, we need to understand how the
game tells its stories.
Destiny’s position in the spectrum of control shows how it is a perennial
experience with quite heavy author control. This authorial power is exercised
in several ways. Destiny employs well-trodden avenues from narrative video
games, like dialogue and cut-scenes and scripted missions—yet often only for
yearly expansions and pivotal moments—combined with depictions of the world
through “lore cards”, item descriptions, and embedded text, which are often
accounts of happenings in the world by fictional characters, either current or
historical. However, the assembly of these narratives into a sensible chronicle
is, for the most part, left to the players, such as when the lorebook “Truth to
Power”, released to the players over months, yet was slowly revealed to be full
of lies [19,26, Patch 2.0.2.1 and onwards]. The players are also given the task of
actuating and propulsing many of the story’s agendas, such as when, during the
“Forsaken” expansion, the area of the “Dreaming City” became cursed through
a spell triggered by the players slaying a dragon. This occurred in September
2018, and is now a historical event (the city is, as of this writing, still cursed)
[26,46,71, Patch 2.0.2.1 and onwards]. This is the primary way Destiny creates
its universal chronicle: Through authored events actuated and enforced by play-
ers. However, sometimes players cause events to feed back into the world such as
the infamous “Loot Cave” [25,44, Original release], a cave of endlessly spawning
enemies players were using to get equipment, which the developers then had to
disable. However, the Loot Cave showed up later in Destiny 2 in 2020 as an
empty cave full of corpses [26,42, Beyond Light, Patch 3.0.0.1]. This is an exam-
ple of an unintended behaviour in the game affecting player behaviour, and then
feeding back to the authors to support the universal chronicle of Destiny. By
reimplementing the cave, the developers are telling the players that the endless
shooting into a cave that happened in 2014 was in fact real, echoing how every
event expands Destiny’s world and moves it forward, regardless who or what
caused it.
Seeing a moment of gameplay of Destiny will not reveal any of this. At its
gameplay core, a player in Destiny shoots aliens. While some parts of the story-
telling is front and center, much of it can be entirely ignored, and is not required
to play the game successfully. Here, the split player/audience is helpful. A player
of Destiny is interacting with the gameplay systems, whereas the audience for
the perennial game of Destiny is experiencing the storytelling. For many play-
ers, this happens concurrently, as it is part of playing for them to experience
the story, read the lore cards and learn about the fictional world. Yet, you could
play without paying attention to the story or the reverse; read lore cards on a
website (like https://www.ishtar-collective.net/) and watch lore recaps by “My
Name is Byf” on YouTube [27] to experience the story without participating in
it. To what extent this happens in reality is out of the scope of this paper, yet it
can still deductively and anecdotally be understood as a possible way to experi-
ence these stories. Furthermore, a player joining later and learning about previ-
ous events other players did in the space, can be said to have a pure audience-
relation to those events, since they did not participate in them.
Wrestling with Destiny: Storytelling in Perennial Games 245
Destiny, for our purposes, is an MMO, despite its shooter trappings. It has a con-
sistent, inhabitable virtual world, rather than a fragmented fiction. MMOs were,
by Bartle and others, not seen as games with traditional storytelling potential:
“...virtual worlds do not have, nor ever can they have, narrative in the conven-
tional sense. They’re places. Players can act out narratives of their own within
them, but the virtual worlds can’t impose a three-act structure or anything like
it.” [6,52,59]. To contrast, here is a story from the universal chronicle of Destiny.
In 2017, early Destiny 2, Mithrax [17] was nothing more than an unnamed NPC
of an enemy faction, who was surprisingly not agressive towards us, the players
[16], but over the years he has slowly revealed his identity and intentions, turning
him from a rogue ally [21], to an insubordinate Fallen helping us overthrow a
new tyrant [23], to recently, where he has fully allied with the players, and been
given asylum in a part of the human city [24]. We hope this example illustrates
how the previous notion is wrong. The story of Mithrax is a quite conventional
story of learning to trust an unexpected ally; it just took 3.5 years to unfold. To
take another example, we today have the power of seeing all the changes that
happened to the “World” of Warcraft in the 17 years since its inception, and it is
no small story. Kings (plural) have died [11–13, Legion, Patch 3.3.0]. Entire con-
tinents upheaved [10,13, Cataclysm]. Ancient trees burned to the ground [13,62,
Patch 8.0.1]. This was not something a player could experience in 2004, when
the game released. This highlights a crucial property of perennial storytelling:
It is slow. Spending mere hours or days with any perennial experience—even
“long” hours by traditional game standards—will not reveal its true nature, and
it is therefore vital to let them take their time: It is only over time their worlds
develop, which is how the storytelling happens. This is what viewing games as
perennial experiences gives us: We can see that what was previously considered
a static world in snapshots has rather obvious parallels with experiences like
wrestling, sports, or serialized television, and there are important lessons from
these we can draw.
One of those lessons is in how perennial games work diegetically. The pace and
temporal continuity of Destiny causes its world to flow interchangeably in and
out of diegesis with the real world, as its events are forever tied to real-world
events—just as the real world has changed over the last 3 years, so has the world
of Destiny. Yet, it gets much weirder. You could (until recently [26, Beyond Light,
Patch 3.0]) go back and re-experience The Red War from 2017, in a motion that
simulates re-watching an old wrestling or sports match for nostalgia or histori-
cal reverence, as it, fictionally, only occurred once. Likewise, as in most online
games, many missions can be repeated ad nauseum for rewards, without explana-
tion for how the enemies reappear (with some rare exceptions, where respawning
is actually fictionalised within the world of Destiny [18], making the instances
where it is not even stranger). Destiny’s temporality also causes its fiction to
246 B. A. Larsen and E. Carstensdottir
be impacted by the real world: Its 2020 expansion got delayed by two months
because of the Covid-19 pandemic, causing the entire fiction of Destiny to “wait”
until it was ready, with no diegetic explanation. Another, recent example is how
the official Twitter account of the developer Bungie got “overtaken” by a fic-
tional character from the world of Destiny, and spoke as if she existed in the real
world [104]. Note how this is similar to wrestling and other long-form transmedia
storytelling: The specific examples of how the diegesis breaks is different but it
is still a fundamental aspect that the dual-diegetic elements arise because of the
perennial nature—specifically the temporal continuity and universal chronicle—
which causes anything that isn’t temporally continuous or part of the univer-
sal chronicle to be inherently odd. However, the fiction-breaking implications
of respawning villains and recurring events are not necessarily any diegetically
stranger than the incorporation of real-life movies into the fiction of wrestling
[73, Monday Night Raw, November 24, 2008]. And in both, the audience does not
seem to care. This is evident purely through the popularity of these experiences:
If diegetic breaks like these were enough to make the storytelling experience fall
apart, they wouldn’t be as consistently popular and well regarded by their fans
as they are. This diegetic power is also in part because of the inherent affordances
perennial games have to create myths about themselves, through continuity and
universality.
The pseudo-historical nature of these games cause a strange effect to happen
upon the veracity of their events. They are still undoubtedly fictional, yet there
is something true about them. To understand what we mean, here is an example
of an event that happened in Destiny: We shot down the Almighty 11 [22]. This
is true. It did happen. Every player (first author included) who was part of this
can attest to the fact that it happened. Therefore, while the world is fictional, as
Krzywinska said, “we nonetheless do “real” things in that world” [56]. This state-
ment is different from most narrative events as it is not an individual experience:
It didn’t happen separately to each person—yes, each player had their unique
perspective on the event, but the event itself is part of the universal chroni-
cle: Every player who played in the seasonal event helped gather the required
resources, and every player who was in Destiny at 10 AM PST on June 6th,
2020 saw the Almighty explode in the sky [22]. And any player who was not can
only hear the story of it. This is how perennial games become historical; become
mythological: They create events which become myth. The destruction of the
Almighty, the fall of the moon “Dalamud” in Final Fantasy XIV [69,95, Pre-
patch 2.0 and transition to A Realm Reborn], the black hole in Fortnite [41,109,
Oct 13-15, 2019], the opening of the gates of Ahn-Qiraj [13,83,114, Patch 1.9.0]
are all examples of the most mythological of these types of events: They were
big, earth-shattering (literally) events that changed the fabric of their universes.
They are foundational myths upon which their universal chronicles stand. These
myths are kept alive through records and retellings and become myths through
player action.
11
A space station on a collision course with Earth.
Wrestling with Destiny: Storytelling in Perennial Games 247
5 Conclusion
This paper has described and defined the “perennial game”, and above it, the
“perennial experience” as an experience that is perpetual, temporally contin-
uous, and has a universal chronicle, with either a shared fiction or a shared
world. Each perennial game exists in relation and negotiation with the authors,
the game itself, and the audience, with players directly interacting with the
game, whereas the audience (who can and do overlap with the players) experi-
ences it. Perennial games exists within the larger scope of perennial experiences,
and the commonalities between them make comparisons useful. A spectrum of
experiences was presented, showing how various examples of this form mapped
onto authorial and audience control. Understanding all of these experiences as
perennial opens the possibility for us to draw teachings from wrestling, sports,
Reality TV, etc. into video games, as the storytelling findings they have from
their decades of run-time could be useful for these types of games.
A case study of Destiny was presented as an example of how perennial games
tell perennial stories over long periods of time. It shows itself through lasting
change in the world, and through social player engagement within it. Perennial
worlds or fictions become mythological through evocative and detailed histories
and mythologically functioning events. The worlds and stories are both real and
not-real, as they fluently flux into and out of reality, through their mimicry of
(and existence within) reality’s time.
Perennial games have truly exploded within the last 10 years, so there is a
much broader wealth of experiences to analyse and study than ever before. They
span the most popular games on the planet right now, and they show little sign
of slowing down. In this paper we outline their definition to create a common
ground upon which we can begin to disseminate and discuss the possibilities
and pitfalls of these games, by showing how their experiences echo other forms
of media. There are still many questions to be explored within this form both in
terms of how they are experienced (what is the effect of real-time and constant
(re)negotiation of the story space for the audience? How much does time affect
the experience?) and how they are made (how to continue to develop new content
for an experienced audience, while also remaining accessible to newcomers?).
The popularity of these games have demonstrated the power of perennial
storytelling experiences, and are therefore of inherent interest to the interactive
storytelling research community. This paper presents a first step in the direction
of trying to wrestle with this form of games, to help answer the questions of
what they are, how they are experienced, and what we can do with them.
248 B. A. Larsen and E. Carstensdottir
References
1. Adams, T., Adams, Z.: Dwarf fortress (2006)
2. Akihiro Yokoi (WiZ), A.M.B.: Tamagotchi (1996)
3. Aylett, R.: Narrative in virtual environments-towards emergent narrative. In: Pro-
ceedings of the AAAI Fall Symposium on Narrative Intelligence, pp. 83–86 (1999)
4. Barthes, R.: The world of wrestling, chap. 2, pp. 23–33. Duke University Press
(1972)
5. Bartle, R.: Hearts, clubs, diamonds, spades: players who suit muds. J. MUD Res.
1(1), 19 (1996). http://www.mud.co.uk/richard/hcds.htm
6. Bartle, R.A.: Designing Virtual Worlds. New Riders, Indianapolis (2004)
7. BBC, Newman, S., Webber, C.E., Wilson, D.: Doctor Who. Television Series on
BBC (1963–Now)
8. BioWare: Mass effect (2007)
9. Birk, M.V., et al.: The effects of social exclusion on play experience and hos-
tile cognitions in digital games. In: Proceedings of the 2016 CHI Conference on
Human Factors in Computing Systems, CHI 2016, pp. 3007–3019. Association
for Computing Machinery, New York (2016). https://doi.org/10.1145/2858036.
2858061, https://doi-org.oca.ucsc.edu/10.1145/2858036.2858061
10. Blizzard: World of warcraft patch notes: Patch 4.0.3a: the shattering of aze-
roth (2010). https://worldofwarcraft.com/en-us/news/1214741/patch-403a-the-
shattering-of-azeroth-now-live
11. Blizzard: Fall of the lich king ending (youtube video) (2011). https://www.
youtube.com/watch?v=qAIrj Vqdfc. Accessed 17 Jan 2021
12. Blizzard: Death of Varian Wrynn- world of warcraft: Legion (youtube
video) (2016). YouTube “Tom’s Hardware”: https://www.youtube.com/watch?
v=T9Z3eLEpxQ8. Accessed 17 Jan 2021
13. Blizzard Entertainment Ltd.: World of warcraft (2004)
14. Bowey, J.T., Mandryk, R.L.: Those are not the stories you are looking for:
using text prototypes to evaluate game narratives early. In: Proceedings of
the Annual Symposium on Computer-Human Interaction in Play, CHI PLAY
2017, New York, NY, USA, pp. 265–276. Association for Computing Machinery
(2017). https://doi.org/10.1145/3116595.3116636, https://doi-org.oca.ucsc.edu/
10.1145/3116595.3116636
15. Bruni, L.E., Baceviciute, S.: Narrative intelligibility and closure in interactive
systems. In: Koenitz, H., Sezen, T.I., Ferri, G., Haahr, M., Sezen, D., C atak, G.
(eds.) ICIDS 2013. LNCS, vol. 8230, pp. 13–24. Springer, Cham (2013). https://
doi.org/10.1007/978-3-319-02756-2 2
16. Bungie: “chances and choices: enemy of my enemy”. From destiny 2 original
release (no longer playable as of beyond light), November 2020. Destiny 2 Story
Mission. Enemy Captain does not fight the player (2017). Video and transcript
can be found at https://www.ishtar-collective.net/transcripts/quest-chances-
and-choices-enemy-of-my-enemy#mithrax by Isthar Collective. Accessed 17 June
2021
17. Bungie: Mithrax (Misraaks). Non-playable character in Destiny 2 (2017–2021).
https://www.ishtar-collective.net/categories/mithrax by Isthar Collective for
overview. Accessed 17 June 2021
Wrestling with Destiny: Storytelling in Perennial Games 249
18. Bungie: Dialogue from destiny 2 strike: The hollowed lair (added in the forsaken
expansion, 2018). Replayable Strike Mission in Destiny 2 (2018). Dialogue Tran-
scripts available at https://commons.ishtar-collective.net/t/the-hollowed-lair-st
rike/2106, by Isthar Commons. Youtube video of relevant dialogue at https://
www.youtube.com/watch?v=rShQFTvPeBw&list=PLS2hBTtCDufSzVXI2EJM
cP4-BCSACYuy&index=17, posted by Destiny Lore Vault. Both URLs Accessed
16 June 2021
19. Bungie: Truth to power (2018). Lorebook from Destiny 2, hosted on https://www.
ishtar-collective.net/categories/book-truth-to-power
20. Bungie: Bungie vidoc - the moon and beyond (2019). https://www.youtube.com/
watch?v=mEvVKCKdtfw
21. Bungie: “zero hour”. From destiny 2’s “season of the drifter”, May 2019 (no longer
playable as of beyond light (2020). Destiny 2 Repeatable Story mission. Mithrax
is passive NPC who starts the mission, and shows up during to help the player
(2019). Video and transcript can be found at https://www.ishtar-collective.net/
transcripts/quest-fallen-transponder-and-zero-hour-outbreak-perfected by Isthar
Collective. Accessed 17 June 2021
22. Bungie: The crash of the almighty. Live event inside Destiny 2, 10 AM PDT,
June 6th, 2020. First author can report being present at the event (2020).
Second-hand video can be seen at https://www.youtube.com/playlist?list=PLS2
hBTtCDufSxtSjyvzjtXIGOPL5VigE-. Article about the event can be seen at
https://www.polygon.com/2020/6/5/21279660/destiny-2-almighty-event-tower-
damage-emblem-seraphs-wings-crash-date-time Both URLs Accessed 16 June
2021
23. Bungie: Quest: Evacuation. Story mission in Destiny 2: Beyond Light (2020).
Player helps evacuate skiff with refugees to Mithrax’s “House of Light”
(2020). Transcript and video can be seen at https://www.ishtar-collective.net/
transcripts/quest-evacuation-beyond-light by Isthar Collective. Accessed 17 June
2021
24. Bungie: The lost splicer. Opening story mission in Destiny 2: Season of the Splicer
(2021). Mithrax is now a fully voiced and singular character (2021). Videos can
be seen at https://www.youtube.com/watch?v=ty9nRvTE26A&list=PLS2hBT
tCDufQgZAAvRPamiMCkJ-r Pef3&index=2 and https://www.youtube.com/wa
tch?v=ZiteIeg3riY&list=PLS2hBTtCDufQgZAAvRPamiMCkJ-r Pef3&index=3
by Destiny Lore Vault. Accessed 17 June 2021
25. Bungie and Activision: Destiny (2014)
26. Bungie and Activision: Destiny 2 (2017), as of 2019, Activision is not affiliated
with Destiny. https://kotaku.com/bungie-splits-with-activision-1831651740
27. My name is Byf: Youtube channel: “my name is byf” (2014–Now). https://www.
youtube.com/c/MynameisByf/videos
28. Cai, W., Chen, M., Leung, V.C.: Toward gaming as a service. IEEE Internet
Comput. 18(3), 12–18 (2014)
29. Canossa, A., Azadvar, A., Harteveld, C., Drachen, A., Deterding, S.: Influencers
in multiplayer online shooters: evidence of social contagion in playtime and social
play, New York, NY, USA, pp. 1–12. Association for Computing Machinery
(2019). https://doi-org.oca.ucsc.edu/10.1145/3290605.3300489
30. Carroll, A.E.: Identity within destiny: virtual selves in virtual worlds. Master’s
thesis, University of Colorado Denver (2016)
31. CCP Games: Eve Online (2008)
32. Clearwater, D.: What defines video game genre? Thinking about genre study after
the great divide. Loading 5(8), 29–49 (2011)
250 B. A. Larsen and E. Carstensdottir
33. Cole, H., Griffiths, M.D.: Social interactions in massively multiplayer online role-
playing gamers. Cyberpsychol. Behav. 10(4), 575–583 (2007)
34. Cryptic Studios: City of heroes (2004)
35. De Garis, L.: The Logic of Professional Wrestling, chap. 9, pp. 192–213. Duke
University Press (2005)
36. dobby rams: Dreaming city dialogue reddit post (2018). https://www.reddit.com/
r/DestinyLore/comments/9z0gtk/dreaming city dialogue/. Accessed 19 Jan 2021
37. Drachen, A., et al.: Guns and guardians: comparative cluster analysis and behav-
ioral profiling in destiny. In: 2016 IEEE Conference on Computational Intelligence
and Games (CIG), pp. 1–8. IEEE (2016)
38. Drescher, C., Wallner, G., Kriglstein, S., Sifa, R., Drachen, A., Pohl, M.: What
moves players? Visual data exploration of twitter and gameplay data. In: Pro-
ceedings of the 2018 CHI Conference on Human Factors in Computing Systems,
CHI 2018, New York, NY, USA, pp. 1–13. Association for Computing Machinery
(2018). https://doi.org/10.1145/3173574.3174134. https://doi-org.oca.ucsc.edu/
10.1145/3173574.3174134
39. Dubois, L.E., Weststar, J.: Games-as-a-service: conflicted identities on the new
front-line of video game development. New Media Soc. (2021). https://doi.org/
10.1177/1461444821995815
40. Emery, K.M., Reinhard, A.: Trading shovels for controllers: a brief exploration
of the portrayal of archaeology in video games. Public Archaeol. 14(2), 137–149
(2015)
41. Epic Games: Fortnite (2017)
42. Gach, E.: They brought back destiny’s loot cave, but not the loot (2020).
Kotaku article. https://kotaku.com/they-brought-back-the-loot-cave-but-not-
the-loot-1845638914. Accessed 30 Apr 2021
43. Gilliam, R.: The history of the league of legends (2018). https://www.
riftherald.com/2018/4/13/17231028/history-lol-league-of-legends-institute-of-
war-summoners. Accessed 17 Jan 2021
44. Good, O.S.: Here’s how to find destiny’s ‘loot cave’ and plunder it for endless
riches (2014). Polygon article. https://www.polygon.com/2014/9/21/6760715/
destiny-loot-cave-engrams-farming. Accessed 30 Apr 2021
45. Gursoy, A.: “We’re excited to chart this unknown territory together”: storytelling
strategies in patch notes documents. In: Proceedings of DiGRA 2020 (2020)
46. Isthar Collective: Ishtar collective’s last wish cinematic transcript. Transcript of
in-game cinematic displayed upon the first canonical completion of the “Last
Wish” Raid, part of the Forsaken campaign (2019). https://www.ishtar-collective.
net/transcripts/last-wish-cinematic?highlight=curse. Accessed 19 Jan 2021
47. Isthar Collective: Ishtar collective’s visions of light transcript. Transcript of in-
game mission during the original Forsaken campaign. The important quote is
“Ghost: The Traveler’s been alive for a year. Why has it waited so long to
speak to us?” (2019). https://www.ishtar-collective.net/transcripts/visions-of-
light. Accessed 19 Jan 2021
48. Jedruszczak, B.: Identity creation and world-building through discourse in video
game narratives. Master’s thesis, Syracuse University (2016)
49. Jenkins, H.: ‘Never Trust a Snake’: WWF Wrestling as Masculine Melodrama,
chap. 3, pp. 33–67. Duke University Press (2005 (1997))
50. Jenkins, H.: Convergence Culture. New York University Press, New York (2006)
Wrestling with Destiny: Storytelling in Perennial Games 251
51. Joshi, R., et al.: A team based player versus player recommender systems frame-
work for player improvement. In: Proceedings of the Australasian Computer Sci-
ence Week Multiconference, ACSW 2019. Association for Computing Machinery,
New York, NY, USA (2019). https://doi.org/10.1145/3290688.3290750
52. Karavatos, A.: “What is Your DPS, Hero?”: Ludonarrative dissonance and player
perception of story and mechanics in MMORPGs. Master’s thesis, University of
Skövde (2017)
53. Klastrup, L.: A poetics of virtual worlds. In: Digital Arts and Culture. RMIT
University (2003)
54. Kleinman, E., Carstensdottir, E., Seif El-Nasr, M.: A model for analyzing diegesis
in digital narrative games. In: Cardona-Rivera, R.E., Sullivan, A., Young, R.M.
(eds.) ICIDS 2019. LNCS, vol. 11869, pp. 8–21. Springer, Cham (2019). https://
doi.org/10.1007/978-3-030-33894-7 2
55. Kokkinakis, A.V., Cowling, P.I., Drachen, A., Wade, A.R.: Exploring the rela-
tionship between video game expertise and fluid intelligence. PLoS One 12(11),
1–15 (2017). https://doi.org/10.1371/journal.pone.0186621
56. Krzywinska, T.: Blood scythes, festivals, quests, and backstories: world creation
and rhetorics of myth in world of warcraft. Games Cult. 1(4), 383–396 (2006)
57. Lee, Y., Armstrong, T., Spataro, J.: Destiny character-animation system and
lessons learned (full text not available). In: ACM SIGGRAPH 2014 Courses,
SIGGRAPH 2014, New York, NY, USA. Association for Computing Machinery
(2014). https://doi.org/10.1145/2614028.2615458, https://doi-org.oca.ucsc.edu/
10.1145/2614028.2615458
58. Linden Lab: Second Life (2003)
59. Lohmann, B.: Storytelling in massively multiplayer online games. Master’s thesis,
IT University of Copenhagen (2008)
60. Louchart, S., Aylett, R.: Solving the narrative paradox in VEs – lessons from
RPGs. In: Rist, T., Aylett, R.S., Ballin, D., Rickel, J. (eds.) IVA 2003. LNCS
(LNAI), vol. 2792, pp. 244–248. Springer, Heidelberg (2003). https://doi.org/10.
1007/978-3-540-39396-2 41
61. Manning, C., Ashwell, S.K.: The gods play dice: procedural content in blase-
ball (2021). Medium Article https://medium.com/the-game-bland/the-gods-
play-dice-procedural-content-in-blaseball-a9487dec0310. Accessed 16 June 2021
62. Marshall, C.: World of warcraft’s latest story campaign has ignited the fanbase
(2018). https://www.polygon.com/2018/8/1/17639398/world-of-warcraft-battle-
for-azeroth-sylvanas-story-war-thorns-campaign
63. Mazer, S.: ‘Real’Wrestling/‘Real’Life, chap. 4, pp. 67–88. Duke University Press
(2005)
64. McArthur, V., Teather, R.J., Jenson, J.: The avatar affordances framework: map-
ping affordances and design trends in character creation interfaces. In: Proceed-
ings of the 2015 Annual Symposium on Computer-Human Interaction in Play,
CHI PLAY 2015, New York, NY, USA, pp. 231–240. Association for Comput-
ing Machinery (2015). https://doi.org/10.1145/2793107.2793121, https://doi-org.
oca.ucsc.edu/10.1145/2793107.2793121
65. Miller, W.: The role of spontaneous order and choice in video games: a case study
of destiny. Available at SSRN 2953791 (2016)
252 B. A. Larsen and E. Carstensdottir
66. Misztal, S., Carbonell, G., Schild, J.: Visual delegates - enhancing player per-
ception by visually delegating player character sensation. In: Proceedings of
the Annual Symposium on Computer-Human Interaction in Play, CHI PLAY
2020, New York, NY, USA, pp. 386–399. Association for Computing Machinery
(2020). https://doi.org/10.1145/3410404.3414238, https://doi-org.oca.ucsc.edu/
10.1145/3410404.3414238
67. Mojang: Minecraft (2011)
68. Nardi, B., Harris, J.: Strangers and friends: collaborative play in world of warcraft.
In: Proceedings of the 2006 20th Anniversary Conference on Computer Supported
Cooperative Work, pp. 149–158 (2006)
69. Nelva, G.: The story of how final fantasy XIV players marched together to the
battle that ended the game (2019). https://twinfinite.net/2019/08/final-fantasy-
xiv-anniversary-march/. Accessed 17 Jan 2021
70. Nitsche, M.: Mapping time in video games. In: DiGRA Conference (2007)
71. O’Connor, A.: Destiny 2’s groundhog day curse is a neat touch that’s easy to
miss. Rock Paper Shotgun (2019). https://www.rockpapershotgun.com/2019/01/
31/destiny-2-the-dreaming-city-curse-cycle/. Accessed 19 Jan 2021
72. Oglesby, B.: Daniel Bryan & the negotiation of kayfabe in professional wrestling.
Master’s thesis, University of South Florida (2017)
73. Oliva, C., Calleja, G.: Fake rules, real fiction: professional wrestling and
videogames. DiGRA (2009)
74. OReilly, D.: Mountain (2014)
75. Pearce, C.: Communities of Play: Emergent Cultures in Multiplayer Games and
Virtual Worlds. MIT Press, Cambridge (2011)
76. Perry, R., et al.: Online-only friends, real-life friends or strangers? Differential
associations with passion and social capital in video game play. Comput. Hum.
Behav. 79, 202–210 (2018)
77. Pirker, J., Rattinger, A., Drachen, A., Sifa, R.: Analyzing player networks in
destiny. Entertain. Comput. 25, 71–83 (2018)
78. Ravari, Y.N., Spronck, P., Sifa, R., Drachen, A.: Predicting victory in a hybrid
online competitive game: the case of destiny. In: Proceedings of the AAAI Con-
ference on Artificial Intelligence and Interactive Digital Entertainment, vol. 13
(2017)
79. Reinhard, A.: Archaeogaming: An Introduction to Archaeology in and of Video
Games. Berghahn Books, New York (2018)
80. Rimington, E., Blount, T.: Lore v. representation: narrative communication of
power with regard to gender in league of legends. In: CEUR Workshop Proceed-
ings, vol. 1628, pp. 1–5 (2016)
81. Riot Games: League of Legends (2009)
82. Riot Games: League of legends universe website (2021). https://universe.
leagueoflegends.com/en US/. Accessed 17 Jan 2021
83. Rousseau, S.: ‘world of warcraft’ recreates one of its biggest events, including bro-
ken servers (2020). https://www.vice.com/en/article/ep4wqj/world-of-warcraft-
recreates-one-of-its-biggest-events-including-broken-servers. Accessed 17 Jan
2021
84. Ryan, J.: Curating simulated storyworlds. Ph.D. thesis, UC Santa Cruz (2018)
85. Ryan, M.L.: Transmedia storytelling: industry buzzword or new narrative expe-
rience? Storyworlds: J. Narrat. Stud. 7(2), 1–19 (2015)
Wrestling with Destiny: Storytelling in Perennial Games 253
86. Schaekermann, M., et al.: Curiously motivated: profiling curiosity with self-reports
and behaviour metrics in the game “destiny”. In: Proceedings of the Annual
Symposium on Computer-Human Interaction in Play. CHI PLAY 2017, pp. 143–
156. Association for Computing Machinery, New York (2017).https://doi.org/10.
1145/3116595.3116603, https://doi-org.oca.ucsc.edu/10.1145/3116595.3116603
87. Schiller, M.H., et al.: Inside the group: investigating social structures in player
groups and their influence on activity. IEEE Trans. Games 11(4), 416–425 (2018)
88. Schreier, J.: Top video game companies won’t stop talking about ‘games as a ser-
vice’ (2017). https://kotaku.com/top-video-game-companies-wont-stop-talking-
about-games-1795663927. Accessed 17 Jan 2021
89. Schwind, V., Wolf, K., Henze, N., Korn, O.: Determining the characteristics of pre-
ferred virtual faces using an avatar generator. In: Proceedings of the 2015 Annual
Symposium on Computer-Human Interaction in Play. CHI PLAY 2015, pp. 221–
230. Association for Computing Machinery, New York (2015). https://doi.org/10.
1145/2793107.2793116, https://doi-org.oca.ucsc.edu/10.1145/2793107.2793116
90. Sciarretta, E.: The use of social media as part of a transmedia storytelling strategy
in WWE’s professional wrestling. In: Meiselwitz, G. (ed.) HCII 2019. LNCS, vol.
11578, pp. 556–570. Springer, Cham (2019). https://doi.org/10.1007/978-3-030-
21902-4 39
91. Sehmby, D.S.: Professional wrestling, whooo!: a cultural con, an athletic dramatic
narrative, and a haven for rebel heroes. Master’s thesis, University of Alberta,
Edmonton (2000)
92. Sifa, R., Drachen, A., Bauckhage, C.: Profiling in games: understanding behavior
from telemetry. In: Social Interactions in Virtual Worlds: An Interdisciplinary
Perspective (2018)
93. Sifa, R., et al.: Controlling the crucible: a novel PvP recommender systems frame-
work for destiny. In: Proceedings of the Australasian Computer Science Week
Multiconference, pp. 1–10. ACM (2018)
94. Sown, H.H., Jeon, S.K.: Storytelling characteristics of the MOBA with a com-
parison to the MMORPG: focused on< league of legends>. Asia Digit. Art Des.
2013(12), 430–433 (2013)
95. Square Enix: Final Fantasy XIV (2013)
96. Stafford, T., Devlin, S., Sifa, R., Drachen, A.: Exploration and skill acquisition
in a major online game. In: The 39th Annual Meeting of the Cognitive Science
Society (CogSci), York (2017)
97. Statt, N.: Fortnite is now one of the biggest games ever with 350 million players.
The Verge (2020). https://www.theverge.com/2020/5/6/21249497/fortnite-350-
million-registered-players-hours-played-april. Accessed 21 Jan 2021
98. Steamcharts: Steamcharts’ top games by current players (2021). https://
steamcharts.com/top. Accessed 17 Jan 2021
99. Tamassia, M., Raffe, W., Sifa, R., Drachen, A., Zambetta, F., Hitchens, M.: Pre-
dicting player churn in destiny: a hidden Markov models approach to predicting
player departure in a major online game. In: 2016 IEEE Conference on Compu-
tational Intelligence and Games (CIG), pp. 1–8. IEEE (2016)
100. Tassi, P.: Destiny 2’s ‘you had to be there’ philosophy is turning off players. Forbes
article (2020). https://www.forbes.com/sites/paultassi/2020/01/22/destiny-2s-
you-had-to-be-there-philosophy-is-turning-off-players/?sh=6e886c641b3a.
Accessed 25 June 2021
254 B. A. Larsen and E. Carstensdottir
101. Tatarchuk, N.: Character creation pipeline and rendering in destiny (full text not
available). In: ACM SIGGRAPH 2014 Courses. SIGGRAPH 2014. Association
for Computing Machinery, New York (2014). https://doi.org/10.1145/2614028.
2615457, https://doi-org.oca.ucsc.edu/10.1145/2614028.2615457
102. The Game Band: Blaseball (2020). https://www.blaseball.com/
103. Toups, Z.O., Hammer, J., Hamilton, W.A., Jarrah, A., Graves, W., Garretson, O.:
A framework for cooperative communication game mechanics from grounded the-
ory. In: Proceedings of the First ACM SIGCHI Annual Symposium on Computer-
Human Interaction in Play. CHI PLAY 2014, pp. 257–266. Association for Com-
puting Machinery, New York (2014). https://doi.org/10.1145/2658537.2658681,
https://doi-org.oca.ucsc.edu/10.1145/2658537.2658681
104. Valle, C.G.D.: ‘destiny 2’ Twitter, Facebook taken over by villain caiatl; zavala
storyline cutscene leaked [video] (2021). Techtimes article https://www.techtimes.
com/articles/257864/20210309/destiny-2-twitter-facebook-taken-over-villain-
caiatl-zavala-storyline.htm. Article has images of tweets in question with cor-
responding avatar icon and name (the acconut has since been rebranded back
to ”Destiny 2” https://twitter.com/DestinyTheGame. Also see Reddit thread
from the time https://www.reddit.com/r/DestinyTheGame/comments/m0kggb/
destinythegames twitter profile has changed his/ and community conversation
about it at https://twitter.com/MyNameIsByf/status/1369991874539311104
105. McMahon, V., World Wrestling Entertainment, Inc.: World wrestling entertain-
ment, Inc (WWE) (1980 (As Titan Sports), 1998 (As WWF), 1999 (as WWE)).
https://www.wwe.com/
106. Wallner, G., Kriglstein, S.: Introducing planet: a tool for visualizing player com-
munities. In: Proceedings of the 13th International Conference on the Foundations
of Digital Games. FDG 2018, Association for Computing Machinery, New York
(2018). https://doi.org/10.1145/3235765.3235808, https://doi-org.oca.ucsc.edu/
10.1145/3235765.3235808
107. Wallner, G., Kriglstein, S., Drachen, A.: Tweeting your destiny: profiling users
in the twitter landscape around an online game. In: 2019 IEEE Conference on
Games (CoG), pp. 1–8. IEEE (2019)
108. Wallner, G., et al.: Beyond the individual: understanding social structures of an
online player matchmaking website. Entertain. Comput. 30, 100284 (2019)
109. Webster, A.: With the apocalyptic black hole event, Fortnite’s storytelling
has grown increasingly ambitious (2019). https://www.theverge.com/2019/10/
14/20913541/fortnite-game-shutdown-black-hole-event-storytelling. Accessed 17
Jan 2021
110. Whalen, Z.: Game/genre: a critique of generic formulas in video games in the
context of ”the real”. Works Days 22(43/44), 289–303 (2004)
111. Williams, M.: Rise of the lifestyle game: Gaming as your second job
(2014). https://www.usgamer.net/articles/rise-of-the-lifestyle-game-gaming-as-
your-second-job. Accessed 17 Jan 2021
112. Wizards of the Coast: Magic: The Gathering (1993)
113. Zagal, J.P., Mateas, M.: Time in video games: a survey and analysis. Simul.
Gaming 41(6), 844–868 (2010)
114. Ziebart, A.: WoW archivist: The gates of Ahn’Qiraj (2011). https://www.
engadget.com/2011-04-19-wow-archivist-the-gates-of-ahnqiraj.html. Accessed 17
Jan 2021
Blabbeur - An Accessible Text
Generation Authoring System for Unity
1
Blabbeur is available here: https://bitbucket.org/lablablab/blabbeur/.
c Springer Nature Switzerland AG 2021
A. Mitchell and M. Vosmeer (Eds.): ICIDS 2021, LNCS 13138, pp. 255–259, 2021.
https://doi.org/10.1007/978-3-030-92300-6_23
256 J. Lessard and Q. Kybartas
2 Blabbeur
Fig. 1. Excerpt from a Blabbeur grammar file used to generate accident descriptions.
defined by separating them with semicolons. In this case, the [generic accident]
symbol can randomly resolve either as “I had a freak accident”, “I messed up
big time”, or “I was minding...”. Symbols can be nested ad infinitum.
Conditional Expressions – Authors can make a symbol conditional by preceding
it with an expression in curly braces. Only the symbols whose conditions are met
will be considered as possible outputs at the moment of generation. In Fig. 1, for
example, the author redirects the resolution of [description] by checking what
kind of accident they are dealing with. If the “TypeOfAccident” variable is set
to “agriculture”, only the second symbol will be considered valid, thus leading to
the resolution of [agriculture]. Currently, the following operators are supported:
!=, ==, <, >, &&, ||
Variables – Variables that have been passed along the generation request can
be used within conditional expressions or surfaced directly within the text. This
is done by placing a condition within angle brackets. In Fig. 1, <Damage> and
<victim.healthpoints> will be replaced with the variables’ values. Values can be
numerical, strings, enumerators, or conditionals.
Comments – Any line preceded with “//” are disregarded as comments.
System Communication – The Blabbeur system is implemented as a singleton
and is accessible at any point of the project’s code. Blabbeur communication is
done through requests, passing the name of the top-level symbol of the desired
grammar, as well as a custom Blabbeur Object containing an arbitrary set of
variables describing the state of the request. The generated text is returned as
a string.
Blabbeur Object – A Blabbeur object is a <string,value> dictionary which
matches a variable string from the grammar to its appropriate value. It is further
possible to “nest” blabbeur objects, e.g. the “victim” shown in Fig. 2, is assigned
to the actor object, which then contains its own dictionary, e.g. name, age, etc.
258 J. Lessard and Q. Kybartas
3 Conclusion
We have been using this system for months now and it has proven easy to use
and efficient. With less than an hour of training, authors can immediately begin
Blabbeur - An Accessible Text Generation Authoring System for Unity 259
producing new content. We already have more than 40 grammars in our system
and they are a key component in communicating with players. As we stress test
Blabbeur, we are also noting requests and comments from authors to increase
usability, focusing on elements such as error feedback, and accessing another
grammar from within a grammar.
References
1. Balousek, M.R.F.: Twincery. https://github.com/mrfb/twinecery. Accessed 07 June
2021
2. Buckenham, G.: Cheap bots done quick. Accessed 07 June 2021
3. Compton, K., Kybartas, B., Mateas, M.: Tracery: an author-focused generative text
tool. In: Schoenau-Fog, H., Bruni, L.E., Louchart, S., Baceviciute, S. (eds.) ICIDS
2015. LNCS, vol. 9445, pp. 154–161. Springer, Cham (2015). https://doi.org/10.
1007/978-3-319-27036-4 14
4. Horswill, I.: Generative text using classical nondeterminism. In: Joint Proceedings
of the AIIDE 2020 Workshops, Worcester, MA (2020)
5. Lessard, J., Brunelle-Leclerc, E., Gottschalk, T., Jetté-Léger, M.A., Prouveur, O.,
Tan, C.: Striving for author-friendly procedural dialogue generation. In: Proceedings
of the International Conference on the Foundations of Digital Games - FDG ’17.
ACM Press (2017). https://doi.org/10.1145/3102071.3116219
6. Ryan, J., Seither, E., Mateas, M., Wardrip-Fruin, N.: Expressionist: an authoring
tool for in-game text generation. In: Nack, F., Gordon, A.S. (eds.) ICIDS 2016.
LNCS, vol. 10045, pp. 221–233. Springer, Cham (2016). https://doi.org/10.1007/
978-3-319-48279-8 20
Enhancing Interactivity in Propp-Based
Narrative Generation
1 Introduction
Since the emergence of computational storytelling, Vladimir Propp’s Morphology
of the Folktale [14] has received great consideration (see, e.g., [4,5,9]) because of
its quite formal nature. In particular, in the context of interactive storytelling
(IS), the early two thousands saw a number of Propp-based systems [3,11,13,16].
After that, however, a phase of decline occurred; in fact a recent overview paper
[8] does not mention any later Propp-based IS system. [18] focuses on the issues
that Propp-based systems suffer from and claims that Propp’s morphology is
inherently unable to provide adequate interactivity. After presenting some back-
ground notions (Sect. 2), we recall the main criticisms of [18] and others (Sect. 3)
and suggest some adaptations to Propp’s morphology aiming at enhancing inter-
activity of Propp-based IS systems (Sect. 4).
player’s input influences the selection of a specific function; (ii) at the plot-level,
it affects the ordering of the function sequence. More specifically, a condition for
interactivity is that the player’s decisions are meaningfully connected to their
outcome [15]. We define a meaningful decision as an affordance [10] that makes it
obvious to the player that the action he takes is a significant intervention on the
function-level and/or on the plot-level. This implies that the player can sensibly
reason about the decision’s consequences both before and after he makes it.
4 An Adapted Morphology
4.1 Negative Reaction Functions
Many of Propp’s functions come in an action/reaction pair, such as the vic-
tory following a struggle. In Propp’s morphology, given one action, only one
reaction function is possible, e.g., the hero always wins a fight. A negative func-
tion is a Propp’s function with its outcome reversed. In the running example of
Table 1, showing an interaction between a player and an ideal IS system based
on our adapted morphology, we introduce, e.g., a defeat function where the hero
(Siegfried) loses the fight against the dragon. Propp himself makes sporadic use
of negative functions in examples, and [7] adopts only a few negative functions
such that the story ends when they occur. We propose to use the negative func-
tions in a systematic way in order to augment the influence of player’s decisions.
Our proposed set of negative reaction functions appears in Table 2.
262 L. Mienhardt and M. Volpe
Table 1. Example of an interaction between a player (P) and an ideal IS system (N).
In [14], Propp suggests that moves can be combined; however he does not inte-
grate a method for that in his morphology. We plan to provide a comprehensive
formalization of move combinations. The basic idea consists in allowing one move
to replace, and thereby expand, one or more functions occurring in another move.
This works especially well for action/reaction pairs, where we can create a new
move with the desired outcome of the reaction function as a lack function. The
liquidation of this lack, if successful, equals a positive outcome of the reaction
function and, if not successful, equals the negative outcome. In Table 1, a replace-
ment move for a map search substitutes the pair donor’s test/hero reaction. By
emphasizing an aspect of the story, a replacement makes it obvious to the player
that his success or failure will have consequences on the plot development. It
also allows the player to have different ways to success, as he is not restricted to
pass the donor’s test to receive the map, but can attack the donor afterwards.
As a special case, replacing the liquidation of misfortune function can have a
slightly different form, to which give the name of complication move. It is based
on the intensification method mentioned in [12] and [10]. To create such an
intensification, we reincorporate [17] the events of a previous move into the next
one. I.e., instead of resolving the first move, we replace the liquidation function
with a second move that uses a further villainy and intensifies the hero’s struggle.
264 L. Mienhardt and M. Volpe
We then release the tension in a climax point when both villainies are resolved.
In our example, Siegfried loses the fight with the dragon. This defeat is built
upon in the next move and consequently he is imprisoned by the dragon. A
causal relation between the two moves is created and transforms the decision to
fight the dragon into a meaningful decision.
4.3 Flashbacks
In order to avoid imminent dead-ends, we propose to use flashbacks. As [6]
points out, a flashback in IS has to be interactive. In our view, this means that
the player, when taken back in time, can either find a solution to the dead-end or
not. In Table 1, Siegfried faces a dragon, but he misses a magical sword to defeat
him. We take Siegfried back to his village before he sets out for the dragon and
give him a chance to get a sword. Flashbacks allow us to deal with dead-ends
only when they occur and to not restrict player decisions beforehand.
5 Conclusion
In this paper we recalled the criticism of [18] and others with respect to the use
of Propp’s morphology in the context of IS systems. The problems that we con-
sider harmful to interactivity are the lack of influential decisions and the rigidity
of the morphology when it comes to plot-level changes. For both problems we
provide ideas on how they can be resolved. In particular, we extend the mor-
phology by negative versions of the original functions in order to allow different
reactions to player decisions and we give such decisions plot-level relevance by
embedding additional moves into the existing story line. We also suggest to use
the narrative device of flashbacks to overcome dead-ends that might occur due
to player interactions. The approach presented in this paper is still at a prelimi-
nary stage. We are currently working on its comprehensive formalization, which
we believe could be used as a foundation for future IS system implementations.
References
1. Barros, L.M., Musse, S.R.: Towards consistency in interactive storytelling: tension
arcs and dead-ends. Comput. Entertain. 6(3), 1–17 (2008)
2. Cameron, A.: Dissimulation: Illusions of interactivity. Millenium Film J. 28 (1995).
http://mfj-online.org/journalPages/MFJ28/Dissimulations.html
3. Fairclough, C.: Story games and the OPIATE system: using case-based planning
for structuring plots with an expert story director agent and enacting them in a
socially simulated game world. Ph.D. thesis, University of Dublin, Trinity College
(2004)
4. Gervás, P.: Propp’s morphology of the folk tale as a grammar for generation.
In: Finlayson, M.A., Fisseni, B., Löwe, B., Meister, J.C. (eds.) 2013 Workshop
on Computational Models of Narrative, CMN 2013, August 4–6, 2013, Hamburg,
Germany, OASICS, vol. 32, pp. 106–122. Schloss Dagstuhl - Leibniz-Zentrum für
Informatik (2013)
Enhancing Interactivity in Propp-Based Narrative Generation 265
5. Gervás, P.: Computational drafting of plot structures for Russian folk tales. Cogn.
Comput. 8(2), 187–203 (2016). https://doi.org/10.1007/s12559-015-9338-8
6. Guy, O., Champagnat, R.: Flashback in interactive storytelling. In: Nijholt, A.,
Romão, T., Reidsma, D. (eds.) ACE 2012. LNCS, vol. 7624, pp. 246–261. Springer,
Heidelberg (2012). https://doi.org/10.1007/978-3-642-34292-9 17
7. Hartmann, K., Hartmann, S., Feustel, M.: Motif definition and classification to
structure non-linear plots and to control the narrative flow in interactive dramas.
In: Subsol, G. (ed.) ICVS 2005. LNCS, vol. 3805, pp. 158–167. Springer, Heidelberg
(2005). https://doi.org/10.1007/11590361 18
8. Kybartas, B.A., Bidarra, R.: A survey on story generation techniques for author-
ing computational narratives. IEEE Trans. Comput. Intell. AI Games 9, 239–253
(2017)
9. Lakoff, G.P.: Structural complexity in fairy tales. Study Man 1, 128–150 (1972)
10. Mateas, A.S.M.: The game design reader: a rules of play anthology. In: Interaction
and Narrative, pp. 642–669. MIT Press (2006)
11. Machado, I., Prada, R., Paiva, A.: Bringing drama into a virtual stage. In: CVE
’00 (2000)
12. Murray, J.: Hamlet on the Holodeck: The Future of Narrative in Cyberspace. MIT
Press, Cambridge (1997)
13. Peinado, F., Gervás, P., et al.: Transferring game mastering laws to interactive
digital storytelling. In: Göbel, S. (ed.) TIDSE 2004. LNCS, vol. 3105, pp. 48–54.
Springer, Heidelberg (2004). https://doi.org/10.1007/978-3-540-27797-2 7
14. Propp, V.: Morphology of the Folktale. University of Texas Press, Austin (1968)
15. Salen, K., Zimmerman, E.: Rules of Play: Game Design Fundamentals. MIT Press,
Cambridge (2003)
16. Spierling, U., Grasbon, D., Braun, N., Iurgel, I.: Setting the scene: playing digital
director in interactive storytelling and creation. Comput. Graph. 26, 31–44 (2002)
17. Tomaszewski, Z.: On the use of reincorporation in interactive drama. In: Intelligent
Narrative Technologies (2011)
18. Tomaszewski, Z., Binsted, K.: The limitations of a Propp-based approach to inter-
active drama. In: AAAI Fall Symposium: Intelligent Narrative Technologies (2007)
The Applicability of Greimassian Semiotics
to Meaningful Procedural Quest Generation
1 Introduction
Procedural Content Generation (PCG) has long since been used as a feature of digital
games, with varying degrees of success. In her historical analysis of PCG, Smith observed
that by delegating the creation of game content to computer systems, “any ability for
providing meaning in the generated characters or consideration for player experience
was lost” [1]. She added that it is “harder to create meaningful content or to understand
the qualities of generated content in terms of player experience” [1] and highlighted the
detrimental effects that using randomness in the generation of encounters, monsters, and
items can have on the players’ experience of a game.
We believe that Smith’s call for meaning refers to the perceived absence of a theme
in procedurally generated content. This idea of theme corresponds to what Howard
called “meaningful action” [2], or an allegorical set of correspondences where game
players find statements about the real world by interacting with a fictional world. Thus,
we raise the question: How can a procedural quest generator provide such a meaning
to its generated stories? To answer it, we turn to the semiotics proposed by Algirdas J.
Greimas [3] as a tool that an algorithm might use to generate quests that mimic Howard’s
meaningful action.
We propose that Greimas’s method of analysis offers a compelling precursor for
an envisioned quest generation algorithm that prioritizes meaning. We believe this for
the following reasons. First, it formally describes the concept of theme as a “relation
of various units of the signified distributed throughout the length of the story” [4] that
follows concrete, procedural rules. Second, Greimas’s equation has been applied as a
method to describe the theme of a complete story and is able to describe and predict
thematically relevant events in a story [5]. Third, it presents a model of narrative grammar
that focuses on theme [6] as well as a way to establish how actions performed by story
actants [7] are related to the story’s theme.
2 Related Work
Among the most influential works to discuss the idea of “meaning” within interactive
narratives are those of Aarseth [8] and Bogost [9]. Aarseth created a conceptual toolset
to understand texts using a function-oriented perspective, and studied them as a machine
for the production and consumption of signs. Aarseth’s work observes elements of texts
and the functions that they serve in the process of communication. The study created
a compelling basis for an ontology of the text, but it does not discuss the processes of
meaning production at an authorial level. In contrast, Bogost’s unit analysis seeks to
assist critics in finding the “discrete meaning-making in texts of all kinds.”
Bogost’s approach is aimed at the critical study of interactive texts. He makes the
claim that critics and creators work with similar tools. In his view, the similarity of these
tools teaches us to read both technology-based works and non-technology-based works
“from the single perspective of their shared procedurality” [9]. This view supports our
position that a detailed analysis of meaning is a precursor of the procedural generation
of meaning.
Eladhari applied the semiotics of Algirdas Greimas and others to the study of mean-
ing production in videogames [10]. Eladhari focused her application of Greimas as a
“conceptualization that breaks down the parts of a story into force fields that make it
possible for the narrative to come into existence” [10]. These “force fields” refer to the
devices of meaning production and the contextualization of actions and events. Eladhari
further wrote that the analysis of the dynamic elements (actions) and static (characters
and settings) helps us observe the semantic syntax which gives a game its meaning. The
view is based on Budaniekiewicz’s syntactic study of action [11], which presupposes
the modal, i.e., motivating factors such as wants, goals and plans, as an antecedent state
of the actualization of actions.
In what is perhaps a more relevant study for interactive computing Yu et al. [12] use
Greimas’s actantial model as a tool in computer-based narrative by studying character
interactions to classify them ontologically by their relationship to the story’s subject.
They conclude that continued tests would further evidence the similarity between human
and system identifications of characters as subjects, helpers, or opponents. Nonetheless,
we find that these studies, originally conceived as tools of analysis, could contribute to
interactive narrative design.
Szilas used elements of structuralism, including Greimas’s canonical narrative
schemata in his study of the narrative act. He described this act as “a type of meta-action
in which the embedded action is one of the core actions of the story” and stated that such
acts “constitute the main sequence (or plot) in the story” [13]. His system, IDtension
[14, 15], works by organizing concrete actions through generated meta-actions. Szila’s
268 L. F. T. Meza and D. Thue
many things unsaid and a large number of open problems”. One of them is an apparent
difficulty to identify functions that allow stories to end in a satisfactory manner; another
is the difficulty in recognizing dependencies between non-consecutive story events [30].
These approaches attempt to add meaning to stories by arranging a sequence of
events following a restrictive formula which only allows permutations at critical points,
while the rest is determined with constrained randomness. This allows only for limited
variations within the same narrative structure, and do not present particularly meaningful
opportunities to engage with a thematic message more than once. To this end, we propose
the further study of Greimassian semiotics.
Greimas’s model attempts to explain how the human mind constructs complex cul-
tural objects [5], starting with simple elements and following a constrained trajectory. In
this section, we will outline this process with an example analysis to introduce the terms
and conceptual tools used by Greimas. This will clarify our discussion section, which
explains how these tools might help an algorithm assemble meaningful computer game
quests.
Describing Greimassian semiotics is best done with examples. Greimas worked with
myths and folktales, whereas Budnikiewicz [11] worked with West’s Miss Lonelyhearts
(a US literature classic), and Hébert [31] worked with biblical stories, Greek classics,
and fairy tales. In our sample analysis, we work with Monterroso’s microstory “The
Dinosaur”; which we will use as a running example throughout the text. The story,
quoted in its entirety is: “Cuando despertó, el dinosaurio seguía allí” [32]. Grossman
translated it as: “When he awoke, the dinosaur was still there” [33]. However, we will not
use this translation given that it ignores the null-subject quality of the original. Instead,
we will work with a more technical (though perhaps less appealing) translation:
generated by a given semantic micro universe” [6]. The second structured stage is surface
structure, which uses the elements described in the deep structure to define the characters,
actions, happenings, and settings that are susceptible to manifestation by means of a
semiotic grammar. The structure of manifestation is the third stage of the process. It
produces the text that is visible to interactors. This last stage is beyond the scope of this
paper, as it would only be observable in a finished game. It is nonetheless useful to be
aware of, as we use it as the starting point of the example analyses that follow.
Our translation in Table 1 attempts to reflect word by word the manifest structure that
Monterroso published in 1959. As the process of translation shows, the ideas: (1) some-
one awakes, and (2) they realize a dinosaur remains at a known place; are independent of
the manifest structure used (such as the original text, or our translation or Monterroso’s
original). Greimas describes the surface structure as a stage in which actants (anthropo-
morphized performers) and their actions follow a syntax to create a meaningful story.
By contrast, the deep structure (which he also called elementary morphology) describes
the thematic significance of story actions or happenings.
but this is not always required; (d) could remain as non-awareness. Contrary terms are
calculated as follows:
d = −a c = −b (1)
(g) and (h) are called deixes by Greimas. They refer to a micro-ontology of actants
(including characters, settings, and events) determined by their relationship to the ele-
mentary concepts of a given story. In “The Dinosaur”, we might classify actants as
belonging to the set (g) because they cause or bring about awareness or absence. On the
other hand, actants organized in (h) will cause or bring to attention the concept either
presence or unawareness in opposition to the actants in set (g). Although these are not
discussed here at length, deixes (g) and (h) form the basis of Greimas’ actantial model,
where (g) includes helpers and opponents and (h) includes senders and receivers.
(i) and (j) are the transformative functions. Greimas originally expressed these as
oriented syntactic operations, in the forms f (a) = a → d or f (d ) = d → a and
conversely f (b) = b → c or f (c) = c → b [6]. These describe the possible actions
that actants (g) and (h) can perform to propose either of the elementary concepts at a
particular point of story time. We depart from Greimas toward adding more mathematical
meaning to the notation. We give (a) a value of 1 when our main character is aware. We
can therefore understand a function (i) such as “he/she/it awoke”, as follows:
This equation represents (i) in Fig. 1, and describes the process by which our person
in (g) awakens, or gains awareness, going from a moment where they are unaware and
towards a moment when they are aware. For example, we can say that at the moment of
awakening, a0 has a value of -1 and the story segment he/she/it awoke adds a contribution
value ut 0 = 2. The result is a1 = 1, indicating that our character is now aware, changing
the state of the meaningful variables of Monterroso’s story.
This narrative unit creates a full transformation from (d = 1) unawareness to (a = 1)
awareness. It is therefore a simple narrative unit, consisting of one function (i), that is, one
272 L. F. T. Meza and D. Thue
narrative utterance. More complex narratives follow a similar process, although the trans-
formative functions (either (i) or (j)) follow a more gradual value change, and therefore
consist of several narrative units. These units may be declarative (assigning or restat-
ing values of our elementary concepts) or transformative (carrying out transformative
functions which an author can manipulate to create distinct dramatic effects).
Because “The Dinosaur” is a very short story, narrative units and narrative utterances
are indiscernible in context. The contrast between them can be seen more easily in the
context of two adaptations:
(Y) “He came back from an uneasy slumber. Slowly, he opened his eyes and began
to rise as he yawned lazily. He didn’t immediately scan the room, but when he did;
the dinosaur remained there.”
(Z) “He came to slowly. He opened his eyes and began to rise tentatively. The
wound in his eye made it difficult for him to look around, so he treated it as best
he could. When he was done, he saw that the dinosaur remained there.
In adaptation (Y), the act of waking up takes 5 distinct actions, whereas in (A), it
takes only one. The individual actions in (Y) are considered narrative utterances, which
comprise the general act (and narrative unit) of awakening. Table 2 compares the three
adaptations, showing that the fluctuation of value (a) as difference narrative utterances
add a contribution value following Eq. 2. to compute a new value of (a) of our actant in
(g), the main character who awakens.
We begin Table 2 by dividing it in three sections, corresponding to our analyzed
statements. We then present the narrative unit type describing the act of awakening as a
function of our actant in (g) going along axis (i) in the semiotic square (see Fig. 1) by
having its current (a) value (-1, indicating the subject is unaware) fluctuate towards our
desired (a) value (1, indicating the subject is aware). We represent the fluctuation with
the addition of variable (ut n ), to which we attribute a value representing the degree in
which a given utterance (with number n) contributes to the value of (a). Variable (ut n )
has an arbitrary value that describes the contribution an utterance makes to the value
with which we describe an actant. Longer narrative units will contain more functions
The Applicability of Greimassian Semiotics 273
with lower values of (ut n ). In (A) the act of awakening takes one narrative utterance,
and therefore (ut 1 ) “when he awoke” has its maximum value (2) when the new value
of (a) is calculated. To contrast this, in (Y), we observe a more gradual awakening, and
therefore (ut n ) has lower values that happen in sequence. Note that every utterance in
(Y) has a value in the positive axis, and are therefore conjunctive functions, that is, they
modify the value of (a) so that it fluctuates towards our desired value.
In (Y), the inclusion of narrative utterance 5 “The wound in his eye made it difficult
to look around” is a disjunctive function, where variable (ut 5 ) has a value in the negative
range, causing the value of (a) to fluctuate towards its value at ut 0 . This potentially
describes a story in which the elementary concept (a) has a more interesting development,
where characters must overcome obstacles, or encounter opponents or situations that
propose (a)’s contrasting value (d).
Figure 2 shows the values of elementary concept (a) in our three versions of “The
Dinosaur”. The three narrative units are superimposed in the same chart, where the
vertical axis reflects the value of (a) awareness, and the horizontal axis represents the
number of utterances in each unit.
From this comparison we present the following conclusions. First, narratives can
be understood as the sequence of narrative utterances which arise from both stated and
274 L. F. T. Meza and D. Thue
1.5
1
0.5
0
-0.5 0 1 2 3 4 5 6 7
-1
-1.5
inferred narrative units. Second, narrative units are transformations of the story world,
represented by the value fluctuation of elementary concepts to their contrary values (1,
-1), and these can be experienced in chains of sequential narrative units that have varying
lengths. Furthermore, a narrative unit describes the way in which actants (characters or
settings classified in deixies (g) or (h)) perform functions (i) or (j) causing the values of
elementary concepts to fluctuate.
ability :
(B) the dinosaur Nu1 ability : Person is aware that
remained there b→c bh = bh + ut1 the state of the
ut = −2 dinosaur in (h) can
change from present
to absent
Nu2 a, b a=1 Person becomes
c = −b b=1 aware of the
d = −a c = −1 dinosaur’s (h) state:
d = −1 present
In Table 2, we presented and analyzed the types of narrative units that can occur in a story
by enlisting the resources employed by Monterroso to communicate “The Dinosaur.”
In it we observed we have a single virtual narrative unit, Nu1 , which communicates
the dinosaur’s ability to become absent. It is here that we turn back to the fact that
the original story gives readers little manifest evidence to discern its true meaning.
Although we instinctively appreciate that the dinosaur can become absent, as most
things are, Monterroso does not fully express the virtuality of the value transformation
of the variable (b) presence. The following information is missing:
These unknowns represent questions that are answered by readers and interpreters.
When they write that the story suggests that “our intellectual slumber prevents us from
exacting social change” [34] Ramírez and Toledo identify the dinosaur as an allegorical
figure that represents political authority and assumes that the subject (a figure represent-
ing society) wishes it to become absent. On the other hand, Aguilar’s cinematographic
adaptation of “The Dinosaur” [35] manifests all the narrative units presented in Table
2 by creating characters, including a drug dealer called “The Dinosaur”, a young man
who purchases drugs, and drugs that cause the young man’s unawareness.
The pronounced differences between Ramírez and Toledo’s and Aguilar’s interpre-
tations can be explained by referring to the process that Greimas calls “investment” [7].
This process defines characters that propose the elementary concepts of the narrative
at the level of the aesthetic (with characters and settings), the virtual (defining wants,
knowledge and abilities) and finally the moral (subjective interpretations).
Contrasting these two interpretations shows us that the aesthetic investment and the
virtual investment need not remain the same for a deeper message to be communicated.
The stories remain recognizable as adaptations of “The Dinosaur” because they follow
the same message of contrast. We present this message in Eq. 3.
describe actions and states in terms of numerical values attributed to elementary con-
cepts of the semiotic square; (iii) A description of narrative unit types at the author level
by their function, including the descriptive, the virtual, and the functional; and (iv) A
description of the component parts of narrative units and narrative utterances, including
a demonstration of how they can present modular variations without compromising the
story’s message.
Having studied these conceptual tools, we argue that a generative system might rely
on them to direct happenings, present optional events to users, and provide players with
quest plots where they may select a role to play and that they can invest with a moral
message. We now sketch the operation of a system that follows these principles.
Furthermore, several of the steps of the system we have sketched will be non-trivial
to automate completely, and thus the support of a substantial body of authored and
annotated content will be required. Finding a viable balance between authoring and
automation remains as future work. In the meantime, we hope that our summary of
Greimas’s tools can support further work that seeks to produce interactive narratives
that convey a message – be they hand-authored, generated, or some combination of the
two.
Looking forward, it would be useful to analyze more works of literature, both tra-
ditional and interactive, with the goal of describing the dramatic resources that human
authors have used to communicate their messages through action. This is because more
complex stories will contain more than one message, or the message might not be as
simple as it is in Monterroso’s “The Dinosaur.” We theorize that observing different mes-
sages with this methodology will allow our theorical system to produce more complex
stories with the use of a library of contrasting elementary concepts.
Another benefit further analysis of stories would be the appreciation of narrative
utterance design. This would allow us to better understand the relationship between a
narrative unit’s emotional effect and the value fluctuation of the ut n variable we intro-
duced to measure each utterance’s contribution to the value of our elementary concepts.
We suspect that works within particular story genres (such as comedy, tragedy or sus-
pense) will present similar patterns in how the values of elementary concepts fluctuate
across their narrative utterances.
References
1. Smith, G.: An analog history of procedural content generation. In: FDG (2015)
2. Howard, J.: Quests: Design, Theory and History in Games and Narratives. A K Peters, Ltd.,
Wellesley, MA (2008)
3. Greimas, A.J.: On Meaning. University of Minesota Press, Minneapolis (1987)
4. Greimas, A.J.: Toward a Semiotics of the Natural world, in On Meaning. University of
Minessota Press, Minneapolis, pp. 17–47 (1987)
5. Greimas, A.J.: The Interaction of Semiotic Constraints, in On Meaning. University of
Minessota Press, Minneapolis, pp. 48–62 (1987)
6. Greimas, A.J.: Elements of a Narrative Grammar, in On Meaning. University of Minnesota
Press, Minneapolis, pp. 48–83 (1987)
7. Greimas, A.J.: Actants, Actors, and Figures, in On Meaning, pp. 106–120. University of
Minessota Press, Minneapolis (1987)
8. Aarseth, E.: Cybertext: Perspective on Ergodic Literautre. JHU Press, Baltimore (1997)
9. Bogost, I.: Unit operations: An Approach to Videogame Criticism. MIT press, Cambridge
(2008)
10. Eladhari, M.P.: Re-tellings: the fourth layer of narrative as an instrument for critique. In:
Rouse, R., Koenitz, H., Haahr, M. (eds.) Interactive Storytelling. ICIDS 2018. LNCS, vol.
11318. Springer, Cham (2018). https://doi.org/10.1007/978-3-030-04028-4_5
11. Budniakiewicz, T.: Fundamentals of Story Logic: Introduction to Greimassian Semiotics.
John Benjamins Publishing, Amsterdam/Philadelphia (1992)
12. Yu, H.-Y., Kim, M.-H., Bae, B.-C.: Actantial model-based character role recognition using
emotional flow graph among characters in text stories. J. Internet Comput. Serv. 22(1), 51–63
(2021)
The Applicability of Greimassian Semiotics 279
13. Szilas, N., et al.: The study of narrative acts with and for digital media. Digit. Scholarsh.
Humanit. 35(4), 904–920 (2020)
14. Szilas, N.: IDtension: a narrative engine for interactive drama. In: Proceedings of the Technolo-
gies for Interactive Digital Storytelling and Entertainment (TIDSE), Darmstadt:Fraunhofer
(2003)
15. Szlias, N.: A computational model of an intelligent narrator for interactive narratives. Appl.
Artif. Intell. 21(8), 187–203 (2007)
16. Chatman, S.: Story and Discourse. Cornell University Press, Ithaca (2021)
17. Riedl, M.O., Young, R.M.: Narrative planning: balancing plot and character. J. Artif. Intell.
Res. 39, 217–268 (2010)
18. Shirvani, A., Ware, S.G.: A formalization of emotional planning for strong-story systems.
In: Proceedings of the Sixteenth AAAI Conference on Artificial Intelligence and Interactive
Digital Entertainment (2020)
19. Shirvani, A., Ware, S.G.: A plan-based personality model for story characters. In: Proceedings
of the AAAI Conference on Artificial Intelligence and Interactive Digital Entertainment,
Atlanta, Georgia (2019)
20. Breault, V., Ouellet, S., Davies, J.: Let CONAN tell you a story: procedural quest generation.
Entertain. Comput. 38, 100422 (2021)
21. Propp, V.: Morphology of the Folktale. University of Texas, Austin (2010)
22. Campbell, J.: The Hero with a Thosand Faces. New World Library, Novato (2008)
23. Tearse, B., Wardrip-Fruin, N., Mateas, M.: Minstrel remixed: procedurally generating stories.
In: Proceedings of the Sixth AAAI conference on Artificial Intelligence and Interactive Digital
Entertainment, vol. 5, no. 1 (2010)
24. Brusentsev, A., Hitchens, M., Richards, D.: An investigation of Vladimir Propp’s 31 functions
and 8 broad character types and how they apply to the analysis of video games. In: Proceedings
of the 8th Australasian Conference on Interactive Entertainment: Playing the System (2012)
25. Turner, S.R.: Minstrel: A Computer Model of Creativity and Storytelling. Diss. University of
California, Los Angeles (1993)
26. Grabson, D., Spierling, U., Braun, N.: Setting the scene: playing digital director in interactive
storytelling and creation. Comput. Graph. 26(1), 31–44 (2002)
27. Ogata, T.: Computational and cognitive approaches to narratology from the perspective of nar-
rative generation. In: Ogata, T., Akimoto, T. (eds.) Computational and Cognitive Approaches
to Narratology. IGI Global, Hershey, pp. 1–74 (2016)
28. Gervás, P., Díaz-Agudo, B., Peinado, F., Hervás, R.: Story plot generation based on CBR. In:
Macintosh, A., Ellis, R., Allen, T. (eds.) Applications and Innovations in Intelligent Systems
XII. SGAI 2004. Springer, London (2005). https://doi.org/10.1007/1-84628-103-2_3
29. Sjöström, J.: Morphology of digital narrative: prototyping digital narratives using the theories
of Vladimir Propp. DiVA Archive (2013)
30. Gervás Gómez-Navarro, P.: Reviewing Propp’s story generation procedure in the light of
computational creativity. In: 40th Annual convention of the Society for the Study of Artificial
Intellingence and the Simulation of Behaviour, London, UK (2014)
31. Hébert, L.: Tools for Text and Image Analysis: An Introduction to Applied Semiotics.
Routledge, New York (2020)
32. Monterroso, A.: Obras Completas (y otros cuentos). Ediciones Era, Mexico, DF (1990)
33. Complete Works and Other Stories. University of Texas Press, Austin, Texas (1996)
34. Ramírez, J., Toledo, J.: El Dinosaurio de Monterroso: La exégesis Política más Breve del
Mundo, Perro Negro, 1 October 2015. http://revistaperronegro.com/el-dinosaurio-de-monter
roso-la-exegesis-politica-mas-breve-del-mundo/. Accessed 19 July 2020
35. Aguilar, S.: Youtube, 8 December 2016. https://www.youtube.com/watch?v=37EPAHbur50.
Accessed 27 July 2021
Interactive Narrative Impact and
Applications
Interviews Towards Designing Support
Tools for TTRPG Game Masters
1 Introduction
In running tabletop roleplaying games, game masters (GMs) lead the game and
take on a variety of roles, including providing challenges to players and improvis-
ing new content based on the player’s character backstories, choices, and actions
in the game world. In order to do this, GMs must be able to adapt the game
on the fly, improvising from both prepared materials and the GM’s imagination
to build new elements of the game world. This can be a difficult task, and one
that is especially daunting to new GMs. In this paper, we use qualitative inter-
views with GMs in order to assess a digital prototype for helping GMs. This
requirements analysis helps us understand how computational tools can help to
support GMs and inform next directions for the design of such tools.
2 Related Work
Currently there exist many different digital commercial and research tools for
helping run TTRPGs. While there are many tools to help GMs facilitate their
games such as virtual tabletop platforms or rulebook references for lookup, we
focus here on the most relevant work related to facilitating GMing. One exam-
ple is Undercurrents, a tool that helps to facilitate hidden information commu-
nication within tabletop roleplaying games by helping the game master share
information with only a single or a few players [3]. This helps maintain hidden
information in what is traditionally an open space of information, and provides a
way of keeping track of what has happened in the game so far. Another example
is Imaginarium, which uses procedural text generation to provide descriptions
that are constrained by the author but still have variations to them, using an
authoring language similar to natural language [6]. Horswill poses this as a casual
authoring tool for game masters to develop semi-randomized content on the fly,
for instance descriptions of monsters.
There has also been previous work theorizing how one might build digital
tools based on these techniques. Bergström uses interviews with TTRPG play-
ers and their own experience with TTRPGs to create categories of “frames”
of storytelling, such as diegetic and non-diegetic communication, using this to
inform the design of digital tools for TTRPG communication [3]. Peinado dis-
cusses how GMing techniques could be applied in a digital domain, such as for
modeling improvisation in a text adventure game. Our work builds on these and
other studies on TTRPGs and using qualitative interviews, such as [8] and [11].
Finally, we use techniques informed by requirements analysis, in which inter-
view insights can help to inform the design and iterative development of digital
tools. Nelson & Mateas [7] provide game design assistive tools to several groups
of game designers with various needs, perform interviews with them about their
needs and how they can use the tools provided to help with those needs, and
iterate on the tool’s design based on the results of these interviews. Grow’s [4]
approach to evaluating AI architecture authoring tools is similar, using three
different case studies to evaluate three different architectures, and looking at
how different architectures require different methods of authoring. Gustafsson,
Holme, and Mackay analyze the play experiences and players’ stories of impor-
tant objects from their play using interviews and questionnaires, using these to
inform the design of new game architectures that provide a greater support for
player narratives [5].
3 Methods
In order to give GMs a baseline from which we could discuss the design and
functionality of a TTRPG digital assistant, we created a digital prototype of a
tool for GMs that had some of the features that we were interested in assessing.
We based this prototype on beginner Dungeons & Dragons 5th edition module
Lost Mine of Phandelver, because it is a scenario meant for new GMs using
a popular TTRPG system, and thus could be the introduction to GMing for
many new GMs. For modeling out the scenario, we focused on Chapters 2 and
3 of Lost Mine because these offered some interesting variance in play style
(social, hub-based quests and combat) and player options. The static version
of the visualization laid out story and character information and connections
Interviews Towards Designing Support Tools for TTRPG Game Masters 285
4 Requirements Analysis
Overall, GMs commented that they liked the tool and were interested in the
potential for computational tools to help with GMing. Many of the GMs liked
that the tool condenses the many pages of information found in the module
into a more accessible form that can be referenced in either planning for a next
session or during play. GMs also talked about how the tool could be used to
keep track of information, such as information the players know, or things that
they have done, which could be used to help map out the rest of the story and
determine effects from previous player actions.
That said, GMs had many recommendations for improvements and features
that they would like to see. One area that GMs wanted to see improved was in
information visualization. The current prototype has some limited information
on NPCs–names, occupations, and a few relationship indicators such as whether
characters are family members–but GMs discussed more features that they would
like to see, and different desired functions, such as being able to sort, filter, and
tag information (1, 2, 5). GMs also wanted more visualizations of information
for storytelling, such as showing relationships between characters or factions
(5). GM #2 added that they would also like to keep track of faction goals–
what members of the factions want, and the next steps that they will take to
accomplish this. This helps to drive the story based around the characters and
their motivations.
GMs also talked about other uses for computation in supporting GMing. One
of the main areas that GMs wanted support for was in being able to swap around
existing content or add their own content, which was partially supported by the
286 D. Acharya et al.
current prototype. Some of this was based on tailoring content to players. For
example, GMs #2, 4, and 6 talked about cutting down on the content that was
provided in the module and swapping elements to help create player investment
such as providing encounters with monsters that the player characters dislike (2),
or using encounters to convey different narrative themes (4). Swapping content
can also be helpful changing content around if needed to help advance the story.
GM #6 gives an example of this from Lost Mine–if Hamun the necromancer
has information that you want the players to know but they decide they don’t
want to pursue that route, you can provide that information on another NPC,
for example having the players find this information after defeating Iarno, the
leader of the Redbrands (6). It would be interesting to further explore how
computational tools could facilitate this content swapping, either manually or
automatically through an AI suggesting changes to tailor scenarios to the players’
backstory, theme, or interests. As GM #1 points out, swapping things around
arbitrarily could lead to more complications down the line as the GM changes
information or key characters for story progression. This could also be a potential
area for computational intervention.
Another area in which GM #1 talked about creating new content for the
game was in helping realize character and faction beliefs in the game. For exam-
ple, if the general of a faction has the belief that the best defense is a good
offense, how is this behavior contextualized and seen during play? GM #1 dis-
cusses how it might be nice to have some provided framing questions in order
to help them better establish the kinds of actions that groups would take in the
world to convey their beliefs. They offer the example of having the tool prompt
the GM with questions–for example, how the given faction might take a fort (by
force, by coercion, by stealth, etc.) or if the town was taken over by a tyrannical
leader, how a member of the faction might deal with that.
Finally, GMs (5, 6) also talk about having improvisational prompts for things
that can happen next. GM #6 discussed how this would be good for beginners,
especially for modules that are large, expansive worlds (the GM gives the exam-
ple of Storm King’s Thunder [1]). In such games, there is a lot of content, but
at any one time players (and the GM) might be at a loss for what to do next,
and there may be large distances (either physically or narratively) between each
chapter of the story, with player potentially missing the plot hooks connecting
them (6). Potentially, a computational system with some knowledge of what has
already happened and story threads could provide prompts for potential events
that could happen next, or stepping stones to help guide players to the next part
of the story.
While there are some limitations to the data collected here, such as the number
of participants interviewed, we believe that this is a strong starting point for
understanding how we can better design computational assistive tools for GMs.
Next steps for this work likely include building out some of these speculative
Interviews Towards Designing Support Tools for TTRPG Game Masters 287
designs as actual functioning tools that GMs can use, and evaluating these tools
through user studies, particularly with a target demographic such as novice GMs.
References
1. Wizards of the Coast, Inc., Perkins, C.: Storm King’s Thunder. Wizards of the
Coast, Renton (2016)
2. Acharya, D., Mateas, M., Wardrip-Fruin, N.: Story improvisation in tabletop role-
playing games: Towards a computational assistant for game masters. In: Conference
on Games (2021)
3. Bergström, K.: Framing storytelling with games. In: Si, M., Thue, D., André,
E., Lester, J.C., Tanenbaum, T.J., Zammitto, V. (eds.) ICIDS 2011. LNCS, vol.
7069, pp. 170–181. Springer, Heidelberg (2011). https://doi.org/10.1007/978-3-
642-25289-1 19
4. Grow, A., Gaudl, S., Gomes, P., Mateas, M., Wardrip-Fruin, N.: A method-
ology for requirements analysis of AI architecture authoring tools. In: Pro-
ceedings of the 9th International Conference on the Foundations of Digital
Games (2014). https://games.soe.ucsc.edu/methodology-requirements-analysis-
ai-architecture-authoring-tools
5. Gustafsson, V., Holme, B., Mackay, W.E.: Narrative substrates: reifying and man-
aging emergent narratives in persistent game worlds. In: International Conference
on the Foundations of Digital Games, pp. 1–12 (2020)
6. Horswill, I.: Imaginarium: a tool for casual constraint-based PCG (2019)
7. Nelson, M.J., Mateas, M.: A requirements analysis for videogame design support
tools. In: Proceedings of the 4th International Conference on Foundations of Digital
Games - FDG 2009. ACM Press, New York (2009)
8. Reyes, M.C.: Measuring user experience on interactive fiction in cinematic vir-
tual reality. In: Rouse, R., Koenitz, H., Haahr, M. (eds.) ICIDS 2018. LNCS, vol.
11318, pp. 295–307. Springer, Cham (2018). https://doi.org/10.1007/978-3-030-
04028-4 33
9. Robinson, O.C.: Sampling in interview-based qualitative research: a theoretical and
practical guide. Qual. Res. Psychol. 11(1), 25–41 (2014)
10. Saldaña, J.: The Coding Manual for Qualitative Researchers. Sage, Thousand Oaks
(2015)
11. Strugnell, J., Berry, M., Zambetta, F., Greuter, S.: Narrative improvisation: sim-
ulating game master choices. In: Rouse, R., Koenitz, H., Haahr, M. (eds.) ICIDS
2018. LNCS, vol. 11318, pp. 428–441. Springer, Cham (2018). https://doi.org/10.
1007/978-3-030-04028-4 50
The Ethics of Virtual Reality Interactive Digital
Narratives in Cultural Heritage
bakk@mome.hu
1 Introduction
The question of the effects of mediated products on their audiences has been studied for
a considerable time [1–3]. As the field of Interactive Digital Narratives (IDNs) matures
[4], questions arise that have been asked of other disciplines [5], and starting to address
these questions helps the field develop further. One such question is the issue of ethics:
what are the ethical considerations of IDNs? The approach taken by this paper is to focus
on ethical questions pertaining to a specific platform and application for IDNs, Virtual
Reality (VR) and cultural heritage. Concretely, we start with a catalogue of general
concerns and then take a bottom-up approach in considering IDNs developed as VR
experiences in cultural heritage. The connection to cultural heritage enables us to take
in insights developed in this more established field, and consider ethical aspects of VR
technology which has been used in the preservation of cultural heritage sites and artefacts
over the past two decades. Bringing together these lessons and our initial catalogue, we
identify specific concerns for VR in cultural heritage applications.
A crucial prerequisite to discussions of ethical questions in IDN design is an under-
standing of its specific qualities in contrast to fixed narratives forms. IDN as a dynamic,
systemic form is principally open to unexpected consequences and unintended uses and
this central aspect shifts a part of the responsibility to the audience as interactors. On this
basis, we can identify questions for ethical implications of IDN productions in general
and VR products for cultural heritage specifically. The following catalogue of questions
is intended as a starting point for a framework of ethical IDN production as part of the
further development of the field.
• How can multi-perspective experiences be created in a way that each perspective gets
a fair treatment?
• How can an IDN scaffold an experience for the interactor in a complex situation so
that they arrive at an understanding of how their actions have resulted in the outcome?
• How can undue simplification and trivialization of a complex situation in an IDN be
prevented?
• How can an IDN design accommodate interactors with different levels of experience
with interactive artifacts?
• How can an IDN design accommodate interactors with different levels of prior
knowledge in regards to the topic of the artefact?
The ability of IDNs to contain multiple perspectives and offer choices make it an attrac-
tive medium through which cultural heritage can be represented. Before we consider the
above ethical concerns, we need to also consider the ethical considerations introduced
by the technological platform of choice. Built to visually and aurally immerse the wearer
into a virtual world, VR technology has been the platform of choice for over two decades
in digital heritage experiences [6]. Examples of such IDN cultural heritage applications
are The Last Goodbye [7], Nefertari: Journey to Eternity [8], Chauvet: The Dawn of Art
[9] and The Book of Distance [10].
In 2016, Michael Madary and Thomas K. Metzinger proposed a possible code of
conduct for VR [11]. The authors emphasize that the main motivation behind their
investigation is VR’s ability to enable the illusion of embodiment and argue that this
can have a manipulative effect, especially if “illusions of embodiment are misused”
[11]. Madary and Metzinger point out the importance of informing the interactors about
potentially lasting psychological effects of VR and “the possibility of using results of
VR research for malicious purposes” [11]. Furthermore, the authors discuss a potential
danger of excessive use of VR, which can lead to a condition where VR users could
“experience the real world and their real bodies as unreal, effectively shifting their sense
of reality exclusively to the virtual environment” [11].
Importantly, IDNs are not just the ‘interactivisation’ of traditional media [12]. In
this regard, Rebecca Rouse reminds of the active role of the audience, created by these
experiences, which traditionally “has been reserved for the curator and exhibit designer”
[13]. The interactivity in IDNs differs considerably from earlier fixed forms of narration
and thus require a novel starting point, such as Koenitz’ SPP model [14]. For the same
reason, we have to reframe the ethical considerations explored above within the framing
of IDN work including VR.
290 J. Barbara et al.
intangible cultural heritage without having the cultural insider’s knowledge necessary
for a full understanding. The vision of VR as the ‘ultimate empathy machine’ [19] is
problematic for several reasons (also see [20]), one of which we can see in The Book
of Distance. What the IDN designer should aim for is instead ‘radical compassion’: ‘an
ethical stance that embraces an openness to understanding and refuses to assimilation
others’ experience into one’s own self’ [21].
• How can the role of the interactor be defined, such that no prior knowledge of a specific
cultural tradition is required, and yet enables an understanding of it?
• How can interactors with different knowledge levels about a specific cultural heritage
tradition be accommodated?
• How can an experience be respectful to a given tradition, but does not submerge
contemporary values to it (e.g. oppression due to gender, origin, religion etc.)?
• How can newly discovered knowledge (e.g. new archeological finds, novel historical
insights) be integrated into an IDN artifact?
• How can undue trivializing of the depicted original behavior be avoided?
• How can the responsibility of the interactor in a cultural heritage experience be clearly
communicated?
4 Conclusion
In this paper, we have provided a starting point for a framework of ethical IDN production
by analyzing ethical considerations arising from applications of IDN in VR for Cultural
Heritage experiences. Building upon ethical aspects of cultural heritage and virtual reality
separately, we have developed a set of ethical questions for IDN design in general and
specifically for the application of IDN in VR representing cultural heritage. From these
questions, initial guidelines can be developed by the community, with the aim to define
a standard applicable to on an international level in the future.
References
1. Benjamin, W.: Das Kunstwerk im Zeitalter seiner technischen Reproduzierbarkeit. Frankfurt
am Main, Suhrkamp (1969)
2. Horkheimer, M., Adorno, T.: The culture industry: enlightenment as mass deception. In:
Durham, M.G., Kellner, D.M. (eds.) Media and Cultural Studies: KeyWorks (2006)
292 J. Barbara et al.
3. Habermas, J.J.: The Structural Transformation of the Public Sphere: An Inquiry into a
Category of Bourgeois Society. (T. Burger, Trans. & F. Lawrence, Trans.) (1989)
4. Koenitz, H.: Thoughts on a discipline for the study of interactive digital narratives. In: Rouse,
R., Koenitz, H., Haahr, M. (eds.) Interactive Storytelling. ICIDS 2018. LNCS, vol. 11318,
pp. 36–49. Springer, Cham (2018). https://doi.org/10.1007/978-3-030-04028-4_3
5. Koenitz, H., Eladhari, M.P.: Challenges of IDN research and teaching. In: Cardona-Rivera,
R., Sullivan, A., Young, R. (eds.) Interactive Storytelling. ICIDS 2019. LNCS, vol. 11869,
pp. 26–39. Springer, Cham (2019). https://doi.org/10.1007/978-3-030-33894-7_4
6. Ch’ng, E., Cai, Y., Thwaites, H.: Special issue on VR for culture and heritage: the experience
of cultural heritage with virtual reality: guest editors’ introduction. PRESENCE Teleoperators
Virtual Environ. 26, iii–vi (2018). https://doi.org/10.1162/pres_e_00302
7. Zalewska, M.: The Last Goodbye (2017): Virtualizing Witness Testimonies of the Holocaust.
8 (2017)
8. Experius VR, Curiosity Stream: Nefertari: Journey to Eternity (2018). https://store.steamp
owered.com/app/861400/Nefertari_Journey_to_Eternity/
9. Tanant, J.: Google Arts & Culture, Atlas V: Chauvet: The Dawn of Art (2020). https://experi
ments.withgoogle.com/chauvet
10. Oppenheim, D., Okita, R.L.: The book of distance: personal storytelling in VR. In: ACM
SIGGRAPH 2020 Immersive Pavilion, pp. 1–2. ACM, Virtual Event USA (2020). https://doi.
org/10.1145/3388536.3407896
11. Madary, M., Metzinger, T.K.: Real virtuality: a code of ethical conduct. Recommendations
for good scientific practice and the consumers of VR-technology. In: Frontiers in Robotics
and AI, vol. 3 (2016)
12. Koenitz, H.: Reframing interactive digital narrative: toward an inclusive open-ended iterative
process for research and practice (2010). https://smartech.gatech.edu/bitstream/handle/1853/
34791/koenitz_hartmut_a_201008_phd.pdf
13. Rouse, R.: Someone else’s story: an ethical approach to interactive narrative design for cultural
heritage. In: Cardona-Rivera, R.E., Sullivan, A., Young, R.M. (eds.) ICIDS 2019. LNCS, vol.
11869, pp. 47–60. Springer, Cham (2019). https://doi.org/10.1007/978-3-030-33894-7_6
14. Koenitz, H.: Towards a specific theory of interactive digital narrative. In: Interactive Digital
Narrative, pp. 91–105 (2015)
15. Mochocki, M.: Heritage sites and video games: questions of authenticity and immersion.
Games Cult. 155541202110053 (2021). https://doi.org/10.1177/15554120211005369
16. Murray, J.H.: Hamlet on the Holodeck: The Future of Narrative in Cyberspace. Simon and
Schuster, New York (1997)
17. Kilteni, K., Groten, R., Slater, M.: The sense of embodiment in virtual reality. Presence
Teleoperators Virtual Environ. 21, 373–387 (2012)
18. Barreda-Ángeles, M., Aleix-Guillaume, S., Pereda-Baños, A.: An empathy machine or a just-
for-the-fun-of-it machine? Effects of immersion in nonfiction 360-video stories on empathy
and enjoyment. Cyberpsychol. Behav. Soc. Netw. 23, 683–688 (2020)
19. Milk, C.: How virtual reality can create the ultimate empathy machine (2015)
20. Fisher, J.A.: Empathic actualities: toward a taxonomy of empathy in virtual reality. In: Nunes,
N., Oakley, I., Nisi, V. (eds.) Interactive Storytelling. ICIDS 2017. Lecture Notes in Computer
Science, vol. 10690, pp. 233–244. Springer, Cham (2017). https://doi.org/10.1007/978-3-319-
71027-3_19
21. Bollmer, G.: Empathy machines. Media Int. Aust. 165, 63–76 (2017)
Exploring Multiple Perspectives
in Citizenship Education with a Serious
Game
1 Introduction
2 Related Work
Among the most notorious barriers to open conversation in the classroom are
sensitive topics, groupthink, and clashing worldviews. We briefly survey existing
research on these specific barriers of open conversation, as well as on how they
can relate to citizenship education and serious games.
Exploring Multiple Perspectives in Citizenship Education 295
3 Game Design
The goal of promoting open conversation around sensitive topics guided the
design of Diermocratie. In order to achieve this goal, many game choices that
were taken are directed towards avoiding groupthink and mitigating the influence
of players’ predispositions.
briefing can be seen in Fig. 2. The briefing wording itself is not recognizable as
the Dutch nitrogen debate, however the link to farmers being the chickens and
environmentalists being the sheep can easily be made.
An important implementation aspect of our use of the metaphor is the
anonymity afforded by a digital game. While many of the gameplay elements
described could translate to a physical tabletop game, role-playing would become
more difficult as there would be a clear relation between actual players and roles.
Since Diermocratie is intended to be played among classmates, anonymizing
players by giving them a virtual avatar both minimizes the effect of pre-existing
relationships and enhances players’ immersion into the game, improving the
role-playing quality.
Fig. 3. Example species exclusive headlines that are paired with the briefing from
Fig. 2.
Exploring Multiple Perspectives in Citizenship Education 301
vote. The headlines and the consequent communication around them, facilitate
multiple diverse perspectives and conversation surrounding a polarizing topic. It
also connects to the argumentation competency of citizenship. After a set time
for gameplay, the final vote is held, requiring the players to act immediately
as time is a limited resource. With the final vote and the subsequent outcome
announcement, the players are informed of the results of their discussion. The
actual tally is not that important, as the game revolves mostly about getting an
understanding of the democratic process as a whole. The results can, however,
be used as input to the final debriefing between teacher and students.
4 Game Evaluation
4.1 Method
In order to evaluate the extent to which the game was effective in its purpose
to promote an open conversation about difficult topics, multiple play sessions
were held with vocational college classes of approximately 20 students each. Due
to current societal constraints, play sessions varied between taking place fully
online, entirely on location or in a half-and-half setting. During each session,
students were given a short presentation to get familiar with the functionality
of the game. After this brief introduction, the students were asked to play the
game.
302 E. Blokland et al.
It was noticeable that students were more involved in the in-person sessions than
when they played online. Debriefing sessions were also more actively attended,
allowing for better discussion on societal issues to match the game purpose. The
survey was filled in by a total of 60 students.
Regarding the communication among players (see Fig. 4), about half of the
players report they were motivated to talk to other players, as can be seen
in Fig. 4a. That said, Fig. 4b indicates that 45% of the players talked to less
than 2 people, which is a clear concern. During the debriefing sessions it came
forward that excessive spamming was one of the major reasons players were
less motivated to talk to each other. Because of this, especially dyslexic people
were having a hard time to keep up with the conversation. Measures should be
taken to prevent spamming and improve the UI so the chat is more organized,
for example by displaying the chat in a sidebar. With these improvements in
More than 5
No 13,3%
27,1%
2 to 5
Indifferent
41,7%
23,7%
20
20
15 15
10 10
5 5
0
0
g
ea
s
er
y
n
s
g
in
on
lit
in
io
ow
pi
th
cy
ism
id
es
er
ity
ng
ch
s
em n
ot
in
at
ua
on
ni
to
io
O
at
ra
us
th
kn
o
ee
ili
om
al
ot
ic
pi
ac
at
eq
N
ni
eb
rd
oc
isc
qu
of
Sp
un
V
O
in
't
pi
Pr
D
Pr
ha
g
D
on
ne
m
m
in
of
om
ri
D
lI
Id
ic
ot
m
us
isc
hn
om
ia
do
isc
ac
Et
Pr
ee
D
R
Fr
(a) What was the game about? (b) What was the goal of the game?
Yes
45,0%
No
55,0%
Fig. 5. Survey results regarding game mechanics, topic and purpose (N = 60)
4.3 Limitations
Here we identify some limitations present in conducting the evaluation of this
research. Firstly, limited playtesting sessions and student interaction during
game development undoubtedly and inevitably impacted the preliminary evalu-
ation of this game. Moreover, during game development, interaction was limited
to only a few educational facilities, and only a small range of course materi-
als could be considered in the scope of this project. The current scope of this
research is therefore very focused, and a long term evaluative effort is much rec-
ommended, in order to extend the range of topics in the game and improve its
impact and effectiveness.
On another dimension, and as a result of the large diversity of subcultures
present, we realized that the metaphors used in the game, including animal
metaphors, can hold very different connotations to situations or activities for
diverse groups of people. As a result, different players are likely to interpret
very differently the game setting, specific characters, and the presented game
scenario. While this challenge was currently experienced as a limitation, future
development scenarios for this game could use this feature, if managed correctly,
as an advantage for fueling diversity of perspectives about sensitive topics.
5 Conclusion
Citizenship education is increasingly regarded as an important pillar in voca-
tional education. However, many educational institutions face important chal-
lenges to adapt to modern demands in this domain.
We designed and developed Diermocratie, an in-classroom game aimed at
promoting open conversations within diverse classes among vocational college
students. The game empowers students to grow in diversity-related competences,
including self-awareness, empathy and argumentation. By means of role-playing
metaphors, players successfully explore their predispositions, becoming aware
of each other’s perspectives. This progress is further supported by a debriefing
session, led by an instructor. As a result of a preliminary evaluation, we have
concluded that the game motivated students to discuss with their colleagues
different perspectives on a complex scenario.
Compared to other serious games around citizenship topics, as e.g. TimeMesh
[5] and LawVille [23], Diermocratie stands out in both the flexibility for potential
discussion topics and the crucial role of player interaction in achieving the game
goals. Diermocratie was not designed to teach a fact-based curriculum, but rather
to promote open conversations around sensitive topics. Conversations between
players, therefore, take a much more central role, and shape the outcome of the
story rather than being simply a tool for players to progress through a story.
We believe that Diermocratie provides a valuable and effective assistance
to vocational college instructors in challenging citizenship education topics. We
therefore expect the game to continue being further developed, extended with
carefully designed scenarios for more specific topics, and rolled out in more
educational institutions.
Exploring Multiple Perspectives in Citizenship Education 305
Acknowledgments. We thank Anouck Wolf and her team at Critical Mass for the
inspiring discussions and feedback throughout this project.
References
1. Abowd, G.D., Dey, A.K., Brown, P.J., Davies, N., Smith, M., Steggles, P.: Towards
a better understanding of context and context-awareness. In: Gellersen, H.-W. (ed.)
HUC 1999. LNCS, vol. 1707, pp. 304–307. Springer, Heidelberg (1999). https://
doi.org/10.1007/3-540-48157-5 29
2. Alaka, S., Cunha, M.L., Vermeer, J., Salamon, N.Z., Balint, J.T., Bidarra, R.:
Stimulating ideation in new teams with the mobile game Grapplenauts. Int. J.
Serious Games 6(4), 87–101 (2019)
3. Alderliesten, D., Valečkaitė, K., Salamon, N.Z., Timothy Balint, J., Bidarra, R.:
MainTrain: a serious game on the complexities of rail maintenance. In: Gentile, M.,
Allegra, M., Söbke, H. (eds.) GALA 2018. LNCS, vol. 11385, pp. 82–89. Springer,
Cham (2019). https://doi.org/10.1007/978-3-030-11548-7 8
4. Aronson, E.: Building empathy, compassion, and achievement in the jigsaw class-
room. In: Improving Academic Achievement, pp. 209–225. Elsevier (2002)
5. Baptista, R., Carvalho, C.V.d.: TimeMesh - a serious game for European citizen-
ship. EAI Endorsed Trans. Game-Based Learn. 1(1), e2 (2013). https://doi.org/
10.4108/trans.gbl.01-06.2013.e2
6. Belman, J., Flanagan, M.: Designing games to foster empathy. Int. J. Cogn. Tech-
nol. 15(1), 11 (2010)
7. Benbassat, J., Baumal, R.: Enhancing self-awareness in medical students: an
overview of teaching approaches. Acad. Med. 80(2), 156–161 (2005)
8. Brandt, M.J., Crawford, J.T.: Worldview conflict and prejudice. In: Advances in
Experimental Social Psychology, vol. 61, pp. 1–66. Elsevier (2020)
9. Chen, A.M., Kiersma, M.E., Yehle, K.S., Plake, K.S.: Impact of the geriatric med-
ication game R on nursing students’ empathy and attitudes toward older adults.
Nurse Educ. Today 35(1), 38–43 (2015)
10. Dawson, V.M., Venville, G.: Teaching strategies for developing students’ argumen-
tation skills about socioscientific issues in high school genetics. Res. Sci. Educ.
40(2), 133–148 (2010). https://doi.org/10.1007/s11165-008-9104-y
11. Dikken, O., et al.: A serious game for changing mindsets about loans for home
retrofitting. In: Marfisi-Schottman, I., Bellotti, F., Hamon, L., Klemke, R. (eds.)
GALA 2020. LNCS, vol. 12517, pp. 347–361. Springer, Cham (2020). https://doi.
org/10.1007/978-3-030-63464-3 33
12. Gee, J.P.: What video games have to teach us about learning and literacy. Comput.
Entertain. (CIE) 1(1), 20–20 (2003)
13. Gerdes, K.E., Segal, E.A., Jackson, K.F., Mullins, J.L.: Teaching empathy: a frame-
work rooted in social cognitive neuroscience and social justice. J. Soc. Work Educ.
47(1), 109–131 (2011)
14. Irvine, J.J.: Complex relationships between multicultural education and special
education: an African American perspective. J. Teach. Educ. 63(4), 268–274 (2012)
15. Janis, I.L.: Groupthink. IEEE Eng. Manag. Rev. 36(1), 36 (2008)
16. Johnson, S.D., Weaver, R.L., II., et al.: Groupthink and the classroom: changing
familiar patterns to encourage critical thought. J. Instr. Psychol. 19(2), 99 (1992)
17. Kello, K.: Sensitive and controversial issues in the classroom: teaching history in a
divided society. Teach. Teach. 22(1), 35–53 (2016)
306 E. Blokland et al.
18. Kneupper, C.W.: Teaching argument: an introduction to the toulmin model. Coll.
Compos. Commun. 29(3), 237–241 (1978)
19. Koenitz, H.: Towards a theoretical framework for interactive digital narrative. In:
Aylett, R., Lim, M.Y., Louchart, S., Petta, P., Riedl, M. (eds.) ICIDS 2010. LNCS,
vol. 6432, pp. 176–185. Springer, Heidelberg (2010). https://doi.org/10.1007/978-
3-642-16638-9 22
20. Kral, T.R., et al.: Neural correlates of video game empathy training in adolescents:
a randomized trial. npj Sci. Learn. 3(1), 1–10 (2018)
21. Leeman, Y., Nieveen, N., de Beer, F., Van der Steen, J.: Teachers as curriculum-
makers: the case of citizenship education in Dutch schools. Curric. J. 31(3), 495–
516 (2020)
22. Leeman, Y., Pels, T.: Citizenship education in the Dutch multiethnic context. Eur.
Educ. 38(2), 64–75 (2006)
23. Lorenzini, C., Brondi, R., Carrozzino, M., Nistico, M., Evangelista, C., Tecchia,
F.: LawVille: a collaborative serious game for citizenship education. In: 2014 6th
International Conference on Games and Virtual Worlds for Serious Applications
(VS-GAMES). IEEE, September 2014. https://doi.org/10.1109/VS-Games.2014.
7012163
24. Raateland, W., et al.: A serious game for students to acquire productivity habits.
In: Marfisi-Schottman, I., Bellotti, F., Hamon, L., Klemke, R. (eds.) GALA 2020.
LNCS, vol. 12517, pp. 335–346. Springer, Cham (2020). https://doi.org/10.1007/
978-3-030-63464-3 32
25. Ruiz, S.: Take action games: Darfur is dying. https://susanaruiz.org/
takeactiongames-darfurisdying
26. Sjölund, A., et al.: Misusing mobile phones to break the ice: the tabletop game
Maze Maestro. In: Proceedings of FDG 2020 - International Conference on the
Foundations of Digital Games (2020)
27. Spohr, D.: Fake news and ideological polarization: filter bubbles and selective expo-
sure on social media. Bus. Inf. Rev. 34(3), 150–160 (2017)
28. Starks, K.: Cognitive behavioral game design: a unified model for designing serious
games. Front. Psychol. 5, 28 (2014)
29. Uleman, J.S., Adil Saribay, S., Gonzalez, C.M.: Spontaneous inferences, implicit
impressions, and implicit theories. Annu. Rev. Psychol. 59, 329–360 (2008)
Digital Narrative and Temporality
e.fulop@lancaster.ac.uk
According to Paul Ricoeur (1984), “Time becomes human to the extent that it is artic-
ulated through a narrative mode, and narrative attains its full meaning when it becomes
a condition of temporal existence”. From this perspective, narrative is our principal tool
for situating ourselves in time—and for experiencing time within ourselves. The digital,
on the other hand, may be characterized as “a tool for the phenomenal deconstruction
of temporality” (Bachimont 2014). This is reflected in its two main tendencies: that of
real time calculation, conveying the impression of immediacy, and that of universality of
access, conveying the impression of availability. The digital could thus lead to constant
present, without any impression of the passage of time. What happens then when we
exploit the particularities of digital technology to tell a story? What kind of temporal
experiences are constructed by new forms of digital narrative, and how are they con-
structed? Reciprocally, what new narrative forms, or even new concepts of narrative, do
these new temporal experiences provided by digital technology offer to us? Through the
example of three types of digital stories, including a smartfiction based on notifications,
a web narrative based on real-time data flow, and the widely used social media feature
of stories, this paper raises the question of the diverging potentialities of the relationship
between the digital, temporality, and narrative.
a real time data flow, it is at the same time a “pure experience of time”, based on the
temporal experience of the fictional character (Chambefort 2020). Chambefort refers
to this work as a “narrative”. If we can speak of a narrative, it is mainly thanks to the
words attributed to Lucette and the unifying force of her perspective. Yet Lucette is no
doubt situated at the frontier of what constitutes a narrative, just like Bury me, my love.
The latter can indeed be categorized as a “narrative game”. Lucette raises the question
as to how far we can stretch digital narratives before they cease to be narratives, and
whether database narratives may be able to help us deal with the digital data flows we
are constantly exposed to, or even to foster a different relationship with them and better
situate them in time.
Here we witness a very different logic, appealing to the senses to provoke simple and
immediate emotional response. Not only does the story fail to impose any criteria for
narrativity, narrativity even seems to be an obstacle to the stories’ immediate efficacy.
4 Further Questions
Our first two examples show that interactivity and dataflows raise questions around
temporality, intertwined with the question of fictionality6 . If the classical narrative aims
for and encourages immersion through movement from real space-time towards fictional
space-time, the new narrative forms allow the emergence of fictional space-time in real
space-time, via a movement that not only crosses, but also deconstructs frontiers.
Our third example invites us to rethink our conception of narrative as it is defined
by classical and post-classical narratology, adopting the term and using it to designate
entirely different phenomena and practices. This radical otherness also expresses itself
through the relation to time: the temporality imposed by the feature, that of the content
it proposes, and that of the users. The function of these stories is no longer to establish
temporal or causal links between events, but simply to include them in what could be
referred to as “the life of the network(s)”. In 1984, Ricoeur observed that “new narrative
forms, which we do not yet know how to name, are already being born, which will
bear witness to the fact that the narrative function can still be metamorphosed, but
not so as to die”. The question today is how far this metamorphosis can go, and how
far we can push the limits of narrative without it losing its traditional use. To what
extent do our consciousness and thinking depend on narratives, and how prepared are
we to consider alternative modes of functioning?7 We may not be facing a total and
definitive disappearance of narratives, of course, but perhaps a relativisation of their
place in our culture, history and theoretical constructions, and a multiplication of modes
of representing and performing time.
References
Bachimont, B.: Availability and the transformation of objects into heritage: digital technology
and the passing of time. In: Dufrêne, B. (ed.) Heritage and Digital Humanities: How Should
Training Practices Evolve?, pp. 49–70. LIT, Berlin (2014)
Bouchardon, S.: Mind the gap! 10 gaps for digital literature?”, Electronic Book Review (2019).
http://electronicbookreview.com/essay/mind-the-gap-10-gaps-for-digital-literature/
Chambefort, F.: Lucette, gare de Clichy (2017). http://fchambef.fr/lucette/
Chambefort, F.: Sortir de l’écran : Lucette, Gare de Clichy. In: Gervais, B., Marcotte, S. (eds.)
Attention à la marche ! Penser la littérature électronique en culture numérique. Montréal: Les
Presses de l’Écureuil - ALN/NT2, pp. 327–338 (2020). ISBN 979-10-384-0004-7
Easterlin, N.: A Biocultural Approach to Literary Theory and Interpretation. Johns Hopkins UP,
Baltimore (2012)
Fludernik, M.: An Introduction to Narratology. Routledge, Londres (2009)
6 On this subject see Fülöp (2021).
7 Galen Strawson, in a well-known article (2008), already suggests that “There are deeply non-
Narrative people and there are good ways to live that are deeply non-Narrative”.
Digital Narrative and Temporality 311
Fülöp, E.: Virtual mirrors: reflexivity in digital literature. In: Fülöp, E. (ed.) In collaboration with
Priest, G. and Saint-Gelais, R., Fictionality, Factuality, and Reflexivity across Discourses and
Media, pp. 229–254. De Gruyter, Berlin (2021)
Herman, D.: Storytelling and the Sciences of Mind. MIT, Cambridge (2013)
Mayer, A., Bouchardon, S.: The digital subject: from narrative identity to poetic identity? Elec-
tronic Book Review, May 2020. https://electronicbookreview.com/essay/the-digital-subject-
from-narrative-identity-to-poetic-identity/
Picard, M.: La smartfiction, quand l’environnement numérique fictionnel s’hybride à
l’environnement numérique personnel. XXIIème congrès de la SFSIC (2021). https://sfsic2
020.sciencesconf.org/333005
Revaz, F.: Introduction à la narratologie. Action et narration. De Boeck, Brussels (2009)
Ricœur, P.: Time and Narrative, vol. 1. University of Chicago Press, Chicago (1984)
Schmid, W.: Elemente der Narratologie. de Gruyter, Berlin (2005)
Strawson, G.: Against narrativity. In: Real Materialism and Other Essays. Clarendon Press, Oxford
(2008)
The Pixel Hunt, Figs, Arte France (2017). Enterre-moi, mon amour. http://enterremoimonamour.
arte.tv/
Zunshine, L.: Why We Read Fiction? Ohio UP, Columbus (2006)
“You Write Your Own Story”: Design
Implications for an Interactive Narrative
Authoring Tool to Support Reflection
for Mental Health in College Students
1 Introduction
The use of narrative and storytelling in mental health therapeutic practice has
been widely documented [1,10–12,15,19,20,22,23,29]. Across techniques used
in mental health therapy, the restructuring or forming of new narratives with
a patient over time is a common theme. One way we engage or restructure our
personal narratives both on our own and in the context of therapy is through
self-reflection, a term we use here to describe any process in which an individual
reviews their life. However, digital self-help tools for mental health therapy often
lack the guidance of a therapist found in a formal therapeutic setting.
For populations such as college students, digital self-help tools for mental
health therapy can be enormously useful since they often face high levels of
anxiety and depression, demand for mental health services continues to outpace
c Springer Nature Switzerland AG 2021
A. Mitchell and M. Vosmeer (Eds.): ICIDS 2021, LNCS 13138, pp. 312–321, 2021.
https://doi.org/10.1007/978-3-030-92300-6_30
“You Write Your Own Story” 313
supply, and they are also familiar with technology [16–18]. The challenge we
perceive thus is: how do we provide a digital intervention for college students
which facilitates the narrative process found in mental health therapy through
a guided reflection they can perform on their own?
Our proposed approach is the design of an interactive story authoring tool
made explicitly to support the reflective process with the support and guidance
found in mental health therapy. The primary aim of interactive story authoring
tools is to provide authors the ability to construct and revise non-linear narra-
tives. This is also necessary in the narrative processes of therapy and reflection,
given that our life narratives are non-linear in the way they branch out into
different future possibilities. This is particularly applicable to college students
who are at a stage of their lives where their personal narratives are in a natural
state of flux, due to a number of stressful life changes [14,28].
Many authoring tools for interactive narratives exist, but they naturally tend
to focus on the final creative output. None are designed to support the process
of reflection on one’s life narrative for mental health support. To design such
authoring tools for college students, we must first understand the nature of how
students engage with their life narratives, and how counselors work with students
to guide their narratives with the goal of improving mental health. This paper
presents findings from a series of interviews that we conducted to fulfill that
purpose. From these findings, we derived design implications for an interactive
story authoring tool for reflection for mental health. Thus, the goals of our study
were as follows: (i) to understand how college students guide their own personal
narratives through reflection and maintain their mental health over time; (ii)
to understand how counselors who treat college students form narratives with
their patients to maintain their care between sessions, and how they guide those
narratives over time; and (iii) to derive design aims for an interactive story
authoring tool to facilitate reflection on college students’ narratives to support
their mental health. We first summarize relevant background and related work,
followed by a description of our chosen methodology, a summary of the interview
study we conducted, and our subsequent analysis. The findings we present focus
on design aims and implications extracted from our analysis.
3 Methodology
Our goals were to understand how college students and counselors interact with
student personal narratives and work to improve their mental health, in order
to derive aims for the design of an authoring tool to facilitate therapeutic reflec-
tion in college students. We seek to better understand college students in this
“You Write Your Own Story” 315
context, as such a tool should align closely with their experiences, to allow for
ease of adapting said tool into their routine, as well account for unique strug-
gles of this population. Wielding the grounded theory method, we conducted
analysis between rounds of data collection to build an understanding of student
experiences, and to assess the need to revise interview questions or collect addi-
tional data. Open-ended interviews with college students were our primary data
source, and we conducted analysis after every 2 to 3 participants. Supplemental
interviews were also conducted with college counselors, in order to understand
from a professional standpoint how care is provided to this population and how
counselors interact with students’ personal narratives. The themes and questions
for both sets of interviews were derived from the first two goals of our study.
4 Study Details
The study consisted of open-ended interviews with college students and col-
lege counselors, lasting a maximum of one hour. We conducted and recorded
these interviews remotely over Zoom, and transcribed them for analysis. Stu-
dents were recruited via word of mouth and through our department’s SONA
recruitment system, which allowed students to participate in our study for extra
credit. Fourteen students in total participated in our study: 10 male, 4 female.
Ages ranged from 17 to 44 years, with an average age of 21.43 years. The eth-
nicities of the student participants were distributed as follows: 2 identified as
Asian, 3 as Hispanic or Latino, 8 as White/Caucasian, and 1 as Two or More
Races. Participants’ majors were primarily in a STEM field, with 11 in Com-
puter Science or Engineering, 1 in Aerospace Engineering, 1 in Physics, and 1 in
a double-major. College counselors were recruited via a digital flyer distributed
to university counseling centers and groups. 2 female counselors participated in
our study, aged 51 and 53 years, and both identified as Hispanic or Latino.
5 Analysis
Our coding process for the interviews followed recommendations made by Char-
maz [7]. First, we performed an Initial Coding of the data, by combing through
interview transcripts for responses relevant to the goals of our analysis, and creat-
ing codes which directly captured what the participant said. Next came Focused
Coding. Initial codes were compared for common themes to create focused codes.
Interpretation of the data began at this stage. Lastly, we performed Theoretical
Coding. Our process for creating theoretical codes was to qualitatively create
relationships between the focused codes. For the student interviews, coding was
jointly performed by 2 researchers for the first few rounds, with 1 researcher
performing the remainder of the analysis. We concluded the study upon estab-
lishing that we had sufficient data to derive significant implications for design.
Upon completion of the interview analysis, theoretical and focused codes were
examined and grouped for common themes to determine general aims for design.
Design aims were constructed with a combination of student and counselor codes,
in order to capture both an individual and professional perspective.
316 S. A. Brown and S. L. Chu
6 Study Findings
Our first two research goals were to understand how college students and college
counselors guide students’ personal narratives and maintain their mental health
over time. Two mappings of themes were extracted from our interviews through
our analysis. The maps consist of our theoretical codes, which draw connections
and establish relationships between these themes, and are supplemented with
further detail by the focused codes that are encapsulated by them. The map
of the college student interviews consists of 30 theoretical codes and 64 focused
codes. The map of the college counselor interviews contains 10 theoretical codes
and 27 focused codes. For brevity, we do not present these mappings, and instead
reference the most relevant codes in the following section.
Our third research goal was to derive design aims for an interactive story author-
ing tool to facilitate the process of reflection on college students’ personal nar-
ratives. Seven aims were extracted from the theoretical and focused codes of our
analysis. As we describe the implications for design associated with each of the
design aims below, we will differentiate theoretical codes and focused codes by
using boldface and italics. We will only include the most relevant codes in our
discussion of the design aims for the sake of brevity and clarity. The presented
implications for design are comparable to sensitizing concepts, and aim to share
relevant observations from our interviews with college students and counselors.
[24]. Thus, their value is less in their prescriptive nature, and more in their ability
to generate novel designs in future work.
Aim 1: The tool should allow the author to externalize and manipu-
late individual experiences in a manner that allows for easy reflection.
The codes Reflecting results in a step-back perspective and Reflecting by compar-
ing past experiences to the present indicate the need to overview the narrative
in a way that allows for easy comparison and isolation of relevant experiences.
Additionally, while an autobiographical retelling of a person’s story lends itself
well to a temporal structure, reflection can involve the comparing of experiences
in causal terms. Causal relationships between events are common in interactive
narratives, however in reflection, there exists a potential for the same memories
to be re-structured depending on the context of a given reflection. The author
should be able to create multiple configurations of life experiences, connected dif-
ferently based on the context of reflection, and visually manipulate the boundary
between past/present experiences to future ones. We propose the term reflection
context to describe a specific configuration of externalized experiences for the
purpose of reflection. Lastly, in the context of existing authoring tool paradigms
(as identified by Green et al. [13]), we suggest that a graph-based format could
best support the process of reflection. This paradigm is perfectly situated to
scaffold the author’s process of extracting specific experiences and restructuring
the narrative, by merit of an interactive visualization of the narrative structure.
“You Write Your Own Story” 317
the counselor interviews suggested that focusing on crises distracted from the
overarching goals of therapy, this does not mean the tool should ignore such a
scenario. Crisis support may come in the form of streamlining the author into the
appropriate reflection context to address the crisis upon an initial assessment, or
by redirecting them to additional resources depending on the situation’s severity.
Aim 5: The tool should capture and be informed by the author’s
identity. The codes under Foci (which contained various foci or priorities in stu-
dents’ narratives), and the counselor codes Counselors help patients determine
what they want/need and Need to account for patient identity suggest that for
our proposed authoring tool, the author’s identity as they undergo the authoring
process is uniquely important, not just the end creative product. This is natural,
given that the story in question is composed of the author’s own life experiences.
The system should thus capture relevant aspects of the author, both in promot-
ing motivation (i.e. Aim 3) and providing guidance. For the latter, there may be
benefits in the system identifying an author’s narrative or behavioral patterns
and bringing attention to them, as a counselor might during a session. This may
require some form of artificial intelligence or natural language processing, as the
system would need to analyze the author’s content and draw appropriate con-
clusions from it. Further research is needed to determine precisely what aspects
of the author should be captured, and how it should be utilized in the system.
Aim 6: The tool should allow the author to share their story. This
goal was informed by codes such as Sharing the story outside of therapy helps
maintain care, which indicated that additional support can stem from the stu-
dent sharing their personal narrative outside of an individual or therapy context.
Thus, the narrative product created by the student should be shareable with oth-
ers, such as their peers and family. However, care should be taken in giving the
author control over what aspects of their life narrative is shared.
Aim 7: The tool should be safe and positive in design. Not everyone
enters reflection in a positive state, as shown by codes such as They struggle
with exaggerating things negatively during reflection and Reflection can involve
thinking about bad times. In therapy, counselors form an important connection
with their patient (E.g. Connecting with their counselor allows students to open
up), and often have to react to serious situations and traumas in the context of
care (E.g. Counselors need to account for patient traumas). Thus, the tool’s user
experience, from visuals to system dialogue, should aim to promote positive
affect and provide a safe space for the author. However, the tool should not
pretend to be as capable as qualified professionals, and include some form of
disclaimer and easily accessible links to appropriate resources.
7 Discussion
Our implications for design describe requirements for an interactive story author-
ing tool primed to support college students through the process of therapeutic
reflection. Unlike prior digital story authoring tools for reflection, we identify
the need for a holistic approach to fully facilitate the process in a standalone,
“You Write Your Own Story” 319
self-help tool. This includes features such as an assessment phase to identify the
general context of reflection and initiate the externalization of life experiences; a
phase which fosters narrative restructuring; and the accounting for the author’s
identity within the tool itself. We also provide design implications which seek to
accommodate struggles our target population of college students has in reflec-
tion and maintaining their mental health, namely the need for features which
promote motivation or lower barriers to engaging in the process of reflection.
Such features augment recommendations from prior work on interactive sto-
rytelling for mental health, and are necessary in the context of a self-help tool,
where a user lacks the support of a counselor to both guide and motivate them.
For instance, while Mobile Story [21] and the 3MR 2 system [27] allowed for
experiences and memories to be externalized, they do not necessarily facilitate
the process of reflection from start to end, and require the intervention of a thera-
pist. On the other hand, existential video games as defined by Chittaro and Sioni
may incite reflection, but do not provide guidance through the reflection pro-
cess [9]. Our design aims instead detail how a system could be designed to fully
cover an ongoing process of narrative reflection in the lives of college students,
without the need for a therapist present. Our findings also expose questions that
future research could address. For instance, What would these narratives look
like? What form would the prose take, and how would they be connected into a
non-linear narrative by student authors? Another question that comes to mind
is How would the various prompts and specifics of the reflection interactions be
designed? We acknowledge that our study was limited by the few counselors who
took participated, due to limited access to this specific population. Lastly, our
student participants were primarily white, male, and enrolled in a STEM major
- future research would benefit from a more diverse participant pool.
8 Conclusion
We contribute aims and implications for the design of an interactive story author-
ing tool that facilitates the process of how college students reflect on their lives
as a narrative, to support their mental well-being. We conducted and analyzed
interviews with college students and college counselors, which resulted in two
maps of codes detailing their experiences in interacting with student narratives
and their mental health. These were examined to form aims for design, which
we use to propose implications for the design. We then discuss how our findings
relate to the existing literature, and identify future avenues for research. Our
work situates itself within existing technologies for mental health as a uniquely
interactive narrative approach, addressing the non-linear nature of the narratives
we construct through life reflection.
References
1. Becvar, D.S., Becvar, R.J.: Storytelling and family therapy. Am. J. Fam. Ther.
21(2), 145–160 (1993)
2. Brown, M., O’Neill, N., van Woerden, H., Eslambolchilar, P., Jones, M., John, A.:
Gamification and adherence to web-based mental health interventions: a systematic
review. JMIR Mental Health 3(3), e39 (2016)
3. Bruner, J.: The narrative construction of reality. Crit. Inq. 18(1), 1–21 (1991)
4. Bruner, J.: Narrative, culture, and mind. Telling stories: language, narrative, and
social life, pp. 45–49 (2010)
5. Butler, S., Guterman, J.T., Rudes, J.: Using puppets with children in narrative
therapy to externalize the problem. J. Mental Health Couns. 31(3) (2009)
6. Carr, A.: Michael white’s narrative therapy. Contemp. Fam. Ther. 20(4), 485–503
(1998)
7. Charmaz, K.: Constructing Grounded Theory. Sage (2014)
8. Cheng, V.W.S., Davenport, T., Johnson, D., Vella, K., Hickie, I.B.: Gamification
in apps and technologies for improving mental health and well-being: systematic
review. JMIR Mental Health 6(6), e13717 (2019)
9. Chittaro, L., Sioni, R.: Existential video games: proposal and evaluation of an
interactive reflection about death. Entertainment Comput. 26, 59–77 (2018)
10. Divinyi, J.: Storytelling: an enjoyable and effective therapeutic tool. Contemp.
Fam. Ther. 17(1), 27–37 (1995)
11. Friedberg, R.D.: Storytelling and cognitive therapy with children. J. Cogn. Psy-
chother. 8(3), 209–217 (1994)
12. Giuliani, F., Couchepin Marchetti, B., El Korh, P.: Is storytelling therapy useful
for children with autism spectrum disorders and severe mental retardation? Adv.
Tech. Biol. Med. 4(1) (2016)
13. Green, D., Hargood, C., Charles, F.: Contemporary issues in interactive storytelling
authoring systems. In: Rouse, R., Koenitz, H., Haahr, M. (eds.) ICIDS 2018. LNCS,
vol. 11318, pp. 501–513. Springer, Cham (2018). https://doi.org/10.1007/978-3-
030-04028-4 59
14. Holmbeck, G.N., Wandrei, M.L.: Individual and relational predictors of adjustment
in first-year college students. J. Couns. Psychol. 40(1), 73 (1993)
15. Kagan, R.M.: Storytelling and game therapy for children in placement. Child Care
Q. 11(4), 280–290 (1982)
16. Kraft, D.P.: One hundred years of college mental health. J. Am. Coll. Health 59(6),
477–481 (2011)
17. Lattie, E., Cohen, K.A., Winquist, N., Mohr, D.C.: Examining an app-based mental
health self-care program, intellicare for college students: single-arm pilot study.
JMIR Mental Health 7(10), e21075 (2020)
18. Lattie, E.G., Lipson, S.K., Eisenberg, D.: Technology and college student mental
health: challenges and opportunities. Front. Psych. 10, 246 (2019)
19. Launer, J.: A narrative approach to mental health in general practice. BMJ
318(7176), 117–119 (1999)
20. Lawlis, G.F.: Storytelling as therapy: implications for medicine. Altern. Ther.
Health Med. 1(2), 40–45 (1995)
21. Matthews, M., Doherty, G.: My mobile story: therapeutic storytelling for children.
In: CHI 2011 Extended Abstracts on Human Factors in Computing Systems, pp.
2059–2064 (2011)
“You Write Your Own Story” 321
22. Parker, T.S., Wampler, K.S.: Changing emotion: the use of therapeutic storytelling.
J. Marital Fam. Ther. 32(2), 155–166 (2006)
23. Reichert, E.: Individual counseling for sexually abused children: a role for animals
and storytelling. Child Adolesc. Soc. Work J. 15(3), 177–185 (1998)
24. Sas, C., Whittaker, S., Dow, S., Forlizzi, J., Zimmerman, J.: Generating implica-
tions for design through design research. In: Proceedings of the SIGCHI Conference
on Human Factors in Computing Systems, pp. 1971–1980 (2014)
25. Shapiro, J.: The use of narrative in the doctor-patient encounter. Family Syst.
Med. 11(1), 47 (1993)
26. Starks, K., Barker, D., Cole, A.: Using twine as a therapeutic writing tool for
creating serious games. In: Marsh, T., Ma, M., Oliveira, M.F., Baalsrud Hauge, J.,
Göbel, S. (eds.) JCSG 2016. LNCS, vol. 9894, pp. 89–103. Springer, Cham (2016).
https://doi.org/10.1007/978-3-319-45841-0 8
27. Tielman, M.L., Neerincx, M.A., Bidarra, R., Kybartas, B., Brinkman, W.P.: A
therapy system for post-traumatic stress disorder using a virtual agent and virtual
storytelling to reconstruct traumatic memories. J. Med. Syst. 41(8), 1–10 (2017)
28. Towbes, L.C., Cohen, L.H.: Chronic stress in the lives of college students: scale
development and prospective prediction of distress. J. Youth Adolesc. 25(2), 199–
217 (1996)
29. Wynne, E.: Storytelling in therapy and counseling. Child. Today 16(2), 11–15
(1987)
Applying Animated Parables to Gamification
in Daily Context: An Expert Evaluation Study
Kenny K. N. Chow(B)
School of Design, The Hong Kong Polytechnic University, Kowloon, Hong Kong
sdknchow@polyu.edu.hk
Abstract. Latest studies regarding gamification for behavior change have sug-
gested using meaningful stories for long-term effects. Drawing on psychological
studies related to narrative transportation and mental simulation, this paper extends
the idea of animated parables and argues that selecting stories to frame gamified
systems should consider perceived causality between the real-life action and the
simulated diegetic effect. This study employs expert evaluation on a collection of
parables regarding perceived usefulness, novelty, and perceived causality. Results
show that parables consensually rated by experts as both useful and novel are
also prominent in perceived causality, which supports the argument that perceived
causality underpins perceived persuasiveness of parables.
1 Introduction
Latest studies regarding gamification have shown that basic game elements (e.g., points,
levels, and leaderboards) can be effective in short-term interventions; for long-term
effects, they have suggested using meaningful stories to frame users’ goals [1–4]. Fic-
tional narratives have proved to be effective in changing beliefs via mental transportation
[5, 6]. Yet, experimental psychology and neuroscience research results seem to cast doubt
on transportation regarding interactive fictions [7]. Interactivity in gamification narra-
tives need to be more implicit. With ubiquitous sensing and visualization technologies,
a gamified system of today can track users’ actions in life routines and present simulated
outcomes blended in relevant contexts. One simply feels that the real-life action causes
the simulated effect in the narrative world.
Chow [8] introduces the idea of animated parables, which refers to a kind of data-
driven interactive visual narrative wherein outcomes depend on real-life actions in
daily context. Animated parables can be applied in gamified systems that support self-
management or habit formation. With proposed design guidelines and evaluation met-
rics, Chow reports the analyses of a collection of about 50 parables generated from a
series of design workshops, arguing that similarities in embodied experiences between
the real-life activity and the narrative likely render the action-outcome links easy to
interpret. This paper reports a study that extends to compare the parables in terms of
perceived causality, novelty, and usefulness. The study invites domain experts to rate the
parables, which are illustrated using the same blending diagram template. Comparison
of the scores shows that parables consensually rated as both useful and novel are also
prominent in perceived causality. Discussion about the results is given.
2 Theoretical Framework
Transportation refers to a reader’s mental journey to the fictional world of a narrative [5].
When one is “absorbed” into a story, one can picture the events, imagine being in them,
feel with the characters in the narrative, want to know how it ends and even “what if.”
Studies show that transportation is positively correlated with story-consistent beliefs,
even though one knows the story is fictional [6]. Transportation entails mental imagery,
which is powerful in persuasion. Imagining a hypothetical event and being the main
character in it has proved to increase perceived likelihood of the event [9] and affect
corresponding intention [10]. The increased likelihood estimates and intention may also
lead to behavior [11]. Hence, reading (or listening to) narratives (e.g., a hypothetical
situation) and constructing corresponding mental images have become an approach,
so-called “mental simulation,” to behavioral interventions [12–17].
Narrative transportation and mental imagery are found positively correlated with per-
ceived realism (individuals’ subjective evaluations of the plausibility of stories) even to
those without relevant prior experience [18]. Yet, some people may lack concrete sensory
images in the “mental library” for constructing particular “unseen world” [19]. Using
interactive visualization technologies can provide artificial sensory images that facili-
tate one’s imagination. For example, Chow and colleagues [20] demonstrate that through
showing an animated character on the mobile screen that becomes tired with increasing
screen time, participants’ reported imagination of mistreating the character is positively
correlated with reduction in daily screen time. Immersing individuals in virtual environ-
ments that simulate hypothetical scenarios (e.g., showering in coal, seeing the old self)
have been found effective in changing attitudes [21–23] and motivating behaviors [24].
Hence, computer-generated simulations (2D images on screens or panoramic images on
head-mounted displays) can be applied to create images of interactive narratives that
facilitate transportation.
Gamified systems with goals framed in stories can simulate outcomes in artificial sensory
images (virtual rewards or penalties) according to one’s real-life actions. Yet, many story-
based gamified apps supporting self-management (e.g., Fortune City [25], Habitica [26],
Forest [27]) rely on users’ self-input of activity logs. Self-input is “cumbersome” to keep
in a daily fashion [28]. With ubiquitous sensing and visualization technologies of today,
324 K. K. N. Chow
a gamified system can track users’ actions in life routines (e.g., flights of stairs climbed)
and present simulated effects (e.g., pictures taken when hiking) in relevant context (e.g.,
on a picture frame in office). The purpose is to let one act in real life and perceive the
simulated diegetic outcomes seemingly blended in the real world.
To maintain narrative transportation, diegetic outcomes should be relevant to real-
life actions based on the user’s knowledge. When one prepares to act, the brain tests
the reality of the action and its outcome. Failing the reality-test disrupts “suspension
of disbelief” and transportation [7]. For instance, walking in real life causing a virtual
fish to grow (e.g., Fish ‘n’ Steps [29]) is easily judged as unreal, because their relation
cannot be easily explained. Yet, if walking in real life rewards a new pair of virtual
shoes (e.g., Fitbit badges [30]), one could interpret that the old pair retired after long
journeys. Interpretable causal links can be “temporarily” treated as true (pass reality-
test) and easily imagined, which make one more confident in the truth of the causality
[31]. Chow names it “blended causality” [30] which links real-life actions to diegetic
outcomes. This blending of the non-diegetic into the diegetic, rather than breaking the
“fourth wall,” actually relocates it and echoes what Conway calls “expansion of the
magic circle” (the fictional game world) which arguably enhances immersion [32].
For interpretable and imaginable action-outcome links in gamification, Chow [8]
proposes blending the real-life activity with an imaginary cause-effect narrative wherein
they have similar causal actions. Chow introduces the idea of animated parables, which
refers to a kind of data-driven interactive visual narrative that presents positive or negative
outcomes metaphorically mapped with real-life actions. Chow reports the generation of
parables from a series of design workshops according to proposed design guidelines,
followed by evaluation based on proposed conceptual metrics. Drawing on embodied
cognition [33], Chow argues with analyses that similarities in embodied experiences
(embodied mapping) between the real-life activity and the simulated narrative render
better parables in terms of the metrics.
Each animated parable in Chow [8] is generated through blending a real-life activity
and a cause-effect narrative. The two inputs share similarities in terms of sensorimotor
(visual, auditory, kinesthetic, etc.) and spatiotemporal (spatial relations like container or
proximity) experiences. Meanwhile, differences between the activity and the narrative
are also crucial for novelty. When the differing parts of the narrative are simulated and
blended into daily context (e.g., a mountain top photo displayed after climbing stairs
to office), the user can interpret the parable through the similarities (e.g., climbing)
meanwhile appreciating the nuances (e.g., different views between mountains and urban
cities), which render the gamified system more intriguing and appealing. Thus, novel
parables benefit system adoption and adherence.
Usefulness not only means that a system effectively meets the user’s needs, but
also that it is easy or convenient to use. Usefulness can be assessed via user evaluation
on minimum viable prototypes. In earlier stages of design processes, perceived useful-
ness [34] can be assessed via expert evaluation on proof-of-concept prototypes (e.g.,
visualized use-case scenarios), which provides health check before proceeding to more
Applying Animated Parables to Gamification in Daily Context 325
3.1 Participants
Seven expert participants have been recruited. Four of them are design practitioners with
relevant work experiences ranging from five to thirty years. Two are researchers in Design
and Technology respectively, with sixteen and fifteen years of relevant experience. One
is educator in Linguistics with fifteen-year experience. Their self-ratings on knowledge
regarding creativity and metaphor fall between three and four on a five-point scale (five
is highest).
There are totally 51 parables in this study, wherein the first parable alludes to the classical
imaginative play “Trashcan Basketball” for comparisons with other parables from Chow
[8]. Each parable is visualized using the author-modified conceptual blending diagram
template. It starts with two horizontally aligned circles respectively representing a real-
life activity (B) and a hypothetical event (C). They are blended into a circle below
them representing the parable (P), which is elaborated into positive and negative action-
outcome paths underneath. Figure 1 is an example parable “Staircase Hiking” that blends
climbing stairs with hiking. It rewards the user with a beautiful mountain view whenever
one walks up the stairs or shows a doddery image in case of using the elevator.
326 K. K. N. Chow
3.3 Measures
Each parable is first presented with the visualizations of (B) and (C), followed by a
statement saying “a product that rewards you with” a positive outcome in the parable
(e.g., a beautiful mountain view) whenever “you” perform a positive action (e.g., walk
up the stairs).
Novelty. The other major criterion in the sociocultural definition of creativity is novelty
[35]. In this survey, expert participants were asked to respond to the question: “How
novel do you judge this idea?” on a seven-point scale ranging from “not novel at all” to
“extremely novel”.
Perceived Causality. The survey then presents visualizations of the positive and nega-
tive action-outcome paths of the parable. Expert participants were then asked to indicate
the level of agreement to the two statements, one about the positive outcome (e.g., “I feel
like walking up the stairs causes a lively body image on the mountain view to appear”)
and the other about the negative outcomes (“I feel like taking the elevator causes only a
doddery image to appear”). Five options were given: not at all, not much, moderately,
quite much, and absolutely. They corresponded to a five-point scale ranging from 1 to 5.
Applying Animated Parables to Gamification in Daily Context 327
4 Results
Responses from the seven expert participants to the survey were collected and aggre-
gated. To detect consensus among our expert participants, parables with high mean
scores and low standard deviations over each measure were selected. On the seven-point
scale, the selection criterion is mean score minus one standard deviation staying above
four. On the five-point scale, the selection criterion is mean score minus one standard
deviation staying above three. Table 1 shows all selected parables with their descriptions
as presented in the survey. Table 2 shows their mean scores and standard deviations over
the corresponding measures.
Table 1. Parables that championed in the expert evaluation and their descriptions.
The juxtaposition across the measures shows that a few parables excel at two or
more attributes. Four parables, namely 12. Ice Breaker Scale, 14. Smoked House, 22.
328 K. K. N. Chow
Table 2. Selected parables with high mean scores and low standard deviations in each measure.
Perceived causality includes positive (+) or negative (−) outcome measures.
Table 2. (continued)
Staircase Hiking, and 29. Tetridge, have high rating in both usefulness (appropriateness
as well) and novelty. Meanwhile, our expert participants are consensually able to feel
the causal links in these four creative parables.
5 Discussion
The list under appropriateness largely overlaps with that under usefulness. The socio-
cultural definition of creativity informs that a creative product should be judged to be
“novel” on one hand, and “appropriate, useful, and valuable” on the other [35]. Appro-
priateness sounds like the basic requirement, followed by usefulness and then value. In
our survey, expert participants were asked to judge the appropriateness of an imagined
product that embodies the parable idea, whereas usefulness was assessed in relation
to target users. With the target users in their minds, expert participants could be more
critical in assessing usefulness.
The four creative parables that are rated highly novel and useful (appropriate as well)
also score high in perceived causality. This aligns with the embodied mapping analyses
by Chow [8]. These four parables consist of structural similarities (i.e., image schemas
[40] e.g., path, containment, proximity, support) directly relating the causal actions (e.g.,
walk up from a low to a high level, fill or empty a confined space, move near to a surface,
step on a plane) to the simulated diegetic outcomes (e.g., a view from a mountain top, the
filled fridge, dust on the picture, cracks on the ice). They also show surface similarities
(by design) (e.g., mountains, tetrominos, dust/dirt, cracks) that prompt the corresponding
narrative frame (e.g., hike, tetris, air pollution, ice cracking) in participants. The nuances,
330 K. K. N. Chow
Results of this study show that parables consensually rated as both useful and novel are
also prominent in perceived causality, which supports that perceived causality underpins
perceived persuasiveness. Discussion suggests that structural similarity can facilitate
perceived causality, and surface similarity by design increases novelty. In this study,
the standard deviations of some measures are high. Future work includes revising the
visualizations of those parables with low perceived causality and examining the profiles
of those participants with scores deviating from the means in hope of identifying key
expertise that might lead to better consensus in assessments.
Acknowledgements. This research benefits from projects supported by The Hong Kong
Polytechnic University.
References
1. Laschke, M., Hassenzahl, M.: Mayor or patron? The difference between a badge and a
meaningful story. In: CHI 2011. ACM (2011)
2. Zuckerman, O., Gal-Oz, A.: Deconstructing gamification: evaluating the effectiveness of con-
tinuous measurement, virtual rewards, and social comparison for promoting physical activity.
Pers. Ubiquit. Comput. 18(7), 1705–1719 (2014). https://doi.org/10.1007/s00779-014-0783-2
3. Mekler, E.D., Brühlmann, F., Tuch, A.N., Opwis, K.: Towards understanding the effects of
individual gamification elements on intrinsic motivation and performance. Comput. Hum.
Behav. 71, 525–534 (2017)
4. Rapp, A.: Drawing Inspiration from world of warcraft: gamification design elements for
behavior change technologies. Interact. Comput. 29, 648–678 (2017)
5. Gerrig, R.J.: Experiencing Narrative Worlds: On the Psychological Activities of Reading.
Yale University Press, New Haven (1993)
6. Green, M.C., Brock, T.C.: The role of transportation in the persuasiveness of public narratives.
J. Pers. Soc. Psychol. 79, 701–721 (2000)
Applying Animated Parables to Gamification in Daily Context 331
7. Holland, N.N.: Spider-Man? Sure! The neuroscience of suspending dislief. Interdisc. Sci.
Rev. 33, 312–320 (2008)
8. Chow, K.K.N.: Crafting animated parables: an embodied approach to representing lifestyle
behaviors for reflection. Digit. Creat. 32, 1–21 (2021)
9. Sherman, S.J., Cialdini, R.B., Schwartzman, D.F., Reynolds, K.D.: Imagining can heighten
or lower the perceived likelihood of contracting a disease: the mediating effect of ease of
imagery. Pers. Soc. Psychol. Bull. 11, 118–127 (1985)
10. Anderson, C.A.: Imagination and expectation: the effect of imagining behavioral scripts on
personal intentions. J. Pers. Soc. Psychol. 45, 293–305 (1983)
11. Gregory, W.L., Cialdini, R.B., Carpenter, K.M.: Self-relevant scenarios as mediators of like-
lihood estimates and compliance: does imagining make it so? J. Pers. Soc. Psychol. 43, 89–99
(1982)
12. Pham, L.B., Taylor, S.E.: From thought to action: effects of process- versus outcome-based
mental simulations on performance. Pers. Soc. Psychol. Bull. 25, 250–260 (1999)
13. Wynd, C.A.: Guided health imagery for smoking cessation and long-term abstinence. J. Nurs.
Scholarsh. 37, 245–250 (2005)
14. Armitage, C.J., Reidy, J.G.: Use of mental simulations to change theory of planned behaviour
variables. Br. J. Health. Psychol. 13, 513–524 (2008)
15. Knäuper, B., McCollam, A., Rosen-Brown, A., Lacaille, J., Kelso, E., Roseman, M.: Fruit-
ful plans: adding targeted mental imagery to implementation intentions increases fruit
consumption. Psychol. Health 26, 601–617 (2011)
16. Chan, C.K.Y., Cameron, L.D.: Promoting physical activity with goal-oriented mental imagery:
a randomized controlled trial. J. Behav. Med. 35(3), 347–363 (2012). https://doi.org/10.1007/
s10865-011-9360-6
17. Conroy, D., Sparks, P., Visser, R.D.: Efficacy of a non-drinking mental simulation intervention
for reducing student alcohol consumption. Br. J. Health Psychol. 20, 688–707 (2015)
18. Green, M.C.: Transportation into narrative worlds: the role of prior knowledge and perceived
realism. Discourse Process. 38, 247–266 (2004)
19. Bailenson, J.: Experience on Demand: What Virtual Reality Is, How It Works, and What It
Can Do. W. W. Norton & Company (2018)
20. Chow, K.K.N., Leong, B.D., Lee, B.Y.H.: Imagining consequences of excessive smartphone
use via a character-based mobile application. Int. J. Ment. Heal. Addict. 16(6), 1420–1434
(2018). https://doi.org/10.1007/s11469-018-9984-7
21. Ahn, S.J.G., Bailenson, J.N., Park, D.: Short- and long-term effects of embodied experiences
in immersive virtual environments on environmental locus of control and behavior. Comput.
Hum. Behav. 39, 235–245 (2014)
22. Bailey, J.O., Bailenson, J.N., Flora, J., Armel, K.C., Voelker, D., Reeves, B.: The impact of
vivid messages on reducing energy consumption related to hot water use. Environ. Behav. 47,
570–592 (2015)
23. Ahn, S.J.G., Bostick, J., Ogle, E., Nowak, K.L., McGillicuddy, K.T., Bailenson, J.N.: Experi-
encing nature: embodying animals in immersive virtual environments increases inclusion of
nature in self and involvement with nature. J. Comput.-Mediat. Commun. 21, 399–419 (2016)
24. Hershfield, H.E., et al.: Increasing saving behavior through age-progressed renderings of the
future self. J. Mark. Res. 48, S23–S37 (2011)
25. https://fortunecityapp.com/en/
26. https://habitica.com/static/home
27. https://www.forestapp.cc/
28. Rapp, A., Cena, F.: Personal informatics for everyday life: how users without prior self-
tracking experience engage with personal data. Int. J. Hum Comput Stud. 94, 1–17 (2016)
332 K. K. N. Chow
29. Lin, J.J., Mamykina, L., Lindtner, S., Delajoux, G., Strub, H.B.: Fish’n’Steps: encouraging
physical activity with an interactive computer game. In: Dourish, P., Friday, A. (eds.) UbiComp
2006. LNCS, vol. 4206, pp. 261–278. Springer, Heidelberg (2006). https://doi.org/10.1007/
11853565_16
30. Chow, K.K.N.: Toward a language of blended causality for transforming behavioral data into
reflective user experiences. In: The ACM Conference on Creativity and Cognition. ACM
Press (2019)
31. Koehler, D.J.: Explanation, imagination, and confidence in judgment. Psychol. Bull. 110,
499–519 (1991)
32. Conway, S.: A circular wall? Reformulating the fourth wall for videogames. J. Gaming Virtual
Worlds 2, 145–155 (2010)
33. Lakoff, G.: Explaining embodied cognition results. Top. Cogn. Sci. 4, 773–785 (2012)
34. Davis, F.D.: Perceived usefulness, perceived ease of use, and user acceptance of information
technology. MIS Q. 13, 319–340 (1999)
35. Sawyer, R.K.: Explaining Creativity: The Science of Human Innovation. Oxford University
Press, New York (2012)
36. Boden, M.A.: The Creative Mind: Myths and Mechanisms. Routledge, London (2004)
37. Boden, M.A.: Computer models of creativity. AI Magazine, vol. Fall 2009 (2009)
38. Amabile, T.M.: Social psychology of creativity: a consensual assessment technique. J. Pers.
Soc. Psychol. 43, 997–1013 (1982)
39. Getzels, J.W., Csikszentmihalyi, M.: Problem finding and creativity. In: The Creative Vision:
A Longitudinal Study of Problem Finding in Art, pp. 236–251. Wiley, New York (1976)
40. Lakoff, G., Johnson, M.: Philosophy in the Flesh: the Embodied Mind and its Challenge to
Western Thought. Basic Books, New York (1999)
Creativity and Collaboration During
COVID-19: Creating an Alternate Reality Game
in the Face of a Pandemic
1 Introduction
In-person collaborative environments have been drastically altered during the COVID-
19 pandemic. In hopes of preserving the spirits of undergraduate students who may be
experiencing university life for the first time amidst a pandemic, the team set out to inspire
collaboration among University of Central Florida (UCF) students in a creative way.
The alternate reality game (ARG) the team designed blurred the lines between the real
world and fantasy by inviting undergraduate students to become members of the Fairy
Investigation Club (FIC), an unofficial club at the university. Alternate reality games
began growing in popularity during the early 2000s [7]. These games deliberately blurred
the lines between the in-game and out-of-game experience for players, which is what the
team hoped to mimic with the creation of the fictional club. In Rethinking the Library
Game: Creating an Alternate Reality with Social Media, the Puppet Masters, or the ones
who typically release the puzzles in response to real-time gameplay, demonstrated that
leveraging social media in ARGs can be beneficial in gaining an audience and creating
a more realistic feel to the game [2]. Likewise, the UCF team took advantage of popular
platforms such as Instagram and Discord to advertise the game and enhance gameplay.
The team took on the task of solving the question: “how could we design an alternate
reality game for undergraduate students that affords them an interactive and entertaining
experience in the face of a pandemic?” To do this, each department focused on what the
ARG would look like, how it would function on a technical level, and how to structure
the puzzles and narrative in a way that challenged players to think creatively about
solutions. It was through an iterative design process that the team refined the story, the
aesthetics, and the overall gameplay experience that players would have [3]. After seven
weeks of ideaphoria, the final game took an additional seven weeks to complete. That
time was spent creating graphical assets, modifying the complexity of the puzzle games,
and integrating everything into an interactive web-based experience. By designing and
branding the game as a “club,” it allowed players to feel a sense of belonging from the
start by fictitiously joining this club during a time where joining a real club on campus
was impossible [4].
To avoid large social gatherings, the team hosted a pre-release scavenger hunt where
3D-printed gnomes were placed in discrete locations on campus with links of the website
on them. Hints to locate the gnomes were left via cryptic captions on Instagram. This
advertising tool was a deliberate design feature that the team included which allowed
players to explore campus in a creative way that did not require face-to-face interactions.
The gnomes were all taken, leaving the team satisfied that whoever discovered them was
exposed to information about the FIC whether they simply stumbled upon them or located
them through Instagram.
Informal user testing was utilized to gain insight into the effectiveness of each game
aspect. Playtests were conducted over the course of a week where members of the team
met virtually with a total of 10 play testers to observe them play through the game
in real-time. Time on task, error rates, and perceived enjoyment of the game were all
recorded during these sessions. The playtest included a 7-point Likert scale pre-survey
Creativity and Collaboration During COVID-19 335
Fig. 1. Flow chart diagram of the narrative blog posts and puzzles in order of which they were
released.
2.3 Gameplay
The FIC took a whimsical design approach while still maintaining the common themes
of mystery and investigation typically found in ARGs [1, 5, 8]. Through the club website,
players were able to familiarize themselves with characters, such as the Club President,
and others who each had a unique personality and role in the FIC. At the start of the
336 D. L. Faverio et al.
game, it appears as though players are simply members of a group aiming to observe
the many mystical fairy sightings on the university’s campus until circumstances grow
increasingly more peculiar as Aerin, the Club President, vanishes and club members must
solve puzzles to bring her home (Fig. 2). The online modality of the ARG contributed
to allowing players to participate in the game whenever and wherever they were located
during this time.
Fig. 2. Screenshots from the FIC blog website depicting two of the puzzles.
The final game consisted of a series of ten tasks, communicated through puzzles
and blog posts that were released over the course of five days, commencing after the
university’s Spring Break in mid-April of 2021. The ARG features a well-known Fae
character, Puck, or Robin Goodfellow, from Shakespeare’s A Midsummer Night’s Dream
as the main antagonist. This adds to the narrative by introducing familiar aspects of fairies
in a contemporary way. Since fairies are historically described as mischievous pranksters,
tying that into the riddles and puzzles was a calculated effort. Comments written from
the viewpoint of the characters were featured in each blog post which adds depth to the
gameplay and plays a role in later puzzles.
3 Conclusion
Through the Fairy Investigation Club ARG, the team was able to create a whimsical
reality for players to enjoy despite the current circumstances of the pandemic. Living
in a time where a global pandemic hindered the traditional ways of collaboration and
participation on college campuses, the team hoped to create a game that could still be
enjoyed and generate a sense of togetherness. The conversations players had on Discord,
such as sharing stories of their fairy encounters, demonstrated to the team a sense of
group interaction. It would be helpful in future games to potentially take a poll of what
site to message on once the game begins to ensure that a majority of people are able to
participate in this aspect of the gameplay more easily. The entire team hopes that despite
the tribulations encountered during the game’s design process, this experience may
challenge other digital media professionals to seek alternate methods of collaborating
with their peers and to think more deeply about how they might create meaningful
experiences for their users.
Creativity and Collaboration During COVID-19 337
References
1. Alice & Smith: The Black Watchmen. Game [PC]. Alice & Smith, Montreal, Quebec, Canada,
25 June 2015 (2015)
2. Battles, J., Battles, J., Glenn, V., Shedd, L.: Rethinking the library game: creating an alternate
reality with social media. J. Web Librariansh. 5(2), 114–131 (2011). https://doi.org/10.1080/
19322909.2011.569922
3. Bonsignore, E., Hansen, D., Kraus, K., Visconti, A., Fraistat, A.: Roles people play: key roles
designed to promote participation and learning in alternate reality games. In: Proceedings of
the 2016 Annual Symposium on Computer-Human Interaction in Play (2016). https://doi.org/
10.1145/2967934.2968108
4. Evans, E., Christopherson, L., Strum, B., King, E., Haefele, C.: Alternate reality games: a
realistic approach to gaming on campus. In: Proceedings of the 38th Annual Fall Conference
on SIGUCCS - SIGUCCS 2010 (2010). https://doi.org/10.1145/1878335.1878376
5. Mihell & Lycos: Forever has Fallen. Game [Web]. Mihell & Lycos, Adelaide, Australia, 5
September 2020 (2020)
6. O’Hara, K., Grain, H., Williams, J.: Participation, collaboration, and spectatorship in an alter-
nate reality game. In: Proceedings of the 20th Australasian Conference on Computer-Human
Interaction Designing for Habitus and Habitat - OZCHI 2008 (2008). https://doi.org/10.1145/
1517744.1517787
7. Piatt, K.: Using alternate reality games to support first year induction with ELGG. Campus-
Wide Inf. Syst. 26(4), 313–322 (2009). https://doi.org/10.1108/10650740910984646
8. O’Connell, W.: Subtext. Game [Mobile]. William O’Connell, Alexandria, Virginia (2018)
Interacting with Climate Change: A Survey
of HCI and Design Projects and Their Use
of Transmedia Storytelling
Abstract. Climate change is arguably one of the most debated issues today. The
scale and global reach of this crisis doesn’t afford a universal solution and requires
widespread global mobilization. Public engagement is essential for the success of
any initiative on this topic. However, sometimes communicating the facts is not
enough. Interactive storytelling and transmedia narratives have an important part
to play in communicating climate change, especially in shifting from a mere trans-
mission of data to a narrative that is more engaging, positive and action focused,
that considers diverse audiences and active participation. Following this premise,
we conducted a survey on climate change applied research projects addressing the
general public to understand how the fields of interactive media, HCI and Design
are using transmedia narratives. The intention of this study is to gather what has
been done regarding these topics and the strategies used, to further the debate
among the community and inform future research.
of the issue [5, 6], but also in associating the topic with the frames of fear, anxiety
and even hopelessness [7, 8]. Climate change is now a much debated and mediated
issue, but arguably what we need now are effective ways of engaging the general public
with positive, action-focused exchanges that will lead to action [9–11]. We argue that
communication techniques such as storytelling and transmedia can play a crucial part in
shifting the dialogue from the negative, defeatist tone into a more positive, entertaining,
engaging, and inclusive one. Therefore, understanding what has been done to date will
help position future work in these areas.
Interactive digital narratives (IDN) is a rapidly evolving area of research [12, 13] and
transmedia narratives can contribute to the development of the field, especially when
dealing with applied research. Previous work [14] has analysed how the HCI research
community has been addressing climate change to the general public, especially their
interaction, storytelling and media choices. In this study, we take this analysis further by
doing a survey on interactive media, HCI and Design applied research focused on climate
change interactions and looking specifically into their media strategies, with a special
focus on transmedia. This analysis used the Grounded Theory Literature Review method
[15]. By looking into the databases of ICIDS, SIGCHI, Interact, DRS, TDJ, JDR, DS,
and She Ji it returned 680 results for the keywords “climate change”, climate crisis” and
“global warming” for Jan. 2010 to Dec. 2020. From these results, we gathered N = 40
interaction projects that used storytelling. A keyword analysis points to a limited focus
in narrative-related topics. Further inquiry on their media strategies concluded that n =
33 use mono-media, n = 2 use multimedia, and n = 5 use transmedia. We focused on
the five transmedia projects to gather their narrative choices based on current proposals
for effective climate change communication.
and immersive media [20]. Environmental topics, including climate change, are already
being prolifically explored through this interactive venue.
The complexity of the issue and the diversity of people being engaged means that no
single story will trigger widespread attention [21]. Likewise, physical vulnerability in
general does not impact people’s risk perceptions, but political or ideological affiliation
and connection to local impacts of climate change does [22]. These results suggest
that crafting a narrative linking climate change to local impacts is more effective than
generalised information.
When discussing the importance of a more positive and action-focused story asso-
ciated with climate change, highlighting paths for action are an important step in the
interaction. If the information is not accompanied by specific recommendations on how
to actually respond, the audience might just accept the fatalism of the situation and won’t
be prompted for action [23]. Furthermore, overdramatization can easily be exploited by
skeptics to discredit the importance of the issue [24]. Pratten [9] advocates for this action
and solution focused approach in a call for transmedia for change (T4C) stating that sto-
rytellers should engage the audience with positive messages that inspire and motivate,
and that they should provide solutions to guide the audience in their path towards change.
Going forth, climate change storytelling is increasingly focusing on this need for
change and action, and the best methods to achieve these goals. The present analy-
sis of applied interaction experiences about climate change intends to understand how
researchers have been developing these narratives, including if the message focuses on
generalized or local issues, if it presents actionable steps, and if the storytelling/narrative
process is a focus of the research.
of people with different social backgrounds, interests, and even political affiliations.
Each part of the story is a possible entry point to the discussion. Henry Jenkins calls
transmedia storytelling “entertainment for the age of media convergence” [27], an age
when computing, communication and content are brought together as a consequence
of the digitization of media content and the ubiquitousness of the Internet. This media
convergence transformed the established media landscape and allowed for the emergence
of entirely new forms of content and storytelling. Furthermore, he presents this type
of transmedia experience as entertainment for the age of collective intelligence – a
shared or group intelligence that comes from collaboration and collective efforts from
many individuals. This prescient notion has taken a deeper meaning recently, with many
scholars arguing for more inclusive, plural, socially aware perspectives in sustainability
and climate change research that account for participation, co-production and collective
action [29–36], and for community and social interaction [37, 38]. The need for systemic
change highlights the importance of these strategies [39–46]. Besides, digital media
practitioners now need to work on services, experiences and networks that take into
consideration these complex socio-technical systems [47].
In any transmedia strategy, the media literacy of the audience in question needs to be
carefully considered. In a world of participatory culture, for example, true engagement
comes only from active participation. If the person addressed does not possess the tools
to actively take part in the exchange, then communication fails. However, if the medium
is adequate for the public it addresses, the possibilities are endless. Users find transmedia
stories engaging and versatile, evoking creativity and collaboration [48]. Participatory
mechanisms allow for constant feedback and participation from the audience. Especially
for social causes, this willingness for participation can be harnessed to help solve prob-
lems [49]. The potential offered by interactive media allows for storytelling to explore
more deeply the intersection between personal experience and community action, and
for a complex topic like climate change, this is a crucial factor to explore.
Nowadays, communicators deal with a complex media landscape where multiple
media channels, shared authorship, social networks, sharing and interaction, are an
everyday reality. Interactivity and entertainment should be leveraged to increase public
engagement and allow for diverse pathways for different audiences to enter the con-
versation. With this in mind, we set out to understand if and how researchers are using
transmedia in climate change related interactions.
and She Ji: The Journal of Design, Economics, and Innovation. These libraries allow for
a multidisciplinary, comprehensive scope with an international representation.
The data selection criteria were: (1) climate change as the main topic of the project; (2)
target audience: a general public outside academia; (3) projects that have an interaction
component; (4) projects that use storytelling to convey a message.
The analysis was restricted to the past decade – from Jan. 2010 to Dec. 2020 – for
timewise relevant results, and to the search terms “climate change”, “climate crisis” or
“global warming” to focus on climate change related narratives.
The initial search returned 680 results, divided as follows: ICIDS: 7; SIGCHI: 395;
Interact: 38; DRS: 102; TDJ: 58; JDR: 17; DS: 16; She Ji: 47. Each result was scrutinized
through its title and abstract. If it mentioned an interaction or communication project, it
was added to the list. This list was then refined by checking if the projects corresponded
to the first three points of the selection criteria, resulting in a final list of 77 projects:
ICIDS: 3; SIGCHI: 43; Interact: 9; DRS: 12; TDJ: 6; JDR: 1; DS: 2; She Ji: 1. Lastly, we
analyzed if the projects used storytelling to convey a message – criteria four – and ended
up with a final data set of 40 projects: ICIDS: 3; SIGCHI: 26; Interact: 2; DRS: 4; TDJ:
4; JDR: 1; DS: 0; She Ji: 0. Therefore, from the initial list of 680 results of papers and
articles that mentioned climate change, 40 (5,9%) mentioned applied projects for the
general public that use storytelling strategies to communicate a message. In the following
section, we present further questions asked of this final dataset with the purpose of better
understanding the storytelling strategies used.
Since our dataset is already a selection of applied research that uses some form of story-
telling to convey information, we intended to better understand the strategies connected
to the use of interactive media. Therefore, we asked of our data set if the projects used
one of the following media strategies: a) mono-media – one media output, either digital
Interacting with Climate Change: A Survey of HCI and Design Projects 343
1. London Phenological Clock [50]: a) Uses analogue clocks and a data visualization
website to represent the annual timing of life cycle events for species in urban
ecosystems. b) Transmedia franchise strategy. c) Local impacts: local plants and
animal’s characteristics and habits. d) Participation component: the data is gathered
by citizen scientists or volunteers from their neighborhood.
2. Vox Populi [51]: a) Uses a card game, software package support, and theatre perfor-
mance to build an interactive narrative about the importance of media for elections,
fake news, and the refugee crisis. This project is part of the ongoing transmedia sto-
ryworld “Shatterland”. b) Transmedia portmanteau strategy. c) Generalized topics
(even though based in two European cities for an European context). d) Participation
component: players have an active role inside the interactive story.
3. Spilltime [52]: a) Formed of three objects, each focusing on a different form of
showcasing and experiencing one’s carbon footprint – a water tank, a wearable,
and a mirror, also with the support of a coach. b) Transmedia franchise strategy. c)
Generalized topics. d) Participation component: data based on user action.
4. Angstfabriek [53]: a) An interactive installation that uses different media to tell a
story about people’s fears, including climate change – VR, corporate-like videos,
QR code scans, analogue elements in lockers, etc., as part of the same interactive
narrative. b) Transmedia portmanteau strategy. c) Generalized topics. d) Participation
component: audience is part of the live experience and plays a part in the story.
5. Econundrum [54]: a) A sculpture and an app to input data and allow for user partic-
ipation. The physical data sculpture represents the small community’s food habits
and the consequent carbon emissions. b) Transmedia portmanteau strategy c) Local
impacts: individual’s choices and their impact. d) Participation component: data
inputted by the participants.
The five projects are quite balanced in terms of transmedia strategy used with n =
3 for Transmedia portmanteau, and n = 2 for Transmedia franchise. Likewise for the
focus on local or generalized topics, with n = 3 for generalized and n = 2 for local. In
terms of audiences, Spilltime considers the advantages of having different objects with
different characteristics for different users – one possibly appealing more to an 80-year-
old person while another to a seven-year-old child. The London Phenological Clock,
being a physical installation and a data visualization website, allows for interaction
344 M. Ferreira et al.
with different types of users in different contexts. The Vox Populi interactive narrative,
the Angstfabriek installation, and the Econundrum sculpture are experiences where the
different media work for the same users. Nevertheless, having different outputs with
varying levels of complexity and participation can help in considering users with varying
media literacies even within the same experience.
References
1. Wuebbles, D.J., Fahey, D.W., Hibbard, K.A., Dokken, D.J., Stewart, B.C., Maycock, T.K.:
Climate science special report: fourth National Climate Assessment. U.S. Global Change
Research Program (2017). https://doi.org/10.7930/J0J964J6
2. IPCC: Special Report on Global Warming of 1.5C (2018)
3. UNEP: Single-Use Plastic: A Roadmap for Sustainability (2018)
4. Climate Outreach: Climate Visuals. https://climatevisuals.org/. Accessed 01 Oct 2020
5. Leiserowitz, A., Maibach, E., Rosenthal, S., Kotcher, J.: Climate Change in the American
Mind, November 2019 (2019)
6. European Commission: Citizen support for climate action—Climate Action. https://ec.eur
opa.eu/clima/citizens/support_en. Accessed 09 Apr 2020
7. Knight, V.: ‘Climate Grief’: Fears About the Planet’s Future Weigh on Americans’ Mental
Health—Kaiser Health News. https://khn.org/news/climate-grief-fears-about-the-planets-fut
ure-weigh-on-americans-mental-health/. Accessed 09 Apr 2020
8. Clayton, S., Manning, C.M., Krygsman, K., Speiser, M.: Mental Health and Our Changing
Climate: Impacts, Implications, and Guidance. American Psychological Association, and
ecoAmerica. American Psychological Association, Washington, D.C. (2017)
9. Pratten, R.: Transmedia for Change. https://www.linkedin.com/pulse/transmedia-change-rob
ert-pratten/
10. Corner, A., Lewandosky, S., Phillips, M., Roberts, O.: The Uncertainty Handbook. University
of Bristol, Bristol (2015)
11. Corner, A., Shaw, C., Clarke, J.: Principles for effective communication and public engage-
ment on climate change: A Handbook for IPCC authors. Climate Outreach, Oxford
(2018)
12. Murray, J.H.: Research into interactive digital narrative: a kaleidoscopic view. In: Rouse, R.,
Koenitz, H., Haahr, M. (eds.) ICIDS 2018. LNCS, vol. 11318, pp. 3–17. Springer, Cham
(2018). https://doi.org/10.1007/978-3-030-04028-4_1
13. Koenitz, H., Dubbelman, T., Knoller, N., Roth, C.: An integrated and iterative research direc-
tion for interactive digital narrative. In: Nack, F., Gordon, A.S. (eds.) ICIDS 2016. LNCS, vol.
10045, pp. 51–60. Springer, Cham (2016). https://doi.org/10.1007/978-3-319-48279-8_5
346 M. Ferreira et al.
14. Ferreira, M., Coelho, M., Nisi, V., Nunes, N.: Climate change communication in HCI: a visual
analysis of the past decade. Presented at the C&C 2021: Creativity and Cognition, June 2021.
https://doi.org/10.1145/3450741.3466774
15. Wolfswinkel, J.F., Furtmueller, E., Wilderom, C.P.M.: Using grounded theory as a method for
rigorously reviewing literature. Eur. J. Inf. Syst. 22, 45–55 (2013). https://doi.org/10.1057/
ejis.2011.51
16. Bruner, J.: Acts of Meaning. Harvard University Press, Cambridge (1990)
17. Spence, A., Pidgeon, N.: Psychology, climate change & sustainable behaviour. Environment
(2009). https://doi.org/10.1080/00139150903337217
18. Moser, S.C.: More bad news: the risk of neglecting emotional responses to climate change
information. In: Creating a Climate for Change (2009). https://doi.org/10.1017/cbo978051
1535871.006
19. Nisbet, M.C.: Framing science: a new paradigm in public engagement. In: Communicating
Science: New Agendas in Communication (2009). https://doi.org/10.4324/9780203867631
20. Games for Change, Inc.: Games for Change. https://www.gamesforchange.org/. Accessed 14
Oct 2021
21. Nisbet, M.C.: Communicating Climate Change: Why Frames Matter (2009)
22. Pidgeon, N.: Climate change risk perception and communication: addressing a critical
moment? Risk Anal. (2012).https://doi.org/10.1111/j.1539-6924.2012.01856.x
23. Maibach, E.W., Roser-Renouf, C., Leiserowitz, A.: Communication and marketing as climate
change-intervention assets. A public health perspective. Am. J. Prev. Med. (2008). https://
doi.org/10.1016/j.amepre.2008.08.016
24. Revkin, A.C.: Climate change as news: Challenges in communicating environmental science.
In: Climate Change: What it Means for Us, Our Children, and Our Grandchildren, Second
Edition (2014). https://doi.org/10.7551/mitpress/9178.003.0008
25. Javanshir, R., Carroll, B., Millard, D.: Structural patterns for transmedia storytelling. PLoS
ONE 15, e0225910 (2020). https://doi.org/10.1371/journal.pone.0225910
26. Jenkins, H.: Transmedia Storytelling 101. http://henryjenkins.org/blog/2007/03/transmedia_s
torytelling_101.html. Accessed 17 June 2021
27. Jenkins, H.: Convergence Culture: Where Old and New Media Collide. New York University
Press, New York and London (2006)
28. Pratten, R.: Getting Started in Transmedia Storytelling: A Practical Guide for Beginners
(2015)
29. Bastian, M. (ed.): Participatory Research in More-than-Human Worlds. Routledge, Taylor &
Francis Group, London, New York (2017)
30. Clarke, R., Heitlinger, S., Light, A., Forlano, L., Foth, M., DiSalvo, C.: More-than-human
participation: design for sustainable smart city futures. Interactions 26, 60–63 (2019). https://
doi.org/10.1145/3319075
31. Tsekleves, E., Darby, A., Ahorlu, C., Pickup, R., de Souza, D., Boakye, D.: Challenges and
opportunities in conducting and applying design research beyond global north to the global
south. Presented at the Design Research Society Conference, 10 September 2020 (2020).
https://doi.org/10.21606/drs.2020.145
32. Wong-Villacres, M., et al.: Decolonizing learning spaces for sociotechnical research and
design. In: Conference Companion Publication of the 2020 on Computer Supported Coop-
erative Work and Social Computing, pp. 519–526. ACM, Virtual Event USA (2020). https://
doi.org/10.1145/3406865.3418592
33. Clarke, R., Heitlinger, S., Foth, M., DiSalvo, C., Light, A., Forlano, L.: More-than-human
urban futures: speculative participatory design to avoid ecocidal smart cities. In: Proceedings
of the 15th Participatory Design Conference: Short Papers, Situated Actions, Workshops and
Tutorial - Volume 2, pp. 1–4. ACM, Hasselt and Genk Belgium (2018). https://doi.org/10.
1145/3210604.3210641
Interacting with Climate Change: A Survey of HCI and Design Projects 347
34. DiSalvo, C.: Critical making as materializing the politics of design. Inf. Soc. 30, 96–105
(2014). https://doi.org/10.1080/01972243.2014.875770
35. DiSalvo, C., Dantec, C.A.L.: Civic design. Interactions 24, 66–69 (2017). https://doi.org/10.
1145/3137097
36. Le Dantec, C.A.: Design through collective action / collective action through design.
Interactions 24, 24–30 (2016). https://doi.org/10.1145/3018005
37. DiSalvo, C.: Community and conflict. Interactions 18, 24–26 (2011). https://doi.org/10.1145/
2029976.2029984
38. Lou, Y.: Designing interactions to counter threats to human survival. She Ji: J. Des. Econ.
Innov. 4, 342–354 (2018). https://doi.org/10.1016/j.sheji.2018.10.001
39. Blevis, E., et al.: Ecological perspectives in HCI: promise, problems, and potential. In: Pro-
ceedings of the 33rd Annual ACM Conference Extended Abstracts on Human Factors in
Computing Systems, pp. 2401–2404. ACM, Seoul Republic of Korea (2015). https://doi.org/
10.1145/2702613.2702634
40. Brynjarsdottir, H., Håkansson, M., Pierce, J., Baumer, E., DiSalvo, C., Sengers, P.: Sustainably
unpersuaded: how persuasion narrows our vision of sustainability. In: Proceedings of the 2012
ACM annual conference on Human Factors in Computing Systems - CHI 2012, p. 947. ACM
Press, Austin (2012). https://doi.org/10.1145/2207676.2208539
41. Dourish, P.: HCI and environmental sustainability: the politics of design and the design of
politics. In: Proceedings of the 8th ACM Conference on Designing Interactive Systems - DIS
2010, p. 1. ACM Press, Aarhus (2010). https://doi.org/10.1145/1858171.1858173
42. Forlizzi, J.: Moving beyond user-centered design. Interactions 25, 22–23 (2018). https://doi.
org/10.1145/3239558
43. Fritsch, J., Loi, D., Light, A.: Designing at the end of the world. In: Companion Publication
of the 2019 on Designing Interactive Systems Conference 2019 Companion, pp. 369–372.
ACM, San Diego (2019). https://doi.org/10.1145/3301019.3319999
44. Hazas, M., Brush, A.J.B., Scott, J.: Sustainability does not begin with the individual.
Interactions 19, 14–17 (2012). https://doi.org/10.1145/2334184.2334189
45. Knowles, B., Bates, O., Håkansson, M.: This changes sustainable HCI. In: Proceedings of the
2018 CHI Conference on Human Factors in Computing Systems, pp. 1–12. ACM, Montreal
(2018). https://doi.org/10.1145/3173574.3174045
46. Boehnert, J.: Anthropocene economics and design: heterodox economics for design transi-
tions. She Ji: J. Des. Econ. Innov. 4, 355–374 (2018). https://doi.org/10.1016/j.sheji.2018.
10.002
47. Ceschin, F., Gaziulusoy, I.: Evolution of design for sustainability: From product design to
design for system innovations and transitions. Des. Stud. 47, 118–163 (2016). https://doi.org/
10.1016/j.destud.2016.09.002
48. Kapadia, M., et al.: JUNGLE: an interactive visual platform for collaborative creation and
consumption of nonlinear transmedia stories. In: Cardona-Rivera, R.E., Sullivan, A., Young,
R.M. (eds.) ICIDS 2019. LNCS, vol. 11869, pp. 250–266. Springer, Cham (2019). https://
doi.org/10.1007/978-3-030-33894-7_26
49. Kim, J., Lee, E., Thomas, T., Dombrowski, C.: Storytelling in new media: the case of alternate
reality games, 2001–2009. First Monday (2009).https://doi.org/10.5210/fm.v14i6.2484
50. Smith, N., Bardzell, S., Bardzell, J.: Designing for cohabitation: naturecultures, hybrids, and
decentering the human in design. In: Proceedings of the 2017 CHI Conference on Human
Factors in Computing Systems, pp. 1714–1725. ACM, Denver (2017). https://doi.org/10.
1145/3025453.3025948
51. Schalk, S.: Vox populi. In: Rouse, R., Koenitz, H., Haahr, M. (eds.) ICIDS 2018. LNCS, vol.
11318, pp. 408–411. Springer, Cham (2018). https://doi.org/10.1007/978-3-030-04028-4_48
348 M. Ferreira et al.
52. Reitsma, L., Wessman, S., Nyström, S.: Spilltime: designing for the relationship between QS,
CO2e and climate goals. Des. J. 22, 1087–1100 (2019). https://doi.org/10.1080/14606925.
2019.1594976
53. Roth, C.: The ‘Angstfabriek’ experience: factoring fear into transformative interactive narra-
tive design. In: Cardona-Rivera, R.E., Sullivan, A., Young, R.M. (eds.) ICIDS 2019. LNCS,
vol. 11869, pp. 101–114. Springer, Cham (2019). https://doi.org/10.1007/978-3-030-33894-
7_11
54. Sauvé, K., Bakker, S., Houben, S.: Econundrum: visualizing the climate impact of dietary
choice through a shared data sculpture. In: Proceedings of the 2020 ACM Designing Interac-
tive Systems Conference, pp. 1287–1300. ACM, Eindhoven (2020). https://doi.org/10.1145/
3357236.3395509
55. Bertolotti, M., Catellani, P.: Effects of message framing in policy communication on climate
change. Eur. J. Soc. Psychol. (2014). https://doi.org/10.1002/ejsp.2033
56. Tonkinwise, C.: Only a god can save us – or at least a good story: i love sustainability (because
necessity no longer has agency). Des. Philos. Pap. 9, 69–80 (2011). https://doi.org/10.2752/
144871311X13968752924554
57. Project Drawdown: The Drawdown Review (2020)
58. Kobayashi, H.H.: Human–computer–biosphere interaction: beyond human - centric inter-
action. In: Streitz, N., Markopoulos, P. (eds.) DAPI 2014. LNCS, vol. 8530, pp. 349–358.
Springer, Cham (2014). https://doi.org/10.1007/978-3-319-07788-8_33
59. Light, A., Powell, A., Shklovski, I.: Design for existential crisis in the anthropocene age. In:
Proceedings of the 8th International Conference on Communities and Technologies, pp. 270–
79 (2017). https://doi.org/10.1145/3083671.3083688
60. Mancini, C., Lehtonen, J.: The emerging nature of participation in multispecies interaction
design. In: Proceedings of the 2018 on Designing Interactive Systems Conference 2018 -
DIS 2018, pp. 907–918. ACM Press, Hong Kong (2018). https://doi.org/10.1145/3196709.
3196785
61. Roudavski, S.: Multispecies cohabitation and future design. Presented at the Design Research
Society Conference, 10 September 2020 (2020). https://doi.org/10.21606/drs.2020.402
62. Forlano, L.: Decentering the human in the design of collaborative cities. Des. Issues 42–54
(2016). https://doi.org/10.1162/DESI_a_00398
63. Dionisio, M., Nisi, V.: Leveraging transmedia storytelling to engage tourists in the under-
standing of the destination’s local heritage. Multimed Tools Appl. (2021).https://doi.org/10.
1007/s11042-021-10949-2
64. Mendes, J., Allison, L.: Bear 71. https://bear71vr.nfb.ca/. Accessed 17 June 2021
65. Bala, P., Dionisio, M., Oliveira, S., Andrade, T., Nisi, V.: Tell a Tail: leveraging XR for a
transmedia on animal welfare. In: Nunes, N.J., Ma, L., Wang, M., Correia, N., Pan, Z. (eds.)
ICEC 2020. LNCS, vol. 12523, pp. 223–239. Springer, Cham (2020). https://doi.org/10.1007/
978-3-030-65736-9_19
66. Dionisio, M., Bala, P., Oliveira, S., Nisi, V.: Tale of T(r)ails: the design of an ar comic book
for an animal welfare transmedia. In: Bosser, A.-G., Millard, D.E., Hargood, C. (eds.) ICIDS
2020. LNCS, vol. 12497, pp. 281–284. Springer, Cham (2020). https://doi.org/10.1007/978-
3-030-62516-0_25
User Evaluation of a Storytelling
Application Assisting Visitors
in Protected Nature Areas
1 Introduction
The use of information technology, in specific digital media, has redefined our
experience of tourism services [10,14]. Different technological tools are now avail-
able to encourage effective, self-motivated methods of engagement. One such
example is technology-driven storytelling: a useful means of exploration, espe-
cially for those looking to expand upon their cultural, historical, and geographical
awareness through meaning-making and self-learning [12,26]. Advancements in
information & media technology have contributed to our meaning- and sense-
making abilities by introducing new delivery mechanisms, media forms, and
tools. The use of geo-referenced and immersive media applications within this
context has a promising future for enriching tourism/adventure experiences. Of
c Springer Nature Switzerland AG 2021
A. Mitchell and M. Vosmeer (Eds.): ICIDS 2021, LNCS 13138, pp. 349–359, 2021.
https://doi.org/10.1007/978-3-030-92300-6_34
350 A. Hameed et al.
2 Related Works
2.1 AR in Location-Based Experiences
Storytelling about places is an old, recognized tool to share personal experi-
ences about places visited. But also, to imagine fictional locations like Calvino’s
descriptive explorations [4]. Paay et al. [28] immersed users in fiction with their
non-goal oriented mobile guide for a city using LBM to weave fictional stories
around existing physical, historical, and environmental features of the city. Infor-
mal post-experience interviews revealed users appreciated how their physical
surroundings were associated to the story. LBM, or locative media, supplement
spatial technologies with context-rich and site-specific information [11]. With
computational advances we are now able to use bodily positions and spatial
movements to provide location-based experiences for creating digitally pervasive
worlds. M-View [7] is a context-aware system where media elements are trig-
gered as users navigate the physical space. Based on their position, users prompt
media instances depicting events that have occurred at the locations where they
appear. Azuma [2] classifies AR storytelling as a subset of the broader location-
based experiences that include ARGs (alternate-reality games), pervasive games,
performance art, transmedia experiences, amongst others.
3 Methods
3.1 Experiment Design
3.2 Setup
Location. The experiment was designed along a trail that led from a car-park
to an erected tipi on the edge of a lake. The trail is approximately 700-meter-long
It was marked by six story points. The test required each participant to follow
the trail from start to finish, stopping at the designated The area surrounding
the trail is a wetland with a variety of vegetation.
Story. This storytelling experience follows the fantastical world of trolls residing
in Trollheimen who are responsible for looking after mother nature. It is their
job to modulate the seasons and manage the surroundings. They also inform
and educate visiting humans on how to behave in Trollheimen.
3.3 Materials
The Game Experience Questionnaire (GEQ) [17] and AttrakDiff [15] were used.
The GEQ [17] is a popular tool used frequently for assessments of gaming experi-
ence [6,27]. It satisfied the needs and requirements of our study and sample size.
User Evaluation of a Storytelling Application Assisting Visitors 353
GEQ has a 3-tier modular structure. This paper looks at 7 core aspects of the
experience and two post-game aspects. Namely, Competence, Immersion, Flow,
Tension, Challenge, Negative Affect, Positive Affect, Tiredness, and Return-to-
Reality. We also used the Short AttrakDiff questionnaire [15], which divides user
experience into pragmatic and hedonic qualities. It provides scores for pragmatic
quality (PQ), hedonic quality from an identity (HQ-I) and a stimulation (HQ-S)
point of view, as well as overall attractiveness (ATT). Users rank word pairs
representing the worst and best on a range from −3 to +3.
4 Results
We find the means and standard deviations for the nine dependent variables
in Table 1 for both conditions. A multivariate analysis of variance (MANOVA)
was conducted to determine significant differences between the two categorical
conditions when considered for the dependent variables, i.e. dimensions of the
GEQ. Our investigation of the two independent variables, TB & AR, achieved
a statistically significant result: F (9, 20) = 3.382, p = 0.011, Wilk’s λ = 0.397,
η 2 p = 0.603. Further univariate ANOVAs for the nine dimensions of the GEQ
across the two conditions, TB & AR, show the following between-subject effects:
Table 1. The mean and standard deviation for the nine dimensions of the GEQ cor-
responding to both conditions.
GEQ aspects
Item Condition Mean (µ) Std. dev N
Competence TB 3.57 0.54 15
AR 3.51 0.75 15
Immersion TB 3.31 0.36 15
AR 3.96 0.68 15
Flow TB 2.60 0.69 15
AR 3.52 0.70 15
Tension TB 1.22 0.35 15
AR 1.31 0.50 15
Challenge TB 1.64 0.49 15
AR 1.83 0.47 15
Negative Aspects TB 1.40 0.43 15
AR 1.33 0.55 15
Positives Aspects TB 3.90 0.50 15
AR 4.15 0.67 15
Tiredness TB 1.63 0.92 15
AR 1.43 0.59 15
Return to Reality TB 1.62 0.62 15
AR 2.02 0.48 15
The addition of AR-based features had a significant effect only on two dimen-
sions of the GEQ. Namely, immersion with a p-value of 0.003, and flow with
a p-value of 0.001. Immersion scores improved from condition-TB (µ = 3.31,
SD = 0.36) to condition-AR (µ = 3.96, SD = 0.68). While that for flow also
increased from condition-TB (µ = 2.60, SD = 0.69) to condition-AR (µ = 3.52,
SD = 0.70). No significant differences were observed in the remaining seven
variables (Fig. 1). An adjusted alpha-value (p <0.025) was used to account for
multiple ANOVAs. Min & Max scores can be seen in Fig. 2. The score deviation
is markedly noticeable towards the maximum. Participants reported a higher
immersion and flow in story experience when delivered using AR-based media.
User Evaluation of a Storytelling Application Assisting Visitors 355
Fig. 1. A comparison of combined means for the nine GEQ dimensions under both
conditions TB & AR
Fig. 2. Min & Max for immersion and flow under TB & AR
Fig. 3. Portfolio chart & average values chart for TB & AR (Color figure online)
356 A. Hameed et al.
5 Discussion
In this paper, we investigated a location-based storytelling application. Our
application used an off-the-shelf commercial tablet in a user study with 30
User Evaluation of a Storytelling Application Assisting Visitors 357
6 Conclusion
A notable outcome from this work has been the implementation of user evalua-
tion in a field experiment. Most user studies are conducted under lab conditions,
which risks the production of unnatural behavior due to the artificial setting, i.e.
low ecological validity. With a field experiment it becomes possible to generalize
findings to a real-life setting. Such evaluations give us a better understanding of
our products. In this case, the application was used by everyday users in a real-
world setting of a natural environment. The results confirm that AR features
are a successful replacement for physical signposts. Future works include mak-
ing improvements to the application. Character development and animation is
currently underway. Additionally increased interactivity features are being intro-
duced to the application, e.g. conversations with the virtual trolls. The inclusion
of video and image formats are also implemented. Further user testing will follow
before the commercial release.
References
1. Aronson, E., Wilson, T.D., Akert, R.M.: Social Psychology, 7th edn. Upper Saddle
River (2010)
2. Azuma, R.: 11 location-based mixed and augmented reality storytelling (2015)
3. Brockmyer, J.H., Fox, C.M., Curtiss, K.A., McBroom, E., Burkhart, K.M.,
Pidruzny, J.N.: The development of the game engagement questionnaire: a measure
of engagement in video game-playing. J. Exp. Soc. Psychol. 45(4), 624–634 (2009)
4. Calvino, I.: Italo Calvino on “invisible cities”. Columbia: J. Lit. Art 37–42 (1983)
5. Carrigy, T., Naliuka, K., Paterson, N., Haahr, M.: Design and evaluation of player
experience of a location-based mobile game. In: Proceedings of the 6th Nordic
Conference on Human-Computer Interaction: Extending Boundaries, pp. 92–101
(2010)
358 A. Hameed et al.
6. Chertoff, D.B., Goldiez, B., LaViola, J.J.: Virtual experience test: a virtual envi-
ronment evaluation questionnaire. In: 2010 IEEE Virtual Reality Conference (VR),
pp. 103–110. IEEE (2010)
7. Crow, D., Pan, P., Kam, L., Davenport, G.: M-views: a system for location-based
storytelling. In: ACM UbiComp (2003)
8. Denisova, A., Nordin, A.I., Cairns, P.: The convergence of player experience ques-
tionnaires. In: Proceedings of the 2016 Annual Symposium on Computer-Human
Interaction in Play, pp. 33–37 (2016)
9. Dionisio, M., et al.: Evaluation of Yasmine’s adventures: exploring the socio-
cultural potential of location aware multimedia stories. In: Schoenau-Fog, H.,
Bruni, L.E., Louchart, S., Baceviciute, S. (eds.) ICIDS 2015. LNCS, vol. 9445, pp.
251–258. Springer, Cham (2015). https://doi.org/10.1007/978-3-319-27036-4 24
10. Dionisio, M., Nisi, V., Correia, N.: Leveraging on transmedia entertainment-
education to offer tourists a meaningful experience. In: Nunes, N., Oakley, I., Nisi,
V. (eds.) ICIDS 2017. LNCS, vol. 10690, pp. 359–362. Springer, Cham (2017).
https://doi.org/10.1007/978-3-319-71027-3 43
11. Farman, J., et al.: Location-based media. In: Dialogues on Mobile Communication,
pp. 161–177. Routledge (2016)
12. Haahr, M.: Reconciling immersion and presence: locative game mechanics and
narrative techniques for cultural heritage. Virtual Creativity 8(1), 23–37 (2018)
13. Hameed, A., Irshad, S., Perkis, A.: Towards a quality framework for immersive
media experiences: a holistic approach. In: Cardona-Rivera, R.E., Sullivan, A.,
Young, R.M. (eds.) ICIDS 2019. LNCS, vol. 11869, pp. 389–394. Springer, Cham
(2019). https://doi.org/10.1007/978-3-030-33894-7 41
14. Han, D.-I.D., Weber, J., Bastiaansen, M., Mitas, O., Lub, X.: Virtual and aug-
mented reality technologies to enhance the visitor experience in cultural tourism.
In: tom Dieck, M.C., Jung, T. (eds.) Augmented Reality and Virtual Reality. PI,
pp. 113–128. Springer, Cham (2019). https://doi.org/10.1007/978-3-030-06246-0 9
15. Hassenzahl, M., Burmester, M., Koller, F.: AttrakDiff: a questionnaire to mea-
sure perceived hedonic and pragmatic quality. In: Szwillus, G., Ziegler, J. (eds.)
Human Computer 2003, pp. 187–196. Springer, Heidelberg (2003). https://doi.org/
10.1007/978-3-322-80058-9 19
16. Högberg, J., Hamari, J., Wästlund, E.: Gameful experience questionnaire (GAME-
FULQUEST): an instrument for measuring the perceived gamefulness of system
use. User Model. User-Adap. Inter. 29(3), 619–660 (2019)
17. IJsselsteijn, W.A., de Kort, Y.A., Poels, K.: The game experience questionnaire.
Eindhoven: Technische Universiteit Eindhoven 46(1) (2013)
18. Jackson, S.A., Eklund, R.C.: Assessing flow in physical activity: the flow state
scale-2 and dispositional flow scale-2. J. Sport Exerc. Psychol. 24(2), 133–150
(2002)
19. Jackson, S.A., Martin, A.J., Eklund, R.C.: Long and short measures of flow: the
construct validity of the FSS-2, DFS-2, and new brief counterparts. J. Sport Exerc.
Psychol. 30(5), 561–587 (2008)
20. Jennett, C., et al.: Measuring and defining the experience of immersion in games.
Int. J. Hum. Comput. Stud. 66(9), 641–661 (2008)
21. Klungre, Ø.S., Hameed, A., Perkis, A.: Narrative’s impact on quality of experience
in digital storytelling. In: SA 2019: SIGGRAPH ASIA Art Gallery/Art Papers
(2019)
22. Lessiter, J., Freeman, J., Keogh, E., Davidoff, J.: A cross-media presence question-
naire: the ITC-sense of presence inventory. Presence: Teleoper. Virtual Environ.
10(3), 282–297 (2001)
User Evaluation of a Storytelling Application Assisting Visitors 359
23. List, J.A.: Field experiments: a bridge between lab and naturally occurring data.
BE J. Econ. Anal. Policy 5(2) (2007)
24. Lombard, M., Ditton, T.B., Weinstein, L.: Measuring presence: the temple presence
inventory. In: Proceedings of the 12th Annual International Workshop on Presence,
pp. 1–15 (2009)
25. Milgram, P., Takemura, H., Utsumi, A., Kishino, F.: Augmented reality: a class
of displays on the reality-virtuality continuum. In: Telemanipulator and Telepres-
ence Technologies, vol. 2351, pp. 282–292. International Society for Optics and
Photonics (1995)
26. Moscardo, G.: Stories and design in tourism. Ann. Tourism Res. 83, 102950 (2020)
27. Norman, K.L.: GEQ (game engagement/experience questionnaire): a review of two
papers. Interact. Comput. 25(4), 278–283 (2013)
28. Paay, J., et al.: Location-based storytelling in the urban environment. In: Pro-
ceedings of the 20th Australasian Conference on Computer-Human Interaction:
Designing for Habitus and Habitat, pp. 122–129 (2008)
29. Perkis, A., et al.: QUALINET white paper on definitions of immersive media expe-
rience (IMEx). arXiv preprint arXiv:2007.07032 (2020)
30. Ryan, R.M., Rigby, C.S., Przybylski, A.: The motivational pull of video games: a
self-determination theory approach. Motiv. Emot. 30(4), 344–360 (2006)
31. Schmidt, S., Zadtootaghaj, S., Möller, S., Metzger, F., Hirth, M., Suznjevic, M.:
Subjective evaluation methods for gaming quality (p. game). CH-Geneva (2018)
32. Schubert, T.W.: The sense of presence in virtual environments: a three-component
scale measuring spatial presence, involvement, and realness. Z. für Medienpsycholo-
gie 15(2), 69–71 (2003)
33. Witmer, B.G., Singer, M.J.: Measuring presence in virtual environments: a presence
questionnaire. Presence 7(3), 225–240 (1998)
The Magic of the In-Between: Mental
Resilience Through Interactive Narrative
1 Introduction
Many interactive narratives exist that are related to mental health. Depression
Quest [13] and Paper Cages [8], for instance, are examples of simple games about
dealing with depression and anxiety. The purpose of these games is, for the most
part, awareness and simulation of these conditions. As such, they do not actually
provide substantial resources for users in need of mental health support.
Adventures with Anxiety is a more complex interactive story that ultimately
encourages the user to work in tandem with their anxiety [5]. A website contain-
ing links to mental health resources is presented at the conclusion of the game.
Prior work [14] has suggested that Adventures with Anxiety is an example of
an interactive narrative with transformative potential to “open our hearts and
change who we are” [11].
During gameplay, Adventures with Anxiety breaks the fourth wall, stating
that “building a healthy relationship with your emotions isn’t as simple as click-
ing buttons on a screen”. While it provides more support than a simulation
game like Paper Cages, it recognizes that real life for its user will be harder
than playing the game itself. Adventures with Anxiety thus points to the next
level of interactive narrative for mental health support: one that could help users
reliably build mental resilience over time.
c Springer Nature Switzerland AG 2021
A. Mitchell and M. Vosmeer (Eds.): ICIDS 2021, LNCS 13138, pp. 360–364, 2021.
https://doi.org/10.1007/978-3-030-92300-6_35
Mental Resilience Through Interactive Narrative 361
of negative affect) when done from a self-distanced perspective [2,3]. In this case,
self-distancing means that the person in question is able to view themselves from
the perspective of an observer during their analysis.
Betwixt provides gradual transitions for users to slowly adjust into and out
of the dreams. Headphones are encouraged as audio and calmly-paced text are
used to enhance immersion with the fantasy world, helping the user to relax
and focus. The overall experience was designed to maximize the potential for
users to self-reflect from an observer’s distance (via dreams), while also actively
involving them in a collaborative telling of their own story. Betwixt maintains
an immersive atmosphere until a resolution point is reached, at which point
the user likely still has questions about the story to keep them interested in
returning. Users gain further agency as they continue, which is intended to have
an empowering effect each time they close the app and return to reality.
Here, we consider three findings relevant to interactive narratives for mental
resilience: (1) the power of immersive yet self-distanced interactive storytelling
for focus and self-reflection, (2) the value of collaborative conversation, and (3)
support for healthy long-term engagement. Preliminary results were gathered
via a survey of 40 Betwixt users. Users were asked to volunteer feedback about
the overall experience to support the further development of the game but were
not required, incentivized, or otherwise compelled to do so (2% response rate).
3 Results
3.1 Immersive Self-distanced Storytelling for Self-reflection
4 Discussion
We have described several of Betwixt’s features that appear to differentiate it
from other interactive narratives for mental health support. In its current state,
Betwixt appeals best to individuals with roughly half an hour available searching
for a relaxing and self-reflective experience. Preliminary feedback suggests it
appears to fulfill these needs, and does so in such a way that users can regularly
learn from the app.
Our results point to features that enhance the power of interactive story-
telling for mental resilience. Relaxing immersive elements and opportunities for
personal, self-distanced reflection that relate to the setting appear to support
healthy, consistent engagement in the self-reflection process rather than “click-
ing through” the experience. Collaborative brainstorming and reframing with
an app that can “listen” and support self-expression and self-connection, rather
than preach or dismiss, was also deemed valuable by users. Future work will con-
tinue to build on our understanding of how to best design interactive narratives
like Betwixt to assist the general population in strengthening mental resilience.
364 S. Harmon et al.
References
1. de Alva, F.E.M., Wadley, G., Lederman, R.: It feels different from real life: users’
opinions of mobile applications for mental health. In: Proceedings of the Annual
Meeting of the Australian Special Interest Group for Computer Human Interaction,
pp. 598–602 (2015)
2. Ayduk, Ö., Kross, E.: Analyzing negative experiences without ruminating: the role
of self-distancing in enabling adaptive self-reflection. Soc. Pers. Psychol. Compass
4(10), 841–854 (2010)
3. Ayduk, Ö., Kross, E.: From a distance: implications of spontaneous self-distancing
for adaptive self-reflection. J. Pers. Soc. Psychol. 98(5), 809 (2010)
4. Baumel, A., Muench, F., Edan, S., Kane, J.M.: Objective user engagement with
mental health apps: systematic search and panel-based usage analysis. J. Med.
Internet Res. 21(9), e14567 (2019)
5. Case, N.: Adventures with anxiety (2019)
6. Cowden, R.G., Meyer-Weitz, A.: Self-reflection and self-insight predict resilience and
stress in competitive tennis. Soc. Behav. Personal. Int. J. 44(7), 1133–1149 (2016)
7. Darcy, A., Daniels, J., Salinger, D., Wicks, P., Robinson, A., et al.: Evidence of
human-level bonds established with a digital conversational agent: cross-sectional,
retrospective observational study. JMIR Formative Res. 5(5), e27868 (2021)
8. Gameworks, P.: Paper cages (2019)
9. Greenberg, L.S.: Emotion-focused therapy. Clin. Psychol. Psychother.: Int. J. The-
ory Pract. 11(1), 3–16 (2004)
10. Litvin, S., Saunders, R., Maier, M.A., Lüttke, S.: Gamification as an approach to
improve resilience and reduce attrition in mobile mental health interventions: a
randomized controlled trial. PloS One 15(9), e0237220 (2020)
11. Murray, J.H.: Hamlet on the Holodeck: The Future of Narrative in Cyberspace.
MIT Press (2017)
12. Nakajima, M., Takano, K., Tanno, Y.: Mindfulness relates to decreased depressive
symptoms via enhancement of self-insight. Mindfulness 10(5), 894–902 (2019)
13. Quinn, Z., Lindsey, P., Schankler, I.: Depression quest: an interactive (non)
fiction about living with depression. DepressionQuest.com (2013). http://www.
depressionquest.com
14. Roth, C., Mekler, E.D., Bowman, N.: Transformative experience through inter-
active narrative. In: ICIDS 2020 International Conference on Interactive Digital
Storytelling (2020)
15. Schneider, T.R., Lyons, J.B., Khazon, S.: Emotional intelligence and resilience.
Pers. Individ. Differ. 55(8), 909–914 (2013)
16. Seligman, M.E.P., Csikszentmihalyi, M.: Positive psychology: an introduction. In:
Csikszentmihalyi, M. (ed.) Flow and the Foundations of Positive Psychology, pp.
279–298. Springer, Dordrecht (2014). https://doi.org/10.1007/978-94-017-9088-
8 18
17. Stein, D., Grant, A.M.: Disentangling the relationships among self-reflection,
insight, and subjective well-being: the role of dysfunctional attitudes and core
self-evaluations. J. Psychol. 148(5), 505–522 (2014)
18. Tabibnia, G.: An affective neuroscience model of boosting resilience in adults.
Neurosci. Biobehav. Rev. 115, 321–350 (2020)
19. Troy, A.S., Mauss, I.B.: Resilience in the face of stress: emotion regulation as a
protective factor. Resilience Mental Health: Challenges Across Lifespan 1(2), 30–44
(2011)
20. Tugade, M.M., Fredrickson, B.L.: Regulation of positive emotions: emotion regu-
lation strategies that promote resilience. J. Happiness Stud. 8(3), 311–333 (2007)
A Transmedia Narrative Framework
for Pediatric Hearing Counseling
1 Introduction
In adult and pediatric health care, communication and person-centredness serve
as the basic elements for health literacy, but also as the main indicators of the
quality of care received by children and their families [3,38]. However, rather than
focusing on promoting more effective child and care provider communication, and
addressing the child’s individual needs for treatment, new pediatric digital tools
have been much more concerned with diagnostic or physical rehabilitative uses.
To help fill this gap in current applications and to support child-centered care in
pediatric health care, we designed and tested a novel narrative transmedia tool
for enhancing child and care provider communication for pediatric audiology and
speech- and language therapy appointments.
3 Design
The tool was designed for initial and follow-up pediatric hearing counseling
appointments and integrates methods from narrative and play therapy, as well
as elements from some of the few existing pediatric hearing counseling tools
[19,23] developed by the Ida Institute. The tool aims to introduce a narrative
communication platform for facilitating more effective communication during
368 N. Kadastik and L. E. Bruni
Fig. 1. Introduction on the tablet computer. On the left, screen capture of the start
screen. On the right, character introduction.
The three environments were chosen based on their pertinence and relevance,
as advised by our participating specialists. After the introductory part, the user is
asked to select one specific communication environment (the setting) they would
like to explore further. The guiding thread is to help the character (Sara), who
suffers from hearing loss to identify potential issues and challenges in the chosen
communication environment, to subsequently encourage exploring and defining
specific communication strategies for that environment. In this initial stage,
through the interactive tasks and questions, the system also obtains relevant
information about the user to personalize the experience, support the selecting
of the most relevant communication environment, and to provide post-session
documentation for the specialist.
Subsequently, a 360-degree video scenario introduces the setting and the
point of view. The user can then experience their previously selected environ-
ment through ‘Sara’s eyes’, using a mobile-VR head-mounted display (HMD)
(See Fig. 3). All three possible scenarios are produced as scripted 2–3 min long
one-shot 360-degree videos. To elicit context-specific emergent narratives, each
360-degree video concurrently conveys a number of typical situations that a
370 N. Kadastik and L. E. Bruni
Fig. 2. Screen capture of interactive tasks and questions in the app on the tablet
computer. On the left, prioritization of the three different communication environments.
On the right, evaluation of the prioritized communication environments. (Given the
original context of the implementation and evaluation, the texts are in Danish language)
child with hearing loss (HL) might encounter on a day-to-day basis, correspond-
ing to the specific communication environment. When designing these three
scenarios, it was important to take into consideration the desired margin for
user-interpretation, as well as the opportunities for achieving this using the 360-
degree video narrative medium. It was hypothesized that allowing more freedom
in the interpretation of the scenarios, instead of introducing explicit narratives,
could better support the therapeutic goals of the system. Although the user was
expected to freely explore the scenario, it was nevertheless important to support
keeping the focus on the specific context of the chosen communication scenario.
This was done through directing the user’s attention in the scenario through con-
current scripted situations. These scenario-specific situations were established
based on an analysis of relevant materials, information gathered from the con-
ducted semi-structured interviews, and observational data from Ida Institute’s
ethnographic videos. To encourage the user in adopting the character’s perspec-
tive during the experience, all 360-degree video scenarios include a first-person
avatar perspective of the character’s virtual body.
Returning to the tablet app, the user is encouraged to reflect on their VR-
experience and use the introduced narrative elements in a play activity on the
tablet computer (See Fig. 4). The play activity aims to elicit and support the
sharing of emergent narratives to meet the embedded counseling goals. The
design for the play activity presents the user with a modifiable ‘canvas’, where
pre-arranged elements that loosely depict some of the situations, characters, and
relevant objects from the selected 360-scenario, but also present the user with
additional context-specific items e.g. hearing aids, an FM-system (the wireless
device that helps enhance hearing aids), and other items. The user can interact
with the elements by dragging them around on the canvas, adding new items from
the menu, or removing them. Additionally, the system also includes a drawing
function, which allows the user to draw missing objects, characters or other
desired elements to support their narratives. Lastly, the system also features a
A Transmedia Narrative Framework for Pediatric Hearing Counseling 371
Fig. 3. Screen capture of one the 360-degree video scenarios, the school environment.
‘tips’ section, providing the user(s) with questions, suggestions and discussion
topics for conversation scaffolding during the activity. After the activity the
user(s) can save their creation and are awarded with a “thank you” animation.
The counseling stages for the specialist include: (1) the exploratory stage (2)
the immersive stage (3) the reflection, strategy, and decision-making stage. In
order to support structuring the session in accordance with the embedded goals,
the tool comes with an additional tool guide. The tool guide acts as a support-
ing guide for the professional to navigate the system and provides suggestions
for conversation scaffolding during each stage. The suggestions for conversation
scaffolding [2] draw inspiration from the Ida Institute tools [19,23], and narrative
scaffolding conversation maps for supporting externalizing, deconstruction, and
unique outcomes, as adapted by [34,35], originally based on Vygotsky’s seminal
work [31,32].
The flow of the three stages go as follows: (1) In the exploratory stage, the
specialist and the child can explore the impacts and potential challenges in the
three communication environments presented, through the initially introduced
narrative elements (context, characters, setting), and the interactive tasks and
questions. By the end of the exploratory stage, the child’s responses to the ques-
tions will be displayed, and the specialist can further examine them: identify
discrepancies, potential ambivalence and attitudes relating to the different com-
372 N. Kadastik and L. E. Bruni
Fig. 4. Play activity on the tablet computer. On the left, screen capture of the play
activity, school environment. On the right, specialist, child and parent engaging in the
play activity, playground.
munication environments. Thereafter, the specialist and the child can select a
communication environment they would like to discuss and explore further.
Fig. 5. On the left, participant exploring one of the VR scenarios, specialist and parent
observing what the child sees on the tablet. On the right, form for the specialist to fill
in comments and observations from the session.
(2) In the second, immersive stage, the specialist can make use of the VR
feature. While the child is experiencing the 360-degree video in VR, the spe-
cialist and the parent(s) can see on the tablet what the child sees in VR (See
Fig. 5). The specialist can use the suggestions in the tool guide to guide the child
in the environment, or let the child experience it more on their own. In combi-
nation with the previous stage, the immersive stage serves as the foundation for
the subsequent stage, helping to establish a common narrative communication
platform.
(3) During the reflection, strategy and decision-making stage, the objective
is to involve the child in the collaborative problem-solving and strategy-building
process through play and emergent storytelling (becoming a ‘co-expert’). Later
A Transmedia Narrative Framework for Pediatric Hearing Counseling 373
in the process, the child’s creation from the activity can be saved and shared
with the child and the parent(s), providing a type of a hand-out, or take-home
resource. The system also allows saving the child’s responses, and provides the
specialist a post-session form for filling in comments and observations from the
session (see Fig. 5). This documentation and the hand-out can be used during
follow-up sessions, for instance, as a conversation starter for discussing progress
and strategies.
4 Evaluation
to synthesize evidence from these multiple cases [22]. As the last step in the
analysis we involved a pediatric hearing care specialist as part of our participatory
analysis approach.
The equipment used during all testing sessions included two Samsung Galaxy
tablets, one Samsung S8 smartphone, a Samsung Gear VR head-mounted-display
(HMD), and an external Wonderboom Bluetooth speaker. One of the Samsung
tablets was used for the tablet-app and the second one was used to mirror the
Samsung S8 smartphone’s VR image during the VR experience, so the specialist
and the parent(s) could follow what the child was seeing in VR. In case the
child did not want to pair their hearing device(s) to the smartphone, an external
Bluetooth speaker was used to amplify the sound during the VR experience.
4.1 Results
In all three cases, the tool supported uncovering essential information about
the functional impacts of the children’s hearing loss and served as an effective
springboard for facilitating communication during the sessions. Introducing dig-
ital technology as part of the appointment, particularly the VR headset, helped
engage the children and the parents from the start. Moreover, as all partici-
pants had previous experience with using tablet computers at home, adopting
the tablet app felt intuitive for all the children.
During the first stage, both specialists used the character and the subsequent
interactive tasks and questions to elicit conversation on important themes and
issues relating to the three presented communication environments. However,
it was especially with the younger children, Participant B and Participant C,
that important information emerged in this phase. For example, when discussing
Participant B’s kindergarten environment, it was brought up by Participant B
that he does not hear very well in the play-hut when other children are speaking.
Participant B’s mother then supported the specialist in further exploring this
issue by asking, “Do you hear some of what they [the other children] are saying?”
to which Participant B replied, “I do not hear anything at all”. While with
Participant C, it emerged that he was highly sensitive to the noise of slamming
doors and the noise of rain falling when at the kindergarten.
The narrative framework embedded in the tool, partially relying on the 360-
video VR scenarios, was observed as most effective with the older participant
(Participant A, 9 y/o). In this case, it supported the child in sharing his per-
spective and helped the specialist in introducing new strategies. During the play
activity, reflecting on the VR scenario to help the character cope better in the
selected environment, Participant A selected the same type of hearing equipment
and the same type of technology for the character as he himself had. When Par-
ticipant A was asked why he gave the character an iPad, he replied that, it was
so the character Sara would not have to focus on what was going on around her,
and could get away from the surrounding noise. It was later suggested by the spe-
cialist that the child likely projected his own coping-strategies and perspectives
on the character. This gave the specialist an opportunity to better understand
the child’s perspective and identify previously unknown issues. The specialist
A Transmedia Narrative Framework for Pediatric Hearing Counseling 375
also used the narrative platform to introduce new strategies in a way that did
not directly involve the child, “Sara is 9 years old. And she just got the phone.
Does she know how to hold it so that she can hear better?” The fact that Partici-
pant A felt it was more of a game where he could also be the expert, encouraged
him to participate in the strategy-building and decision-making process more
actively, while allowing the specialist to steer the conversation naturally to the
child’s own habits. Specialist 1 assessed the role of the VR feature in establishing
the narrative communication platform with Participant A as an essential part of
the tool, “It was something he was captivated by. When he gets captivated by it,
we can then also talk about it. Had he [Participant A] thought it was too boring
or had not liked it, we also would not have been able to talk about it later”. In
this case, not only did the VR feature help make the session more exciting and
engaging for the Participant A, but it also had direct implications for helping
facilitate the conversation during the later phases.
On the other hand, in the case of Participant B and Participant C, the narra-
tive framework did not work as well as intended. It was clear that both younger
participants found it difficult to apprehend the VR experience, subsequently
making it more challenging in terms of bridging the experience with the latter
play activity on the tablet. Although both enjoyed exploring the functionalities
of the system, they also needed much more support from the specialist in terms of
contextualizing the play activity based on the previously presented information.
Furthermore, despite the suggestions provided in the tool guide, and both spe-
cialists seeing the VR feature as a valuable addition to the tool, it was also clear
that the specialists were new to this type of technology, and found it challenging
to put to use in terms of serving the goals of the session. Therefore, to encour-
age further exploration, and for the VR experience not to become underutilized
during the session, this feature would require either an instructional debriefing,
or some other type of an embedded protocol for the specialist.
Nevertheless, the findings suggest that the tool offered a structured and time-
framed approach for the specialists to explore the various impacts of the chil-
dren’s hearing loss. And while the tool was demonstrated as most effective for
exploring the functional impacts, it was evident that it also provided ways for
scaffolding psycho-social impacts. Both participating specialists were interested
in adopting the tool in their practice.
5 Conclusion
The present study was designed to explore the possibilities and potential of
a narrative transmedia counseling tool that integrates smartphone-based VR
for encouraging a more child-centered approach during pediatric hearing care
appointments. The communication tool was based on a transmedia-based emer-
gent narrative system, and was directly tied to the counseling goals, integrat-
ing methods from narrative and play therapy. The tool was evaluated through
three case studies in which two speech- and language specialists tested the tool
with three children diagnosed with hearing loss. The findings suggest that the
376 N. Kadastik and L. E. Bruni
tool was most effective in supporting the specialists in exploring the functional
impacts of the children’s hearing loss, while allowing for scaffolding psycho-
social impacts. In all cases, by establishing a communication platform based on
context-specific themes and narrative elements, the tool provided a way for the
children to actively engage in conversation about their individual perspectives,
promoting the children’s self-advocacy skills during the session.
While the novelty of the VR feature managed to equally engage all the partici-
pants, the younger participants found the experience more difficult to apprehend.
Hence, the narrative communication platform integrating the VR feature, was
more effective and beneficial when used with the older child. In this case, the nar-
rative communication platform provided a way to better understand the child’s
perspective but also offered an additional way for the specialist to introduce and
collaboratively define new strategies without putting unnecessary pressure on
the child. Nevertheless, for the VR feature not to become underutilized during
the session, it would require an instructional debriefing, or some other type of
an embedded protocol for guiding the child in the VR environment.
In this study, the tool was tested by two speech and language specialists,
who have different roles and approaches in the pediatric hearing care process,
therefore, it could also be interesting to establish the potential of the tool in a
more clinical context, for instance, when used as part of an audiological assess-
ment. Moreover, future work should focus on providing more opportunities for
personalization i.e. adding more characters, the children being able to themselves
choose the character’s gender, age, hearing equipment. While the narrative sys-
tem was considered most successful when used with an older child, the different
elements of the tool showed high potential with younger children. The fact that
the specialists participating in the study were interested in adopting the tool into
their practice, further demonstrates the potentiality of this type of a solution
as part of a pediatric hearing care session for enhancing child and care provider
communication.
References
1. Aylett, R.: Narrative in virtual environments towards emergent narrative. In: Pro-
ceedings of AAAI Symposium on Narrative Intelligence (1999)
2. Baceviciute, S., Rützou Albæk, K.R., Arsovski, A., Bruni, L.E.: Digital interactive
narrative tools for facilitating communication with children during counseling: a
case for audiology. In: Oyarzun, D., Peinado, F., Young, R.M., Elizalde, A., Mén-
dez, G. (eds.) ICIDS 2012. LNCS, vol. 7648, pp. 48–59. Springer, Heidelberg (2012).
https://doi.org/10.1007/978-3-642-34851-8_5
3. Bell, J., Condren, M.: Communication strategies for empowering and protecting
children. Pediatr. Pharm. Advocacy Group 21(2), 176–184 (2016). https://doi.org/
10.5863/1551-6776-21.2.176
4. Bennett, L.: Narrative methods and children: theoretical explanations and practice
issues. J. Child Adolesc. Psychiatr. Nurs. 21(1), 13–23 (2008). https://doi.org/10.
1111/j.1744-6171.2008.00125.x
A Transmedia Narrative Framework for Pediatric Hearing Counseling 377
5. Bräne, A.: User Experience Design for Children: Developing and Testing a UX
Framework (Dissertation) (2016). http://urn.kb.se/resolve?urn=urn:nbn:se:umu:
diva-125935. Accessed 12 Oct 2021
6. Bryant, L., Brunner, M., Hemsley, B.: A review of virtual reality technologies
in the field of communication disability: implications for practice and research.
Disabil. Rehabil. Assist. Technol. 15(4), 365–372 (2020). https://doi.org/10.1080/
17483107.2018.1549276
7. Chakraborty, K., Bhide, A.: General principles for psychotherapeutic interven-
tions in children and adolescents. Indian J. Psychiatry (2020). http://www.
indianjpsychiatry.org/text.asp?2020/62/8/299/276107. Accessed 23 July 2020
8. Chow, E.O.: Narrative therapy an evaluated intervention to improve stroke sur-
vivors’ social and emotional adaptation. Clin. Rehabil. 29(4), 315–326 (2015).
https://doi.org/10.1177/0269215514544039
9. Chu, S.L., Quek, F., Tanenbaum, T.J.: Performative Authoring: nurturing story-
telling in children through imaginative enactment. In: Koenitz, H., Sezen, T.I.,
Ferri, G., Haahr, M., Sezen, D., C˛atak, G. (eds.) ICIDS 2013. LNCS, vol. 8230, pp.
144–155. Springer, Cham (2013). https://doi.org/10.1007/978-3-319-02756-2_18
10. Coyne, I., Holmström, I., Söderböck, M.: Centeredness in healthcare: a concept
synthesis of family-centered care, person-centered care and child-centered care. J.
Pediatr. Nurs. 42, 45–56 (2018). https://doi.org/10.1016/j.pedn.2018.07.001
11. Furlonger, B.: Narrative therapy and children with hearing impairments. Am. Ann.
Deaf 144, 325–333 (1999)
12. Georgieva, I., Georgiev, G.: Reconstructing personal stories in virtual reality as a
mechanism to recover the self. Int. J. Environ. Res. Public Health 17(26), 5 (2019).
https://doi.org/10.3390/ijerph17010026
13. Gonçalves, M., Matos, M., Santos, A.: Narrative therapy and the nature of “Inno-
vative Moments” in the construction of change. J. Constructivist Psychol. 22(1),
1–23 (2009). https://doi.org/10.1080/10720530802500748
14. Gottschalk, F.: Impacts of technology use on children: exploring literature on the
brain, cognition and well-being (2019). https://www.oecd.org/officialdocuments/
publicdisplaydocumentpdf/?cote=EDU/WKP%282019%293&docLanguage=En.
Accessed 29 July 2021
15. Guterman, J., Martin, C.: Using puppets with aggressive children to externalize
the problem in narrative therapy. In: Drewes, A., Schaefer, C. (eds) Play Therapy
In Middle Childhood, pp. 135–151. American Psychological Association (2015).
https://doi.org/10.1037/14776-008
16. Jenkins, H.: Convergence Culture: Where Old and New Media Collide. New York
University Press (2006)
17. Jenny, E.: Projection. In: Goldstein, S., Naglieri, J.A. (eds.) Encyclopedia of Child
Behavior and Development, pp. 1162–1163. Springer, Boston (2011)
18. Kraft, P., Yardley, L.: Current issues and new directions in Psychology and Health:
what is the future of digital interventions for health behaviour change? Psychol.
Health 24(6), 615–618 (2009). https://doi.org/10.1080/08870440903068581
19. Living Well for Teens and Tweens Digital Pediatric Counseling Tool: Ida Insti-
tute (2018). https://idainstitute.com/tools/living_well_for_teens_and_tweens/.
Accessed 12 Oct 2021
20. Louchart, S., Aylett, R.: Narrative theory and emergent interactive narrative. Int.
J. Continuing Eng. Educ. Life Long Learn. 14(6) (2004)
21. Lundtofte, T.E.: Young children’s tablet computer play. Am. J. Play
12(2), 216–232. (2020). https://www.journalofplay.org/sites/www.journalofplay.
org/files/pdf-articles/AJP-12-2-Article5-Tablet.pdf. Accessed 12 Oct 2021
378 N. Kadastik and L. E. Bruni
22. Mills, A.J., Durepos, G., Wiebe, E.: Encyclopedia of Case Study Research (vols.
1–0). SAGE Publications, Inc., Thousand Oaks (2010). https://doi.org/10.4135/
9781412957397
23. My World Digital Pediatric Counseling Tool: Ida Institute (2018). https://
idainstitute.com/tools/my_world/. Accessed 12 Oct 2021
24. O’Connor, T.: Narrative therapy. In: Leeming, D.A. (ed.) Encyclopedia of Psychol-
ogy and Religion, pp. 1182–1185. Springer, Heidelberg (2014). https://doi.org/10.
1007/978-1-4614-6086-2_455
25. Parsons, T.D., et al.: Virtual reality in pediatric psychology. Pediatrics 140(2),
86–91 (2017). https://doi.org/10.1542/peds.2016-1758I
26. Pattison, S., Harris, B.: Counselling children and young people: a review of the
evidence for its effectiveness. Couns. Psychother. Res. 6, 233–237 (2006). https://
doi.org/10.1080/14733140601022659
27. Round, J.L.: The need and uses of the Ida ‘My World’ counselling tool in paedi-
atric rehabilitation for hearing-impaired children (2013). https://idainstitute.com/
fileadmin/user_upload/Downloads/My%20World%20Tool%20Thesis%20Jessica
%20Round.pdf. Accessed 10 Oct 2020
28. Si, M., Marsella, S., Miller, L.: Interactive stories for health interventions. In:
Aylett, R., Lim, M.Y., Louchart, S., Petta, P., Riedl, M. (eds.) ICIDS 2010. LNCS,
vol. 6432, pp. 291–292. Springer, Heidelberg (2010). https://doi.org/10.1007/978-
3-642-16638-9_46
29. Sweeney, D.S., Landreth, G.L.: Child-centered play therapy. In: O’Connor, K.J.,
Braverman, L.D. (eds.) Play Therapy Theory and Practice: Comparing Theories
and Techniques, pp. 123–162. Wiley (2009)
30. Tabbaa, L., Ang, C., Siriaraya, P., She, W., Prigerson, H.: A reflection on vir-
tual reality design for psychological, cognitive and behavioral interventions: design
needs, opportunities and challenges. Int. J. Hum.-Comput. Interact. 37(9), 815–866
(2020). https://doi.org/10.1080/10447318.2020.1848161
31. Vygotsky, L.S.: Mind in Society: The Development of Higher Psychological Pro-
cesses. Harvard University Press, Cambridge (1978)
32. Vygotsky, L.S.: Thinking and speech. In: Rievery, R.W., Carton, A.S. (eds.) The
Collected Works of L. S. Vygotsky, Volume 1: Problems of General Psychology, pp.
39–285. Plenum Press, New York (1987)
33. Westby, C.: Resource review: my world. Word Mouth 29(2), 12–13 (2017)
34. White, M.: Externalizing conversations revisited. In: Morgan, A., White, M. (eds.)
Narrative Therapy with Children and Their Families, pp. 2–56. Dulwich Centre
Publications, Adelaide (2006)
35. White, M.: Maps of Narrative Practice. W.W. Norton & Company, New York
(2007)
36. Wilson, D.K., et al.: Exploring the role of digital storytelling in pediatric oncology
patients’ perspectives regarding diagnosis: a literature review. SAGE Open (2015).
https://doi.org/10.1177/2158244015572099
37. Wyatt, T., Hauenstein, E.: Enhancing children’s health through digital story. CIN:
Comput. Inform. Nurs. 26(3), 142–148 (2008)
38. Wynia, M., Osborn, C.: Health literacy and communication quality in health care
organizations. J. Health Commun. 15(sup2), 102–115 (2010). https://doi.org/10.
1080/10810730.2010.499981
39. Zaman, B., Mifsud, C.: Young children’s use of digital media and parental media-
tion. Cyberpsychol.: J. Psychosoc. Res. Cyberspace 11(3) (2017). https://doi.org/
10.5817/CP2017-3-xx
“What’s Important to You, Max?”: The
Influence of Goals on Engagement
in an Interactive Narrative for Adolescent
Health Behavior Change
Abstract. Interactive narrative technologies for preventive health care offer sig-
nificant potential for promoting health behavior change in adolescents. By improv-
ing adolescents’ knowledge, personal efficacy, and self-regulatory skills these
technologies hold great promise for realizing positive impacts on adolescent
health. These potential benefits are enabled through story-centric learning expe-
riences that provide opportunities for adolescents to practice strategies to reduce
risky health behaviors in engaging game-based environments. A distinctive feature
of interactive narrative that promotes engagement is players’ ability to influence the
story through the choices they make. In this paper, we present initial work investi-
gating engagement in an interactive narrative that focuses on reducing adolescents’
risky behaviors around alcohol use. Specifically, we consider how the short-term
and long-term goals adolescents choose as being important to the protagonist
character relates to their engagement with the interactive narrative. Leveraging
interaction log data from a pilot study with 20 adolescents, we conduct a cluster-
based analysis of the goals players selected. We then examine how engagement
differs between the identified clusters. Results indicate that adolescents’ choices
for the protagonist’s short-term and long-term goals can significantly impact their
engagement with the interactive narrative.
1 Introduction
Many adolescents engage in behaviors that increase morbidity and mortality during
adolescence [1]. In the United States, automobile accidents and unintentional injuries
are the most prominent causes of adolescent deaths, often occurring in conjunction
with using alcohol or other substances [2]. The majority of adolescent health problems
are amenable to behavior change interventions and a wide array of health information
technologies are being explored to promote improved health outcomes in adolescents [3–
8]. Given the popularity of digital games as a source of entertainment for adolescents,
recent years have seen growing interest in leveraging games to help address health
care challenges [e.g., 9–12]. Interactive narrative technologies enable the creation of
rich, story-centric game experiences in which players are active participants shaping the
events of an unfolding story [e.g., 13–17]. Designing game-based interactive narratives
to enhance adolescents’ knowledge, personal efficacy, and self-regulatory skills holds
significant potential for bringing about positive impacts on adolescent health [8].
Inspire is a game-based interactive narrative for adolescent health behavior change
that targets reducing risky behaviors in high school students (ages 14 to 18) [8]. Inspire
features a storyline with gameplay that is grounded in social cognitive theory for behavior
change [18]. By situating players in a prototypical high school party setting, Inspire pro-
vides players with relatable characters that model a broad range of behaviors to promote
learning, challenging situations to practice strategies, and guidance in developing skills
for reducing alcohol use [8]. Inspire aims to foster increased self-regulation through
a variety of mechanisms, including setting and monitoring goals (both short-term and
long-term) [8]. Our previous work on Inspire has shown that adolescents find it to be
engaging, believable, and relevant to their lives [8].
This paper reports on an exploratory cluster-based analysis of how players’ choices
of short-term goals (i.e., what are the protagonist’s goals for the evening) and long-term
goals (i.e., what is important to the protagonist) influence their engagement with Inspire.
Specifically, we investigate the following research questions:
RQ1: What groupings of players automatically emerge based upon the short-term and
long-term goals players select as being important to the protagonist?
RQ2: How does reported engagement with the interactive narrative differ between the
identified groups of players?
To answer these questions, we use data collected in a preliminary pilot study with
20 adolescents. We perform a cluster-based analysis using the short-term and long-
term goals players selected. The clustering algorithm automatically determines the opti-
mal number of clusters and partitioned players into a four-cluster group based on their
selected short-term goals and a two-cluster group based on their selected long-term goals.
We then explore how engagement differs between the clusters in these groups. Find-
ings suggest that the choices adolescents make early in the game for the protagonist’s
short-term and long-term goals can influence their overall engagement with the game.
2 Related Work
Digital games have for many years been a staple in the lives of adolescents. With 97%
of children and adolescents playing games at least one hour per day, there is increasing
interest in understanding the benefits of playing video games [19]. This underscores the
potential of games to serve as a beneficial platform for reaching adolescents. Traditional
entertainment focused games such as “Life is Strange” (2015) by Dontnod Entertain-
ment have appropriated interactive narrative techniques to enhance players’ experiences.
Interactive narrative technologies allow players to have deeper interactions with narra-
tives through the choices they make which can directly impact the outcome of the story.
“What’s Important to You, Max?” 381
Mawhorter and colleagues introduced the choice poetics framework that examines how
choices and their structure impact user experience with a narrative [20]. The presen-
tation of choices can also influence players’ narrative experience [21]. Choices can
help increase players’ agency, by enabling them to have greater responsibility over the
unfolding story to generate meaningful outcomes [22].
In recent years there has been growing interest in the potential of digital games
to promote health behavior change. Digital games for health have tackled a variety of
issues including providing emotional support [9], preventing disease [23, 24], facilitating
communication [25], and teaching healthcare professionals [10, 26]. Digital games for
health have also been designed to promote health behavior change for adolescents. For
example, “Escape from Diab” is an action-adventure game designed to reduce childhood
obesity where the athletic Deejay and his friends must escape from Diab, a world where
it is difficult to acquire healthy food [23]. A key challenge to achieving successful health
behavior change is ensuring players remain engaged with the intervention. Researchers
have started investigating techniques to address this issue by providing tasks to keep
players on track as well as purposely selecting narrative content to maintain engage-
ment [27]. Inspire utilizes a familiar narrative-centric game design as well as impactful
choices to promote health behavior change for adolescents.
Fig. 1. The Inspire interactive narrative for adolescent health behavior change.
382 M. Mott et al.
Inspire features a mystery involving a high school student named Max who discovers
that his little sister is missing the morning after a small get together with his friends,
which he was hosting at his house while his parents were away, turns into an out-of-
control party. In the game, players “relive” the events of the night before. They experience
issues dealing with peer pressure, social norms, and the consequences of alcohol use,
while practicing strategies for reducing risky health behaviors. The narrative sees Max
solve the mystery of where his sister has gone before Max’s parents arrive home.
The Inspire story opens with a short trailer that sets the stage for the events taking
place in the game (Fig. 2). The trailer introduces Max, his little sister Mia, and his close
friends Nikki and Jay who he invites over, while his parents are away for the weekend
at his grandparents. Jay, in turn, asks his friend Hailey to join them. Hailey tells Jay
that it sounds like fun and asks for the address. The introduction ends with a foreboding
message about the night not going as planned.
Fig. 2. Screenshots from the Inspire backstory trailer that introduces adolescents to the game.
The interactive narrative is set in Max’s house where players assume the role of Max,
the story’s protagonist. As Inspire begins, players see Max awakening to the aftermath
of the events from the previous night upon receiving a text message from his parents
letting him know that they will be home earlier than expected (Fig. 3).
After getting out of bed, a short gameplay tutorial introduces players to navigating
Max around the house and interacting with objects in the environment. The tutorial
guides Max to look for his little sister to let her know that their parents will be home
early, but she is not in her room. After hearing a noise, he heads downstairs where Max,
expecting to find his sister, is startled by a stranger in his kitchen. Max remembers seeing
Mia downstairs early on during the previous night, which was shortly before the party
spiraled out of control. However, Max cannot remember where he last saw her. At this
point, players are transported back to the beginning of the night where they are tasked
with reliving the events of the party.
“What’s Important to You, Max?” 383
Fig. 3. The Inspire protagonist awakening the morning after the party.
Fig. 4. Goal-setting interface in Inspire for long-term goals (left) and short-term goals (right).
As adolescents navigate the events of the night, they experience a series of problem
scenarios in the interactive narrative that are realized through branching dialogue with
non-player characters (Fig. 5, left). Players’ interactions with non-player characters
feature 2 to 4 dialogue responses, as well as a timer, that require adolescents to assess
the situation and make choices under time pressure that affect the outcome of the story.
The Inspire storyline features two types of branching narrative structures. The first type
of choice has short-term impact on the story through the immediate reaction of non-
player characters (e.g., Nikki responding enthusiastically to Max after he approves of her
decision to invite additional guests), while the second type has long-term consequences
on the outcome of the story (e.g., Nikki’s excessive drinking resulting in accidentally
breaking a porcelain vase). In addition to the two narrative structures, players are asked
384 M. Mott et al.
Fig. 5. Interfaces for players to select dialogue responses to non-player characters to advance the
narrative: without goal monitoring (left) and with goal monitoring (right).
4 Pilot Study
In this work, we use data from a pilot study that was conducted with Inspire to under-
stand how adolescents engage with the game [8]. A total of 20 adolescents between the
ages of 14 and 19 (M = 16.63, SD = 1.36) from public and private high schools and
community-based after school programs in San Francisco, California participated in the
pilot study. The sample included a diverse group of participants from a range of ethnic
and socioeconomic backgrounds as shown in Table 1 based on data from 19 partici-
pants who provided demographic data. Informed parental consent and adolescent assent
was obtained from all participants in the pilot study under a human subjects approved
protocol.
Gender Race
Female 53% (10) Asian 16% (3)
Male 42% (8) Hispanic or Latino 37% (7)
Transmale 5% (1) Multiracial 26% (5)
White 21% (4)
“What’s Important to You, Max?” 385
4.1 Procedure
4.2 Data
The pre-survey questionnaire asked participants about their video game playing expe-
rience, such as how frequently do you play video games. It also asked about alcohol
use, such as have you used alcohol in the past 12 months. The post-survey questionnaire
included 10 items from a narrative transportation measure [30], which included items
such as “I could picture myself in the scene of the events presented in the narrative”
and “The narrative affected me emotionally.” These items were measured on a 7-point
scale ranging from “Not at all” to “Very much.” The post-survey also included 15 items
from a user engagement measure for video gameplay that focused on satisfaction and
perceived usability, which included items such as “The gaming experience was fun” and
“The gaming experience was demanding” [31]. These items were measured on a 7-point
Likert scale ranging from “Strongly disagree” to “Strongly agree.”
Of the 19 participants who responded to the pre-survey, 5.3% reported playing games
not at all, 36.8% rarely, 31.6% occasionally, 15.8% frequently, and 10.5% very fre-
quently. Only 18 participants chose to report their alcohol use on the pre-survey with
61.1% indicating that they had used alcohol in the past 12 months and 33.3% reporting
that they had at least one drink in the past 30 days. On the post-survey, adolescents
reported a mean score of 4.16 (SD = 0.80) on the narrative transportation items mea-
sured on a 7-point scale, with the average item scores ranging from a high of 6.25 (SD
= 1.16) on “I wanted to learn how the narrative ended” to a low of 2.95 (SD = 1.90) on
“The events in the narrative are relevant to my everyday life.” The mean score on the
engagement measure was 4.93 (SD = 0.78) on a 7-point scale, with the average item
scores ranging from a high of 5.65 (SD = 1.18) on “I felt interested in the game” to a
low of 4.00 (SD = 1.52) on “I found the game confusing to use.”
In addition to the data collected using the pre- and post-surveys, as participants
interacted with Inspire the software recorded all of their interactions to log files on their
tablet device. These interaction logs include records of all the events that occurred during
the interactive narrative, such as navigating from one room to another, interacting with
objects, selecting goals, and making dialogue choices. Across the 20 participants in the
pilot study, over 70,000 events were recorded in the interaction logs.
1 One participant did not complete the pre-survey questionnaire; however, they did complete the
interactive narrative and post-survey questionnaire, so their data is included in the analysis.
386 M. Mott et al.
5 Results
Using the data from the pilot study, we conducted a cluster-based analysis to examine
groups of players based upon the short-term and long-term goals players select as being
important to the protagonist (RQ1) as well as how engagement differs between the
identified groups (RQ2).
Fig. 6. Relative importance of long-term goals for identifying the two clusters.
In order to help interpret the derived clusters, a decision tree classifier was trained
using the 8 binary variables as features and the assigned cluster as the class label. The
decision tree was trained using Weka’s J48 algorithm [33], which resulted in a decision
tree of size 5 that partitions all of the participants into their assigned clusters (Fig. 7).
The decision tree leverages two long-term goals in deciding how to properly label
each participant, including “Having fun” and “Trying new things.” All but one of the
participants is correctly classified by looking at whether “Having fun” was selected by
the participant as a long-term goal for the protagonist (LTC2) or not (LTC1). To correctly
label the remaining participant the “Trying new things” long-term goal is examined. The
difference between these two clusters is further illustrated by the stacked column graph
in Fig. 8, which shows “Having fun” and “Trying new things” as being selected by
participants in LTC2, while “Being healthy” and “Being a good friend” are primarily
associated with participants in LTC1. In short, LTC1 consists of participants who did not
“What’s Important to You, Max?” 387
Fig. 7. Trained decision tree for predicting cluster assignment based on long-term goals.
Fig. 8. Number of participants selecting each long-term goal across clusters LTC1 and LTC2.
select “Having fun” as a goal for the protagonist, while LTC2 consists of participants
who selected “Having fun” or “Trying new things” as a goal.
Using the same clustering approach as presented in Sect. 5.1 above, we also identified
clusters based on the short-term goals participants selected as being important to the
protagonist. The inputs to the clustering algorithm consisted of 8 binary variables (one
for each potential short-term goal) indicating if the participant selected the correspond-
ing short-term goal (Fig. 4, right). Four clusters were automatically identified by the
clustering algorithm: STC1 with a size of 4, STC2 with a size of 8, STC3 with a size of
3, and STC4 with a size of 5. In determining the clusters, the “Stay in control” goal was
identified as the most important variable by the clustering algorithm, while the “Avoid
alcohol” goal was least important (Fig. 9). Another decision tree classifier was trained
using the 8 binary variables for the short-term goals as features and the assigned cluster
as the class label. This resulted in a decision tree of size 7 that correctly classified all of
the participants into their assigned short-term goal clusters (Fig. 10).
The decision tree uses three short-term goals in labeling each participant, including
“Stay in control,” “Resist peer pressure,” and “Look out for your friends.” At the top-level
of the tree, “Stay in control” splits the participants into a group that is either in cluster
STC1 or STC2 or a group that is in either clusters STC3 or STC4. Within the first group the
388 M. Mott et al.
Fig. 9. Relative importance of the short-term goals for identifying the four clusters.
Fig. 10. Trained decision tree for predicting cluster assignment based on short-term goals.
decision tree uses “Resist peer pressure” to correctly partition the participants into either
STC1 or STC2. In the second group, “Look out for your friends” is consulted to decide
between STC3 or STC4. The stacked column graph in Fig. 11 shows the breakdown of
short-term goals participants selected across the four clusters.
Fig. 11. Number of participants selecting each short-term goal across clusters.
Participants in cluster STC1 wanted the protagonist to resist peer pressure and stay
in control, while having a good time. Cluster STC2 consisted of participants who wanted
the protagonist to stay in control and not get in trouble, while looking out for friends and
having a good time with limited alcohol use. Participants in cluster STC3 wanted the
protagonist to stay out of trouble and look out for friends, while resisting peer pressure
and having a good time. Finally, cluster STC4 was composed of participants who wanted
the protagonist to have a good time, while not getting into trouble.
“What’s Important to You, Max?” 389
To answer RQ2 we examined how engagement differed between the clusters identified
based on the short-term goals and long-term goals using independent-samples t-tests.
Prior to using the parametric tests in the analysis, normality was confirmed with the
Shapiro-Wilk’s test.
Participants’ self-reported engagement ranged from 3.00 to 6.27 with a mean of
4.93 (SD = 0.78) on a 7-point scale. Looking at engagement based on short-term goals,
participants in cluster STC1 has an engagement mean score of 4.77 (SD = 0.67), STC2
had a mean score of 4.84 (SD = 0.94), STC3 had a mean score of 5.76 (SD = 0.45), and
STC4 had a mean score of 4.71 (SD = 0.53). A series of independent-samples t-tests
confirmed that engagement scores were significantly different only between participants
in clusters STC3 and STC4, t(6) = 2.854, p = 0.029, indicating that adolescents in STC3,
who did not select “Stay in control” but did select “Look out for your friends”, reported
higher engagement than adolescents in STC4, who did not selected “Stay in control” or
“Look out for your friends” as a short-term goal for the protagonist.
Examining engagement based on long-term goals, participants in cluster LTC1
reported engagement with a mean score of 5.33 (SD = 0.64), while participants in LTC2
reported engagement with a mean score of 4.53 (SD = 0.72). An independent-samples
t-test confirms that the engagement scores were significantly different between the two
groups, t(18) = 2.654, p = 0.016, indicating that adolescents in LTC1, who did not select
“Having fun” as a long-term goal for the protagonist, reported higher engagement than
adolescents in LTC2, who selected “Having fun” or “Trying new things.”
6 Discussion
Through the self-report questionnaires as well as our prior work on Inspire [8], pre-
liminary evidence suggests that a diverse group of adolescents—with-respect-to race,
gender, gaming experience, and alcohol use—found Inspire to be engaging. The story-
line of the game appears to have successfully promoted adolescents’ interest in the game
as measured by items on both the narrative transportation and engagement scales, which
will help drive continued use of the game. This will play an important role in supporting
adolescents in developing mastery across a broad range of situations involving alcohol
use and its effects.
Although adolescents overall reported being engaged with the game (M = 4.93, SD
= 0.78 on a 7-point scale), the cluster analysis based on long-term goals identified a
unique partitioning of players into two groups that reported a statistically significant
difference in their engagement with the game. Deciding how to partition the players
based on their selected long-term goals centered around the “Having fun” and “Trying
new things” goals, while the “Being healthy” and “Being a good friend” goals were
also identified as important. This suggests that one group of players (LTC2) approached
the game wanting the protagonist to focus more on enjoyment and adventure, while the
other group desired the protagonist to focus more on health and friends (LTC1). This
later group reported significantly increased engagement with the game than the group
focused on enjoyment and adventure.
390 M. Mott et al.
Using the short-term goals that players selected as being important to the protagonist,
the cluster analysis partitioned players into four unique groups. Two of these groups
reported a statistically significant difference in their engagement with the game. Both of
these two groups did not select “Stay in control” as a short-term goal, while one (STC3)
selected “Lookout for your friends” and the other (STC4) did not. The analysis of the
clusters based on short-term goals suggests that players who wanted the protagonist to
look out for their friends were more engaged with the interactive narrative than players
who wanted the protagonist to have a good time.
There are two limitations to the findings presented in this work. First, only 20 adoles-
cents participated in the pilot study. This limited the available statistical power, inhibited
some of the analysis we could conduct, and resulted in the sizes of the clusters based on
short-term goals being somewhat small. Second, the pilot study focused on measuring
adolescents’ perceptions of (and engagement with) the interactive narrative since the
primary objective of the pilot was to inform the continued development of the interac-
tive narrative. The ultimate outcomes of changes in adolescents’ self-efficacy and risky
behaviors related to alcohol use have not been examined.
7 Conclusions
Acknowledgements. This research was primarily supported by the National Science Foundation
through grants IIS-1344803 and IIS-1344670 and the National Cancer Institute through grant
R01CA247705. Any opinions, findings, and conclusions expressed in this material are those of
the authors and do not necessarily reflect the views of the National Science Foundation or the
National Cancer Institute.
“What’s Important to You, Max?” 391
References
1. Ozer, E., Irwin, C.: Adolescents and young adult health: from basic health status to clinical
interventions. In: Lerner, L.M., Steinberg, L. (eds.) Handbook of Adolescent Psychology, 3rd
edn. (2009)
2. Kann, L., et al.: Youth risk behavior surveillance. Morb. Mortal. Weekly Report-Surveill.
Summ. 67(8), 1–114 (2018)
3. Beale, I.L., Kato, P.M., Marin-Bowling, V.M., Guthrie, N., Cole, S.W.: Improvement in
cancer-related knowledge following use of a psychoeducational video game for adolescents
and young adults with cancer. J. Adolesc. Health 41(3), 263–270 (2007)
4. Bickmore, T., Zhang, Z., Reichert, M., Julce, C., Jack, B.: Promotion of preconception care
among adolescents and young adults by conversational agent. J. Adolesc. Health 67(2), S45–
S51 (2020)
5. Harris, S.K., et al.: Research on clinical preventive services for adolescents and young adults:
where are we and where do we need to go? J. Adolesc. Health 60(3), 249–260 (2017)
6. Majeed-Ariss, R., et al.: Apps and adolescents: a systematic review of adolescents’ use of
mobile phone and tablet apps that support personal management of their chronic or long-term
physical conditions. J. Med. Internet Res. 17(12) (2015)
7. Rowe, J.P., Lester, J.C.: Artificial intelligence for personalized preventive adolescent
healthcare. J. Adolesc. Health 67(2), S52–S58 (2020)
8. Ozer, E.M., et al.: Fostering engagement in health behavior change: iterative development
of an interactive narrative environment to enhance adolescent preventive health services. J.
Adolesc. Health 67, S34–S44 (2020)
9. Bowman, A.: Creating a virtual support group in an interactive narrative: a companionship
game for cancer patients. In: Rouse, R., Koenitz, H., Haahr, M. (eds.) ICIDS 2018. LNCS,
vol. 11318, pp. 662–665. Springer, Cham (2018). https://doi.org/10.1007/978-3-030-04028-
4_80
10. Christman, G.P., Schrager, S.M., Callahan, K.: Using interactive fiction to teach pediatricians-
in-training about child abuse. In: Nunes, N., Oakley, I., Nisi, V. (eds.) ICIDS 2017. LNCS,
vol. 10690, pp. 278–281. Springer, Cham (2017). https://doi.org/10.1007/978-3-319-71027-
3_25
11. Jurdi, S., Montaner, J., Garcia-Sanjuan, F., Jaen, J., Nacher, V.: A systematic review of game
technologies for pediatric patients. Comput. Biol. Med. 97, 89–112 (2018)
12. Rodriguez, D.M., Teesson, M., Newton, N.C.: A systematic review of computerised serious
educational games about alcohol and other drugs for adolescents. Drug Alcohol Rev. 33(2),
129–135 (2014)
13. Ammanabrolu, P., Cheung, W., Tu, D., Broniec, W., Riedl, M.: Bringing stories alive: gen-
erating interactive fiction worlds. In: Proceedings of the AAAI Conference on Artificial
Intelligence and Interactive Digital Entertainment, vol. 16, no. 1, pp. 3–9 (2020)
14. Kreminski, M., Wardrip-Fruin, N.: Throwing bottles at God: predictive text as a game
mechanic in an AI-based narrative game. In: Rouse, R., Koenitz, H., Haahr, M. (eds.) ICIDS
2018. LNCS, vol. 11318, pp. 275–279. Springer, Cham (2018). https://doi.org/10.1007/978-
3-030-04028-4_29
15. Miller, C., Dighe, M., Martens, C., Jhala, A.: Crafting interactive narrative games with adver-
sarial planning agents from simulations. In: Bosser, A.-G., Millard, D.E., Hargood, C. (eds.)
ICIDS 2020. LNCS, vol. 12497, pp. 44–57. Springer, Cham (2020). https://doi.org/10.1007/
978-3-030-62516-0_4
16. Marc Cavazza, R., Young, M.: Introduction to interactive storytelling. In: Nakatsu, R., Rauter-
berg, M., Ciancarini, P. (eds.) Handbook of Digital Games and Entertainment Technologies,
pp. 377–392. Springer, Singapore (2017). https://doi.org/10.1007/978-981-4560-50-4_55
392 M. Mott et al.
17. Ware, S.G., Garcia, E.T., Shirvani, A., Farrell, R.: Multi-agent narrative experience man-
agement as story graph pruning. In: Proceedings of the AAAI Conference on Artificial
Intelligence and Interactive Digital Entertainment, pp. 87–93 (2019)
18. Bandura, A.: Social cognitive theory for personal and social change by enabling media.
In: Singhal, A., Cody, M.J., Rogers, E.M., Sabido, M. (eds.) Entertainment-Education and
Social Change: History, Research, and Practice, pp. 75–96. Lawrence Erlbaum Associates
Publishers, Mahwah (2004)
19. Granic, I., Lobel, A., Engels, R.C.: The benefits of playing video games. Am. Psychol. 69(1),
66–78 (2014)
20. Mawhorter, P., Mateas, M., Wardrip-Fruin, N., Jhala, A.: Towards a theory of choice poetics.
In: Proceedings of the 9th International Conference on the Foundations of Digital Game
(2014)
21. Estupiñán, S., Maret, B., Andkjaer, K., Szilas, N.: A multidimensional classification of choice
presentation in interactive narrative. In: Rouse, R., Koenitz, H., Haahr, M. (eds.) ICIDS 2018.
LNCS, vol. 11318, pp. 149–153. Springer, Cham (2018). https://doi.org/10.1007/978-3-030-
04028-4_12
22. Rezk, A.M., Haahr, M.: The case for invisibility: understanding and improving agency in
Black Mirror’s Bandersnatch and other interactive digital narrative works. In: Bosser, A.-G.,
Millard, D.E., Hargood, C. (eds.) ICIDS 2020. LNCS, vol. 12497, pp. 178–189. Springer,
Cham (2020). https://doi.org/10.1007/978-3-030-62516-0_16
23. Thompson, D., et al.: Serious video games for health how behavioral science guided the
development of a serious video game. Simul. Gaming 41(4), 587–606 (2010)
24. Miller, L.C., Appleby, P.R., Anderson, A.N., Christensen, S., Marsella, S.: SOLVE-IT:
socially optimized learning in virtual environments: a web delivered HIV prevention 3D
game intervention for young at-risk MSM. J. Mob. Technol. Med. 1(4S), 10 (2012)
25. Baceviciute, S., Albæk, K.R.R., Arsovski, A., Bruni, L.E.: Digital interactive narrative tools
for facilitating communication with children during counseling: a case for audiology. In:
Oyarzun, D., Federico Peinado, R., Young, M., Elizalde, A., Méndez, G. (eds.) ICIDS 2012.
LNCS, vol. 7648, pp. 48–59. Springer, Heidelberg (2012). https://doi.org/10.1007/978-3-642-
34851-8_5
26. Morningstar-Kywi, N., Kim, R.E.: Using interactive fiction to teach clinical decision-making
in a PharmD curriculum. Med. Sci. Educ. 31(2), 687–695 (2021)
27. Langxuan, Y., Bickmore, T., Montfort, N.: An interactive narrative system for narrative-based
games for health. In: Intelligent Narrative Technologies 7: Papers from the 2014 Workshop,
pp. 45–51 (2014)
28. Bandura, A.: Self-efficacy mechanism in human agency. Am. Psychol. 37(2), 122–147 (1982)
29. Ozer, E.M., Bandura, A.: Mechanisms governing empowerment effects: a self-efficacy
analysis. J. Pers. Soc. Psychol. 58(3), 472–486 (1990)
30. Green, M.C., Brock, T.C.: The role of transportation in the persuasiveness of public narratives.
J. Pers. Soc. Psychol. 79(5), 701–721 (2000)
31. Wiebe, E.N., Lamb, A., Hardy, M., Sharek, D.: Measuring engagement in video game-based
environments: investigation of the User Engagement Scale. Comput. Hum. Behav. 32, 123–
132 (2014)
32. Chiu, T., Fang, D., Chen, J., Wang, Y.: A robust and scalable clustering algorithm for mixed
type attributes in large database environment. In: Proceedings of the 7th ACM SIGKDD
International Conference on Knowledge Discovery and Data Mining, pp. 263–268 (2001)
33. Hall, M., Frank, E., Holmes, G., Pfahringer, B., Reutemann, P., Witten, I.H.: The WEKA data
mining software: an update. ACM SIGKDD Explor. Newsl. 11(1), 10–18 (2009)
Transfordance: The Decentering Effect
of Transformative Affordances in Virtual
Reality in The Hollow Reach
When you look through the Virtual Reality (VR) lenses of The Hollow Reach (THR),
the deck of a starship, reminiscent of Star Trek, stretches around you; at your side,
an officer at attention. Your task: to complete a set of exercises to practice using your
artificial limb. But the serene background of lightspeed is interrupted by your arrival to
a new system. Appearing on the viewscreen is a disabled ship, drifting and identified
as the Pelops. An enemy warship appears, the officer requests your orders. As a player,
in any new interactive digital narrative (IDN), the question is the same: what can I do?
The experience focuses on recovery from trauma through both mechanical and virtual
prosthesis. A sequence of challenges connect interactions with characters, where they
respond to the player’s gaze or tone. The player’s actions are also not reliable, departing
from VR conventions. The player’s speech content in dialogue choices is less important
than the tone and gaze tendency when determining how characters respond.
The piece sets out to ask “How can we experience differently-abled bodies in an
immersive medium?” What unique opportunities do VR IDNs offer to shape player
expectations and understandings? This paper describes the design of THR and introduces
the concepts and theories used in its development. The project is available for download1
and runs on any OpenXR compatible headset.
1.1 VR Narratives
Participants have many ways of performing actions and navigating in simulated envi-
ronments. Perhaps they interact by walking and talking in a cartoonish 3-D world, as in
the AR adaptation of Façade or move through breathing in the scuba-inspired works of
Char Davies, Osmose (1995) and Ephèmere (1998). However, most of these VR experi-
ences map movement and other interaction onto a simulation of an able-bodied human
moving through space, no matter how fantastic or infinite the virtual reality. Neither
the virtual reality nor the embodied experience of it are equivalents to lived reality, so
designers determine how to represent a boundless virtual space exists within a normative,
embodied sensory experience.
Since the physical space around a VR user is limited, one technique is to constrain
the visual environment. For example, in I Expect You to Die [1], an escape-room homage
to the James Bond franchise, the player finds themself in tight places such as the interior
of a car or an office. On the other hand, works such as Virtual Virtual Reality [2] and The
Under Presents [3] offer innovative transportation mechanics, casting about as if fishing
and dragging themselves. In Hot Dogs, Horseshoes, and Hand Grenades [4], players
jog through the Weiner-filled realm by pumping their arms as if running.
While embodied movement is a necessary and obvious part of most VR experiences,
we argue that designed interactions and input benefit from narrative framing. However,
in our IDN, our goal was not to portray, but to decenter the normative sensory experi-
ence. We accomplished this by subverting existing interaction expectations, which led
us to invent the term transfordance. Transfordance describes the interactions that decen-
ter normative embodied experience for expressive purposes through narrative framing.
Transfordance is a portmanteau of “transformative affordance.” An affordance is “the
relationship between a physical object and a person” [7]. Rather than just a sensorimotor
substitution, a “transfordance” is an affordance, that evokes a re-assessment by the user
of previous affordances. In other words, a transfordance decenters what was previously
treated as “normal” affordances.
2 Narrative-Based Explorations
Direct manipulation is well studied in VR user interfaces [15], so our narrative centers
on the experience of an unwieldy prosthesis. This frame was inspired by the use of VR
to aid in rehabilitation of sensorimotor cognition [16]. To be clear, most VR systems can
register at least some hand or arm movement. In THR, we narrate that exploration of a
character’s experiences with disability through the fiction of a prosthesis, as the main
character must overcome physical and emotional traumatic loss of their arms through
gaining acumen with prosthesis. At the start, we frame inability to control this virtual
limb as the player character’s lack of sensorimotor control.
Players in IDNs take actions such as making dialogue choices, using items, or navigating
spaces. VR relies on gestures that imitate real-world interactions, such as grasping or
pointing. Instead of pressing X, players can reach and grab an object by performing an
embodied actions. The goal of THR is for players to experience a narrative of embodied
otherness through the following mechanics:
Destabilizing Bodies. In the diegetic world of THR, the player’s use of their limb does
not directly relate to their controller position. After the first period of practice, they knock
a coffee cup to the ground, shattering at once the cup and the illusion of the diegetic
virtual world. The player is jolted into the “real world” of the game, a desolate, empty
apartment. The loss of control is a false choice [17], such as used in Depression Quest
396 J. T. Murray and M. C. Marino
[18], that enacts a lack of player-character agency. The perceived limitation in agency
becomes the means of delivering a non-normative subjective experience.
Gaze as Input. In THR, the player’s head direction factors into the dialogue in subtle
ways, signifying the player’s attention to the world or a character. In THR, the environ-
ment is designed to distract from a character’s gaze. The player’s love interest, Jordan,
commands the player to confront her emotional pain by meeting her gaze while also
looking at (and examining) all that the player has so painfully lost physically and emo-
tionally in order, of course, to overcome that loss. If the player chooses not to, the
character observes and responds to that choice as if the player acted more explicitly.
The player’s head position is considered a latent part of the experience of VR, since “look-
ing around” is a fundamental capability. However, in THR, your gaze is an involuntary
expression of attention, and as in social situations, a means of building or damaging
relationships. For instance, confronting their hostile challenge of the alien commander
or an offer of help from a loved one in their traumatized state. In other words, we treated
gaze as an act. Another example is the player’s inability to control the prosthetic limb,
the subversion of direct control integrated into the player’s character arc. Inability to
control a prosthesis in VR becomes the narrative struggle with disability and pulls the
player in through the control that is often taken for granted. This can be extended to
speech. When our player character’s love interest asks, “Why are you shutting me out?”
the player must attempt some verbal defense. The response is determined not by rec-
ognizing words, but by the emotions recognized [20], using Arezina’s technique [21].
Now, as in the case of the gaze and the prosthetic arm, the player must grapple with an
input whose narrative implications are so overdetermined and entangled as to make it
the dialogue equivalent of a prosthesis that cannot be manipulated with certainty. We
frame the player’s struggle to make themselves recognized by the system through the
narrative as a struggle to communicate difficult and delicate emotional state. It is that
tangled and tortured human condition of plays, novels, and games such as Façade built
into the mechanic but rendered meaningful through the narrative.
Our demo highlights how a designed difference between desired interaction and
user experience doesn’t have to be either a failure of interface design nor a novel way to
aestheticize frustration. The Hollow Reach demonstrates one way the gap can be narrated
and rendered meaningful through transfordance. The player’s negotiations of virtual
reality are rendered narratively significant by being framed as a character’s struggle to
Transfordance: The Decentering Effect of Transformative Affordances 397
come to terms with physical reality. Transfordant design offers opportunities for further
narrative exploration of the difference between hegemonic embodied experience and
alternatives as a means of drawing the player into the story while they are drawn into
the mechanisms of the game.
References
1. Schell, J.: Making Great VR: Six Lessons Learned From I Expect You To
Die. https://www.gamasutra.com/blogs/JesseSchell/20150626/247113/Making_Great_VR_
Six_Lessons_Learned_From_I_Expect_You_To_Die.php. Accessed 18 July 2018
2. Virtual Virtual Reality. Tender Claws (2017)
3. The Under Presents. Tneder Claws (2019)
4. Hot Dogs, Horseshoes & Hand Grenades
5. LaViola, J.J., Jr., Kruijff, E., McMahan, R.P., Bowman, D., Poupyrev, I.P.: 3D User Interfaces:
Theory and Practice. Addison-Wesley Professional, Boston (2017)
6. Azmandian, M., Hancock, M., Benko, H., Ofek, E., Wilson, A.D.: Haptic retargeting: dynamic
repurposing of passive haptics for enhanced virtual reality experiences. In: Proceedings of
the 2016 CHI Conference on Human Factors in Computing Systems, pp. 1968–1979 (2016)
7. Bergström, J., Mottelson, A., Knibbe, J.: Resized grasping in VR: estimating thresholds for
object discrimination. In: Proceedings of the 32nd Annual ACM Symposium on User Interface
Software and Technology, pp. 1175–1183 (2019)
8. Dürr, M., Weber, R., Pfeil, U., Reiterer, H.: EGuide: investigating different visual appear-
ances and guidance techniques for egocentric guidance visualizations. In: Proceedings of
the Fourteenth International Conference on Tangible, Embedded, and Embodied Interaction,
pp. 311–322. Association for Computing Machinery, New York (2020). https://doi.org/10.
1145/3374920.3374945
9. Gonzalez-Franco, M., Cohn, B., Ofek, E., Burin, D., Maselli, A.: The self-avatar follower
effect in virtual reality. In: 2020 IEEE Conference on Virtual Reality and 3D User Interfaces
(VR), pp. 18–25. IEEE (2020)
10. Mateas, M., Wardrip-Fruin, N.: Defining operational logics. In: Proceedings of the Digital
Games Research Association, pp. 1–8 (2009)
11. Osborn, J.C., Wardrip-Fruin, N., Mateas, M.: Refining operational logics. In: Proceedings
of the 12th International Conference on the Foundations of Digital Games, pp. 1–10. ACM,
Hyannis Massachusetts (2017). https://doi.org/10.1145/3102071.3102107
12. Douglass, J.: Command Lines: Aesthetics and Technique in Interactive Fiction and New
Media. ProQuest (2007)
13. Bertrand, P.: The machine to be another - embodied telepresence using human performers, 7
14. Cárdenas, M., Head, C., Margolis, T., Greco, K.: Becoming dragon: a mixed reality durational
performance in second life. In: The Engineering Reality of Virtual Reality 2009, p. 723807.
International Society for Optics and Photonics (2009). https://doi.org/10.1117/12.806260
15. Lilija, K., Kyllingsbæk, S., Hornbæk, K.: Correction of avatar hand movements supports
learning of a motor skill. In: 2021 IEEE Virtual Reality and 3D User Interfaces (VR), pp. 1–8
(2021). https://doi.org/10.1109/VR50410.2021.00069
16. Tieri, G., Morone, G., Paolucci, S., Iosa, M.: Virtual reality in cognitive and motor rehabili-
tation: facts, fiction and fallacies. Expert Rev. Med. Dev. 15, 107–117 (2018). https://doi.org/
10.1080/17434440.2018.1425613
17. Mawhorter, P., Zegura, C., Gray, A., Jhala, A., Mateas, M., Wardrip-Fruin, N.: Choice poetics
by example. Arts. 7, 47 (2018). https://doi.org/10.3390/arts7030047
18. Quinn, Z.: Depression Quest (2013)
398 J. T. Murray and M. C. Marino
19. Maloney, M.: Out on a Limb: Practical Approaches to Branching Story (2017)
20. Mustaqeem, Kwon, S.: A CNN-assisted enhanced audio signal processing for speech emotion
recognition. Sensors 20, 183 (2019). https://doi.org/10.3390/s20010183
21. Arezina, K.: Live Speech Emotion Categorization. https://medium.com/nerd-for-tech/live-
speech-emotion-categorization-7692933037ba. Accessed 15 Oct 2021
Contextualization of Design Qualities
in Interactive Story-Based Visualization Applied
to Engineering
1 Introduction
Growing data complexity and abundance, coupled with extreme performance require-
ments of pressure-filled, mission-critical, or other engineering-specific applications,
increase the demand for robust data exploration tools. Visualization is the most widely
employed technique for rapid knowledge extraction, insight enhancement, and informa-
tion retention. It takes advantage of the human perceptual system to mitigate cognitive
overload and ameliorate comprehension. It has been extensively used to aid engineers
and scientists in a variety of tasks, such as design, simulation, monitoring and mainte-
nance, remote collaboration, training, evaluation and optimization, risk modeling and
assessment, decision-making, and presentation. Visualization techniques are constantly
evolving to fulfill the requirements of their applied field and keep up with technological
interactivity, and engagement are mutually reshaped when the same immersive tech-
nologies are applied to different domains [22]. They compared different disciplines in
media entertainment (video games and cinematography). For industrial visualization,
which is a disparate experience form, this means a further re-evaluation of established
concepts. Ma et al. [4] emphasized the differences in key narrative elements (setting,
character, and plot) of traditional and scientific storytelling. According to them, in tra-
ditional media, such as theatre, film, or gaming, the setting is treated as the physical
or virtual environment in which the story is set, while in visualization it is the back-
ground information that provides context. Characters, which would normally be people
or protagonists, are visual elements or data in visualization. Visualization’s plot is the
comparison, interaction, and evolution of visual elements.
In IN theory, the narrative progression is controlled by the author, observer, or both.
Interactivity is determined by varying degrees of autonomy provided to the user. Wohl-
fart and Hauser proposed a model for different kinds of story consumption based on
multiple levels of control between the author and the observer [23]. The example of
full author-driven control is passive, non-interactive visualization; while the extreme
opposite is liberal data exploration, dictated by the end-user. Segel and Heer explored
the balance between author and reader-driven experiences through the identification of
different genres of narrative visualization [24]. In evolving systems that employ robust
analytics and learning techniques, visualization can also progress algorithmically. This
type of storyline development takes data iterations as input. Algorithmic changes in
visualization’s narrative progression can be treated as an author-driven approach, as
AI-generated work can be extrapolated to be human-generated [25].
extensive list of visualization qualities and it should be treated as a starting point for
further research.
3.1 Aesthetics
Beauty can be used as a guiding principle in the practice of sciences and systems engi-
neering [26]. The concept of aesthetics, its usage, and effect on storytelling and visual-
ization has been extensively discussed in the literature. It has caused polarized opinions
among visualization researchers. Iliinsky outlined this dichotomy, noting that beauty in
visualization can be achieved through compelling graphical construction of visual ele-
ments, or as another opportunity to increase the utility of the visualization [27]. While
merely decorative elements might not be in the interest of practitioners from engineering
and sciences, they might appreciate aesthetics for its potential to add to the visualiza-
tion’s efficacy. For Krzywinski, form with function is beautiful [28]. Kyndrup argued
that beauty is an independent quality and should, therefore, be assessed separately from
the primary goals of the visualization [29]. He did, however, express the concern that
too beautiful visualizations can disturb the viewer’s attention and perception. Beauty
is a subjective judgment that is difficult to quantify, so the appropriate level of beauty
for visualizing data remains abstract. Benefits of aesthetics for data representation have
been mostly addressed through its relationship with usability, which is the principal pre-
requisite for implementation of interactive experiences. Studies have shown a positive
correlation between aesthetics and usability, a strong relationship between the perceived
appearance and the usability of the interface [18, 30], and that users handle aesthetic
visualizations more patiently and profoundly [17]. However, a recent study by Flan-
gas et al. has failed to prove that aesthetic elements of the visual interface affect user
performance or decision fatigue [31].
In interactive digital narrative (IDN) theory, aesthetics is understood through Mur-
ray’s seminal model [32]. She proposed three aesthetic categories for interactive story-
telling analysis: immersion, agency, and transformation, which will be briefly discussed
in the context of narrative visualization.
Immersion is the sense of physical or mental absorption in content or medium, which
is experienced by an interactor. It is a sensation and experiential pleasure of taking part
in a synthetic reality in which the user’s actions impact the immersive environment and
follow its logic. Raja et al. showed that immersion in visualization results in better user
performance, especially when viewing large datasets [33]. Aesthetic visualizations can
facilitate a greater mental immersion into the data [9].
Agency denotes the user’s ability to take meaningful action, which will happen in a
dynamically responsive world. The changes should not be a result of arbitrary exploration
of utilities, they rather have to reflect the user’s intentions. While in IDN theory agency
originally refers to user’s responses within the domain of the digital environment, in
narrative visualization the concept can be extended to the physical world, too. In some
engineering applications (e.g., cyber-physical systems), the user can act as a bond within
the real and the virtual world and send input to both ends of the system.
The beauty of narrative also comes from the transformational experience that can
happen on multiple levels. Transformation can manifest through variations on the theme
Contextualization of Design Qualities 403
or role play. For narrative visualization, this can help the viewer gain a better under-
standing of data by providing insight in various forms (e.g., layering the data, providing
different viewing angles, or using diverse graphic formats). Personal transformation is
the ability of storytelling to change the viewer. It is crucial for the success of serious
applications [34], such as engineering or scientific applications. In story-based visual-
ization, personal transformation can be achieved through learning, knowledge discovery,
and gaining a novel outlook on the subject matter.
3.2 Usability
Usability is tremendously important for interactive narrative visualizations, as it is a
prerequisite for data exploration and message extraction. Lack of usability in engineering
applications poses health, safety, malfunction, and error risks, which can have serious
financial or fatal consequences. Usability has been widely evaluated as a dimension of
user experience for IDNs [35]. From a human-computer interaction perspective, usability
can be understood as the user’s experience with both hardware and software. For an
interactive experience, system usability is essential for its potential to influence efficacy,
autonomy, and meet user expectations [34]. Roth and Koenitz identified usability as a
subcategory of agency using Murray’s model. Thus, it can be interpreted as an aesthetic
experiential category of IDNs. We decided to treat the terms aesthetics and usability
separately, as they bear different meanings in visualization research.
Usability can be interpreted as readability or visual literacy in visualization systems.
It manifests as the user’s ability to receive and disentangle content. As a visual explo-
ration technique, visualization is prone to misinterpretation and mismatching between
the creator’s intent and the viewer’s understanding.
Some limitations of hardware interaction arise from the user’s unfamiliarity with the
technology or adverse effects on the user (such as cyber or motion sickness). Failure to
adequately interface with the software comes from the user’s inability to navigate, oper-
ate, and manipulate the system and results in misapprehension, confusion, frustration,
and task abandonment.
3.3 Novelty
In the context of story-based interactive visualization, novelty refers to perceived inno-
vation and originality in different elements of the experience [35]. Novelty is powerful
for narrative visualization because it triggers interest, facilitates learning, and elicits cog-
nitive pleasure [36]. Unfamiliar or unexpected visualizations and interactions are more
memorable to the user [37, 38]. A user study by Kolhoff and Nack [39] showed that nov-
elty initiated the engagement of users with interactive narrative content. Cawthon and
Moere compared various visualization methods and disproved the notion that familiar
examples, which people are more exposed to, accustomed to, and trained to comprehend,
result in high effectiveness [17].
Original visualization techniques can, however, be repelling if they aggravate the
readability of the visual narrative. Novelty can negatively affect usability, which is a
trade-off that has to be carefully considered for serious applications. According to the
theory of preference-for-prototypes, people are biased towards the most familiar choices
404 J. Plavšić and I. Mišković
because they minimize the risk of the unknown [10, 40]. In time-sensitive and mission-
critical engineering applications, decision-makers will want to avoid unpredicted sce-
narios and they might perform better in a well-known environment. It is easier to perform
tasks in computer interfaces that take advantage of previously learned techniques [41].
This is particularly relevant to non-tech-savvy users, who will need to be trained to
maneuver novel hardware or software solutions. Higgins and Howell explained that a
stimulus that is excessively novel triggers a displeasure response and results in disen-
gagement with the stimuli [42]. They also noted that the threshold for unpleasurable
arousal is subjective and it differs between individuals. In the engineering environment,
it will be important to determine how comfortable the users are with novelty and in
which domains. It will be crucial to mitigating the possibility of an adverse reaction that
can lead to negative feelings, errors, and task abandonment.
3.4 Complexity
The complexity of IN for the visual articulation of data is expressed as the degree of
cognitive effort needed to extract and receive information, the quantity of content, or
the number of options the narrative provides. We approached the complexity problem
from two viewpoints: data complexity (e.g., types of data, layers of data, tasks to be
performed), and narrative complexity (e.g., the number of decision points or autonomy
levels).
First, we will consider the concept of data complexity. Presentation of complex and
massive amounts of real-time information is extremely difficult, even when using visual
media [43]. Data has to be organized and structured efficiently, so it doesn’t overload
the receiver. The complexity and scale of data are increasing, as insight gained from
analytics is at the core of the Industry 4.0 revolution. For real-time engineering appli-
cations, this means that optimization techniques need to be employed to increase the
visualization’s efficiency and prevent bottlenecks. Some data complexity issues can be
resolved using custom optimization techniques (e.g., limiting the polygon count and
introducing different levels of detail in a virtual environment for immersive applica-
tions), while others are dependent on new generations of robust technology (e.g., using
5G network to instantaneously send and process data from the physical to the virtual
world, and vice versa). Excessive intricacy can be also solved by carefully designing
and organizing information. The rule of the thumb is to manage it by layering and dis-
playing only the necessary information at a time [44]. Even though additional visual
information can increase the system performance and mental effort requirements, it can
aid the knowledge-grasping process [41]. Preference for complexity or simplicity varies
between individuals. Some people tend to enjoy cognitive effort [42] and they will enjoy
performing complex tasks more than those in a lower need for cognition, who opt for
simple tasks [45]. Personal preferences and mental capabilities of users need to be taken
into account when designing narratives for interactive engineering visualization.
In this paragraph, we will discuss the complexity of the narrative. Using IDNs to
comprehend complex phenomena has received more attention in recent years, and it is
best seen through initiatives such as INDCOR1 . Complexity provokes interest [20], so it
1 https://indcor.eu/.
Contextualization of Design Qualities 405
initiates involvement with the narrative and motivates users to spend more time engaging
in it. In the context of IDNs, complexity is best correlated with interactivity. Creators
have to balance the interactive complexity and the quality of the IDN artifact [35]. Giving
the users too much freedom can result in unexpected behavior in which users can, for
instance, navigate away from the area of interest. Complexity can also be related to
autonomy, the independence to choose from a series of options without the feeling of
being pulled in a single direction [34]. The number and the quality of available options to
influence the narrative has to satisfy the user’s expectations and meet the author’s intent.
Too much content or available options result in increased anxiety [46]. Limitations are
also necessary for designing a usable system, as complex systems and interfaces are
harder to use [47].
The goal of this paper was to examine design qualities relevant for visualization research
and evaluate them in the context of interactive narrative visualization for engineering
applications. Our motivation was to aid practitioners in the design process by setting up
a theoretical foundation based on existing literature. We identified four design qualities
(aesthetics, usability, novelty, and complexity) that are relevant for visualization research
and which have a great potential for interactive story-based visualization in engineer-
ing. We discussed their interrelationships and influence on visualization efficiency (see
Fig. 1). Our aim was not to create a comprehensive list, but rather to provoke critical
thinking and provide a starting point for further development of this nascent field.
Fig. 1. Diagram presenting the four design qualities of interactive narrative visualization for
engineering applications and their interrelationships.
1. Terminology needs to be re-evaluated in the new context. In the first part of this
paper, we discussed the idea of narrative visualization and identified the need to re-
evaluate IN concepts in the new context. We further compared the shared concepts
of aesthetics, usability, novelty, and complexity in visualization and IN research, and
showed similarities and differences in their usage and understanding.
406 J. Plavšić and I. Mišković
2. All four design dimensions we analyzed are subjective and prone to personal inter-
pretation. Different reactions, comfort levels, and preferences among users are asso-
ciated with factors such as demographics, previous experiences, intellectuality, per-
sonal taste, etc. Designing a good visualization is a complex process and while some
guidelines exist, there is no guarantee that they will be successful in a particular
context. More user studies will need to be conducted as the field develops.
3. All four design qualities can be both positive and negative, depending on the con-
text. Closely related to the first statement, presented concepts can be beneficial or
disadvantageous as a result of a subjective judgment. There is a point beyond which
subjects start experiencing negative effects and finding that threshold for every indi-
vidual might be crucial. Also, different applications will require different design
principles. Our research focused on the needs of the engineering industry, which can
differ from those of other industries. Differences can be also made between different
engineering subdisciplines and specific applications.
4. Trade-off between design qualities has to be made. Design qualities can also posi-
tively or negatively affect each other. Visualization designers will need to determine
the end goal of visualization and give priority to features that best reflect the desired
intent. This network of mutual relationships of visualization or story components
will also vary with the application.
5. There is a lack of completed case studies in the field. Despite the great potential
of interactive storytelling for visualizing engineering data, the lack of completed
case studies is evident. While work in the field of narrative visualization has been
addressed in the literature, it only touches upon engineering applications. More
practical visualization solutions are needed to expand our list and set up a theoretical
foundation of the field.
The greatest limitations of our work come from a lack of available case studies and
a small number of qualities compared. Different conclusions might be drawn from ana-
lyzing practical visualization examples and expanding the list of design qualities. Also,
some of the research conducted in the visualization field is specific for 2D, static infor-
mation visualization, and might not be directly applicable to more complex visualization
formats, and vice versa. We interpreted the term visualization broadly because interac-
tive narrative visualization for engineering applications combines various visualization
methods. We believe our research is the first step to establish this emerging visualization
field, identify its requirements, and conceptualize adequate design practices. We hope
these results will help to bridge the gap between storytellers, visualization designers,
and engineers and expedite collaborative efforts to employ interactive narratives for
visualizing engineering data.
References
1. Ghidini, E., Santos, C.Q., Manssour, I., Silveira, M.S.: Analyzing design strategies for nar-
rative visualization. In: Proceedings of the XVI Brazilian Symposium on Human Factors in
Contextualization of Design Qualities 407
21. Furnell, S.: Usability versus complexity – striking the balance in end-user security. Netw.
Secur. 2010, 13–17 (2010). https://doi.org/10.1016/S1353-4858(10)70147-1
22. Vosmeer, M., Schouten, B.: Interactive cinema: engagement and interaction. In: Mitchell, A.,
Fernández-Vara, C., Thue, D. (eds.) ICIDS 2014. LNCS, vol. 8832, pp. 140–147. Springer,
Cham (2014). https://doi.org/10.1007/978-3-319-12337-0_14
23. Wohlfart, M., Hauser, H.: Story telling for presentation in volume visualization. In:
Eurographics/IEEE-VGTC Symposium on Visualization, pp. 91–98 (2007). https://doi.org/
10.2312/VISSYM/EUROVIS07/091-098
24. Segel, E., Heer, J.: Narrative visualization: telling stories with data. IEEE Trans. Visual.
Comput. Graph. 16, 1139–1148 (2010). https://doi.org/10.1109/TVCG.2010.179
25. Eshraghian, J.K.: Human ownership of artificial creativity. Nat. Mach. Intell. 2, 157–160
(2020). https://doi.org/10.1038/s42256-020-0161-x
26. Devaney, K.: Beauty as a guiding principle for systems engineering. In: INCOSE International
Symposium, vol. 26, pp. 1448–1462 (2016). https://doi.org/10.1002/j.2334-5837.2016.002
38.x
27. Iliinsky, N.: On beauty. In: Steele, J., Iliinsky, N. (eds.) Beautiful Visualization, pp. 1–13.
O’Reilly Media, Inc.
28. Krzywinski, M.: Scientific data visualization: aesthetic for diagrammatic clarity. In: Philipsen,
L., Kjærgaard, R.S. (eds.) The Aesthetics of Scientific Data Representation More than Pretty
Pictures, pp. 22–35. Routledge (2017)
29. Kyndrup, M.: “Facts” - and representational acts. In: Philipsen, L., Kjærgaard, R.S. (eds.)
The Aesthetics of Scientific Data Representation More than Pretty Pictures, pp. 123–132.
Routledge (2017)
30. Tractinsky, N., Katz, A.S., Ikar, D.: What is beautiful is usable. Interact. Comput. 13, 127–145
(2000). https://doi.org/10.1016/S0953-5438(00)00031-X
31. Flangas, A., Tudor, A.R., Harris, F.C., Dascalu, S.: Preventing decision fatigue with aestheti-
cally engaging information buttons. In: Yamamoto, S., Mori, H. (eds.) HCII 2021. LNCS, vol.
12765, pp. 28–39. Springer, Cham (2021). https://doi.org/10.1007/978-3-030-78321-1_3
32. Murray, J.H.: Hamlet on the Holodeck: The Future of Narrative in Cyberspace. MIT Press,
Cambridge (1998)
33. Raja, D., Bowman, D., Lucas, J., North, C.: Exploring the benefits of immersion in
abstract information visualization. In: Proceedings of the Immersive Projection Technology
Workshop, pp. 61–69 (2004)
34. Roth, C., Koenitz, H.: Evaluating the user experience of interactive digital narrative. In:
Proceedings of the 1st International Workshop on Multimedia Alternate Realities, Amsterdam,
The Netherlands, pp. 31–36. ACM (2016). https://doi.org/10.1145/2983298.2983302
35. Revi, A.T., Millard, D.E., Middleton, S.E.: A systematic analysis of user experience dimen-
sions for interactive digital narratives. In: Bosser, A.-G., Millard, D.E., Hargood, C. (eds.)
ICIDS 2020. LNCS, vol. 12497, pp. 58–74. Springer, Cham (2020). https://doi.org/10.1007/
978-3-030-62516-0_5
36. Biederman, I., Vessel, E.: Perceptual pleasure and the brain. Am. Sci. 94, 247–253 (2006).
https://doi.org/10.1511/2006.59.247
37. Gough, P.: From the analytical to the artistic: a review of literature on information
visualization. Leonardo 50, 47–52 (2017). https://doi.org/10.1162/LEON_a_00959
38. Larsen, B.A., Schoenau-Fog, H.: “Well, That was Quick” – towards storyworld adaptivity that
reacts to players as people. In: Cardona-Rivera, R.E., Sullivan, A., Young, R.M. (eds.) ICIDS
2019. LNCS, vol. 11869, pp. 204–213. Springer, Cham (2019). https://doi.org/10.1007/978-
3-030-33894-7_22
39. Kolhoff, L., Nack, F.: How relevant is your choice? In: Cardona-Rivera, R.E., Sullivan, A.,
Young, R.M. (eds.) ICIDS 2019. LNCS, vol. 11869, pp. 73–85. Springer, Cham (2019).
https://doi.org/10.1007/978-3-030-33894-7_9
Contextualization of Design Qualities 409
Hélène Sellier(B)
Dating back from the 2000’s Art Game trend [1] and amplified with the expansion of
the Indie Games at the end of the decade [2] as well as the rise of accessible authoring
tools such as Unity [3] and Twine [4] (which gave birth to different artistic endeavours
such as Serious Games [5], Games for Change or Queer Games [6]), there is a growing
wave of games that tackle social issues concerning cultural diversity and the inclusion
of varied identities by delivering a message or painting a personal experience. Thanks
to their ludo-narrative structure, they aim to change society by raising awareness about
existing problems, questioning social norms or empowering marginalized individuals.
The idea of a discourse embedded in the game was famously theorized by Ian Bogost
in his book Persuasive Games [7] and his conclusion infiltrated the industry, leading to
debates such as the ludo-narrative dissonance debate [8].
The video games which emerged from this background often give a specific opinion
or share a particular perspective on the social issues they discuss. However, the socio-
cultural fabric is made of a variety of points of view and a multiplicity of ways to
deal with them (stereotypes as a cognitive model, internalized self-prejudices, cognitive
dissonance, etc.). For example, even within groups of people invested in the same issue
(such as gender equality or disability), there are several conflicting views [9, 10].
To depict issues regarding cultural diversity, inclusiveness of different identities and
social diversity, it seems adequate to find a way to allow “a genuine polyphony of fully
valid voices” [11] than adopting a single teaching stance, a moralizing discourse, or a
dominating perspective. This essay presents a research project, undertaken by the team
of the indie game studio The Seed Crew, which aims to conceptualize and build a ludo-
narrative model that enables the expression of different feelings and thoughts on multi-
factorial and sensitive issues. Based on the idea that a consensus can emerge from the
dissensus [12] it intends to bring people together and to foster social interconnectedness.
Following the recent developments of Game Studies as well as the evolution of
the concerns in the professional world, a solution might be found with the approaches
that emphasize the role of the player in the construction of the game’s meaning. The
project is thus grounded in several frameworks: firstly, Post-classical Narratology theo-
ries that focus on the reader’s interpretation [13, 14]; secondly, Play Studies that posit
the importance of the player’s experience [15] and their applications, such as Expressive
Games [16]; lastly, Narrative Design since this creative discipline focuses on the user’s
involvement [17].
In this essay, I will begin by describing the methodology of the project, then I will
explain its current state and conclude by establishing the plan for further research.
To obtain a ludo-narrative model which is both conceptually interesting and pragmat-
ically viable, methods from the field of research-creation [18, 19] are used. The project
links theory, which is actualized in scientific publications and communications, and prac-
tice, which leads to the development of RecovR1 . This video game, intended to be a tool
for professional training, is divided into several modules, each related to a social issue
(sexism, ableism – i.e. prejudices and discriminations against people with disabilities…)
and made up of several episodes. According to Chapman and Sawchuck’s categorization
[20], the whole project falls mainly under the practices of “research-for-creation” and
“research-from-creation”.
On one hand, reading and interpreting scientific literature on media culture, game
studies, social psychology, sociology, and cultural studies allows us to conceptualize
a part of the game. Since most of the fictional situations in the game are based on
facts taken from research papers, it could be said that the project also falls under the
category that Chapman and Sawchuck call “creative presentation of research”. However,
individually, each topic is not our area of scientific expertise, so we don’t define this
feature as fundamentally part of our work.
On the other hand, the game is created in order to generate data that can be analyzed.
Indeed, we designed the game to test several hypotheses regarding the narrativity and
the expressive potential of video games, such as the power of seriality to encourage
dialogue between the players. When finished, at least partially, the production can thus
be analyzed in different ways (formal analysis, user experience…) to establish tentative
results. Indeed, our work is an iterative process in which a phase of scientific discovery
leads to a time of creation which, in return, introduces a stage of theorization.
As of June 2021, the first module of the game, focusing on sexism in the business
world, was developed in a bit more than a year and is published. It was designed to
enable the player’s expression of themself in several ways. In the following paragraphs,
I will focus on the main features: each episode basically consists of four parts.
Playing Charlie, a young woman recently hired, the user is first invited to explore a 3D
environment depicting an open office space. They can choose to listen to rumors which
are spread all around. Some “barks” [21] are sexist comments which pollute the shared
space; others are only quick pictures of everyday life. Similar to a kind of wandering
[22] this type of play is completely optional and embraces the necessary random aspect
of learning [23]. The player can choose how many complex and potentially disturbing
situations they want to face.
Each exploration phase is followed by a phase of mandatory dialogue depicted in 2D
in which the player can learn about the life of Charlie’s colleagues, the challenges she
faces, and the adversity of corporate life. Using curiosity, surprise, and suspense effects
[14] as well as techniques particular to branching narratives (such as narrative puzzles
or hidden paths), those moments aim to immerse the player in the fictional world by
creating a narrative tension. If the user doesn’t have the choice to endure the situation
(be it a neutral, mundane, comment or a sexist microaggression), they can decide how
they handle their response. Based on the two main behaviours identified by the research
in psychology about sexism [24] and inspired by the Jakobson’s functions of language
[25], the dialogue options allow the player to either confront the Non-Player Characters
(passionately or rationally) or dodge their observation (being vague or using humor).
Since there is no right or wrong way to answer, but only different consequences that
depend on the interlocutor’s personality and occupation, this feature allows the player
to freely express themself and to experiment their opinions.
If the narrative structure makes room for the player’s voice, their avatar also has her
own views. The text of every interface is diegetic: for example, the beginning and the
ending menu are written like a personal journal in which Charlie sums up what hap-
pened to her but also expresses her feelings toward her colleagues’ sexist comments.
These moments invite the player to discover another interior life and assume that empa-
thy can lead to a growing awareness of someone else’s problems [26, 27]. Since the
player-character is not only an interface allowing fictional immersion, but also a tool
encouraging a reflexive attitude [28], we decided to harness the potential of this ambiva-
lence to design moments of conflicting point of views. In the exploration or the dialogue
phases, the player can push a red button each time they think the character’s comment
is sexist. The player is thus invited to adopt an analytical thinking. If their reasoning
matches with Charlie’s, her interpretation of the situation is unlocked in the ending
screen. The player is encouraged to express their opinion, but they also face another
consideration of the interpersonal situation.
The game expects the player to be alternatively immersed in the story and critical
of its implications, but it also provides an emotional outlet. In each episode, there is a
turn-based combat phase in which the player has to perform different QTEs (Quick Time
Events), such as pushing frantically and repeatedly on one button or, on the contrary, aim
at the right time in the right place. On one hand, this feature corresponds to the funda-
mental autotelic and non-productive characteristics of games [29] since it doesn’t help
learning about sexism and reconnects with the idea of play as fun. However, if the player
wants to invest themself in an interpretation, this part of the game is also meaningful on
several levels. Symbolically, the combat phase corresponds to the psychological fight
Charlie endures when she can’t take it anymore and thus is very different aesthetically.
The combat stances coincide with the dialogue options, and there are correspondences
between the player’s movement and the meaning of the chosen attitude. For example,
Encouraging Self-expression and Debate in RecovR 413
the passionate and confronting answer is linked with the successive hits. This feature
encourages the player to verbally express their feelings, be it as a mumble for themself
or an exclamation of triumph, when playing in a group.
The game was created to be played in two main contexts: either alone as a self-
examination experience or with other people as a shared experience. However, in each
case, the episodic format tends to stimulate the dialogue after the moments of play,
since the users ponder upon the past fictional events and imagine what will happen next
[30]. The seriality of the project thus aims to create an environment for the players to
collectively talk about sexism and to give them the opportunity to gather around this
social issue to find suitable and sustainable solutions in their interpersonal context.
RecovR confronts the player to various beliefs and behaviors, the NPCs preju-
dices, their avatar’s views, and their own judgment. The game also urges the player
to alternate between several opposite attitudes: immersion and reflexivity, empathy and
self-expression, cathartic fun and analytical gaze… The ideal experience unsettles the
player, who is constantly invited to interpret the game, to give their own perception and
understanding of the fictional events.
Our current hypothesis is the following: To create a deep and sustainable impact on
social issues such as sexism and ableism, games need to be catalysts for human relations,
they need to let the players express their own feelings and thoughts on multifactorial and
sensitive issues. For a game to foster exchange between diverse points of view, a solution
seems to be to not try to impose a unique scope, but rather to create an experience based
on tensions between diametrically opposed poles. Rather than focusing on designing
harmony and consensus, we may seek to draft a chaotic structure and trust the player to
take into their own hands the process to make sense of it.
This line of thought is close to the vibrant field of contemporary research on narrative
structures which leave interpretative room for the players. More particularly, it is inspired
by Rémi Cayatte’s work [31] in which, focusing on procedural narratives, he argues that
the expressive power of video games resides precisely in the moments when the player
takes in charge the operating of the content of the game to inject meaning in the gaps in
the ludo-narrative structure.
The preliminary conclusions obtained thanks to the conception of the first module
of RecovR and the subsequent formal analysis of the created object form a knowledge
which we need to question and build upon. Firstly, the complementary ideas of a balanced
experience between converse elements and a breeding ground for debate will be tested in
another module: the second part of RecovR will tackle the discriminations linked with the
experience of disability and focus the discussion on the issue of ableism. Secondly, we
have set up a series of user tests, in which a close observation of play will be followed by
semi-directive interviews in order to determine if the theoretical hypotheses are verified
in concrete uses. Thirdly, by communicating provisional results [32], we hope to enrich
our understanding of the link between interpersonal communication and sustainable
social impact through video gaming practices thanks to the feedback of the scientific
and professional communities.
414 H. Sellier
References
1. Sharp, J.: Works of Game: On the Aesthetics of Games and Art. MIT Press, Cambridge (2015)
2. Sharp, J.: Independent game. In: Lowood, H., Guins, R. (eds.) Debuggin Game History: A
Critical Lexicon. MIT Press, Cambridge (2016)
3. Nicoll, B., Keogh, B.: The Unity Game Engine and the Circuits of Cultural Software. Palgrave
Macmillan, London (2019)
4. Anthropy, A.: Rise of the Videogame Zinesters: How Freaks, Normals, Amateurs, Artists,
Dreamers, Drop-outs, Queers, Housewives, and People Like You Are Taking Back an Art
Form. Seven Stories Press, New York (2012)
5. Alvarez, J., Djaouti, D.: Introduction au Serious Game. Questions théoriques, Paris (2012)
6. Ruberg, B., Shaw, A. (eds.): Queer Game Studies. University of Minnesota Press, Minneapolis
(2017)
7. Bogost, I.: Persuasive Games. The MIT Press, Cambridge (2007)
8. Séraphine, F.: Ludonarrative dissonance: Is storytelling about reaching harmony? (2016).
http://www.fredericseraphine.com
9. Freedman, E.: No Turning Back: The History of Feminism and the Future of Women.
Ballantine Books, New York City (2003)
10. Goodley, D.: Disability Studies An Interdisciplinary Introduction. Sage Publications Limited,
Thousand Oaks (2011)
11. Bakhtin, M.M.: Problems of Dostoevsky’s Poetics. University of Minnesota Press, Minneapo-
lis (1984)
12. Heraclite: Fragments. Seuil, Paris (1998)
13. Eco, U.: Lector in Fabula. Editions Grasset et Fasquelle, Paris (1979)
14. Baroni, R.: La Tension Narrative. Seuil, Paris (2007)
15. Sicart, M.: Against procedurality. Game Stud. 11(3) (2011)
16. Genvo, S.: Comprendre et développer le potentiel expressif. Hermès, La Revue 62(1) (2012)
17. Posey, J.: Narrative design. In: Despain, W. (ed.) Professional Techniques for Video Game
Writing. AK Peters/CRC Press, Natik (2008)
18. Gosselin, P.: La recherche en pratique artistique. In: Le Cogiec, E., Gosselin, P. (eds.) La
recherche création. Pour une compréhension de la recherche en pratique artistique. Presses
de l’Université du Québec, Québec (2006)
19. Paquin, L.-C., Noury, C.: Définir la recherche-création ou cartographier ses pratiques? In:
Association francophone pour le savoir (2018)
20. Chapman, O., Sawchuck, K.: Research-creation: intervention, analysis and “Family Resem-
blances”. Canadian J. Commun. 37 (2012)
21. Beaulieu, S.: How a character says hello: Writing Barks for video games (2020). https://sarah-
beaulieu.com
22. Flawinne, V.: L’œil vidéoludique et le nouveau flâneur. Mémoire de master, Université de
Liège (2007)
23. Lavigne, M.: Les instrumentalisations du jeu numérique. Une approche critique de l’emploi
contemporain de la notion de jeu au travers de ses manifestations numériques à vocation non
distractives. Mémoire d’habilitation à diriger les recherches, Université de Limoges (2020)
24. Melotte, P.: Les femmes face au sexisme: confrontation ou évitement? Étude des proces-
sus émotionnels et cognitifs lors d’interactions comprenant une remarque sexiste. Thèse de
doctorat, Université Libre de Bruxelles (2017)
25. Jakobson, R.: Essais de linguistique générale. Éditions de Minuit, Paris (1963)
26. Zanna, O.: Le corps dans la relation aux autres. Pour une éducation à l’empathie. Presses
universitaires de Rennes, Rennes (2015)
Encouraging Self-expression and Debate in RecovR 415
27. Immordino-Yang, M.H.: Emotion, sociality, and the brain’s default mode network: insights
for educational practice and policy. Policy Insights Behav. Brain Sci. 3(2) (2016)
28. Barnabé, F., Delbouille, J.: Aux frontières de la fiction: l’avatar comme opérateur de réflexivité.
Sciences du jeu 9 (2018)
29. Huizinga, J.: Homo Ludens: essai sur la fonction sociale du jeu. Gallimard, Paris (1951)
30. Goudmand, A.: Récits en partage. Expériences de la sérialité narrative en culture médiatique.
Thèse de doctorat, Université Paris Sciences et Lettres (2018)
31. Cayatte, R.: L’émotion comme moteur de jeu: mettre des mots sur une expressivité
vidéoludique. Conference Les langages du jeu vidéo: codes, discours et images en jeu,
Université de Lausanne (2019)
32. Bouchardon, S.: La Valeur heuristique de la littérature numérique. Mémoire d’habilitation à
diriger des recherches, Université de Technologie de Compiègne (2012)
Supporting Interactive Storytelling
with Block-Based Narrative Programming
{ahubbar,cringst}@wested.org
Abstract. Recent years have seen growing interest in utilizing digital storytelling,
where students create short narratives around a topic, as a means of creating moti-
vating problem-solving activities in K-12 education. At the same time, there is
increasing awareness of the need to engage students as young as elementary school
in complex topics such as physical science and computational thinking. Building
on previous research investigating block-based programming activities for story-
telling, we present an approach to block-based programming for interactive digital
storytelling to engage upper elementary students (ages 9 to 10) in computational
thinking and narrative skill development. We describe both the learning environ-
ment that combines block-based narrative programming with a rich, interactive
visualization engine designed to produce animations of student generated stories,
as well as an analysis of students using the system to create narratives. Student gen-
erated stories are evaluated from both a story quality perspective as well as from
their ability to communicate and demonstrate computational thinking and physical
science concepts and practices. We also explore student behaviors during the story
creation process and discuss potential improvements for future interventions.
1 Introduction
Recent years have seen growing evidence that engaging students in digital storytelling
activities is an effective tool for promoting meaningful learning across a variety of
subjects [1–3]. While these activities have been shown to facilitate positive outcomes
such as creative exploration of scientific phenomena [4, 5], studies have also shown
that the benefits of digital storytelling can vary greatly between individual students. This
highlights a critical need for a better understanding of how to effectively support students
as they create and present stories.
Another barrier to large-scale adoption of digital storytelling activities in classrooms
is lack of perceived alignment with existing curricular goals, specifically in areas such as
computational thinking and physical science [6]. At the elementary level, many teachers
have limited, if any, instructional time allocated to these topics, and may see digital
storytelling activities as an inefficient use of that limited resource. However, there is a
growing recognition of the similarities between the digital storytelling process and com-
putational thinking and science competencies [7, 8]. Additionally, teachers who often
lack significant training in science and computational thinking may be more comfortable
supporting a more familiar task such as narrative construction.
To address these challenges, we have developed the InfuseCS narrative-centered
learning environment, to engage students in deep, meaningful physical science and com-
putational thinking learning through the creation of interactive narratives. InfuseCS
utilizes a custom-built narrative programming environment, where students utilize a
block-based programming interface to create, revise, and visualize interactive narratives.
Building on research exploring design best practices for block-based programming for
younger learners [9, 10], we created a learning environment to enable upper elemen-
tary students o create rich interactive narratives while also engaging with concepts and
competencies from physical science and computational thinking. In this paper, we use
InfuseCS to specifically focus on two research questions:
RQ1: How effectively are students able to use block-based programming to create
interactive narratives, when evaluated from a story quality perspective?
RQ2: How do students exhibit knowledge of physical science and computational
thinking concepts through their created interactive narratives?
To answer these research questions, we conducted a pilot study with students in the
southeastern United States as part of an after school program. Initial results show students
were able to effectively use the tool to create stories, while also demonstrating evidence of
computational thinking concepts such as debugging and sequential execution. However,
not all student generated stories met all story quality criteria, and all students struggled
to integrated physical science concepts into their narratives.
2 Related Work
Narrative experiences offer an exceptionally promising tool for engaging students in
computationally-rich problem solving. By leveraging narrative’s unique ability to help
us understand the world around us [11] and communicate conceptual understanding to
others [12], storytelling has significant potential well beyond the traditional educational
context of language arts. Specifically, digital storytelling has shown great potential to
leverage the creativity and effectiveness of narrative for domains such as science [1, 13]
through activities such as creating a multimedia presentation of a story. Other digital
storytelling activities have shown positive effects for both cognitive measures of visual
memory and writing skills [14], affective measures such as student engagement [2], and
improved 21st century skills such as problem solving, argumentation, and cooperation
[15].
However, the benefits of digital storytelling interventions are dependent on the abil-
ity of the student to construct a narrative. This has led to a broad range of research
418 A. Smith et al.
into how to best design and support digital storytelling interventions for students of a
variety of age ranges, including approaches that seek to leverage the synergies between
storytelling and computational thinking through block-based programming. A modified
version of the popular Scratch environment was used to enable students to create ani-
mated stories involving placing and moving sprites, as well as including audio clips and
other events responsive to user inputs [16]. Other research has investigated using block-
based programming in non-traditional methods such as a tangible, sticker-based block
language used as part of an interactive storybook [17], or as a method for introducing
computational thinking strategies into English language learning [8]. This work extends
these efforts by designing and investigating story quality and demonstration of concep-
tual knowledge using a block-based programming environment focused primarily on
interactive storytelling augmented by rich visualizations.
avoiding event-driven programming. Dialogue and Stage Direction blocks utilize drop
downs for character names and a starter story is provided to the students to encourage
customization and editing rather than a purely generative activity. Finally, the language
on the blocks is designed to limit complicated syntax and vocabulary.
The set of custom blocks are separated into four categories corresponding to the
main features of visual interactive narratives: describing a scene, moving characters in
and out of the scene, character dialogue (including narration), and branching.
The Scene blocks, shown in Fig. 2a, allows students to set the location where the
events in the narrative take place. Location blocks are defined before the activity to fit
with the motivating scenario framing a particular digital storytelling activity, and to align
with art assets for the visualization system described below. This also allows InfuseCS
to utilize type-checking, and prevent students from attaching incompatible blocks, i.e.
text fields, a feature also afforded by the color-coding scheme of blocks.
The Stage Direction blocks allow for students to move characters in and out of the
scene of their story. The Enter block, shown in Fig. 2b, allows students to select which
character is entering the scene, with fixed location blocks that can be attached similar to
the locations for the Scene block. The other Stage Direction block currently implemented
is the Exit block, where students can specify a character to leave the scene.
The Dialogue blocks, shown in Fig. 2c, are the core blocks of the storytelling sys-
tem, as the majority of the content in the student-generated stories is dialogue between
characters or statements by the narrator. A line of dialogue is generated through the
combination of a Character block and a Text block. The name property of the Character
block is filled with a list of characters that have been previously introduced to the student
earlier in the activity. Additionally, a Narrator option can be chosen to provide narration
for the story.
The Ask Audience block, Fig. 2d, is designed to allow students to incorporate branch-
ing into their narratives, while also providing them an opportunity to experiment with
computational thinking concepts like conditionals and flow control. The Ask Audience
block contains three properties that must be defined by learners. The first property is
the question that will be asked to the audience, and then the two possible responses the
audience can respond with to the question. Below each response, students can then place
story blocks that will execute if that choice is selected.
At any point in the story authoring process, students can press a run button at the
top of the editor to play an animation of their story. First, the block-based representation
is converted into Ink script (https://www.inklestudios.com/ink/), a narrative scripting
420 A. Smith et al.
language developed primarily for game applications. Next, the generated ink script is
passed to the visualization engine, which displays it using graphical assets in Unity. The
visualization pauses after each dialogue utterance to give the audience time to read the
story, and advances when a button is pressed. At the completion of the visualization, the
student returns to the narrative editor where they can continue to iterate and revise their
story until they are satisfied.
4 Pilot Study
To understand how InfuseCS supports digital storytelling and computational thinking
in elementary classrooms, a pilot study was conducted at an after-school program in
the southeast United States. Participants in the pilot study included 6 fourth grade stu-
dents ages 9–10, including 4 males and 2 females. Four of the students were Black or
African-American, one was White, and one was Multiracial. The 6 students worked
independently, and students reported a range of previous experience with block-based
coding environments with 1 student reporting no prior experience, 2 students reporting
some prior experience, and 3 students reporting that they frequently used block-based
coding environments.
The pilot study took place over 4 days, with approximately 30–45 min spent each
day. Students and coordinators for the after-school program were together in person,
while research staff observed remotely using video conferencing software. The first
day students were given a brief introduction to the activity, completed a demographic
survey, and watched an introductory video introducing the motivating scenario of a
group of scientists who were shipwrecked on a remote tropical island. On the second
day, students were presented with the task of solving how the scientists would power
their makeshift village and engaged with physical science content focusing on types of
energy and energy conversion. After completing the science content, students were given
a planning worksheet to assist them with planning their story. On the third day, students
completed their planning worksheet and began creating their stories using the narrative
programming interface, and on the fourth day the students spent the entire time period
working on their stories.
For all students, a brief starter story was provided in the block-based programming
interface, which provided an example of each type of block. The learning environment
logged detailed trace logs of student actions in the environment, as well as logging their
story workspace each time the story was visualized.
All 6 students were able to use the system to create short narratives in the allotted time.
Each student effectively used at least one of each type of block with stories ranging from
12–26 “lines of code” (i.e. two connected components of a Dialogue block counting as 1
“line”), consisting mostly of narration and dialogue between characters. Overall student
stories averaged 17.3 total blocks (M = 17.3, s.d. = 5.64), 4 scene blocks per story (M
= 4, s.d. = 1.78), 4.2 Stage Direction blocks per story (M = 4.2, s.d. = 1.94), 1.8 Ask
Audience blocks (M = 1.8, s.d. = 0.98), and 7.3 Dialogue blocks (M = 7.3, s.d. = 1.97).
Additionally, students used an average of 3.33 characters in their story. This represents
a large increase compared to previous versions of the system containing a similar set of
story blocks but lacking the visualization functionality of the current system [21].
Supporting Interactive Storytelling with Block-Based Narrative Programming 421
We also evaluated the student stories based on a set of story quality metrics, as well as
for evidence of curricular competencies from both physical science and computational
thinking. We first investigate the research question:
RQ1: How effectively are students able to use block-based programming to create
interactive narratives, when evaluated from a story quality perspective?
To measure story quality, we devised a scoring rubric to evaluate the students’ sto-
ries based on the Common Core State Standards (http://www.corestandards.org/) for
4th grade English Language Arts Writing. Common Core State Standards are a set of
standards for various subjects such as writing and mathematics that have been adopted
by many states in the United States. Students’ stories were evaluated based on six cri-
teria in alignment with the standards and one additional criterion we added specifically
focusing on how branching was utilized in the story to facilitate its organization and
development. These criteria are described in Table 1 below. Each criterion was rated on
a 0–2 scale, with 0 as having little to no evidence, 1 as having some evidence, and 2 as
having sufficient evidence of the given criteria.
Criteria Description
Criteria 1 Establishes a situation that introduces reader to characters/setting
Criteria 2 Presents/establishes an organized event that unfolds naturally (including a distinct
beginning, middle, and end)
Criteria 3 Includes dialogue and description that develops experiences and events or shows
the responses of characters to situations
Criteria 4 Uses a variety of transitional words and phrases to manage the sequence of events
Criteria 5 Uses concrete words and phrases and sensory details to convey experiences and
events precisely
Criteria 6 Provide a conclusion that follows from the narrated experiences or events
Criteria 7 Integrates branching to organize and develop the story
Analysis of student narratives using these criteria is shown below in Table 2. Overall,
the custom blocks facilitated Criteria 1, 2, and 3 for all students. Student stories performed
worst on Criteria 5 and 6. Summing across all criteria yielded an average of 6.67 per
story, with a high score of 12 and a low score of 4.
RQ2: How do students exhibit knowledge of physical science and computational
thinking concepts through their created interactive narratives?
To evaluate physical science content, we used frequency counts to measure how many
science concepts from the physical content were referenced in their stories. Overall, the
number of science concepts students utilized in their stories ranged from 0 to 3 with four
of the students not integrating any science concepts in their stories.
422 A. Smith et al.
For computational thinking we looked at both the final story, as well as behaviors
exhibited while creating the story. Using the K-12 Computer Science Framework (https://
k12cs.org/) we identified concepts and practices that overlap with the story creation
process. In addition to the concept of Creating Computational Artifacts, students also
exhibited evidence of troubleshooting/debugging behaviors through the iterative running
and modifying of their stories as they were created. Overall, the 6 students averaged 42
runs of their story (minimum = 12, maximum = 61). Students showed evidence of
understanding control structures and sequential execution of their stories through their
usage of the Ask Audience block. Multiple students stored segments of code on the
workspace not connected to the Start block, showing evidence of their understanding of
the single thread of execution.
5 Discussion
Overall, the InfuseCS learning environment enabled students to create interactive narra-
tives incorporating features such as dialog, setting changes, and branching. Compared to
a previous version of the system, the new system with a revised set of custom blocks and
a story visualization system appears to have effectively supported longer, higher quality
stories, as well as encouraged desirable debugging and troubleshooting behaviors. This
was particularly noticeable in the students’ improved understanding and usage of the
Ask Audience block to incorporate branching into their narratives.
A noticeable shortcoming was the lack of physical science content included in the sto-
ries. Discussions with students suggested adding props representing the science content
to the block palette may encourage their inclusion in the resulting narratives. Similarly,
more explicitly including science concepts in the planning process may also encour-
age their usage in the resulting narratives. From a computational thinking perspective,
further analysis of the students’ debugging and troubleshooting behaviors could help
ensure that the visualization functionality is being used in a productive manner, rather
than a distracting one. Inclusion of more complex command structures, such as loops
or event-based execution, could provide more opportunities for students to demonstrate
richer understanding. Finally, including more traditional assessments (i.e. validated sur-
vey measures) could help determine if lack of understanding, or a shortcoming of the
interface is responsible for the omission of concepts included in the narratives.
Supporting Interactive Storytelling with Block-Based Narrative Programming 423
Finally, there are two key limitations to these findings. First, we acknowledge the
small sample size makes it difficult to generalize the findings too broadly. Secondly,
while researchers were able to observe the activities via video conferencing, lack of in
person observations due to the ongoing pandemic makes qualitative analysis of the story
creation process more difficult.
Acknowledgments. This research was supported by the National Science Foundation under
Grants DRL-1921495 and DRL-1921503. Any opinions, findings, and conclusions expressed
in this material are those of the authors and do not necessarily reflect the views of the National
Science Foundation.
References
1. Robin, B.R.: Digital storytelling: a powerful technology tool for the 21st century classroom.
Theory Into Pract. 47, 220–228 (2008). https://doi.org/10.1080/00405840802153916
2. Smeda, N., Dakich, E., Sharda, N.: The effectiveness of digital storytelling in the classrooms:
a comprehensive study. Smart Learn. Environ. 1(1), 1–21 (2014). https://doi.org/10.1186/s40
561-014-0006-3
3. Yang, Y.T.C., Wu, W.C.I.: Digital storytelling for enhancing student academic achievement,
critical thinking, and learning motivation: a year-long experimental study. Comput. Educ. 59,
339–352 (2012). https://doi.org/10.1016/j.compedu.2011.12.012
4. Henriksen, D.: Full STEAM ahead: creativity in excellent STEM teaching practices. STEAM
J. 1, 1–9 (2014). https://doi.org/10.5642/steam.20140102.15
424 A. Smith et al.
5. Henriksen, D., Mishra, P., Fisser, P.: Infusing creativity and technology in 21st century
education: a systemic view for change. Educ. Technol. Soc. 19, 27–37 (2016)
6. Tan, M., Lee, S.-S., Hung, D.W.L.: Digital storytelling and the nature of knowledge. Educ.
Inf. Technol. 19(3), 623–635 (2013). https://doi.org/10.1007/s10639-013-9280-x
7. Lee, I., et al.: Computational Thinking for Youth in Practice (2011)
8. Parsazadeh, N., Cheng, P.Y., Wu, T.T., Huang, Y.M.: Integrating computational thinking
concept into digital storytelling to improve learners’ motivation and performance. J. Educ.
Comput. Res. 59, 470–495 (2021). https://doi.org/10.1177/0735633120967315
9. Hill, C., Dwyer, H.A., Martinez, T., Harlow, D., Franklin, D.: Floors and flexibility: design-
ing a programming environment for 4th-6th grade classrooms. In: Proceedings of the 46th
ACM Technical Symposium on Computer Science Education (SIGCSE), pp. 546–551 (2015).
https://doi.org/10.1145/2676723.2677275
10. Weintrop, D., Hansen, A.K., Harlow, D.B., Franklin, D.: Starting from scratch: outcomes of
early computer science learning experiences implications for what comes next. In: Proceed-
ings of the 2018 ACM Conference on International Computing Education Research (ICER),
pp. 142–150 (2018). https://doi.org/10.1145/3230977.3230988
11. Bruner, J.: Acts of Meaning. Harvard University Press, Cambridge (1990)
12. Avraamidou, L., Osborne, J.: The role of narrative in communicating science. Int. J. Sci. Educ.
31, 1683–1707 (2009). https://doi.org/10.1080/09500690802380695
13. Robin, B.R.: The power of digital storytelling to support teaching and learning. Digit. Educ.
Rev. 17–29 (2016). https://doi.org/10.1344/der.2016.30.17-29
14. Sarica, H.Ç., Usluel, Y.K.: The effect of digital storytelling on visual memory and writing
skills. Comput. Educ. 94, 298–309 (2016). https://doi.org/10.1016/j.compedu.2015.11.016
15. Niemi, H., Multisilta, J.: Digital storytelling promoting twenty-first century skills and stu-
dent engagement. Technol. Pedagog. Educ. 25, 451–468 (2016). https://doi.org/10.1080/147
5939X.2015.1074610
16. Franklin, D., et al.: Using upper-elementary student performance to understand conceptual
sequencing in a blocks-based curriculum. In: Proceedings of the Conference on Integrating
Technology into Computer Science Education, ITiCSE, pp. 231–236 (2017). https://doi.org/
10.1145/3017680.3017760
17. Horn, M.S., AlSulaiman, S., Koh, J.: Translating Roberto to Omar: computational literacy,
stickerbooks, and cultural forms. In: Proceedings of the 12th International Conference on
Interaction Design and Children, pp. 120–127 (2013). https://doi.org/10.1145/2485760.248
5773
18. Franklin, D., Hill, C., Dwyer, H.A., Hansen, A.K., Iveland, A., Harlow, D.B.: Initialization in
scratch: seeking knowledge transfer. In: Proceedings of the 47th ACM Technical Symposium
on Computing Science Education, SIGCSE 2016, pp. 217–222 (2016). https://doi.org/10.
1145/2839509.2844569
19. Weintrop, D., Wilensky, U.: Using commutative assessments to compare conceptual under-
standing in blocks-based and text-based programs. In: Proceedings of the 11th Annual Interna-
tional Conference on International Computing Education Research (ICER 2015), pp. 101–110
(2015). https://doi.org/10.1145/2787622.2787721
20. Dwyer, H., Hill, C., Hansen, A., Iveland, A., Franklin, D., Harlow, D.: Fourth grade students
reading block-based programs: predictions, visual cues, and affordances. In: Proceedings of
the 2015 ACM Conference on International Computing Education Research (ICER), pp. 111–
120 (2015). https://doi.org/10.1145/2787622.2787729
21. Smith, A., et al.: Toward a block-based programming approach to interactive storytelling for
upper elementary students. In: Bosser, A.-G., Millard, D.E., Hargood, C. (eds.) ICIDS 2020.
LNCS, vol. 12497, pp. 111–119. Springer, Cham (2020). https://doi.org/10.1007/978-3-030-
62516-0_10
Climate Influence: Implicit Game-Based
Interactive Storytelling for Climate Action
Purpose
1 Introduction
Despite the mass of scientific evidence on climate change [16], a gap remains between
public awareness [17] and actions that people take. People often see climate change as
distant and impersonal [11, 21], making it difficult to capture the attention and affect the
actions of people using logical arguments, especially climate change skeptics distrustful
of science. Climate change communication requires motivating strategies to cultivate
long-term actions, using storytelling and interactive experiences designed for specific
social goals [10] from a pro-environment mindset [3].
Stories have the potential to make climate change real, so that audiences may have
greater emotional connection [12]. Previous work has shown that collaborative story-
telling provides platforms for behavioral change [8, 10]. However, our previous narrative
designs are Pavlovian [7] in the sense that they equate certain types of goals like future-
looking orientation and individual responsibility with a positive outcome but do not ask
V. Ruijters—Independent Artist.
the reader to take actions interactively. However, players are reinforced or punished in an
operant conditioning context in an interactive environment like games [18], thus directly
promoting certain actions. Games place players in scenarios and prompt them to analyze
situations from different perspectives [20]. These scenarios enable the target audience to
resonate with their physical situations and worldviews and create an emotional relation-
ship with characters in the game [20]. In particular, analyzing risks inherent in climate
change [19] and adapting to future climate change scenarios [13] have utilized serious
games, whereas explicit forms of entertainment involving saving worlds from environ-
mental catastrophe have populated efforts to influence climate action in players [5, 9,
15]. However, these games with explicit climate change goals can alienate those who
don’t align with climate action goals, so a more effective design would involve implicit
forms of storytelling based on psychological goals rather than directly adding climate
change themes into games [1]. To apply these operant interactive strategies that reinforce
pro-climate behaviors, we created a game that uses the Tamagotchi to carry forth the
care-taking metaphor to a climate change issue that is not explicitly told to players.
Fig. 1. (Left) Comic magazine Drizzle as narrative influence for pro-climate action. (Right) The
serious game Chikyuchi promotes climate action by aligning players with pro-climate goal.
2 Designed Intervention
To understand the effect of storytelling on audiences, we designed five stories based on
Booker’s [2] story structure for particular human phenomena that lead to climate change,
including immediate gratification, the idea that individual effort doesn’t matter, myopia,
etc. One story from the collection is New Revolia, which uses the plot of shipwreck on
the Antarctic ocean to narrate the idea that distributing resources with others leads to
self benefits (Fig. 1), a pro-climate action goal. While visual narratives can frame stories
that promote actions consistent with climate action, they cannot reinforce behaviors that
actively lead to pro-climate choices. To create an immersive experience that reinforces
audience behavior to align with climate action goals, we created and exhibited a game
that reinforces the notion of “caring for environment” without explicitly revealing the
concept. The game uses a metaphor to reinforce players who take care of an avatar that
implicitly represents the Amazon forest.
The main avatar in the Chikyuchi game anthropomorphizes a natural resource (trees
of the Amazon) that is declining at the rate of a hectare a minute in reality, leading to
Climate Influence 427
increased surface temperature and reduced rainfall [14]. The Chikyuchi is connected to
the deforestation API, reporting the actual size decreasing 2 km2 with every second in the
statistics menu. Chikyuchi is determined by its health and mood (1 to 4 hearts). Health is
always low due to the situation of deforestation. Although users are able to temporarily
cheer up the Chikyuchis with thematically related food and games, increasing their
mood status, it won’t change their critical health status. Chikyuchi also periodically
chats about each crisis by interrupting the game. It does this by using text generated
with the transformer language model GPT-2 pretrained on tweets containing the words
“deforestation” and “global warming” over a 3 day period (temperature = 0.8 during
generation, 6500 epochs training). The text generated are implicit forms of persuasion
rather than directly lobbying for climate action.
Fig. 2. (Left) Game Device. (Top Right) Feeding and threat interactions. (Middle Right) Info
flow: name, age, health, declining rainforest size. (Bottom Right) The umbrella/sunglasses game
play flow (the player must guess whether the next day will be sunny or raining).
There are three lower buttons (select, execute and cancel) that allow participants to
make in-game decisions (Fig. 2). Participants could interact with the virtual pet through
six functions: food, clean defecation, play game, medicine, information and attention.
Each in-game decision affects Chikyuchi’s mood in the short term but does not improve
its health. For example, playing a guessing game with the weather about whether it
should bring an umbrella (it’ll rain) or sunglasses (it’ll be sunny) can raise its mood
if the guesses are correct. The random weather outcome also implicitly informs us
about climate change. The stats menu gives the current health, mood, age, and size of
Chikyuchi. In summary, Chikyuchi uses the interactive gaming paradigm to promote
care-taking behavior from players in relation to climate change [4].
To allow engagement with Chikyuchi, we created an installation consisting of a live
video-feed for real-time interaction of audiences in Tokyo and Hong Kong, two game
devices, photos that hint at the climate topic, and a rusted bronze cast of Chikyuchi
that represents destroyed natural resources as dead virtual pets. The interaction with
Tamagotchi [6] is shown as a live video feed that brings people together to develop a
common practice of caring across different places in the world.
428 Z. Song et al.
Fig. 3. (Top Left) Exhibition. (Middle Left) Real-time audience interaction. (Bottom Left)
Bronze cast showing Chikyuchi’s decline. (Right) Survey of effectiveness of intervention and of
audience perception. (To see text of survey questions used: sites.google.com/view/chsu/app).
We studied the use of implicit influences in purposive interactive storytelling for cli-
mate communication. Narrative interactive games are proposed to reinforce the mental
connection that alters climate awareness in the long run and take pro-climate actions.
Indeed, evaluation showed audiences understood the caretaking aspect of the game sig-
nificantly more than the actual information about resources conveyed. Future work would
involve evaluating audience perception of comics for climate action to compare between
interactive and non-interactive strategies for climate influence.
References
1. Abraham, B., Jayemanne, D.: Where are all the climate change games? Locating digital
games’ response to climate change. 21 (2017)
Climate Influence 429
2. Booker, C.: The Seven Basic PLOTS: Why We Tell Stories. A&C Black (2004)
3. Boykoff, M.: Creative (Climate) Communications: Productive Pathways for Science, Policy
and Society. Cambridge University Press, Cambridge (2019)
4. Caillois, R., Barash, M.: Man, Play, and Games. University of Illinois Press, Urbana (2001)
5. Chen, Y., Bodicherla, D., Scott, B., Whittinghill, D.: Meltdown: a serious game for envi-
ronmental awareness on climate change, pp. 388–394. Association for the Advancement of
Computing in Education (AACE) (2014)
6. Clyde, A.: Electronic pets. Emerg. Libr. 25(5), 34–36 (1998)
7. Domjan, M.: Pavlovian conditioning: a functional perspective. Annu. Rev. Psychol. 56(1),
179–206 (2005)
8. Farinella, M.: The potential of comics in science communication. J. Sci. Commun. 17(1),
Y01 (2018)
9. Fox, J., McKnight, J., Sun, Y., Maung, D., Crawfis, R.: Using a serious game to communicate
risk and minimize psychological distance regarding environmental pollution. Telemat. Inform.
46, 101320 (2020)
10. Lc, R., Mizuno, D.: Designing for narrative influence: speculative storytelling for social
good in times of public health and climate crises. In: Extended Abstracts of the 2021 CHI
Conference on Human Factors in Computing Systems, pp. 1–13. Association for Computing
Machinery, New York (2021)
11. Leiserowitz, A.: Climate change risk perception and policy preferences: the role of affect,
imagery, and values. Clim. Change 77(1), 45–72 (2006)
12. Moezzi, M., Janda, K.B., Rotmann, S.: Using stories, narratives, and storytelling in energy
and climate change research. Energy Res. Soc. Sci. 31, 1–10 (2017)
13. Neset, T.-S., Andersson, L., Uhrqvist, O., Navarra, C.: Serious gaming for climate adap-
tation—assessing the potential and challenges of a digital serious game for urban climate
adaptation. Sustainability 12(5), 1789 (2020)
14. Nobre, P., Malagutti, M., Urbano, D.F., de Almeida, R.A.F., Giarolla, E.: Amazon defor-
estation and climate change in a coupled model simulation. J. Clim. 22(21), 5686–5697
(2009)
15. Onencan, A., Van de Walle, B., Enserink, B., Chelang’a, J., Kulei, F.: WeShareIt game:
strategic foresight for climate-change induced disaster risk reduction. Procedia Eng. 159,
307–315 (2016)
16. Pachauri, R.K., Allen, M.R., Barros, V.R., et al.: Climate Change 2014: Synthesis Report.
Contribution of Working Groups I, II and III to the Fifth Assessment Report of the
Intergovernmental Panel on Climate Change. IPCC, Geneva, Switzerland (2014)
17. Rhodes, E., Axsen, J., Jaccard, M.: Exploring citizen support for different types of climate
policy. Ecol. Econ. 137, 56–69 (2017)
18. Staddon, J.E.R., Cerutti, D.T.: Operant conditioning. Annu. Rev. Psychol. 54(1), 115–144
(2003)
19. Undorf, S., Tett, S.F.B., Hagg, J., et al.: Understanding interdependent climate change risks
using a serious game. Bull. Am. Meteor. Soc. 101(8), E1279–E1300 (2020)
20. Villar, M.E.: Community engagement and co-creation of strategic health and environmental
communication: collaborative storytelling and game-building. J. Sci. Commun. 20(01), C08
(2021)
21. Weber, E.U.: Experience-based and description-based perceptions of long-term risk: why
global warming does not scare us (Yet). Clim. Change 77(1), 103–120 (2006)
VR for Diversity: Amelia’s Dream
Mirjam Vosmeer(B)
Abstract. This demo is the first outcome of the research project VR for Diversity.
The theoretical backgrounds for the project are shortly discussed and the concept
for Amelia’s Dream is presented. Amelia’s Dream is a VR experience that is
filmed using volumetric video capture technology, in which a young woman shares
some of her dreams and concerns, relating to issues of gender equality. Focusing
on how parasocial and physical interaction may impact the persuasive effects of
VR, the research plan shortly elaborates on how the installation will be used for
experimental studies into the possibilities of VR as a perspective shifter.
1 Introduction
The VR experience that is demonstrated here is the first outcome of the VR for Diversity
research project. The goal of VR for Diversity is to investigate the ‘perspective shifting’
affordances of VR by testing whether an interactive VR installation can have a positive
impact on users’ knowledge and attitude towards a topic. Our ambitions with the VR
production Amelia’s Dream are threefold: 1. it will be used for research into the effects of
interactive virtual reality, 2. it enables the producers to experiment with the possibilities
of volumetric video capture technology, 3. it will be presented to a general public at
demo sessions and festivals, to inform them about the topic of gender equality. With this
research project, we continuate our investigations into interactive cinema and storytelling
for VR [17–23].
Volumetric video capture technology enables the producer to experiment with the size
of the character in relation to the user: while Amelia will sometimes appear to be of
‘normal’ size, she will also be featured in a much larger scale as her face appears behind
a window, or much smaller when she performs as a miniature ballerina.
3 VR as a Perspective Shifter
Since the re-introduction of VR in 2014 there has been considerable attention to the
possible positive effects of the medium, with many initiatives investigating the ways VR
may be used to contribute to a better world. Big companies founded projects such as VR
for Impact [24] and VR for Good [25] in which all kinds of installations were presented
that explored the possibilities of VR to change attitudes or behavior among the audience.
By 2021, a whole range of VR productions have been presented, that according to the
Digital Catapult report on Immersive Content Formats for Future Audiences [1], can be
included in the category ‘VR as Perspective Shifter’. In most productions that fall into
the category the user enters someone else’s life circumstances, either via the simulation
of inhabiting their body or by ‘meeting’ them. The result the creators aim for in their
audience is a shift in perspective. This could be an invitation to reflect on a particular
theme, contributing to a shift in world view, an increase in understanding, the fostering
of empathy for a particular group of people, or simply a change in opinion or better
understanding of an issue.
However, after analyzing a representative sample of 150 non-fiction VR titles
released between 2012 and 2018, Bevan et al. [2] stated that the degree to which VR is
capable of evoking emotional engagement and how audience members respond to con-
tent is far from clear. The authors claim that the biggest current challenge for VR content
producers therefore is to recognize that providing an immersive 360° environment may
in itself not be sufficient to make a viewer feel present in a story. The Digital Catapult
report seems to underwrite this insight. It describes how in most of the installations that
aim to work as a perspective shifter, the user rarely has any agency beyond looking
around.
4 Interaction in VR
The early promise of VR was that it would allow audiences to experience and connect
to reality in ways beyond what could be achieved with traditional two-dimensional film.
In this vision, audiences would no longer be limited to being passive observers of the
story; they could be embodied and feel present within the story world, potentially taking
a much more active, interactive role. Bevan et al. [2] state that perhaps the greatest
surprise of their exploration of non-fiction VR was that the amount of content that
attempts to directly fulfill this ambition is actually very small. They conclude that in
their sample the role of the viewer in VR was still mainly passive and observant, instead
of actively participatory. Rose [12] has also pointed out that the potential for embodied
interaction has not yet been widely adopted for non-fiction VR. She therefore claims that
the possibility of VR as a medium that would enable the user to ‘do something’ within
432 M. Vosmeer
the mediated world, instead of simply watching a troubling situation, could be exactly
the distinctive feature that would allow us to think of VR as an experiential medium.
The term parasocial interaction is also defined as the ‘pseudointerpersonal interac-
tion’ that occurs between mediated performers and their audiences [5, 7]. Mateer points
out that the way viewers connect to characters in VR can be defined as social presence
[10]. Cummings and Bailenson [3] state that the relationship between the immersive
quality of a mediated environment and the level of (social) presence experienced by
the user is often predicated on the assumption that greater system immersion causes
greater user presence, which, in turn, enhances the applied effectiveness of the mediated
environment. In their meta-analysis, they point out that immersive features such as the
possibility to take action may be an important factor for users to perceive themselves
as being located within a virtual space. In an overview of the literature around design
for user engagement in virtual environments Sutcliffe points out that interactive features
such as sliders, responsive objects and pop-up features allow the user to explore and
control the virtual world and to become more “present” or immersed [16]. Hudson et al.
[9] showed that immersion mediates the effects of interaction between users and the
virtual environments with satisfaction and loyalty, supporting the notion that designing
virtual environments with interactive features will lead to more immersive experiences.
Based upon these insights, we hypothesize that physical and parasocial interaction can
be expected to heighten the ‘effectiveness’ of a virtual environment.
5 Research Plan
With this installation, we aim to study a number of different concepts. During the exper-
iments, the VR experience will be presented in two different versions: version A will be
interactive, requiring the user to perform little tasks to move the narrative forward, such
as opening a box to see the miniature ballerina, starting a sequence by pushing a button, or
moving objects to facilitate the protagonist. In version B, the scenes will progress in the
exact same way, but without user interaction. In this A-B test set up we intend to evaluate
the user experience on the following dimensions, using scales that are based on the work
of Roth and Koenitz [13]: social presence; spatial presence, agency/effectance, autonomy
(perceived action possibilities), curiosity, enjoyment, satisfaction, (cognitive/emotional),
empathy, affect and character believability. By comparing the outcomes for A and B, we
hope to determine the impact of physical interaction in VR.
A second aim for our research is to verify the impact of eye gaze and attention:
in some fragments the protagonist will look straight into the camera while sharing her
stories, while in other fragments she will look in a different direction, for instance at
her own reflection in a mirror. With this part of our research, we follow up on the work
of Hartmann and Goldhoorn on the viewers’ experience of parasocial interaction [8]
and explorations into eye-gaze-based interaction for immersive virtual reality by [11].
The attention of the user will be operationalized by measuring head movement, which
will enable us to determine whether mutual eye-gaze and shared attention leads to better
story recall.
The third aim for our research is related to the persuasive impact of interactive VR.
In their study into VR, presence and attitude change, Tussyadiah et al. [15] substantiate
VR for Diversity: Amelia’s Dream 433
the persuasive role of the medium, suggesting that the subjective experience of presence
in VR can translate into real world attitude and induce behavioral change. With previous
explorations into the persuasive effects of VR in the context of marketing and advertising
in mind [14, 26], we intend to establish the user’s knowledge and attitude towards issues
of gender equality, before and after experiencing Amelia’s Dream, to verify whether an
interactive VR experience may have a persuasive effect on its users. For this goal we
intend to implement items from the questionnaire as presented by Faddoul and Chatterjee
[4].
Acknowledgements. The research project VR for Diversity is led by the Amsterdam University
of Applied Sciences, runs from October 2020 until October 2022 and is funded by SIA RAAK. The
consortium consists of: Amsterdam University of Applied Sciences, Utrecht University of the Arts,
Vrije Universiteit, WeMakeVR, IJsfontein, &Samhoud media, UC 360, The Virtual Dutchmen,
Submarine Channel, VR Days Europe, Bostheater, VR Amsterdam, Interactieve Producenten
Nederland, Federatie Dutch Creative Industries and ARDIN.
References
1. Allen, C., Tucker, D.: Immersive content formats for future audiences. Digital Catapult (2019)
2. Bevan, C., et al.: Behind the curtain of the “Ultimate Empathy Machine”: on the composition
of virtual reality nonfiction experiences. In: Proceedings of the 2019 CHI Conference on
Human Factors in Computing Systems (506) (2019)
3. Cummings, J.J., Bailenson, J.N.: How immersive is enough? A meta-analysis of the effect of
immersive technology on user presence. Media Psychol. 19(2), 272–309 (2016)
4. Faddoul, G., Chatterjee, S.: A quantitative measurement model for persuasive technologies
using storytelling via a virtual narrator. Int. J. Hum.–Comput. Interact. 36(17) (2020)
5. Giles, D.: Parasocial interaction: a review of the literature and a model for future research.
Media Psychol. 4, 279–305 (2002)
6. Hackl, C.: What is volumetric video and why it matters to the enterprise (2020). https://
www.forbes.com/sites/cathyhackl/2020/09/27/what-is-volumetric-video--why-it-matters-to-
the-enterprise/. Accessed 14 Sept 2021
7. Hartmann, T.: Parasocial interactions and paracommunication with new media characters.
Mediated Interpersonal Commun. 177, 199 (2008)
8. Hartmann, T., Goldhoorn, C.: Horton and Wohl revisited: exploring viewers’ experience of
parasocial interaction. J. Commun. 61(6), 1104–1121 (2011)
9. Hudson, S., Matson-Barkat, S., Pallamin, N., Jegou, G.: With or without you? Interaction and
immersion in a virtual reality experience. J. Bus. Res. 100, 459–468 (2019)
10. Mateer, J.: Directing for Cinematic Virtual Reality: how the traditional film director’s craft
applies to immersive environments and notions of presence. J. Media Pract. 18(1), 14–25
(2017)
11. Piumsomboon, T., Lee, G., Lindeman, R.W., Billinghurst, M.: Exploring natural eye-gaze-
based interaction for immersive virtual reality. In: 2017 IEEE Symposium on 3D User
Interfaces (3DUI), pp. 36–39. IEEE (2017)
12. Rose, M.: The immersive turn: hype and hope in the emergence of virtual reality as a nonfiction
platform. Stud. Documentary Film 12(2), 132–149 (2018)
13. Roth, C., Koenitz, H.: Evaluating the user experience of interactive digital narrative. In:
Proceedings of the 1st International Workshop on Multimedia Alternate Realities, pp. 31–36
(2016)
434 M. Vosmeer
14. Tussyadiah, I.P., Wang, D., Jia, C.H.: Exploring the persuasive power of virtual reality imagery
for destination marketing (2016)
15. Tussyadiah, I.P., Wang, D., Jung, T.H., tom Dieck, M.C.: Virtual reality, presence, and attitude
change: empirical evidence from tourism. Tourism Manag. 66, 140–154 (2018)
16. Sutcliffe, A.: Designing for user experience and engagement. In: O’Brien, H., Cairns, P. (eds.)
Why Engagement Matters, pp. 105–126. Springer, Cham (2016). https://doi.org/10.1007/978-
3-319-27446-1_5
17. Verouden, N., Vosmeer, M., Sandovar, A.: Sleep tight johnny idaho a multicultural explo-
ration into virtual reality. In: Rouse, R., Koenitz, H., Haahr, M. (eds.) Interactive Storytelling:
11th International Conference on Interactive Digital Storytelling, ICIDS 2018, pp. 356–358.
Springer, Cham (2018). https://doi.org/10.1007/978-3-030-04028-4_41
18. Vosmeer, M., Sandovar, A., Schouten, B.: From literary novel to radio drama to VR project:
the thousand autumns of Jacob de Zoet. In: Rouse, R., Koenitz, H., Haahr, M. (eds.) Interactive
Storytelling: 11th International Conference on Interactive Digital Storytelling, ICIDS 2018,
pp. 392–400. Springer, Cham (2018). https://doi.org/10.1007/978-3-030-04028-4_46
19. Vosmeer, M., Sandovar, A.: Circus noel: a case study into interaction and interface design
for cinematic VR. In: Clua, E., Roque, L., Lugmayr, A., Tuomi, P. (eds.) Entertainment
Computing – ICEC 2018, pp. 223–227. Springer, Cham (2018). https://doi.org/10.1007/978-
3-319-99426-0_21
20. Vosmeer, M., Roth, C., Koenitz, H.: Who are you? Voice-over perspective in surround video.
In: Nunes, N., Oakley, I., Nisi, V. (eds.) ICIDS 2017. LNCS, vol. 10690, pp. 221–232. Springer,
Cham (2017). https://doi.org/10.1007/978-3-319-71027-3_18
21. Vosmeer, M., Schouten, B.: Project orpheus: a research study into 360° cinematic VR. In:
Proceedings of the 2017 ACM International Conference on Interactive Experiences for TV
and Online Video 2017, pp. 85–90. ACM (2017)
22. Vosmeer, M., Roth, C., Schouten, B.: Interaction in surround video: the effect of auditory
feedback on enjoyment. In: Schoenau-Fog, H., Bruni, L.E., Louchart, S., Baceviciute, S.
(eds.) ICIDS 2015. LNCS, vol. 9445, pp. 202–210. Springer, Cham (2015). https://doi.org/
10.1007/978-3-319-27036-4_19
23. Vosmeer, M., Schouten, B.: Interactive cinema: engagement and interaction. In: Mitchell, A.,
Fernández-Vara, C., Thue, D. (eds.) ICIDS 2014. LNCS, vol. 8832, pp. 140–147. Springer,
Cham (2014). https://doi.org/10.1007/978-3-319-12337-0_14
24. VR for Impact. https://vr4impact.com/. Accessed 26 July 2021
25. VR for Good. https://www.oculus.com/vr-for-good/. Accessed 26 July 2021
26. Wu, D., Lin, J.: Ways of seeing matter: the impact of a naturally mapped perceptual system
on the persuasive effects of immersive virtual reality advertising. Commun. Res. Rep. 35(5),
434–444 (2018)
Exploring Narrative Novelties in VR
christian.roth@hku.nl
1 Introduction
It was not that long ago that the most recent revival of virtual reality – now generally
referred to as the third wave of VR [2] - managed to thrill the tech world and even
the most critical platforms uninhibitedly expressed their high expectations for the new
medium. Technology magazine Wired, for instance, declared early in 2014 that the first
Oculus Rift demos did not only give rise to grand anticipations about the way VR
would revolutionize video game play, but that the format was also highly probable to
dramatically change the way we watch movies [40].
The possibility of unlimited interaction was one of the features that visionaries
openly fantasized about. Cinematic VR would not only allow the viewer to walk around
in movies, but also provide the opportunity to fully interact with the characters inhabiting
the story world, thus creating one’s own unique story by taking part in narrative fiction
in a way that would ultimately be indistinguishable from real life.
Looking back though, while these assumptions often seem to have been based on
pre-existing ideas and technological fantasies about VR, rather than on actual technical
possibilities, these statements certainly influenced the public opinion and expectations
about the medium. Graves [14] pointed out how the fictional representations of VR as
a wondrous technology affected how the public perceived the medium and warned that
since these fictional accounts on the possibilities of VR were often strongly positive, the
public may become disappointed with their actual experiences of VR.
Janet Murray [30] stated that one of the most striking responses to ‘the advent
of digital modes of representation in the second half of the 20th century’ has been the
anticipation – framed sometimes as a hope, sometimes as a fear – that digitally generated
forms will someday be indistinguishable from reality. By default, these expectations
include the possibility to interact and communicate with the inhabitants of digital realities
in a seemingly infinite domain of interactive narrativity. She pointed out that this attitude
of omnipotent representational powers leading to a replacement of the real world with
the virtual world has led non-scientists to overestimate the present and future of VR
experiences.
In the current study, we therefore aim to take a closer look at the actual possibilities
that ‘real’ contemporary VR offers in terms of interactive narrative and character inter-
action. Rather than be inspired by over-enthusiastic technological predictions, future
fantasies or science fiction, we intend to stick to what is actually available to the general
public and can be experienced with increasingly affordable and obtainable headsets. We
aim to present a selection of contemporary VR cases that feature a range of interest-
ing narrative novelties that may eventually inspire new perspectives for the design and
creation of interactive digital narratives.
2 Theoretical Considerations
In futuristic accounts of VR, interactors would have the opportunity to fully engage
with the characters that inhabit the story world, in a seemingly boundless interactive
narrative, with characters responding in ways that would feel so natural that it would be
indistinguishable from reality. Lombard and Ditton [25] proposed to use the term ‘per-
ceptual illusion of non-mediation’ for this level of technology that reaches a maximum
level of immersion, eventually leading to a situation in which interactors are no longer
aware of the fact that there is technology involved. In ‘real’ reality however, this level
of technological perfection is still mostly a theoretical thought experiment, as the actual
production of ‘flawless’ VR has turned out to be considerably more complicated [38].
One of the first industry accounts exploring the nature of the relation between users
of VR technology and the characters that may be encountered in virtual environments
was published on the blog of the Oculus Story Studio [4]. In his report, Matt Burdette
described the uncomfortable sensation that was detected during the making of their
early works such as Henry [17] and Lost [26]. Burdette describes how the team noticed
that in previous versions of these experiences, the team had noticed a distinct lack of
connection to the characters and the environment, and in turn, the story. Because of the
ghost-like feeling that one may have with this lack of acknowledgement and the sensation
of having no tangible relationship with your surroundings despite feeling present in the
Exploring Narrative Novelties in VR 437
world, Burdette coined the ‘Swayze Effect’ in a reference to the movie classic Ghost.
With Henry, the team discovered that by having the character ‘lock eyes’ with the user
- by looking right into the camera and seemingly acknowledging their existence - this
effect could be partially eliminated.
Since then, it has become clear that interactors in VR may develop dramatically
different connections to the characters and the virtual environment they inhabit than
they would with characters in traditional ‘flat’ media. As of yet however, research and
theory on this specific aspect of VR is still scarce.
2.1 Identification
Mainly referring to traditional media such as cinema and television shows, the possibility
to connect or interact with characters and the way audience members react to these char-
acters has been studied from different perspectives. A first perspective is identification,
described by Livingstone [24] as ‘imagining being in someone else’s shoes and seeing
the world through his or her eyes’. In his extensive theoretical discussion of the topic,
Cohen defined identification as ‘an imaginative process through which an audience mem-
ber assumes the identity, goals, and perspective of a character’ [6]. The author defines
identification as a mechanism through which audience members experience mediated
events from the inside, as if the events were actually happening to them. On first glance,
both descriptions seem to fit quite well to what users experience in cinematic VR, when
they are invited to take the perspective of someone else’s world as recorded by the 360°
camera, as was for instance described by Chris Milk: “It’s a machine, but inside of
it, it feels like real life. And you feel present with the world you are inside, and you
feel present with the people that you are inside of it with.” [32]. Another aspect of the
broader notion of identification with mediated characters has been explored within the
field of video game studies. Hefner, Klimmt and Vorderer [16] discussed identification
with game characters as a mechanism for computer game enjoyment and specifically
proposed game interactivity as an important facilitator of strong identification.
A second perspective that is considered with relations between viewers and media char-
acters is parasocial interaction (PSI). The concept of parasocial interaction refers to
audience members interacting with media personae as they would in face-to-face inter-
actions [22]. This concept was first discussed in reference to television personalities by
Horton and Wohl, who proposed to consider individual audience members as actively
relating to media performers in a psychological perspective, “as if they were involved in
an active face-to-face exchange rather than in passive observation” [in 15]. Livingston
[23] clarified that compared to identification, PSI adds an interactional component to
the relation between viewers and media personae that the former is lacking. However,
as Konijn and Hoorn [22] pointed out, the term parasocial interaction was coined in
a period when the audience was considered to passively consume mass media. Cohen
[6] argued that PSI applies to media figures who directly address the viewer, such as
newscasters and presenters, while Giles [13] proposed to differentiate between media
438 M. Vosmeer and C. Roth
figures who are direct representations of real people such as newscasters, and dramatic
characters played by actors and fantasy figures such as game characters.
3 Case Studies
For our current exploration, we have selected four contemporary VR case studies that
feature different examples of interactive storytelling. All four are available in the Oculus
Exploring Narrative Novelties in VR 439
Store, can be viewed with the Oculus Quest headset and were released between 2019 and
2021. And importantly, they all have been praised by the international VR community as
featuring innovative insights into narrative, gameplay and immersion, either by authors
on VR review sites or by juries of VR and new media festivals. After shortly describing
the VR experiences, we will discuss whether and how the concepts of identification,
parasocial interaction, ‘breaking the fourth wall’ and presence can be accounted for in
the selected sample.
take part in other game-related actions. This way, The Line shows that it is possible to
tell an interactive story to a user who is not involved with a body or a character, but still
be very active and emotionally engaged in the narrative.
4 Discussion
A first essential difference between the definitions by Livingstone and Cohen for identifi-
cation with characters in traditional media and the process of identification in VR seems
Exploring Narrative Novelties in VR 441
to be the level of psychological engagement that is required from the user. Identification
with a television character is based on a psychological attachment between the viewer
and a character, but rather than leading to interaction with the character, it leads to imag-
ining being the character. Precisely this last psychological process is partly superfluous
in 360° VR, as the technology facilitates the step between watching a fictional character
and imagining being in the shoes of that character. In the four cases that are presented in
the current overview, this particular mode of relating to a character in VR is developed
in a number of different ways.
Down the Rabbit Hole features a number of short scenes in which the gameplay shifts
from the usual third person to a first-person perspective, in which the player temporarily
takes up the role of the protagonist and is accordingly addressed by the other characters
in the story world. In the biggest part of the experience, though, the role of the user is
quite suitably described by Tekaia [37]: “You assume the role of what is essentially a
disembodied spirit floating in the center of the rabbit hole, with its curved subterranean
walls encircling you on all sides.” And while Down the Rabbit Hole does not offer any
surprising new insights into interactive storytelling, the particular narrative novelty of
this experience lies in the fact that the traditional fourth wall is provoked in an intriguing
way: not by breaking it but by transforming it into an interactive mobile cylinder.
In The Line, identification with the characters does not happen by taking up their
role, since the user is placed outside of the story world as a spectator who nonetheless
needs to perform certain actions to move the story forward. This mechanic, that in itself
already causes a certain commitment to the characters and their love story, places the
user in an interesting lacuna between presence and narrative engagement: while they are
not taking part in the story world, at the moment of play they are no longer in the ‘real’
reality either, as they are surrounded by the spatial reality of the virtual room in which
the diorama is placed. Near the end of the experience, the producers have inserted a small
innovative interaction when Pedro the puppet falls out of the diorama and ‘contacts’ the
user with gestures and facial expressions, begging them to help him get up again. Here, a
fourth wall is broken, with the user and the protagonist engaging in an active parasocial
relationship, temporarily taking part in the same performance and working together to
progress the narrative towards a happy ending.
While the gameplay is fairly simplistic and the narrative comes across as a classic
children’s story, the fusion of experience and interactivity in A Fisherman’s Tale show
what VR as a narrative medium may be capable of in terms of providing new possi-
bilities of interactive storytelling and character involvement. The user takes up the role
of fisherman Bob in a more or less classical first-person perspective, but by adding the
sensation of simultaneously being confronted with a giant and a miniature doppelgänger,
and at the same time being able to interact with the objects in their worlds, the sense of
embodiment in this experience is considerably stretched. While the use of a narrator’s
voice over ensures that users keep a clear cognitive distance between themselves and the
protagonist, the sense of spatial presence that is heightened by the use of mise en abyme
causes a distinct and surprisingly physical sense of engagement.
In Wolves in the Walls Lucy’s imagination permeates the experience and the resulting
dissonance between what is imagined and what could be real evokes a sense of curiosity
and engagement. Furthermore, sharing and actively living Lucy’s fantasies creates a
442 M. Vosmeer and C. Roth
bond with her and the player, leading to empathic immersion and social presence. The
player’s role undergoes an important transformation when Lucy draws them a pair of
hands, thereby granting agency to interact with the world. By taking pictures, shining
light in dark places or writing notes on the walls, they are from that moment on fully
involved in the narrative, not by guiding or controlling the protagonist, but by actively
cooperating with her. This coherence of narrative and gameplay mechanics creates what
could be understood as ludonarrative harmony and supports the player’s meaning-making
process [33]. Remarkably, in this experience the role of the player sometimes resembles
the ghost-like feeling that was described by Burdette [4], when the player’s existence
is not acknowledged by the other family members in the house. But the fact that Lucy
is constantly talking to them, interacting and actively seeking the player’s assistance
and collaboration, and the distinct ability to interact with the surroundings, together
create a sense of spatial presence as well as narrative engagement that show how the
establishment of media conventions for VR is still in constant development.
5 Conclusions
A characteristic that the four reviewed case studies have in common, is the fact that
while they all make use of technologically advanced interaction techniques, providing
remarkable insights into the creative possibilities of the medium, their narrative content
is mostly targeted at a children’s audience and derived from children’s books. And while
the reviewed cases are to a certain level emotionally engaging, they mostly seem to serve
as an exercise into what the medium eventually will be able to contain. However, the
concerns that were formulated in early reviews on the medium, and that were mostly
informed by future fantasies or science fiction, are not expected to easily become reality.
In her essay Virtual/reality: how to tell the difference, Janet Murray states that when
users enter a well-crafted VR experience, they will suspend disbelief and get lost in it,
the same way as they do in movies or video games. However, after taking the headset
off, a viewer returns to the one reality we all live in, where we are responsible for our
actions. The author points out that it is not hard now to tell the difference between the
real and the virtual, ‘and it will not get any harder in the future because that is not
how a medium of technology develops’ [30, p. 11]. She suggests viewing VR as an
emerging medium within an evolving community that is beginning to develop the media
conventions to support sustained interaction and immersion. With the current discussion
of four selected VR case studies, showing how the application of theoretical conventions
such as identification, parasocial interaction and ‘breaking the fourth wall’ may slightly
shift when used for VR, we hope to have contributed some rudimentary insights into the
range of possible media conventions that narrative virtual reality may contain.
Acknowledgement. This study is part of the VR for Diversity research project at the Amsterdam
University of Applied Sciences, funded by SIA RAAK.
References
1. Auter, P.J.: Psychometric: TV that talks back: an experimental validation of a parasocial
interaction scale. J. Broadcast. Electron. Media 36(2), 173–181 (1992)
Exploring Narrative Novelties in VR 443
2. Belisle, B., Roquet, P.: Guest Editors’ Introduction: Virtual reality: immersion and empathy
(2020)
3. Bell, E.: Theories of Performance. Sage, Los Angeles (2008)
4. Burdette, M.: The Swayze Effect (2015). https://storystudio.oculus.com/en-us/blog/the-swa
yze-effect/. Accessed 26 July 2021
5. Carroll, L.: Alice’s Adventures in Wonderland. Macmillan & Co. (1865)
6. Cohen, J.: Defining identification: a theoretical look at the identification of audiences with
media characters. Mass Commun. Soc. 4(3), 245–264 (2001)
7. Conway, S.: A circular wall? Reformulating the fourth wall for video games. J. Gaming
Virtual Worlds 2(2), 145–155 (2010)
8. Collins, K.: Breaking the fourth wall? User-generated sonic content in virtual worlds. In: The
Oxford Handbook of Virtuality, pp. 351–363. Oxford University Press, Oxford (2014)
9. Down the Rabbit Hole, Cortopia Studio’s (2020)
10. Feltham, J.: (2019). https://uploadvr.com/fishermans-tale-review/. Accessed 26 July 2021
11. A Fisherman’s Tale. Vertigo Games (2019)
12. Gaiman, N., McKean, D., Friend, B.: The Wolves in the Walls. HarperCollins, New York
(2003)
13. Giles, D.C.: Parasocial interaction: a review of the literature and a model for future research.
Media Psychol. 4(3), 279–305 (2002)
14. Graves, E.: Media influence on expectations of virtual reality. Refractory 30 (2018)
15. Hartmann, T.: Parasocial interactions and paracommunication with new media characters.
Mediated Interpersonal Commun. 177, 199 (2008)
16. Hefner, D., Klimmt, C., Vorderer, P.: Identification with the player character as determinant of
video game enjoyment. In: Ma, L., Rauterberg, M., Nakatsu, R. (eds.) ICEC 2007. LNCS, vol.
4740, pp. 39–48. Springer, Heidelberg (2007). https://doi.org/10.1007/978-3-540-74873-1_6
17. Henry. Oculus Story Studio (2015)
18. IndieSamAdonis (2021). https://indiesamadonis.com/2021/04/13/the-line-vr-review-the-pup
pets-of-sao-paulo-help-us-fall-in-love-with-vr/
19. Jefferson, A.: Mise en abyme and the Prophetic in Narrative. Style 196–208 (1983)
20. Kilteni, K., Groten, R., Slater, M.: The sense of embodiment in virtual reality. Presence:
Teleoper. Virtual Environ. 21, 373–387 (2012)
21. Ko, D., Ryu, H., Kim, J.: Making new narrative structures with actor’s eye-contact in cinematic
virtual reality (CVR). In: Rouse, R., Koenitz, H., Haahr, M. (eds.) ICIDS 2018. LNCS, vol.
11318, pp. 343–347. Springer, Cham (2018). https://doi.org/10.1007/978-3-030-04028-4_38
22. Konijn, E.A., Hoorn, J.F.: Parasocial interaction and beyond: media personae and affective
bonding. In: The International Encyclopedia of Media Effects, pp. 1–15 (2017)
23. The Line, Arvore Immersive Games Inc. (2020)
24. Livingstone, S.M.: Making Sense of Television: The Psychology of Audience Interpretation.
Routledge, New York (1998)
25. Lombard, M., Ditton, T.: At the heart of it all: the concept of presence. J. Comput.-Mediated
Commun. 3(2) (1997)
26. Lost. Oculus Story Studio (2016)
27. Ma, Z.: Effects of immersive stories on prosocial attitudes and willingness to help: testing
psychological mechanisms. Media Psychol. 23(6), 865–890 (2020)
28. McGee, American: American McGee’s Alice. Electronic Arts (2000)
29. McGee, American: Alice: Madness Returns. Electronic Arts (2011)
30. Murray, J.H.: Virtual/reality: how to tell the difference. J. Vis. Cult. 19(1), 11–27 (2020)
31. Pressgrove, G., Bowman, N.D.: From immersion to intention? Exploring advances in prosocial
storytelling. J. Philanthropy Market. 26(2), e1689 (2021)
32. Rose, M.: The immersive turn: hype and hope in the emergence of virtual reality as a nonfiction
platform. Stud. Documentary Film 12(2), 132–149 (2018)
444 M. Vosmeer and C. Roth
33. Roth, C., van Nuenen, T., Koenitz, H.: Ludonarrative hermeneutics: a way out and the narrative
paradox. In: Rouse, R., Koenitz, H., Haahr, M. (eds.) Interactive Storytelling, ICIDS 2018,
pp. 93–106. Springer, Cham (2018). https://doi.org/10.1007/978-3-030-04028-4_7
34. Shafer, D.M., Carbonara, C.P., Korpi, M.F.: Exploring enjoyment of cinematic narratives in
virtual reality: a comparison study. Int. J. Virtual Real. 18(1), 1–18 (2018)
35. Slater, M: A Note on Presence Terminology (2003). https://www.researchgate.net/public
ation/242608507_A_Note_on_Presence_Terminology
36. Slater, M.: Immersion and the illusion of presence in virtual reality. Br. J. Psychol. 109(3),
431–433 (2018)
37. Tekaia, P.: Down the Rabbit Hole review (2020). https://adventuregamers.com/articles/view/
40467. Accessed 30 July 2021
38. Vosmeer, M., Schouten, B.: Project Orpheus a research study into 360 cinematic VR. In:
Proceedings of the 2017 ACM International Conference on Interactive Experiences for TV
and Online Video, pp. 85–90 (2017)
39. Vosmeer, M., Roth, C., Koenitz, H.: Who are you? Voice-over perspective in surround video.
In: Nunes, N., Oakley, I., Nisi, V. (eds.) ICIDS 2017. LNCS, vol. 10690, pp. 221–232. Springer,
Cham (2017). https://doi.org/10.1007/978-3-319-71027-3_18
40. Watercutter, A.: Oculus is awesome for games, but it’s the future of movies. Wired (2014).
http://www.wired.com/2014/01/oculus-movies/. Accessed 26 July 2021
41. Wolves in the Walls, Fable Studio Inc. (2019)
42. Zibrek, K., Kokkinara, E., McDonnell, R.: The effect of realistic appearance of virtual char-
acters in immersive environments-does the character’s personality play a role? IEEE Trans.
Visual Comput. Graph. 24(4), 1681–1690 (2018)
The Interactive Narrative Research
Discipline and Contemporary Practice
A Preliminary Survey on Story
Interestingness: Focusing on Cognitive
and Emotional Interest
1 Introduction
Storytelling has diverse roles. Through storytelling activities, we can learn lan-
guages and develop our literacy skills, including reading and writing [20]. Lan-
guage and literacy learning is particularly effective in digital and online sto-
rytelling [28,38]. Storytelling research is also widely conducted for entertain-
ment and other purposes such as transmedia storytelling [14], empathy building
through collaborative storytelling [37] and Game AI [30].
In video games and interactive storytelling, interactivity is a key motivational
factor for the player to continue playing [26,32]. Rigby and Ryan [32] introduce
the Player Experience of Need Satisfaction (PENS) model to understand better
the relationship between the players’ motivation and their well-being due to
game playing. In the PENS model, three core desires - competence, autonomy
(i.e., choice and violation), and relatedness (i.e., companionship) - are suggested
as motivations for game playing, bringing the player immersion. Similarly, in the
interactive digital narrative, agency (i.e., “the user’s ability to control aspects
of the narrative”) must make sense and be meaningful for user satisfaction [26].
Due to the time and space limit, we focus on linear narrative in this paper,
leaving out interactivity as future work.
Narrative understanding entails a “mental mechanism” including cognitive
and emotional processes [24,44], and interestingness is of great importance both
c Springer Nature Switzerland AG 2021
A. Mitchell and M. Vosmeer (Eds.): ICIDS 2021, LNCS 13138, pp. 447–453, 2021.
https://doi.org/10.1007/978-3-030-92300-6_45
448 B.-C. Bae et al.
– Schema Violation: The reader can use schema to understand a story, where
a schema refers to “a data structure for representing the generic concepts
stored in memory” [33]. The schema violation resulting from unexpectedness
or incongruity can lead to story interest [22,34]. Furthermore, the violated
unexpectedness or incongruity requires proper resolution, possibly with con-
sistency and novelty. These schema violation factors are closely connected
with surprise as an aesthetic emotion.
2.3 Discussion
In addition to the cognitive and emotional factors of story interestingness,
another two aspects can be explored - interactivity and narrative factors. First,
as mentioned earlier in the Introduction section with the PENS model [32],
interactive narrative allows the reader to feel more immersive in a story world
by giving a sense of control to the reader. Thus appropriate indicators such as
choice ranges and choice variabilities for measuring interactive narrative experi-
ences will be necessary [41].
Second, the suggested cognitive and emotional factors of story interestingness
are related to story understanding to some extent. Narrative factors - such as
plot structure, character types, narrative devices (e.g., temporal devices such as
flashback and foreshadowing [4,45]), story conventions and clichè [21] - can be
explored further.
3 Conclusion
In this paper we explore the two factors of story interestingness - cognitive
interest and emotional interest. Based on the outlined literature review, we
present our incipient framework for measuring story interestingness. The pro-
posed framework consists of two factors - cognitive factors mainly for comprehen-
sion (goal, novelty, inference, schema violation) and emotional factors (empathy,
external emotions, humor, outcome valence).
A Preliminary Survey on Story Interestingness 451
Our work is in its early stage and has limitations, such as a lack of com-
putational metrics for measuring story interest. We plan to extend our current
framework with two further steps. First, we intend to analyze the factors of
story interestingness from the narrative theories perspective, centering around
plot and characters. Second, we will investigate computational approaches to
the proposed cognitive, emotional, and narrative factors, detailing the presented
framework.
Acknowledgements. This work was supported by NCSOFT NLP Center. This work
was also supported in part by Institute for Information & communications Technol-
ogy Promotion (IITP) grant funded by the Korea government (MSIT) (No. 2017-
0-01772, Development of QA systems for Video Story Understanding to pass the
Video Turing Test) and the National Research Foundation of Korea (NRF) grant
(2021R1A2C1012377).
References
1. Alabdulkarim, A., Li, S., Peng, X.: Automatic story generation: challenges and
attempts. In: Proceedings of the Third Workshop on Narrative Understanding,
pp. 72–83. Association for Computational Linguistics, Virtual (2021). https://doi.
org/10.18653/v1/2021.nuse-1.8
2. Alhussain, A., Azmi, A.: Automatic story generation: a survey of approaches. ACM
Comput. Surv. 54, 1–38 (2021). https://doi.org/10.1145/3453156
3. Bae, B.C., Jang, S.J., Ahn, D.K., Seo, G.: A VR interactive story using POV and
flashback for empathy. In: 2019 IEEE Conference on Virtual Reality and 3D User
Interfaces (VR), pp. 844–845 (2019). https://doi.org/10.1109/VR.2019.8797899
4. Bae, B.-C., Young, R.M.: A use of flashback and foreshadowing for surprise arousal
in narrative using a plan-based approach. In: Spierling, U., Szilas, N. (eds.) ICIDS
2008. LNCS, vol. 5334, pp. 156–167. Springer, Heidelberg (2008). https://doi.org/
10.1007/978-3-540-89454-4 22
5. Behrooz, M., Robertson, J., Jhala, A.: Story quality as a matter of perception:
using word embeddings to estimate cognitive interest. In: Proceedings of the AAAI
Conference on Artificial Intelligence and Interactive Digital Entertainment, vol. 15,
no. 1, pp. 3–9 (2019). https://ojs.aaai.org/index.php/AIIDE/article/view/5217
6. Brewer, W.F., Lichtenstein, E.H.: Stories are to entertain: a structural-affect the-
ory of stories. J. Pragmat. 6(5), 473–486 (1982). https://doi.org/10.1016/0378-
2166(82)90021-2
7. Cummings, J.J., Tsay-Vogel, M., Cahill, T.J., Zhang, L.: Effects of immersive sto-
rytelling on affective, cognitive, and associative empathy: the mediating role of
presence. New Media Soc. (2021). https://doi.org/10.1177/1461444820986816
8. Dijkstra, K., Zwaan, R.A., Graesser, A.C., Magliano, J.P.: Character and reader
emotions in literary texts. Poetics 23(1), 139–157 (1995). https://doi.org/10.1016/
0304-422X(94)00009-U
9. Gerrig, R.J.: Is there a paradox of suspense? A reply to Yanal. Br. J. Aesthetics
37(2), 168–174 (1997). https://doi.org/10.1093/bjaesthetics/37.2.168
10. Gordon, M.: What makes humor aesthetic. Int. J. Human. Soc. Sci. 2(1), 62–70
(2012)
452 B.-C. Bae et al.
11. Hidi, S., Baird, W.: Interestingness–a neglected variable in discourse processing.
Cogn. Sci. 10(2), 179–194 (1986). https://doi.org/10.1016/S0364-0213(86)80003-9
12. Hoeken, H., Vliet, M.: Suspense, curiosity, and surprise: How discourse structure
influences the affective and cognitive processing of a story. Poetics 27, 277–286
(2000). https://doi.org/10.1016/S0304-422X(99)00021-2
13. Iran-Nejad, A.: Cognitive and affective causes of interest and liking. J. Educ. Psy-
chol. 79, 120–130 (1987)
14. Jenkins, H.: Transmedia storytelling and entertainment: an annotated syllabus.
Continuum 24(6), 943–958 (2010). https://doi.org/10.1080/10304312.2010.510599
15. Jose, P.E.: Story interestingness: goal importance or goal attainment difficulty?
Paper presented at the Annual Meeting of the American Education Research Asso-
ciation (1984)
16. Keen, S.: Empathy and the Novel. Oxford University Press, Oxford (2007)
17. Kim, S.I.: Causal bridging inference: a cause of story interestingness. Br. J. Psychol.
90(1), 57–71 (1999). https://doi.org/10.1348/000712699161260
18. Kintsch, W.: Learning from text, levels of comprehension, or: why anyone would
read a story anyway. Poetics 9(1–3), 87–98 (1980)
19. Kybartas, B., Bidarra, R.: A survey on story generation techniques for authoring
computational narratives. IEEE Trans. Comput. Intell. AI Games 9(3), 239–253
(2017). https://doi.org/10.1109/TCIAIG.2016.2546063
20. Lucarevschi, C.R.: The role of storytelling on language learning: a literature review
(2016)
21. MacDowell, J.: Happy Endings in Hollywood Cinema. Edinburgh University Press,
Edinburgh (2014)
22. Mandler, G.: The structure of value: accounting for taste. CHIP report, Center for
Human Information Processing, Department of Psychology, University of Califor-
nia, San Diego (1981)
23. Manney, P.: Empathy in the time of technology: how storytelling is the key to
empathy. J. Evol. Technol. 19(1), 51–61 (2008)
24. Matravers, D.: Fiction and Narrative, 1st edn. Oxford University Press, Oxford
(2014)
25. Meyer, J.C.: Humor as a double-edged sword: four functions of humor in communi-
cation. Commun. Theory 10(3), 310–331 (2000). https://doi.org/10.1111/j.1468-
2885.2000.tb00194.x
26. Miller, C.: Digital Storytelling: A Creator’s Guide to Interactive Entertainment,
4th edn. CRC Press, Boco Raton (2019)
27. Oatley, K.: A taxonomy of the emotions of literary response and a theory of iden-
tification in fictional narrative. Poetics 23(1–2), 53–74 (1995). https://doi.org/10.
1016/0304-422X(94)P4296-S
28. Rahimi, M., Yadollahi, S.: Effects of offline vs. online digital storytelling on the
development of EFL learners’ literacy skills. Cogent Educ. 4(1), 1285531 (2017).
https://doi.org/10.1080/2331186X.2017.1285531
29. Randall, W.: Narrative intelligence and the novelty of our lives. J. Aging Stud. 13,
11–28 (1999). https://doi.org/10.1016/S0890-4065(99)80003-6
30. Riedl, M., Thue, D., Bulitko, V.: Game AI as storytelling. In: González-Calero,
P.A., Gómez-Martı́n, M.A. (eds.) Artificial Intelligence for Computer Games, pp.
125–150. Springer, New York (2011). https://doi.org/10.1007/978-1-4419-8188-2 6
31. Riedl, M.O.: Computational narrative intelligence: a human-centered goal for arti-
ficial intelligence (2016)
32. Rigby, S., Ryan, R.: Glued to Games: How Video Games Draw Us in and Hold Us
Spellbound. New directions in media, Praeger (2011)
A Preliminary Survey on Story Interestingness 453
33. Rumelhart, D.E.: Understanding Understanding. Taylor & Francis, Milton Park
(2013)
34. Schank, R.C.: Interestingness: controlling inferences. Artif. Intell. 12, 273–297
(1979)
35. Schindler, I., et al.: Measuring aesthetic emotions: a review of the literature and
a new assessment tool. PLOS ONE 12(6), 1–45 (2017). https://doi.org/10.1371/
journal.pone.0178899
36. Silvia, P.: Looking past pleasure: anger, confusion, disgust, pride, surprise, and
other unusual aesthetic emotions. Psychol. Aesthetics Creativity Arts 3, 48–51
(2009). https://doi.org/10.1037/a0014632
37. Skaraas, S.B., Gomez, J., Jaccheri, L.: Playing with empathy through a collabo-
rative storytelling game. In: Clua, E., Roque, L., Lugmayr, A., Tuomi, P. (eds.)
ICEC 2018. LNCS, vol. 11112, pp. 254–259. Springer, Cham (2018). https://doi.
org/10.1007/978-3-319-99426-0 26
38. Smeda, N., Dakich, E., Sharda, N.: The effectiveness of digital storytelling in
the classrooms: a comprehensive study. Smart Learn. Environ. 1(1), 1–21 (2014).
https://doi.org/10.1186/s40561-014-0006-3
39. Smith, A.: Cognitive empathy and emotional empathy in human behavior and
evolution. Psychol. Rec. 56(1), 3–21 (2006). https://doi.org/10.1007/BF03395534
40. Snyder, B.: Save the Cat!: The Last Book on Screenwriting You’ll Ever Need.
Cinema/Writing, M. Wiese Productions (2005)
41. Szilas, N., Ilea, I.: Objective metrics for interactive narrative. In: Mitchell, A.,
Fernández-Vara, C., Thue, D. (eds.) ICIDS 2014. LNCS, vol. 8832, pp. 91–102.
Springer, Cham (2014). https://doi.org/10.1007/978-3-319-12337-0 9
42. Vandaele, J.: Humor mechanisms in film comedy: incongruity and superiority. Poet-
ics Today 23(2), 221–249 (2002). https://doi.org/10.1215/03335372-23-2-221
43. Vandaele, J.: Narrative humor (i): enter perspective. Poetics Today 31, 721–785
(2010). https://doi.org/10.2307/25835337
44. Worth, S.E.: Narrative understanding and understanding narrative. Contemp. Aes-
thetics 2(2), 9 (2004)
45. Wu, H.Y., Young, M., Christie, M.: A cognitive-based model of flashbacks for
computational narratives, vol. 12, pp. 239–245 (2016). https://ojs.aaai.org/index.
php/AIIDE/article/view/12873
46. Yanal, R.J.: The paradox of suspense. Br. J. Aesthetics 36(2), 146–159 (1996)
Imagining the Other for Interactive Digital
Narrative Design Learning in Real Time
in Sherlock
jtm@ucf.edu
foregrounding the importance of collaboration between designers and players and for
making the study of collaborative behaviors scalable and seamless. We also define the
term authoring-other exchange to describe the process where authors gain knowledge
through IDN authoring and sharing reflections in real time (synchronously) while using
the same tool and exchanging feedback. We are particularly interested in examining how
peers of similar and different sociocultural affiliations exchange viewpoints through such
feedback. This paper presents the design considerations, architecture and preliminary
findings of a pilot conducted on an online media research platform modified to facilitate
the workshop.
This paper is organized as follows. First, in this section, we describe the background
of the Imaging the Other project and its connection to previous research. Next, we
describe extensions we developed for an existing media research platform and, specif-
ically, the Twine module created for it. We then introduce the Imagining the Other
workshop series, describe the coding schema used for analysis, and provide a sample
analysis and insights from a preliminary pilot study. Finally, we discuss the implications
for future research directions.
1.1 Background
A prior study in a physical lab setting showed how exchanging the designer–player
roles was productive to show how novice peers were able to incorporate feedback into
their authoring process [1]. However, the methods of using unmodified tools proved too
cumbersome to implement for educational practice and research. Imagining the Other
focuses on scaling such research because of the importance of studying how peers of
similar sociocultural backgrounds (ethnicity, gender, language) work when acting as
authors and audience members. In the new design, peers not only learn an authoring
tool for the first time but are also providing and receiving feedback synchronously with
another player-author in alternating turns. This real time learning, designing, and playing
process, we argue, has the potential to illustrate the inherently social nature of IDN and
the ways in which different social interactions impact development of the interactive
narrative. To better study these interactions, we also needed a practice-based research tool
to accommodate hundreds of students across a large urban university system. We turned
to Sherlock, an existing media research tool [2] that previously supported collecting,
analyzing and visualizing research data on IDN traversals through a web interface. We
extended the platform with a Twine module that enables participants to both author stories
and exchange feedback through a Sherlock web client. In addition to collecting the data,
we further identified the need to manage participant participation through scheduling
and communication in order to minimize errors and drop outs.
provides an example of one such research project, offering guidelines for how to shape
and encourage collaboration among larger groups [4]. Recently, Millard et al. call for
additional research on the authoring process itself as they report on interviewing 20
different digital interactive authors about the process [5]. Our approach focuses on the
initial introduction to authoring for interactive digital narratives but shares many of
the concerns with understanding the nature of authoring through observation and tool-
building, and further argue that the community needs to collaborate on shared platforms
to enable additional research in this area.
The Imagining the Other project required supporting multi-user interaction on Twine.
The previous work identified several challenges in how the study was conducted, includ-
ing physical steps required to share the output of one participant with the other. Twine
supports exporting a project as HTML, but sharing the file requires participants to under-
stand and employ another method, such as email or other channel, and be able to identify
the recipient. This proved challenging, as participants would need to receive additional
instructions and could lose work due to user error. Internally, Twine saves story data
using local browser storage, which is available to other JavaScript applications running
from the same domain. With this knowledge, we developed code to watch for changes
and record the saved browser data as a story was edited.
2.1 Sherlock
Twine was designed for a single author to create interactive stories and publish them
as a web page. Recording and transferring Twine data required participant action in
the default version of Twine, such as exporting the work as an HTML file, slowing
down the overall process. In the current version of Twine, there is also no support for
recording events or histories of edits made to the story. These two limitations of the
Imagining the Other for Interactive Digital Narrative Design Learning 457
open-source tool necessitated modifications both to the Twine authoring tool itself as
well as additional support for recording data to study.
For the Imagining the Other project, we developed a special module for Sherlock that
wraps and adds features to Twine. We record events generated by the Twine editor, such as
creating, editing, and deleting passages, from a modified version of Twine inside an inline
frame element and share it through the server with the partner, including the story content
typically stored in the session storage. This allows us to save timestamped snapshots of
the story, which effectively transforms Twine from a single-user tool into a multi-user
environment, and further allowed partners to annotate passages using their titles with
their comments as they played the story. Using the existing API within Sherlock, this
data is sent to the server and stored in the database and published to other clients.
In addition to embedding both the Twine interface and runtime into the Sherlock web
application, we implemented additional logic in JavaScript to manage the overall protocol
of the study, such as advancing to the next step and changing the mode of the interface
as well as publishing events and data to the server. This automation enabled participants
to remain in sync during the session. These features were critical for making the real-
time exchange of comments available to the peers and supporting the authoring-other
exchange process.
next authoring turn, each participant is prompted to “Consider your player’s reflections
and continue your IDN design.”
Exit survey
Player Player Player +
Twine Design turn Design turn Design turn Design turn
reflection reflection reflection certificate
tutorial 1 2 3 4
turn 1 turn 2 turn 3 of
completion
To evaluate the feasibility of the series, we applied a coding schema to the data
produced by the workshop sessions. These data included authored Twine content, the
comments from the author and their partner, and how the author acted on the comments
across several design-play-edit turns. In the next subsection, we describe the data analysis
categories as they were implemented from the pilot and a sample analysis in the following
subsection.
process and platform underwent changes, as appropriate for a pilot run of such a complex
system. We also explored how features of the authoring-exchange platform, unavailable
offline, might suggest dynamic analysis categories linking all three strands of the data.
Because of the time stamps in the research panel shown in Fig. 2, we explored possible
intra-dyad creative synergies at those moments, as detailed in the next subsection.
“intersubjective design pivot”, we argue, is a key finding and central to ongoing research
using Sherlock and the application of the coding schema.
techniques, though, including Unity and Ink, and in the future, XR content authoring
tools such as that found in Facebook Horizons [11].
One thing we have found is that there is an urgent need to attend to the needs
of researchers studying the authoring tools and the use of the authoring tools. The
complexity of the content and the availability of open-source versions of many of the
tools provides an opportunity to collect detailed data that would allow understanding
of how collaborative IDN authoring can be used for various ends, such to offer insights
for IDN authoring pedagogy and for the field to elaborate the social justice potential of
IDN design, such as creating and supporting creation with someone of a very different
background and identity.
Acknowledgments. The authors thank Jack J. Wright for developing the design structuremeasure
and Fedor Marchenko for assistance with early phases of the pilot.
References
1. Daiute, C., Duncan, R.O., Marchenko, F.: Meta-communication between designers and play-
ers of interactive digital narratives. In: Rouse, R., Koenitz, H., Haahr, M. (eds.) ICIDS 2018.
LNCS, vol. 11318, pp. 134–142. Springer, Cham (2018). https://doi.org/10.1007/978-3-030-
04028-4_10
2. Murray, J.T.: Telltale hearts: encoding cinematic choice-based adventure games (2018).
https://escholarship.org/uc/item/1n02n02z
3. Mitchell, A., McGee, K.: Designing hypertext tools to facilitate authoring multiple points-
of-view stories. In: Proceedings of the 20th ACM conference on Hypertext and hypermedia
- HT ’09. pp. 309. ACM Press, Torino (2009). https://doi.org/10.1145/1557914.1557966
4. Rouse, R.: Someone else’s story: an ethical approach to interactive narrative design for cultural
heritage. In: Cardona-Rivera, R.E., Sullivan, A., Young, R.M. (eds.) ICIDS 2019. LNCS, vol.
11869, pp. 47–60. Springer, Cham (2019). https://doi.org/10.1007/978-3-030-33894-7_6
5. Kitromili, S., Jordan, J., Millard, D.E.: What authors think about hypertext authoring. In:
Proceedings of the 31st ACM Conference on Hypertext and Social Media, pp. 9–16. ACM,
Virtual Event USA (2020). https://doi.org/10.1145/3372923.3404798
6. Telltale Games: The Wolf Among Us. (2013)
7. Daiute, C.: Narrative Inquiry: A Dynamic Approach. Sage Publications (2013)
8. Koenitz, H., Dubbelman, T., Knoller, N., Roth, C.: An integrated and iterative research direc-
tion for interactive digital narrative. In: Nack, F., Gordon, A.S. (eds.) ICIDS 2016. LNCS, vol.
10045, pp. 51–60. Springer, Cham (2016). https://doi.org/10.1007/978-3-319-48279-8_5
9. Murray, J., Mateas, M., Wardrip-Fruin, N.: Proposal for analyzing player emotions in an
interactive narrative using story intention graphs. In: Proceedings of the 2017 International
Conference on the Foundations of Digital Games (2017)
10. Atlas.ti. Scientific Software Development GmbH
11. Murray, J.T., Johnson, E.K.: XR Content Authoring Challenges: The Creator-Developer
Divide. CRC Press (2021). https://doi.org/10.1201/9781003052838-16
A Proposed Curriculum for an Introductory
Course on Interactive Digital Narratives
in Virtual Reality
Abstract. Immersive Media programs of study are being developed and enacted
at many higher education institutions. It is proposed that a course on Interactive
Digital Narratives (IDN) in Virtual Reality (VR) can familiarize undergraduate stu-
dents of diverse backgrounds with the foundational technical, design, and develop-
ment tenets of immersive storytelling. Course curriculum balances IDN design and
immersive storytelling strategies with VR project management, user experience
and interface design, spatial audio, digital scenography, introductory program-
ming, and rudimentary artificial intelligence. The course connects technical and
media affordances to theories of IDN to provide an introductory understanding of
IDN in VR. The proposed course ran in the spring of 2021 at a small liberal arts
college. The paper presents the course’s 15-week curriculum. An evaluation that
includes student work, insights lessons, and resources is provided.
building, the main objective for the proposed course was for students to become adept
at developing IDN experiences for VR.
In academia, the desire to standardize IDN language, design strategies, and theories
has been emphasized by Hartmut Koenitz and other members of ARDIN [1, 2, 4, 10,
11]. The authors of a recent presentation by the Immersive Research Learning Network
(IRLN), “The State of XR and Immersive Learning Outlook 2021” expressed a desire
for more immersive storytelling in XR curricula as a credible pathway toward standard-
ization [12]. Yet, as of this writing, in the archived proceedings of the IRLN there is not
a single article proposing a curriculum for a course that teaches IDN for VR [13].
The VR audience, which grew by 60% in 2020 (headset purchases) [14], is a testa-
ment to industry’s need for capable developers, designers, and storytellers. To prepare
undergraduate students to succeed in these areas, this paper proposes a 15-week curricu-
lum for an introductory course on creating IDNs in VR. Note that by introductory, we do
not mean that this course is suitable for first-year students. Instead, we propose this as
an introduction to the practice of crafting IDNs in VR. The course was evaluated by an
informal midterm evaluation and an end-of-the-semester formal evaluation. Due to the
small sample size, the results are preliminary and should not be generalized. Lessons,
outcomes, and insights for future research and curriculum development follow.
2 The Course
As part of a new Immersive Media BA and minor program at Columbia College Chicago,
a series of intensive courses were being developed to replace two survey courses. The
two survey courses introduced students to immersive media generally, from augmented
to mixed reality and projection mapping to immersive theater. They provided no tech-
nical training or theory. To give form to the discipline as part of the degree program, the
proposed course was designed and developed. The course goals were: (1) Develop an
understanding of the design and development techniques for VR experiences (Virtual
Environment Design, Unity programming, Spatial Audio, VR UI, VR UX, and Inter-
action Design), (2) Construct a VR experience with best practice and artist-led tech-
niques, (3) Understand IDN story structures and implement them in a VR experience
(4) Articulate and receive clear feedback through usability studies.
With the establishment of the Immersive Media program. It was deemed critical
that theory and practice be well-balanced and blended within its curricula. As discussed
in previous work on IDN pedagogy, the creation of IDNs requires a “practice-based
education” [1]. Accordingly, the course was scheduled to meet once a week for three
hours to accommodate both direct instruction (lectures) and practice-based learning.
The course time was broken into 3 one-hour sections. In the first five weeks, two
hours were spent on lectures on IDNs in VR. One hour was spent on design, coding, or
development activities. In the second five weeks, time was split evenly between lecture
and studio activities. In the final five weeks, students spent class time doing work on
their VR IDNs and working through usability studies to produce the final experience.
The topics covered in the course were C# programming in Unity, locomotion, spatial
audio, virtual environment design and digital scenography, interaction design, VR UI and
UX, usability tests and feedback, IDN story structures and writing strategies. Critically,
464 J. A. Fisher and J. T. Samuels
each of these topics needed to scaffold effectively over 15 weeks for students to produce
a compelling portfolio piece [15].
Table 1. IDN in VR Course Schedule. Classes were broken up into lectures, instructor-led
activities, and studio time. Subjects and practices would often meld from one class to the next.
Interactive Digital Narrative Overview. Early lectures on IDN surveyed the field’s
history [17], what differentiates IDN from traditional narrative[18], dramatic agency[19],
and canonical IDN works suggested by Hartmut Koenitz and Mirjam Palosaari Eladhari
[4]: namely Façade [20], afternoon, a story[21], and Save the Date [22]. Further, VR IDNs
from SideQuest were assigned. A list of these follows this section. Marie-laure Ryan’s
work on textual architectures is also presented by Week 3 [23]. These architectures were
then used in the next weeks to begin producing IDNs. With only 15 weeks, the goal was
to get students building as quickly as possible.
Interactive Digital Narrative Writing Strategies. Ryan’s story structures [24], were
used as worksheets and templates. Specifically, the Vector with Side Branches structure
was presented as a plot to travel through in a Spatial Storyworld Architecture [24, 25].
This encouraged the students to think of each of the vector’s branches as a different VR
space. As Asim Hameed and Andrew Perkis discuss, from each space come moments for
dramatic agency and storytelling [26]. Following the suggestion of Colette Daiute and
Hartmut Koenitz, students were led in character development exercises [3]. These Non-
Player Characters (NPC) became nodes for branching dialogue and the narrative within
A Proposed Curriculum for an Introductory Course on Interactive Digital Narratives 465
the VR spaces. Concepts of worldbuilding were introduced along with immersion [18,
25] and Sense of Presence (SoP) [27–29]. Students created their storyworld, the spaces
within that world, and what interactions would increase and maintain immersion and
SoP. These were then reviewed and critiqued by peers in class before being revised.
Basic Interaction and Locomotion Design. VR IDNs that achieve immersion, support
an SoP for interactors that enforces an affective experience [30], and provides them dra-
matic agency requires consistent, transparent, and meaningful interactions [18, 23, 31,
32]. At this point in the semester, students were taught to code and design basic interac-
tions. Tutorials were foundational and implemented in Unity. Advanced narrative game
mechanics were discussed in a later module. Students were taught the importance of
interaction design, both as being meaningful to the narrative [19] and as an opportunity
for joy and play [33] to support immersion. Video tutorials for designing foundational
grabbing and pointing interactions were provided along with Unity Packages of pre-
coded and designed scenes. A 3rd party interaction framework was provided later in the
semester to help students develop and design faster.
Locomotion Design is critical for creating immersion and for exploring an IDN with a
Spatial Storyworld architecture [33–36]. Simultaneously, locomotion in VR is an issue of
accessibility wherein an interactor’s natural mode of locomotion must be considered [37].
Students are taught to implement multiple forms of locomotion including (1) impossible
space architectures [38], (2) joystick, (3) teleportation, (4) flight, (5) climbing, (6) and
vehicle or platform based [39].
Virtual Environment Design and Digital Scenography. For the course’s prescribed
story architecture, Virtual Environment (VE) Design and Digital Scenography were
taught to achieve spatial immersion, a focus on game spaces [33] that tell stories. Level
design material was based on the work of Miriam Bellard of Rockstar North [40] and
others [41–43]. There is a fair amount of semantic overlap between narrative in games,
experience, and level design. Students were taught to collapse these concepts into a
form of digital scenography. Digital Scenography is the practice of using the design and
composition of space to tell stories and support immersion [26, 44]. These practices
included how to use light [45], compose spatial sound [46], and the use of color to direct
user attention [47, 48]. Activities on how to create these environments—the technical
processes and conceptual work—were done in class and for homework.
in the semester. Given the unique nature of each IDN, materials and resources for these
mechanics were produced through one-on-one instruction with peer review.
Students may not have had the opportunity to experience VR. Accordingly, they need
to become familiar with how people tell stories with the medium. The following are
suggested experiences. They are available on SideQuest and the Oculus store.
• Vanishing Grace [61]: A narrative puzzle game that uses spatial exploration, two
different modes of locomotion, rich voice acting, and interactions that are both ludic
and narratological. It is also a case study in limiting scope, standardizing interactions
in the world, and designing meaningful spaces.
• Moss [62]: A 3rd Person VR adventure experience that upsets expectations for VR
narratives. The interactor plays a god-like character that directs Quill, a young mouse.
The spaces are lush, but more importantly, Moss is a case study in how to build a
social sense of presence for the interactor with a character. When the interactor is
not active, Quill will look up at them and respond to their inactivity. This interaction
helps to teach the importance of character development and social presence and their
impacts on immersion.
A Proposed Curriculum for an Introductory Course on Interactive Digital Narratives 467
• Q1 (4.8/5): (A1) I want more discussion of the VR experiences and best practices.
(A2) Some of the videos are less useful than others. I would prefer having a smaller
core of videos that best demonstrate learning concepts and then having a list of extra
videos we can choose to watch for more inspiration.
A Proposed Curriculum for an Introductory Course on Interactive Digital Narratives 469
• Q2 (4.5/5): (A1) The lectures on immersion and presence felt rushed. (A2) If I had
known this course was about storytelling and not games I might not have taken it
(A3) Too much focus on narrative and not VR. (A4) The lectures are a little heavy,
however, the classes always provide interesting and useful content that I think will
help me in the future.
• Q3 (4.2/5): (A1) Slower tutorials please! (A2) There are not enough tutorials for
my story. (A3) Class playthroughs for feedback are long and boring. (A4) The C#
programming is very difficult. Provide more documentation.
• Q4 (4.9/5): N/A
• Q1 (5/5): (A1) The class was tough, but it felt worth it to learn. (A2) This course
certainly wasn’t for the faint of heart. It was incredibly challenging, but also very
fulfilling. (A3) In the future I think it is very important to do some pre-evaluation on
the skill level and knowledge of the students. This course was incredibly challenging,
and I felt way in over my head. (A4) Students need to know the level of work before
signing up.
• Q2 (5/5): (A1) When I joined the course I was told that as long as I followed tutorials
I would be able to have a finished project enough to do well. This was not the case for
me. (A2) In studio collab, I feel that since the project would not be my own, I would
not feel as motivated as I did in this class.
• Q3 (5/5): (A1) I had to put in probably 20 h a week on this course. Students need to
know and be prepared for the level of work they are signing up for. (A2) My experience
in this course was probably different from other students in the course, simply because
I had no previous experience with game design or Unity, as well as very very basic
understanding of C#. (A3) I’m really really proud of myself for sticking with it and
producing the final product.
• Q4 (5/5): (A1) Teach the class about SideQuest earlier. (A2) Once we started to
build our world the guidance and tutorials didn’t feel like enough. (A3) Presenting the
470 J. A. Fisher and J. T. Samuels
students with tools and things they can utilize to build their worlds would be beneficial
before starting to get into the conception phase. If I knew how difficult some of the
world & interaction building be I would probably start with something a whole lot
simpler for myself.
The responses to the questions during the final evaluation indicate several changes
for the future course. As indicated by all respondents, the class was difficult and required
a major investment outside of course time. The college where the class was run expects
students to spend 9 h outside of class on material. Clearly, as discussed in Q3 A1, some
students did more than double that amount of work. And, as in Q1, some students felt
wholly unprepared for the scope of work required to complete an IDN in VR. In response
to these challenges, future courses will provide a wide range of design and code templates
to ease the production demands for students. Plug-and-play templates might be the most
effective even if it reduces the diversity of IDNs created. Options like Quill [71], the
VR illustration and animation tool, might be more accessible and are frequently used
for sequential narratives.
In response to questions and responses about the production demands (Q2 A1, Q4
A2 and A3), an established 3rd party framework for interactions will be implemented
in the future course earlier in the semester. There are several 3rd party frameworks that
can be implemented in Unity for the Oculus Quest (VR Easy [72], VRTK [73], and
NewtonVR [74] to name a few). These templates provide a breadth of interactions with
comprehensive documentation. The future course will utilize The VR Interaction Frame-
work produced by Bearded Ninja Games [75]. The framework has documentation and
tutorials for grabbing, grabbable events, climbing, teleportation, joystick locomotion,
ziplines, platform movement, buttons, switches, levers, inventory management, tools,
damageable items, drawing, UI elements, and more.
As discussed by Q1 A1, Q2 A1 and A3 during the midterm evaluation and Q4 A1-3
in the final evaluation, a stronger connection between technical practices in VR and IDN
storytelling needs to be established earlier on in the semester. Not only will this help
with project scope (Q4 A3), but it may result in interactions that have stronger dramatic
agency and connection to the narrative. IDNs in VR should be presented as case studies
that students can emulate using provided frameworks and templates.
minutes long). These experiences also had multiple branching dialogue options that
expanded the storyworld, and in one case, changed the outcome of the story.
The best student work is presented here with summaries of their narratives as well as
stills. Links to their trailers are also provided.
Corbit. Corbit, in Fig. 1, “takes place in a whimsical world in the center of the earth,
where every cog has their purpose in the machine. Corbit aims to explore existential and
environmental concepts in a fun, hopeful and engaging way.” The interactor embodies
a Cog, part of a system of cogs and gears, that help to keep earth alive while humanity
causes damage. Over the course of five spaces, interactors learn who they are in this
world and what their role is, how humans are destroying the earth, and what they can
do to fix it. In conclusion, the interactor learns that their work is Sisyphean—that the
machine is made for the earth, not humans, and will exist long after they are gone.
Corbit uses digital scenography melded with UI elements to maintain immersion while
supporting the user’s SoP. The narrative is grounded within contemporary issues and
anxieties which works to support SoP and engagement. Corbit effectively utilized the
antechamber VR design method [76] to script the interactor before the experience began.
Not only did the student capably weave diegetic information into the antechamber, but
the interactions they scripted in the space modified the world.
Fig. 1. From left to right: (1) The interactor rides a platform through the world of the machine. It
is a sprawling space of massive scale intentionally designed for immersion. (2) NPCs in the space
provide branching dialogue options and diegetic information. (3) UI to access the IDN content is
rendered in world on screens or consoles that do not break immersion. (4) Scenographic clues are
given to encourage user interactions that impact the world. The trailer is here: https://youtu.be/
QRdEI5mdfss
The Spotted Journey. In this experience, shown in Fig. 2, the interactor inhabits the
role of a bully who is transformed into a gecko for their immoral acts. The student loosely
based the narrative off a myth about the goddess Demeter’s punishment for a man that
mocked her as she ate and drank. The interactor must do good deeds while collecting
crystals from rooms within their own house. Once the interactor is transformed into
a gecko, the bugs and small creatures within the space take the opportunity to scold
472 J. A. Fisher and J. T. Samuels
them for their bad behavior (when large). The Spotted Journey relies strongly on scale
and climbing locomotion to make the player feel small and feckless. As the interactor
goes from room to room, the climbing challenges require larger body movements. Such
movements have a beneficial impact on SoP. NPCs are quick to let the interactor know
how rude they were when they were a fully-sized human inhabiting the space with them.
Fig. 2. From left to right: (1) When the experience starts, the interactor is a young child. They
are then scaled down into a gecko for the experience. (2) Interactors are made to traverse large
rooms to encourage feelings of helplessness. (3) Interactors must climb on furniture to get magical
crystals and do good deeds. (4) At the end of the experience, if the interactor has chosen the right
dialogue options, NPCs that travel with them throughout the experience are present when they are
turned back into humans. The trailer is here: https://youtu.be/X-LLmQ3bkZk
Fig. 3. From left to right: (1) In the antechamber space, interactors are introduced to the world
and are taught how to scan “bytes” to see if they are infected or not. (2) The interactor’s bedroom
showcases the dystopian and dark world of the narrative. The billboard says, “Remember their
Names”. (3) The main interactions occur in the security hall where bytes are scanned. In the
background, an infected byte is escorted away. The interactor hears the byte get deleted in the
distance. (4) Inside of their befriended colleague’s head, the interactor learns that they are infected
and witnesses their friend execute a fellow byte. The trailer is here: https://youtu.be/jXitKdS3o4w
produced a playable IDN in VR. Almost every student met the challenges of the course,
learned a variety of different technical skills, and became more literate in the VR design
and development space. Given that a creative writer, with no prior Unity experience,
produced the most comprehensive and emotionally satisfying experience is a testament to
how scaffolding, templates, and direction can help novices achieve their goals. Templates
and resources from the course are available here: https://github.com/jadlerfisher/IDN-
in-VR-Resources. These resources will allow students to focus more on their stories than
their technical implementation.
This course’s failures fall into two domains. The first domain is connected to IDN and
VR Literacy. Students were introduced to the free experiences available on SideQuest too
late in the semester. This delayed learning and then implementing conventions. Further,
not connecting these same experiences to technical demos made it difficult for students to
connect their own IDNs to VR affordances. In short, teaching narrative game mechanics
came too late. The second domain of failures connects to the scope and variety of work
required to complete a comprehensive and compelling experience. There was simply too
much for a single student. They often felt overwhelmed. Templates and tutorials were
produced on an ad hoc basis to meet needs and skill levels for various narrative game
mechanics. Additionally, introducing the user study prototype pair protocol discussed
by Koenitz and colleagues would have been beneficial both from a research perspective
and to produce better stories [1]. Lastly, only one student produced an IDN wherein
dramatic agency produced a new ending for the story. Most students produced branching
narratives wherein all choices resulted in a common ending. So, while students did gain
a working knowledge of IDN, the majority were unable to produce either IDNs with
multiple endings or kaleidoscopic stories [77].
standardize the discipline’s language, history, and canon of experiences. This is critical
and exciting work to prepare the next generation of immersive storytellers.
References
1. Koenitz, H., Roth, C., Dubbleman, T., Knoller, N.: Interactive narrative design beyond the
secret art status: a method to verify design conventions for interactive narrative. 6, 107–119
(2018). https://doi.org/10.14195/2182
2. Dubbelman, T.: Teaching narrative design on the importance of narrative game mechanics.
In: Suter, B., Bauer, R., Kocher, M. (eds.) Narrative Mechanics: Strategies and Meanings in
Games and Real Life, pp. 79–89. Transcript Verlag, Bielefeld (2021)
3. Daiute, C., Koenitz, H.: What is shared? - A pedagogical perspective on interactive digital
narrative and literary narrative. In: Nack, F., Gordon, A.S. (eds.) ICIDS 2016. LNCS, vol.
10045, pp. 407–410. Springer, Cham (2016). https://doi.org/10.1007/978-3-319-48279-8_37
4. Koenitz, H., Eladhari, M.P.: Challenges of IDN research and teaching. In: Cardona-Rivera,
R.E., Sullivan, A., Young, R.M. (eds.) ICIDS 2019. LNCS, vol. 11869, pp. 26–39. Springer,
Cham (2019). https://doi.org/10.1007/978-3-030-33894-7_4
5. Roth, C., Koenitz, H.: Towards creating a body of evidence-based interactive digital narrative
design knowledge: approaches and challenges. In: AltMM 2017 - Proceedings of the 2nd
International Workshop on Multimedia Alternate Realities, co-located with MM 2017, pp. 19–
24 (2017). https://doi.org/10.1145/3132361.3133942
6. Shibolet, Y., Knoller, N., Koenitz, H.: A framework for classifying and describing authoring
tools for interactive digital narrative. In: Rouse, R., Koenitz, H., Haahr, M. (eds.) ICIDS 2018.
LNCS, vol. 11318, pp. 523–533. Springer, Cham (2018). https://doi.org/10.1007/978-3-030-
04028-4_61
7. Murray, J.T., Johnson, E.K.: XR content authoring challenges: the creator-developer divide.
In: Fisher, J.A. (ed.) Augmented and Mixed Reality for Communities, pp. 245–264. CRC
Press, Boca Raton (2021)
8. Rouse, R.: Media of attraction: a media archeology approach to panoramas, kinematography,
mixed reality and beyond. In: Nack, F., Gordon, A. (eds.) Interactive Storytelling. ICIDS
2016. LNCS, vol. 10045, pp. 97–107. Springer, Cham (2016).https://doi.org/10.1007/978-3-
319-48279-8
9. Association for Research in Digital Interactive Narratives: Committees and Task Forces.
https://ardin.online/committees/. Accessed 7 Sep 2021
10. Koenitz, H.: Design approaches for interactive digital narrative. In: Koenitz, H. (ed.) Inter-
national Conference on Interactive Digital Storytelling, pp. 50–57. Springer, Cham, Nov
2015
11. Koenitz, H.: Five theses for interactive digital narrative. In: Fernández-Vara, C., Mitchell, A.,
Thue, D. (eds.) Interactive Storytelling: 7th International Conference on Interactive Digital
Storytelling, pp. 134–139. Springer International Publishing, Singapore (2014). https://doi.
org/10.1007/978-3-319-12337-0
12. Lee, M.J.W., Georgieva, M., Alexander, B., Craig, E., Richter, J.: State of XR & Immersive
Learning 2021 Outlook Report. Immersive Learning Research Network, Walnut (2021)
13. Immersive Learning Research Network: Immersive Learning Research Network Digital
Library.
14. Grand View Research: Virtual Reality Market Size, Share & Trends Analysis Report By Tech-
nology (Semi & Fully Immersive, Non-immersive), By Device (HMD, GTD), By Component
(Hardware, Software), By Application, And Segment Forecasts, 2021–2028. San Francisco
(2021)
A Proposed Curriculum for an Introductory Course on Interactive Digital Narratives 475
15. Nurbekova, Z., Grinshkun, V., Aimicheva, G., Nurbekov, B., Tuenbaeva, K.: Project-based
learning approach for teaching mobile application development using visualization technol-
ogy. Int. J. Emerg. Technol. Learn. 15, 130–143 (2020). https://doi.org/10.3991/IJET.V15I08.
12335
16. Dubbelman, T.: Narrative game mechanics. In: Nack, F., Gordon, A.S. (eds.) ICIDS 2016.
LNCS, vol. 10045, pp. 39–50. Springer, Cham (2016). https://doi.org/10.1007/978-3-319-
48279-8_4
17. Carey, B.P.: The Architect of Forking Paths: Developing Key Writing Strategies for Interactive
Writers (2018)
18. Murray, J.H.: Hamlet on the Holodeck: The Future of Narrative in Cyberspace. MIT Press,
Cambridge (2017)
19. Murray, J.: Agency. In: Hamlet on the holodeck, pp. 159–189. MIT Press, Cambridge (2017)
20. Mateas, M., Stern, A.: Façade (2005)
21. Michael, J.: Afternoon, a story (1987)
22. Cornell, C.: Save the Date (2013). http://paperdino.com/save-the-date/
23. Ryan, M.L.: Interactive narrative, plot types, and interpersonal relations. In: Spierling, U.,
Szilas, N. (eds.) Interactive Storytelling. ICIDS 2008. Lecture Notes in Computer Science,
vol. 5334, pp. 6–13. Springer, Berlin, Heidelberg (2008). https://doi.org/10.1007/978-3-540-
89454-4_2
24. Ryan, M.: Toward an interactive narratology. In: Avatars of Story, pp. 97–125. University of
Minnesota Press, Minneapolis (1997)
25. Ryan, M.-L.: The text as world: theories of immersion. In: Narrative as Virtual reality 2:
Revisiting Immersion and Interactivity in Literature and Electronic Media, pp. 61–84. JHU
Press, Baltimore (2015)
26. Hameed, A., Perkis, A.: Spatial storytelling: finding interdisciplinary immersion. In: Rouse,
R., Koenitz, H., Haahr, M. (eds.) Interactive Storytelling. ICIDS 2018. LNCS, vol. 11318,
pp. 323–332. Springer, Cham (2018). https://doi.org/10.1007/978-3-030-04028-4_35
27. Engberg, M., Bolter, J.D.: The aesthetics of reality media. J. Vis. Cult. 19, 81–95 (2020).
https://doi.org/10.1177/1470412920906264
28. Sundar, S.S., Kang, J., Oprean, D.: Being There in the Midst of the Story: How Immersive
Journalism Affects Our Perceptions and Cognitions. Cyberpsychology, Behav. Soc. Netw.
20(11), 672–682 (2017). https://doi.org/10.1089/cyber.2017.0271
29. Sallnäs, E.-L.: Effects of communication mode on social presence, virtual presence, and
performance in collaborative virtual environments. Presence Teleoperators Virtual Environ.
14, 434–449 (2005). https://doi.org/10.1162/105474605774785253
30. Quesnel, D., Riecke, B.E.: Awestruck: natural interaction with virtual reality on eliciting awe.
In: 2017 IEEE Symposium on 3D User Interfaces (3DUI), pp. 205–206. IEEE (2017)
31. Facebook: Interactions | Oculus Developers. https://developer.oculus.com/learn/hands-des
ign-interactions/. Accessed 11 July 2021
32. Murray, J.H.: Inventing the Medium (2003)
33. Thon, J.: Immersion revisited: on the value of a contested concept. Extending Exp. Struct.
Anal. Des. Comput. Game Play. Exp. 29–43 (2008)
34. Slater, M., Usoh, M., Steed, A.: Taking steps: the influence of a walking technique on presence
in virtual reality. ACM Trans. Comput. Interact. 2, 201–219 (1995)
35. Soler-Domínguez, J.L., de Juan, C., Contero, M., Alcañiz, M.: I walk, therefore I am: a
multidimensional study on the influence of the locomotion method upon presence in virtual
reality. J. Comput. Des. Eng. 7, 577–590 (2020)
36. Nilsson, N.C., Serafin, S., Steinicke, F., Nordahl, R.: Natural walking in virtual reality: a
review. Comput. Entertain. 16, 1–22 (2018)
37. Boletsis, C.: The new era of virtual reality locomotion: a systematic literature review of
techniques and a proposed typology. Multimodal Technol. Interact. 1, 24 (2017)
476 J. A. Fisher and J. T. Samuels
38. Fisher, J.A., Garg, A., Singh, K.P., Wang, W.: Designing intentional impossible spaces in
virtual reality narratives: a case study. In: Proceedings - IEEE Virtual Reality (2017). https://
doi.org/10.1109/VR.2017.7892335
39. Di Luca, M., Seifi, H.: Locomotion vault: the extra mile in analyzing vr locomotion techniques.
In: Conference Human Factors Computer System - Proceedings (2021). https://doi.org/10.
1145/3411764.3445319
40. Bellard, M.: Environment Design as Spatial Cinematography: Theory and Practice. https://
www.youtube.com/watch?v=L27Qb20AYmc&t=53s. Accessed 3 Jan 2021
41. Taylor, D.: Ten Principles of Good Level Design.
42. Feil, J., Scattergood, M.: Beginning Game Level Design. Thomson Course Technology,
Boston (2005)
43. Blom, K.J., Beckhaus, S.: The design space of dynamic interactive virtual environments.
Virtual Real. 18(2), 101–116 (2013). https://doi.org/10.1007/s10055-013-0232-y
44. Howard, P.: What is Scenography? Routledge, Abingdon-on-Thames (2009)
45. Larionow, D.: Space, light and sound in performances of Leszek M’dzik’s Scena Plastyczna
KUL as the innovation in theatre design, vol. 1, pp. 1–3. Scenography International (2001)
46. Gruber, C.: Scenography of virtual sound-stages. Perform. Res. 18, 1972013https://doi.org/
10.1080/13528165.2013.818338
47. McKinney, J.: Empathy and exchange: audience experience of scenography. In: Kinesthetic
Empathy in Creative and Cultural Practices, pp. 221–235 (2012)
48. McKinney, J.: The role of theatre design: towards a bibliographical and practical accomoda-
tion, vol. 1, pp. 1–13. Scenography International (1998)
49. Dzardanova, E., Kasapakis, V., Gavalas, D.: Affective impact of social presence in immersive
3D virtual worlds. In: 2017 IEEE Symposium on Computers and Communications (ISCC),
pp. 6–11. IEEE (2017)
50. Kilteni, K., Groten, R., Slater, M.: The sense of embodiment in virtual reality. Presence
Teleoperators Virtual Environ. 21, 373–387 (2012). https://doi.org/10.1162/PRES_a_00124
51. Feng, C., Bartram, L., Riecke, B.E.: Evaluating affective features of 3D motionscapes. In:
Proceedings of the ACM Symposium on Applied Perception, pp. 23–30. Association for
Computing Machinery, New York, NY, USA (2014). https://doi.org/10.1145/2628257.262
8264
52. Tagiuri, R.: Movement as a cue in person perception. In: David, H.P., Brengelmann, J.C.
(eds.) Perspectives in Personality Research, pp. 175–195. Springer, Berlin, Heidelberg (1960).
https://doi.org/10.1007/978-3-662-39598-1_9
53. Gorini, A., Capideville, C.S., De Leo, G., Mantovani, F., Riva, G.: The role of immersion and
narrative in mediated presence: the virtual hospital experience. Cyberpsychol. Behav. Soc.
Netw. 14, 99–105 (2011)
54. LiminalVR: Home - Liminal VR. https://liminalvr.com/. Accessed 11 July 2021
55. Murray, J.: Hamlet on the Holodeck: The Future of Narrative in Cyberspace. The MIT Press,
Cambridge (1998)
56. Bolter, J., Gromala, D.: Transparency and reflectivity: digital art and the aesthetics of interface
design. Aesthetic Comput. 7 (2004)
57. Bolter, J.D., Grusin, R.: Introduction: the double logic of remidiation. Remediat. Underst.
New Media 2, 3–15 (1999)
58. Bolter, J.: Transference and transparency: digital technology and the remediation of cinema.
Intermédialités Hist. théorie des arts, des lettres des Tech. Hist. Theory Arts, Lit. Technol.
13–26 (2005)
59. Kuter, U., Yilmaz, C.: Survey methods: questionnaires and interviews. Choos. Human-
Computer Interact. (HCI) Appropr. Res. Methods. (2001)
60. Desurvire, H., El-Nasr, M.S.: Methods for game user research: studying player behavior to
enhance game design. IEEE Comput. Graph. Appl. 33, 82–87 (2013)
A Proposed Curriculum for an Introductory Course on Interactive Digital Narratives 477
dennis@ranj.com
1 Introduction
In recent years, there has been a growing interest in, as well as concern about, the societal
challenge posed by the increasing complexity of the world and in the potential of digital
mediation to address that challenge. In [1], the authors interviewed a variety of experts
and drew stark conclusions about the severity of the challenge and the need to promote
complexity thinking to address it, recommending the use of playable models, as well
as calling for “new narratives”. The examples discussed, however, were serious games,
understood as simulations of complex dynamic systems. These were shown to have
helped people recognise that certain issues were more complex than they had anticipated.
Yet the report did not explain how playable models constituted new narratives. If playable
simulations appeal to a logico-scientific rather than a narrative mode of thinking [2], they
might fail to employ a crucial resource, since simulation without narrative lacks specific
explanatory power [3, 4], and since narrative can organise knowledge into patterns that
are more meaningful [2] and self-relevant [5].
Popular authors [6, 7] have also been emphasising the crucial role of storytelling
in society, pointing out that our ability, as humans, to collaborate in large numbers has
always depended on our ability to imagine and share common stories and storyworlds.
Narrative is more than an individual cognitive process, being, as cognitive narratologist
David Herman has pointed out in [8], “anchored in collaborative discourse processes
- how the mind is grounded in participants’ relations with one another and with their
surrounding social and material environment” (p. 320). In this he has endorsed folklorist
and anthropologist Richard Bauman’s analytic strategy [9], based on ethnographic stud-
ies of the performative aspects of oral storytelling, that took it to be constituted as an
“indissoluble unity of text, narrated event and narrative event” (p. 7), in which story-
world events (narrated events) are emergent in performance. Thus, cognitive narrative
processes cannot fruitfully be understood to be uncoupled from their social and mate-
rial environment, or indeed from their deep relation to the practices and communicative
and social functions of storytelling. Complexity thinking, as a cognitive process, might
require more than just individual shifts brought about by engagement with playable
models. It is common stories and storyworlds that bind our individual experiences of the
natural, social, cultural, or technological environment together into a meaningful shared
experience.
Reviewing [1] and sharing the concerns it raised, Knoller [10] has therefore argued for
a more explicitly narrative approach, and suggested that interactively narrating complex-
ity may be more likely to further an understanding of complexity, and that this might be a
challenge for Interactive Digital Storytelling and Narrative (IDS&N). Narrative Userly
Texts [10, 11] in various forms - narrative serious games, interactive documentaries,
digitally immersive theatre, as well as other forms - can develop to become the way
humans make (common) sense of the world, and of being and acting within and upon it,
through a developed narrative understanding of complex systems. Creators of such sys-
tems can draw on the affordances of digital media that distinguish them from pre-digital
media [12]: the ability to store vast amounts of data, to disseminate narratives instanta-
neously around the world, to allow audiences to participate in the narrative experience
as interactors and to try out different options. This may make it possible to represent,
communicate, experience, and grasp complex systems, dynamics and processes, and to
experience how to be agents within them (individually as well as collaboratively) in
ways that are distinct from preceding media.
This has led Knoller [10] to reframe the societal challenge for IDS&N as the potential
to design and employ narrative userly texts as semiotic scaffolds for a cognitive reduction
of complexity (p. 107) - a step further than the realisation, scaffolded by playable models,
that something is complex [1]. This potential, as well as the urgency of the societal
challenge, have gained further recognition with the creation of the EU-funded INDCOR
network [13], whose aim is “the interdisciplinary study of the potential interactive digital
narrative has as a means to addressing complexity as a societal challenge by representing,
experiencing and comprehending complex phenomena”.
Given the potential, urgency, and institutional recognition, it is both necessary and
possible to concentrate efforts on looking for evidence for the effectiveness of such
an approach. In this paper, we begin to address this by focusing on the conceptual
challenge of describing complexity. Such description is necessary as a preliminary step
480 N. Knoller et al.
a. the relative scale of outcomes (small actions can lead to large effects);
b. the cascading, ripple or 2nd order effects of an action, beyond a local effect;
c. the time scale: changes can be immediate, delayed or develop over a period.
To measure the level of learners’ understanding, for each of the six complex-
systems ideas, Yoon further used a four-level scale of complexity that she had devel-
oped and validated earlier [19], ranging from “completely clockwork” to “completely
complex”. “Clockwork” describes systems as linear, single-cause, non-networked, cen-
tralised, static or. “Complex” describes systems as nonlinear, networked, multiple-cause,
dynamic, decentralised, or non-deterministic. This results in a general 6 by 4 matrix that
can be used to describe:
(i) The six complexity dimensions of any complex system S. These correspond to the
primary complexity in the complexity triad mentioned in the introduction.
(ii) Responses by learners to questions asking them to describe a complex system S.
These correspond to the tertiary complexity in the complexity triad.
Measuring the distance between the two descriptions can be considered as an approx-
imation of the cognitive reduction of the complexity of system S: the smaller the distance,
the more effective the cognitive reduction.
Alert readers will have noticed that a component of the complexity triad is missing:
(ii) The secondary complexity, that of the mediating artefact, the userly text.
The learning progression framework is media agnostic. It is not interested in the
media specificity of the learning materials or methods, and does not discuss whether
the media used have anything to do with how complexity is being represented or under-
stood. But the mediatic complexity of artefacts is significant. Interactive Digital Narrative
artefacts have not yet been studied to the extent that we possess a wealth of empirical
evidence to support any theory about how they might function, or a validated method-
ology according to which such evidence might be collected and assessed so that it can
then be used to inform theory and even provide design guidelines.
While it allows for a comparison primary and tertiary complexity and thus for a mea-
surement of learning outcomes, the original framework cannot provide us with insight
into how this evident effectiveness is scaffolded by mediating artefacts. In the next section
we will discuss the Userly Text model, how it addresses the mediation and scaffolding
of complexity, and how this leads us to extend the framework.
482 N. Knoller et al.
The Encoded Storyworld. This construct has three levels of abstraction: (1) The most
basic elements are the units of diegetic materials (locations, agents, and events) and form
(units of media, either remediated assets such as media files or digitally native units of
code). (2) A middle level establishes Discourse Relations between elements: causality,
temporal order, spatial location, formal relations such as symmetry, colour or rhyming,
semantic relations, statistical pattern matching and so on. (3) Discursive strategies build
on the relations of the second level. The narrative strategy, which emphasises causal and
spatiotemporal relations between events involving agents in locations, is one possibility,
but other storytelling strategies are available. In film, for example, materials may be
organised along the linear temporal dimension according to the discursive strategy of
narrative form (establishing relations of space, time, and causality), or thematically, or
alphabetically, or according to formal-abstract qualities such as rhythm and composition,
or to construct a rhetorical argument [21].
The Interaction Model. The second construct manifests as the hardware user-input
and user-output components of the system, as well as software driving the hardware
and interpreting its operations to produce affordances for interaction, and software that
models the user(s) and interpreting their operation, and thus translating userly perfor-
mance to encoded-storyworld terms. The authoring of an interactive digital storytelling
experience also entails the authoring of the interaction model. The interface, which medi-
ates the semiotic and affective feedback loops between the encoded storyworld and the
user’s body-mind, is very much part of the userly text. The creator of such an experience
is authoring a complex system of communicative exchanges. Authoring an interaction
model entails – even if implicitly - a choice of hardware platform that determines which
information can be communicated between the human system (and which parts of the
human body are treated as the human system’s interface) and the computer system(s).
The Complexity Analysis Matrix 483
Authoring the interaction model also entails structuring the software procedures that
model the possibilities of interaction that define and constrain the potentials of userly
performance.
3.2 Userly Performance and Complexity: Skill, Repetition, and the Hermeneutic
Spiral
Userly performance is the specific mode of reception that userly texts afford. It is this
procedural, embodied-cognitive activity that can activate the potential of narrative userly
texts to offer scaffolding for cognitively reducing complexity.
The term userly performance welds together senses of the performative or perfor-
mantial [22] in both the performing arts and sociology. Artistic performance connotes
the execution of a score or a script, and it requires skill, which is developed through repe-
tition. The skill involved in userly performance is not merely sensorimotor in nature, but
can also be hermeneutic, connoting the artistic sense of performance as interpretation
(which can also develop further into expressive interpretation).
Narrative userly texts can be structured to encourage replay, and thus a hermeneutic
spiral: a process of gradually accumulated double-interpretation, in which the interactor
needs to make sense both of the encoded storyworld and of the interaction model, as well
as of both the current plot they are performing and the relation of this plot-performance
to other plot-performances. Whereas traditional narratology speaks of a narrative circle
[23], and traditional hermeneutics of a hermeneutic circle [24] moving between parts
and the whole text, the metaphor of the spiral reflects the fact that userly texts appear as
materially different every time they are “read” (in the semiotic sense, which is extended
here to refer to the performance of interactive reception). As has also been theorised in
Koenitz’s SPP model [25], plots, as well as their performances, exist in the system as
potentials, requiring instantiation contingent on input from a user and further procedural
processing, enacting a variability of performed instantiations.1 This flexible variability of
outputs is constitutive of the userly text. A hermeneutic spiral might allow interpretation
to connect the different performances and outputs together into a meaningful subjective,
enactive representation of systemic complexity.
First- and Second-Order Subject Positions: Observation vs. Enaction. The distinc-
tion between the encoded storyworld and the interaction model foregrounds a distinc-
tion within narrative userly texts for complexity between first-order and second-order
complex representations of complex systems, which entail different subject positions.
1 The encoded storyworld is similar to Koenitz’ protostory [25], and differs in its internal abstrac-
tions, since the userly text model attempts to also account, through the added strategic abstrac-
tion level, for the instantiation of non-narrative discursive strategies and for the potential of a
text to shift between different discursive strategies (for example, from chronological recitation
to stream of consciousness) in response to patterns of user input. This aspect is particularly
relevant for complexity representation in a post-PC landscape because it allows for a descrip-
tion of how different interaction models might be coupled with one encoded storyworld, not
just within a single artefact (shifting strategies, but now of interaction), but also across distinct
artefacts (as could be the case in a transmedial implementation).
484 N. Knoller et al.
Considering the above, the original framework proposed by Yoon et al. needs to be
extended to account for some additional complexities that narrative userly texts can
represent. The original framework, accounting for natural-scientific complex systems,
addresses learners only as observers. How do we account for the enactive position and
the second-order consequences made possible by narrative userly texts?
2 See [30, 31] for different takes on double hermeneutics in games and [32] for an extended
discussion of double hermeneutic circles and spirals.
The Complexity Analysis Matrix 485
complex skills training, and represent complexities using additional types of knowledge
[33]. The original framework considers only declarative knowledge, knowledge about
a complex system, observational knowledge that can be described and reproduced using
statements in language. Enactment, however, also concerns the procedural knowledge
type, knowledge how, knowledge-in-use, which a learner may be able to successfully
apply despite not being able to consciously verbalise.
The implications of the userly text model for extending the original framework are
ultimately not limited to the description of the secondary complexity in the triad, that
of the mediating userly text itself. Depending on the type of system and the mode of
instruction, the description of the tertiary, subjective complexity might need to incorpo-
rate skills, understood as procedural knowledge. Likewise, the description of the primary
complexity, the one to be reduced, requires an extension to describe the subjective and
reflexive roles of agents within the complex system - particularly when interactively
narrating complex systems involving human agents.
After examining the limitations of the original matrix and the requirements to extend
it, we decided to split each of the six ideas between first-order complexity and second-
order effects, both of which can then independently be described as completely clock-
work, somewhat clockwork, somewhat complex or completely complex. This results in
a double-matrix of 6 complexity dimensions (in total – 12 dimension) by 4 levels of
complexity. The matrix can be used, if necessary, for each of the components of the
complexity triad. It can initially be used to describe the primary complexity of the phe-
nomenon. This may be useful as a benchmark to then critique or validate the quality of
a userly text as a representation of the primary complexity3 . Once that is established, it
becomes possible to create a specific coding matrix for a narrative userly text to evaluate
the tertiary, subjective complexity, and thus to measure its effectiveness in achieving a
cognitive reduction of complexity. This can be achieved by finding correlations between
(1) the design features that distinguish A/B versions, (2) the actual performance of the
learner (measurable, for example by the system or through observation), (3) the sub-
jective experience of the learner (using either objective measures or self-reporting), (4)
the resulting declarative knowledge outcomes (using common testing methods), and (5)
the procedural knowledge outcomes (measurable using a post-test separate simulation
or role playing assessment task) – all of which can then be described in terms of both
first- and second-order complexity understanding, on a 4-level scale.
Acknowledgments. This research was supported by a grant from the Dutch Creative Industries
Fund.
3 Visit [14] for further resources, including a filled-out matrix describing the serious training
game Mission Zhobia [34].
486 N. Knoller et al.
References
1. Rajeski, D., Chaplin, H., Olson, R.: Addressing complexity with playable models. Paper,
Science and Technology Innovation Program. Wilson Center, Washington DC (2015)
2. Bruner, J.: The narrative construal of reality. In: Bruner, J. (ed.) The Culture of Education.
Harvard University Press, Cambridge (1996)
3. Hayles, N. K.: Making the cut: the interplay of narrative and system, or what systems theory
can’t see. In: Cultural Critique, 30: The Politics of Systems and Environments, Part 1 (Spring
1995), pp. 71–100. University of Minnesota Press (1995)
4. Walsh, R.: Narrative theory at the limit. In: Middeke, M., Reinfandt, C. (eds.) Theory Matters:
The Place of Theory in Literature and Cultural Studies, pp. 265–279. Palgrave-Macmillan,
London (2016)
5. Ricoeur, P.: Time and Narrative. University of Chicago Press, Chicago (1984–1988)
6. Gottschal, J.: The Storytelling Animal. Houghton Miffling, Boston (2012)
7. Harari, Y.N.: Sapiens, a Brief History of Humankind. Harvill Secker, London (2014)
8. Herman, D.: Storytelling and the sciences of mind: cognitive narratology, discursive
psychology, and narratives in face-to-face interaction. Narrative 15(3), 306–334 (2007)
9. Bauman, R.: Story, Performance, and Event: Contextual Studies of Oral Narrative. Cambridge
University Press, Cambridge (1986)
10. Knoller, N.: Complexity and the userly text. In: Grishakova, M., Poulaki, M. (eds.) Narrative
Complexity: Cognition, Embodiment, Evolution, pp. 98–120. Nebraska University Press,
Lincoln (2019)
11. Knoller, N.: The expressive space of IDS-as-art. In: Oyarzun, D., Peinado, F., Young,
R.M., Elizalde, A., Méndez, G. (eds.) ICIDS 2012. LNCS, vol. 7648, pp. 30–41. Springer,
Heidelberg (2012). https://doi.org/10.1007/978-3-642-34851-8_3
12. Murray, J.H.: Inventing the Medium. MIT Press, Cambridge and London (2012)
13. INDCOR – COST Action CA18230 – Interactive Narrative Design for Complexity Repre-
sentations. https://www.indcor.eu. Accessed 15 Oct 2021
14. Mission Zhobia: Validating a Narrative Game for Complexity. https://www.userly.tech/res
earch/zhobia/. Accessed 5 Oct 2020
15. Knoller, N.: The complexity triad and two+ systemic models of IDS/N (2020). https://doi.
org/10.13140/RG.2.2.14721.58729. Accessed 07 July 2021
16. Gee, J.P.: What Video Games Have to Teach Us About Literacy and Learning. Palgrave
Macmillan, New York (2003)
17. van’t Riet, J., Meeuwes, A.C., van der Voorden, L., Jansz, J.: Investigating the effects of a
persuasive digital game on immersion, identification, and willingness to help. Basic Appl.
Soc. Psychol. 40(4), 180–194 (2018)
18. Yoon, S.A., Goh, S.-E., Yang, Z.: Toward a learning progression of complex systems
understanding. Complicity: Int. J. Complexity Educ. 16(1), 1–19 (2019)
19. Yoon, S.A.: An evolutionary approach to harnessing complex systems thinking in the science
and technology classroom. Int. J. Sci. Educ. 30(1), 1–32 (2008)
20. Barthes, R.: S/Z. Hill and Wang, New York (1974)
21. Bordwell, D., Thompson, K.: Film Art: and Introduction, 5th edn. McGraw-Hill, New York
(1997)
22. Ryan, M.L.: Multivariate narrative. In: Schreibman, S., Siemens, R., Unsworth, J (eds.) A
Companion to Digital Humanities. Blackwell, Oxford (2004). http://www.digitalhumanities.
org/companion/. Accessed 30 July 2021
23. Bal, M.: Narratology, 2nd edn. University of Toronto Press, Toronto, Buffalo and London
(2004)
24. Gadamer, H.G.: Truth and Method. Crossroads, New York (1985)
The Complexity Analysis Matrix 487
25. Koenitz, H.: Towards a specific theory of interactive digital narrative. In: Koenitz, H., Ferri,
G., Hahr, M., Sezen, D., Sezen, T.İ. (eds.) Interactive Digital Narrative - History, Theory and
Practice, pp. 91–105. Routledge, New York and London (2015)
26. Clarke, B.: Heinz von Foerster’s demons - the emergence of second-order systems theory. In:
Clarke, B., Hansen, M.B.N. (eds.) Emergence and Embodiment: New Essays on Second-Order
Systems Theory, pp. 34–61. Duke University Press, Durham and London (2009)
27. Bredillet, C.: A discourse on the non-method. In: Drouin, N., Muller, R., Sankaran, S. (eds.)
Novel Approaches to Organizational Project Management Research: Translational and Trans-
formational [Volume 29: Advances in Organization Studies], pp. 56–94, Copenhagen Business
School (2013)
28. Giddens, A.: Social Theory and Modern Sociology. Stanford University Press, Stanford (1987)
29. Tsoukas, H., Hatch, M.J.: Complex thinking, complex practice: the case for a narrative
approach to organizational complexity. Hum. Relat. 54(8), 979–1013 (2001)
30. Karhulahti, V. M.: Double fine adventure and the double hermeneutic videogame. In: Fun and
Games’ 2012, Toulouse, France—4–6 September 2012, pp. 19–26. ACM (2012)
31. Roth, C., van Nuenen, T., Koenitz, H.: Ludonarrative hermeneutics: a way out and the narrative
paradox. In: Rouse, R., Koenitz, H., Haahr, M. (eds.) ICIDS 2018. LNCS, vol. 11318, pp. 93–
106. Springer, Cham (2018). https://doi.org/10.1007/978-3-030-04028-4_7
32. Knoller, N.: How hermeneutic spirals may reduce complexity to narrative schemata - expand-
ing on “complexity and the userly text” (2021). https://hal.archives-ouvertes.fr/hal-03254233.
Accessed 15 Oct 2021
33. de Jong, T., Fergusson-Hessler, M.G.M: Types and qualities of knowledge. Educ. Psychol.
31(2), 105–113 (1996)
34. &Ranj: Mission Zhobia - Winning the Peace [online game]. PeaceNexus (2017)
Interactive Digital Narratives (IDN)
as Representations of Complexity: Lineage,
Opportunities and Future Work
mirjam@dsv.su.se
1 Introduction
to have societal impact. This is the interdisciplinary endeavor of the COST Action IND-
COR1 . In this overview paper, we will focus on conceptual aspects, describe the history
of complexity and its representations, the need for new narratives, provide a conceptual
framing and suggest focal points for future work.
Complexity as a concern for scholarly enquiry emerged in the natural sciences from
the realization that the Newtonian, mechanistic way of considering the universe is not
sufficient to explain the complex world around us - an insight reflected at least since
Einstein’s Relativity theory [2], further developed in quantum physics (most famously
explained in Schrödinger’s ‘cat’ thought experiment [3]), and more generally applied
in approaches such as cybernetics [4, 5], system theory [6–8] and chaos theory [9–11].
An interactive map of complexity sciences is provided by Castellani [12]. What these
perspectives share is the insight that many phenomena cannot be explained with simple
linear one-to-one mappings of cause and effect, that combinatorics, unpredictability and
randomness are important factors. This understanding is at the heart of research on com-
plexity. Accordingly, [13] defines complexity as resulting “from the inter-relationship,
inter-action and inter-connectivity of elements within a system and between a system
and its environment” (p. 1). In addition, the behavior across time of such systems [14]
is a crucial aspect when the focus is on systemic change.
Complexity as an academic field is an umbrella term for a number of different
perspectives and approaches, e.g. in the guise of complexity sciences concerned with
problems in Physics and other natural sciences [14, 15] or as complexity theory (for an
overview see [16]). One exemplar approach in complexity theory resulted from a multi-
disciplinary collaboration pursuing lines of enquiry characterized by emergent complex
behaviors of multi-level systems [13]. Concrete applications of complexity sciences
include weather forecasts [17], the analysis of ecological systems [18] and accident
investigations [19]. Resilience theory in ecology is based on complexity science [20]
while complexity science is being suggested as an alternative approach to traditional
economics [21], especially for forecasting [22] as well as a crucial element in order to
reach the goals of the UN 2030 Agenda [23].
However, the recognition of complexity is not limited to the natural sciences. Many
developments in the 20th century in the social sciences can be understood similarly as
ways to acknowledge and analyze complexity. In this vein we can for example under-
stand approaches such as psychoanalysis [24], feminism [25], system theory applied to
society [26], queer theory [27], critical race theory [28] and intersectionality [29]. The
latter, a method of analysis that takes discrimination as a combination of multiple factors,
including gender, race, age, class, sexuality, physical appearance, nationality and ability,
is clearly an example of complexity - in this instance of the complex societal combina-
torics behind oppression and discrimination. Earlier, the advent of psychoanalysis was
predicated on the complex relationship between the conscious and the unconscious [30],
while feminism [31] and queer theory [27] investigate the complexity of gender relations
1 https://indcor.eu.
490 H. Koenitz et al.
and gender roles [32, 33]. Critical race theory started with legal scholars investigating
the complex relationship between discrimination and laws in the United States [28].
Concrete applications of these approaches are in national and internal policies and laws
to support equality, diversity, and assure equity, but also in movements such as Black
Lives Matter [34] and Fridays for Future [35]. Another important application is found
in psychotherapy [36].
Equally, complexity is integral to many humanities approaches. For example, post-
structuralism has left behind the fixed structural analysis of literary texts and other phe-
nomena [37], while phenomenology [38] concerns itself with the relationship between
body and mind, leading to recent discussions on embodiment [39]. Postmodernism [40]
is the recognition that the unifying grand narratives of old (nation, religion, class etc.)
no longer represent a complex reality [41]. These aspects are captured in concepts such
as amalgam, dispositive and heterotopia. Complexity is thus an issue that all academic
disciplines have been concerned with for a considerable time.
3 Depictions of Complexity
The arts were equally affected by the scientific developments at the turn of the last
century, but also by societal changes due to increased industrialization, mass culture
and world events such as both world wars, the Great Depression and the holocaust.
Different art movements found new approaches to express the complexity of contempo-
rary human experience. Examples in this vein include surrealism (meant to express the
unconscious), gothic novels and magic realism (expressing the supernatural), abstract
art and ‘new music’ (expressing sentiments rather than depicting objects) postmodern
literature (expressing the multiplicity of meanings), the theatre of the absurd (expressing
the complex relationship between cultural veneer and human nature) as well as partici-
patory forms such as Happenings [42] and participatory theater which put the complex
relationship between planned event and emerging participation front and center. Later
examples include cybernetic art [43] and system art.
Non-fiction forms of expression have so far struggled to find ways to express complex-
ity. Instead, documentary filmmaking and modern journalism established practices to
improve the truthfulness and objectivity of reporting – for example, by clearly distin-
guishing opinion from “objective” reporting, by attempting to minimize the impact of
the documentary filmmaker (cf. in the ‘cinema verite’ movement [44]) and by improv-
ing ways of verifying the depicted content through fact-checking. More recently, linked
sources, multimedia content, infographics and interactive databases became standard.
However, even with these additions, non-fiction forms are still focused on linear commu-
nication, on reducing the complexity so it can be expressed with the established means
of the newspaper article and the TV broadcast. A concrete instance of this overarching
issue is in what Widholm and Appelgren [45] have termed the “data journalism paradox”,
the difficulty of connecting the individual perspectives enabled by interactive forms of
narratives with journalistic practices predicated on collective representations.
Interactive Digital Narratives (IDN) as Representations of Complexity 491
strengthen democracy. Systemic thinking has to become the norm and unduly simplistic
narratives have to be branded as inadequate and thus problematic, regardless from which
ideological side they come from.
To understand the new narratives of IDN, a model is necessary that emphasizes the speci-
ficity of interactive narratives. Koenitz’ SPP model (system, process, product) [51] does
that, understanding a dynamic system as the central element, and describing the interac-
tive process which results in instantiated narrative products. The SPP model includes the
notion of a protostory - a space of potential narratives combining fixed and dynamic ele-
ments. This kind of narrative design can feature complexity in the form of an underlying
network of rules and contrasting claims, implementing for example the initial conditions
and backstories in cultural heritage [52], or the rules of a particular political system. The
engagement of the interactor with the dynamic system of the protostory in the process
moves the narrative forward, facilitating the personal interpretation of the narrative expe-
rience. The SPP model has been related to three separate aspects of complexity [53], that
of environment (the real-world complexity implemented in the protostory), of represen-
tation (the complexity of the artifact) and of message (the complexity of interpretation
and interactors’ actions resulting in recorded playthroughs and retellings [54]) and is a
basis of INDCOR’s effort in creating a shared vocabulary [53]. IDN afford multilinear
and multi-perspective narratives (including conflicting views) that trigger choices and
evoke audience’s reflections. through their ‘double hermeneutic’ of interpreting both the
current narrative trace and their opportunities for interaction [55, 56]. The increase in
Interactive Digital Narratives (IDN) as Representations of Complexity 493
1. There are scientific ways to handle complexity, but these need to be translated to
make them understandable to general audiences.
2. IDN has the potential to do that, once we understand its specific affordances, in
particular its ability to represent dynamic systems
3. In order to use IDN as a tool in society to understand complexity, audience education
is needed
494 H. Koenitz et al.
4. For journalism to seriously pick up IDN, journalists need training in the new form,
while new workflows as well as business models need to be established
5. Research has an important role in supporting the development and adoption of IDN.
The problem space highlights aspects that need scholarly attention, in particular the
design of IDNs (approach, methods and building blocks), evaluation to verify effective-
ness, professional training of creators as well as the education of both adult audiences
as well as school-age children.
In terms of design, an adequate paradigm is system building [64], improving upon
‘Newtonian mechanics’ in game design, as exemplified in the popular MDA model
(mechanics, dynamics, aesthetics) [65] as well as Brenda Romero’s pronouncement of
the mechanics as the message [66]. Conversely, narrative design has often been misunder-
stood in a mechanical way, as the combination of fixed narrative forms with interactivity
(i.e. cinematic cutscenes in games). With the dynamic systems of IDN, new challenges
arise, as the creator is now the designer of a procedural and participatory system. IDN
design thus means to accommodate audiences’ new role as interactors as well as the
unpredictability that is introduced by unexpected combinatorics and interactors’ behav-
ior [67]. In addition, IDN design means to consider the different aspects of complexity as
shown in Fig. 1. The development of specific evaluation methods to verify the effective-
ness of IDN complexity designs (e.g. the further development of Roth’s toolbox [68]) is
an important aspect of IDN design research.
The implementation of a system building paradigm in actual production, especially
when it comes to integrating into the workflows of journalism organizations pose con-
siderable challenges. Some steps in this regard have been taken, for example, in the
Interactive Digital Narratives (IDN) as Representations of Complexity 495
JOLT project3 which investigated the combination of game design and journalism and
collected resources on the topic. Of particular concern in this regard is the tendency
documented in the project to see journalistic ‘interactives’ mostly as add-ons to ‘normal
journalistic activity’, as a means to drive subscribers to newspapers’ websites. Instead,
IDN design needs to become a standard journalistic activity which requires (re-)training
of journalists as IDN system builders. Else, there is a danger that interactive forms are
relegated to a position of cheap gimmick, a possible development Bogost et al. have
been warning about for newsgames [69] (also see [70]). In this regard IDN, with its
focus on narrative, has the potential to bridge the divide between games and traditional
news stories. Furthermore, as a multi-linear and multivariate form, IDN can also address
shortcomings of newsgames, in particular a forced “quick closure and a problematic
binary (good versus bad behavior)” [71] as well as a tendency of ‘wanting to win’.
Instead, IDN can focus on complexity. As there is also an ethical dimension in terms
of representing complexity as a contribution for public discourse, the development of
a code of conduct, similar to the ones for journalism, is an important topic for further
investigation. Another necessary aspect is the consideration of how the production of
IDN complexity products will be financed - possible solutions include public funding
as a service akin to public broadcasting, but also business models similar to the ones
used by commercial video games or commercial newspapers. Finally, the education of
audiences about IDN is an area in need of attention. IDN can be used in schools to teach
about complex subjects, but there is also the need for high-profile critique (similar to
film critics) educating adult audiences about IDN.
6 Conclusion
In this paper, we traced the lineage of complexity and discussed the need for complexity
representations in democratic societies. Then, we explained the potential of IDN in
representing complexity, applying the SPP model and its emphasis on dynamic systems
and interactive processes. After describing a range of examples, we discuss the problem
space of IDN and complexity, identifying areas for future work in order to realize the
potential of IDN as a means to help citizens in understanding the many complex issues
we are faced with - from global warming to issues of personal mental health.
References
1. Murray, J.H.: Hamlet on the Holodeck: the Future of Narrative in Cyberspace. Free Press,
New York (1997)
2. Einstein, A.: Die Grundlage der allgemeinen relativitätstheorie. Analen Phys. 49, 769–822
(1916)
3 https://edspace.american.edu/jolt/.
496 H. Koenitz et al.
32. Haraway, D.: Situated knowledges: the science question in feminism and the privilege of
partial perspective. Fem. Stud. 14, 575 (1988)
33. Butler, J.: Gender Trouble. Routledge, London, New York (1999)
34. Chase, G.: The early history of the black lives matter movement, and the implications thereof.
Nev. LJ 18, 1091 (2017)
35. Spannring, R., Hawke, S.: Anthropocene challenges for youth research: understanding agency
and change through complex, adaptive systems. J. Youth Stud., 1–17 (2021). https://doi.org/
10.1080/13676261.2021.1929886
36. Coburn, W.J.: Attitudes in psychoanalytic complexity: an alternative to postmodernism in
psychoanalysis. In: Beyond Postmodernism, pp. 195–212. Routledge (2013)
37. Dillon, M.: Poststructuralism, complexity and poetics. Theory Cult. Soc. 17, 1–26 (2000)
38. Husserl, E.: The Idea of Phenomenology. Springer, Heidelberg (1964)
39. Csordas, T.J.: Embodiment and cultural phenomenology. na (1999)
40. Lyotard, J.-F.: The Postmodern Condition. University of Minnesota Press, Minneapolis (1984)
41. Cilliers, P.: Problems with representation. In: Complexity and Postmodernism, pp. 68–98.
Routledge (2002)
42. Kaprow, A.: “Happenings” in the New York scene. In: The New Media Reader, pp. 1–6. MIT
Press (2003)
43. Ascott, R.: The cybernetic stance: my process and purpose. Leonardo 1, 105 (1968)
44. Mamber, S.: Cinema Verite in America: Studies in Uncontrolled Documentary. Lightning
Source (1974)
45. Widholm, A., Appelgren, E.: A softer kind of hard news? Data journalism and the digital
renewal of public service news in Sweden. New Media Soc. 1461444820975411 (2020)
46. Habermas, J., Lennox, S., Lennox, F.: The public sphere: an encyclopedia article (1964). New
German Critique 49–55 (1974)
47. Rejeski, D., Chaplin, H., Olson, R.: Addressing Complexity with Playable Models. Wilson
Center (2015)
48. Rosen, R.: On complex systems. Eur. J. Oper. Res. 30, 129–134 (1987)
49. Luhman, J.T., Boje, D.M.: What is complexity science? A possible answer from narrative
research. Emerg. J. Complexity Issues Organ. Manag. 3, 158–168 (2001)
50. Uprichard, E., Byrne, D.: Representing complex places: a narrative approach. Environ. Plan
A 38, 665–676 (2006)
51. Koenitz, H.: Towards a specific theory of interactive digital narrative. In: Interactive Digital
Narrative: History, Theory, and Practice. New York (2015)
52. Vanoverschelde, F.: No Story without a backstory: the role and importance of the backstory
in an augmented reality application for cultural heritage. Presented at the Proceedings of the
8th International Workshop on Narrative and Hypertext (2019)
53. Koenitz, H., Eladhari, M.P., Louchart, S., Nack, F.: INDCOR white paper 1: a shared
vocabulary for IDN (Interactive Digital Narratives) (2020)
54. Eladhari, M.P.: Re-tellings: the fourth layer of narrative as an instrument for critique. In:
Rouse, R., Koenitz, H., Haahr, M. (eds.) ICIDS 2018. LNCS, vol. 11318, pp. 65–78. Springer,
Cham (2018). https://doi.org/10.1007/978-3-030-04028-4_5
55. Karhulahti, V.-M.: Double fine adventure and the double hermeneutic videogame. Presented
at the Foundations of Digital Games, New York, New York, USA (2012)
56. Roth, C., van Nuenen, T., Koenitz, H.: Ludonarrative hermeneutics: a way out and the narrative
paradox. In: Rouse, R., Koenitz, H., Haahr, M. (eds.) ICIDS 2018. LNCS, vol. 11318, pp. 93–
106. Springer, Cham (2018). https://doi.org/10.1007/978-3-030-04028-4_7
57. Mateas, M.: A preliminary poetics for interactive drama and games. Digit. Creat. 12, 140–152
(2001)
58. Duijn, M., Wolting, F.: Co-director documentary. In: Pallotta, T. (ed.) Co-Director Documen-
tary: Last Hijack Interactive (2014)
498 H. Koenitz et al.
1 Narrative Instruments
virtuosically by more practiced players; may be played solely for the player’s own
enjoyment, or for a wider audience; are often played as part of a larger ensemble,
in concert with other instruments; and may be modified or creatively misused
by their players to achieve novel or unexpected effects.
Why instruments? We take inspiration in the use of this analogy from several
other scholars who have tried to characterize what makes instruments special—
distinguishing them from tools on one side and from toys on the other. Writing
in the context of creativity support tools [31], or computational tools intended to
support human creative practices, Nakakoji [23] contends that some creativity
support systems may be better characterized as instruments. For Nakakoji, a
creativity support system may be more of an instrument than a tool if it is often
used playfully and if its designers prioritize the creation of a particular user
experience over maximal efficiency. Tanaka [34] further unpacks the distinction
between instruments and tools, suggesting that musical instruments succeed not
by maximizing the efficiency of musical creation, but by contributing a particular
desirable “personality” or “voice” to the music they are used to create:
The term tool implies that an apparatus takes on a specific task, utilitarian
in nature, carried out in an efficient manner. A tool can be improved to
be more efficient, can take on new features to help in realizing its task,
and can even take on other, new tasks not part of the original design
specification. In the ideal case, a tool expands the limits of what it can do.
It should be easy to use, and be accessible to a wide range of naive users.
Limitations or defaults are seen as aspects that can be improved upon.
A musical instrument’s raison-d’etre, on the other hand, is not at all utili-
tarian. It is not meant to carry out a single well defined task in the way that
a tool is. Instead, a musical instrument often changes context, withstand-
ing changes of musical style played on it while maintaining its identity. A
tool gets better as it attains perfection in realizing its tasks. The evolu-
tion of an instrument is less driven by practical concerns, and is motivated
instead by the quality of sound the instrument produces. In this regard, it
is not so necessary for an instrument to be perfect as much as it is impor-
tant for it to display distinguishing characteristics, or “personality”. What
might be considered imperfections or limitations from the perspective of
tool design often contribute to a “voice” of a musical instrument.
This argument for instruments as succeeding or failing on the basis of the
characteristic voice they provide may help to explain why games that exhibit
strong, recurring narrative texture across multiple playthroughs—such as the
recurring “gradual rise followed by sudden precipitous decline” arc of many
Dwarf Fortress stories—do not disqualify these games from use by players as
story-making tools. What might be a weakness from a perspective that privileges
a tool’s generality—the way that these stories bend characteristically toward
disaster—may actually represent a key desirable attribute of Dwarf Fortress as
an instrument. From the narrative instruments perspective, the perceptibility of
an instrument’s grain in the stories that it is used to create marks not a failure
of generality, but a success of voice.
Toward Narrative Instruments 501
2 Case Studies
To explore the implications of treating IEN systems as narrative instruments,
consider the following brief case studies of narrative instruments in action.
1
Though the systems Wardrip-Fruin highlights here are textual, they are not narra-
tive; therefore we depart from his term in attempting to characterize the class of
instrumental playable systems that produce narrative structure as narrative, rather
than textual, instruments.
502 M. Kreminski and M. Mateas
through the simulated world, they may speak to any of the simulated charac-
ters they encounter; during a conversation with the participant, these characters
are embodied by the actor, who improvises a personality and dialogue for each
character based on the character’s role and traits within the simulation. During
conversation, the wizard provides the actor with a live feed of relevant infor-
mation about the identity of the character they are playing and the state of
the simulated storyworld—including the network of social relationships and the
history of the town’s development—with the goal of subtly directing the partic-
ipant toward discovery of narratively charged situations currently active within
the town.
The Bad News performers make use of at least three distinct narrative instru-
ments: the Talk of the Town simulation, which is run at the start of a performance
to generate the storyworld in which the performance will take place; the “wizard
console”, a command line-based sifting tool used to investigate the storyworld,
operated backstage by the wizard during a performance; and the actor interface
operated by the actor, which allows them to quickly access information about
the character they are currently playing and chat with the wizard to request
additional information as needed. These instruments were designed to be played
in concert, and each has a crucial role in enabling Bad News to be performed.
The importance of the wizard in this performance context stems from the fact
that leaving a simulation to run (regardless of the simulation’s narrative potency)
does not in and of itself produce compelling narrative directly. Instead, some
agent—often a human, as in the case of Bad News—is required to sift through
the storyworld state to surface and narrativize the interesting situations that
emerge. Since improvisationally performing as a character based on a relatively
limited amount of background information requires the actor’s full attention,
the wizard is needed to perform this narrativization function. This division of
labor suggests that a wide variety of new narrative authorship play experiences
may be enabled by a willingness to examine and divide up the tasks that existing
tabletop roleplaying games (for instance) tend to bundle up within a single player
role. Additionally, it is interesting to examine the Bad News performance team as
something like a touring narrative band. Due to the high degree of skill involved
in performing the actor and wizard roles, almost all past performances of Bad
News have featured the same two highly-trained individuals (Ben Samuel and
James Ryan) in the actor and wizard roles respectively. This performance crew
has traveled the world to perform Bad News at a diverse array of venues, and
they have become renowned, virtuosic operators of their narrative instruments
in the process.
2.3 Blaseball
Blaseball [35] is a live narrative idlegame driven by a simulation of a surrealist
fantasy baseball league. Uncommonly for an emergent narrative game, Blaseball
provides a single simulation instance that is shared between all players, rather
than spinning up a new simulation instance for each game or playthrough. Every
week of real time represents a single “season” of storyworld time, with a sea-
son consisting of approximately 1000 simulated baseball games between different
pairs of teams. Characters in the storyworld are procedurally generated baseball
players, frequently with humorous names (e.g., “Jessica Telephone”, “Gerund
Pantheocide”), and have a mix of visible and hidden numerical “stats” that run
the gamut from the practical (e.g., “baserunning”) to the absurd (e.g., “Shake-
spearianism”). The simulation juxtaposes typical baseball game events (like a
simulated baseball player scoring a run or striking out) with much stranger events
(such as players being incinerated by “rogue umpires”, swapped to the opposing
team mid-game due to “weather conditions”, or trapped in giant peanut shells).
Additionally, members of the game’s fan community are given the chance to
earn virtual money by betting on the outcomes of simulated games, which they
can then spend to increase the likelihood that their favored team will receive cer-
tain “blessings” at the start of the next season; to temporarily change the rules
of baseball as they apply to specific simulated teams; or to pursue various kinds
of collective community progression in the game’s overarching metanarrative.
The blessings and metanarrative options available in the end-of-season election,
504 M. Kreminski and M. Mateas
of digital tools like Imaginarium [10] to define and use new, simple constraint-
based procedural content generators (for things like encounters with enemies,
NPCs, and so on) as the course of the story demands. In this way, tabletop
roleplaying groups focused on storytelling may serve as natural testing grounds
for new kinds of narrative instruments. TTRPG groups can perhaps be viewed as
the narrative equivalent of garage bands: their members gather to co-construct
narrative on an ongoing basis largely due to enjoyment of the process, but they
may occasionally produce narrative artifacts that are suitable for consumption
by a wider audience.
In addition, the importance of a characteristic voice to the success or fail-
ure of an instrument may help to explain why later and more restricted story-
making tabletop games (such as Fiasco [4], Microscope, and The Quiet Year )
have proven more successful as narrative instruments than the apparently more
general story-making tools presented by games like the earlier Universalis [19]. In
its aspirations to generality, Universalis attempts to avoid fixing any part of the
storyworld or narrative structure in place, instead leaving everything up to the
players—in sharp contrast to later systems, which all impart a certain distinctive
texture on the stories they are used to construct.
All of these case studies demonstrate the active use of IEN systems by players
with the explicit goal of crafting stories: in other words, the use of IEN systems
as narrative instruments. In the context of Louchart and Aylett’s taxonomy of
user roles in emergent narrative [18], the systems with which we are concerned
primarily position the user as an author of narrative, rather than a spectator or
participant.
From a narrative instruments perspective, non-interactive emergent narrative
systems that position the user as a spectator (such as non-interactive story
generators like Tale-Spin [20]) resemble the narrative equivalent of windchimes:
they produce a kind of pretty but uncomplicated ambient background narrativity
that fades in and out of the spectator’s awareness, with most generated stories
or events failing to arouse much interest because of their great similarity to one
another. These narrative generation systems are at an inherent disadvantage due
to their need to compete for attention with stories that a human author had some
role in crafting; unlike participatory or authorship-focused emergent narrative
experiences, they can’t easily trade on the interactor’s sense of involvement to
make the stories they produce seem special. Nevertheless, low-stakes ambient
narrativity is worth exploring further, especially in concert with other forms of
narrativity that may demand or benefit from more active player involvement.
The play-pleasures of participatory IEN are more like the pleasures of going
to a concert. At the lower end of involvement, you might listen more or less
passively. But you might also dance, mosh, headbang, or otherwise move along
with the music; sing along with familiar lyrics; participate in call-and-response
rituals led by the band; call for the performance of specific songs from the band’s
506 M. Kreminski and M. Mateas
own back catalog, or for covers of songs by other bands; capture images or video
of key moments in the show, as a sharable souvenir; exchange shouted dialogue
with the band members between songs; or generally “vibe with” the band in a
wide variety of ways. The band, in turn, rarely ignores the audience completely:
instead, they pay attention to the energy of the crowd, allowing it to bleed into
the music in various ways, and engage with the audience in ways that foster a
sense of involvement or participation without ceding the stage entirely.
And then there are the play-pleasures of authorship, which are most closely
analogous to the play-pleasures of making music yourself. It is here that we
want to particularly focus our attention. Both spectatorship and participation
have formed the groundwork for a number of compelling emergent narrative
experiences—but what we are most interested in is exposing more people to the
joys of making narrative, and in expanding the set of instruments available for
casual narrative play. Existing story-making tabletop games have begun to map
out the contours of this design space, but the prominence and growth of IEN-
driven retelling practices also indicates player demand for narrative instruments
that leverage digital computation to provide forms of creativity support that
would not be possible or feasible without it.
The narrative instruments framing perhaps helps to clarify why participa-
tion tends to bleed into spectatorship at one end of the involvement spectrum
and into authorship at the other. Some bands are much more interactive or
responsive toward the audience than others, and in extreme cases a band may
either ignore the audience completely (leading to an experience that resembles
spectatorship) or permit the audience to play a significant role in determin-
ing the tone and flow of the concert (leading to an experience that resembles
authorship). Moreover, nothing can stop a sufficiently dedicated member of the
narrative avant-garde from grabbing a set of narrative windchimes and operating
them in some unexpected way, leveraging a system designed for spectatorship
as an unlikely narrative instrument to produce an experience of authorship for
themselves. But narrative instruments, like musical instruments, are nevertheless
designed to be used in certain ways—and the design of narrative instruments to
afford novel forms of narrative expression has as many nuances and complexities
as the design of musical instruments to afford novel forms of musical expression.3
More broadly, we find the idea of narrative instruments compelling. As an
explanatory framework, it helps us make sense of several recent IEN projects
that challenge our traditional design categories, but that nevertheless seem to
be compelling to players. As a design metaphor, it suggests future directions
for the development of new IEN systems. And as a provocative genre label,
it centers play practices and experiences that had previously been treated as
marginal in interactive narrative research communities. Altogether, we believe
3
In fact, there exists an entire academic conference—New Interfaces for Musical
Expression (NIME)—dedicated to the development of experiential new musical
instruments. Parrish’s New Interfaces for Textual Expression project [24] extends
the NIME ethos to the development of textual instruments, much like those called
for by Wardrip-Fruin in his own writing on the subject.
Toward Narrative Instruments 507
References
1. Alder, A.: The Quiet Year (2013). https://buriedwithoutceremony.com/the-quiet-
year
2. Bay 12 Games: Dwarf Fortress (2006). https://bay12games.com/dwarves
3. Blaseball Wiki contributors: Category: Players - Blaseball Wiki (2020). https://
www.blaseball.wiki/w/Category:Players
4. Bully Pulpit Games: Fiasco (2009). https://bullypulpitgames.com/games/fiasco
5. Denee, T.: Bronzemurder. https://www.timdenee.com/bronzemurder
6. Denee, T.: Oilfurnace. https://www.timdenee.com/oilfurnace
7. Eladhari, M.P.: Re-tellings: the fourth layer of narrative as an instrument for cri-
tique. In: Rouse, R., Koenitz, H., Haahr, M. (eds.) ICIDS 2018. LNCS, vol. 11318,
pp. 65–78. Springer, Cham (2018). https://doi.org/10.1007/978-3-030-04028-4 5
8. Garbe, J.: Simulation of history and recursive narrative scaffolding (2018). http://
project.jacobgarbe.com/simulation-of-history-and-recursive-narrative-scaffolding
9. Guzdial, M., et al.: Tabletop roleplaying games as procedural content generators.
In: International Conference on the Foundations of Digital Games (2020)
10. Horswill, I.: A declarative PCG tool for casual users. In: Proceedings of the AAAI
Conference on Artificial Intelligence and Interactive Digital Entertainment, vol.
16, pp. 81–87 (2020)
11. Kreminski, M., et al.: Cozy Mystery Construction Kit: prototyping toward an
AI-assisted collaborative storytelling mystery game. In: Proceedings of the 14th
International Conference on the Foundations of Digital Games (2019)
12. Kreminski, M., Dickinson, M., Mateas, M., Wardrip-Fruin, N.: Why are we like
this?: Exploring writing mechanics for an AI-augmented storytelling game. In:
Electronic Literature Organization Conference (2020)
13. Kreminski, M., Dickinson, M., Mateas, M., Wardrip-Fruin, N.: Why are we like
this?: The AI architecture of a co-creative storytelling game. In: International
Conference on the Foundations of Digital Games (2020)
14. Kreminski, M., Samuel, B., Melcer, E., Wardrip-Fruin, N.: Evaluating AI-based
games through retellings. In: Proceedings of the AAAI Conference on Artificial
Intelligence and Interactive Digital Entertainment, vol. 15, pp. 45–51 (2019)
15. Kreminski, M., Wardrip-Fruin, N.: Generative games as storytelling partners. In:
Proceedings of the 14th International Conference on the Foundations of Digital
Games (2019)
16. Kromtec: Legends Viewer (2015). https://github.com/Kromtec/LegendsViewer
17. Kruggsmash: Kruggsmash - YouTube. https://www.youtube.com/user/kruggsmash
18. Louchart, S., Aylett, R.: The emergent narrative theoretical investigation. In: Nar-
rative and Interactive Learning Environments Conference, pp. 21–28 (2004)
19. Mazza, R., Holmes, M.: Universalis: The Game of Unlimited Stories (2002)
20. Meehan, J.R.: TALE-SPIN, an interactive program that writes stories. In: Inter-
national Joint Conference on Artificial Intelligence (1977)
21. Murnane, E.: Emergent narrative: stories of play, playing with stories. Ph.D. thesis,
University of Central Florida (2018)
508 M. Kreminski and M. Mateas
22. Murray, J.H.: Hamlet on the Holodeck: The Future of Narrative in Cyberspace.
MIT Press, Cambridge (1997)
23. Nakakoji, K.: Meanings of tools, support, and uses for creative design processes.
In: International Design Research Symposium, vol. 6, pp. 156–165 (2006)
24. Parrish, A.: New interfaces for textual expression. Master’s thesis, New York Uni-
versity (2008)
25. Robbins, B.: Microscope: a fractal role-playing game of epic histories (2011).
https://lamemage.com/microscope
26. Ryan, J.: Curating simulated storyworlds. Ph.D. thesis, University of California,
Santa Cruz (2018)
27. Ryan, J., Mateas, M.: Simulating character knowledge phenomena in talk of the
town. In: Game AI Pro 360, pp. 135–150. CRC Press (2019)
28. Ryder-Hanrahan, G.: The Book of Ages (2018). https://site.pelgranepress.com/
index.php/the-book-of-ages
29. Samuel, B., Mateas, M., Wardrip-Fruin, N.: The design of Writing Buddy: a mixed-
initiative approach towards computational story collaboration. In: Nack, F., Gor-
don, A.S. (eds.) ICIDS 2016. LNCS, vol. 10045, pp. 388–396. Springer, Cham
(2016). https://doi.org/10.1007/978-3-319-48279-8 34
30. Samuel, B., Ryan, J., Summerville, A.J., Mateas, M., Wardrip-Fruin, N.: Bad
News: an experiment in computationally assisted performance. In: Nack, F., Gor-
don, A.S. (eds.) ICIDS 2016. LNCS, vol. 10045, pp. 108–120. Springer, Cham
(2016). https://doi.org/10.1007/978-3-319-48279-8 10
31. Shneiderman, B.: Creativity support tools: accelerating discovery and innovation.
Commun. ACM 50(12), 20–32 (2007)
32. Snow, K., Kavallines, G., Ferkol, T., McClure, A.: Matul remrit (2013). http://
www.bravemule.com/matulremrit
33. Sych, S.: When the fourth layer meets the fourth wall: the case for critical game
retellings. In: Bosser, A.-G., Millard, D.E., Hargood, C. (eds.) ICIDS 2020. LNCS,
vol. 12497, pp. 203–211. Springer, Cham (2020). https://doi.org/10.1007/978-3-
030-62516-0 18
34. Tanaka, A.: Interaction, experience and the future of music. In: O’Hara, K., Brown,
B. (eds.) Consuming Music Together. Computer Supported Cooperative Work, vol.
35, pp. 267–288. Springer, Dordrecht (2006). https://doi.org/10.1007/1-4020-4097-
0 13
35. The Game Band: Blaseball (2020). https://www.blaseball.com/
36. Wardrip-Fruin, N.: Playable media and textual instruments. Dichtung Digital
(2005)
Hypertext as a Lens into Interactive
Digital Narrative
1 Introduction
Interactive Digital Narrative (IDN) has its foundations in both early text adven-
ture games and the hypertext fiction of the 80s and 90s. Hypertext is therefore
often seen as a precursor to modern IDN systems, or (as manifest in popular
tools such as Twine) as a particular subset of IDN, based on textual nodes and
navigational links. However, Hypertext research has continued over the last few
decades, with many contributions around hypertext models, applications, usage
in the wild, tools and standards. This knowledge could be a valuable resource
c Springer Nature Switzerland AG 2021
A. Mitchell and M. Vosmeer (Eds.): ICIDS 2021, LNCS 13138, pp. 509–524, 2021.
https://doi.org/10.1007/978-3-030-92300-6_51
510 D. E. Millard and C. Hargood
to the IDN community if the relationship between Hypertext and modern IDN
systems could be defined more clearly.
Rather than try to rationalise IDN as Hypertext, or vice-versa, in this paper
we instead consider “Hypertext as Method”, an approach that uses Hypertext
as a method of inquiry to understand different types of systems [5]. In this
way Hypertext acts as a lens through which we can understand the common
hypertextuality of different IDN systems, creating an analytical tool which can
show how IDN forms are related to one another.
We also use Thue’s Interactive Process Model [43] as a way to scope the
lens and understand the boundaries of that hypertextuality. Rather than setting
out hypertext as a restrictive view of IDN, we thus hope that this approach
is a liberating one that could help us to understand how, through this common
hypertextuality, theories developed for one form of IDN might translate to others.
2 Background
Hypertext has its roots in the work of its pioneers, Doug Engelbart who cre-
ated NLS/Augment and considered hypertext as ‘augmenting man’s intellect’
[14], and Ted Nelson and his conceptual system Xanadu, that imagined a global
‘Permascroll’ with digital technologies to seamlessly navigate and explore [34].
In both cases the central idea was to allow readers to navigate between text
(or other media) by traversing navigational structures (such as links or trails).
By the 1980s there were many digital implementations available such as IBIS,
Intermedia, NoteCards, Hyperties, and ZOG [11], and by the end of the 1990s
hypertext was an established research area, with distinct sub-communities work-
ing both on hypertext as a knowledge tool and hypertext as digital literature
[45].
There were also attempts, working across both sub-communities, to formalise
hypertext and create agreed models so that systems might interoperate [12,13].
This led to an appreciation that there were different domains of hypertext, sets of
models and behaviour focused on a particular task. Navigational hypertext based
around nodes (containing media) connected via navigable links, spatial hypertext
focused on spatial structures like lists or sets that could be dynamically arranged
and identified by a spatial parser, and taxonomic hypertexts where conceptual
hierarchies are arranged into alternative views that can be traversed [31].
In the last two decades Hypertext has gone feral, adopted and adapted by
thousands of Internet communities [44], it has spawned studies in folksonomies
and semantic graphs [27], been applied to the expanding web and social networks
[4,22], as well as the real world via both augmented and mixed reality [17,42],
and it continues to struggle with its own form and poetics [8,35].
Interactive Digital Narratives share some of this early history, with key works
published in early hypertext systems such as Notecards, and dedicated platforms
such as Storyspace [7]. However, a mirror heritage in parser-based games, and
a focus on high level narrative and content rather than low level associative
structure, has given IDN research its own flavour, and ultimately a distinct
Hypertext as a Lens on IDN 511
community. Popular free platforms have been established such as Twine and
Inform, encouraging a wider audience and experimentation that has driven craft
knowledge [15]. In the last decade advances in game development frameworks
has resulted in an explosion of independent narrative games, unconstrained by
traditional forms, that are pushing IDN in a myriad of directions [16,33,38].
Not surprising then, that many in the IDN community consider hypertext
to be either something from the history books or at best a tiny subset of what
IDN can be. Millard attempts to reconcile the two, seeing ‘literary hypertext as
a subset of games, but narrative games as fundamentally hypertextual.’ [32] But
what does this hypertextuality actually mean?
Atzenbeck and Nürnberg suggest the approach that we use in this paper.
They argue that Hypertext is a method of inquiry; a way of thinking about
systems that can provide new insights [5]. They identify three hypertext per-
spectives: first class structure, context dependent structure, and open ended
structure, and point out that ‘other fields may adopt features of some of these
perspectives, but the primacy the hypertext literature places upon them is distin-
guishing.’ They tend towards viewing hypertext as a knowledge tool, but in this
paper we explore the other hypertext tradition, hypertext as digital literature.
We also draw on Thue’s model of an interactive narrative process [43], this
views IDN systems as three functions that control respectively observation (what
the reader sees), action (what the reader does), and transitions (how those
actions change narrative state). Our approach is to use the hypertext lens to
analyse the transition functions of different IDN forms, arguing that the hyper-
text focus on structure helps to understand and contextualise those forms. By
examining the observation and action functions we can also scope the lens, and
see what important elements lie beyond the hypertextual analysis.
3 Hypertext as a Lens
In this section we systematically examine a series of IDN forms, looking at how
a hypertext lens might help explain how they function and how they are similar
or distinct to one another. We do this by exploring the way in which IDNs are
structured, using hypertext terms and structural models.
When we use the term hypertext, most people will think of the navigational
domain of hypertext, defined with nodes and links. Where these are explicitly
defined this has also been called Calligraphic Hypertext, as the links are drawn
deliberately between nodes to create navigational paths between them [6]. Links
are emblematic of hypertext systems, so much so that Halasz referred to the
‘Tyranny of the Link’ to complain about the extent to which they exclude other
types of structures (such as trails, or virtual documents) [19].
This hyperstructure is concerned with defining the possible states of the
hypertext, and the ways in which a reader might move between them. It is
512 D. E. Millard and C. Hargood
1
Although we use the term lexia this need not be textual content, and in fact our
emphasis on lexia representing state changes means that it is their purpose within
the overall narrative that is important rather than their form, making our use of the
term very similar to Mateas’ notion of ‘dramatic beats’ [30].
2
The adaptive hypertext literature refers to a user model, as an amalgam of user
preferences and current reading state, but in IDN non-diegetic user preferences (from
outside of the user’s interaction with the hypertext, e.g. age or expertise level) are
less important, so we adopt the term story state to describe the same concept.
Hypertext as a Lens on IDN 513
Fig. 1. One high level Storylet captures the behaviour of three low level Storylets.
Like sculptural hypertexts, parser fiction uses story state to manage its nar-
rative, only here the interaction engine relies on interpreted typed commands
rather than clickable links. These parser fictions often take the form of a tra-
ditional text adventure and express the story state in terms of a world model,
a schema or super structure that provides a framework for a coherent (and
more complete) description of the story world. This allows authors to express
themselves at a higher level than a set of variables - for example, in Inform we
declare a set of rooms and their relationships, objects and their locations within
rooms, and scenes in which story unfolds. There is then a grammar for interact-
ing with this world model (moving between rooms, picking up and inspecting
objects, etc.) The query interface of database narratives can also be considered
as a simple grammar, but the superstructure of the world model makes a more
complex grammar possible, as valid actions can be defined against whole classes
of objects. Parser fiction also often uses a different presentation engine to tra-
ditional hypertext, as it doesn’t show you explicit choices and instead expects
your choices to be made diegetically by interacting with the world model [28].
While parser based adventure games adopt new interaction and presentation
engines, narrative games can take this even further. Whilst still built on a story
engine (and often a world model) their interactions can include rich ludonarra-
tive mechanics [2], with elements of the story being delivered through a variety
of channels, such as cutscenes, scripted events/character barks, log entries, inter-
active dialog, and environmental storytelling (reminiscent of transmedia [23]).
The element of environmental storytelling means that, like database narra-
tives, narrative games and parsers do not necessarily track and manage all of
the state through their story engines, and that there is the possibility that the
transition function is shared between player and machine. To some extent then,
all three forms are epistemic, not in Ryan’s motivational sense (in that players
are “driven by the need to know”) [40] but rather they are at least partially
driven by what the player knows.
As we have considered these different forms of IDN - Calligraphic, Adaptive,
Sculptural, Database, Parsers, and Games - we have moved through and then
beyond the hypertext lens. Particularly in these last two forms, where there is
increased importance of the observation and action functions as ways of deliver-
ing story and managing the experience. Thus a design theory of narrative games
cannot be solely based on hypertext, but these hypertextual models can be used
to understand the transition function - the story engine and underlying struc-
tural models for such experiences. Hypertext models can therefore be used as
analytical tools to understand how narrative games work, and hypertext systems
can be used as authoring tools for much of their narrative structure.
516 D. E. Millard and C. Hargood
4 Discussion
Through our hypertextual description of IDN forms we have developed a simple
layer model of the Hypertext Lens (shown in Fig. 2).
Central to this is a finite state model where each lexia that the reader encoun-
ters represents a transition between states of the narrative. Note that lexia repre-
sent transitions, not states (see Sect. 3.1). However, it is possible that narratives
are managed solely via lexia state, in which case there is a one-to-one correlation
between the lexia state and the state of the narrative (e.g. the state might be:
‘You have just read Page 1’). Calligraphic Hypertext exists at this level, read-
ing a lexia in the hypertext transitions you to a matching state, and links in
that lexia represent the next possible lexia and state transitions. Authors are
effectively directly writing the narrative state graph. Because of this one-to-one
match Calligraphic Hypertext is easy to understand and is therefore one of the
most accessible forms of IDN, although if the narrative is very open (meaning
that the reader has many choices) then the density of the state graph can become
high, and the hypertext becomes difficult to author and visualise.
A way to solve this problem is to model story state explicitly, and to use
this and not lexia state to model the narrative. Figure 3 shows a simple story
sequence modelled as a calligraphic hypertext (using only lexia state) and as
a sculptural hypertext (using only story state). Lexia still act as transitions
between story states (e.g. reading ‘You take the key.’ leads to the state: ‘has
key’) but now, assuming that not all of the information in the lexia will affect
the onward story4 , different lexia can transition you to the same story states,
4
In our example we are not interested in modelling the state of the guard.
Hypertext as a Lens on IDN 517
which simplifies the state graph (in Fig. 3, Original Story, states 2 and 3 in
the calligraphic model have been consolidated into one state in the sculptural
model). This is the basis of Sculptural Hypertext, which exclusively uses story
state to manage the reader’s progress.
Fig. 3. Calligraphic Hypertext relies on Lexia State and authors directly write the state
model, whereas Sculptural Hypertext uses Story State and users write at a higher level
5
Even in calligraphic hypertext, the observation and interaction engines convey mean-
ing, e.g. Mason and Bernstein’s work on the poetics of contemporary link usage [29].
Hypertext as a Lens on IDN 519
layers, and places hypertext in a central ‘Storage Layer’. The Lens we have set
out (lexia state, story state, world model, and story engine) correspond to this
‘Storage Layer’. Whereas interaction engines are part of the ‘Runtime Layer’
and presentation engines part of the ‘Within Component Layer’, both of which
Dexter considers to be outside of the core hypertext design.
The Hypertext Lens does not therefore cover the entirety of IDN systems.
However, it helps to explain what Millard called a ‘core hypertextuality’ in
narrative games [32], evidenced in attempts to characterise Bandersnatch as a
Gauntlet [37] or to map the structure of The Walking Dead [24]. We go beyond
this claim and have begun to unpack this hypertextuality and to relate it to
different IDN forms. Thue’s model of an interactive narrative process helps
to explain that the Hypertext Lens describes the transition function of IDN,
whereas games design better covers the observation and action functions. This
means that when those functions are relatively simple the hypertext lens is a
more complete description than when they are complex. While the explanatory
power of the lens is focused on the interactive narrative elements of games, it
does not follow that it is limited to games where narrative is seen as a minor
element, just that its scope is limited within those games. In these cases the lens
provides clarity precisely because it allows you to focus on one part of a more
complex medium.
The Hypertext Lens also reveals the fluidity of IDN forms. This might inspire
us to consider IDNs that mix forms together. For example, we might imagine
database stories with some sculptural elements, sculptural stories with a more
robust world model, or calligraphic stories that include some free text query
and therefore push some story state outside of the story engine. We have also
seen how hypertext can help explain complex IDNs that already mix up these
forms. This is true even in the case of commercial narrative games, for example
Supergiant’s narrative roguelike Hades could be seen as a sculptural hypertext
using large grain storylets with calligraphic internal structure, coupled with a
presentation engine that selects storylets based on a priority selection strategy.
Understanding the hypertextual commonalities of these forms also enables
them to benefit from theories and craft knowledge developed elsewhere. Database
narratives may seem like unknown design spaces, but much of the craft knowledge
from storylets could be applied to them (e.g. by utilising sculptural patterns
in their design [20]). Understanding whether the narrative model of a game is
more calligraphic or sculptural enables us to apply the right narrative design
principles, and to pick an appropriate IDN tool in which to write the script.
Hypertext is a structure-centric view of IDN (after all, Atzenbeck argued that
a focus on structure was the defining perspective of hypertext [5]), and while it
does not fully capture all of the elements it does allow us to think about the
structures separately from the other parts of the experience. Thus it has real
value as a thought pattern, a way of approaching IDN that sits alongside similar
analytical models such as SPP [25] or the Double-Hermeneutic Circle [39] and
can be used both descriptively for analysis and prescriptively for design.
520 D. E. Millard and C. Hargood
5 Conclusion
In this paper we have adopted Atzenbeck’s approach of using Hypertext as a
Method of Inquiry in order to analyse IDN forms through a hypertext lens.
We have also used Thue’s Interactive Process Model as a means to scope that
lens, and have shown that it helps to deconstruct the transition function of IDN
systems (how the system manages narrative state). In doing so we have revealed
how different IDN forms relate to one another. Calligraphic hypertext purely
uses lexia states and transitions (resulting in a one to one mapping between
lexia and narrative state), sculptural hypertext (a set of storylets) purely uses
story states and transitions, and adaptive hypertexts mix the two, using lexia
state to manage most of the narrative, with story state employed to simplify
overly complex sections. In all three cases a story engine manages the transition
function. The lens also shows that database IDNs are sculptural in nature, but
instead of a story engine the transition function occurs within the mind of the
reader; and it reveals parser fiction as structurally sculptural but defined against
a world model (which makes authoring complex state easier). Throughout, we
have given small or abstract examples of the lens in action, and in future work
we hope to use the lens to explore specific IDN works in more detail.
The hypertext lens is a thought pattern that puts structure first, it thus
moves the focus away from the presentation and interaction engines that fulfil
the observation and action functions (how narrative and choices are presented,
and the ways in which the reader/player interacts with them). In calligraphic and
adaptive hypertext these are relatively simple, and the lens provides an almost
complete picture, but in sculptural systems a more complex presentation engine
makes decisions about which storylets to reveal next, and in parser fiction a more
sophisticated interaction engine manages a complex grammar made possible by
the world model. In narrative games there is a high level of experimentation
with presentation and interaction engines - with multiple channels reminiscent
of transmedia, and complex mechanics that yield only to game design theory.
Nevertheless, the hypertext lens continues to usefully highlight the structures
within the transition function and the activity of the game’s story engine.
This coherent hypertextual view of IDN suggests a fluidity of forms, with
the potential for theory to be applied consistently across them. When new ideas
emerge (such as database narratives) if they can be explained in terms of exist-
ing hypertext models they immediately benefit from design theory that already
exists. Hybrid approaches that seem at first esoteric and strange (for example,
the located and contextual nodes of StoryPlaces) can be explained in terms of
what has gone before. It also implies a core set of narrative design skills that
could be taught, and would be relevant across many different forms.
Finally it is worth acknowledging that the hypertext lens is a technical decon-
struction of IDN, it does not tell us how to write dramatic plots, how to manage
agency effectively, or how to build believable characters, but it does explain the
structural narrative architecture of IDN forms and reveals the fluidity between
them. This comprehensive explanation of a part of IDN is valuable. If we under-
stand the hypertextual similarities between IDNs then when we see best practice
Hypertext as a Lens on IDN 521
in one form we can translate it to another, even though the lens doesn’t tell you
what that best practice itself might be. We hope that by using hypertext to
set out a coherent view of IDN forms, and by describing a lens that can be
used in future to analyse new works and approaches, that we will empower the
next generation of writers, designers, and tool developers to explore these other
questions, and in doing so, develop answers that can be applied widely.
Acknowledgements. The authors would like to thank David Thue and Joey Jones
for providing feedback on an earlier draft of this paper.
References
1. Storynexus: Reference guide, version 1.04. Technical report, Failbetter Games
(2012). http://wiki.failbettergames.com/start
2. Aarseth, E.: A narrative theory of games. In: Proceedings of the International Con-
ference on the Foundations of Digital Games, FDG ’12, pp. 129–133. Association
for Computing Machinery, New York (2012). https://doi.org/10.1145/2282338.
2282365
3. Adams, E.: Fundamentals of Game Design, 3rd edn. New Riders Publishing, New
York (2014)
4. Anderson, M., Carr, L., Millard, D.E.: There and here: patterns of content tran-
sclusion in wikipedia. In: Proceedings of the 28th ACM Conference on Hypertext
and Social Media, pp. 115–124 (2017)
5. Atzenbeck, C., Nürnberg, P.J.: Hypertext as method. In: Proceedings of the 30th
ACM Conference on Hypertext and Social Media, HT ’19, pp. 29–38. Association
for Computing Machinery, New York (2019). https://doi.org/10.1145/3342220.
3343669
6. Bernstein, M.: Card shark and thespis: Exotic tools for hypertext narrative. In:
Proceedings of the 12th ACM Conference on Hypertext and Hypermedia, HYPER-
TEXT ’01, pp. 41–50. Association for Computing Machinery, New York (2001).
https://doi.org/10.1145/504216.504233
7. Bernstein, M.: Storyspace 1. In: Proceedings of the Thirteenth ACM Conference
on Hypertext and Hypermedia, HYPERTEXT ’02, pp. 172–181. Association for
Computing Machinery, New York (2002). https://doi.org/10.1145/513338.513383
8. Bernstein, M.: On hypertext narrative. In: Proceedings of the 20th ACM Confer-
ence on Hypertext and Hypermedia, pp. 5–14 (2009)
9. Brusilovsky, P.: Adaptive hypermedia. User Model. User-Adap. Inter. 11(1), 87–
110 (2001)
10. Brusilovsky, P., Kobsa, A., Vassileva, J.: Adaptive Hypertext and Hypermedia.
Springer, Heidelberg (1998). https://doi.org/10.1007/978-94-017-0617-9
11. Conklin, J.: Hypertext: an introduction and survey. Computer 20, 17–41 (1987).
https://doi.org/10.1109/MC.1987.1663693
12. Davis, H.C., et al.: Interoperability between hypermedia systems: the standardisa-
tion work of the OHSWG. In: Proceedings of the Tenth ACM Conference on Hyper-
text and Hypermedia: Returning to Our Diverse Roots: Returning to Our Diverse
Roots, HYPERTEXT ’99, pp. 201–202. Association for Computing Machinery,
New York (1999). https://doi.org/10.1145/294469.294904
522 D. E. Millard and C. Hargood
13. Davis, H., Reich, S., Millard, D.: A proposal for a common navigational hypertext
protocol. Technical report, Dept. of Electronics and Computer Science, University
of Southampton (1997)
14. Engelbart, D.C.: Augmenting human intellect: a conceptual framework. Menlo
Park, CA (1962)
15. Ford, M.: Writing Interactive Fiction with Twine. Que Publishing, Seattle (2016)
16. Gasque, T.M., Tang, K., Rittenhouse, B., Murray, J.: Gated story structure and
dramatic agency in Sam Barlow’s Telling Lies. In: Bosser, A.-G., Millard, D.E.,
Hargood, C. (eds.) ICIDS 2020. LNCS, vol. 12497, pp. 314–326. Springer, Cham
(2020). https://doi.org/10.1007/978-3-030-62516-0 28
17. Grønbæk, K., Kristensen, J.F., Ørbæk, P., Eriksen, M.A.: “Physical hypermedia”
organising collections of mixed physical and digital material. In: Proceedings of
the Fourteenth ACM Conference on Hypertext and Hypermedia, pp. 10–19 (2003)
18. Halasz, F., Schwartz, M., Grønbæk, K., Trigg, R.H.: The dexter hypertext refer-
ence model. Commun. ACM 37(2), 30–39 (1994). https://doi.org/10.1145/175235.
175237
19. Halasz, F.G.: “Seven issues”: revisited, hypertext’91 closing plenary. In: Proceed-
ings of Third ACM Conference on Hypertext (Hypertext’91) (1991)
20. Hargood, C., Hunt, V., Weal, M., Millard, D.: Patterns of sculptural hypertext in
location based narratives. In: HT’16: Proceedings of the 27th ACM Conference on
Hypertext and Social Media, pp. 61–70. ACM (2016). https://eprints.soton.ac.uk/
390748/
21. Hargood, C., Weal, M., Millard, D.: The storyplaces platform: building a web-based
locative hypertext system. In: HT ’18 Proceedings of the 29th ACM Conference
on Hypertext and Social Media, pp. 128–135. ACM (2018). https://eprints.soton.
ac.uk/421122/
22. Helic, D., Strohmaier, M., Granitzer, M., Scherer, R.: Models of human navigation
in information networks based on decentralized search. In: Proceedings of the 24th
ACM Conference on Hypertext and Social Media, pp. 89–98 (2013)
23. Javanshir, R., Millard, D., Carroll, E.: Structural patterns for transmedia story-
telling. PLoS ONE 15(1) (2020). https://eprints.soton.ac.uk/438130/
24. Killham, E.: Here’s a chart of every choice in the walking dead: season
1 (image) (2013). https://venturebeat.com/2013/03/31/the-walking-dead-season-
one-plot-graph/
25. Koenitz, H.: Towards a theoretical framework for interactive digital narrative. In:
Aylett, R., Lim, M.Y., Louchart, S., Petta, P., Riedl, M. (eds.) ICIDS 2010. LNCS,
vol. 6432, pp. 176–185. Springer, Heidelberg (2010). https://doi.org/10.1007/978-
3-642-16638-9 22
26. Kreminski, M., Wardrip-Fruin, N.: Sketching a map of the storylets design space.
In: Rouse, R., Koenitz, H., Haahr, M. (eds.) ICIDS 2018. LNCS, vol. 11318, pp.
160–164. Springer, Cham (2018). https://doi.org/10.1007/978-3-030-04028-4 14
27. Marlow, C., Naaman, M., Boyd, D., Davis, M.: HT06, tagging paper, taxonomy,
flickr, academic article, to read. In: Proceedings of the Seventeenth Conference on
Hypertext and Hypermedia, pp. 31–40 (2006)
28. Mason, S.: On games and links: extending the vocabulary of agency and immersion
in interactive narratives. In: Koenitz, H., Sezen, T.I., Ferri, G., Haahr, M., Sezen,
D., C atak, G. (eds.) ICIDS 2013. LNCS, vol. 8230, pp. 25–34. Springer, Cham
(2013). https://doi.org/10.1007/978-3-319-02756-2 3
Hypertext as a Lens on IDN 523
29. Mason, S., Bernstein, M.: On links: exercises in style. In: Proceedings of the 30th
ACM Conference on Hypertext and Social Media, HT ’19, pp. 103–110. Association
for Computing Machinery, New York (2019). https://doi.org/10.1145/3342220.
3343665
30. Mateas, M., Stern, A.: Structuring content in the façade interactive drama archi-
tecture. In: Proceedings of the First AAAI Conference on Artificial Intelligence
and Interactive Digital Entertainment, AIIDE’05, pp. 93–98. AAAI Press (2005)
31. Millard, D.E., Moreau, L., Davis, H.C., Reich, S.: FOHM: a fundamental open
hypertext model for investigating interoperability between hypertext domains. In:
Proceedings of the Eleventh ACM on Hypertext and Hypermedia, HYPERTEXT
’00, pp. 93–102. Association for Computing Machinery, New York (2000). https://
doi.org/10.1145/336296.336334
32. Millard, D.: Games/hypertext. In: ACM Conference on Hypertext and Social
Media (13/07/20 - 15/07/20), pp. 123–126 (2020). https://eprints.soton.ac.uk/
442652/
33. Di Pastena, A., Jansen, D., de Lint, B., Moss, A.: The link out. In: Rouse, R.,
Koenitz, H., Haahr, M. (eds.) ICIDS 2018. LNCS, vol. 11318, pp. 206–216. Springer,
Cham (2018). https://doi.org/10.1007/978-3-030-04028-4 21
34. Nelson, T.H., et al.: Literary machines: The report on, and of, project Xanadu, con-
cerning word processing, electronic publishing, hypertext, thinkertoys, tomorrow’s
intellectual revolution, and certain other topics including knowledge, education and
freedom (1981)
35. Pisarski, M.: New plots for hypertext? Towards poetics of a hypertext node. In:
Proceedings of the 22nd ACM Conference on Hypertext and Hypermedia, pp. 313–
318 (2011)
36. Reed, A.A.: Telling stories with maps and rules: using the interactive fiction lan-
guage “inform 7” in a creative writing workshop. In: Creative Writing in the Digital
Age: Theory, Practice, and Pedagogy, pp. 141–152 (2015)
37. Rezk, A.M., Haahr, M.: The case for invisibility: understanding and improving
agency in black mirror’s bandersnatch and other interactive digital narrative works.
In: Bosser, A.-G., Millard, D.E., Hargood, C. (eds.) ICIDS 2020. LNCS, vol. 12497,
pp. 178–189. Springer, Cham (2020). https://doi.org/10.1007/978-3-030-62516-
0 16
38. Roth, C., Koenitz, H.: Bandersnatch, yea or nay? Reception and user experience
of an interactive digital narrative video. In: Proceedings of the 2019 ACM Interna-
tional Conference on Interactive Experiences for TV and Online Video, pp. 247–254
(2019)
39. Roth, C., van Nuenen, T., Koenitz, H.: Ludonarrative hermeneutics: A Way Out
and the narrative paradox. In: Rouse, R., Koenitz, H., Haahr, M. (eds.) ICIDS
2018. LNCS, vol. 11318, pp. 93–106. Springer, Cham (2018). https://doi.org/10.
1007/978-3-030-04028-4 7
40. Ryan, M.-L.: Interactive narrative, plot types, and interpersonal relations. In:
Spierling, U., Szilas, N. (eds.) ICIDS 2008. LNCS, vol. 5334, pp. 6–13. Springer,
Heidelberg (2008). https://doi.org/10.1007/978-3-540-89454-4 2
41. Short, E.: Beyond branching: quality-based, salience-based, and waypoint
narrative structure (2016). https://emshort.blog/2016/04/12/beyond-branching-
quality-based-and-salience-based-narrative-structures/
42. Sinclair, P., Martinez, K., Millard, D.E., Weal, M.J.: Links in the palm of your
hand: tangible hypermedia using augmented reality. In: Proceedings of the Thir-
teenth ACM Conference on Hypertext and Hypermedia, pp. 127–136 (2002)
524 D. E. Millard and C. Hargood
43. Thue, D.: What might an action do? Toward a grounded view of actions in interac-
tive storytelling. In: Bosser, A.-G., Millard, D.E., Hargood, C. (eds.) ICIDS 2020.
LNCS, vol. 12497, pp. 212–220. Springer, Cham (2020). https://doi.org/10.1007/
978-3-030-62516-0 19
44. Walker, J.: Feral hypertext: when hypertext literature escapes control. In: Proceed-
ings of the Sixteenth ACM Conference on Hypertext and Hypermedia, pp. 46–53
(2005)
45. Wardrip-Fruin, N.: What hypertext is. In: Proceedings of the Fifteenth ACM Con-
ference on Hypertext and Hypermedia, HYPERTEXT ’04, pp. 126–127. Asso-
ciation for Computing Machinery, New York (2004). https://doi.org/10.1145/
1012807.1012844
46. Weal, M.J., Millard, D.E., Michaelides, D.T., Roure, D.C.D.: Building narrative
structures using context based linking. In: In Hypertext ’01. Proceedings of the
Twelfth ACM Conference on Hypertext, Aarhus, Denmark. (01/08/01), pp. 37–38
(2001), https://eprints.soton.ac.uk/256136/
A Pilot Study on Analyzing Critical Retellings
Using Digital Humanities Tools
Abstract. This paper presents the early findings of a pilot study on the analysis of
longform video essays as critical retellings of video game narratives. Using web-
based reading and analysis environment Voyant, the study explores the indicators
of how players critically approach and discuss game narrative design.
1 Introduction
Steven Sych [5] suggests the concept of “critical retellings” which he describes a sub-
set of retellings that do not directly track the success of a narrative system but instead
take a deliberately critical stand towards it. Critical retellings, according to Sych, pro-
vide commentary on the flaws in a game’s system, tend to be more anecdotal, target an
experienced and knowing audience, use irony in their criticism, and explicitly reference
and reflect on the mechanics of the narrative system. Sych’s description emphasizes the
role of subjectivity, self-reflexivity, and satirical position while experiencing games, and
analyzing and interpreting narrative outcomes of game systems. While Sych describes
critical retellings as short, sharp, ironical texts; the elaborate, personal, and creative
longform video essays retell game experiences with a deliberately critical stand towards
videogame systems as well. Considered one of the most prominent contemporary forms
of non-scholarly videogame criticism [6, 7], longform video essays on video games can
be seen as a variation of critical retellings. Not solely narrative products [8] instantiated
in one uninterrupted play through, longform video essays shared on social media plat-
forms provide in-depth discussions and interpretations exemplifying open scholarship
in videogame studies [7].
Game vlogger Noah Caldwell-Gervais’ video essays can be seen as examples of
longform video essays as critical retellings. Compared to film critique Roger Ebert for
legitimizing the form of videogames and demonstrating its potential [13], he produces
in-depth, several hours long analyses and critiques of videogames. In his two-hour long
video essay titled How does the Last of Us Part 2 compare to the Last of Us Remas-
tered? Caldwell-Gervais [14] presents his experiences in playing both incarnations of
the videogame series TLoU and discusses the connections between gameplay, narrative
design, and the story. TLoU [15] and its sequel TLoU Part II [9], are action-adventure
games set in a post-apocalyptic future. Featuring elements of the survival horror genre,
they share a character-driven linear narrative design. TLoU was praised [16] for its rep-
resentation of its protagonists Joel and Ellie, a hardened man, and a young girl, who
throughout the game develop a father-daughter relationship, leading Joel to commit
questionable acts of violence to protect Ellie at the end of the game. TLoU Part II starts
with the murder of Joel by the daughter of one of his victims from the previous game,
Abby, and then follows the crossing paths of now vengeful Ellie and Abby with their
respected companions Dina and Lev, till a final confrontation. Both the choice of killing
off Joel and making Abby a protagonist alongside a hostile Ellie in TLoU Part II has
caused negative reactions in some fan communities [17]. Providing a nuanced reaction
to TLoU Part II, Caldwell-Gervais’ retelling [14] in comparison describes the narrative
direction of the series as creatively unique and genuinely surprising, as it allows the
player to discover the other side of the same story from the opponents’ perspective. He
argues that the designers expected the players to feel anger towards Abby, and at the
same time allowed them to explore her perspective and invited them to re-evaluate the
story of the previous game critically [14].
The Voyant analysis focusing on frequencies and collocates reveals how Caldwell-
Gervais discusses the story of TLoU Part II gradually in his retelling and shows how he
constructs his arguments around characters.
4 Conclusion
As detailed, informed, and critical discussions of videogames, longform video essays
hold the potential to provide an in-depth record of players’ meaning-making processes.
Digital humanities tools may prove useful in analyzing how such retellings are con-
structed by their authors. The Voyant analysis provided a look on how Caldwell-Gervais
perceives game characters and their relationships. Further comparative research on a
larger sample size is required to understand further advantages of this methodology
and may reveal situatedness of game narrative perception. Inclusion of social aspects
of shared retellings such as comments and discussion threads may also prove useful.
Using custom categories in text mining such as gameplay and narrative design and using
further tools focusing on networks formed around retellings may provide deeper insights
on how videogame narratives are perceived and discussed among players as well.
References
1. Fernandez-Vara, C.: Introduction to Game Analysis. Routledge, New York (2014)
A Pilot Study on Analyzing Critical Retellings 529
2. Mukherje, S.: Videogames as minor literature: reading videogame stories through paratexts.
GRAMMA J. Theory Crit. 23, 60–75 (2016)
3. Eladhari, M.P.: Re-tellings: the fourth layer of narrative as an instrument for critique. In:
Rouse, R., Koenitz, H., Haahr, M. (eds.) ICIDS 2018. LNCS, vol. 11318, pp. 65–78. Springer,
Cham (2018). https://doi.org/10.1007/978-3-030-04028-4_5
4. Kerttula, T.: What an eccentric performance: storytelling in online let’s plays. Games Cult.
14(3), 236–255 (2016)
5. Sych, S.: When the fourth layer meets the fourth wall: the case for critical game retellings. In:
Bosser, A.-G., Millard, D.E., Hargood, C. (eds.) ICIDS 2020. LNCS, vol. 12497, pp. 203–211.
Springer, Cham (2020). https://doi.org/10.1007/978-3-030-62516-0_18
6. Cramer, M.: How We Talk About Games. https://www.youtube.com/watch?v=JVN9h-
5UHMk&t=3438s. Accessed 5 June 2021
7. Saklofske, J.: Gaming the Publishing Industry: Exploring Diverse Open Scholarship Models
in Digital Games Studies. https://popjournal.ca/issue02/saklofske. Accessed 5 June 2021
8. Koenitz, H.: Towards a specific theory of interactive digital narrative. In: Koenitz, H., Ferri,
G., Haahr, M., Sezen, D., Sezen, T.I. (eds.) Interactive Digital Narrative: History, Theory and
Practice, pp 91–105. Routledge, New York (2015)
9. Naughty Dog: The Last of Us Part II. Sony Interactive Entertainment (2020)
10. Kreminski, M., Samuel, B., Melcer, E., Wardrip-Fruin, N.: Evaluating AI-based games
through retellings. In: Proceedings of Fifteenth AAAI Conference on Artificial Intelligence
and Interactive Digital Entertainment, pp 45–51. The AAAI Press, Palo Alto (2019)
11. Toh, W.: The player experience of bioshock: a theory of ludonarrative relationships. In:
Ensslin, A., Balteiro, I. (eds.) Approaches to Videogame Discourse: Lexis, Interaction,
Textuality, pp 247–268. Bloomsbury Academic, New York (2019)
12. Roth, C., van Nuenen, T., Koenitz, H.: Ludonarrative hermeneutics: a way out and the narrative
paradox. In: Rouse, R., Koenitz, H., Haahr, M. (eds.) Interactive Storytelling. ICIDS 2018.
LNCS, vol. 11318, pp. 93–106. Springer, Cham (2018). https://doi.org/10.1007/978-3-030-
04028-4_7
13. Good, N.: This Oregon-Based Vlogger Might Be the Closest Thing Video Games Have
to Roger Ebert. https://www.wweek.com/technology/2020/05/20/this-oregon-based-vlogger-
might-be-the-closest-thing-video-games-has-to-roger-ebert/. Accessed 5 June 2021
14. Caldwell-Gervais, N.: How does the Last of Us Part 2 compare to the Last of Us Remastered?
https://www.youtube.com/watch?v=Bat38vErWr4. Accessed 5 June 2021
15. Dog, N.: The Last of Us. Sony Interactive Entertainment (2013)
16. Anyó, L., Colom, À.: Emotional ambivalence in the last of us. emotion in video games,
between narrative complexity and player loyalty. L’Atalante: Revista de estudios cinematográ-
ficos 31, 85–101 (2021)
17. Vargas, J.A.: AngryJoeShow: The Last of Us Part II - Angry Review. https://www.youtube.
com/watch?v=_-sTlYUeT8o&t=3s. Accessed 5 June 2021
18. Sinclair, S., Rockwell, G.: Voyant Tools. http://voyant-tools.org/. Accessed 5 June 2021
Author Index