Michael Burmester, Daniela Gerhard,

Frank Thissen (eds.)

Digital Game Based Learning
Proceedings of the 4th International
Symposium for Information Design
2nd of June 2005 at Stuttgart Media University
Digital Game
Based Learning

Proceedings of the 4th International

Symposium for Information Design
2nd of June 2005 at Stuttgart Media University

Michael Burmester
Daniela Gerhard
Frank Thissen
(eds.)
Impressum

Universitätsverlag Karlsruhe
c/o Universitätsbibliothek
Straße am Forum 2
D-76131 Karlsruhe
www.uvka.de

Dieses Werk ist unter folgender Creative Commons-Lizenz

lizenziert: http://creativecommons.org/licenses/by-nc-nd/2.0/de/

Universitätsverlag Karlsruhe 2006

Print on Demand

ISBN 3-86644-010-3
Editorial

Learning and gaming: does this fit together? Can we use the idea of computer games
to make use for learning purposes? Who will benefit from this? Does learner learn
better or more motivated, when they play games in order to acquire content? Is it
more fun to learn by games? According to Marc Prensky’s new book children sitting
at their digital learning game would say “Don’t Bother Me Mom — I’m Learning!”
Will that become true?

The European project SIG-GLUE, the “Special Interest Group for Game-based
Learning in Universities and lifElong Learning” (www.sig-glue.net) tries to bring
together experts and practitioners in the field of digital game based learning in order
to discuss and exchange information concerning the questions mentioned above. The
main goal is to promote more and better use of better learning games. Furthermore,
the project would like to convince non users of digital games of their educational
value, to encourage and support game developers in the creation of better educational
games, and to make educators aware of how to use games more effectively in education.
On the 2nd of June 2005, the 4th International Symposium for Information Design
at the Stuttgart Media University in Stuttgart, Germany, was organised in cooperation
with the SIG-GLUE project in order to give experts from research and game design
the chance to present their ideas on “Digital Game Based Learning”. This book contains
the presentations of that Symposium.

The authors of this book present an overview of ongoing research and development
projects in the field of digital game based learning. At the beginning of the book we
will get an introduction into SIG-GLUE and the SIG-GLUE community. Maja Pivec
(coordinator of SIG-GLUE) and Anastasia Sfiri from FH Joanneum Graz, Austria,
introduce the SIG-GLUE project, showing the objectives and the possibilities to
participate in the SIG-GLUE community. The central platform for information ex-
change and discussion is the SIG-GLUE web site www.sig-glue.net. Ioannis Antonellis,
Christos Bouras, Apostolos Gkamas and Vassilis Poulopoulos from the Research
Academic Computer Technology Institute (CTI) from Patras, Greece, describe the
intentions and the technical implementation of that communication platform.

5
A very good introduction in basic considerations of digital game based learning is
done by the following three contributions. The paper “Categorising and investigating
Gender-based Neurocognitive Propensities influencing Gameplay: An Interactions-
oriented approach” of Philip Bonanno from the University of Malta describes in very
precise ways important foundations of gender-based neuro-cognitive propensities
influencing game-play. Boys and girls as well as men and women show significant
differences in selecting and using games. Fiona Littleton, Jeff Haywood and Hamish
Macleod from the School of Education at the University of Edinburgh in the UK
show in their chapter, whether videogame play influence the student’s approach to
learning. Results of a comprehensive study are presented and discussed on the basis
of the relevant research literature. In the healthcare education virtual patients are in-
creasingly used. Michael Begg, Rachel Ellaway, David Dewhurst, Hamish Macleod
from the University of Edinburgh show that narrative and game play are an important
consideration in making virtual patients useful for students.

Intelligent and creative use of modern information and communication technology

in order to facilitate learning experience is a central and very important basis for
design of digital game based learning. Ulrike Spierling from University of Applied
Sciences Erfurt in Germany describes the potential of digital agents for educational
applications. The agents are imbedded in simulations, games, and storytelling. The
types of agents used in learning applications are described by showing several project
examples. Anja Hoffmann from the Design Service Team of SAP AG in Germany
and Ido Iurgel and Felicitas Becker from Computer Graphics Center (ZGDV) in
Darmstadt, Germany, describe applications of story-telling for learning purposes.
They do it on the basis of several very interesting research projects. The interesting
European research project YoungNet is presented by Fabian Kempf from the company
VITERO GmbH and Karin Hamann from the Fraunhofer Institute for Industrial
Engineering IAO. Both organisations are located in Stuttgart, Germany. The Young
Net is a virtual learning community platform for pupils in schools. The pupils can
play and learn with other pupils at other schools via the YoungNet platform. But, is
digital games based learning possible only on the basis of advanced technology? No, is
the answer of Lotte Krisper-Ullyett from the factline.com company in Austria, Johann
Ortner, professor at different universities in Austria and Barbara Buchegger from the
University of Natural Resources and Applied Life Sciences in Vienna. They show that
on the basis of simple internet platforms 1000 “Low Tech Games” can be played for
learning purposes.

6
What is the effect on pupils and students when learning with digital games?
Chris Brannigan and Angela Owen from the software company Caspian Learning in
Sunderland, UK, present a case study and show how pupils and educators react to
game based learning applications. An extensive evaluation study with students at
university level is presented by Michael Burmester, Daniela Gerhard and Frank
Thissen from the Stuttgart Media University. The used a dramaturgic approach to
e-learning in an e-learning course with over 130 participants, and found what is
good and what could be better when using drama in story based e-learning.

After having presented theories, foundations, applications and studies concerning

digital games based learning it is time to talk about quality standards. Claudio Dondi
and Michela Moretti from the research centre and service provider organisation
SCIENTER in Bologna, Italy, focus in their contribution on quality in e-learning
and quality of digital learning games.

We hope that you gain insight of digital games based learning and that you enjoy the
interesting contributions of all the authors. Last but not least we hope to see you in
the SIG-GLUE community on www.sig-glue.net. It would be nice to discuss and to
exchange interesting information on digital game based learning with you.

Michael Burmester, Daniela Gerhard, Frank Thissen

Stuttgart, May 2006

Acknowledgements
First of all, we thank all the authors for providing so brilliant and interesting
contributions. In our opinion they enable a broad view into diverse aspects of
digital games based learning. Thank you for all your hard work.

We would like to thank the Universitätsverlag Karlsruhe to support us in publishing

this book. Their pragmatic and supportive approach to all our questions and require-
ments was very helpful for us in planning and producing this book.

A special thank is for Carsten Heyler for his splendid layout work and his patience
with all our changes and refinements until the print version was finished.
Thanks to Richard Slade for reading and correcting the English of all the contributions.

7
8
SIG-GLUE: A Special Interest Group for
Game-based Learning in Universities and
Lifelong Learning
Dr. Maja Pivec and Anastasia Sfiri

Abstract
The SIG-GLUE community has been formed to provide a world wide communi-
cation and exchange platform for game-based learning in the field of universities
and lifelong learning. The community could be seen as “knowledge pool” i.e. a
resource of guidelines and practical examples that can be taken over thus facil-
itating the application of game-based learning. In the paper we describe formation
process, theoretical background, tools, activities and resources of the SIG-GLUE.

Key Words
Game-based learning, special interest group, communities of practice

1. Introduction
Although games are part of children’s growing up and formal education, digital game-
based learning is a novel approach in the area of Universities and Lifelong learning. In
the universities’ search for new positioning in the changing setting of lifelong learning,
gaming is becoming a new form of interactive content, worthy of exploration.

One of the European Projects, exploring this topic is Minerva project UniGame:
Game-based Learning in Universities and Lifelong Learning (UniGame). Goals of the
UniGame project were as follows: to promote digital game-based learning in Europe,
to test different educational games within different subjects in various European
countries and to focus on social game forms that include virtual communities and
collaborative learning. More details on research results are sampled in guidelines for
game based learning (Pivec et al., 2004).

9
A follow up project in the area of games for learning is SIG-GLUE (SIG-GLUE).
The aim of SIG-GLUE is to foster more and better use of better learning games as
well as to strengthen competencies and support practitioners in the application of
games for learning.

The aim of SIG-GLUE is establishing structured collaboration and research in the

game-based learning area, exchange of knowledge, experience in the game-based
learning, monitoring the quality and establishing a quality stamp for game-based
learning resources, contributing to innovation of the European and Worldwide
institutions and Universities.

Main objectives of the SIG-GLUE project are as follows:

• To research and promote game-based learning and other related novel and
innovative approaches to learning in the field of formal and informal education
i.e. universities and lifelong learning.
• To foster production of educational games that are based on pedagogical and
didactical principles.
• To improve quality of educational game products by defining a clear specifi-
cation of educational and gaming elements and by establishing a quality stamp.
• To organize events where community members could meet in person and
exchange ideas and experiences.
• To organize workshops on educational games application, to disseminate the
ideas of game-based learning and show best practices.

The synergy of SIG-GLUE project with UniGame, provides a unique opportunity

to continue to focus on game based learning and start with an active promotion of
game-based learning approach worldwide. SIG-GLUE is designed to offer trainers,
educators, teachers, students, game developers and designers and policy making
institutions a technological and a social environment that not only supports and
allows interactions among people who belong to a Special Interest Group, but an
environment that also encourages the formation, development and sustainability of
a virtual community in the future. The SIG-GLUE mission statement, the tools for
resources and knowledge sharing among the SIG-GLUE members as well as the
strategies for group work and collaboration have been developed with this implicit
aim in mind.

10
2. Formation of the SIG-GLUE
Creation of the community is a process of establishing a theoretical basis for the
community, defining possible features and services for the users, defining community
committees i.e. scientific committee, defining tools and setting implementation
priorities, outlining sustainability issues. Essential parts are outlined in this section.

2.1 Theoretical Background

Wenger’s Community of Practice (CoP) model is one of the theoretical models on
the notion of learning and professional development within a community. According
to this model there are three dimensions which we need to support in order to assist the
development of a virtual community. (1) The domain is the area of interest, which
creates the common ground and a sense of belonging. (2) The practice is the body of
shared knowledge and resources, which enables community members to develop and
deal efficiently with the domain and (3) the community dimension is the social frame
and arena for the learning and knowledge exchange.

The domain, community and practice dimensions refer to areas, which must be
acknowledged in order to create the right “ecology” to encourage CoPs. In the
same way that knowledge is emergent and organic so are CoPs. This also means
that CoPs cannot be managed in the traditional sense of ordering. Instead one
can encourage, cultivate and nurture these three tightly interconnected, dynamic
dimensions. Domain, practice and community help identify potential communities
and leverage points, which help implement CoP inspired initiatives (Sfiri et al.,
2003; Pauschenwein et al., 2003).

2.1.1 Domain
At the building stage of a community it is important to define what the community is
going to be about and how it is going to define its role. “Without commitment to the
domain, a community is just a group of friends. A shared domain creates a sense of
accountability to a body of knowledge and therefore, to the development of a practice.”
(McDermott et al., 2002).

11
Although the domain reflects the members’ own view of what they think is impor-
tant, it is not a static dimension of a community. If new problems or more pessing
questions arise, perhaps under the pressure of external changes the members of a CoP
redefine the domain. Within the SIG-GLUE community of practice the domain
question is already well defined in the area of Game-based Learning for Universities
and Lifelong Learning. However, what needs to be defined and communicated is the
role the community is going to play within this domain. This role, which should cor-
respond to the user requirements, can be expressed in a mission statement, indicating
the purpose for the community. This statement will determine who will join; it will
influence members’ behaviour and the practice.

SIG-GLUE mission statement is “More and better use of better learning games”. The
steps of achieving the mission as follows:

• Convince non-users of games of their educational value. (here we contact the

potential users of games, pointing out innovative approaches for learning and
their educational value)
• Encourage and support game developers in the creation of better educational
games. (eventually also pointing out curriculum relevant issues that could
easily be integrated into the commercial product, or bringing together educators
and content developers and enable a fruitful exchange)
• Make educators aware of how to use games more effectively in education.
(exchange of good practice worldwide thus making easier to apply new ideas
in their own classes)

2.1.2 Practice
Whereas the domain is the area of expertise, which the community focuses on, the
practice is the specific knowledge the community shares, develops and maintains
(Wenger, 1999; McDermott et al., 2002).

Successful practice development depends on a balance between joint activities, in

which members explore ideas together, thus encouraging tacit knowledge types and
the production of explicit tools such as documents or Websites. If CoPs do nothing
more than discuss current member problems, without documenting the insights they
develop, they run the risk of developing “amnesia”. This is to say that because they
have no record of insights they have already developed, when similar issues arise they

12
rework ideas, which they have already discussed. In the long run this “déjà vu effect”
can be deadly to a community, since it makes participation seem unproductive. More-
over, sharing too much tacit knowledge without documentation excludes non-members
from benefiting from the knowledge, which the community has already developed.
Tacit knowledge needs to be externalized in a concrete way in order to be a useful
resource to more people. On the other hand, too heavy a focus on gathering explicit
knowledge and “documentalism” is another danger. It is wrong for a CoP to spend
vast amounts of time updating entries in databases rather than concentrating on
sharing knowledge. The body of shared knowledge and resources enables the CoP
to develop and deal efficiently with its domain.

CoP members are peers in the execution of “real work” and what holds them together
is a common sense of purpose and a real need to know what each other knows.
Developing the practice also develops members’ own communal resources, which
include both tacit and explicit aspects of the community knowledge. They range from
documents and other objects, which make the practice explicit, to tacit knowledge
types such as a common perspective or a certain thinking style. To develop a practice,
members of a CoP must learn to talk to one another using words in the same way
and communicating with a common vocabulary.

2.1.3 Community
The community element presents the social frame and the arena for the learning
processes to take place. It is a very important element since the human relationship
is the basis for learning within this social-cultural perspective. A strong community
element is essential for a CoP to be effective, because the construction of knowledge
as well as the knowledge sharing processes depend on a well-functioning relationship
among the learners. It is “people” who drive the whole structure, not best practices or
websites. This is also why members need to be aware of how to behave towards one
another. A CoP can be seen as a group of people who interact, learn together, build
relationships and in the process develop a sense of belonging and mutual commit-
ment. Core members of a CoP need time to learn about one another and discover
how to operate as a community.

The challenging work for SIG-GLUE has been the application of the three
dimensions in order to encourage the formation of a working virtual community.

13
2.2 User Analysis
User Analysis is process of identifying all possible stakeholders and grouping them
into the user groups. User analysis forms a solid base for further actions i.e. creation
of the community, defining IT tools, providing features and services, dissemination
plan and activities.

The phase of Special Interest Groups (SIG) user analysis is focused on research for
potential SIG members (i.e. practitioners, educational institutions, game developers,
researchers, etc.), to be able in further steps to contact potential members and to
inform them about the formation of the SIG, publicly announce the SIG along with
intention of the SIG and roadmap, get responses from the stakeholders, analyse and
classify them according to topic of interests.

Enclosed is the list of questions related to the SIG users that formed the basis for the
more detailed user analysis:

• Who are the target groups and what do they do? Collection of information
and characterization in relation to Game Based Learning (GBL).
• What is important for them? What are their views and their needs?
• Who provides them with support for their needs?
• What kind of service would they like?
• How can SIG-GLUE provide this to them?
• What can they offer SIG-GLUE?
• How will we approach them?

By providing answers to these questions we also define areas of interest of each parti-
cular user and certain user groups and win-win situation and information exchange
between SIG-GLUE and users.

To elaborate a user analysis different various creative techniques such as brainstorming

i.e. thinking creatively on-demand, user analysis in groups and elaboration of personas
were carried out. Brainstorming gave us a list of different kinds of users. We grouped
them into two general categories: active and potential users. The project team decided
to focus on active users in the first place.

14
Active SIG-GLUE users can be further classified into categories as follows:
(1) Super users i.e. policy makers, organizations that are interested in innovation
of educational process and organizations fostering ICT and education
(2) Students and researchers and other users of learning games
(3) Game developers, educational game developers and digital learning
content providers

To build a clear picture of users’ needs, questions and expectations from SIG-GLUE,
we made a set of personas for each of user categories listed above. The Personas method
was first presented by Alan Cooper in his book The Inmates Are Running the Asylum
(Cooper et al. 1999). In his book Alan Cooper claims “… Whenever I hear the phrase
‘the user’, it sounds to me like ‘the elastic user’. The elastic user must bend and stretch
and adapt to the needs of the moment. However, our goal is to design software that
will bend and stretch and adapt to the user’s needs. […] In our design process, we
never refer to the ‘user’. Instead, we refer to a very special individual: a persona.”
When designing a UI, Alan Cooper suggests defining user personas, i.e. virtual
persons who represent typical people belonging to the product’s target group(s).
This persona will play the user’s role in the UI design process. It is for him or her
that the User Interface is designed. The reason for using personas in the design process
is primarily that the persona only has collective, average properties and need, is
always available and after a while will be well-known by the designers and engineers.

Based on detailed user analysis we created personas, i.e. virtual persons with specific
needs for different users and user groups. Based on the user analysis and created
personas, SIG-GLUE services and tools for specific user groups can be defined.

2.3 Community tools

SIG-GLUE offers a range of community tools for effective communication and informa-
tion exchange of stakeholders. SIG-GLUE is an open source web-based community,
where trainees, educators, game designers and game developers can collaborate in the
innovation of learning in terms “make learning fun” and for construction of better
educational games. The platform is based on collaborative and community knowledge
building paradigms. The following tools are merged within the SIG-GLUE site:

15
 (1) Dissemination of information: Announcements, bi-monthly Newsletter,
Glossary of game based learning, Libraries (of papers, good practice),
Games (list of games for learning, game providers, etc.)
(2) Synchronous and asynchronous communication and collaboration tools,
such as the forum (public and private, that can be used for closed sessions of
working groups), personal messages, polls, comments, e-mail, and similar.

The majority of the SIG-GLUE community contents are open to anyone interested,
however to be able to participate in the discussions, moderate discussions, initiate
new working groups and access to all libraries, one has to register. As registered user
one gets optionally also an automatic e-mail announcement about the new issue of
newsletter.

3. SIG-GLUE community
The essential focus of the SIG-GLUE project is to foster exchange of good practice in
game-based learning and innovative learning approaches as well as the promotion of
game-based learning approach per se. Based on the project goals, theoretical principles
and appropriate user analysis, a SIG-GLUE community page was developed. SIG-
GLUE community page provides on one hand access to community tools and on the
other hand gives information about the project and team of people involved in this
project. The main activities of the SIG-GLUE community are as follows:

• Newsletter is a regular bi-monthly SIG-GLUE electronic publication to inform

community members about ongoing research, discussions, coming events, etc.
• Discussion within different moderated working groups (formed by project
stakeholders) by means of community tools. Communication and information
exchange in the community helps education and lifelong learning practitioners
in :
(i) Getting and contributing information on the pedagogical and
didactical issues related to game-based learning,
(ii) Assessing quality of the games from educational point of view,
(iii) Utilizing the game and e-learning platforms for playing educational
games,
(iv) Getting support in the selection, development and modification of
games for learning.

16
 • Quality stamp for the games suitable for learning purposes. Individuals and
game producers can submit their game prototypes or fully developed games for
the quality review. The review is carried out by a list of independent evaluators
following the quality criteria for learning games. More details about quality
stamp, submission process and quality criteria are outlined in the community
section Games.
• SIG-GLUE events on national and international level, fostering the knowledge
exchange (in form of symposia, workshops, etc.) and creating possibilities for
networking.

Apart from discussions within the working groups, SIG-GLUE offers also other
resources, such as e.g. a glossary of game-based learning, where one can look up
definitions and examples. There is also a possibility to comment and extend the
definition. Another pool of resources is the library, where members can find and/or
contribute book reviews, good practice examples, etc. A separate module is focused
on games and provides collection of games, game providers, interesting game links,
actors in this area, etc.

SIG-GLUE is an open community where everyone is invited and welcome to part-

icipate, contribute and organize an activity. Various community areas, benefits and
ways of participation for different user groups are presented in more detail on the
SIG-GLUE walkthrough DVD that is enclosed in this publication.

17
4. Conclusions
SIG-GLUE promotes game based learning approach worldwide. However we are
aware that there are many practitioners for whom discussion in English could be a
barrier. Therefore we created also “SIG-GLUE around the world” discussion, where
we invite and support people to establish work groups in their own language e.g.
working groups in Spanish, Polish, etc.

SIG-GLUE and related projects contribute to an active promotion of game-based

learning approach worldwide and give inspiration and encouragement for the
development of innovative learning approaches by practitioners. We hope that game
based learning will get more support from policymakers e.g. national ministries for
education in the form of including game-based learning into curricula thus supporting
directly the work of practitioners. We hope that game based learning will remain a
priority of technology enhanced learning foundation programmes, thus improving
practice and contributing to educational innovation.

5. Acknowledgements
SIG-GLUE: Special Interest Group for Game-based Learning in Universities and
Lifelong Learning is an EC eLearning initiative Project (Agreement No: 2003-4704/
001-001 EDU ELEARN). Many thanks to everyone involved in the project for their
contributions, fruitful discussions and excellent work that contributed to the progress
of the project.

6. Authors

Maja Pivec: Her research is focused on game based learning, adaptable e-learning
systems, knowledge management, and multimedia knowledge modules. For her research

18
achievements Maja Pivec received in 2001 the Herta Firnberg Award (Austria) in the
field of computer science. She was involved in the international learning standard
development carried out within the IEEE LTSC committee.

Anastasia Sfiri: Her work in the area of computer supported learning includes research
and development on pedagogical and didactical approaches for eLearning, Computer
Supported Collaborative Learning, Communities of Practice and Game-based Learning.
Anastasia studied Psychology (BSc) at the University of Wales, Bangor, and graduated
the Human Communication and Computing course (MSc) at the University of Bath
in Great Britain. Since 2003 she works for the ZML—Innovative Learning Scenarios
at the FH Joanneum GmbH in Graz, Austria. She is the project manager and admi-
nistrator of the SIG-GLUE virtual community.

7. References:
[Cooper et al. 1999] Cooper, A., Saffo, P.: “The Inmates Are Running the Asylum:
Why High Tech Products Drive Us Crazy and How To Restore The Sanity”,
Sams, 1999.

[McDermott et al. 2002] McDermott, R., Snyder, W. & Wenger, E. (2002): A Guide
to Managing Knowledge, Cultivating Communities of Practice, Boston

[Pauschenwein et al. 2003] Pauschenwein, J., Matzer, M., Sfiri, A. (2003). Building
and Sustaining Communities of Practice in the Virtual World: From Theory to
Practice. Proceedings of the International Workshop for Interactive Computer
Aided Learning (ICL), CD-ROM Ed., Carinthia Tech Institute, Villach/Austria.

[Pivec et al. 2004] M. Pivec, A. Koubek, C. Dondi, (Eds.): Guidelines for Game-
Based Learning (Pabst Science Publishers, 2004)

[Sfiri et al. 2003] Sfiri, A., Matzer, M., Pauschenwein, J., Shaw, M., Sime, J.A.
(2003). VirRAD: A New Paradigm for Technology Enhanced Learning. Kerres,
M. & Voss, B. (Ed.). Proceedings of the 8th European Conference on Media in
Higher Education (GMW03), University Duisburg-Essen/Germany. Digitaler
Campus, Münster: Waxman Verlag. ISBN 3-8309-1288-9

19
[SIG-GLUE] SIG-GLUE: Special Interest Group for Game-based Learning in
Universities and Lifelong Learning; project web-page. Retrieved 15. 03. 2006,
from (http://www.sig-glue.net)

[UniGame] UniGame: Game-based Learning in Universities and Lifelong Learning;

project web-page. Retrieved 15. 03. 2006, from (http://www.unigame.net)

[Wenger 1999] Wenger, E. (1999). Communities of Practice: Learning, Meaning

and Identity, New York.

This paper was presented in part at the ED-MEDIA 05 conference.

20
21
22
Implementing and supporting a game
based learning related community
I. Antonellis, C. Bouras, A. Gkamas, V. Poulopoulos

Abstract
A web-based community aims at providing communication and collaboration
tools for a special interest group. This paper describes the functionality and archi-
tecture issues of a community whose aim is to bring together users who are interested
in the field of game based learning and lifelong learning. The members of the
community are provided with tools in order to share their knowledge and experi-
ence in game based learning through enhanced forums and chats, to read news or
receive a newsletter concerning the aforementioned issue, arrange meetings, and
make use of shared spaces. All these services are enhanced in order to meet the
needs of this special interest group which is unique as it includes the collaboration
of game developers, pedagogues and users. Furthermore, we describe a method-
ology to build a fully functional community with tools for communication and
collaboration starting from a simple template and using core content management
techniques.

Keywords
Online community, collaboration environment, communication environment,
phpnuke, mobile users

1. Introduction
The constant expansion of the web has affected almost any on-line community.
People feel the need to communicate, to collaborate, to share their knowledge and
express their opinions and thoughts. The growth of on-line communities is great, not
only in the number of members but also in the quality of the services they provide.
Previous concerns that online communication is hostile, divisive and uninhibited
(Kiesler and Sproull) have been overcome over the time. People are very familiar with

23
the use of devices like personal computers, mobile phones, PDAs etc for their daily
communication needs.

This need for communication is one of the most important reasons for the creation
of web based communities. The ease of being member of such a community and
the simplicity of communication and collaboration between so many people makes
communities a must for most internet users.

All the above led to many attempts that try to motivate users to act as groups. The
idea is really simple: create a web site which is the base of the community and support
this site with web service that can promote communication and collaboration. In
order to construct a successful community, “members must be able to fulfil their
purpose and accomplish those goals and interact with other members” (Ferguson et al.
2002). Now existing communities have reached outrageous numbers and the services
they provide are at least high level. People want to learn about the communities and
seek communities that best match their needs.

One community on game based learning is Special Interest Group for Game based
Learning in Universities and lifelong learning (SIG-GLUE1). Game-Based Learning
is an issue that concerns game developers, educators and trainees. Teaching methods
based on educational games are expected to be extremely attractive to either University
students or people who are involved with Lifelong Learning. In addition, the social
and educational aspect of this type of communities (Bouras et al., 2003) is becoming
increasingly interesting both from a technological and social perspective. Besides, the
Internet’s characteristics facilitate the development of unique forms of interpersonal
and group interaction. (Oren et al., 2002). Therefore, we intend to create tools for an
on-line community where the users can be organized into groups to discuss their ideas
and exchange their knowledge.

SIG-GLUE is a web based community for people who are concerned about game
based learning in universities and lifelong learning. The primary objective of the
community is to inform the public about issues of game based learning. In a second
phase the community wants to bring together educators, trainees, game designers and
game developers who will collaborate and communicate in order to produce better
educational games.

1 SIG-GLUE, Special Interest Group for Game based Learning in Universities and lifElong learning, , eLEARNING INITIATIVE, European Union.

24
After some months of use the community seems to be able to operate in an autonomous
way, without the support of the creators. More than 200 people are members of the
community and they communicate daily, expressing their thoughts and sharing their
knowledge.

This paper is structured as follows. Firstly, we present related work. Secondly, the
community of SIG-GLUE design issues is analysed. Afterwards, the basic architecture
is explained and the more important issues are indicated. After the architecture, the
implementation issues are reported and the changes that can be made in order to
enhance the community and provide to the users with more attractive services. In
section 6, we present the necessary changes and infrastructure iso that SIG-GLUE can
support mobile users. After that in section 7 we present the ideal scenario regarding
the functionality community of SIG-GLUE. Finally, some concluding remarks and
proposals for future work are provided in section 8 and 9.

2. Related work on game based learning

Many attempts have been made in order to motivate internet users to act as groups.
The basic idea is to create a web-based community aiming to bring together Internet
users that have similar interests. Many online communities, supporting high-level
communication tools, have tried to bring users together but failed owing to negligible
participation of the members. “In order to construct a successful community, members
must be able to fulfill their purpose and accomplish those goals and interact with
other members” (Ferguson et al., 2002). Consequently, the tools used to develop
a community are less important than the target group and the common interests
of the members of the community. The case of EverQuest, an online role-playing
game developed by SONY had huge success when 430,000 subscribers joined the
game in 2003 (Rowan, 2003). This means that there is huge interest of people in
“collaborative” playing. In our case the interest of the members is focused on the
promotion of Game-Based Learning in Universities and Lifelong Learning.

Game-Based Learning is an issue that concerns game developers, educators and trai-
nees. Our main goal is to create a community organized and adopted by the groups
listet above aiming to promote the idea of Game-Based Learning. The teaching method
based on educational games is expected to be extremely attractive to either University
students or people who are involved in Lifelong Learning. In addition, the social

25
and educational aspects of this type of community (Bouras et al., 2004) is becoming
increasingly interesting both from a technological and social perspective. Therefore,
we intend to create an on-line community where people, who are involved in the
fields mentioned above can be organized in groups in order to discuss their ideas,
exchange the knowledge of their own fields and finally participate in a Special Interest
Group (SIG) with the intention of constructing better Educational Games.

Many attempts have been made by developers, game providers and pedagogues to
create communities or games that can be used for educational aspects. TopSIM
(2002) by TERTIA Edusoft provides different business games which have been used
in business education and advanced training. With the use of TopSIM someone can
simulate different scenarios (Management, Logistic, Banking, etc) and create educational
scenarios. For example the management simulation is a model representing a company
or functions within the firm. The participants of the simulation take over the
management role of a company. They compete in groups against other simulated
companies in a simulated market. Through their decisions, the participants influence
the success of their company. Thus, they learn about the interrelatedness of a company
and about the internal and external factors that affect its economic success. The
par-ticipants learn through experience the connections within the organization and
the internal and external factors that influence the enterprise’s profitability. The
target groups are senior and junior managers, administration and authority, employees
in technical areas, students of business administration and industrial engineering and
trainees. Myzel (2002) is another online community game. The rules of the game are
created by the players themselves. The players have to select a role and try to survive
in the virtual world of Myzel with its various planets and complex social and political
life. Environmental Detectives (2002) was developed by MIT and Microsoft within
the Games-to-Teach project. It is a handheld PC game where players play as scientists
investigating a rash of health problems in their city stemming from point-source
pollution problems. Players learn the science behind contaminants such as source
pollution problems. Finally, Unigame (2002) is a project that introduces the concept
of game-based learning with a focus on higher education sector and lifelong learning.

All the above, introduce some attempts to bring Internet users together. But none of
them, except Unigame (2002) include the collaboration of developers, game providers
and pedagogists. These projects focus on specific perspectives, either the technological
(developing the games) or the social (pedagogical models). Our intention is to create a

26
community that will try to bring together Internet users with the intention to exchange
ideas, knowledge and experience in order to produce more efficient educational games,
and it will be focused on both technological and pedagogical aspects.

3. Design issues
This community will provide users with educational aspects on game-based learning
in universities and lifelong learning. It can be therefore seen as a web-based learning
tool that can contribute to knowledge acquisition and should respect a number of
instructional supporting measures (Cohen, 1994; Slavin, 1996). These measures have
been developed to stimulate learning–favourable activities and have been proven as
suitable for face-to-face situations as well as have been partially tested in order to
promote web-based collaborative education (Hron et al., 2000; Weinberg et al.,
2002). The characteristics of each community differ so, it is important to point out
the unique characteristics of the SIG-GLUE community and eamine them (A. Hron
& H.F. Friedrich, 2003).

The basic needs and requirement for communication and collaboration of the SIG-
GLUE community are organized into the following categories (Figure 1):
• need for communication with each other or in groups
• need for collaboration in groups
• need for shared spaces to store and archive information
• need for making available / disseminating information to the public

Figure 1: Web community structure.

27
3.1 Communication tools
SIG-GLUE members need to communicate among themselves in order to exchange
opinions, ask questions, offer advice etc. Communication should be both direct and
indirect, and both private and public:
• Direct: Direct communication requires that all participants participate at
the same time and that communication is exchanged instantly (i.e. with no
or little perceived delay). Direct communication is required when one needs
an immediate personal conversation with a specific person or with a group of
people.
• Indirect: This type of communication implies that the participants do not
have to participate at the same time in a conversation and that the messages
may be exchanged with a (possibly large) delay. In this type of communication
the time factor is not counted at all and therefore it is used when delay in
communication is not crucial.
• Person to Person and Group Communication: SIG-GLUE is a community
where communication between groups is of great importance. Besides, the
community is enhanced with the functionality of person to person com-
munication in order to achieve a greater level of agreement between two sides.
• Private, Public and Semi-public: Communication is called private when it
includes only the parties that are intended to participate in a communication.
It is very important not to reveal any of this information to the public. Private
communication is needed for exchanges that are of private nature, or do not
concern others and might overload them with useless information. As people
within SIG-GLUE work together it is highly probable that they might want
to communicate in private for various reasons without sharing with the whole
community (or parts of it).

However, taking into consideration that the main purpose of SIG-GLUE is to allow a
large number of people interested in Game-based Learning to communicate and
exchange results, the most important need is for public communication. All related
members—game producers, game developers, content designers and pedagogues need
to easily access all communication functionalities and information exchanges within the
community and gain the most from this. In addition to its public nature, this type of
communication needs to be archived for future reference.

28
Between private and public communication, there is a need for semi-public commu-
nication. Semi-public communication is the one that is revealed to a special group of
people, larger than the individuals communicating, but not to everyone.

3.2 Collaboration Environment

Members of the SIG-GLUE community will need tools to support their co-operative
work. This is especially true for the SIG-GLUE Working Groups where in each group
the members will have a task to execute co-operatively. After the community’s expan-
sion the collaboration environment will be converted into a shared working environ-
ment for the SIG-GLUE “family”. This means that this environment should cover
specific requirements in order to provide its members with adequate functionality:
• Document sharing: A very common task that all the members will face is the
need to produce documents for their work. The documents should be classified
and collected into a place in order to create an archive for future needs and
reference. Therefore, document sharing is essential for co-operation between
the SIG-GLUE members. This makes their work more practical and easier.
Access control is also necessary, as there will be situations where only some
members of the group could alter the documents while the rest will just read them.
• Calendar: A calendar will be useful to keep the members informed about
scheduled conferences, online meetings, and other events or milestones.
For the Working Groups a calendar can be used for scheduling tasks and
monitoring work progress. In addition each member might benefit from a
personal calendar. All these various levels of calendars (personal, Working
Group (WG)-wide, community-wide) should be mergeable so that one
may see in one’s calendar the entries one is interested in and/or pertains
to one’s work within SIG-GLUE.
• Support for workgroups: Members of the community must be organized
into workgroups. The existence of the workgroups will result in better co-
operation, more efficient work and better results. The environment must
group members of the same group together, support the super-members and
moderators of each workgroup and provide services relevant to the members
of the whole workgroup.

29
3.3 Shared Spaces
As the numbers of SIG-GLUE members increase the use of attachments to exchange
information becomes infeasible. In addition forums do not usually support attach-
ments to keep the space requirements down. An alternative is to have one (or more)
shared spaces where files can be uploaded and made available to SIG-GLUE mem-
bers. The shared spaces must include extended functionalities and specific roles for
the members because the shared spaces contribute both to communication and colla-
boration. Depending on the scope of members having access to shared spaces can be
public, or private.
• Public folders: Public folders will be virtual on-line folders that will contain
documents, deliverables and all other information that intend to be available
for the public. Availability of these folders may be a specific service of SIG-
GLUE or it can be provided as part of each workgroup area (one public folder
per workgroup). Access to the data of these folders will be provided without
any constraint.
• Private folders: Private folders will contain information and documents that
will be provided to specific users. These users may be all registered users or
specific members of a workgroup (WG). Access to these folders is controlled by
access rights management system. A special case of these private shared folders is
each WG’s shared folder.
• Databases: Databases contain structured information of various kinds e.g.
an educational games database, a provider’s database, a member’s database, a
links database etc. In these shared databases each member of the community
(subject to having rights to do so) can submit, edit, delete and view information.
• Announcements database: A special kind of such a database is a news database.
As the main purpose of a community is the easy access to news, articles, results
and the exchange of knowledge, a well organized news database where every
member of the community can submit links and articles of public interest is
necessary. The news may be divided into various categories with the possibility
that a moderator could add new categories.

30
4. Architecture
In this paragraph we present the architecture which will accommodate the above
mentioned design and functionality issues. First we present the basic platform in
which our system is based and after that we present the extensions which we have
added to the basic platform.

4.1 Basic platform

In order to build the online community in a manner that would provide all the above
communication and collaboration tools in a unique platform and allow efficient
administration and customization by the administrator we had to decide on the basic
platform to deploy and use. As many available web tools exist for implementing many
of the collaboration and communication functions the fundamental issue is to unify
all these in a manner that would result in a complete environment with unified look-
and-feel as well as functionality.

PhpNuke2 is an open source content management system for the web that allows
the easy creation of web portals that support web-communities. What really makes
it suitable for use is the big open source community that supports it by providing
additional modules and fixing possible existing bugs in the code. Despite this
expanding community PhpNuke is considered to be not a proper choice for a web
portal as it lacks security.

Taking these issues into account, we decided only to use the core mechanism of
PhpNuke that deals with the “dirty” work of content management and implement
new or modify the existing modules so as to provide the required functionality as
well as the necessary security. We describe the core mechanism of PhpNuke that we
used to construct the web community of SIG-GLUE and later in this paper we give
a methodology of using it in order to transform any static web site with communi-
cational needs into a fully functional online community.

2 http://phpnuke.org

31
4.2 Extending the basic platform
The expansion of the web has been enormous and users of the internet have become
familiar with the idea of using the World Wide Web almost for every task, simple or
not. The use of the Internet by users is changing according to technological habits
and current trends. Communities has become a very easy way to bring lots of users
together without their physical existence to a specific place or time and without having
to accomplish a task or even have to be present when a discussion takes place.

In order to achieve all these, very simple tools and lot of patience are required. Users
are now very familiar with web developments and the existing tools do not seem to be
enough for them. Most users love trying new methods of communication. They seem
to be extremely attracted by communities and web services which are not “conservative”,
but include an alternative way of doing something, even if it is a simple task.

4.2.1 The hidden power

In a previous section we referred to some reasons why some communities after a
period of growth begin to stop functioning, with most of the users being bored.
Therefore, we should find some special characteristics that may enhance the commu-
nity. A way to do this is to meet the hidden or subconscious needs of the members
of the community.

One way to enhance a community by attracting its members is by using creativity

techniques. Creativity techniques aim of generating ideas within a group of users and
promoting the creativity of each group member (Woerndl et. al., 2005). From the
ordinary users of a community to the most advanced users and to those who are very
keen on using collaboration and communication tools, creativity is always a challenge.
Especially when we refer to a community like SIG-GLUE where pedagogues, trainees,
game designers and game developers meet creativity seems to be the essential ingredient
of success. The creativity techniques can be classified into categories. The two main
categories that derive from the classification are the creative method in which ideas
are created by spontaneous thinking and the systematic method in which the ideas are
created by systematic and analytical ideas. The creative method is supported by areas
where the users can write just their ideas or express some kind of knowledge and could

32
be areas like, announcements, library, games and the forum. The systematic method is
supported by areas where one can find feedback from users. This can be mainly the
forum where we have lots of discussion on many topics.

The aforementioned issue introduces a hidden technique in order to enhance a

community. Another technique is to reassure the members about the stable and
unified structure of the community. One should think first of the community as a
super organization and as a social system, because many online socio-technical systems
function and operate similarly to offline systems, despite differences in implementation
(Flor et. al., 2005). This means that the community’s members act like they do in
their real life. After this, it is easily understood that the members of the community,
and therefore of the online society, would like to be treated as unique individuals with
different needs and obligations. They would like to feel like they are in their real life
and treat the web-based society like the offline society. In addition the role of the
administrator must be hidden and he should interfere only in special occasions like
the teachers role in a virtual class (Sotillo, 2000). So, it seems to be very important to
reassure the members of the community on hidden issues like stability and consistency.
This means that the users must be able to see that whenever they visit a different area
of the community, their actions remain the same. For the ordinary visitors of the
community every area is locked to submissions, while registered members have full
access to submit their contribution to the different areas of the community in the same
way whatever the module is. This means that members have the power to access every
area of the community in the same way they are not distracted and they actually
experience the unified framework on which the community is built.

Having a community treated as a super organization and as an offline society means

that the members need to follow some rules. But, most web-based communities do
not state clearly any rules. This leads us to the result that the community’s policies
must by implied so that members would be informed about the policies without
interfering with their use of the community. Policies are an important component of
any virtual community since they represent the foundation upon which the com-
munity can actually work. Because policies are crucial for the proper working of a
community, techniques must be followed to ensure the security and confidentiality.
(Squicciarini et. al., 2005) It is important that the members of the community are
reassured about the security of the community, which means correct flow of
information, no disclosure of personal information, encryption of private conversations
and private groups documents etc. In this way the members will be attracted to use

33
the community without any doubt or fear. All the above issues concern the “hidden
power” of a community’s elements. In the next section we will describe more obvious,
to the user, issues about enhancing a community.

4.2.2 The power of learning

There is no doubt that we live in a time where information and communication
technologies are of high importance. This factor means that more and more people
are informed about issues concerning everyone and everywhere in the world. Besides
that, the community of the World Wide Web does not only want to be informed.
The users are attracted by the learning character that has been given to the internet.
More about web-based communities and the information society can be found in
(Pyati, 2005, Ahmed and Blustein, 2005, Macher and Pathak, 2005).

Regarding a web community and more specifically the SIG-GLUE community,

its power can derive from the needs of people to learn. A web community must be
able to foster science and promote a kind of education. This may result in a very
limited web community, but we should take into consideration the positive points of
this issue. An example of this can be the community of ICTP (International Centre
for Theoretical Physics). This community addresses a quite limited audience but it
grows and becomes bigger year by year. This derives from the issue that the com-
munity has declared and confirmed its scientific and multicultural character (Fonda
et. al., 2005) which attracts more and more people. These techniques result in a specific
and especially a sophisticated character of the community which enhances it.

Learning through a community is actually an attractive feature, but how can this
learning character achieved? This can be achieved through the learning object paradigm.
A learning object can be defined as “any digital resource that supports learning”
(Wiley, 2002). The learning object paradigm has emerged during the past few years
in order to accommodate the need for sharing and exchanging reusable learning
resources on the Web (Chatzinotas et. al., 2005). In this way the members of the
community will be able to retrieve and offer knowledge through learning objects,
knowing that this knowledge is simultaneously converted into experience by the
reusability feature of the learning objects.

34
4.2.3 The power of mobility to the atom
The issues that are discussed in the two previous sections highlight some issues that
are actually known to be very attractive for users of the internet and members of a
community. Besides all these a community must follow the trends of the technology,
and must be flexible enough in order to support all kinds of users.

The latest trend of the technology is “mobility”. It refers to what its name says: provide
the users mobile access and functionality. A web-based community is actually build
on the internet. No access to the internet means no life for a community. In addition,
the members of the online community can interact with each other (communicate
and collaborate) only when they are in front of their personal computer and at the
same they must be connected to the Web. A large flourishing community will have
enthusiastic discussions and will change from hour to hour. This means that if a
member of the community has not logged in the last two or three days to the com-
munity web site he may become isolated from the other members because he has
become out of touch.

Such a community must be enhanced with mobile features. The members of the
community should be empowered with the ability to connect to the web-based
community through a mobile device like a phone or a PDA. In addition they should
be able to participate in most of the features of the community through the mobile
devices. This means that firstly the community must support small screen devices,
following the open standards, and secondly promote the functionalities in such a
way that a mobile user can access and use them with the limited tools (no mouse
and no full keyboard) that the mobile device offers. Besides that the users may be
also informed about postings in the community via text messages or multimedia
messages to their mobile devices.

The mobility seems to be crucial and we think it is an issue that will concern future
web communities. The active members of a community really need to be always
informed about every action in the community and they actually demand universal
access.

35
5. Implementing issues
In order to provide SIG-GLUE users with the above described services using the plat-
form that PhpNuke offers, we had to develop specific new modules and edit the exis-
ting modules to meet our specific needs. What we actually used from PHPNUKE was
its core content management mechanism.

5.1 Inside PhpNuke

PhpNuke has a main module that is responsible for the management of the site
content. The functional characteristics of such a module depend on the definition
of the content that it must manage and its key features as well as the fulfilment of
specific goals concerning the administration of the system and its easy expansion.
That is why in order to understand how PhpNuke’s core works we need to describe
these parameters:
• General page layout: The system that PhpNuke generates consists of web pages
with the wire frame layout shown in Figure 2. Every page consists of five main
elements, the header, the footer, the left section of blocks, the right section of
blocks and the main module.

Figure 2: Page Layout.

36
 • Expansion capability: The architecture of PhpNuke supports the easy addition
of new blocks and modules. The system is responsible for identifying a new
contribution and automatically properly initializes all relevant database data.
From a developer’s point of view, code guidelines are provided in order to
successfully integrate an individual web application into a module or a block and
it is possible to embody specific PhpNuke’s entities (such as registered users or site
administrators) into any module using the provided function library. In addition,
specific instructions are provided in order to integrate module’s administration
functions into PhpNuke’s administration panel.
• Site’s Administrator Role: The entire system is administered through an
administration panel to which specific users (called site administrators) have
access. Administration permits customization of general attributes of the site
(name, description etc), specification of the visible blocks and modules, their
position and the users groups that have access as well as per-module and per-
block administration utilities (defined from each module/block itself ).

5.2 SIG-GLUE services and implementation

The SIG-GLUE final platform that we implemented started from a simple template
and through a specific procedure (Figure 3) resulted in the creation of the services.
It is clearly shown that any menu-driven web template can easily be converted into a
full functional web community. Implemented modules supply services for Forums,
Chats, (communication tools), Calendars and on-line libraries (Collaboration Tools),
news database and public-private folder (shared spaces).

Figure 3: Basic Platform.

37
5.3 Hiding the power into the architecture
What is to be discussed in this paragraph is the hidden power that should be included
in the architecture. Talking about hidden power, we refer to the techniques that should be
hidden in such a way that it will not interfere with the users of the community’s interface.

The architecture should support personalization and unification of the modules of the
community. The community should have a stable and unified graphic user interface
with integrity of information and security. What we did in order to enhance the com-
munity’s integrity and security issues was to omit any unused functions, and modify
all the functions that may contain any malicious code. We changed the interface in
order to achieve unified and stable layout and concluded with the layout that is
shown in figure 4.

Figure 4: Screenshot from the main page of the community.

As we can see from figure 4 we moved the right hand menu of the PhpNuke platform
to the top of the page in order to achieve a unified layout throughout the different
modules of the community, because we faced occasions where modules covered the
right hand menu. The menu to the left is the first level menu with the second level
menu coming right after the title of the page as we can see from figure 5.

38
 Figure 5: Announcements module with submenu right after title: “Announcements”.

After making the layout unified we had to change the module functionality in order
to give life to our super-organization to act as a standalone system. The community
became an “open community” where the moderators were not needed to vet the con-
tent from members. The members are free to use any functionality of the community
apart from these that could harm the content or the integrity of information. On the
other hand special attention was given to the “private” spaces of the community in
order to reassure members about the high security level. In these ways we empowered
the community’s core system.

5.4 Promoting the “learning” characteristic

SIG-GLUE is a community that aims to bring together people who are concerned
about game based leaning (GBL) in universities and lifelong learning. Considering
who might be in this group of people, we concluded that they are mostly people who
are concerned about science and research, so it could be useful for the growth of the
community to give a more sophisticated atmosphere to the community. Scientists
and Researchers need to work in a sophisticated environment.

39
Firstly, we created places where one can learn about the community, in order to
convince people who are aware of GBL that the community’s goals include GBL.
In addition we created means of synchronous and asynchronous communication in
order to begin the discussion groups for the people who want to learn through the
community. You can see a means of communication in figure 6. Besides that it was
essential to create databases and shared spaces in order to store the learning objects
and convert them into experience and references.

Figure 6: Discussion Area-Forum.

The expansion capability of PhpNuke helped us creating the appropriate tools for
promoting the learning characteristic. The Glossary, the useful Links, the Good
Practices that can be found in the community imply a place of knowledge spread and
transfer more and more people.

The community is separated into places that can be classified from the access that the
users have and from the type of content. Creating something that can be compared to
a real library, the library of the community of SIG-GLUE includes almost everything
that can be useful for someone who wants to be fully informed about GBL in universi-
ties and lifelong learning.

40
6. Supporting mobile users
The next step to be taken is to support mobile users. An easy way to inform users
about the change of the community and the support of mobile devices is information
through mobile devices. The members of the community will be informed through
SMS or MMS about news of the community, new discussions in the forum, new
forum posts that possibly concern them, upcoming events etc. This first step seems
to be crucial and will stimulate a debate. From the results of this “debate” we will
understand whether the users will be attracted by the community’s mobile character
or not.

If the results are positive, a redesigning of the community will begin. The redesigning
will concern only the code of the community because it will have to be changed in
order to be supported by small screen devices. This means that either the html pages
will be converted into a language supported by all the mobile devices or that the most
crucial parts of the code will be rewritten in order to become visible and fully functional
on a small device.

Most of the functionality must be changed because the mobile devices are not easy
to handle when browsing the internet. This means that one can post a reply to the
forum by sending a voice message, or synchronizing the calendar of his/her PDA
with the calendar of the community, downloading the events that may concern
him/her.

6.1 Transforming the Community

In trrying to investigate how all these needs of traditional Internet users can be
transformed into requirements of mobile users we meet two main types of obstacles.
Firstly, the interface of the communication tools should be changed so as to take into
account small screen restrictions. Moreover, community services, that is communication
and collaboration tools, as well as shared spaces must be revised in order to permit
easy access through mobile networks, utilizing more efficient techniques in order to
meet low bandwidth restrictions.

As SIG-GLUE’s main aim is to inform its members about current trends in game
design, we can easily apply transformation techniques derived from mechanisms

41
to push news into mobile devices. Such mechanisms take into consideration both
restrictions and extra capabilities of the mobile interface and the mobile network
resulting in an optimum use of the community tools. Moreover, these techniques
have also designed in order to exploit mobility behaviour of mobile users (Albert and
Kim, 2000; Sharples, 2000).

6.2 Characteristics of mobile device’s interface

Mobile devices introduce a very different interface for their user. Small screen size,
slow text input facilities, small storage capacity, limited battery life, low bandwidth
network capabilities and slow CPU speed are main characteristics of such modern
devices. In particular, screen size limitations directly affect the user’s behaviour. Recent
studies (Kawachiya and Ishikawa, 1999; Buyukkoten et al., 2000) on the effect of
screen size on completing task related with browsing, show that mobile users tend to
follow links less frequently than traditional Internet users. This behaviour comes from
the more conservative link exploration trend of small screen users. This implies that
we must provide user with information within the first two or three links he visits in
order not to lose him.

In addition, as many different small devices exist equipped with web browsers that
support different content types (HTML, xHTML, wml, multipart, mixed), it is also
important to be able to identify the user agent that a mobile device users and provide
it with the corresponding edition of the community pages.

6.3 Content Transformation Techniques and Mobile Services

Mobile edition of the aforementioned communication tools should take into account
special requirements of mobile devices as well as the expected change of mobile users’
behavior while browsing through community services. It is clear that both structural
design of the community and community’s services must be transformed in order to
meet very different needs.

The major challenge of displaying community pages into small screen devices such
as PDAs or smart mobile phones is the content transformation into an appropriate
layout that will enable users to access it easily. Standard transformation techniques

42
have been widely studied (Alberts, 2000a; Alberts, 2000b; Bickmore, 1997; Buyukkokten
et al., 2000; Buyukkokten et al., 2001a; Buyukkokten et al., 2001b; Danielson, 2002;
Dyson and Haselgrove, 2001; Jones and Marsden, 1997; Jones et al., 1999; Feiner, 1998)
and include direct migration methods, data modification techniques as well as data
suppression migration methods.

Direct migration does not require human or system intervention as there are no trans-
formations made to the original web page. It does require more effort to navigate the
pages by users as only a small part of the page is visible at one time (Alberts and Kim,
2000a; Dyson and Haselgrove, 2001; Jameson et al., 1998)

Data modification techniques involve the creation at source of duplicate Web pages
that are intended for use on the small screen by reducing image’s size or summarizing
text. Data suppression methods remove data to make small-screen versions display
only skeleton information of the source data resulting in simplified navigation and
reduced disorientation.

The selected transformation technique is very similar to the Gateway methodology

(MacKay, 2003). According to this method, whenever users visit a web page, they
construct a mental model about the structure and the content of it. What is critical
while transforming the content is to minimize the adaptation of the existing mental
model when migrating between devices so as to reduce volatility. Characteristic factors
that form this mental model include page layout, scrolling attributes, distance to
information (path of links that must be followed), graphical interface and information
density (Danielson, 2002; Spence, 2001).

What Gateway tries to do is display the page designed for a large screen on the small
screen by reducing the web page in scale to fit the screen. The same ‘look and feel’ is
replicated using same layout and design. However, blocks of information are identi-
fied so as to allow users navigate included information as they would on the large
screen (Figure 7).

43
 (a) (b)

(c) (d)

Figure 7: SIG-GLUE community seen from the Symbian Opera browser of the smart phone Sony Ericsson P800.
(a) without any transformation, (b) after transformation according to the Visual Gateway schema.
Figures (c) and (d) show the corresponding content of a visual block of the main page with and without zoom.

The transformation mechanism of SIG-GLUE’s community interface can be

summarized as shown on Figure 8.

Figure 8: Content transformation mechanism.

44
7. Functionalities—The “ideal scenario”
7.1 Community functionality
Concerning the functionalities of the web-based community it is important to analyze
and address the interactive features. Advanced features and innovative services attract
members of the community to participate and collaborate. The collaboration of more
members has another positive point. The more people are concerned about the issue
of Game-Based Learning; the more people express their ideas, opinions and experience.

An important issue of an on-line community is the Graphic User Interface. It should

be simple enough so as not to confuse the users but also intelligent enough in order
to guide the users through the advanced features and consequently encourage them
to use the tools which promote communication and collaboration.

SIG-GLUE is a web-based community aiming to bring together people whose

interests are focused on game based learning. Consequently, the graphic user
interface should be game like. This means that it should be colourful and cheerful,
without transcending the limits of the formality of its educational goal. The interface
should have an unequivocal hierarchy to help the users navigate the community. As
a result of all the above, the community has menu based navigation, with main
menus, some of which have a submenu.

The Home button leads the user to the first page of the community. Some extra
information can be found in the second section which is the Information about SIG-
GLUE. Following is the Announcements. In this section the users will be directed
to news concerning the community and technological advances in their field of interest.
Besides, there is the Newsletter of the community. Interaction is focused on the articles
of the news or the newsletter, as the members can submit their own articles. The
Forum button leads to the interactive communication. There are the forums of the
community which are separated into the general discussion forums and the work-
groups’ forums. After this section there is the library. It contains a submenu, which
has a link to the shared space, to the glossary and the link “Links” which leads
members to some useful links concerning game-based learning in Universities and
Lifelong Learning. Its red color indicates the power of interaction which is produced
by uploading and downloading files. The library area is followed by the games are. In
this area one can find useful information about Games that are related to game based

45
learning, the “actors” of the field, and game developers. The contact button is at
the end of the main menu of the community website and has contact details for the
community administrators. Figure 9 shows the left menu which includes all the
above links

Some extra areas that are not shown in figure 9, but are very useful for the community,
are the “Who’s Who” and “private messages” area. The “Who’s Who” area includes a
full list of the members of the community where people can get to know each other
better by viewing each other’s profile. In the private messages area the users can send
each other private messages in order to communicate in a “private” manner about a
specific issue.

Figure 9: Left menu of the SIG-GLUE community

In the following paragraphs we will describe the interaction of each group of users
with the community web-site. Firstly, we should consider the types of users that will
be members of the community. There are three types of users: anonymous users,
registered users (members), and the administrators. This hierarchy of community
users implies the level of participation of each user group that will be using the
community web-site.

Anonymous users have access only to information which refers to community itself.
In addition they have limited access to sections of the community that are interactive,
such as the forum. The community is aiming at to encourage anonymous users to
become members of the community.

An anonymous user can become member of the community by registering at the

community web-site. Little information is required for the registration such as the

46
name and a valid mail of the user. After submission of the registration form the user
becomes automatically a member of the SIG-GLUE community. This membership
allows user to interact with the collaboration and communication tools. From the
“Forum” section a member has access to the forum. A member is able to view all the
conversations in the forum and post his answer to any topic. Entering the “Newsletter”
section, a member can view all the news and newsletters of the community. Members
become subscribers of the community newsletter after submitting the registration
form. Moreover, they may send their own articles for publication to the news section,
or to the newsletter. The community library gives members their own shared space
where they can upload files concerning game-based learning and lifelong learning
and download other users’ files.

The administrators of the site are actually special members, who supervise the whole
community. There are also some administrators who are responsible for specific
sections of the community. An administrator is responsible for the preservation of
the consistency of the community’s sections. He/She has special access to any service
without any restriction and in addition to the administrator section of the site which
is an extended version of the administration section of PhPNuke. From these pages
the administrator can manage the content of the community, the services and the
members. The administrator is also responsible for the shared spaces. The security of
the shared space is advanced but the administrator should vet every file added.

The hierarchy of the users is an extremely important issue for the development of the
community. The users can interact with the system without exceeding the limits and
restrictions which are necessary for any type of community. The users should act as
they do in their real life where they have obligations and rights, which are implied
through the restrictions to the services of the community. In this way, it is expected
that the on-line community will flourish.

7.2 The “ideal scenario”

After discussing issues on how to enhance the community it is the right time to explain
the “ideal scenario” of the empowered community. The functionality issues and the
first scenario of the community can be found in (Antonellis et. al., 2005). The scenario
that will be described contains a user and more specific a game designer but in general

47
the user could be an educator, a trainee, a game designer or a game developer and the
scenario would be still the same.

The dissemination of information of the SIG-GLUE community through MMS

advertisements leads a game designer to visit the community. She is attracted by the
way she learned about the community (via MMS), reads the information about the
SIG-GLUE community and decides that it is a good idea to become a member of this
community in order to share their experience and acquire knowledge.

She registers for the community and creates her personal information and interests. An
email is sent to the game designer welcoming her to the community and is followed by
some information on what could be useful for her in the community according to the
interests she submitted earlier. After browsing the pages of the community that were
suggested automatically she decides to become a member of the game designers group.
She creates a group membership account and the group moderator gives the game
designer full access to the semi-public functionality of the group.

The game designer can be now become an active member of the community. She
submits events to the calendar, takes part to the events and remains synchronized
with the SIG-GLUE calendar by synchronizing her mobile device. She takes part in
events of her group and post replies to the discussions. If, for any reason, she is away
from a computer she is able to take part in events, discussions, brainstorming by using
the features that are designed for her mobile device. In addition she uses the “library”
of the community in order to upload content and she downloads any useful content
acquiring knowledge and gaining experience.

After a short period of time, and many hours spent in the community, she has a pro-
totype idea on how to create a better educational game. She feels secure enough to
share her idea with the members of the community. A new section in the discussions
section of her workgroup begins and opinions from all members are heard. A small
survey begins on the different ideas that derived from the discussion. The members
of the group are called to vote for the best idea either through their personal computer
or mobile device (SMS, MMS). The results are posted to the community and the
ideas are presented to the other working groups. The presentations can be found in
the public sections of the community.

48
A considerable number of people discuss the idea, trying to enhance it and support it
by evidence which is uploaded to the community’s shared spaces. The idea is widely
disseminated amongst the community. Many papers and practices on the idea are
written and a universal survey and discussion begins in the community in order to
finalize the idea. The results are posted and the final version of the idea becomes a
complete proposal. The company of the game designer will design the game. Companies
from the members of the groups will deal with the rest (practices, scenario, game
developing, testing and production).

The “idea” of the game designer has become an educational game and many Univers-
ities use the game because they learned about it through the community’s dissemination
features. The game designer could have created the game alone with her company,
but through the aforementioned procedure a learning game was designed with the
collaboration of trainees, educators, game designers and game developers.

The exampled that was described introduces the ideal scenario of the community
and seems to be a chimera, but it should be noted that the community does support
all the aforementioned functionality and if one wants to follow this scenario he/she
will be able to do it.

8. Conclusions
This paper presented the technological functionality and architectural issues of a
web-based collaborative environment enhanced to support Game Based Learning.
The features of this environment include Shared Spaces, Communication tools and
Collaboration tools. These features are implemented by creating an extended platform
based on PhpNuke. PhpNuke and generally the use of PHP-MySQL technology
was chosen because we believe its open source style harmonizes with the style and
tone of the open web-based community we intend to create. This platform includes
all the necessary tools that are needed by the members of a community. In addition to
the standard tools, some integration is necessary to produce adequate features for the
members. Besides the basic tools, many more modules and blocks were created in
order to provide static pages or features that are not implemented in PhpNuke. We
analysed the demands of the community and specified the needs of communication
and collaboration of the members. Together with the above we described the role
model of the members in the community thoroughly, because we believe that the

49
success of a community is based on the cognition of the target group rather than the
use of advanced and perfect features. This means that firstly we focused on establishing
the target group and its special needs and after that modified the basic architecture.
In addition, this paper presented techno-logical functionality and architectural issues
concerning game based learning for mobile users. The mobile specific edition of a
web community with communication and collaboration tools as well as shared spaces
was presented and mobile specific services were analysed. Web content transformation
techniques for mobile devices although useful need to be altered in order to adapt to
mobile specific services. Enabling mobile game designers to access such a community
extends will contribute to easier access to the services.

9. Future Work
In the future we will focus on the needs that will arise through the use of the
community, trying to cover the needs of the members in communication and
collaboration issues. In addition, following new technology architectures we could
upgrade all the features and the services aiming to faster communication system and
more productive collaboration features. Features, concerning video conferencing,
voice conferencing, application sharing and smart notes tools could be added as
advanced modules of our platform or as stand alone sections which will use the same
database with the platform. In conclusion it is essential that a web-based community
should follow new technology issues in order to become, not only a huge community
in terms of numbers of members, but also to promote the collaboration and the
remote communication of Internet users in an optimum way. Moreover, we will
focus on the needs that will arise through the use of these services. It will be given
more emphasis on content personalizing techniques, that will conserve bandwidth
by providing mobile users only with content close to their interests. Construction of
the user profile will exploit implicit information about the mobile device as well as
user-defined interests.

50
10. Authors

Ioannis Antonellis was born in Patras, Greece in 1983. He graduated from ‘Peiramatiko’
High School of Patras with ‘Excellence of perfomance’ during all of his school-years.
While in High school he was honoured by the Hellenic Mathematical Union for his
perfomance in local contests as well as the Hellenic Mathematical Olympiads (1997,
1999, 2000). He entered the Computer Engineering and Informatics Department in
2001 and he is now in the 4th year of his studies. He has been awarded the National
Foundation of Scholarships during his three years of studies (2nd , 3rd, 1st-expected
among 200) and he is also scholar of Zosima Scholarship Foundation. During his
three years of undergraduate studies he has been working as network operator and
system administrator to the Networks Operating Group of University of Patras
(www.upnet.gr) and the Computer Center of CEID. He is also member of the Web-
Databases Group of Computer Center of CEID and responsible for the Students’
portal (my.ceid. upatras.gr). He has worked as web developer in WestNet A.E. and
he is a part time engineer in Bytemobile Inc. European Development Center (www.
bytemobile.com). During 2004 he joined RD6 of CTI and he is now working for
the SIG-GLUE project.

Christos Bouras obtained his Diploma and PhD from the Department Of Computer
Engineering and Informatics of Patras University (Greece). He is currently an Associate
Professor in the above department. He is also a scientific advisor to Research Unit 6
in the Research Academic Computer Technology Institute (CTI), Patras, Greece. His
research interests include Analysis of Performance of Networking and Computer
Systems, Computer Networks and Protocols, Telematics and New Services, QoS and
Pricing for Networks and Services, e-Learning Networked Virtual Environments and
WWW Issues. He has extensive professional experience in Design and Analysis of
Networks, Protocols, Telematics and New Services. He has published 200 papers in
various well-known refereed conferences and journals. He is a co-author of seven
books in Greek. He has been a PC member and referee in various international journals
and conferences. He has participated in R&D projects such as RACE, ESPRIT,

51
TELEMATICS, EDUCATIONAL MULTIMEDIA, ISPO, EMPLOYMENT,
ADAPT, STRIDE, EUROFORM, IST, GROWTH and others. He is also member
of experts in the Greek Research and Technology Network (GRNET), Advisory
Committee Member to the World Wide Web Consortium (W3C), Member of
WG3.3 and WG6.4 of IFIP , Task Force for Broadband Access in Greece, ACM,
IEEE, EDEN, AACE and New York Academy of Sciences.

Apostolos Gkamas obtained his Diploma, Masters Degree and Ph.D from the Computer
Engineering and Informatics Department of Patras University (Greece). He is currently
an R&D Computer Engineer at the Research Unit 6 of the Research Academic
Computer Technology Institute, Patras, Greece. His research interests include
Computer Networks, Telematics, Distributed Systems, Multimedia and Hypermedia.
In particular he is interested in transmission of multimedia data over networks and
multicast congestion control. He has published more than 40 papers in international
Journals and well-known refereed conferences. He is also co-author of three books (one
on Multimedia and Computer Networks; one on Special Network Issues and one
on IPv6). He has participated in R&D projects such as OSYDD, RTS-GUNET, ODL-
UP, ODL-OTE, TODAY’S STORIES, ATMNet, “Technical Consultant to Pedagogical
Institute”, ELECTRA, Teleteaching Service for Greek PTT, ASP-NG (ASP-New
Game—IST-2001-35354) and 6NET (LARGE-SCALE INTERNATIONAL IPv6
TESTBED—IST-2001-32603) and is currently involved in projects IPv6—GRNET,
SIG-GLUE (Special Interest Group for Game-based Learning for Universities and
lifelong Learning-Agreement No: 2003-4704/001-001 EDU ELEARN) GN2 (Multi-
Gigabit European Academic Network—Contract Number: 511082) and Technical
Support for the implementation of Metropolitan Broadband Networks based on
Optical Fibres in the Region of Western Greece.

Vassilis Poulopoulos was born in Kalamata, Greece in 1982. In 1998 he participated

in the local contest of the Hellenic Mathematical society (Thales) and he passed in
the National contest (Archimedes) of the society above. In 2000 he entered the
Computer Science and Engineering Department of Patras University (Greece). He
obtained his diploma on July 2005 and he has been a member of Research Unit 6
since December 2001. He is responsible for the management of the web site of RU6.
His basic fields of interest are : ASP, PHP(mysql) and JSP language programming,
HTML programming, C and Java programming. He has participated in two projects
(ASP-NG, Broadband Promotion). He is currently participating in the SIG-GLUE
project running by Computer Technology Institute (CTI), Patras, Greece.

52
11. Acknowledgements
Work was supported by SIG-GLUE / Elearning Initiative, http://www.sig-glue.net
Ioannis Antonellis’s work was partially supported by a Zosima Foundation
Scholarship under grant 1039944/ 891/ B0011 /21-04-2003 (joint decision of
Ministry of Education and Ministry of Economics)

12. References
Ahmed, I., Blustein, J., Navigation in Information Space: How does Spatial Ability
play a part?, In Proceedings of Web Based Communities 2005 (IADIS), Algarve,
Portugal, 23 – 25 February 2005.

Alberts, M. and L. Kim 2000a, User Web Characteristics Using Palm Handhelds for
Information Retrieval. In Proc. of IPCC/SIGDOC Technology and Teamwork.
Cambridge, MA (2000), 125-135

Antonellis I., Bouras Ch., Kapoulas V., Poulopoulos V., Design and Implementation
of a Game Based Learning Related Community, In Proceedings of Web Based
Communities 2005 (IADIS), Algarve, Portugal, 23 – 25 February 2005.

Bickmore T.W., and B.N. Schilit, “Digestor: Device independent Access to the
World-Wide Web.”, In Proc. of 6th Int. World-Wide Web Conf., 1997.

Buyukkokten, O., H. Garcia-Molina, A. Paepcke, and T. Winograd, “Power Browser:

Efficient Web Browsing for PDAs.”, In Proc. of the Conf. on Human Factors in
Computing Systems, CHI’00, 2000, pp. 430-437.

Buyukkokten, O. H. Garcia-Molina, A. Paepcke, 2001a “Accordion Summarization

for End-Game Browsing on PDAs and Cellular Phones.”, In Proc. of the Conf. on
Human Factors in Computing Systems, CHI’01, 2001.

Buyukkokten O., H. Garcia-Molina, and A. Paepcke. 2001b “Seeing the Whole in

Parts: Text Summarization for Web Browsing on Handheld Devices.”, In Proc.
WWW10, pages 652—662, 2001.

Bouras C., Igglesis V., Kapoulas V., Tsiatsos T., A Web Based Virtual Community:
Funcionality and Architecture Issues, IADIS International Conference-Web
Based Communities 2004, Lisbon, Portugal, March 23 – 26 2004, pp. 59 – 66

Chatzinotas S., Sampson D., Kastradas K., A Web-Based Architecture for

53
 Empowering Learning Communities through the Learning Object Paradigm, In
Proceedings of Web Based Communities 2005 (IADIS), Algarve, Portugal,
23 – 25 February 2005.

Cohen, E.G., 1994. Restructuring the classroom: Conditions for productive small
group. Review of educational research, Vol. 64, No 1, pp 1-35.

Danielson, D.R. Transitional, “Volatility in Web Navigation: Usability Metrics and

User Behavior.”, M.S. Thesis, Stanford University, 2002.
Dyson, M. and M. Haselgrove, The Influence of Reading Speed and Line Length on
the Effectiveness of Reading from Screen. International Journal Human-
Computer Studies, (2001), 54 (4): 585-612

Environmental Detectives Project 2002,

http://cms.mit.edu/games/education/Handheld/Intro.htm

Feiner, S., “Seeing the Forest for the Trees: Hierarchical Display of Hypertext
Structure.”, Conf. on Office Information Systems, New York: ACM, 1988,
pp. 205-212.

Ferguson, J.D., Mohamed, R., Weir, G. and Wilson, J., Professional Development
On-line, Computers and Advanced Technology in Education (CATE 2002)
Cancun, Mexico, 2002.

Flor, N., Coulson, Sean., Maglio, P. Schema Blending and Sable Structure in Online
Social Systems, In Proceedings of Web Based Communities 2005 (IADIS),
Algarve, Portugal, 23 – 25 February 2005.

Fonda, C., Zennaro, M., Canessa, E., Sreenivasan, K.R., A Web Community to
Foster Science in Developing Countries: www.ictp.it, In Proceedings of Web Based
Communities 2005 (IADIS), Algarve, Portugal, 23 – 25 February 2005.

Hron, A. et al, 2000. Implicit and explicit dialogue structuring in virtual learning
groups. British Journal of Educational Psychology, Vol. 70, No 1, pp 51–64.

Hron, A. and Friedrich, H.F., 2003. A review of web-based collaborative learning:

factors beyond technology. Journal of Computer Assisted Learning, Vol. 19, No 1,
pp 70-79.

Jones, M., G. Marsden, N. Mohd-Nasir, K. Boone and G. Buchanan, “Improving

Web Interaction on Small Displays.”, In Proc. of 8th Int. World-Wide Web Conf.,
1999, pp. 51-59.

54
Jones, M. and G. Marsden, 1997. “From the Large Screen to the Small Screen—
Retaining the Designer’s Design for Effective User Interaction.”, IEEE Colloquium
on Issues for Networked Interpersonal Communicators. 139( 3): 1-4.

Kawachiya, K. and Ishikawa, H. “Improving Web interaction on Small Displays.”, In

Proceedings of 8th International WWW Conference, 51-59, 1999.

Kiesler, S. and Sproull, L. 1992. Group decision making and communication

technology. Organizational Behavior and Human Decision Processes, 52:96-123.
MacKay, B. and C.Watters. “The Impact of Migration of Data to Small Screens on
Navigation.”, IT & Society. 2003. 1(3). pp 90-101

MacKay B., Carolyn R. Watters, Jack Duffy: “Web Page Transformation When
Switching Devices.” In Proceedings of Mobile HCI, 2004: 228-239

Macher, E., Pathak, P., Role of Digital Repositories in Supporting Web Based
Communities, In Proceedings of Web Based Communities 2005 (IADIS),
Algarve, Portugal, 23 – 25 February 2005.

MIT and Microsoft. 2002. Environmental Detectives (by MIT and Microsoft).
http://cms.mit.edu/games/education/Handheld/Intro.htm

Myzel. 2002, Myzel: Online community game (by Institut für Gestaltungs und
Wirkungsforschung Technische Univeristät Wien), http://www.myzel.org

Oren, A., Mioduser, D., Nachmias, R., 2002, The Development of Social Climate
in Virtual Learning Discussion Groups, International Review of Research in
Open and Distance Learning, April 2002.

Pyati, A.K., Web-Based Communities and the Information Society: Alternatives

and Resistance, In Proceedings of Web Based Communities 2005 (IADIS),
Algarve, Portugal, 23 – 25 February 2005.

Rowan D. (2003), The EverQuest widows,

http://www.davidrowan.com/2003/02/times-everquest-widows.html
Sharples, M. (2000). “The Design of Personal Mobile Technologies for Lifelong
Learning.” Computers and Education 34: 177-193.

Slavin, R.E., 1996. Research on cooperative learning and achievement: What we

know, what we need to know. Contemporary Educational Psychology, Vol. 21,
No 1, pp 43-69.

55
Spence, R.. Information Visualization. 2001. ACM Press: N.Y.

Sotillo, M. S. (2000). Discourse functions and syntactic complexity in synchronous

and asynchronous communication. Language learning & Technology, 4(1),
p. 82-119.

Squicciarini, A.C., Paci, F., Bertino, E., A Secure Publishing Protocol for Virtual
Community Contracts, In Proceedings of Web Based Communities 2005
(IADIS), Algarve, Portugal, 23 – 25 February 2005.

TopSim 2002, http://www.topsim.com/en/

Unigame Project (2002) http://www.unigame.net/

Weinberger, A. et al, 2002. Fostering computer supported collaborative learning

with cooperation scripts and scaffolds. In Proceedings of the Conference on
Computer Support for Collaborative Learning (CSCL 2002) (ed. G. Stahl).
Lawrence Erlbaum, Hillsdale, NJ, pp 573–574.

Wiley D., 2002. Connecting Learning Objects to Instructional Design Theory:

A Definition, a Metaphor, and a Taxonomy. The Instructional Use of Learning
Objects (Bloomington, IN: Agency for Instructional Technology)

Woerndl, W., Eicker, D., Gruban, P. Creativity Techniques Meet the Web, In
Proceedings of Web Based Communities 2005 (IADIS), Algarve, Portugal,
23 – 25 February 2005.

56
57
58
Gender-based Neurocognitive Propensities
influencing Gameplay: An Interactions-
oriented approach
Philip Bonanno

Research interests
Educational Technology; Instructional Psychology; Cognitive Neuroscience;
Differentiated and Flexible Learning; Metacognition; Science Education.

Based on ‘work-in-progress’ investigating interactions during solitary and

collaborative gaming.

Abstract:
This paper reviews the literature on gender-based propensities in the use of ICT.
It also outlines the methodology and preliminary results obtained by the author
investigating interactions in collaborative gaming.

Using recently published reports and a wide spectrum of research, gender-based

neurocognitive propensities are identified and categorised. These include gender-
based differences in psychomotor skills, neurocognitive abilities and social skills.
Their manifestation and role in determining typology and frequency of interactions
in group-based cooperative gaming contexts are identified through a taxonomic
model that classifies interactions along three dimensions (Domain, Technology
and Community) and across three pedagogical levels (Acquisition, Participatory
and Contributory). Each dimension has a receptive cognitive component and a
reactive meta-learning one. This model underpins the experimental methodology
for a number of investigations carried out with college students organised in small
study groups according to personal data obtained from preliminary surveys. Results
include gender-related interaction profiles demonstrating underlying neurocognitive

59
 propensities employed during solitary and collaborative gaming. A concluding
section discusses the pedagogical implications of such profiles.

Keywords
Game-based learning, Cognitive Neuroscience, Pedagogy, Process-oriented
approaches, Gender-related issues, Technology-intensive collaborative learning
environments, Group dynamics.

1. Introduction: Gender Differences in the use of Information and

Communication Technology (ICT).
This paper reviews the literature related to research about gender-based propensities
in the use of ICT. It also outlines the methodology and preliminary results obtained
by the author investigating interactions in collaborative gaming.
Research with different age groups and ICT applications consistently point to differences
in the behaviour of males and females when using ICT. The Eurydice report (October
2005) “How boys and girls in Europe are finding their way with information and
communication technology”, points to a number of gender-related differences. There
is no difference between boys and girls regarding communication and word processing
but significant differences in the case of games and programming. In all countries,
boys use more graphics programs and the internet (for collaboration or downloading)
more frequently. In all countries, boys say that they have mainly learnt how to use
computers with friends or on their own, whereas girls report that they have mainly
done so at school or with their family. The same observation applies to use of the
Internet although the differences are less marked. Boys claim to be more competent
when performing complex activities. Differences between girls and boys become more
clear-cut in the case of complex activities grouped together under the heading file
management. Boys show a higher degree of facility in activities such as plotting graphs
using spreadsheets or creating presentations using programs such as PowerPoint.

The fact that these trends are apparent in all countries shows that neither culture nor
the level of school computerization appears to influence such gender-based patterns.
Eurydice concludes that “Whatever the country or level of school computerisation, the
attitudes of girls and boys vis-à-vis the use of ICT seem to differ in the same way: boys
are more attracted to ICT and use it more freely”, (Pg. 8).

60
Colley (2003) confirms that boys obtain greater experience with computers at home
than girls quoting Comber et al. (1997) and Shashaani (1994) and asserts that much
of this experience is acquired with computer games, which boys play to a far greater
extent than girls, quoting Van Schie & Wiegman (1997), Yelland & Lloyd (2001)
and Bonanno & Kommers (2005). Besides providing entertainment Colley points
to a number of effects of using computers and computer games. Such interactions
encourage familiarity in using computer hardware and software. Quoting Gailey
(1993), Colley claims that such activity promotes learning of the programming skills
necessary for particular games, thus providing transferable expertise in the types of
logic and operations used by many types of software. Boys develop positive computer
attitudes, Kirkman (1993) and Levine & Donitsa-Schmidt (1998) and higher levels
of confidence, Brosnan (1998), Comber et al. (1997) and Todman & Dick (1993).

Such exposure to ICT tools promotes a different approach to computers. Through

playing games, boys learn to regard computers as toys (Gailey 1993) or as technology
to be mastered and thus interact with software in a more playful and exploratory
manner, (Giacquinta et al. 1993). Girls approach computers differently and tend
to use them as tools to assist them in producing work having a preference for using
routines they know. These different approaches are also present among adults—women
use the Internet as a tool rather than as technology for play or mastery, (Singh, 2001).

Woman made more tool-like references than experienced men, who made more
personal references indicating their greater sense of control, (Hall & Cooper 1991).
Gender-related perceived control give rise to stylistic tendencies when interacting with
computers. Turkle (1984) describes boys as “hard masters”, exercising control with
analytical rigor giving them an advantage in programming. She categorizes girls as
“soft masters” who take a more concrete approach. They tend to treat computers as
entities with which to negotiate in order to create a product such as when using
modern software and the Internet for exploration and creativity.

These trends in childhood and adolescent years evolve into characteristic gender-based
interaction patterns and approaches with ICT in adulthood. The Pew Internet Project
(2005) reports that, while roughly the same percentage of men and women in the US
are serious internet users, they differ in fundamental aspects. Men value the net for

61
the freedom it gives them to try new ways of doing things and for obtaining all kinds
of information ranging from sports results, job offers and taking part in online fantasy
sports leagues. By contrast women like the opportunities the net gives them to make
and maintain human connections and for gathering information about health, map
directions and religious material. Gender-related differences are also reported in the
use email. While men using it as a way to maintain links with organisations rather
than individuals, females make greater use of e-mail for exchanges with individuals
and support groups.

The differences between male and female approaches and interactions with computers
are highly relevant to understanding how boys and girls use computers in all settings,
including education and gaming, whether solitary or collaborative. A review of the
literature points to a number of physical, cognitive and social gender-related differences
that clearly show the evolution of biological underlying mechanisms determining
individual and collective interactions within specific socio-cultural environments. This
paper categorises these gender-related differences and extends their implications to the
psycho-social climate within collaborative gaming contexts.

2. Theoretical background: Research in Gender-related Neurocognitive

Propensities
Evolutionary pressures in humans led to gender differences in physical adaptations,
neurocognitive processing tendencies and social comportment. Frost (1998; 2002)
argues that sexual division of labour resulted in the emergence of sexually differentiated
skills. Through the riskier, hunting activity man evolved better navigational abilities
as a consequence of travelling over a wide territory with much solitary scouting and
reconnoitering. Females taking care of food gathering developed superior social and
linguistic skills as gathering necessitate continual social interaction among individuals
concentrated within a smaller area. Geary (1998; 2004) attributes evolutionary
differentiation to male-to-male competition for fertile females. Besides enhanced
navigational skills, males evolved to be superior in the use of projectile (e.g., spears),
blunt force (e.g., clubs), tracking and dodging thrown objects. Such biologically
determined primary abilities are eloquently manifested and exploited during action
and fighting games showing this highly evolved “aggression management” mechanism
in males.

62
Cognitive gender-related differences include visuo-spatial (mathematical) abilities,
linguistic propensities and information processing in working memory. Most games
capitalise on visuo-spatial skills like orientation, manipulation of three-dimensional
objects, target following, localisation, maze following and navigation. Geary (1998)
uses the principles of sexual selection as an organizing framework for interpreting
cross-national patterns of sex differences in biologically-primary cognitive abilities
(those that have been directly shaped by evolutionary pressures i.e. by means of either
natural or sexual selection). Primary abilities, such as language and mathematical
abilities are found pan culturally and are supported by neurobiological architectures
designed to process domain-specific information and develop in natural contexts,
that is, during children’s play or social activities (Geary 1995; Pinker & Bloom 1990;
Witelson 1987).

Cross-national studies suggest that there are no sex differences in biologically primary
language abilities including phonological, semantic, language-generation and articulatory
mechanisms. Together with mathematical abilities, these show moderate heritability
estimates and are found in all cultures as well as in nonhuman primates. Sex differences
in several biologically secondary linguistic and mathematical domains, such as those
related to reading, writing and computation emerge primarily in school or in ICT-
intensive environments, are found throughout the industrialized world. In particular,
males consistently outperform females in the solving of mathematical word problems
and geometry as a result of greater elaboration in males than in females of the neuro-
cognitive systems that support navigation in three-dimensional space.

Research shows that males and females differ systematically on tests of spatial cognition
(Halpern 1992; Kimura 1996; Voyer, & Bryden 1995). On average men have an
advantage in tasks of spatial cognition such as the Viewfinding Task, (Watson &
Kimura, 1991), localisation and orientation (Gordon & Lee, 1986), and mental
rotation, (Gouchie & Kimura, 1991; Moffat & Hampson, 1996). Men also have
an advantage in tasks that require both target-directed motor skills such as finger
dexterity and spatial cognition involving three dimensional object rotation, such as
catching an object, (Watson & Kimura, 1991), and guiding or intercepting projectiles
such as throwing darts, (Watson & Kimura, 1989, 1991). These cognitive and psycho-
motor skills are strongly influenced by early and current hormonal levels, (Margolis &
Fisher, 2002). Gazzaniga et al (1998) discusses hormonal influence on spatial learning
tasks in both human and rodents. High levels of oestrogen in females are associated
with poorer performance, while high levels of testosterone optimize spatial abilities in

63
males. Janowsky et al. (1994) show how increased levels of testosterone from testoste-
rone supplement through scrotal patches, to regulate bone and mineral metabolism,
resulted in increased ability in spatial manipulation of objects but not in motor
dexterity, verbal memory and cognitive flexibility.

One of the largest between-sex cognitive differences favouring males, that is extensively
exploited by digital games, concerns visuospatial tasks that require transformations in
visuospatial working memory, (Krikorian, Bartok, & Gay, 1996). Males have an
advantage in visuospatial reasoning being more adept at performing disembedding
and internal spatial transformations required by a task such as mental rotation of
shapes, shape recall, geometry, maze learning and map reading, (Gazzaniga et al., 1998;
Halpern, 1986; Wilson et al., 1975; Coren et al., 1994). The differences hold up
whether one is dealing with simple or complex patterns (Bryden & George, 1990).
Males are either faster or more accurate when completing such tasks. Males also differ in
increased aggression and greater risk taking. The male advantage in spatial reasoning
may be one reason why they score higher on maths tests, why most chess masters are
men and why males use computers more as a medium for entertainment, learning,
communication and personal expression than females.

On average, women have an advantage on tasks requiring perceptual speed and fine
motor skills, such as identification of pictures, (Watson & Kimura, 1991) or sequenced
hand movements, (Nicholson & Kimura, 1996). Numerous studies, (Inglis & Lawson,
1982; MacCoby & Jacklin, 1974) consistently found gender differences in language
and visuospatial skills. On average, girls begin to acquire language about one month
prior to boys, and by age 11, they consistently perform better on tests of verbal
abilities, with the gender gap growing at least until adulthood. Female superiority
is seen on tests of both receptive and productive language, and on more complex
tasks such as making analogies and creative writing. The gender difference is not
large, showing extensive overlap between the two groups, but it is consistent. In
general, women are reported to be more verbally fluent than men, (Stumpf, 1995),
although mixed results have also been obtained. For instance, studies have shown a
female advantage for quickly producing words from a particular semantic category
(e.g., foods), (Gordon & Lee, 1986). Girls are much better than boys at generating
sentences when given the initial letter of each word but there is no sex difference for
rapidly producing words beginning with a particular letter, (Gordon & Lee, 1986),
for either type of fluency measure (e.g., (Moffat & Hampson, 1996)) or task of rapid
articulation, (Gouchie & Kimura, 1991). This explains the observed preference by

64
females for word puzzle games (Bonanno & Kommers, 2005). Girls also show better
interpretation of emotion cues, and a higher tendency to comply with parents and peers.

Baenninger and Newcombe (1995) make a significant suggestion very relevant to

competence in the use of digital media. They claim that many gender differences may
be decreasing in recent years, citing a number of meta-analyses that have shown this
pattern. Crawford et al. (1995) also stress a decline in gender differences in spatial
visualization skills in the past 40 years. Stumpf (1995) acknowledges the reduction in
some gender differences with time but points out that gender differences across a
number of cognitive dimensions are still quite robust despite the reduction over time.
Such studies point out the importance of psychosocial factors on sex differences in
cognitive abilities. Finally, current models, (Casey, 1996; Halpern & Tan, 2001)
propose that the interaction between life experience and inherited biological propen-
sities builds sex difference in spatial cognition. Sherry et al. (2002) points to the
importance of game playing in developing such skills quoting Subrahmanyam &
Greenfield (1994) who argue that children learn cognitive skills, such as how to orient
things in space, by playing electronic games and that these skills differ between girls
and boys because of differential exposure to electronic games.

It is clear that different genres of games such as First/third person shooters, Fighting
and Sports games capitalize on these biologically primary psychomotor skills that
are an intimate component of the male neurocognitive constitution and that are
pan culturally distributed. Interaction with artefacts within a given cultural context,
such as digital games, gives rise to biologically secondary psychomotor skills such as
increased perceptual acuity, eye-to-hand co-ordination, reaction speed, parallel process-
ing and quicker information processing that characterize highly competent gamers.

Gender-based propensities in Social Skills.

Sexual division of labour, mainly hunting for males and food gathering and child
rearing for females, together with competition for mates, resulted in the emergence
of sexually differentiated social skills. Over time, these differing task requirements
led to the emergence in men of better navigational abilities, aggressive behaviours
and defensive skills. On the other hand evolutionary pressures equipped women
with superior social and linguistic skills (Geary, 1998).

65
Geary also elaborates on the survival value of coalition-based competition. Play appears
to provide the practice necessary to develop the capacities needed for survival and
reproduction in adulthood (Fagen, 1981). The pattern of human sex differences in
play activities is consistent with the hypothesis that coalition-based male-male com-
petition was an important feature of human evolution. Sex differences in rough-and-
tumble play, or play fighting, are especially evident in species where conflict in adult-
hood is often resolved through physical aggression (Smith 1982). In polygynous
primates, males engage in play fighting more frequently and more vigorously than
conspecific females, a sex difference that is related, at least in part, to prenatal exposure
to androgens (Collaer & Hines, 1995; Wallen, 1996). Although boys sometimes
engage in play hunting, studies of the spontaneous play behaviour of children
reveal that boys more frequently organize themselves into groups and compete against
groups of other boys and engage in this form of play three times more frequently than
do same-age girls (e.g., (Lever, 1978)). More often, boys compete as members of teams
and must simultaneously coordinate their actions with those of their team mates
while taking into account the action and strategies of their opponents. Boys interviewed
expressed finding gratification in acting as a representative of a collectivity; the approval
or disapproval of one’s team mates accentuates the importance of contributing to a
group victory. (Lever, 1978, p. 478).

Most of these competitive games require many of the same physical, social, and
cognitive capacities that are involved in coalition-based warfare (Geary 1995)
characterized by in-groups and out-groups formations, the strategic coordination
of the activities of in-group members as related to competition with the out-group,
the throwing of projectiles at specific targets, and the tracking and reacting to the
movement of these projectiles. These activities can be exemplified by team-based
sports games and by the game of Throw at Each Other with Mud, played by Sioux
Indians 200 years ago (Hassrick, 1964), among many other athletic games.

Contiguous and virtual collaborative or competitive gaming contexts are the most
recent, eloquent manifestation of these coalition-based interactions where males tend
to dominate. Multiplayer LAN parties and massive multiplayer on-line games are
exclusively male domains as are international gaming competitions like the Cyber X
Games and the World Cyber Games. ‘The imbalance between men and women was
clear to see in the Cyber X Games with almost 50 male teams taking part, compared
with just seven female squads. And as in the world of sport, there were separate com-

66
petitions for each gender. On-line games are overwhelmingly male (Counter-Strike is
99% male) so females that do come into this world feel overpowered or that they are
not welcomed,’ (Hermida, 2004).

Conclusion
This paper elaborates on individual neuro-cognitive processes identified above and
extends their implications to the psycho-social dimension within collaborative gaming
scenarios. The typologies, frequency and directionality of task and person-oriented
interactions will be quantified and discussed in relation to the above mentioned
gender-related neuro-cognitive mechanisms and propensities.

3. Methodology
3.1 An Interactions approach
An interactions model that classifies interactions along three dimensions (Domain,
Technology and Community) and across three pedagogical levels (Acquisition,
Participatory and Contributory) is being proposed. Each dimension has a receptive
cognitive and a reactive, meta-learning component. The proposed interactions-oriented
approach is organized around an elaborated version of Gilroy’s (2001) model:
Valued Learning Experience = F [Pedagogy; Content; Community; Metacognition]
The domain/content dimension involves interactions leading to the acquisition of
relevant knowledge and skills. The social dimension captures the socio-emotional
climate and ensuing interpersonal interactions within a group that lead to the forma-
tion of bonds, relationships and roles. Interactions also occur at the technological
level showing increased knowledge and refined skills in the use of digital tools.
Through metacognitive activity, participants organize and control individual and
collective learning along these dimensions.

A more detailed discussion of this model may be found in Bonanno (2005), where it
is applied to identify learning profiles of participants in Web-based communities.

67
3.2 Experimental Methodology
3.2.1 Subjects
This investigation was carried out with college students enrolled for a course in
Biology at advanced level. The gender distribution was 35.1% male and 64.9 %
female. This was the mean for different student samples involved in various investiga-
tions.

3.2.2 Procedure
The investigation involved two stages. First, data about participating students were
analysed to determine the prevalence of Neurocognitive propensities in Solitary
Gaming. The second stage involved investigating the effect of Gender-based
Neurocognitive propensities on interactions in Collaborative gaming.

Allocation of Subjects to Study Groups and Data Collection

This stage of the investigation involved a group of students interacting co-operatively
using one computer on which games were run. The procedure included the following
steps:
I. Administering questionnaire to collect data about participants’
gaming tendencies.
II. Setting up study groups with different composition and
experimental conditions.
III. Recording gaming sessions.
IV. Analysing recorded student and game-play videos for type and
intensity of interactions.
V. Computing typologies of interactions on to data sheets.
VI. Entering and analysing data into SPSS, according to research questions.
3.2.3 Recording Gaming sessions
Separate recordings captured: (i) students interacting and (ii) corresponding simulta-
neous game-play as seen on computer monitor. The game was played on a Laptop
PC connected through a converter card to a VCR. The VCR was connected to a TV
so that students could follow game play both from the Laptop monitor and from

68
the TV. At the same time a camcorder fitted on a tripod was placed in front of the
students to record group activity. After a study session, the group recordings were
transferred from the camcorder cassette to a VHS cassette to be used for analysis.

3.2.4 Games Used

Different game genres were used to create diverse interactive scenarios and hence
experimental conditions. Sid Meier’s Alpha Centauri (AC) was used as a strategy game
based on science and technology. Empire Earth (EE), Age of Empires and Rise
of Nations have a historical orientation with an attractive intermix between narrative
and game play. Mind Maze (MM), from Microsoft’s Encarta package, was used as a
language-based puzzle game. SIMS Deluxe was used for its emphasis on the dynamics
of human relationships. Need for Speed was used as an action/sports game.

4. Results
4.1 Neurocognitive Propensities in Solitary Gaming
Results obtained in this investigation are discussed in more detail (Bonanno & Kommers,
2005). From this analysis it is very evident that games offer the context for gender-
related behaviour patterns arising from underlying neurocognitive tendencies. This
investigation about the use of digital games within a Maltese context is in line with
findings from similar investigations. Though the Computer is the most popular
gaming device in use, the tendency in females to play puzzle games leads to a wider
use of the mobile phone. Maltese male college students spend more time playing
games than females. The pattern for average time (hours per week) obtained for
Maltese males (M= 6.712), for females (M=2.492) and for the total group (M=3.941)
is comparable to that obtained in other studies—sample mean (M= 5.00) for British
Columbia teens (Media Lab Analysis report, 1988) and (Shelly et al., 2002) College
males (M = 13.77), females (M =4.65). The lower values can be attributed to a
number of factors. The lower mean for the whole group is a consequence of the
limited time available for playing digital games (and in fact for all other leisure
activities), because Maltese students are enrolled in a very intensive two-year course
that compels students to dedicate most of their time to study. The overt resistance
to or anxiety about playing digital games from a good percentage of female students
confirms this. Predominant gender scripts in relation to this medium amplify this
trend.

69
The above analysis of gaming tendencies discloses a number of underlying cognitive
and motivational gender-related trends. The high percentage of females opting for
the Puzzle, Adventure, Fighting and Managerial games confirms Shelley’s et al. (2002)
findings that females’ top reasons for playing include challenge and arousal. Males’
preference for First Person Shooters, Role Playing games, Sport and Strategy games
indicates gratification of different needs—challenge and social interaction. These
tendencies can also be seen as a process of accommodation to different underlying
gender-related neurocognitive processes. The indicated preferred games by females
capitalize on their natural propensities and skills such as perceptual speed, fine motor
skills and sequenced hand movements, (Watson & Kimura, 1991). Games preferred
by males demand a higher visuospatial ability involving localization, orientation, mental
rotation, target-directed motor skills, greater reaction speed, increased aggression and
greater risk taking. This confirms Halpern & Wright (1996) and Casey (1996) claim
that males excel at tasks requiring maintenance and manipulation of information in
working memory, while females excel at tasks which require rapid access and retrieval
of information from stored memory. While males excel at tasks that benefit from
combining new strategies, such as mental rotation ability, females tend to draw on
memory of past algorithms or knowledge such as verbal fluency, rather than inventing
new approaches. These underlying neurocognitive processes manifest themselves
externally as gender-related processing styles. Referring to Turkle and Papert, Rommes
(2002) claims that males prefer command structure approaches in computers, as they
want to ‘feel in control’ and they are not afraid of taking risks in learning how to use
a computer, thus preferring a ‘learning by doing’ approach. Females favour a more
concrete, contextualized, intimate ‘bricolage’ approach and they do not like to take
risks while learning (Turkle, 1988; Turkle and Papert, 1991). Game genre preference
is yet another context where these tendencies are expressed.

4.2 Influence of Neurocognitive propensities in collaborative gaming.

4.2.1 Identifying Factors determining Interactions.
Experiments conducted to understand the mechanisms for efficient collaboration
include approaches that decomposed ‘work alone’ as a dependent variable into several
other measures of performance, such as the improvement of monitoring and regulation

70
skills (Brown & Palincsar, 1989; Blaye & Chambres, 1991). Dillenbourg et al. (1996)
conclude that collaboration is in itself neither efficient nor inefficient. Collaboration
works under some conditions, and it is the aim of research to determine the conditions
under which collaboration is efficient. For this purpose one has to vary these conditions
systematically along two dimensions, both dependent (work alone / collaborative
learning) and independent variables (the composition of the group, the features of the
task, the context of collaboration and the medium available for communication). The
composition of the group covers several other independent variables such as the
number of members, their gender and the differences between participants. These
significant research conclusions are being integrated with results from the first stage
of this investigation to quantify gender-related neurocognitive and social propensities.
These will be analyzed through the following independent variables: Preferred social
context for gaming; separate Task-oriented interactions (TOI: Focused reception,
Interacting with Game, Imitating Game actions, Referring to gaming Biography,
Asking for help, Giving help, Responding, Sharing, Providing feedback, Confirming,
Suggesting); separate Person-oriented interactions (POI: Pleased look, Neutral
look, Disagree / Rejection, Jubilant Expression, Approving gestures, Disapproving,
Disengaged, Hostile, Recommends game, Censures game).

4.2.2 Results for effect of neurocognitive propensities in

collaborative gaming.
A highly significant difference regarding the preferred gaming environment was obtained:
Pearson Chi Square p < 0.001. 69.2% of the males prefer solitary gaming compared to
30.8% who prefer playing games in company with others. The reverse tendency applies
for girls who prefer gaming with others (57%) to playing games alone (43%).

Different games utilize different gender-related neurocognitive processes that manifest

themselves through characteristic types and frequencies of interactions. Preliminary
analysis of data about interactions in various gaming contexts shows different
gender-related interactions profiles. For the science and technology-oriented, real
time strategy (RTS) game, Alpha Centauri, males and females show different
interaction repertoires. At the beginning of the game males show high levels of
cognitive activity manifested through higher ‘Interaction with game’, ‘Focused
Reception’, ‘Sharing’ and ‘Providing Feedback’. As the game proceeds it establishes
a repetitive game-play pattern inhibiting male exploratory or risk-taking activities.

71
This de-motivates males who start ‘Censuring the game’. Female rejection of such
game play is manifested through lack of interaction. Throughout the gaming session,
females show reduced cognitive activity regarding ‘Interaction with Game’, Task-
oriented interactions (including ‘Sharing’ and ‘Providing feedback’) and reduced
interpersonal communication.

For the game Empire Earth females show a higher level of cognitive activity manifested
through longer periods of ‘Focused Reception’, show higher level of response and
positive reactions. Males show a greater tendency to be ‘Disengaged’ thus showing
that most males are not intellectually stimulated by rehearsal-oriented RTS games.
Still, when asked for help, males contributed by ‘Giving Help’ mainly to females who
show a higher level of ‘Responding’ due to their more reactive rather than proactive
approach. This manifests a clear distinction in attitude to games between males and
females. Males assume a superior competence status considering themselves as a refe-
rence point when females find themselves in difficulty. On the other hand females
tend to accept a subordinate competence status. Tacitly males consider RTS games as
‘chicken feed’ for them. Their opinion is that true gaming challenge is only found in
Action, Shooting, Racing and Sports games, but these types of games exploit their
highly developed visuo-spatial skills. Male disengagement is actually their reaction to
the rehearsing strategy demanded by RTS games which happens to match female
neurocognitive propensities. When faced with difficulty females prefer to ask assistance
to males, perceived as more competent in solving game related problems. This
contrasts with the approach adopted by males when faced with difficulty. Males adopt
a tinkering approach, not afraid of taking risks, rather than asking for help from their
colleagues.

For the Puzzle game Mindmaze there is no overall gender-related significant effect.
The game seems to be appealing to both sexes but in different ways. Males are
stimulated by practicing their navigational skills while manoeuvring through the
labyrinth of passages and doorways in the virtual medieval castle. Females exploit
their linguistic abilities being enthused by recalling and corroborating information
when prompted by quiz questions. This clearly demonstrates that males and females
adapt to gaming contexts that use their natural strategic propensities, obtaining
gratification if this matches their gender-related neuro-cognitive propensities.

An investigation was also carried to explore possible female preference for socially
supportive, intimate gaming contexts through analysis of interactions at the socio-

72
emotional level assuming that increased interactivity is an expression of preferred
socio-emotional context. Females tend to show higher cognitive and emotional
interactivity in groups composed solely of female participants rather then in mixed-
gender groups. In the presence of males, females tend to be more restrained in ex-
pressing themselves thus contributing much less in building a positive socio-emotional
atmosphere.

4.2.3 Conclusion: Pedagogical Implications

Analysing different aspects of collaborative gaming contexts from an interactions per-
spective is a rich and valuable research experience that can provide insight into poten-
tial pedagogical applications and possibly a paradigm shift in the conception of design-
ing technology intensive collaborative learning environments. The proposed interactions
approach, involving the technology (game and gaming hardware), the subject or domain
and the community dimensions, is a more systematic approach in developing a
pedagogy for digital game-based learning. Instead of designing learning solely based
on content and task analysis, it is a more valid approach to identify and promote
interactions both at a cognitive and metacognitive level within the identified dimensions.
A process-oriented approach also fosters an innovative methodology for evaluating
the instructional and educational potential of games. The use of games in identified
learning or training situations should be determined by the type of neuro-cognitive
processes that are employed by user/s. Different game genres utilize and develop
different neurocognitive skills. Hence it is the job of teachers, instructors and game
designers to identify and exploit these in instruction or game design.

Design activities should be sensitive to gender-based neurocognitive propensities.

Both when designing games or game-based learning, such propensities should be
respected, exploited and challenged. Male and female learners should be made aware
of their natural tendencies in the use (or avoidance of use) of digital games or game-
based learning. Relevant pedagogical support should accompany such activities,
pointing to congenial tendencies both in patterns of games used in solitary contexts but
also in collaborative scenarios. Patterns of use based on uncongenial neurocognitive
processing, stereotypes or evaded group roles during gaming should be challenged to
develop the necessary compensatory skills.

73
This interactions model can also be applied in the evaluation of the various gaming
contexts—PC based gaming, console-based, mobile gaming, solitary, cooperative or
competitive gaming. The instructional or educational potential of such contexts can
only be holistically assessed through the use of interaction metrics and not content-
oriented approaches. Educational games and game-based learning should be analysed
for the degree of domain oriented interaction—how much and in what ways does a
particular game promote knowledge, skills, insight and reflection about the domain
in question. It should also be assessed for its potential in developing gaming and tech-
nology related skills (gaming gestalts) and knowledge (awareness of surface and deep
game structure), together with corresponding attitudinal change. The pedagogical
value of games and gaming contexts should also be evaluated through the level of
participation and contribution that the experience triggers in users.

5. Author

In 1982 graduated from the University of Malta (UOM) with a bachelors in Education
specializing in Science Education. Between 1982–1995, taught science and biology in
Secondary state schools. In 2001 graduated with a Master of Philosophy by research
from the Centre for Communication Technologies, UOM, specializing in Individual
Differences in Learning and Instruction. Presently reading for a PhD in Instructional
Psychology and Technology at University of Joensuu, Finland. Lectures in Biology at
a pre-university college, and in Flexible Learning at the UOM. Currently involved in
various EU projects involving Educational Technology.

74
6. References

Baenninger, M., & Newcombe, N. (1995). Environmental input to the development

of sex related differences in spatial and mathematical ability. Learning and
Individual Differences, 7, 363–379.

Berger , P.L. and Luckman, T. (1966). The Social Construction of Reality, Penguin
Books, Middlesex, England.

Blaye, A., & Chambres, P. (1991). Hypercard as a psychological research tool-

Experimental studies. In A. Oliveira (Ed.) Hypermedia couseware: structures
of communication and intelligent help. Berlin: Springer-Verlag.

Bonanno, Ph. (2005). Developing Learning Profiles for Web-Based Communities:

Towards an Interactions-Oriented Model. Int. J. of Web Based Communities,
Vol. 1, No. 3, 2005.

Bonanno, Ph. & Kommers, P.A.M. (2005). Gender Differences and Styles in the
Use of Digital Games. Educational Psychology. Vol: 25 No. 1 2005.

Brosnan, M. (1998). The role of psychological gender in the computer-related

attitudes and attainments of primary school children (aged 6–11). Computers
and Education, 30, 203–208.

Brown, A.L., & Palincsar, A.S. (1989). Guided cooperative learning and individual
knowledge acquisition. In L.B. Resnick (Ed). Knowing, learning and instruction,
essays in honor of Robert Glaser. Hillsdale, NJ: Lawrence Erlbaum Publisher.

Bryden, M. P., & George, J. (1990). Sex differences and the role of figural
complexity in determining the rate of mental rotation. Perceptual and Motor
Skills, 70, 467-477.

Casey, M.B. (1996). Gender, Sex, and Cognition: Considering the Interrelationship
between Biological and Environmental Factors. Learning and Individual
Differences, Volume 8, Number 1, 1996, pages 39-53.

Coles, S. (2005). Bill Gates plots a Windows future. BBC Technology News. 21
January. URL: http://news.bbc.co.uk/1/hi/programmes/click_online/4195177.stm

75
Collaer, M. L. & Hines, M. (1995) Human behavioral sex differences: A role for
gonadal hormones during early development? Psychological Bulletin 118:55–107.

Colley, A. (2003). Gender differences in adolescents’ perceptions of the best and

worst aspects of computing at school. Computers in Human Behaviour. Vol. 19,
Issue 6, Pages 653-797 (November 2003), Pages 673-682.

Collis, B. & Moonen, J. (2001). Flexible Learning in a Digital World: Experiences

and Expectations. London: Kogan Page Limited, UK.

Comber, C., Colley, A., Hargreaves, D. J., & Dorn, L. (1997). The Effects of age,
gender and computer experience upon computer attitudes. Educational Research,
39, 123– 133. Department for Education and Employment. (1997). Connecting
the learning society. London: HMSO.

Coren, S., Ward, L.M., & Enns, J.T.(1994). Sensation and Perception, 4th edition.
Ft. Worth, TX: Harcourt Brace Collge Publishers.

Crawford, M., Chaffin, R., & Fitton, L. (1995). Cognition in social context.
Learning and Individual Differences, 7, 341–362.

Dillenbourg, P., Baker, M., Blaye, A. & O’Malley, C. (1996) The evolution of
research on collaborative learning. In E. Spada & P. Reiman (Eds) Learning in
Humans and Machine: Towards an interdisciplinary learning science. (Pp. 189-
211). Oxford: Elsevier.

Eurydice Report. (2005).

http://www.eurydice.org/Documents/Eurydice_en_bref/EN/8_PAGES

Fagen, R. M. (1981) Animal play behavior. Oxford University Press.

Frost, P. (1998). Commentary on David C. Geary (1996). Sexual selection and sex

differences in mathematical abilities. BBS 19:229–84. Behavioral and Brain

Sciences (1998) 21: 443-448.

Frost, P. (2002). Review of Desolate Landscapes: Ice-Age Settlement in Eastern

Europe by John F. Hoffecker. Human Nature Review. 2: 156-163.

76
Gailey, C. (1993). Mediated messages: gender, class and cosmos in home video games.
Journal of Popular Culture, 27, 81–97.

Gazzaniga, M. S., Ivry, R. B., & Mangun, G. R. (1998). Cognitive Neuroscience:

The Biology of the Mind. W. W. Norton & Company. New York. London.

Geary, D. C. (1995) Sexual selection and sex differences in spatial cognition.

Learning and Individual Differences 7:289–301.

Geary, C. (1998). Male, female: The evolution of human sex differences.

Washington, DC: American Psychological Association.

Geary, C. (2004). Origin of Mind: Evolution of Brain, Cognition, and General

Intelligence. American Psychological Association (APA).

Giacquinta, J. B., Bauer, J. A., & Levin, J. E. (1993). Beyond technology’s

promise: examination of children’s computing at home. Cambridge: Cambridge
University Press.

Gilroy, K. (2001). Collaborative E-learning: the right approach. ArsDigita Systems

Journal. Availble: http://www.arsdigita.com/asj/elearning/ [accessed 17 June 2002].

Gordon, H. W., & Lee, P. A. (1986). A relationship between gonadotropins and

visuospatial function. Neuro-psychologia, 24, 563–576.

Gouchie, C., & Kimura, D. (1991). The relationship between testosterone levels
and cognitive ability patterns. Psychoneuroendocrinology, 16, 323–334.

Guttormsen Schär, S., Schluep, S., Schierz, C., Krueger, H. (2000). Interaction for
Computer-Aided Learning. Interactive Multimedia Electronic Journal of
Computer-Enhanced Learning. Volume 2, No. 1, April, 2000. URL:
http://www.imej.wfu.edu/articles/2000/1/03/index.asp

Hall, J., & Cooper, J. (1991). Gender, experience and attributions to the computer.
Journal of Educational Computing Research, 7, 51–60.

Halpern, D. F. (1992). Sex differences in cognitive abilities (2nd ed.). Hillsdale,

NJ:Erlbaum.

77
Halpern, D., & Tan, U. (2001). Stereotypes and steroids: using a psychobiosocial
models to understand cognitive sex differences. Brain and Cognition, 45,
344–392.

Halpern, D. F., & Wright, T. M. (1996). A process oriented model of cognitive sex
differences. Learning and Individual Differences, 8, 3–24.

Rommes, E. (2002). Gender Scripts and the Internet. Twente University Press,
Enschede.

Hardy, I. (2005). Handheld Battle Starts in 2005. BBC Technology News. 14

January, 2005. URL: http://news.bbc.co.uk/1/hi/programmes/click_online/
4175141.stm

Hassrick, R. B. (1964) The Sioux: Life and customs of a warrior society. University of
Oklahoma Press.

Hermida, A. (2004). Girl gamers strike at the boys. http://news.bbc.co.uk/1/hi/

technology/3496963.stm

Inglis, J. & Lawson, J.S. (1982). A meta-analysis of sex differences in the effects of
unilateral brain damage on intelligence test results. Canadian Journal of
Psychology. 36: 670-683.

International Federation for Information Processing. (2005). Stellenbosch

Declaration. http://www.ifip.or.at/

Janowsky, J.S., Oviatt, S.K., and Orwoll, E.S. (1994) Testosterone influences spatial
cognition in older man. Behavioural Neuroscience 108: 325 – 332.

Jonassen, D. H. (1991). Context is everything. Educational Technology,

31(6), 35–37.

Jonassen, D. H. (1994). Toward a constructivist design model. Educational

Technology, 34(4), 34–37.

Keen, K. (1992). Competence: what is it and how can it be developed?

In J. Lowyck, P. de Potter, J. Elen (Eds.), Instructional design: implementation
issues (pp. 111–122). Brussels, Belgium: IBM Education Center.

78
Kimura, D. (1996). Sex, sexual orientation and sex hormones influence
human cognitive function. Current Opinion in Neurobiology, 6, 259–263.

Kirkman, C. (1993). Computer experience and attitudes of 12-year old students:

implications for the UK National Curriculum. Journal of Computer Assisted
Learning, 9, 51–62.

Krikorian, R., Bartok, J. A., Gay, N. (1996). Immediate memory capacity for
nonsequential information: the configural attention test. Neuropsychology,
10, 352-356.

Lever, J. (1978). Sex differences in the complexity of children’s play and games.
American Sociological Review, 43,471-483.

Levine, T., & Donitsa-Schmidt, S. (1998). Computer use, confidence, attitudes, and
knowledge: a causal analysis. Computers in Human Behaviour, 14, 125–146.

MacCoby, E., & Jacklin, C. (1974). The Psychology of Sex Differences. Stanford,
CA: Stanford University Press.

Margolis, J. & Fisher, A. (2002) Unlocking the Clubhouse: Women in Computing,

Cambridge: MIT Press.

Moffat, S. D., & Hampson, E. (1996). A curvilinear relationship between

testosterone and spatial cognition in humans: possible influence of hand prefe-
rence. Psychoneuroendocrinology, 21, 323–337.

Myers, I.B., & McCaulley, M.H. (1985). Manual: A guide to the development
and use of the Myers-Briggs Type Indicator. Palo Alto, CA: Consulting
Psychologists Press.

Nicholson, K. G., & Kimura, D. (1996). Sex differences for speech and manual skill.
Perceptual and Motor Skills, 82, 3–13..Osterrieth, P. A. (1944). Le teste de copie
d’une figure complexe: contribution a l’etude de la perception et de la memoire.
Archives de Psychologie, 30, 206–259.

Perkins, D. N. (1986). Knowledge as Design. Hillsdale, New Jersey: Lawrence

Erlbaum Associates.

79
Pew Internet Project (2005): http://news.bbc.co.uk/1/hi/technology/4555370.stm

Pinker, S. & P. Bloom. (1990). “Natural language and natural selection.” Behavioral
and Brain Sciences, 23, 707-784.

Prensky, M. (2001). Digital Game-Based Learning. New York: McGraw-Hill.

Richtel, M. (2004). Game Sales Thrive Thanks to the Big Kids (in Their 20’s). New
York Times Technology News, 27 December, 2004. URL:
http://www.nytimes.com/2004/12/27/business/media/27game.html?oref=login&th

Salomon, G (1990). Studying the Flute And the Orchestra: Controlled VS.
Classroom Research on Computers. International Journal of Educational
Research, Vol. 14, No. 6, 1990.

Salomon, G & Perkins, D.N. (1998). Individual and Social Aspects of Learning.
Research in Education, Volume 23.

Shashaani, L. (1994). Gender differences in computer experience and its influence on

computer attitudes. Journal of Educational Computing Research, 11, 347–367.

Sherry, D.F., Jacobs, L.F., and Gaulin, S.J. (1992). Spatial memory and adaptive spe-
cialization of the hippocampus. Trends Neuroscience. 15:298-303.

Sherry, J., de Souza, R., Greenberg, B.S., & Lachlan, K. (2002a). Age and video game
U & G: Why do Adolescents Play Video Games? Developmental Stages Predicts
Video Game Uses and Gratifications, Game Preference, and Amount of Time
Spent in Play. On-line document: http://web.ics.purdue.edu/~sherryj/videoga-
mes/papers.htm, Retrieved on: 24/03/03.

Singh, S. (2001). Gender and the use of the Internet at home. New Media and
Society, 3, 395–415.

Smith, P. K. (1982) Does play matter? Functional and evolutionary aspects of animal
and human play. Behavioral and Brain Sciences 5:139–84.

Stumpf, H. (1995). Gender differences in performance on tests of cognitive

abilities:experimental design issues and empirical results. Learning and
Individual Differences, 7, 275–287.

80
Subrahmanyam, K. and Greenfield, P.M. (1994) Effects of Video Game Practice on
Spatial Skills in Girls and Boys, Journal of Applied Developmental
Psychology15(1): 13_32.

Todman, J., & Dick, G. (1993). Primary children and teachers attitudes to
computers. Computers in Education, 20, 199–203.

Turkle, S. (1984). The second self: computers and the human spirit. New York:
Simon & Shuster.

Turkle, S. (1988). Computational reticence: why women fear the intimate machine.
In Kramarae, C. (Ed.): Technology and Women’s Voices. Routledge & Kegan
Paul:41-61. New York/London.

Turkle, S. & Papert, S. (1991). Epistemological Pluralism: Styles and Voices within
the Computer Culture. In I. Harel & S. Papert (Eds.), Constructionism
(pp. 161-193). Norwood, NJ: Ablex.

Van Schie, E. G. M., & Wiegman, O. (1997). Children and videogames: leisure
activities, aggression, social integration, and school performance. Journal of
Applied Social Psychology, 27, 1175–1194.

Voyer, D., Voyer, S., & Bryden, M. P. (1995). Magnitude of sex differences in spatial
abilities: a meta-analysis and consideration of critical variables. Psychological
Bulletin, 117, 250–270.

Wallen, K. (1996) Nature needs nurture: The interaction of hormonal and social
influences on the development of behavioral sex differences in rhesus monkeys.
Hormones and Behavior 30:364–78.

Watson, N. V., & Kimura, D. (1989). Right-hand superiority for throwing but not
for intercepting. Neuropsychologia, 27, 1399-1414.

Watson, N. V., & Kimura, D. (1991). Nontrivial sex differences in throwing and
intercepting: relation to psycho-metrically-defined spatial functions. Personality
and Individual Differences, 12, 375–385.

81
Wilson, J.R., DeFries, J.C., McClearn, G.C., Vanderberg, S.G., Johnson, R.C. &
Rashad, M.N. (1975). Cognitive abilities: Use of family data as a control to assess
sex and age differences in two ethnic groups. International Journal of Aging and
Human Development, 6, 261-275.

Witelson, S. E. (1987). “Neurobiological aspects of language in children.” Child

Development, 58,653-688.

Yelland, N., & Lloyd, M. (2001). Virtual kids of the 21st century: understanding
children in Schools today. Information Technology in Childhood education
Annual, 13, 175–192.

82
83
84
Influence of videogame play on a student’s
approach to learning?
Fiona Littleton, Jeff Haywood and Hamish Macleod

Keywords
learning, higher education, student, videogames

1. Abstract
From over thirty years ago when a Pong prototype was first switched on in Andy
Capp’s bar in California, computer, arcade and videogames (from here on referred
to collectively as videogames) have made a significant cultural, social, economic,
political, and technological impact on society (Newman, 2004). Since the launch
of Pong in the 1970s, computer and videogames have grown into a $30 billion
worldwide industry. Between 2000 and 2004, the UK’s consumption of videogame
products and activities grew by over £15 billion, with the recent successful UK
and US launches of Microsoft’s Xbox 360 and Sony’s PlayStation Portable (PSP)
not yet included in those figures. In recent years there has been an upsurge in the
number of researchers studying gaming. Topics have included investigations of the
reasons why people play videogames, the potential of games in educational settings,
and the effects which these games, and the culture which surrounds them, may be
having on society. As videogame playing becomes more widespread, and increasing
numbers of videogame playing students enter higher education, educational
research into the impact of extensive videogame playing on student approaches
to learning and attitudes towards higher education is vital. The chapter begins
by presenting an introduction to the culture of videogames and continues with a
discussion on the relevant literature in this research area. An overview of our
research project currently being carried out in the School of Education at the
University of Edinburgh in the UK will be provided. The project is seeking to
explore the experience of videogame play among students in higher education, and
to identify the differences which might exist between gamers and non-gamers in
their approaches to learning and study, and in their attitudes towards higher edu-

85
 cation. Finally, key results of our project will be presented and discussed in the
context of current research.

2. “Everything bad is good…”

Videogames represent the fastest-growing sector of the global media and entertainment
industry and dominate much of young people’s leisure time. Presently, the videogame
industry’s revenue surpasses that of the movie industry. In 2004, the average amount
of time eight to eighteen year olds spent playing games per day on a computer, console
or handheld was 49 minutes, up from an average of 26 minutes per day in 1999. Of
the 2000 children surveyed, 22% played videogames for more than an hour per day
(Kaiser Family Foundation Report, 2005). By the age of twenty-one the average
American has played 10,000 hours of videogames compared to 3,000 hours of book
reading (Prensky, 2001a).

From their launch, videogames have attracted widely disparate responses from media,
educators, parents, academic researchers, and even religious groups and politicians.
“In America in the early 1980s President Ronal Reagan extolled the virtues of
videogames to create a generation of highly skilled cold war warriors, while at
he same time the US Surgeon General C. Everett Koop proclaimed videogames
among the top health risks facing Americans. To be sure, such extreme cultural
reactions to technological and cultural innovations are hardly new; mid twentieth-
century critics feared that television watchers would become addicted to television,
never leaving their homes, and critics before them feared that film would pervert
viewers”
(Squire, 2002).

Koop claimed that the sole objective of most videogames seemed to be kill, destroy
and eliminate the enemy and that they contained nothing constructive. Koop eventu-
ally explained that he was not basing his opinion on any scientific evidence and that
it was merely personal opinion, yet the debate continues even today. In the summer
of 2005, following the Grand Theft Auto ‘Hot Coffee’ controversy, in which hidden
code within the videogames’ programme could be unlocked to reveal pornographic
scenes, Senator Hillary Clinton called on the US Federal Trade Commission to instigate
a 90 million dollar study on the effects of videogames on children. Steven Johnson,
author of Everything bad is good for you: how popular culture is making us smarter,

86
responded with an open letter in the Los Angeles Times addressed to Senator Clinton,
refuting the claim that popular culture is dumbing-down our society. Johnson (2005a)
states that on the contrary “popular culture has been growing increasingly complex
over the past few decades, exercising our minds in powerful new ways” (p13) and
highlights videogames as “the best example of brain-boosting media” (Johnson, 2005b).
Here in the UK, Prime Minister Tony Blair, under intense pressure from politicians
and parents in 2004 following an enormous media frenzy surrounding the content
of the videogame ‘Manhunt’, commissioned a new research study into a possible
connection between videogames and violence, which is due for publication in 2006.

With his influential book, What video games have to teach us about learning and
literacy, James Paul Gee (2003a) finally gave intellectual respectability to academic
reflection on videogame studies. He points out that “the phenomenon of the videogame
as an agent of mental training is largely unstudied: more often, games are denigrated
for being violent or they’re just plain ignored” (Gee, 2003b). He describes videogames
as media that require deep thinking and complex problem solving skills and which
are fun, immersive, and spectacularly successful at engaging players. In his research,
Seymour Papert (1996) claims “what is best about the best games is that they draw
kids into some very hard learning” (p88). He writes about children using the term
‘hard fun’ to describe videogames, and he immediately explains, “[the kids] don’t
mean [the videogame] is fun in spite of being hard, they mean it is fun because it is
hard” (ibid: p88). Therefore, videogames are played because they are essentially diffi-
cult and challenging. As any successful games designer knows, if videogames are too
simple and easy they will not be played, and if they are too hard and complicated
they will also be avoided. Gee (2003b) states that “each level [of a successful game]
dances around the outer limits of the player’s abilities, seeking at every point to be
challenging enough to be hard but doable”, never moving so far beyond a player’s
comfort zone that they will give up. This he says “is referred to in cognitive science,
as the regime of competence principle, which results in a feeling of simultaneous
pleasure and frustration” (Gee, 2003b). He believes that “learning works best when
challenges are pleasantly frustrating” (ibid). He does not appear to be advocating the
implementation of videogames in classrooms, but identifies the key design principles
within some of the most popular videogames and presents these as learning principles
that should be applied and integrated into educational curricula. Gee’s current re-
search would seem to be foreshadowed by the work of Tom Malone in the 1980s,
who argued that the education community had much to learn from videogames,
and from the enthusiastic engagement that they motivate in people (Malone, 1981).

87
3. The changing nature of students
Educating our students is a primary goal of colleges and universities. Reaching that goal,
however, depends on understanding our learners, which is an essential component in
facilitating learning. Gaining an understanding of our students’ behaviours, attitudes
and expectations gives us more options for engaging students in the learning process
(Oblinger, 2003).

In parallel with the observations of authors such as Johnson, there exists a burgeoning
interest in the changing nature of students in higher education, with many researchers
attributing this change to the impact of new media (internet, videogames, mobile
phones) on students. Students who are engaged with these types of media were born
in the 1980s or later and are often referred to as the ‘Net Generation’ or ‘Net Gen’
(Oblinger, 2003). Today’s students are accustomed to the ubiquitous presence of
digital technology in their lives, for example, computers, videogames, video cameras,
mobile phones and digital music players, all of which demand ‘new media literacy’.
The students currently in higher education are different from their predecessors in
many ways, but particularly because they are the first generation to grow up surrounded
by digital media. The rate of change between generations is perceived as accelerating,
with technology at the heart of that acceleration. Marshall McLuhan (1964) coined
the term ‘electric speed’ to describe this phenomenon. Johnson (2005a) explains that
“[technology] introduces new platforms and genres at an accelerating rate. We
had thirty years to adapt to the new storytelling possibilities of cinema; then
another twenty for radio; then twenty years of present-tense television [...] five
years to acclimate to the VCR and video games; then e-mail, online chats, DVDs,
[...] the Web—all becoming staples of the pop culture diet in the space of a decade”
(p175).

Authors are beginning to claim that as a result of this ubiquitous digital environment,
the minds of the students in this generation have literally been altered (‘rewired’ is a
popular term often used). Marc Prensky (2001a) claims “people who undergo different
inputs from the media and culture that surround them can, and do, think differently”
(p43). He describes all students as “native speakers of the digital language of computers,
games and the Internet” (Prensky, 2001b, p1), and he uses the term ‘digital native’ to
describe these students collectively. He believes that inputs such as computer games,
television and mobile phones change the way digital natives think and behave. They
make decisions faster, at so-called ‘twitch speed’, and can easily parallel process (the

88
art of taking in information from several sources at once) and multi-task. He refers
to those who were not born into and did not grow up in a digital world, as ‘digital
immigrants’, those with a foot in the past, and an ‘accent’ they will never lose, which
is constantly visible and identifiable through their behaviours, such as going to the
internet second rather than first in search of information and the printing of their
email. He claims that digital natives, who are empowered in their personal lives by
and immersed in interactive technology, find old teaching methods ‘horribly boring’
and have to ‘power down’ at school. Prensky maintains that the “single biggest problem
facing educators today is that our digital immigrant instructors, who speak an out-
dated language (that of the pre-digital age), are struggling to teach a population that
speaks an entirely new language” (Prensky, 2001b, p1). He believes that the best way
to ignite this missing ‘spark’ of learning is to bring computer games into classrooms.

In contrast, Squire (2002) writes “some advocates of digital game based learning imply
that developing educational games is a moral imperative, as kids of the ‘videogame
generation’ do not respond to traditional instruction” and he continues by “cautioning
against overexuberance about the potential of digital games to transform education”
(Squire, 2002). He believes that “games are very particular kinds of experiences and
that playing games does not appeal to everyone, and certainly no one game (or more
appropriately game experience) appeals to everyone” (Squire, 2005). It follows that
even if students have been exposed and had access to various new types of media, one
cannot presume, that these students will have chosen or choose to be game players
(past or present). The dichotomous metaphor of ‘digital natives, digital immigrants’
seems to imply a presumption that all young people make up a homogeneous group,
a group who are immersed in the culture of digital technology and videogames, and
that therefore, all young people can be classified collectively as ‘digital natives’, a
presumption that we are challenging through our research. The variance that exists
within this generation (i.e. ‘net generation’) needs to be acknowledged and understood.

While research into the ways in which new media affects a young person’s cognitive
development is ongoing, it seems less clear how differences based on previous game
experience might impact on students in higher education. John Beck and Mitchell
Wade (2004) looked to young adults within the game generation to observe the
impact of new media (videogames) on their orientation towards business. They point
out that the prevalence of games has shaped how younger adults and children, both
males and females, think and learn (Beck and Wade, 2004). The authors report that
people who have intense experiences with digital media, in this case videogames, in

89
their formative years, will be “a different kind of future employee—people who are
confident with risk and surprise, who regard simulation and fantasy as a useful tool
not a distraction, and who see themselves as the heroes of their own narratives” (Beck
and Wade, 2004).

4. Videogame Research in Higher Education

It is against this background and in parallel with Beck and Wade’s research that we
are conducting our research at the University of Edinburgh. We are investigating the
gaming experiences of student childhoods to explore whether there is a relationship
between a student’s past experiences of (i.e. engagement with) videogames and their
current approaches to learning and study and their attitudes towards specific aspects
of higher education (e.g. feedback and collaboration). From the research discussed
earlier in this chapter further investigation is needed into the connection between
the two elements. In the context of our research we have defined videogames to be
‘all types of electronic, handheld, console and computer video games’. Our research
question is:
What impact or influence does extensive videogame play in pre-university
years have on the attitudes of students to education and specifically does it
appear to influence their learning styles?

The method used in the first step of the project’s data collection was a survey, which
sampled over 1200 full time undergraduate students from a wide range of disciplines
(e.g. Archaeology, Nursing, Divinity, Electronics) across every year of their degree
(bachelor) courses. An option of online survey or paper-based survey was offered to
students in an effort to ensure high completion rates. It was vital to our research that
all students in our data analysis were of a certain age category so that they would have
had potentially similar media backgrounds; therefore, only young adults between the
age of 17 and 24 years old were included in the analysis of the data. For the purposes
of our analysis, rather than grouping the students by subject area, of which there were
a very large number, we grouped them into four broad cognate areas which we have
found useful in the past (i.e. Medicine and Veterinary Medicine, Humanities, Science
and Engineering and Social Science).

90
The survey was designed to gather data from each student on three major themes:
• Attitudes towards higher education
• Approaches to learning and study
• Videogame play experience.

It was clearly not possible to completely capture the rich diversity of students’ approaches
to learning and study and their attitudes towards higher education within a simple
survey and so we focussed on some key elements to provide a baseline so that any
differences between student groups could be highlighted based on their previous video
game engagement. Emerging themes from the analysis of the survey (discussed in the
results section) could then be analysed further using qualitative methods. These key
elements likely to have an influence on students’ approaches to learning and study
and attitudes towards various aspects of higher education were identified from previous
research (Marton et al, 1997). The majority of responses were elicited using a 5-point
Likert scale. The students were asked to rate their general approaches to learning and
study, rather than for specific approaches used towards one course or course module.

Please indicate your level of agreement with each of the following statements on a scale of strong
agreement to strong disagreement.

For the most part, I manage to hold my concentration when I attend lectures ≤ ≤ ≤ ≤ ≤
I concentrate on learning just the information I need to know to pass ≤ ≤ ≤ ≤ ≤

In general, I prefer to study on my own rather than with other students ≤ ≤ ≤ ≤ ≤

On the whole, I am quite systematic and organised in my study ≤ ≤ ≤ ≤ ≤

Figure 1: Sample of survey questions

Data were gathered on previous videogame play experiences of each student. Judging
that it would be difficult for students to recall their exact age when they first played
videogames, which for some students could be almost twenty years ago, we asked
students to recall at which stage in their childhood they played videogames. Students
were also asked to recall how often they played videogames at each of these stages.

91
 Can you recall the period(s) in your life during which you played a video game? How frequently did
you play?
Every Most Most Relatively Never
Day Days Weeks Infrequently

During Primary School (Elementary School) ≤ ≤ ≤ ≤ ≤

Early Years of Secondary School (Junior High School) ≤ ≤ ≤ ≤ ≤

Late Years of Secondary School (Senior High School) ≤ ≤ ≤ ≤ ≤

In University ≤ ≤ ≤ ≤ ≤
Figure 2: Sample survey question on videogame play experience

Questions were included that focussed on other aspects of videogame play experiences,
such as how frequently students participated in online gaming communities, how
frequently students purchased videogame-related materials and what type of video-
game they played most often (e.g. adventure, strategy or sports). To allow for potential
constraints of time, parental control and/or monetary issues in relation to the fre-
quency of a student’s videogame play, we also asked the students if they could recall
ever having the desire to play videogames more frequently than opportunities (e.g.
parents/guardian/teacher or time) allowed. We believe that data gathered on these other
aspects of videogame play might play an important role in future research analysis.

5. Emerging Themes
The sample analysed included 1239 full time, undergraduate University of Edinburgh
students aged between 17 and 24 years old. The majority of respondents (64%) were
female, and over 60% of all students completed the paper-based survey. Almost half
the respondents (46%) were from Science and Engineering, with 14% from Medicine
and Veterinary, 24% from Social Science and the remainder from Humanities. Around
40% of students were entry-level undergraduate students and over 35% of students
were in the second year of their course.

All Students Male Female

Ever Played Videogames 95% 99% 94%

Currently Play Videogames 39% 69% 22%

Table 1: Videogame play (by percentage of gender)

92
Students were asked had they ever played a videogame and 1182 students (95%) said
yes, with an almost even distribution between the percentage of males and the percen-
tage of females (see table 1). However, when asked if they currently played videogames
only 39% of all students said yes, with 69% of all males surveyed currently playing
videogames but only a mere 22% of females a fact which requires further investigating
using qualitative methods.

One of the key steps for this research is to produce empirically derived definitions of
‘formative years’ and ‘intense engagement with videogames’ that can be used to observe
the influence of previous videogame play on students’ approaches to learning and
study and their attitudes towards higher education. Drawing on the work of Jean
Piaget in the 1950s and through our data analysis, we produced the following defini-
tions. We define formative years as ‘the childhood period up to and including the age
of 14 or 15 years old, i.e. the end of a child’s early secondary school period (junior high
school)’ and we define a person who has an intense engagement with videogames as
someone who ‘played videogames every day or most days at some stage within their
formative years (as defined)’. The definitions are, however, still being refined. Using
these definitions we assigned each student to one of two videogame player categories:
High Gamer or Low Gamer.
• High gamer includes all students who at some stage in their formative years
(either in primary or early secondary or both) had an intense engagement
(played every day or most days) with videogames.
• Low gamer includes all students who never played videogames in their formative
years and students who played videogames relatively infrequently or most weeks
at some stage in their formative years.

Total Number Total Percent Male Female

Low Gamer 753 61% 156 597

High Gamer 486 39% 294 192

Total 1239 450 789

Table 2: Videogame player categories (by gender)

93
Thirty-nine percent of all students surveyed were classified as high gamers, of which
60% were male. The majority of male students were high gamers (65% of all males
surveyed were high gamers) and the majority of female students were low gamers
(76% of all females surveyed were low gamers). As detailed earlier, we also recorded
data on various other aspects of their videogame play experiences that might prove
influential in future research. When asked if they could recall ever having the desire
to play videogames more frequently than opportunities allowed, over 69% of high
gamers recalled having a desire to play more often, as compared to 28% of low gamers
and 19% of high gamers recalled having this desire frequently, as compared to a mere
3% of low gamers. Over 25% of high gamers reported having participated in online
videogame communities, but less than 6% participated on a frequent basis. Over 50%
of high gamers recalled taking part in videogame-related activities (i.e. purchasing
videogame magazines and visiting videogame-related websites) with less than 10%
of high gamers participating on a frequent basis.

Some of the specific questions used in the survey were detailed earlier in this chapter.
During our analysis each question on approaches to learning and study and attitudes
towards higher education was ranked with a score between 1 and 5. The higher the
student scored the stronger they agreed with the statement given. Through detailed
data analysis, key components which focussed on student approaches to learning and
study and attitudes towards aspects of higher education emerged: organised approach
to study, studying for understanding (relating ideas while studying), exerting minimal
effort to pass, attitude towards collaboration (group work), difficulty concentrating in
lectures and attitude towards qualitative feedback.

So for example in relation to the component ‘organised approaches to study’, the

higher a student scored, the more organised they reported themselves to be in their
approaches to study.

During our analysis we compared the means of low gamers and high gamers (using
the statistical t-test) in relation to the key elements detailed earlier. We also carried
out tests of independence asking ‘is there a relationship between previous videogame
play (each videogame category) and each key finding?’ If the test was found to be
statistically significant (p £ 0.05) then there exists strong evidence to imply that a
relationship may exist between the two. As our project is still ongoing it is only possible
at this stage to present key findings from the survey analysis, all of which require
further investigating using qualitative methods.

94
 Low Gamer High Gamer Difference % Difference Tests of
N = 753 N = 486 +/- (Low gamer to Independence
(Mean) (Mean) High gamer) (p)

Organised approaches to study 2.98 2.79 -.19 -6.37 P<0.0005

Exerting minimal effort to pass 3.07 3.25 .18 +5.86 P<0.0005

Difficulty concentrating in lectures 3.28 3.09 .19 -5.79 P=0.002

Willingness to collaborate 3.22 3.22 .00 0.00 Not Sig

Table 3: Some Key Findings from our survey analysis

Low gamers scored higher when reporting how organised they were in their study,
which implies that low gamers are more organised in their study, as compared to high
gamers. The percentage difference between low gamers and high gamers is 6.37, and
the test of independence (detailed earlier) is found to be highly significant (p < 0.0005),
which gives strong implications that a relationship may exist between the two elements
and therefore we can state that students who had low engagement with videogames
in their formative years are currently more organised in their study, as compared to
students who had intense engagement with videogames in their formative years.

The second finding highlighted in Table 3, refers to another aspect of the student’s
approach to study, which we named ‘exerting minimal effort to pass’. The higher a
student scores on this factor the more inclined they are to do the least amount of
study possible while still passing. When comparing the means of each group, high
gamers scored higher than low gamers, with a percentage difference of 5.86. This was
also found to be highly significant (p < 0.0005) and therefore we can conclude that
students who had intense engagement with videogames in their formative years exert
the least amount of effort in their study, as compared to students who had low engage-
ment with videogames in their formative years.

When asked if they find it difficult to concentrate in lectures low gamers scored higher
than high gamers, with a percentage difference of 5.79. This finding was also found
to be statistically significant (p = 0.002) and therefore we can conclude that students
who had low engagement with videogames in their formative years reported finding
it more difficult to concentrate in lectures, as compared to students who had intense
engagement with videogames in their formative years, a key finding that would seem
contrary to the US based research that was discussed earlier.

95
Finally, in relation to a student’s willingness to collaborate with their fellow students,
no difference, significant or otherwise, was found between the low gamers and high
gamers. Both high gamers and low gamers would prefer to work alone than with
other students. Essentially it should be noted that students who had high engagement
with videogames in their formative years are no more likely to be willing to work
collaboratively than students who had low engagement with videogames in their
formative years.

During the course of our analysis we also worked with other videogame player cate-
gories which we referred to as ‘Hardcore Gamer’ and ‘Non-Gamer’. Hardcore gamer
includes all students who at some stage in their formative years played videogames
every day. Non-gamer includes all students who had never ever played videogames
in their formative years. Assigning students to the categories of hardcore gamer and
non-gamer decreased our sample size by over half, with over 700 students not belon-
ging to either category. However when we tested these exclusionary gaming categories
against the key elements of the survey (highlighted in table 3) the exact same findings
emerged. The percentage difference between the two groups (non-gamer to hardcore
gamer) did increase (as compared to the percentage difference between low gamer to
high gamer), but the direction of the difference and the significance of each was the
exact same. We feel that finding differences between groups that are closer on the
videogame player spectrum (i.e. high gamer and low gamer) is more important for
the purposes of our research than finding differences between such extreme groups
as hardcore gamers and non gamers, where differences might be expected to occur.

6. Future Directions
It is important to bear in mind that the research project is still ongoing and that the
findings highlighted above are an element in a larger research project. It is possible
that larger differences do exist between aspects of approaches to learning and study
and attitudes towards higher education that have not been highlighted here.
Interviews and focus groups with high gamers and low gamers will constitute the next
stage of this research, with the aim of further exploring the survey findings.

96
7. Authors

Fiona Littleton is a research assistant and PhD student. Jeff Haywood and Hamish
Macleod are senior academic staff members. All authors work within the Department
of Higher and Community Education, School of Education, University of Edinburgh,
United Kingdom. Their shared research interests include student learning, higher
education, videogames and popular culture.

8. Acknowledgements
The authors would like to thank the following people who have assisted with the
research project and the publication of this chapter: Si‚n Bayne, Michael Begg,
Andrea Christoforou, Denise Haywood, Velda McCune and David Tosh.

9. References
Beck, J. C. and Wade, M. (2004). Got Game: how a new generation of gamers is re-
shaping business forever. Boston, Massachusetts, Harvard Business School Press.

Frand, J. (2000). The Information Age Mindset: changes in students and implications
for higher education. Educause Review 35(5): 15-24.

Gee, J. P. (2003a). What video games have to teach us about learning and literacy.
New York, Palgrave Macmillan.

Gee, J. P. (2003b). Games, not school, are teaching kids to think. Retrieved Jan 30,
2006 from http://www.wired.com/wired/archive/11.05/view.html.

Johnson, S. (2005a). Everything bad is good for you: how today’s popular culture is
actually making us smarter. New York, Riverhead Books.

97
Johnson, S. (2005b). Dome Improvement. Retrieved Jan 30, 2006 from
http://www.wired.com/wired/archive/13.05/flynn.html.

The Kaiser Family Foundation Report. (2005). Generation M: Media in the lives of
8-18 Year-olds. Retrieved Jan 30, 2006 from http://www.kff.org/entmedia/7251.cfm.

Malone, T. W. (1981). Towards a Theory of Intrinsic Motivation. Cognitive Science

5(4): 333-369.

Marton, F., Hounsell, D. and Entwistle, N., Eds. (1997). The experience of learning:
implications for teaching and studying in higher education. Edinburgh, Scottish
Academic Press.

McLuhan, M. (1964). Understanding Media. London, Routledge.

Newman, J. (2004). Videogames. London, Routledge.

Oblinger, D. (2003). Boomers, Gen-Xers and Millennials: Understanding the New

Students. Educause Review 38(4): 37-47.

Papert, S. (1996). The Connected Family. Atlanta, Longstreet Press.

Piaget, J. (1953). The Origin of Intelligence in the Child. London, Routledge.

Prensky, M. (2001a). Digital game-based learning. New York, McGraw Hill.

Prensky, M. (2001b). Digital Natives, Digital Immigrants. On the Horizon 9(5): 1-6.

Squire, K (2002). Cultural framing of computer/video games. Retrieved Jan 30, 2006
from http://www.gamestudies.org/0102/squire.

Squire (2005). “Changing the game: what happens when video games enter the
classroom?” Retrieved Jan 30, 2006 from
http://www.innovateonline.info/index.php?view=article&id=82.

98
99
100
Virtual Patients: considerations of
narrative and game play
Michael Begg, Rachel Ellaway, David Dewhurst and Hamish Macleod

1. Introduction
Virtual patients (VP) and virtual cases (one or more virtual patients or other participants
in a contextualised setting or scenario) are increasingly common tools in healthcare
education. Intended to be used as a proxy of a real patient a VP represents those aspects
of the patient that are relevant to the educational context in which it is used. Thus
two uses of the same VP may use significantly different data or involve the VP in very
different ways (Ellaway, 2004).

Although the increasing use of web technologies in healthcare education is one driver
for their use, a more important one is that opportunities for medical students to access
real patients are rapidly decreasing since the trend in healthcare is for patients to spend
less time in hospital and for many procedures to be carried out entirely on an outpatient
basis. This has meant that viable alternatives are urgently required to maintain the
quality and efficacy of the medical education process. It also suggests that the primary
role for the use of VPs is one of prosthesis; providing an imitation of the real thing.
This, it will be suggested in this chapter, need not necessarily be the best approach
to considering the development and use of VPs.

While VPs simulate aspects of individual processes and procedures, it may be argued
that they presently offer little in the way of extending this simulation to the complex
decision making challenges that are an essential part of professional clinical practice.

The literature has indicated how successful gameplay and experiential learning
opportunities share common aspects (Gee, 2003). It would seem that in the paradigm
of the clinical case, in which there may be a degree of contextual role play (interactions
which carry consequences for the scenario, or narrative engagement with the learner),
the VP offers an opportunity for ‘game informed learning’ (Begg, Dewhurst et al., 2005).

101
Currently, most VPs are linear constructs that are developed around single scenarios
and linear decision paths with limited scope for immersive engagement for the student.
Choices are largely constrained and student’s choices do not substantially affect the
predetermined outcomes. In this chapter it is suggested that by adopting the idea of
the ‘virtual as potential’ as an alternative to the ‘virtual as prosthesis’ this may suggest
a means of developing VP activities that have requisite complexity (Greif, 1991),
more meaningful engagement for learners, and that more effectively promote the
development of professional decision making skills. This approach, it is also suggested,
is also more conceptually in keeping with the underlying architectural principals of
VP data structure—that is, abstracted, object classes that remain open to a multiplicity
of possible contexts and forms of manipulation.

In order to do this the idea of the virtual and how it presently relates to contemporary
healthcare education will be examined. The importance of the opportunities presented
to healthcare education through the use of VPs—principally the increasing emphasis
placed on experiential and situated learning and the need for the development of
decision making skills will be highlighted. The chapter will also examine the potential
benefits of realigning our appreciation of the virtual from prosthesis to potential and
how this may influence and encourage learning applications that are not only more
rewarding in terms of complexity and engagement than those currently in use, but
are also more closely aligned to the flexibility of choice and multiplicity of contexts
encouraged by emerging VP interoperability and modelling standards and specifications.

While it remains outside the scope of this work to develop a taxonomy of game-
informed elements for inclusion in VP data frameworks, it is anticipated that a case
can be made for a more substantive consideration of game and narrative influences
within the essentially fictive construct of the VP case.

2. The Virtual
Given the popularity of the word virtual in our culture, it will serve us well to equip
ourselves with an awareness of the origins of the term. It may be very well to have a
virtual learning environment, populated with VP cases that we discuss in virtual chat
rooms with virtual friends represented by virtual avatars, but the simple fact of the
term’s ubiquity in the contemporary academic vocabulary does not teach us about the
essence of the term.

102
Through the evolution of our understanding or interpretation of what the virtual is,
many artists have borne witness to the complexity of the concept. From the tantalisingly
tangible yet resolutely flat trickery of trompe l’oeil painting to the richly textured
journeys into memory within Proust’s A la recherche du temps perdu, from Borges’s
Labyrinths, to Nicholson Baker’s involved adventures in frozen moments of time,
whether it be explicitly concerned with space or time, the virtual has—and continues
to—inspire, entrance, captivate, engage and enlighten.

The concept of the virtual has been with us for many centuries. The word itself has
its origins in the latin virtus, meaning strength or power and evolved to virtualis
signifying something or someone of outstanding quality: the virtual person being
what we would now commonly understand as the virtuous person (Shields, 2003).
More recently, Baudrillard suggested that as a culture we have become so involved
with representations, simulations and signs of the real, that the real has been to all
intents and purposes lost (Baudrillard, 1995).

An alternative approach, however, (arising largely from Deleuze), considers the virtual
to be somewhat more than merely a prosthetic for a missing reality.

The suggestion is that it would be a mistake to consider the virtual as in any way
opposed to the real. Pierre Levy’s Becoming Virtual offers this:

“The virtual, strictly defined, has little relationship to that which is false, illusory
or imaginary. The virtual is by no means the opposite of the real. On the contrary,
it is a fecund and powerful mode of being that expands the process of creation,
opens up the future, injects a core of meaning beneath the platitude of immediate
physical presence (Levy, 1998)”

Ryan observed that the relationship between the actual and the virtual is a one-to-
many relationship. The virtual has the potential to be realised in any number of ways
depending on the individual context of use. From an educational point of view this
squares conveniently with learning technologists’ appreciation that a primary strength
of online learning is its capacity to support individual learning styles within a single
system. Anyone with a passing familiarity with the contemporary drive within learning
technology for reusable, multi-contextual components and aggregates cannot help
but be drawn in as Ryan continues:

103
 “The concept of virtualisation is an extremely powerful one. It involves any mental
operation that leads from the here and now, the singular, the usable for once-and-
for-all, and the solidly embodies to the timeless, abstract, general, multiple, versatile,
repeatable, ubiquitous, immaterial, and morphologically fluid… It is through the
consideration of the virtual as potential that the mind puts together representations
that can act upon the world (Ryan, 2001).”

3. Narrative, videogames and the Virtual

The near ubiquity of personal computers in more recent years has brought forward a
further multiplicity of contexts—in the home, in the work place, in education—for
the term virtual. We can play a part in virtual societies and become active in virtual
communities that promise to help to counter the alienating, fractured nature of our
post-modern lives (Rheingold, 1994). Equally, we can make use of virtual reality to
learn how to fly aeroplanes, develop virtual characters and engage in social and
gaming behaviour with other virtual characters, earn spare cash in virtual action
rooms, and lose all these earnings in virtual casinos and shopping malls.

Computers are in essence virtual machines. Internally they do little more than respond
to binary instructions. However, when given a set of instructions within a specific
context of use and using appropriate interfaces they become much more complex
phenomena; metaphorical representations, simulations, a prosthetic, perhaps even,
as Turkle has suggested, a part of our selves (Turkle, 1984; Turkle, 1997).

Emergent narrative, as described by Murray (1997) suggests a new form of narrative,

one that is constructed directly through interaction rather than entirely predefined.
Video game developers realised that their developments—and consequently the
experience of players -was enriched by making the play personal. No two players
should have the same experience of a single game world. A single player should have
a different game experience depending on how they approach the game. As far back
as the early 1980s Malone identified dynamic feedback as being a significant factor in
a game’s ability to engage a player (Malone, 1981). If feedback is not instantaneous,
and is completely irrelevant to the move the player has made, then engagement is in
danger of being lost. Consequential agency—the feeling that a player can have a real
substantive effect upon the game world would also seem to be central to the success
of these environments. Bruckman has reported on how players react negatively when

104
they feel unfairly or unrealistically constrained by a game—they try to break out of
that environment! (Bruckman, 1992). The game or emerging narrative must respond
appropriately and believably then to the player’s interactions. The game narrative is, as
it were, a world which the player inhabits and should not begin to consider leaving.

Emotional simulation—a condition that, in narrative contexts, calls upon the empathy
of the reader for the character, and the emotional performance that this empathy
encourages, is another factor suggested by Ryan as being significant in being able
to successfully bring about prolonged engagement with the narrative (Ryan, 2001).

So far, in this chapter, a number of issues surrounding the essence of the virtual,
particularly within a cultural framework in which computer mediation of contexts
of work and play is increasingly commonplace have been addressed. It will now turn
specifically towards healthcare education and the use of VPs.

4. The virtual in healthcare education

The increasingly widespread use of integrated educational support systems such as
virtual learning environments (VLEs) has marked a significant growth in our
dependence on new technologies to support and structure our learning frameworks
(Browne and Jenkins, 2003). At the same time the imperative for the healthcare services
is to reduce the amount of time patients stay in hospital. With fewer patients spending
less time in hospitals and clinics, opportunities for building learning activities around
real patients have decreased and various forms of representative simulation have become
an increasingly common alternative in healthcare education. However, such virtual
interventions commonly address single, linear challenges or processes; very little is
currently on offer to account for the loss of opportunity to pick up the tacit knowledge
of clinical practice that arises from the experience of actual involvement in a real
clinical scenario.

Even after graduation medical students will undertake many years of further training
before they may call themselves a full professional. This postgraduate training is
undertaken mostly in real clinical contexts, where knowledge, skills and understanding
are acquired through the experience itself. It would seem, therefore, that much of the
education of healthcare professionals depends upon the learning opportunities offered
through the actual practicing of skills in context.

105
As has already been suggested, VPs are increasing in popularity partly as a result of the
reduction in opportunities for this practical engagement with clinical practice and so
it would perhaps seem appropriate to suggest that we attempt to introduce as much
of the essence of real clinical scenarios as possible into VP structures.

5. Virtual Patients
Healthcare students are no strangers to VPs and they are encountered in a variety of
contexts: actors performing the role of patients in clinical examinations; mannequin
dummies in activities relating to, for instance, anaesthesiology and resuscitation; web-
based computer assisted learning (CAL) packages, in which students can interact with
case based self assessment exercises online, to name just a few.

These clinical simulations, simulated patients, CAL sequences, and so on already

exhibit certain characteristics that could be considered analogous to many of the
factors outlined above. For instance;
• Students may be encouraged, often indirectly, to adopt a character role through
which they can engage with the activity—most commonly that of a professional
clinical practitioner.
• Students are presented with situations that require them to interact directly with
them (from answering multiple choice questions in a CAL sequence to responding
to changes in physiological status in clinical simulations)
• The patients are often presented as characters, with names, personal and social
histories, in essence a story
• Student responses are often assessed and feedback given, enabling personal
reflection on how well they have performed.

It is also perhaps true to say, however, that the narratives that accompany many current
VP scenarios are simplistic and linear. The underlying frameworks supporting patient
data and associated resources are largely directed towards the convenience of teachers
and those creating the scenarios. The efforts being made to promote multiple contexts
for VP elements is a largely administration driven process. There would appear to be
comparatively little effort being made at using such initiatives to address the individual
experience of the learner.

106
At the University of Edinburgh there is an ever-growing range of web based VPs that
deal directly with numerous curriculum topics. The following are two brief illustrati-
ons of VPs currently used by undergraduate medical students:
• George is a VP in his mid fifties who medical undergraduates encounter first in
year one. Throughout the subsequent five years of their undergraduate programme
they will encounter George numerous times as his condition grows ever more
complex thereby matching their current levels of study. George offers real time
engagement and is a valuable platform for integrating learning tasks that had
previously been taught in isolation, such as social and cultural factors of health,
communication skills
• Hannah is another VP developed to provide learning aids in reproductive health.
The student is invited to learn more about Hannah on her journey from discovering
she is pregnant to the birth of her child. Students are quizzed on a variety of
associated topics and are offered numerous resources such as scans, lab reports,
case notes, etc to enhance the learning experience.

These applications, and others of their kind are popular with students. They offer
elements of a realistic context in which the students can interact with a variety of data
and resource types. However, as outlined above, they are currently tightly constrained
in how much reality they can represent. If a student offers an inappropriate response
to a question, say, regarding George’s prescription medication, there is no consequence
to that, merely feedback to indicate that their answer is wrong, an indication of what
the correct answer should have been and a prompt to move on to the next page. The
same would be true of Hannah, and other web based VPs.

It has been suggested that there is already much in VP cases that are analogous to
other narrative driven applications—videogames in particular. It is appropriate tp
highlight possible factors where a more considered application of narrative/game
informed elements could actively enhance the learner experience of interacting with VPs:

6. The learner as consequential agent

How might the environment allow the learner to feel that their interactions have real
consequence? Three possibly influential factors might be;

107
 • Emergent narrative: Murray’s (1997) description of emergent narrative suggests
that it is a form defined by interaction. The progress of the story is—as much as
possible—defined by the choices that the interactor makes. How might we enrich
the possible navigational pathways in VP applications to allow the learner to feel
that it is their own choices that are driving the progression through the case?
• The responsive environment: The learner will reasonably expect the environment
to respond to his/her input. These expectations should not necessarily be limited
to one right path and a few alternatives that lead quickly to a dead end. If a learner
can take an individual path through a case, which allows for reasonable deviations,
and that does not result in a termination point, the environment will likely prove
to be more engaging, more responsive, and may even allow for further knowledge
assimilation through information picked up on the individual’s choice of path.
How might we approach accounting for these diversions within VP applications?
• The psycho-social moratorium: Gee (2003) has illustrated how games allow for
the making of poor choices without any detrimental effect upon the richness of
engagement. The player—and for our purposes the learner—having failed to
meet a core objective, can (and does) return to try again. Successive attempts
can be made to attain the core objective—each attempt increasingly informed by
knowledge that has been successfully acquired through previous attempts. Once
successful, it is not unusual, in gaming contexts, for the player to return purely
in order to improve their performance. How might we structure our VP case
architectures to accommodate appropriate break points that would: allow learners
to make repeated attempts at core tasks; allow for a dynamic grading of per-
formance; encourage learners to try repeatedly to find their optimum path to
the desired outcome?

7. Labyrinth
‘Labyrinth’ is a new development initiative within the College of Medicine and
Veterinary Medicine’s Learning Technology Section at the University of Edinburgh.
The Labyrinth application is a content authoring and delivery platform for learning
activities such as VP case scenarios. Like the examples offered above, the application
can present fictive cases in which characters are introduced and a variety of tasks are
offered to the student, who will also be identified with a consequential role within the
application—commonly that of the health professional. Unlike the previous examples,

108
however, the emphasis within Labyrinth is on decision-making. The decisions made
by the student directly influence how the material and subsequent options emerge.

For example, in a module concerned with clinical admissions the student may enter
into a scenario in which they are contextualised as an agent in charge of an admissions
unit at the start of a nightshift. The student will read a basic introduction text then be
asked to make a decision as to what to do next: consult with registrar; establish what
resources are available; approach a patient; or make a cup of tea. Clearly, some choices
from this array would be more appropriate than others. The critical point is that no
matter what choice the student makes, the narrative continues unbroken. Electing ‘to
make a cup of tea’ will result in a text message indicating that your attitude may appear
questionable and that the registrar seems to be giving you a disapproving look.
Choices will then be made asking you whether you would now like to: consult with
the registrar; approach a patient; or establish what resources are available. Similarly, in
another scenario in which the student performs as a community GP on call, a situation
is presented in which the GP receives a telephone call at 8.30 in the evening from
the mother of a baby, concerned that her child is not settling and has a persistent
cough. You are asked to decide whether to: go and see the child immediately; instruct
the mother to administer Calpol and phone the surgery in the morning; or call an
ambulance. Again, each choice comes with its own consequences and these consequences
inform the subsequent narrative chain of events.

By logging all interactions it is possible to track an individuals progress through a case

and, if appropriate, map their choices against optimum pathways suggested by the
case authors. As well as the dynamic instant feedback received by students as they
progress through the case, they may also be presented with summary reports that
indicate where their interactions have differed from those which the author has
established as good, or even best practice.

This brief illustration offers an indication of how factors associated with successful
gameplay have been introduced to the basic VP activity: specifically, presenting an
environment that reacts dynamically and plausibly to the student’s interactions and
the emergence of a narrative that arises directly as a consequence of the student’s
decisions.

109
8. Discussion
Medical students do need to interact with real patients—this is what they are being
trained to do. It is apparent that the opportunities for such interaction are becoming
fewer. The use of VPs has grown in popularity as a means of presenting learning
opportunities that have the potential to situate learning within realistic contexts and
employ a range of data and resource types.

However, the present structuring of VPs limits interaction to a single structured

pathway, where the way in which the student responds to questions posed has no
impact upon the unfolding of the scenario.

The perceived function of present VP applications would seem to be largely

prosthetic; filling in the gaps left by the unavailability of human subjects. However,
as we have described, the realism of these situated cases is greatly limited by the lack
of consequential decision making opportunities.

It is suggested here that the essence of the VP should perhaps be more concerned
with potentiality than prosthesis. A deeper embedding within VP meta frameworks
of procedural aspects of narrative may impact positively upon the learner experience
of VPs—just as they have been shown to be a contributing factor to the success of
videogames.

The impact may be just as profound upon teaching practices, in particular content
authoring. It would seem more appropriate to create several potential pathways
through a clinical scenario producing a learning activity which will prove more
challenging than that of a single path. Authoring interfaces may be able to help
alleviate some of the challenges here, but there may be further potential in multiple
authorship and the interleaving of cases (bad decisions made by a student in treating,
say, a pharmacology case could result in a straying into a case addressing overdose, or
critical resuscitation skills, for example).

It is almost ironic to observe, however, that the kind of content structures that would
allow for multiple pathways to be taken through case based procedures are conceptually
in keeping with the imperative to maintain flexibility in the meta structures of VPs to
allow them to be used in part or in whole across multiple contexts (this sentence does
not make a whole lot of sense to me).

110
Making narratively rich, game informed decisions about how to present what information
and when to present it in the unfolding of the narrative may lead to the development
of compelling applications in which the learner feels driven to compete, to succeed, to
know—who feels caught up in a rich environment where there are critical choices to
be made—and where those choices have real consequence. While it seems likely that
the sense of engagement would be enhanced by the kinds of measures outlined in this
chapter another important new question is opened: is there any correlation between
learners sense of engagement with a computer mediated learning application and the
pedagogical impact of that application?

9. Authors

Michael Begg is the special projects manager for e-learning within the College of
Medicine and Veterinary Medicine’s Learning Technology Section (LTS) at the
University of Edinburgh, UK.

Dr Rachel Ellaway is the e-Learning Manager for the Learning Technology Section
in the College of Medicine and Veterinary Medicine at the University of Edinburgh
and is an e-Learning Subject Specialist Advisor to the HE Academy Subject Centre
for Medicine, Dentistry and Veterinary Medicine.

David Dewhurst, BSc, PhD, is a professor of e-learning at the University of Edinburgh,

UK. As an assistant principal of e-learning and e-health, he has a university-wide role
in supporting these areas. He is also the director of the Learning Technology Section
(LTS) within the College of Medicine and Veterinary Medicine

Hamish MacLeod is a senior lecturer within the University of Edinburgh’s Center

for Teaching, Learning, and Assessment (TLA).

111
10. References
Baudrillard, J. (1995). “Simulacra and Simulation (The Body, In Theory: Histories
of Cultural Materialism)”. Michigan, University of Michigan Press

Begg, M., Dewhurst, D. and Macleod, H. (2005). “Game Informed Learning:

Applying computer game processes to Higher Education.” Innovate 1(6).
http://innovateonline.info/index.php?view=issue&id=9

Browne, T. and Jenkins, M. (2003). “VLE Surveys: a longitudinal perspective between

March 2001 and March 2003 for Higher Education in the United Kingdom”.
UK, UCISA. http://www.ucisa.ac.uk/groups/tlig/vle/vle2003.doc

Bruckman, A. (1992). “Identity Workshop: Emergent social and psychological

phenomena in text-based virtual reality”. MIT. Cambridge, Massachusetts.
http://www.cc.gatech.edu/~asb/papers/old-papers.html#IW

Ellaway, R. (2004). “Modeling Virtual Patients and Virtual Cases.” Journal

Volume(Issue): Pages. http://meld.medbiq.org/primers/virtual_patients_cases_
ellaway.htm 13th October 2005

Gee, J. P. (2003). “What video games have to teach us about learning and literacy”.
New York, USA, Palgrave MacMillan

Greif, S. (1991). “The Role of German Work Psychology in the Design of Artifacts”.
Designing interaction: psychology at the human-computer interface. Cambridge
UK, Cambridge University Press.

Levy, P. (1998). “Becoming Virtual: Reality in the Digital Age”, Plenum Trade

Malone, T. (1981). “Heuristics for designing enjoyable user interfaces: lessons

from computer games”, Association for computing machinery

Murray, J. (1997). “Hamlet on the Holodeck: The Future of Narrative in

Cyberspace”. Cambridege, Massachucetts, MIT Press

Rheingold, H. (1994). “The Virtual Community: Finding connection in a

computerized world”

112
Ryan, M.-L. (2001). “Narrative as Virtual Reality: Immersion and Interactivity in
Literature and Electronic Media”. Baltimore, Johns Hopkins University Press

Shields, R. (2003). “The Virtual”. London, Routledge

Turkle, S. (1984). “The Second Self; computers and the human spirit”.
London, Granada

Turkle, S. (1997). “Life on the screen: Identity in the age of the internet”.
London, UK, Weidenfeld & Nicolson

113
114
Learning with Digital Agents—Integration
of Simulations, Games, and Storytelling
Ulrike Spierling

Keywords
Game-based Learning, Interactive Digital Storytelling, Storytelling Agents, Gaming
Simulation

Abstract
Digital agent technologies provide extensive opportunities for learning applications.
Special emphasis in this article is on a critical review of the chances, limitations
and risks associated with the use of anthropomorphic avatars and digital agent
representations. Depending on the different conceptual models and connected
metaphors, various interactive experiences can be shaped, as illustrated by several
examples. Different content types and different learning effects are discussed
that can be realized by a varying mixture of storytelling, gaming simulation and
interface design aspects. The main argument made is that the metaphor of a help
agent, appearing in a superior role towards a human user, is counterproductive to
wide acceptance, and that other applications of the agent, such as in a role of a
digital companion, are more likely to be successfully established. Finally, scenarios
are provided that show how distinct forms can be used in diverse learning situations
dealing with social skills, and in a constructive approach to learning.

1. Introduction: Digital agents in the context of learning

A “Digital Agent” is a concept that has been raising increasing interest over the last
15 years. In general terms, a digital agent is software, working on behalf of a user or
several users, and is sometimes also referred to as “software agent technology”. Commonly,
the field of software agent technology also includes the specification of agent-oriented

115
software architectures and programming languages, where an agent works on behalf
of other software components. The defining attributes of software agents, in contrast
to conventional objects or programs, are described in terms of their behaviour, such
as autonomy, intelligence, or reaction to the environment (Franklin and Graesser, 1996).
More specifically, these are models of human-like features similar to mental states, for
example beliefs, desires, and intentions, as well as internal world representations, formed
by ontological descriptions. In this article, the focus is on applications that make use
of the humanoid properties of digital agents for their interaction with users, supporting
specific conceptual models, metaphors and associated expectations. The user types
under consideration here are end-users of learning applications (learners), as well as
teachers or learning content authors.

1.1 Visions and motivation

In Human-Computer Interaction (HCI), the employment of digital agents in a
graphical interface follows the primary vision to achieve a bridging between the
communication styles of people and of computer technology. Instead of forcing the
user to learn and adapt to the vocabulary and styles of machines, the familiar symbol
sets that are used for human-to-human communication are used by the machine.
This includes natural language as well as nonverbal signs that are not used consciously
by people, such as facial expressions, gestures, voice pitch, and presence. The vision
resulted in the modelling of the “avatar”1 as an anthropomorphic approximation
of the computer to a human being. Avatars are perceivable embodiments of “software
agents”—in a sense applying the metaphorical division of “body” and “mind” to
software. “Embodied Conversational Agents” (ECA’s) form a current research topic
addressing the definition of their parameter sets, markup languages, and user
evaluations (Cassell, Sullivan, Prevost and Churchill, 2000). In summary, user interface
agents and their visible avatars serve as a conceptual bridge. The initial goal has been
to make technology more comprehensible and also to enhance it through new
emotional qualities—information technology obtains a human face. However,
anthropomorphism in the interface has also met criticism from the outset, for
example see a discussion of Don, Brennan, Laurel and Shneiderman (1992), and
will be discussed in section 2.

1 Avatar: (Sanskrit source: Embodiment, Incarnation) Originally, an avatar is the manifestation of a Hindu Deity (especially Vishnu) in a human,
superhuman or animal form. Today the term has been introduced in the context of computing to either refer to a visual representation of a user in a
multi-user environment, or, as in this article, of an intelligent software component for human-like communication with users.

116
Beyond their application as a proxy appearance for “the computer program”, avatars
as well as software agents have been applied in the storytelling and gaming industry
for years. Through Computer Graphics, animated characters have been brought to
life on the virtual stage, at first to tell linear stories2. In contrast to pre-produced
imagery in movies, virtual figures that appear to respond to a user’s actions in a
video game have to actually be rendered in “real time”, with the help of underlying
“behaviour” controllers. These are software agents, managing their avatar’s movements,
depending on user inputs, context information and internal world representations.
As a joint effort in Computer Graphics and Artificial Intelligence, mental states such
as personality traits, emotions, as well as conversational knowledge are added to build
“Virtual Humans”. Assuming that these agents are additionally equipped with more
sophisticated models of mimetic competence and plot coherence, I refer to them
as “Interactive Storytelling Agents”—a breed of digital agents that are capable of
responding in a meaningful way to a user’s request while retaining an overall story or
action plan defined by an author of a story.

The topic of “Interactive Digital Storytelling” is also a new interdisciplinary research

field, gaining interest from the media and games industries as well as computer
scientists dealing with agent technology. More than merely enabling human-like
interactions with ECA’s, dedicated story agents provide a digital presentation with
narrative coherence of content and dramatic tension, while it responds to users’
inputs. There are various and contradictory definitions of Interactive Storytelling,
depending on the starting point from which the approach is made. For example,
according to game designer Chris Crawford (2004), Interactive Storytelling is a “new
kind of computer gaming” through adding story complexity to a game, whereas
Andrew Glassner (2004) as a writer refers to it as “responsive narrative”. In either
case, it constitutes a hybrid form, containing elements of games, simulated humanoid
agents and storytelling. General conceptual issues of Interactive Digital Storytelling
applications are discussed in (Spierling, 2005a).

Finally, a class of applications have been conceived and developed from a computer
science perspective, which form the category of “Pedagogical Agent” technologies.
As a special case of ECA’s, these agents are semi-autonomous animated virtual figures,

2 A quick overview of the history can be found in Wikipedia’s “Timeline of CGI in movies—Pioneering uses of computer-generated imagery in
film and television”, http://en.wikipedia.org/wiki/Timeline_of_CGI_in_film_and_television

117
capable of showing and telling by using gestures and language. They are designed
“to teach”, namely to be able to demonstrate complex tasks within a tutorial-like
communication, and to motivate by conveying emotional responses. A survey of
these approaches to technology, which are still very much in their infancy, is provided
by Johnson, Rickel and Lester (2000).

1.2 Hypotheses
Games and storytelling can both be of major help in teaching situations and are
widely used for learning, with or without a computer. E-learning applications in
general should be easy to use, to let the learner focus on the content matter instead
of struggling with the interface. Therefore, the first hypothesis raised here claims that
digital agents with avatars can be used as entertaining and useful sparring partners for
learning in an E-learning application, while satisfying several learner/user interests
simultaneously: enhancing the human-computer interface, introducing actors that can
tell a story and present a learning topic, and that constitute believable game characters.

However, against the background of visions that once led to their development, and
regarding recently occuring concrete artefacts, the assignment of agents and avatars
also raises acceptance problems. On the one hand, this is due to a current state of
technology that is not yet mature enough to enable the desirable and necessary
performance. On the other hand, the thesis made here is that users often develop
misleading conceptual models of assumed virtual humans that hamper the establishment
of feasible and sustainable associations and expectations. For example, the predominant
existing role model for virtual agents has been their conceived occurrence as digital
“guide”, “teacher”, or “adviser”. Any metaphor of an “intelligent help agent” implies
a human-agent relationship of the agent in a way outmatching a human user. In a
learning application, such an artefact is likely to be criticized immediately, as soon as
its behaviour is not satisfactory—independent of its effective functionality. Therefore,
the major argument made in this article stresses the importance of designing a suitable
conceptual model, which should always place human users into the “driving seat”.
The most useful digital agents then appear to be digital companions acting as pawns
in a game, or in other words, acting as digital operators with limited tasks in strictly
defined scenarios.

In order to illustrate several application possibilities, examples and content types

are presented first. Then, their promises and provoked expectations for learning are

118
explored, while also discussing conceptual aspects of storytelling and simulations
involved in digital agents. Finally, learning scenarios with digital agents, beyond
their function as virtual guides, are outlined.

2. Examples of conceptual models and applications of digital agents

The term “conceptual model” is well established within HCI, constituting a metaphor
or mental image of a known concept such as an activity or a physical entity, leading
to certain expectations of users. It is explained as “a set of integrated ideas […] about
what it should do, behave and look like, that will be understandable by the users in
the manner intended” (Preece, Rogers and Sharp, 2002, p.40). Donald Norman (1988)
pointed out the distinction between the design of a conceptual model (the “designer’s
model”) and a user’s understanding of it (the “user’s model”). The characteristic problems
discussed arise when a software designer develops a certain conceptual model of a tool
to build, but the resulting appearance of the artefact, together with contextual conditions,
provoke a different one in the imagination of a user. The following is at first a discussion
of general conceptual models that can be observed with the use of virtual characters,
followed by specific example applications.

2.1 Different conceptual models of digital agents

User interface help elements
One of the most familiar examples of a help agent is the Microsoft paper clip, “offering”
help while using a tool-like office application. While its conceived purpose is beneficial,
and there are a lot of similarities to the first “guiding” avatars designed for learning,
user sympathy is ambivalent, and it rather often meets with criticism. Reasons have
been researched for example by Swartz (2003). For the context of this paper, two
characteristics shall be pointed out:
1. The agent occurs proactively on top of the work, without being asked for help,
setting the input focus to its widget, forcing the user to interact with it. The
designer’s conceptual model actually underlying this behaviour is that of a User
Interface element, particularly a dialog box. Unfortunately, it doesn’t look like a
dialog box—its appearance rather supports a mental model of an agent.
2. The visualization suggests the possibility to perform free text input of natural
language, presenting the verbal invitation to “ask questions”. In fact the number

119
 of questions that the system can answer is very limited. It only can refer to
indexes in a catalogue of simple keywords. It looks similar to a chatbot’s interface,
but without a chatbot’s functionality.

As a conclusion, one major reason for the unpopularity of the paper clip could be that
there is an inapt mixture of different possible ways to frame its concept. Basically, in
its role model of a “helper”, it appears as if it is not smart enough in its behaviour,
although the mere knowledge it has in its database should be satisfactory. It is just not
able to do small talk, which is what people would expect in reaction to its appearance.

Chatbots as virtual beings

Chatbots, in fact, are another example worth investigating. The term “chatbot” (also:
“chatterbot”) refers to a software robot program which attempts to maintain a textual
conversation with a person. The most important challenge that is faced by developers
is to create entertaining small talk in order to keep the conversation going as long as
possible. Technically, chatbots use simplest pattern matching of written user input to
find suitable answer patterns in a database, also called “knowledge base”. In addition
to prepared answer sentences completely written by a human author (the “botmaster”),
the chatbot can generate sentences using the typical techniques of “Eliza”, Weizenbaum’s
famous program from 1966 (Weizenbaum, 1966). Eliza is able to turn a user’s input
sentence into a counterquestion, by replacing sentence parts such as “I am” with
“Why are you” and reflecting the rest of the sentence back to the interlocutor. This
behaviour has proved to tempt people to believe that Eliza “understands” them, and
they judge it as intelligent behaviour—at least as long as they don’t become aware of
its rule-based and stereotypical nature. By annual competitions, such as the Loebner
Prize 3 and the Chatterbox Challenge4, chatbots are compared and tested with a method
introduced by Alan Turing in 1950 as the “imitation game” (Turing, 1950). Criteria
for winning the original “Turing Test” were simply that the program could pass as a
human. The current competition awards the best achieved indistinguishability to a
human interlocutor, while conversations between bots and real humans are compared
by a jury and the bots have to be detected by the judges. The remaining typical problem
with pure chatbot conversations is the lack of a coherent storyline or conversational
plan to be achieved, since sentence-based pattern matching is all there is.

3 http://www.loebner.net/Prizef/loebner-prize.html
4 http://www.chatterboxchallenge.com/

120
Chatbot technology recently finds an increasing interest in commercial websites for
advertising, or as a pre-sorting mechanism, answering first questions within an online
call center for customer service. Very often in this context, people take chatbots as
virtual beings that pretend to be human and immediately join in a testing mode
similar to the imitation game. Chatbots experience being asked very personal questions
that completely stretch the boundaries of their task and purpose and entertain comments
considered annoying for a human. For the purpose of entertainment, most bots are
designed to answer these questions up to a certain point. It is remarkable, that users
typically seem to develop a conceptual model of having a “being” as a counterpart,
but not as a fellow human being they would respect and encounter using the usual
rules of politeness. At the end of the day, humans clearly commit themselves to a
social interaction with the agents, assigning their perceived behaviour to apparent
character traits.

Virtual actors, puppets and dolls

In animated movies, puppet theatre and recently also in video games, virtual characters
are used as actors to play a drama on the virtual stage. The underlying conceptual
model is clearly not comparable to the function of autonomous agents, but rather
of agents serving as a medium, which is used to convey a story on behalf of an author.
The undeceived audience should be aware of the “fake” authenticity of the apparition.
However, in mass media, especially in TV series with introduced characters, “parasocial”
behaviour of the audience towards the fictional figures can be recognized (for exam-
ple, identification with the fictional character, or a one-sided affective relationship).
In stories, the goal is not to create for the mastery of an imitation game, but rather
for the “willing suspension of disbelief ” of the audience, a term coined by the poet
S. T. Coleridge. Interestingly, the attributes that lead to success are less the photo-
realistic visualization of the virtual actor than, again, its behaviour, especially in
the perceivable coherence of actions. This is typically obtained and supported by a
compelling story, with a clearly modelled goal and character of the actors. The parasocial
interaction is different from real social interaction, which has been identified for
interactive media and computers (Nass and Sundar, 1994).

Video games are a special case for the function of virtual actors. In a sense, they
resemble aspects of each of the last two mentioned metaphors (a “being” or an
“actor”), depending on the game genre that applies best. In the case of a simulation
game, the suitable metaphor rather is a “doll” than an actor, but with additional

121
mimetic talent compared to physical dolls. The main difference is that control
of a doll is completely up to the player instead of the author. An example game is
the best selling computer game “The SIMS”5, which allows players to simulate
social life while influencing parameters of single virtual dolls—including the possibility
to abuse them, treat them badly and make weird experiments, just like with real
dolls. The simulator will calculate their fate, based on the states of their parameter
set. In simulation games, there is no storytelling author controlling the coherence
of events. The only storytelling, if any, happens at the beginning, in order to explain a
mission or an outset. Actually, this is done partially by the players, who in fact are in
charge of creating their own stories with their dolls.

Virtual Human
The first attempts at building Artificial Intelligence (AI) entities to converse with,
as in the ones mentioned in the context of chatbots, provoked the vision and fostered
research into the possibility of making a machine “really understand”, to build machines
with consciousness. The philosophical debate between the notions of “strong AI” and
“weak AI” on whether or not this is possible will not be covered further6. Instead, the
practicality of the metaphor of a “virtual human” shall be critically investigated, as it
is a very likely conceptual model activated through a pedagogical agent appearing as
a surrogate teacher — even more so, since current research in computer graphics and
speech technologies also strives to perfect this vision. The idea of a fully-fledged
virtual human is a controversial concept as alarmed reactions of teachers have shown
in this context (Spierling, 2005b), fearing a substitution by automation. Additionaly,
expectations have been raised so high that the results are very likely to be a dis-
appointment, last but not least because of a phenomenon called the “uncanny valley”7.
This valley occurs as a drop in the plotted curve of empathic emotional responses
to increasing anthropomorphism in a robot, just when the “almost-human” stage is
reached, resulting in negative feelings such as weirdness. Bickmore and Picard (2004)
have reported insightful research results that show how especially designed agents
can influence the perception—or rather the illusion—of “caring”, perceived even by

5 The SIMS 2, produced by Electronic Arts, http://thesims2.ea.com/

6 In AI philosophy, this debate is still alive and goes back to John Searle’s argument that true consciousness can not be achieved by formal logic
systems, as illustrated in (Searle, 1980) by the “Chinese Room Argument”.
7 The term “Uncanny Valley” was coined by the robotics researcher Masahiro Mori in 1970, compare
http://en.wikipedia.org/wiki/Uncanny_Valley

122
computer-savvy persons who know that machines don’t experience genuine caring,
and pointed out the possible undesirable implications. As a result of these findings,
and especially as machine consciousness is far from reality, a user’s conceptual model
is preferable that raises lower technical expectations, as well as gives control and
comprehension to the user.

2.2 Example applications

Pedagogical Agents and instructional storytelling
The current state of the art in animated Pedagogical Agents aims at a new paradigm
for learning systems (Johnson, Rickel and Lester, 2000). In principle, the emphasis is
on face-to-face interactions between an agent and a student in a virtual, interactive
learning environment. Some classic examples are the applications Steve (University of
Southern California, CARTE Institute), Cosmo (North Carolina State University)
and PPP Persona (DFKI, German Research Center for Artificial Intelligence)8. Steve
has been applied to naval training tasks such as instructions for engine operating.
Students share a virtual environment with Steve by 3D stereoscopic vision and audio.
The Steve 3D-model is able to perform gaze directions and pointing gestures while
talking with an electronic voice. The model is visualized as an upper body floating
through virtual space to reach locations to point at or talk about. A similar approach
of combined gesture, locomotion and speech, referring to objects in the environment
while delivering problem-solving advice, is followed with Cosmo. While students use
the learning system about network topology as a construction game to solve tasks and
learn actively about network routing mechanisms, Cosmo’s task is to help, offer advice
and to explain factual knowledge. Its visualization is a similar functional abstraction
such as for Steve, however less anthropomorphic. PPP Persona guides a learner through
web-based material, by combining text with abstract 2D pointing visualizations.
The presentation is pre-planned but can be adapted and repositioned dynamically
according to the user’s actions. The user is mainly busy performing a task with the
mouse and keyboard.

All mentioned examples are supplementary to the performed task and produce
deictic references to particular objects that are part of the virtual world, together with
textual information, such as speech. They can be used for:

8 All three examples are explained in (Johnson, Rickel and Lester, 2000)

123
 • Demonstrating physical tasks interactively by animation
• Navigational guidance, problem solving guidance
• Nonverbal signals for conversational control and emotional feedback
• Conversational interface to the computer, conversational assistance

Comparative evaluation studies have shown that the agents contribute to the learning
effect at least through a higher form of engagement and motivation of a learner (Johnson,
Rickel and Lester, 2000). Another study for conversational assistance through avatars
showed a neutral effect concerning the comprehension of instructions, but a combined
higher degree of entertainment and positive acceptance (Krämer and Bente, 2004).
However, the results concerning the preferences for a certain visual appearance were
highly diverse, depending on personal biases. Furthermore, all mentioned examples
were tested in research environments, and a similar broad reality check as with the paper
clip example (see above: Swartz, 2003) is still outstanding. The resulting perceived
conceptual model might vary, depending on the concrete application, between a help
element and a performing actor.

Simulations
Simulation games have been discussed as ideal learning tools for the comprehension
of system dynamics, such as in SimCity9 (Bos, 2001), or Mobility10. Particularly in the
context of planning and construction, as well as for the control of machines, compu-
ter simulations are wide-spread. The basis of each simulation game is an abstract
model of the system that underlies the learning subject. The application of
simulations on the base of abstract models in the domain of human or social systems
is rather in its infancy. Two examples shall be emphasized: The best-selling computer
game “The SIMS”11 for play and entertainment, and the learning tool “Virtual
Leader”12. In “The SIMS”, each virtual doll has an extensive set of parameters
describing its properties (e.g. needs: hunger, comfort, social, etc.) and behaviour (e.g.
skills: cooking, creativity, charisma, etc.). The complex interface of many buttons and
sliders allows players to invest resources and make detailed day-to-day decisions. There
are no objectives for the player, it is rather a household or family simulation which

9 The first version of SimCity was produced by Maxis in 1989. The current official website is http://simcity.ea.com/
10 http://www.mobility-online.de
11 http://thesims2.ea.com/
12 Virtual Leader has been produced by Simulearn, http://www.simulearn.net/

124
can be shaped freely, and the player organizes the life of the agents so that they
achieve personal goals. There is no winning or loosing, and it has no educational
purpose.

In contrast to “The SIMS”, “Virtual Leader” is an early example of an application

of digital agents in a simulation game as a dedicated learning tool. It is a leadership
training simulation, giving the learner exercises in different scenarios of a business
meeting with virtual employees, where biases of employee ideas, financial performance
and customer satisfaction are at stake. The user interacts with the mouse and slider-
like graphical tools, rating persons as well as their upcoming ideas during the conver-
sation. Depending on the decisions of the learner, the system gives feedback on three
different aspects of leadership performance: power, tension, and ideas.

The function of the digital agents in the case of the simulations are:
• Demonstration of communication behaviour (verbal and nonverbal) of a
counterpart in a conversation, and representing the virtual population of
the scenario
• Improving the immersion and “suspension of disbelief ” in the scenario by
rendering the avatars as believable entities
• Apart from their rendered appearance, the most important function of the
digital agents is to “be” the abstract computational model of humans with
assigned variables and calculated states, in order to provide awareness and
cognition about a social system

The creator Clark Aldrich (2003) admits that a major problem is due to the fact that
these simulations are not at all realistic in every aspect. There is a general recognition
issue with simulations that is similar to any of the other problems of digital agents:
where the necessary abstraction in a simulation model of a city or a traffic system
does not lead to acceptance problems—when human concerns are at stake, things
are different. This does not only concern the appreciation by the target group. The
phenomenon is even debated in philosophy and sociology, asking if quantitative
methods from natural sciences and mathematics can be applied to cultural and social
issues. For successful functioning of the agents’ state machine generating the simulation,
quantifications are indeed a pre-requisite, resulting in the impact on (and expression
of ) emotions, personality and skills of the agents by mere calculation and computer
logic. There is the argument that “binding” causal explanations would additionally
have to take into account at least the temporal metric of the physical world (likewise,

125
extensively complex parameter sets) in order to do justice to real-world cognitive
processes.13

Interactive Digital Storytelling

Beyond the face-to-face interaction of a single pedagogical agent and a learner, there
have been recent research experiments in the context of “Interactive Digital Storytelling”,
building the foundation for most of the findings in this article. While simulations add
an underlying world model to the concept of digital agents, here, particular structures
from storytelling domains are added. In general, this is any representation of a plot,
optionally further divided into narrative sub-structures such as acts and scenes. Most
important is a dramatic character constellation, giving each character a clear goal in
the context of the told story. Stories and simulations show some resemblances at their
starting points, since character constellations can also be modelled as parameter sets.
Their differences are revealed during the unfolding of the plot. Put simply, the power
of a told story is to explain the cause for an effect and its special circumstances.
Therefore, a specific ending along with specific major turning points is most important,
especially when the end is surprising, since it is often the motivation to tell this
particular story at all. In contrast, pure simulations are open-ended—if at all, they
can be considered to be “what-if ”-stories. After having run the simulation, “cause-
and-effect”- stories can be told about it. However, some simulated endings might be
completely uninteresting to tell later. In Interactive Digital Storytelling, there is the
notion of “emergent narrative”, pointing out the open-endedness of such a plot that
only comes into being through interactions of users, and therefore cannot be
completely predefined by an author. Formulating general acceptance criteria and
design rules to maintain dramatic tension is still an open research issue, as the whole
field hasn’t yet produced mainstream products.

The following investigations have been performed in edutainment projects, which

employ digital conversations of a user/player with virtual characters on a stage
to convey information and to entertain. Within the game design community, a similar
integration of simulation and storytelling was used in the project “Façade” by Mateas
and Stern (2003), however, not for educational purposes. In Façade, two virtual
characters perform a conversation along a plot outline, in a situation of a couple
having a severe argument. The resulting role play assigns a role to the player as a

13 This argument on using simulations is discussed in a comprehensive way, from an epistemological point of view, in (Peschl and Scheutz, 2001).

126
friend stopping by their apartment and getting involved in the quarrel—either
to calm things down or to witness their breakup, by typing contributions to their
conversation on a keyboard.

With the edutainment projects, various approaches were explored to combine similar
plot-based interactive storytelling with character-based emergent conversations. They
differ in their user interfaces and in their bias on either more storytelling or more
simulation—resulting in the degree of user activity, versus pre-authored plot played
out by the agents. In all examples, several animated characters converse digitally with
each other and with one or more users, who either type text with the keyboard, or
apply choices, for example, with special hardware interfaces (compare Figure 1, right).

“Geist”14 (Kretschmer et. al., 2001) is a scenario in a real world setting explored with
a mobile Augmented Reality (AR) platform and GPS location tracking. User interacti-
on occurs by walking through the historic site of the Heidelberg castle, provided with
the “magic equipment”: a mobile computer. The goal is to track down virtual
ghosts, visible through magic binoculars, who live at real places and ask the tourist
for help (see Figure 1, left). Each of them tells a part of a story about the past. The
tracking system provides information about the location and the line of vision of
users. In effect, tourists cannot alter the story outcome; but they can change the order
of scenes slightly, depending on the route they choose. The main effect here of the
agents for learning is the motivating adventure of having a live experience with ghost
characters attached to a real historical site.

Figure 1: Applying choices through special interfaces. Left: Geist—in the Mobile-AR scenario, interaction occurs through
finding ghosts by walking in real locations. Right: art-E-fact—video-tracked pointing gestures select parts of a painting,
while special props define the meaning of the interaction.

14 The term “Geist” is German for “ghost”

127
In the project “art-E-fact”, the employment of digital agents has a similar motivation,
they entertain and provide people in a museum with initial interest in the topic of
art history (Spierling and Iurgel, 2003). Two characters perform a conversation, taking
different roles, starting for example an argument about the question if a Madonna-
and-Child topic of the Renaissance can be considered “kitsch” or not, or a detective
challenge. Each actor explains different facts in the context of the painting, for example
about comparative art history or particular painting techniques, and they can involve
the visitor into the discussion and ask him for arbitration in their quarrel. For specific
interaction with the painting in the museum, for example closer inspection or painting
exercises, toy-like props have been developed as interaction devices for pointing, such
as a magnifying glass or a brush. The learning topic is on the details of the painting
style, and the agents deliver a suspenseful story framing the content to generate
greater motivation.

The function of the digital agents in the edutainment scenarios are:

• Enhancing motivation and interest by introducing a compelling story
or fictional scenario
• Playing out a conversation, story or conflict among the agents, while
representing character roles as an initiation of a role play to join in later
• Prompting the learner for interaction with direct manipulation devices

Figure 2: Example screen with text interaction: Scenejo—Emerging small talk between several virtual
agents and one user.

128
Finally, the experimental platform “Scenejo” (Müller, Spierling and Weiß, 2005) offers
an almost free chat with several virtual agents and users (compare Figure 2). While
users type text on a keyboard to contribute to the conversation, the 3d-modelled talking
heads pronounce the answer text through a speech synthesizer, lip-synchronized with
their facial animation. A future task would be the integration of a speech recognition
system for spoken text input. The dialogues between the actors are pre-modelled by a
chat knowledge base, which primarily contains a huge amount of text patterns that a
user or an agent can potentially express, along with associated lines for the bots to
answer. The result is an emerging dialogue that is partially but not completely predefined
by an author, and which can have surprising turns, in contrast to the aforementioned
pedagogical or storytelling agents.

The two projects art-E-fact and Scenejo have similar goals while they approach a
middle ground of narrative presentation and the interactive conversation based on
user input. In Scenejo, the emphasis of activity is more on the user’s side. As a platform,
this enables the integration of simulation-like structures. As shown later in chapter 4,
Scenejo is used as a technical base for an interactive conversation, which combines the
general underlying idea of the “Virtual Leader” simulation with free verbal conversations.
It offers several levels of conditional states, processed by a drama manager, for example
changing abstract parameters such as emotions, planning of dialogue acts and specific
wording. For the description of the verbal knowledge base, the mark-up language
AIML of the Open Source project A.L.I.C.E. is used15. Figure 3 shows a sketch of
Scenejo’s design levels and an interface screenshot of the authoring view on the plot
graph through a transition network.

Figure 3: Scenejo (Drama manager & Authoring tool). Left: sketch of the parameters on several abstract levels, proces-
sed by the drama manager, Right: screenshot of a modelled plot graph

15 http://www.alicebot.org

129
3. Expectations for Learning
The second section briefly discussed selected existing examples and applications of
digital agents to illustrate the hypotheses made in the introduction. To a large extent,
alongside with most recent discussions about the topic of digital agents for learning, a
pedagogical agent obtains the role of a “help agent”, or the role of a guide, a teacher,
or simply provokes an expectation that implies a superior relationship towards a
human user. This expectation is similar to the yet unfulfilled dream that a chatbot
could counsel a human client in a meaningful way. Up to now this has only led to
partially satisfactory results. The main technical as well as conceptual challenge is not
to let the agent tell something, but to make the agent listen to a human user carefully.

The interactive scenarios presented show different ways of letting a user interact with
a digital agent. For example, a user can walk through a site and explore places where
she listens to pieces of a story presented by an actor—or the user might be actively
occupied typing sentences in a rapid interchange with the chat agent—or the interface
provides buttons and switches that let a user unhurriedly and deliberately control the
states of a complex parameter set in a simulation. While virtual actors playing out
content are widespread, there is little discussion about how the user’s interaction is
part of the learning process.

3.1 Three aspects of learning content

In the following, three learning aspects of the created media are discussed, which are
all present at the same time—however possibly with changing relevance (see Figure 4).
The aspects apply to any medium that can be used for learning, for example to paper
and pencil, as well as to digital agents, and have been discussed in (Spierling, 2005b).
Clark Aldrich (2003) noticed three “primary colours in the palette of educational
material” (linear, cyclic and open-ended), which as well can be mapped easily onto
the three aspects, in the mentioned sequence. Given the example of the paper-and-
pencil medium: It can be used for communication between humans, e.g. to make a
drawing to explain and to instruct; it can be used to train drawing and handwriting;
finally it can be used for communication of a human with herself—in the form of
extended memory used for drafting and sketching hypotheses.

130
 Figure 4: Three aspects of learning content: 1. Mediated communication of Human to Human, 2. Experienced
actions between Human and Media, 3. Medium for thinking, and communication of Human with Self

Communication
At first, opportunities for learning with digital agents exist within the “storytelling”
function: factual information can be conveyed through narration, using the agents as
a medium, as puppets on a stage. Since they have been equipped through an author
with the control about what they can tell, users are in the position of selecting what
they want to hear next. It conforms to mediated communication (compare Figure 4,
left) from a human author to a human user, essentially by more or less linear educa-
tional content. Even the initial concept of pedagogical agents emphasizes presenting
animated instructions.

Interface
The second aspect regards the (physical and virtual) interface and its resulting activity
patterns of cyclic repetitive actions, which have a training effect. For many of the
examples, pointing and choosing is the predominant standard of interaction—be it
through a mouse, or another spatial pointing device such as the physical prop in Figure
1. Pointing is the favoured action to be performed with a mouse—the most popular
interaction device. Indeed, with many actions being reduced to pointing, one can “only”
select from a limited choice. From the point of view of human-computer interaction,
this constraint has resulted in better usability, since possible human errors can be
reduced. Unfortunately, the point-and-click paradigm also dominates areas where
actually free exploration, including the possibility to make mistakes to learn from,
could contribute to learning. Instead, interaction with digital agents is rather often
still reduced to operating a virtual video player, playing out stories and sequences.

131
While there is no doubt raised about the need for computer usability, the “selection”
interaction style is an action that a learner is used to training over and over again.
Skilled computer users accomplish a lot while “clicking around” without much
reflection and thought. The need to phrase questions on one’s own is less convenient
if seen only as an action performed to use the computer. In the context of active
learning, however, the requirement to formulate one’s own utterances is desirable.
Repeated hand-eye-coordination such as aiming and shooting gradually improves
marksmanship, frequent talking improves pronunciation as well as verbal skills, as
does typing, which also improves typing speed.

Model
The third aspect of learning media gives people a medium to think and to test hypo-
theses, and its structure is open-ended—in fact providing an abstract environment
that provides a limited system of forces to explore. For example, proportional aspects
of architectural design can be drafted on a sheet of paper, whereas the third dimension,
the material properties, and gravity are ignored. While a sheet of paper provides little
predefined structure, a toy construction kit can serve as a more realistic model of
reality, and a today’s flight simulator is an almost perfect mapping of reality to a
computer model. A simulation model can consist of only a set of rules how to use a
paper and a pencil, but can also be implemented in a computer that calculates states
in a virtual world, based on the rules. Simulation models are tools, built to allow
testing and especially failure, which in reality would lead to undesirable events.
Digital agents are abstract simulation models of human matters, such as a visible body
with locomotion and expressions, as well as a mind, containing world knowledge,
personality traits, and emotions—including a rule set defining the behaviour towards
user interactions. One of the opportunities for learning is to enable failure without
severe consequences in interactions with the abstract simulated environment. Hence, a
simulation model with digital agents would be meaningful in topics about so-called
soft skills, embedded in social settings. Instead of working as a pedagogical agent, a
digital agent rather occupies the role of a learning companion or a sparring partner.

The few examples of giving the user a more active role towards agents are actually
either chatbots, or simulation games. In the current state of these applications, agents
show verbal and/or animated behaviour, but it is still hard to converse with them. For
example, “Virtual Leader” (Aldrich, 2003), doesn’t yet make the underlying model
really transparent during play. Agents are mainly used as actors to play out a script,
which, in the most interactive case, is adapted dynamically, depending on user choi-

132
ces. Reasons for the current infrequency of simulation agents are the degree of diffi-
culty with implementation, as well as accessibility and acceptance problems as men-
tioned in section 2.

Authoring aspects
Though authoring aspects shall not be in the focus of this article, it has to be mentioned
that the majority of existing examples described in section 2 have been created mostly
by computer scientists or at least by programmers. The current state of the art needs
further development in suitable methods and tools for authors, who are teachers or
experts in their field of study, independently making use of the agents. Then, considering
the aforementioned three aspects, authoring aspects are also threefold—possibly with
tasks distributed in a team.

Certain members of the author team can be considered to be mainly storytellers and
to have a primary focus on shaping and scripting the dialogues and visual behaviour
of the agents. The second possible focus of attention in authoring is the interface
design. This comprises the conception of the actual interface that the learner directly
holds and uses (typing, speaking, drawing, etc.), including visualized affordances, the
design of the pacing, the turn taking, the timing (such as time pressure), the design
of the immediate feedback, or possible constraints in the interface. Finally, the third
aspect for authors is game design, starting with the creation of a simulation model.
The model may contain actors, resources, and parameter-value pairs—states that can
change through the user’s interaction. Finally the rules used by the computational
state machine have to be framed—according to experience not only once, since part
of the game design is “tuning and testing”, until the behaviour is properly balanced.
This can only be done by people, who know the learning domain very well—being a
programmer alone doesn’t qualify for it.

Part of the reason why authoring of emergent media is still hard, is next to the lack of
tools, the sluggish advancement of interdisciplinary development of concepts. Due to
the nature of agents, there is a case of automation involved that leads to a “substitution
myth”, resulting in reluctance, for example of teachers, to engage with it. Christoffersen
and Woods (2002) have formulated a concept to overcome the substitution myth by
always including the human in the loop through the aspects of “observability” and
“directability”. Automated agents are seen as new team players, engaging in a joint
system with human team members. For authors, it is necessary to see (“observe”) what
they are doing, what their plans are, and how to intervene (“direct”).

133
3.2 Use of digital agents as educational engines
In “Engines for Education”, Schank and Cleary (1995) have formulated the thesis
that in American schools, the natural learning process is reversed, in the sense that
answers are provided before a student has asked questions. They suggested several
learning architectures, some of which are briefly discussed here to possibly benefit
from digital agents. He called them (among others) “Learning by Doing”, “Case-
based Teaching”, “Learning by Reflection” and “Learning by Exploring”.

Learning by Doing
Roger Schank argues that the main obstacle to implementing “Learning by Doing” is
economics, and that a computer simulation of an environment that allows users to
perform tasks in a simulated system of modelled forces can be a solution. He points
out that the simulation can never be as rich as the real world. However, in contrast to
reality, simulations can offer a degree of abstraction that particularly emphasizes
situations causing students to experience well-known traps and to work their way
through them. It is important that the student is supported in the “mapping” process
of the results of their simulation experience to reality. As shown with the example
of “Virtual Leader” as well as the “Interparolo” project mentioned in the next section,
digital agents can offer valuable functionality for the creation of abstract simulations.

Learning by Reflection
Another suggestion is the architecture of “Reflection”, which consists mainly of a
“sounding board” helping students to formalize their ideas, their thoughts as well as
questions, and reflect them with the computer as a virtual partner. Schank argues that
“what computers lack in empathy, they make up for patience”, and suggests a system
of questions to be asked in response to thoughts of the student. In its structure, these
questions are not dependent from expert knowledge in the learning domain concerned,
and therefore can be pre-phrased as generally useful questions uttered by an agent. A
digital agent as a textual chatbot, containing a question base in a general chatbot-
style, could possibly be useful without any claim to appear as a virtual consultant.
The written conversation as such would be an exercise in reflection—in the sense of
a diary communication.

Case-based Teaching
According to the argument, people learn from failure. According to Schank, experts
could tell a lot of cases how failing can occur—so-called “war stories”. Cases can be

134
constructed as stories together with tasks that the student has to work through,
including possible failure. Interactive Digital Storytelling, introducing emotional
agents acting as main characters with a problem to be solved by a student, offers
possibilities here. However, there is a warning attached to Schank’s vision, that case-
based teaching only works well if there are a lot of cases, not just a few. There has to
be a variety of possible ways to fail in the system, incorporating tailored advice, otherwise
it wouldn’t be case-based teaching. This warning is reminiscent of an inherent problem
in interactive storytelling, namely to create the required complexity of variations in a
story to be presented interactively.

Learning by Exploring
Another important idea, mentioning the idea of agents explicitly, is the Learning by
Exploring architecture. Schank mentioned 1995 the necessity of search agents finding
relevant information of many experts in story archives on the Internet, and making it
accessible in classrooms. Meanwhile this has become a reality. Search engines, Wikis,
Blogs and even recommendation systems based on demographics contribute to a great
pool of expert knowledge on the net. Rather than modelling a virtual expert to ask as
the one-stop agent, we are aware of a rather social and integrated conceptual model of
knowledge systems.

The conclusion is there are many ways of using agents in a role which is not the expert,
the educationalist or the pedagogue, making further developments worthwhile.

4. Project scenarios, chances, limitations and risks

The arguments for and against digital agents tackled before, point in a different direction
from seeing the pedagogical agent guiding the user through factual knowledge as the
only application possibility. In this section, two scenarios are sketched that assign roles
to digital agents that are different from the “guide” metaphor. The first is being worked
on in a current project at the FH Erfurt, University of Applied Sciences—the project
“Interparolo”. The second is a vision for future digital agents as personal learning
companions. In both scenarios, the metaphor of a “dollhouse” is superior compared
to a “Virtual Human”. In the conclusion of this article, the “Virtual Human” metaphor
is abandoned for the context of contemporary learning applications.

135
4.1 Interparolo: Conversations with digital characters
The project “Interparolo” is an interdisciplinary project concerned with the endeavour
to build e-learning content that offers interactive text chat dialogues for learning.
“Interparolo” is an Esperanto term for “conversation”. Accordingly, conversations
build the focus of exploration, not just the means of transferring the knowledge. The
learning topic of “moderation and mediation” is a course at the FH Erfurt, University
of Applied Sciences, within the faculty of transport and communications. Students
learn how to moderate a discussion between several stakeholders in the context of
urban planning. Naturally, these types of discussions bring together people of contrasting,
even antagonized positions and with varying skills in expression and discussion. As a
moderator of a meeting, one can run into situations that are difficult to master. These
situations include deadlocked positions that make discussion impossible, dealing with
difficult people, or with time pressure, just to name a few. The existing course material
includes a collection of instructions and work sheets presenting factual background
knowledge. However, the core skills that have to be achieved benefit at first from tacit
knowledge, including the competence to identify situations and the ability to react
accordingly. Hence, the traditional learning methods within the seminar concentrate
a lot on learning by doing, for example by employing live role playing games, which
allow simulations of cases and situations.

Within the project Interparolo, media content for this seminar topic is created and
delivered via e-learning. First course material includes text, as well as videos which
show live situations. Beyond these presentations, we explore the use of conversations
with digital agents based on chatbot technology in two ways: first a virtual character
was added to the digital learning environment platform “metacoon”16 being used
within the project. It works as a digital chat partner to answer glossary-like questions
and to point to existing text material, following a guide agent metaphor. In a further
development, several chatbots are implemented so that they can converse with each other
and the user. We explore the adequacy of this technology for serving as conversational
sparring partners in a digital role playing game.

16 http://www.metacoon.de

136
A digital role playing game simulates a live role playing game for learning up to a
certain point, while showing several significant differences. These include disadvantages,
since nothing is as suited to train for a real situation as the training with a live action
role playing game. However, there are also advantages using simulated environments
over real ones. Therefore, there is no intention to substitute existing methods by a
virtual game, rather to enhance the learning material by additional possibilities.

Design considerations
Technically, the virtual game is designed by connecting several chatbots in a conversa-
tional loop of turn taking. As a prototype, the platform Scenejo (compare Figure 2) is
used. The user is able to interrupt the conversation of the virtual characters, either by
typing text or by using control elements of the graphical user interface.
Figure 5 shows an abstract sketch of the conceived visible technical elements, which
allow the design of several sub-games. User utterances, which are input during runtime
of the game by typing on a computer keyboard, are compared with a prepared database
of pieces of dialogue covering possible verbal interactions. The matched patterns
influence a set of parameters of each character by applying rules, which in turn are
taken into account while the drama manager selects agent utterances from the scripted
dialogue base.

Figure 5: Design sketch of the partially visible elements of the “killer

phrase” game, such as the parameter states of the virtual participants

137
The first learning game being built in the context of the course “moderation and
mediation” tackles the topic of how to identify and react to so-called “killer-phrases”
within a discussion. Killer phrases are “creativity killers”, often used in a knee-jerk
manner and destroying new ideas before they are discussed with solid arguments. The
designed game assumes a scenario with two parties of planners and residents, arguing
about novel plans for an airport expansion. The scripted conversation between the
two parties, carried out across the table, contains killer phrases. The learner plays the
role of the moderator and has to manage the meeting. There are several stages of com-
plexity:
1. Modelling, without a game element: The agents model an ideal or a dysfunctional
interaction of the moderator, providing context information, with no interaction.
2. Situation-Awareness: The player/learner has to simply identify occurring killer
phrases, by hitting a “buzzer”-like button and reading more information about
the situation, earning points on hits and failures.
3. Coaching: The player/learner plays moderator and has to interrupt and phrase
verbal reactions to the virtual characters, in order to influence the meeting in a
positive way. In a first “learning system”-version, several game features can help
the learner, like wild cards. These are a time-out function to gain more time to
phrase an utterance, as well as a Help-function, calling another agent working
as a kind of coach or prompter. Game-like features include gathering points for
managing a situation.
4. Simulation: The player/learner does the same, except that there is a simulation
of a real meeting including the decision whether or not to interrupt, the time
constraints for actions, and the experience of the outcome not measured by
gained points, but by the reactions of the simulated meeting participants.
Optionally, the influenced parameters are explicitly made visible, or hidden in
order to let the player focus on the reactions only (such as in a real situation).
5. Reflection: Players/learners are able to replay their finished simulation, again
with the option to either visualize internal parameter states or not. This
can be embedded in a debriefing phase linked to the classroom course.

The aforementioned stages of possible virtual game designs with the created electronic
chatbot material relate to an often-used instructional design approach of “Cognitive
Apprenticeship” (CA). The step of the simulation, however, as a task of exploring the

138
territory, occurs at a rather early stage compared to CA methods17. It instead
fosters the constructivist exploring of unknown territory with the opportunity to
make mistakes in a risk-free environment. Stages 2 to 4 particularly benefit from
digital material including agents, while the digital material is tolerant, uncomplaining,
and repeated interaction with it is harmless. The digital material, however, cannot
stand alone and has to be embedded in the course curriculum within a blended
learning strategy.

Advantages and disadvantages of using the digital storytelling agents in this scenario
The project also investigates the suitability of using such a learning game for moderation
and mediation in the University environment. Several strengths and opportunities, as
well as weaknesses and risks, were identified in the early design phase of the game.

The advantages of the application of digital chatbot agents to the given topic are the
following:
• Doll playing in general allows users to test boundaries more than in real live
role play which occurs in a social setting. The “magic circle” with digital agents
and thus the safe space of “just playing” is more clearly defined than with live
participants in a role playing game.
• This reflects the attitude of people dealing with chatbot conversations:
ignoring the usual rules of polite behaviour and trying to test the boundaries.
In fact, this phenomenon is debated within the project. On the one hand, this
can be a chance to experiment with situations like this, and experience the
outcome directly. The counter-argument is the concern that students don’t
take it seriously and prefer to make fun of the annoying situations. Given the
background that the games would be embedded in a classroom course, this
objection is given minor importance during the design phase.
• The digital environment allows for adjustment of timing and pacing, for example
to stop and to play single scenes for more emphasis, to increase the time pressure
intentionally, or to replay achievements and played games for reflection and
debriefing.

17 According to Collins, Brown and Holum (1991), the instructional design method of “Cognitive Apprenticeship” sequences the
learning content by increasing complexity through the steps 1) Modeling, 2) Coaching, 3) Scaffolding/Fading, 4) Articulation,
5) Reflection, 6) Exploration.

139
 • Through the text-only interface, phrasing can be judged explicitly, since
parameters from the real world are missing that could have a disguising effect
for recognition. Dealing with verbal language is the subject of interest being
taught, so the choice of words plays an important role. The repetitive learning
aspect addresses typing quickly and phrasing carefully. While the first one is only a
side training effect and does clearly not support the real situation, the second
one is focused upon.
• For a live action gaming simulation, a classroom setting with participants and
a proper preparation is needed. Classroom realities dictate that this is not
always convenient, especially when repetition is desired. Computer simulations
can be beneficial as additional opportunities or for preparation of the live
game in the preliminary stages. They are also useful for experiencing extreme
situations which are hard to create in real life and repetition of a given situation
is possible.

The identified disadvantages include:

• The argument against simulations in general also applies here: A simulation
never represents the real situation properly with all its complexity. A mapping
of the experiences in the simulated world to the real situation has to be
achieved (Schank and Cleary, 1995). The computer simulation is even more
abstract than the live role playing game, so it may also confuse the issue more
than contributing to a successful transfer.
• Nonverbal cues such as body language are difficult to implement or non
existent as in our first prototype. Verbal factors are presented in isolation,
while life is more complex. Several emotional levels are left out, and the
stress a learner experiences in the live situation is not part of the system.
Voices of the characters are currently only represented electronically (by
text-to-speech, which facilitates rapid prototyping during the design phase),
or not at all (with text in speech bubbles).
• The work involved in the creation of the necessary dialogues for the chatbot
knowledge base, before a game can successfully be played, is significant.
The technology base used to run the chatbot dialogues still needs technical
development.
• There is an underlying risk, that the typical flaws known from chatbot inter-
actions also apply to the learning game, and that they may hamper the outcome
of the game. The only way to work against it is an increased effort of scripting,
in order to be prepared for many possible situations.

140
 • Risks can also be identified in the lack of experience of teachers as simulation
designers. The quality of the simulation model is crucial to the success of the
learning experience. This model can only be built in a cooperative effort
between experts of the domain and experts in gaming simulation. In this sense,
the greatest risk is that the designed model can be plain wrong, which results in
learning mistakes.

The goal of the project Interparolo, next to creating material for evaluation, is the
building of examples of reference that can be used to expand the general idea to further
content scenarios. During the design phase, one obvious learning effect has been
the increased reflection on the dependencies within the system dynamics, particularly
experienced by the game designers and creators of the modelled dialogues.

4.2 Looking to the future: Digital learning companions

Almost all conversations with digital characters looked at so far, follow a pattern that
assumes no long-term relationship between the agents and the users. Moreover, the
last conclusion that a significant learning effect has been experienced by the creators
of the chatbot scenario and dialogues, leads to a radical constructivist thought: the
people who make the digital agents smart, that is in our case the group of chatbot
authors, go through a learning procedure, or at least have to apply knowledge in a
creative and explicit way. The resulting chatbot reflects the applied “knowledge”
directly, while reacting more or less appropriately to the questions asked or the utterances
confronted with. These considerations result in a different concept for using digital
conversational agents in learning environments: not as intelligent teachers or instructors,
but as initially clueless companions with empty knowledge-containers, who need the
learner to make them smart and to fill them with responses.

The learners’ task, during an authoring mode, would be to create pieces of dialogue
which gradually, during the learning process, fill the memory of the agent—teaching
the companion, instead of being taught by it. In a playback mode, the agents’ role
would be to reflect what has been “taught”. In a sense, one could build a digital “Alter
Ego” that mirrors verbal knowledge that has been explicitly articulated by a learner, in
a playful way. From an instructional design perspective, also in accordance with the
Cognitive Apprenticeship method, a further level of complexity could be added:

141
the “articulation” stage. The resulting genre of a simulation game would be to put
the learner into the position of a puppet master, being responsible for the puppet’s
appearance during playback.

While only using chatbot-like pattern matching, the dimensions of applied knowledge
are naturally as shallow as the pattern matching algorithm, focusing on explicit
utterances. The full scenario, instead, requires a more sophisticated architecture.
Depending on a context (situations, relationships, dialogue history, etc.), concrete
language would have to be made adaptable to cases (compare illustration in Figure 6).
This would also require the articulation of contextual rules, as a significant part of the
knowledge that could be applied and tested during a playback mode.

The digital learning companion is a combined metaphor of a digital doll with an

electronic diary function, a medium used by a single person for self-communication
and reflection (compare section 3). Even more than the games within the “Interparolo”
project, it fosters the active construction of knowledge in a creative way.

Figure 6: Scenario of a digital agent as an “Alter Ego” for construction and reflection of verbal knowledge

5. Conclusion
All discussions and examples presented here have focused on the conception of various
forms of digital games, including storytelling and simulations, based on digital agent

142
technology. They have roughly illustrated the huge potential offered by these new
technological endeavours. However, there are also limitations and risks involved that
have led to initial acceptance problems.

The main argument made here is, that by only working on better technology, in the
sense that digital agents get more sophisticated, intelligent and finally human-like, the
acceptance problems will not completely be solved. The hypothesis is, that if a
digital agents adopts a superior role to a human, there is always the probability that
the basic trust that is needed to take advice will not be established towards a synthetic
character, even if there is no doubt about the evidence of social relationships with
technology. The thesis is supported by experiential phenomena from storytelling
and robotics, such as the notion of the “uncanny valley”.

On the other hand, when embedded in a game or abstract simulation taking place
with clearly communicated limitations and goals, even doll-like visualizations,
behavioural abstractions or repetitions don’t produce a problem. Within the “magic
circle” of a game, the rules are set by the game and not by expectations of reality.
“Suspension of disbelief ” will work for abstraction, and behaviour just has to be
consistent with its own rules to be believable. Therefore, it is necessary that clear
conceptual models forming realistic expectations are established by a suitable game
design. Even the current state-of-the-art agent technology provides means to create
simple games that can be applied to learning situations, as demonstrated by the
“Interparolo” project.

The more technological “intelligence” is realized in applied agent architecture, the

more challenging is the creation of suitable content. This is especially true when we
assume that creators of the learning material don’t have to have a diploma in Artificial
Intelligence or even programming skills. Digital agents have the potential to be a
welcome help for teachers, given that responsible authors will be able to define the
agents’ behaviour and tasks easily. Instead, doubts remain that the conceptual model
of an intelligent pedagogical agent, represented on a screen by an almost realistic
“Virtual Human”, will be appreciated, when it looks like a substitution of a real teacher,
or anything that delivers better results when done “in reality”. Finally, the claimed key
to success is that active construction by learners is involved as a goal to achieve by the
game design, and that there are humans in the loop who take on responsibility and
are enabled to do so since they are in the driving seat.

143
6. Author

Ulrike Spierling has been a professor of Media Design at the University of Applied
Sciences (FH) in Erfurt, Germany, since 2002. Previously, she was the head of the
Research Department of Digital Storytelling at the Computer Graphics Center
(ZGDV e.V.) in Darmstadt, Germany, where she developed a research agenda for
interdisciplinary approaches to the design of technology for interactive storytelling
and conversational interfaces. Her current research includes the development of
authoring concepts, models and tools for non-programmers, employing conversational
agents for interactive digital storytelling and the design of knowledge media.

7. References
Aldrich, C. (2003). Simulations and the Future of Learning. Pfeiffer & Company
ISBN: 0-7879-6962-1

Bickmore, T. W. & Picard, R. W. (2004). Towards Caring Machines. In: CHI 2004
Extended Abstracts on Human factors in computing systems (pp. 1489 – 1492).
ACM Press, New York.

Bos, N. (2001). What Do Game Designers Know About Scaffolding? Borrowing

SimCity Design Principles For Education. Playspace – An Examination Of
Learning In Multicultural, Digital Play Environments. URL. Retrieved June 15,
2005 from http://playspace.concord.org/papers.html.

Cassell, J., Sullivan, J., Prevost, S., Churchill, E. (2000). Embodied Conversational
Agents, MIT Press, Cambridge MA, USA

Crawford, C. (2004). Chris Crawford on Interactive Storytelling. New Riders

Publishing.

144
Christoffersen, K. & Woods, D. D. (2002). How to Make Automated Systems Team
Players. In: E. Salas (Ed.), Advances in Human Performance and Cognitive
Engineering Research, Volume 2 (pp. 1–12). JAI Press, Elsevier Science.

Collins, A., Brown, J. S., & Holum, A. (1991). Cognitive Apprenticeship: Making
Thinking Visible. American Educator, Winter Issue 1991 (pp. 6 – 11, 38 – 46).
URL. Retrieved May 04, 2006 from http://www.21learn.org/arch/articles/
brown_seely.html

Don, A., Brennan, S. E., Laurel, B., Shneiderman, B. (1992). Anthropomorphism:

from ELIZA to Terminator 2. In: Striking a Balance: Proceedings of the 1992
ACM/SIGCHI Conference on Human Factors in Computing Systems (pp. 67-
70). ACM Press, New York.

Franklin, S. & Graesser, A. (1996). Is it an Agent, or just a Program?: A Taxonomy

for Autonomous Agents. In: Proceedings of the Third International Workshop on
Agent Theories, Architectures and Languages, Springer-Verlag, London. URL.
Retrieved Mar 01, 2006 from http://www.msci.memphis.edu/~franklin/
AgentProg.html.

Glassner, A. (2004). Interactive Storytelling: Techniques for 21st Century Fiction. AK

Peters, Ltd.

Johnson, W.L., Rickel, J.W., Lester, J.C. (2000). Animated Pedagogical Agents: Face-
to-Face Interaction in Interactive Learning Environments. International Journal of
Artificial Intelligence in Education 11, 47-78.

Krämer, N.C. & Bente, G. (in press). Assistenz durch Virtuelle Helfer. In: H.
Wandke & E. Wetzenstein (Eds.), Der Umgang mit Technik im Alltag: Assistenz
für Benutzer. Springer-VDI, Berlin.

Kretschmer, U., Coors, V., Spierling, U., Grasbon, D., Schneider, K., Rojas, I.,
Malaka, R. (2001). Meeting the Spirit of History. In: Proceedings of the
International Symposium on Virtual Reality, Archaeology and Cultural Heritage,
VAST 2001 (pp. 161-172). Glyfada, Greece.

Mateas, M. & Stern, A. (2003). Integrating Plot, Character and Natural Language
Processing in the Interactive Drama Façade. In Technologies for Interactive Digital
Storytelling and Entertainment, Conference Proceedings, Darmstadt.

145
Müller, W., Spierling, U. & Weiß, S. (2005). Synchronizing Natural Language
Conversation between Users and Multiple Agents in Interactive Storytelling
Applications. In: TESI 2005 Conference Proceedings CD-ROM, Training,
Education & Simulation International, Maastricht.

Nass, C., & Sundar, S.S. (1994). Is human-computer interaction social or parasocial?
Stanford University. Submitted to Human Communication Research, 1994. URL.
Retrieved March 04, 2006 from http://www.stanford.edu/group/commdept/olds-
tuff/srct_pages/Social-Parasocial.html.

Norman, D. A. (1988). The Design of Everyday Things. Basic Books, New York.

Peschl, M. & Scheutz, M. (2001). Some Thoughts on Computation and Simulation

in Cognitive Science. Philosophie, Wissenschaft, Wirtschaft (pp. 534 – 540).
öbt&hbt Verlagsgesellschaft, Wien. URL. Retrieved May 04, 2006 from
http://www.nd.edu/~mscheutz/publications/scheutzpeschl00linz.pdf.

Preece, J., Rogers, Y. & Sharp, H. (2002). Interaction Design: Beyond Human-
Computer Interaction. John Wiley & Sons, New York.

Schank, R. C. & Cleary, C. (1995). Engines for Education. Lawrence Erlbaum

Associates, Hillsdale.

Searle, J. (1980). Minds, Brains and Programs. The Behavioral and Brain Sciences,
Vol. 3 (pp. 417-424). Cambridge University Press.

Spierling, U. & Iurgel, I. (2003). “Just Talking About Art”—Creating Virtual

Storytelling Experiences in Mixed Reality. In Virtual Storytelling, Proceedings
ICVS 2003, Springer Berlin-Heidelberg.

Spierling, U. (2005a). Interactive Digital Storytelling: Towards a Hybrid Conceptual

Approach. In: Conference Proceedings, Selected Papers of DIGRA 2005, Digital
Games Research Association’s 2nd International Conference “Changing Views:
Worlds in Play”, Vancouver.

146
Spierling, U. (2005b). Beyond Virtual Tutors—Semi-Autonomous Characters as
Learning Companions. In: SIGGRAPH 2005 Educators Program, ACM
International Conference on Computer Graphics and Interactive Techniques, Full
Conference DVD-ROM, Conference Select CD-ROM, Los Angeles.

Swartz, L. (2003). Why People Hate the Paperclip: Labels, Appearance, Behavior, and
Social Responses to User Interface Agents. Symbolic Systems Program, Stanford
University. URL. Retrieved Mar 04, 2006 from http://xenon.stanford.edu/
~lswartz/paperclip/.

Turing, A. M. (1950). Computing Machinery And Intelligence. Mind, Volume 59,

433-460. URL. Retrieved June 15, 2005 from http://www.loebner.net/Prizef/
TuringArticle.html.

Weizenbaum, J. (1966). ELIZA—a computer program for the study of natural

language communication between man and machine. Communications of the
ACM, Vol. 9/1.

147
148
Interactive Drama and Learning
Experiences
Anja Hoffmann, Ido Iurgel, Felicitas Becker

Keywords
Learning environments, emotional and social context of learning, virtual characters,
narrative models, soft skills, story generators

1. Introduction
Even though computer-based training is becoming highly popular, e-learning applica-
tions are still dealing with high failure and dropout rates. This may be because
e-learning applications are designed in a way that only address cognitive issues. For
learning applications to be most effective, however, both cognition and emotion have
to be addressed. The idea is to enable enhanced learning experiences that are more
motivating and compelling through the integration of storytelling and game play.
These learning environments will support both the acquisition of knowledge and
the training of key qualifications and soft skills. In the following examples, narrative
learning applications are introduced and the StoryGenerator which enables these
enhanced learning applications is presented.

2. Point of Departure
Today’s learners—especially the younger generation—are “digital natives” who are
used to using digital applications for communicating, sharing experiences, creating
personal artefacts, searching for information, exchanging opinions, and gaming—
anytime, online and collaboratively. In contrast, most common learning resources
disregard these factors which can make any application convincing and fascinating to
a young audience. At the 2005 Learntec Conference, Marc Prensky gave an impression
of the prerequisites of today’s learners: They don’t want to study Rome—they just

149
build it! (Prensky, 2005) His point was that engagement must come before content,
and that detailed and boring learning applications are not sufficient.

Using narration can be an answer to this problem. Stories are the first and most
important tools used to share not only knowledge, but also feelings, emotions, cultural
and human context, and diversity. Children learn from the stories their parents tell.
Beyond this, stories are the structures which underlie every learning process. While
stories will address the emotional side of the learning process, game play can be used
to ensure that the learning application is fun and easy to follow. As a future trend we
can be sure that games will become part of our daily life. They will assume a place
comparable to today’s television, and will detach from the computer and its single
monitor as their place of origin.

Each of these elements—learning, storytelling and game play, overlap in certain areas.
Storytelling and gaming come together in 2D or 3D virtual environments—especially
in adventure games but also in other game genres. Interestingly, the story part of these
games is becoming increasingly important which is shown in the merging of films and
computer games: On the one hand, films are based on computer games (Tomb Raider,
Resident Evil, Final Fantasy etc.) and on the other, computer games are often released
along with a corresponding film (Star Wars, Indiana Jones).

The combination of learning and storytelling can already be found in very different
areas. Many books—such as “Sophie’s World“ by Jostein Gaarder or “The Name of
the Rose“ by Umberto Eco—combine learning material with a story. Museums and
other cultural institutions often try to wrap their exhibits into stories to appeal to
younger visitors—as do television shows like Sesame Street.
The question is how to combine these aspects of learning, storytelling and gaming?

Figure 1: Overlapping areas of Learning,

Storytelling and Gaming

150
3. Digital Storytelling Projects
The challenge of combining learning, storytelling and gaming is approached by digital
storytelling projects. The following gives an overview of learning related projects of
the Digital Storytelling department at ZGDV in Darmstadt, Germany.

3.1 art-E-fact
art-E-fact1 is an EU-funded project which aims at providing culturally enriching and
entertaining experiences to museum visitors through interactive installations related to
the artworks.

The art-E-fact installations are composed of stories that engage visitors in dialogues
with virtual characters. The visitor interacts with the characters through keyboard
input, by gesturing at the screen, or using dedicated interaction tools. Interactive
devices, such as virtual and physical tools, allow the visitor to inspect the artwork
more closely using unconventional means. By choosing a virtual x-ray-tool, for example,
the visitor can scan the painting to discover hidden layers below the image surface.
The visitor can also choose a torch or a magnifying glass. This way, visitors of art
galleries acquire a better understanding of the artwork.

Figure 2: Pointing Gestures in art-E-fact: Magnifying Glass / Torch

1 IST-2001 37924, cf. www.art-E-fact.org

151
The knowledge of the multifaceted and deeply personal aspects of art is transferred to
the visitor through conversations with well informed virtual characters. The discuss-
ions between these characters are embedded in a story which unfolds against the per-
sonal background of each character. This is a model for a new kind of interactive sto-
rytelling which emphasizes a narrative, interactive presentation of a theme through a
discussion group (Iurgel, 2004). The gaming aspect lies in the playful interaction with
the artwork. Further, at the end of the story the visitor must resolve a conflict where
the fate of the hero lies in the visitor’s hands. It is as much a question of winning or
losing the game as it is a moral challenge.

3.2 Servingo
Servingo (2006) is funded by the Federal Ministry of Economics and Technology
(BMWi) in Germany. The project will ensure an efficient and innovative flow of
personal communication at the FIFA World Championships 2006. The main goal
of the project is to establish an interactive platform as a logistic tool for all sorts of
information about the championships.

One of the services provided by the platform is a personalized diary in which the user
can store his impressions in the form of photos, videos, or audio files. An interface
from this diary leads to the StoryGenerator, which combines the user’s material with
metadata like rankings, scorings, or categories to generate a short movie clip the user
can send to his friends vie e-mail. In order to have his story generated, the user choos-
es a genre or story model which will then be used to structure his material. If, for
example, the user wants to create a thriller with his material, he will choose the
“thriller” genre. If, instead, he wants a show host and a guest to present his impressi-
ons, he can choose a “show” model. He can then decide which of the meta data
should be used for the story. Once he has made these simple choices, the movie
will be automatically generated, adding system content such as virtual characters
and pre-implemented audio to the user’s data.

It is expected that story generators equivalent to the StoryGenerator used in Servingo

will become more and more popular, and perhaps evolve into a helpful tool for teachers
to create stories with learning content for their students. Without the need to know
anything about story structure or about how best to integrate knowledge into a story,
the teacher can simply choose from a wide range of personalized data and then have a
story generated automatically. Another possibility is to let students use the

152
StoryGenerator themselves, simply giving them a certain pool of information from
which to choose. Not only will they create an enjoyable presentation of the
content, they will also have worked with the information beforehand—having read
each piece of information to be able to choose from the data etc.

Figure 3: StoryGenerator on Servingo: Choosing personalized data from the diary

3.3 Virtual Human

The project Virtual Human is funded by the Ministry of Education and Science
(BMBF) in Germany (Virtual Human 2006). Virtual humans are human-like inter-
action agents who are employed as personal dialogue partners in certain application
contexts. A virtual human can engage in dialogues both with other virtual humans
and with human users, showing credible emotional behaviour and employing non-
verbal communication forms. A future research challenge of the project consists of
the creation of convincing virtual actors. In one of the application scenarios of Virtual
Human, the user becomes a football coach and can decide the success of the national
football team. Improvisational abilities of virtual characters will enable them to tell
jokes and to tease the participants, demonstrating the interaction of novelists and of
an improved authoring process.

The learning aspects concern information about different teams, individual players,
and which strategy will be the most efficient for winning the match. A so-called

153
“Trainer-Test” will show the effectiveness of the team line-up chosen by the user,
including comments of two virtual experts on football.

Figure 4: Virtual Human scenario: two virtual experts discussing the user’s team line-up
(Drawings by rmh new media gmbh, Cologne)

The scenario could be easily adapted to learning. First of all, the social abilities make
the virtual characters more life-like and—even with their faults and moods, or maybe
especially because of them—they become likeable. The user can actually bond with
them and is emotionally satisfied. We expect applications to become more absorbing
and rewarding this way, becoming more motivating and fun at the same time, all the
while making learning easier. The content could of course be altered in any way.
Instead of being football experts, the virtual characters could be high profile doctors
or palaeontologists evaluating the choices the user is making. In a medical context,
the user could be asked to write prescriptions to patients while listening to comments
from the professional doctors. Similar to a “Trainer-Test”, a “Treatment-Test” could
show how the patient’s illness will proceed in relation to the user’s decisions.

3.4 Ask & Answer

Ask & Answer is an educational narrative game that demonstrates how virtual charac-
ters and narration can be employed to enhance efficiency, fun and motivation in lear-
ning (Iurgel/Ziegler 2005). It adopts the formats of familiar TV-shows for kids and
families, where two teams compete, and scores are given for correct answers. In Ask &
Answer two teams of 2-8 participants play over the network. A team scores when it
knows the right answer to a question asked by the system. Each participant looks at
his own screen and has its own speech enabled avatar. The demonstration’s current
domains of learning are history and geography.

154
The single autonomous virtual character has a decisive role in the game. It knows
the right answer to every question of the game, and can give hints. But it will only
help if the player assists it with its story related problems. The virtual character,
called Fritz, can only be heard by the team it is near to. If the team does not succeed
in endearing themselves, it might refrain from delivering the answer, or even worse,
it might leave the neglecting team and move to the other team. But even if Fritz is
satisfied with a group, it might first want to talk about some favourite topic (typically
additional material related to the question), before giving away any solution. Thus,
the players have to establish social bonds to Fritz, and have to dwell on further aspects
of the theme that they are learning, in order get the hints and win the game.

Figure 5: Two teams confront each other, Fritz is in-between.

The answer is usually entered with the keyboard. Any participant may post an answer,
but he risks negative scores for his team if the answer is wrong. Therefore, in order to
coordinate the answers, the participants can communicate through their avatars.
Avatars that are within a specific area, the territory of the group, can only be heard by
members of the group that are located within this area. The players are allowed to use
other material to find out the answers, e.g. textbooks.

Thus, this game involves social elements in a complex way:

• members of a team must cooperate to acquire knowledge and coordinate
answers, in order to defeat the other team;
• a team must develop a common strategy to endear themselves to Fritz,
for example by nominating a “Fritz-Friend” to exploit this source of
knowledge, and
• the relationship to Fritz is a social relationship that pays.

155
Coping with the virtual character’s peculiarities and establishing social bonds with it
thus becomes part of the challenge of the game. Many edutainment and entertainment
applications could be conceived from this starting point: The focus need not be in
human-like behaviour of virtual characters; rather, the focus can be on the human-
like situation where it is profitable to adapt to the social expectations and emotional
peculiarities of the (virtual) other.

4. Future Prospects
What will the future bring us? We expect to see mixed groups of virtual and real
humans interacting and learning together, in several directions, soon. Group formation
of hybrid groups, in the sense of a feeling of belonging, is an important research
issue. A hybrid group will go through exciting experiences, structured according to
narrative models, and face challenges and learn together. The user will help virtual
and real characters in need, compete, get assistance, create social bonds, and thus
will have an emotional and social context for his learning experiences. These projects
describe a step towards those novel learning experiences.

Due to the social embedding of learning, learning of soft skills, of social behaviour,
emotional understanding, of empathy, and even self-growing and guided personal
development are natural and important aspects of the research direction described
above.

It is also expected that in the future stories will be automatically generated. As

described in context with Servingo, story generators will give teachers, students, and
users of any other field the chance to generate stories from their own personalized
data or for that matter, any data at all.

Beyond story generators research and development focuses on authoring tools,

which enable professionals as well as laymen to write interactive stories with self-made
characters, individual interaction modalities, story-lines and content. By encouraging
and assisting users to create content in the form of Interactive Storytelling, narrative
learning applications will become more accessible and universalized.

156
5. Authors

Anja Hoffmann received her diploma in Media System Design from the University of
Applied Sciences in Darmstadt. She has been working in the area of Interactive Story-
telling since 1998 and was a research associate at ZGDV Darmstadt. Now she is part
of SAP’s Design Services Team, in the Office of the CEO, and works on innovative
concepts for user experience design.

Ido A. Iurgel is a scientist at ZGDV Darmstadt working on concepts for interactive

drama and edutainment, employing virtual characters and novel authoring tools. He
has a Dipl.-Inform. in Computer Science from the Technical University of Darmstadt,
and a M.A. of Philosophy, Social Psychology and Linguistics from the University of
Bochum.

Felicitas Becker is a full time scientist at the ZGDV since 2005, and has a Masters
degree in film studies, communication and psychology from the Johannes Gutenberg-
University of Mainz. She has worked as a camera operator for various TV stations and
as an editor for Radio and TV. Before becoming a full time scientist she worked at the
ZGDV as a freelancer and as a student assistant. Her special interests are story models
and the dramatization of behavior and emotions of film characters and virtual characters.

157
5. References
art-E-fact Project (2006), project website. Retrieved March 11, 2006 from
www.art-E-fact.org.

Iurgel, I. (2004). From Another Point of View: art-E-fact. In: Göbel, S.; Spierling, U.;
Hoffmann, A.; Iurgel, I.; Schneider, O.; Dechau, J.; Feix, A. (Eds.), Proceedings of
2nd International Conference onTechnologies for Interactive Digital Storytelling
and Entertainment, TIDSE 2004 (pp. 26-35). Darmstadt: Lecture Notes in
Computer Science Vol. 3105.

Iurgel, I., Ziegler, M. (2005). Ask and Answer: An Educational Game Where It Pays
to Endear Your Capiricious Virtual Companion. In: Panayiotopoulos, Themis
(Ed.) u.a.: Intelligent Virtual Agents. Berlin; Heidelberg (Lecture Notes in
Artificial Intelligence (LNAI) 3661) : Springer, 15-24.

Prensky, M. (2005). Digital Game Based Learning. Talk at Learntec Conference,

Karlsruhe 2005.

Servingo Project (2006), project website. Retrieved March 11, 2006 from
www.servingo.org.

Virtual Human Project (2006), project website. Retrieved March 11, 2006 from
www.virtual-human.org.

158
159
160
YoungNet—a Virtual Learning Community
Platform for Youngsters
Fabian Kempf and Karin Hamann

Keywords
Edutainment, Avatars, 3D Multi-User Environment, Synchronous Communication,
Voice over IP, Learning, Playing, School

Abstract
e-Learning has become a major buzz word said to make education and training
more efficient in business. It is being introduced to universities on a larger scale.
However, the situation at schools is a bit different: two key distinctive aspects are
motivation and communication.

YoungNet is a two-year European funded project. It aims to help schools to

introduce e-Learning during and in addition to their lessons by providing
technology as well as content and guidelines. Game-based learning is a central
element in the learning community.

e-Learning communities highlight the fact that learning may also be a joint activity.
YoungNet offers an integrated virtual learning community which is accessible to
various schools and individuals from different countries via the internet. They
study, play, meet, exchange ideas about the nature, way of life, and culture in their
respective countries. At a technical level, it offers both synchronous and asynchronous
communication based on leading edge ICT like Virtual Reality, 3D representations,
audio communications, shared applications and provides a fascinating and
“edutaining” multi-user environment. The use of this platform is encouraged by
intrinsic motivation and thereby overcoming the division between learning
(school) and leisure.

161
1. Why YoungNet
Networked learning and virtual education has been already accepted not only as a
new but also as a more efficient way of education and training in business especially
at corporate universities. Academic universities have started to adopt e-Learning on a
larger scale by centrally organizing the introduction of learning management systems.
Pilot projects have readily been absorbed into mainstream use. At schools the situation
still looks a little bit different. Money and resources are often restricted, therefore
highly motivated teachers are a prerequisite.

However especially for young people such as pupils it is important to develop learning
skills and cope with new learning environments and innovative learning materials as
early as possible. Through them they can not only learn factual knowledge but also
gain the ability to communicate, co-operate, learn from one another, play educational
games together and acquire knowledge of foreign countries and cultures which is
more important than ever in our globalized society. Furthermore positive experiences
in youth will facilitate the development of a positive attitude towards international-
ization and help students to deal with foreign cultures and to understand the need
to be able to express themselves in a foreign language. In fact all these are key factors
for future competitiveness.

These needs are addressed by the presented game-based virtual learning community
for young people (8 – 14 years old) featuring a virtual meeting place where joint
educational projects can be undertaken, educational games can be played, and informal
information on day to day life can be exchanged across borders.

2. Objectives
In order to help achieve learning objectives, the YoungNet project is developing a
virtual e-Learning community and playground for young people where they can
meet and interact with other young people from different countries.

The major features of YoungNet are as follows:

Feature 1: A fascinating and edutaining multi-user learning environment with clearly

defined learning objectives.

162
 The multimedia game market is the driving force in the technological develop-
ment. Good multimedia games—especially multi-user co-operative games—are
very attractive to young people. The market potential undoubtedly is enormous.
However, the objectives underlying these games are not always educationally
sound (e.g. combat games). YoungNet combines this attractiveness with well-
defined educational goals, e.g. the application of higher level cognitive goals,
fostering problem-solving, creativity and specifically team work aspects.

Feature 2: Enable learners and teachers to communicate (including audio

communication) and co-operate across national borders and cultures.

In an increasingly globalized world, the ability to communicate and co-operate

with people from other countries is becoming a key skill for both individuals and
societies as a whole. The traditional educational system can not sufficiently teach
these communication skills. A stay in a foreign country is usually limited to a short
period of time This ability can only be developed and strengthened by a continuous
opportunity to communicate and co-operate with peers in other countries on a
more regular basis than is possible in traditional classroom lessons.

Feature 3: Overcome the boundaries between school time and leisure time by
intrinsic motivation created through attractive features of the YoungNet virtual
meeting place

Most of the existing internet communities merely offer chat rooms and bulletin
boards. A successful virtual community needs to take a comprehensive approach
by incorporating the virtual environment, educational games (challenges), commu-
nication features and the navigational controls as integral components of the same
system that depicts the real world in a virtual world (Soliman 2001).

Feature 4: Continuous evaluation by the schools involved

Teachers of 50 schools in various European countries were involved in different

phases of the project. They are regarded as the core specialists for evaluating the
YoungNet functionalities, tools, services and contents from the perspective of
pedagogical specialists and experts in the conditions in their schools and their
students.

163
The following sections describe the architecture of the YoungNet community which
is composed of two major components the YoungNet system and the YoungNet con-
tents and which fulfils the objectives mentioned above.

3. The YoungNet System

The YoungNet system is an integrated Internet/ Intranet solution for communication
and administration tailored to the needs of teachers and their students, which combines
the following technical components:
• An application of the Hyperwave eLearning Suite for communication and
administration
• A 3D-Environment for synchronous communication and gaming.
• An audio component for authentic Internet communication

A Learning Management System (LMS) is the basis for the community platform.
The LMS is constructed as a virtual training centre consisting of
• several rooms, for different purposes (such as a course room, a café, an
administration centre, a private study room which works as a personalized
learning portal etc.)
• tools of various types (asynchronous and synchronous communication
and collaboration, information retrieval, authoring and structuring,
administration and reporting tools) which complement the functionality
of the rooms,
• different types and roles of users (such as students/trainees, authors, tutors
and trainers, administrators) and
• strong knowledge and content management functionality (such as fine
grained access control, automatic link management, distributed web based
editing with version control and release management etc.).

The virtual 3D Environment of YoungNet is based on the successful results of the

completed EU-project VIRLAN . IAT and IAO as the main technical project mem-
bers developed a virtual community with the focus on collaborative language learning
(Kempf, 1998). As in VIRLAN every user is represented as an avatar and the possibi-
lities for interaction especially for youngsters are fascinating.

164
The VIRLAN system (see
Figure 1) has been suc-
cessfully tested with a
restricted number of users
(Kempf, 1998; Kempf,
Müller, 1999). The testing
focused on functionality
and technical issues, and
Figure 1: VIRLAN, avatars with speech bubbles
the overall result has been
very promising. YoungNet takes up parts of the technology of VIRLAN and shifts the
focus away from mere technical development towards more concessive content and at
the same widens the range of applications, develops new innovative functionality and
offers a platform to the general young audience.

The design of the unified system will support a paradigm shift from pure instructivistic/
behaviouristic to constructivistic learning (Baumgartner, 2001) where teachers become
a facilitator of learning rather than an instructor delivering factual knowledge.
Communication and collaboration is offered on an individual level between teachers/
pupils by appropriate communication tools, meeting places (e.g. teachers’ room) and
games for leisure time activities. Teachers and students are free to use these features on
a national or on an international level for establishing contacts.

However it is very important that the system is provided with guidelines, best practice
examples and material for use during the lessons, otherwise it would not be very helpful.
The chance to use it in the most efficient and planned way would be low. Therefore
the project also includes the development of the YoungNet content. Some of the
partners (especially the publisher Klett) have already gained a lot of experience in
former EU-projects (Eurogame, Eurodelphes, see http://www.educational-concepts.de)
which they have contributed to the content and usage guidelines.

4. The YoungNet Content

The YoungNet content is designed for use in school. It is a combination of materials
optimized for multimedia-supported lessons, including:

165
 • Multi-user educational content specific games which can be played alone
or in teams
• Content specific teaching and methodological guidelines for teachers
and students
• Multimedia materials (e.g. sounds, pictures, audio files, videos, etc.)

The content reflects subjects taught by multidisciplinary teachers involved, at

moment Geography and English because they are the partners’ specialism and both
subjects are easily woven into a good overall story which is the background for the
content.

Collaborative contents are exemplified by the “Travel Book” story board: here every
school produces their own travel book in class concerning where they live: famous
places, famous people, buildings, the landscape, mountains, seas, etc.

Documents are designed with a HTML-editor with the help of the teachers and
templates, then uploaded to the YoungNet server, where other schools can look at
the documents. In the next step they prepare questions for their travel book with
several similar sounding answers, and these are also put in a special quiz engine on
the server. All of the schools prepare for the national quiz contest which will be
similar to “Who Wants To Be A Millionare”. Afterwards the best schools participate in
an international quiz game.

This is a good example of how the integration of the communication and collabor-
ation between national and international classes into a scenario (story) works. The
topics chosen have to motivate the classes involved and encourage collaboration with
other classes. Generally speaking an integrated scenario has to be worked out, parts
of this scenario are:
1. Work within the class (individually and in groups)
2. Asynchronous communication between different schools
3. Controlled synchronous events such as a competition

Experiences showed that synchronous events particularly must be well prepared to work
properly because it is quite difficult to organize them between schools especially if
they are in different countries.

166
Furthermore the YoungNet system enables the teachers to create their own projects in
any subject they wish. They can upload their own instruction guidelines, create their
own content by using the comprehensive administration and communication features.

5. Design Aspects
The actual work of implementing the YoungNet System was done after an intense
phase of analysing the needs of the involved schools as well as the available technolo-
gies. Special consideration was given to:
• Graphical User Interface design
• Pedagogical guidelines

5.1 Graphical User Interface

When you go into YoungNet (see figure 2), you have the choice between the following
areas: “All Classes” and “Edutainment Area”; teachers also have the “Topic Description”
Area . Under these areas you will find the respective contents that you can click on
to get to the corresponding areas. In “All Classes Area” you will find the public folders
of all classes as well as that of your own class. Click on a document or open a folder
to see individual documents. The class representative can include new documents.
The “Topic Descriptions” are only visible to the teachers. Here they will find the
relevant descriptions for each project topic. In the “Edutainment Area” all available
games are listed.

Figure 2: Look and feel

of the YoungNet

167
We took several considerations into account. The first precondition is the
YoungNet Navigation develops according to the needs of the users. The YoungNet
System includes all features needed for convenient communication between the
users and administration of files and accounts. To facilitate easy navigation the
number of features is minimized.

The look and feel of the system has to be attractive for youngsters. Therefore the
LMS was adapted accordingly (see Figure 2: Look and Feel of the YoungNet System)

5.2 Pedagogical guidelines

YoungNet provides the involved schools with guidelines for project work with pre-
defined content ideas. As the project includes international aspects intercultural
communication will be needed and supported by the system. For language training
especially synchronous audio communication is a very useful tool.

The games have a pedagogical background. By synchronous communication in virtual

3D multi-user environments the users are enabled to experience nearly authentic
situations. The motivation for learning increases through the facility to design their
own games according by simply uploading of individual multimedia files onto the
YoungNet server.

6. Edutainment Area
In the Edutainment area students have the possibility to learn and deepen learning
using games, some in 3D. The games can be played alone or with other YoungNet
Users. At a technical level, it offers both synchronous and asynchronous communication
based on leading edge ICT like Virtual Reality, 3D representations, audio communi-
cations, shared applications and provides a fascinating and “edutaining” multi-user
environment. Users are self motivated to use the Edutainment Area thereby overco-
ming the borderline between learning (school) and leisure. And to keep in mind,
the students have a strong incentive to practise a foreign language: they want to win
the game!

168
6.1 Virtual Home—The Playing-table
In the Virtual Home there is a playing-table. The game leader sits on the purple chair.
On one hand, the playing-table gives the possibility to opportunity and discuss project
results as a presentation visible to all. On the other hand different games can be played
together, where the participants are represented by avatars sitting around the virtual
table. Here the students use Audio-Chat for communication.

Figure 3: YoungNet Playing-Table

Figure 3 shows the following: The Virtual Home Screen is displayed in two sections.
In the upper part (3D area) you can navigate within the house that you have selected.
Beneath there are different buttons and an entry line for text-chat. Every Virtual
Home includes a playing table, through which you can start the YoungNet games
and a picture gallery, where you can show the pictures, which you have uploaded to
the YoungNet System. Around the table there are 2-5 seats. The purple seat in the
middle is for the game leader. Only the user, who sits on the game leader seat, can
call up the different functions, e.g. start a game.

169
6.2 Pairs
One of the YoungNet games is Pairs (see figure 4). Here, it is possible to play games
that have been designed by other schools or by the YoungNet system. Teachers have
the facility to create their own game variations (Game-Editor). Pairs can be played
using different difficulty levels and three different modes: picture-picture, word-word,
pic-word. To practice, pairs can be played alone or with other students. To communicate
with each other, the students use text or an audiochat (Voice Over IP).

Figure 4: YoungNet Pairs

6.3 EuroMap
EuroMap is a virtual treasure trail for teams (see figure 5). At the start of the game
there is a mission with three parts: Where was Beethoven born? In which city U2 was
founded? The people in what country eats the most sweets in Europe? One of the
players is the driver, who has to find the mission destinations on a three dimensional
Europe map. The team-mates (navigators) will see the map from a bird’s-eye view and
will help the driver to find the necessary destinations. The players use the text chat to
communicate with each other.

The game has two difficulty levels. At level one you just have to find the flag of the
destination country. At level two, you have to put a ball into a hole at the flag using
a virtual car (just like golf ).

170
Figure 5: YoungNet EuroMap

6.4 Quiz
In the YoungNet project, every class has the chance to qualify for the YoungNet
Champions League. This is a live quiz involving the best four classes from a year (see
figure 6). The qualification takes place in the EuroCup, where each class has to per-
form in an online quiz during a period of one week. To practice the quiz module is
available to every student in YoungNet. Teachers have the possibility to create their
own quiz with the Game-Editor. The participants communicate via Audio-Chat.

Figure 6: YoungNet Quiz

171
6.5 Virtual Home
The Virtual Home is the personal home in YoungNet (see figure 7), where 3D worlds
underwater, in the air or underground are on offer. Friends from YoungNet can be
invited to the Virtual Home to communicate and to play. In the Teachers’ Virtual
Home, planned events take place at defined times.

Figure 7: YoungNet Virtual Home

Figure 7 shows the following: Every YoungNet member has their own personal virtual
house within the YoungNet System, called the Virtual Home. A maximum of five
users can go into each Virtual Home. Each time you enter your Virtual Home you
can choose between three different Virtual Home basic types (Water, Earth, Sky).
Depending on the Virtual Home you have chosen different Avatars are available. An
Avatar is the pawn in a game, which represents you in the Virtual Home. Once you
have chosen a Virtual Home and an avatar, you can enter the Home. In your Virtual
Home you can insert pictures in jpg- or gif-format, to show to your friends and

172
pupils from other classes. You can also play all of the YoungNet games in your Virtual
Home. You also decide who gets permission to enter to your Virtual home and who
does not.

6.6 Virtual Home—Avatars

The Avatar is the personal representative in the Virtual Home (see figure 8). In the
underwater world for example you can move with a submarine. Three different
expression (smiling, angry, normal) and five gestures support communication.

Figure 8: YoungNet Avatars

An Avatar is a kind of pawn in a game, which represents you in the Virtual Home
(see figure 8). If you have uploaded a passport photo, individual Avatars, which
include your photo are also available.

6.7 Game Editor

Using the game-editor teachers have the possibility to create new or to edit available
learning games. They can choose from the games Pairs and Quiz. In addition the
teacher can choose if he wants to publish the game for everybody or only for his class.

173
Figure 9: YoungNet Game Editor

With the “Quest-Editor” you can create or modify quiz games (see figure 9). You can
edit or modify questions and the corresponding answers. For a live game at least 5
questions with at least two answers have to be defined. With the “Picture-pair-game-
editor” you can create and/or modify picture-picture-pairs or picture-word-pairs. For
a live game at least 20 picture-pairs with a vocabulary have to be created.

7. Summary
We have shown the outline of a concept to bring networked learning to schools and
youngsters by offering an integrated real-time virtual education environment with
both synchronous and asynchronous communication based on leading edge ICT
technology. More specifically the objectives are to

• provide a community platform that enables young people up to 14 years of

age to communicate and co-operate across national borders and cultures;
• offer a fascinating and edutaining multi-user environment with innovative
services and learning materials with clearly defined, sensible content;
• provide a platform students are intrinsically motivated to use and thereby
overcoming the division between learning (school) and leisure.

174
8. Outlook
The project promoted by the European Commission met with impressively positive
response in the schools that were involved. Both teachers and pupils in the participating
schools were enthusiastic about Youngnet and perceived a considerable added value in
it as a worthwhile supplement to and even a worthwhile part of school teaching.
Youngnet was an extremely successful project which was partly funded by the European
Commission (European Commission Information Society Technologies Programme
action line “Schools of Tomorrow”). Unfortunately, it was not possible to sustain the
Youngnet System without follow-up financing—much to the regret of the schools
(there were many schools which were registered on a Youngnet waiting list besides
those that participated) and the Consortium. It is a little comfort that some suggestions
from the Youngnet project were resumed in the product VITERO (a web conferencing
and live e-learning software by the Fraunhofer spin-off VITERO GmbH). With
financial support it might be possible to revive Youngnet.

9. Authors

Fabian Kempf studied engineering at the University of Stuttgart. He’s been employed
at Fraunhofer IAO since 1995 as a project manager of industrial consulting mandates
as well as national and international research and development projects. His work
comprises the topics Learning Communities, Live-E-Learning, Cooperation Tech-
nologies and Desktop Virtual Reality. In September 2005 he became Managing
Director of the Fraunhofer Spin-Off VITERO GmbH.

Karin Hamann studied industrial and organisational psychology at the university

Tübingen.1995-1999 she worked as research assistant at the Fraunhofer IAO, and
since 2001 has worked there as a scientist. The main topics of her current projects
are e-learning e-collaboration and knowledge work management in projects with
interdisciplinary and distributed teams.

175
10. References
Baumgartner P. (2001). Webbasierte Lernumgebungen—Neue Ansätze zum
Politiklernen. Retrieved April 8, 3006 from http://www.mrge.de/lehrer/
sigismund/software/hotpot/links/VLEs-lang1.pdf

Kempf, F (1998). VIRLAN—Language Learning Community for Young Children.

In: Proceedings of ED-Media / ED-Telecom, Freiburg in Breisgau; June 20-25.

Kempf, F. & Müller, A. (1999). VIRLAN—VR Community zum Sprachenlernen.

In: Proceedings of Learntec, Karlsruhe; February 9-11,1999.

Soliman, F. (2001). Role of VR in e-Learning. In: Proceedings of VRIC (Virtual

Reality International Conference), May 16th-18th 2001, Laval, Laval, France.

176
177
178
Games based learning or learning based
games? A case study
Chris Brannigan and Angela Owen

Chris Brannigan
Chris is CEO and one of the founders of Caspian Learning. Prior to founding Caspian,
Chris spent 8 years in strategy and consultancy roles within the Banking and Infor-
mation Technology sectors. Chris undertook his first degree in Psychology at Cardiff
University. This was followed by post graduate research at the University of St Andrews
into the electrophysiological correlates of long term memory.

Angela is Head of Research and one of the founders of Caspian Learning. Prior to
founding Caspian, Angela worked for BT Labs in the UK, leading work to develop
image processing software based upon the characteristics of the human visual system.
Angela undertook her first degree in Psychology at Cardiff University. This was follo-
wed by doctoral research into motion perception within the human visual system.

Abstract
Educators have concerns over the learning value of educational computer games,
this is a key barrier to the widespread adoption of such technologies within educa-
tion. We describe our work at Caspian Learning in approaching this problem by
embedding pedagogy, content mapping and assessment into learning based games.
This article explores the reactions of educators and students to learning based
games in a formal classroom situation. Tangible examples are provided within the
article, demonstrating that the learning hurdle can be overcome. Learning based
games can be powerful tools that clearly aid teachers and students, and are valued
as such. When such conditions are met, educators are ready to embrace these tech-
nologies. The use of gaming technology in itself is not a barrier to achieving credi-
bility and acceptance within education. There are a number of issues that are
cited as barriers to the use of games for learning; “learning value” need not be one
of them.

179
Games based learning or learning based games? A case study
For all of the excitement among educators around the potential for using computer
games technologies for learning, there remain considerable hurdles to their widespread
application in the classroom. These obstacles have been highlighted by many authors
(Lunce 2004, Egenfeldt-Nielsen 2005) and include: very high cost; technical comple-
xity; time to develop, the practicalities of classroom integration, teacher skills and the
low specification technology in schools.

However, while technical and practical obstacles exist, the fundamental barrier to
using games in the classroom concerns learning effectiveness. This is a frequent criti-
cism levelled by educators, who are not convinced of the actual learning benefits that
games based learning may offer (Kirriemuir and McFarlane 2004). For many, games
based learning seems a bit ‘frivolous’. Okan (2003) refers directly to this problem by
questioning the value of ‘edutainment’ and the expectation that learning should al-
ways be “colourful and fun”. Specifically, edutainment can be a diversion away from
the “need to work, serious study and the development of cognitive structures”.

In our experience, there are at least 3 major concerns for educators:

• Pedagogy. Where is the learning?
• Content. How to integrate learning content and map to a curriculum?
• Assessment. How can learning be measured?

In this article I will describe our work at Caspian Learning, and the work of other
practitioners in approaching this problem. I aim to demonstrate, by means of tangible
examples, that the ‘learning hurdle’ can be overcome. Furthermore, educators are
more than ready to embrace games based learning methodologies, not for novelty
value, but for their value as powerful learning aids in the classroom. It is possible to
utilise games based learning methods to achieve learning goals that would be more
difficult using other methods. Fun is only part of it!

Games Based Learning or Learning Based Games?

In the learning games community, we often talk about developing ‘Serious Games’.
There is no clear definition of what this means, but it generally indicates an experience
where the student is clearly not just enjoying a game but is deriving obvious learning

180
benefits. In attempting to utilise or test serious games in the classroom many researchers
start from the gaming perspective. They observe some popular commercial off the
shelf game titles, and attempt to identify the factors that facilitate elements that are
interesting from a learning perspective, such as, engagement, collaboration, competi-
tion, and the ever elusive ‘systems thinking’ output. The majority of writings in this
field and classroom based case studies undertaken have taken this approach.

More recently, a number of researchers have begun to design and create games based
learning applications from scratch. These researchers will often start from the learning
perspective. That is, to consider what are known to be key factors underpinning suc-
cessful learning, and then to utilise the tools and technologies from the paradigm of
interactive 3D games, to enhance those factors.

Starting from the learning perspective

There already are some notable examples of this approach.

Immersive Education, a UK company formed out of research from Oxford University,

target ‘story telling’ as a learning method. This method engages students in the
construction of scenarios, and the retelling of events in their own words, or worlds
in this case! Immersive’s ‘Kar2ouche™’ technology, involves the student in an active
process of story development, and scenario creation. To do this, the student must
construct rich 3D scenes, and then, add in complex interactions. The student, or
class can collaborate and share their content.

Immersive have published many titles using this approach covering all curriculum areas.
This approach has been very successful, with deployments in over 5,000 schools in the
UK, and is now also being used in the United States. http://www.kar2ouche.co.uk/

Team Play Learning Dynamics (TPLD), a company working with the University of
Dundee in Scotland, have focused upon ‘teamwork’ dynamics and capabilities. Their
approach is founded upon psychological research work into this area. TPLD have
integrated this research as scenarios to be experienced within multi player 3D compu-
ter gaming environments. Their products are now being used in corporate training.
http://www.tpld.ltd.uk

181
PIXELearning have applied computer game and simulation techniques to create
business education applications. The learning application ‘The Business Game’ places
the student in the position of running a business or business department. The game
takes the learner through various stages from initial product selection and market
research through to manufacturing and selling.

During the game the student will face many different scenarios and they must manage
many different elements simultaneously if they are to fight off the market competition
and make a healthy profit. Integrated performance information provides constant
feedback to the student and educator during gameplay.

PIXELearning have also developed a platform called ‘LearningBeans’ that allows

educators to create their own business running simulations using simple wizards.
http://www.pixelearning.com/

Caspian Learning case study

At Caspian Learning our development work has focused upon thinking skills—an
individuals ability to process information in different ways to achieve different lear-
ning outcomes. The objective was to engage students in the process of applying diffe-
rent thinking skills to manipulate subject content and to think about different thin-
king processes. For example, to be able to reflect upon the process involved within a
structured enquiry and to understand how and when to apply that process. To achieve
this goal we developed two complimentary technolgies for the classroom:
1. Subject specific learning applications based upon a technology named
Qcognition®. This is a set of distinct thinking processes which can be
embedded into a game as routines, options and actions.
2. Thinking Worlds™. A content authoring and classroom education tool.
Using this tool, educators and students can use the Qcognition® thinking
processes to edit, create, publish and share their own 3D learning games.

The technologies are complimentary. Educators can use the learning applications to
engage students in different thinking processes around their subject content. Then
educators and students can use the Thinking Worlds authoring tool to target meta
level thinking as students can edit thinking tasks and content or create whole new
thinking tasks within the subject. These technologies are described in more detail below.

182
1. Learning applications
At the outset, we wanted to leverage the real time interactivity of the games engine to
create a range of different thinking processes that would support the learning processes
of students. In relation to the research literature the thinking processes act as ‘cognitive
tools’. Van Joolingen (1999) describes a cognitive tool as an “instrument that is part
of the learning environment that supports or performs an identifiable cognitive process”.

The cognitive tools in our methodology are embedded as routines and options that a
student must actively direct within a 3D environment, such as identifying information,
sequencing information, evaluating an argument, or assembling a presentation. We
describe these as ‘cognitive tasks’. During a specific task, the student is set a learning
objective determined by the embedded thinking process. For instance, a frog may ask
the student to explain the food chain in the local habitat and the student (player)
must then assemble a presentation to give to the frog.

Within a cognitive task, the student would have the opportunity to perform multiple
behaviours all related to a specific cognitive processing operation. For example,

In one low level cognitive

task, namely a knowledge
gathering task, the student
would undertake simple
knowledge gathering acti-
vities. They would interact
with objects that would
hold subject relevant infor-
mation, such as meeting
Louis XVI in a French
Revolution scene, or a cell
within the human respira-
tory system. The student
has the option to receive Figure 1: 'French Revolution' in game screenshot
different amounts of information and must collect multiple-choice questions, storing
questions and answering them at any point. They can then search out objects that
may hold information to help them.

183
In one higher order cogni-
tive task, namely an infor-
mation integration task,
the student must integrate
new information with old
information in order to
satisfy an outcome. For
instance, the student would
be given information about
the normal conditions of a
component within an
electrical device and an
outcome showing a prin-
tout of a fault. Students Figure 2: 'Electronics' in game screenshot

must then assemble the additional conditions that could give rise to the fault outco-
me. To do this, they may interact with different components, selecting different infor-
mation elements to integrate with the starting information to satisfy the outcome.

The Qcognition® cognitive tasks can be aligned along Blooms taxonomy of cognitive
skills (Bloom 1964) as shown in Figure 3.

For example, Blooms Knowledge category of cognitive skill relates the the recall and
simple retention of information. This covers such activities as: recognising; gathering;
recalling; identifying; selecting; and labelling. Qcognition® includes a number of
cognitive tasks, such as Recognition, Elimination and Knowledge Gathering that
embodie these types of activities and so can be positioned against the Knowledge
category within the taxonomy.

Blooms Evaluation category of cognitive skill covers the ability to evaluate and make
judgements about the value of information / materials for a given purpose. This covers
such activities as: contrasting; criticising; arguing; appraising; evaluating; describing;
and explaining. Qcognition® includes a number of cognitive tasks, such as Structured
Investigation, Integrate New and Old and Assess Predictions that embodie these types
of activities and so can be positioned against the Evaluation category within the
taxonomy.

184
Figure 3: Qcognition® cognitive tasks mapped to categories of cognitive skill along Blooms (1964) taxonomy

Content and Curriculum Mapping

For games based learning methodologies to work effectively in the classroom, they
must be actively embraced by teachers. Teachers in all education systems are under
ever greater pressure to deliver results. These target ‘results’ are often set centrally.
There will be specific learning objectives for the students to achieve. Curriculum
guidelines will often also stipulate activities and methods that can be used to deliver
these objectives. It is for these reasons that in many cases teachers are reluctant to
work outside of the central curriculum. In this situation, to confidently utilise games
based learning applications in the classroom, teachers need to be sure that the activi-
ties that they are undertaking are directly linked to their objectives.

Mapping to the Curriculum and Integrating Educational Content

The Caspian cognitive tasks can be directly mapped on to the cognitive learning ob-
jectives specified in a curriculum subject area. For example, in the English National
Curriculum, for “World War 2—The Home Front”, section 5 covers ‘what did people
eat during the war’ and has the following required learning objectives:
• Infer causes of rationing

185
 • Identify some similarities and differences between foods available
during the war and today

We can map these objectives onto Blooms taxonomy of cognitive abilities and identify
appropriate cognitive tasks to target these learning objectives. In this example the tasks
are self evident as the learning processes are directly specified—‘make inferences’ and
‘compare and contrast’ processes.

The next step is to map the learning content by choosing the appropriate the 3D
environment and subject specific content that will feature within the cognitive task.

Once tasks have been completed to cover a subject area, they are then formed into an
overall games based learning application. Figure 4 shows the application menu for the
WW2 application. To date this process has been employed to create applications for
21 different subject areas.

Figure 4: Front screen from WW2 application. The application is divided into sections based upon the English
curriculum. Each section contains different cognitive tasks that target learning outcomes specified in the curriculum.

186
Assessment
The capturing of rich performance information for use in scaffolding processes and
assessment is in our view one of the undiscovered oppotunities that games technolo-
gies offer educators. Games engines can measure any specified event within a volume-
tric space over time. The ability to be able to precisely observe students directing
events and actions within a 3D world provides many opportunities for data capture.
When this 3D world is contains embedded pedagogy where the user is undertaking
targeted learning behaviours to achieve a learning outcome, then the opportunity is
magnified. Within the Qcognition® architecture mechanisms are integrated within
each cognitive task in order to record direct and indirect learning behaviours.

During the task we use this

data to feed performance
information back to the
student, so as to indicate:
correct behaviours; incor-
rect behaviours; informati-
on objects remaining; time
taken.

This data is also analysed

and fed into other mecha-
nisms designed to provide
cognitive scaffolding for
the student. Such as:
• Rules which are built into each task to guide students learning process and
performance
• A Revision System following completion of a task which comprises of three
elements, namely:
1. Quantitative Information Screen—factual performance information
(see Figure 5).
2. Verbal Feedback Screen—performance specific advice.
3. Revision Materials Screen—Targeted content is fed back to the user
based upon the errors that they have made during the task.

187
Figure 5: Screenshot from WW2 application showing the Revision Screen functionality. The page is open on the
quantitative feedback. In this case the student has completed a Labelling task.

The information captured is also utilised to support the student and educator outside
of the immediate lesson. When playing the applications the students cover alot of
information and make many learning judgments. After each task, information on the
task and the students performance is written into a personalised electronic workbook
for the student.

Detailed information is captured within a performance database. This provides several

standard reports and an exporting facility. The teacher is able to compare performance
of individuals, classes and over time.

2. Thinking Worlds™ authoring tool

The development of the Thinking Worlds™ tool was motivated by the desire to target
meta level thinking in students by actively involving them in the process of creating,
editing and refining thinking tasks. In educational terms, students are asked to under-
take the role of educators where they have to edit or create a learning application so
that it can be used by other learners to achieve specific learning outcomes. Educators
can use the tool to faciliate this learning experience. The second objective was to

188
provide educators with the tools with which to create their own learning games based
upon Qcognition® that could then be used in lessons.

Teaching thinking skills and meta cognition has traditionally been very difficult in the
classroom. By using games technologies we aimed to make this process highly enga-
ging, very visual, collaborative and active. By embedding pedagogy in the authoring
tool we were focused upon keeping the whole authoring experience focused upon the
design and application of thinking processes. Authors can edit and customise a library
of 3D assets, however to add interactivity and make a ‘game’ the author must create a
thinking task.

Authors have access to a suite of

interactive Thinking Behaviours
that can be populated with sub-
ject content and then combined
and arranged into Thinking
Tasks.

The author can edit the rules,

mechanics and measurement
outcomes of thinking behavio-
urs used in the overall task.

Figure 6: Screenshot from Thinking Worlds™ editing screen.

The author can customise all

content including: 3D worlds,
2D, text, sound and movies.

Authors are free to use an ever

increasing library of developed
Thinking Worlds covering a
multitude of subjects and sce-
narios.

Figure 7: Screenshot from Thinking Worlds™ menu screen.

189
Classroom Feedback
The technology was tested and refined by Caspian Learning during six months of
piloting with students across Gateshead Local Education Authority in the UK1.
Additional case studies have been undertaken by Oldham City Learning Center2 and
The Learning Lab under the guidance of Professor Steve Molyneux3.

Across all studies students displayed enormous enthusiasm toward using the applicati-
ons which extended beyond novelty engagement. Students exhibited great demand
for the applications to be used as a self-directed learning tool in the home and at lunch
time. However, the best results were seen when the technology was used as a class-
room facilitation tool with the teacher using the cognitive tasks, learning content and
recording methods to direct the learning of the students.

Engagement of students is the principal

reason why practitioners contemplate
using games based learning. Some
studies using games in the classroom
have noted student dissatisfaction
either in not liking the game or in not
seeing the relevance with their school
work. In testing applications from a
range of curriculum areas with students
aged between 8 and 15 years old and
across all ability levels, we saw consis- Figure 8: Screen shot from pilot classroom.
tently high observed and self reported levels of engagement:
• 98% students agreed that they would like to have the chance to use the
software to learn in this way more often.
• 94% of students agreed that they preferred using the software to normal
methods of learning in the subject area.
• 82% of students agreed that using the software made it easier to learn.

1 The Gateshead pilot involved over 380 students ranging in age from 9 years to 16 years and over 2500 student hours of testing.
2 The City Learning Centre in Oldham has been using Caspian software since the start of the Summer Term 2005. In that time, eight

different secondary schools and around three times as many primary schools tested the software have as have Oldham's Youth Service.
http://www.caspianlearning.co.uk/Oldham%20Case%20Study.pdf
3 Learning Lab research undertaken on behalf of the North West Grid for Learning as part of the DiDA project.

http://www.caspianlearning.co.uk/learninglab.pdf

190
During the pilot we monitored student opinions to maintain high engagement levels.
As a consequence we layered additional engagement elements onto the technology:
• Greater task variety.
• Richer 3D environments.
• Greater choice of characters, and the use of vehicles.
• Elements of humour and surprise, for example, A.I. Chickens.

Each additional engagement element was monitored for adverse learning consequen-
ces to eliminate distractors.

Students perceptions of the ‘game’

We were concerned that the level of thinking challenge and volume of subject specific
content in the applications might disengage the students. Happily, we were wrong. In
debriefs with individual students it became apparent that students clearly distinguished
between the Caspian game based learning applications and games that they played
while at home for entertainment purposes. Students in the main were comparing the
learning applications to other methods of instruction typically used in the classroom,
not to off the shelf commercial games.

Secondly, from the first engagement with the application the students were actively
involved in the task mechanics and in discussing the learning content. Surprisingly,
these elements were as important to the students as the graphics, fun elements and
interactivity of the game. This was also borne out through live observation where the
buzz of student conversation and interaction related to learning elements—answers and
tips on performing tasks—rather than the thrill of the experience. Our preconceptions
were coloured by a generalised view that modern students require entertainment and
fun that maybe at the expense of hard work. The pilot testing quickly demonstrated
to us that students were very interested in learning and being challenged. The key was to
deliver the learning activities and challenge within an environment that stimulated them.

Hard work and challenge

A key criticism of games based learning has been the perception of this sort of activity
as time wasting or frivolous. Games are seen as the ‘easy option’ where students can

191
pass idle time and avoid focused learning. We have used the term ‘learning intensity’
to describe the level of learning effort that a student may expend on a subject matter
within a given time frame. Using the gaming environment and embedded pedagogy
we were able to get some insight into levels of learning intensity by precisely monito-
ring the learning behaviours performed by the student during the taks and the diffi-
culty of cognitive activity that the student was undertaking.

For example, these typical figures are taken from whole class averages within a
classification type of task.

In a 20 minute period, the average student: made 3 attempts, had observed

26 information elements about which they had to make a classification deci-
sion, made 17 classification decisions between categories 5 of which would
have been incorrect, 10 decisions that information did not apply to any cate-
gory of which 3 would have been incorrect. On 5 occasions they would have
observed information and not made any decision about its properties.

This was typical of student effort levels and participation and was reinforced by obser-
vations from teachers. The students wasted very little time during the lessons. The
learning games were the focus of intense learning activity. As we shall describe below,
the integration of the learning applications into the lesson also provided a focus for
enhanced participation in traditional lesson activities.

Competition and collaboration

Competition between students and groups of students was immediate. Prior to the
pilot a few educators had identified competition as a potential concern that may
negatively effect performance and student participation. The opposite was found to
be true. The students inherently competed for higher scores, greater understanding of
task mechanics and the acquisition of rewards, such as vehicles for their characters.
This was a positive influence on student performance. It was made positive because of
the high levels of collaboration that it also engendered. Competition and collaborati-
on occurrred simultaneously. Whether working as individuals or in small groups,
students would spontaneously compare their experiences, sharing answers, work
arounds, methods and examples. Within a short period a whole class of students
would be aware of ‘best practices’ within thinking tasks.

192
The collaboration and competition interactions were overwhelmingly focused upon
learning outcomes. It was exciting for everyone to observe students that were usually
extremely disinterested in the French Revolution to be actively discussing the diffe-
rences between social groups within the Revolution, and directing their friends to
approach characters from the Clergy or Aristocracy to learn more.

As the pilot progressed teachers running the lessons began make use of competition
and collaboration more formally in their lesson plan, for example by dividing the class
into groups to focus upon different elements of a thinking task before reporting back
to the class.

Teacher Feedback
During the pre-pilot briefing, the elements of pedagogy and content mapping were
identified as key benefits by teachers who understood that the students would be
undertaking valuable learning activities with content and tasks precisely mapped to
the curriculum objectives. This gave teachers the confidence to invest their time and
resources into the project. This was reflected in the teacher comments during the ses-
sions:
“I’ve never been a great fan of games-based learning, but this has captured
our attention because it is based on research into learning, not simply on the
motivational (gaming) aspects of using a 3D games environment. Listening
to pupils talking to each other during Caspian sessions it is striking how
much pupils talk about the subject, for example, the means of reducing the
impact of earthquakes—not just about the motivational features like gold
coins and skateboards”. Dave Barter Oldham City Learning Center.

Teachers as learning facilitators

The temptation with a learning game is to assume that it is founded upon a model of
purely self directed learning at the expense of the expertise of the teacher. During the
pilot work and in subsequent case studies the role of the teacher has been central to
the success of the learning game in the classroom. We found that key elements of the
learning game provided crucial support for teachers in adopting the role of facilitating
learning among their students.

193
Firstly, the level of student engagement was an enormous benefit for teachers. The
high student engagement levels freed the teachers to do what they wanted to do
—teach.

During the pilot, teachers utilised the cognitive task structures as a foundation for
‘learning conversations’ with their students. For instance, using the cognitive tasks to
undertake a structured enquiry, the teacher was able talk through the specific behavio-
urs that the student would carry out in the 3D environment, observe and assist the
students as they performed the behaviours, and post-task hold a class discussion on
the different elements of the structured enquiry. This overcame real practical difficul-
ties normally associated with teaching higher order thinking skills to a large group of
students.

Teacher feedback also showed that assessment mechanisms and task structures enabled
focused learning interventions during the lesson, aimed at either individuals or to the
whole class.

Often a teacher would observe an interesting learning behaviour on a students

computer. The teacher would then pause the class and convene a plenary session
to discuss before returning to the task.
“I always have an insight into the students performance, how they are
working through the task and how they have done. Knowing my students,
I can give appropriate feedback to them there and then, based upon exactly
how they are performing.”—Tony Hannah, Primary School Teacher.

These practices quickly became established as the norm during the sessions.

Refinements
Throughout the development testing we refined and added different elements to the
software based upon teacher feedback.

The cognitive tasks were all timed to last 10 minutes or less. This enabled teachers to
easily integrate the tasks into their overall lesson plans and to tailor their use to diffe-
rent ability level students.

194
The development of the electronic ‘student workbook’ (described above) that captu-
red a detailed record of the students performance occurred early in the pilot. One of
the reasons for this was simply to remove administrative burden from teachers. The
main reason however was to support learning. Teachers would ask students to email
their workbooks to review progress. The workbooks were sometimes used as foundations
for homework assignments. Students also used the workbooks for revision purposes,
that gave them an insight into their strengths and weaknesses within a subject matter.

The Performance Information system was used extensively post lessons to assess indi-
vidual and class performance. However, it became evident that the breadth and detail
of information captured within the applications was far more extensive than required.
For example, within a compare and contrast task knowing the time taken to observe
different information elements categorised by the action that the student should perform
on them was too much information to be of practical use to the teachers. The reports
were therfore refined to provide summary information for teachers.

Update
Since the pilot work, applications built using the software have been deployed in
schools across the UK. The Thinking Worlds authoring tool has been piloted in
Alpha version in January 2006 in Sunderland UK with groups of gifted and talented
and also disengaged students. The Thinking Worlds tool will be deployed in a global
beta trial in May 2006.

Conclusion
We have described applications founded upon research into cognitive processing
methods that utilise interactive 3D games technologies to enhance these methods.
These applications have successfully embedded pedagogy, content mapping and
assessment elements into the design.

These “Leaning Based Games”, have been shown to engender very high levels of stu-
dent engagement and attention effort in the classroom. However, this effort is neither
‘frivolous’, nor just ‘colourful and fun’. The applications map students efforts directly
on to relevant learning content and activities.

195
Within each application, data captured about learning behaviours was analysed and
fed back to support assessment, revision and other cognitive scaffolding mechanisms.
Teachers made use of these mechanisms to make make real time learning interventions
with students. The cognitive tasks structures were used to facilitate ‘learning conversa-
tions’ where teachers were able to direct students through higher order thinking tasks
that would otherwise have been difficult within normal classroom conditions.

Learning Based Games can be powerful tools that clearly aid teachers and students,
and are valued as such. When such conditions are met, then educators are more than
ready to embrace these technologies. The use of games technologies in itself is not a
barrier to credibility and acceptance within education. There are a number of issues
that are cited as barriers to the use of games for learning; “learning value” need not be
one of them.

References
Bengamin S. Bloom, Bertram B. Mesia, and David R. Krathwohl (1964). Taxonomy
of Educational Objectives (two vols: The Affective Domain & The Cognitive
Domain). New York. David Mckay.

Egenfeldt-Nielsen (2005). Unpublished PhD thesis. Beyond Edutainment

Exploring the Educational Potential of Computer Games (available at http://
www.it-c.dk/people/sen/egenfeldt.pdf February 2006).

Zühal Okan (June 2003). Edutainment: is learning at risk? British Journal of

Educational Technology, Volume 34, Issue 3, Page 255.

Wouter van Joolingen (1999). Cognitive tools for discovery learning. International
Journal of Artificial Intelligence in Education, 10, 385-397 .

John Kirriemuir, and Angela McFarlane (2004). Literature Review in Games and
Learning. A Report for NESTA Futurelab.
Les, M. Lunce (Oct 2004). Computer Simulations in Distance Education.
http://itdl.org/Journal/Oct_04/artilce02.htm

196
197
198
Field Report: Collaborative Blended Learning
meets Game-based Learning or how Inter-
net Platforms allow us to play 1000 “Low
Tech Games”
Lotte Krisper-Ullyett, Johann Ortner, Barbara Buchegger

Preface
In mid-2004, I (Lotte Krisper-Ullyett) was invited to submit a paper on my practical
experiences with game-based learning to an Austrian e-learning conference1. Since I
had played such games with several groups of students during the previous academic
semester I felt this was a good opportunity to reflect on what had happened, structure
my thoughts and find out what others had to say on the topic. I must admit, I had
expected to find like-minded tutors and lecturers using forums, platforms, chat rooms
etc. to experiment and play games with their students or, to be more exact, to let their
students play games with each other. I was very surprised to find that most of the
other examples showed a world which was very alien to me: a world full of avatars, 3-
D landscapes, beautifully designed games and simulations. All games which must
have taken several man years to design and build. My first thought was one of minor
panic. Was I the right person for this conference? After all, all I had to show was a
forum thread and some simple children’s games I had adapted for use on the Internet.

Nonetheless, I submitted my paper, gave my presentation and found that—despite

my initial misgivings—I received a warm welcome in the game-based learning com-
munity. It seemed that although the origins of game-based learning lay in a different
domain to my area of focus (collaborative blended learning), the community was just
starting to embrace the idea of ”low tech games“, and the examples I provided happe-
ned to be “just the thing they wanted to hear”.

1 9th Business Meeting, TU Vienna, 2 and 3. December 2004, “Game-Based Learning und e-Assessment”,

organized by bm:bwk und fnm-austria, http://serverprojekt.fh-joanneum.at/sp/

199
About this paper
This paper begins with a discussion of why we might want to play such games in the
first place. This is followed by a detailed description of two of the games we (Lotte
Krisper-Ullyett and Johann Ortner) have played with our students. The paper conclu-
des with a summary of our (the authors) findings and the proposal that some of these
findings in the field of collaborative blended learning could well also be applicable to
the world of game-based learning.

1. Why play online games?

The first issue we will look at is the question of why we might want to play online
games with our students in the first place? What kind of problems do we seek to
resolve using online games?

To answer this question, we would like to give the following two examples of situat-
ions in which we felt it appropriate to play a game with our students.

• In a course on “Personal Networking”, it came to our attention that whereas

students tended to judge other people’s use of personal networks as “unfair”,
they saw their own use of their own personal networks as “quite acceptable”
or “totally natural”. In some cases, students simply did not even recognize the
fact they were actually using personal networks. We realized we needed to find
some way of helping them to get rid of these blind spots, i.e. to make them
aware of the difference between their “theories in use” and “theories espoused”
(as Chris Argyris2 would describe this phenomenon).

• We also noticed that students identified with the use of ”scientific terminology“,
even if they at times only had a very superficial level of understanding of what
this terminology might mean. The German word Schaumschlägerei describes
this phenomenon perfectly. Translated literally, Schaumschlägerei means to
“whip up foam”, but it also translates into English as “humbug”, “bragging”3,
“hype” or even “blowing hot air”. How could we make our students better aware
of their use of language, encourage them to seek the original meaning of a term,
phrase or statement (e.g. “win-win-situation”, “intrinsic motivation”, etc.) and
even refer to the person who had coined the term in the first place?

2 Argyris, C. (1996): Knowledge for Action. A Guide to Overcome Barriers to Organizational Change. San Francisco: Jossey-Bass
3 http://dict.leo.org, http://www.dict.cc

200
In both of the above cases, our solution was to let them play a game and draw their
own conclusions for themselves! That way they would teach each other about the
situation, and we as tutors would remain on the sidelines.

We felt that game-based learning was a suitable solution, since we were aware that
there was little use in simply “telling“ the students to change their attitudes or behav-
iour (an approach they would have found moralizing and boring). We could literally
sense the resistance we would have faced if we had tried to lecture them on these
issues. We also felt they would be more interested in what their peers had to say about
these issues, especially since they concerned values, norms or attitudes. They would
rather match themselves with their peers than with their teachers. Last not least, we
also felt it would be fun to play a game and watch something “evolve” with the group
as a whole in charge, not the teachers. The game would also provide us with plenty of
pertinent, real-life examples for our next face-to-face session and give us the back-
ground material we needed to discuss how the picture might have changed (e.g. how
85% of the students had used their personal contacts to get their last summer job or
how easy it is to fool people with buzzwords and statements which sound “mega-
scientific” but are, in fact, nonsensical).

2. Description of two “low-tech online games”

In this section we will describe the way we designed the games/tasks to stimulate
interaction between the students.

Case A: “Personal Networking”

To illustrate the use of personal networks, we asked the students to tell the story of how
they had found their last summer job. In a first round, they were asked to analyse
their stories in accordance with the following structure:
• the participants involved in the process
• their relationship to these participants
• the actions involved
• visualization of the process.

In the next step, the students were asked to generalise their stories using an abstract
statement, a model or a theory. In the final round, students were asked to award a
total of 10 points to those statements they felt were most relevant and based on actual

201
evidence. Not only did it become clear that (with a few outstanding exceptions) every
student had used his/her own personal networks (family, schoolmates or friends) to
find a job, the group also found the examples highly relevant when we later went on
to discuss well known theories and concepts such as “the strength of weak ties“,
”structural holes“, etc.

Even though the final task “visualization of the process” (intended to help the stu-
dents identify the essential factors) was on voluntary basis, one third of students con-
tributed visualizations of their stories: a knock-on effect that began when one student
scribbled something down quickly on a piece of paper and posted a scanned copy in
the system. As a result, others were inspired to visualize their stories (with increasing
levels of complexity).

Figure 1: Student visualisations of their experiences with personal networking

This “level of sophistication” in turn inspired another student to post a comical visual-
ization in which he pictured himself as the actor Brad Pitt, claiming it had been his
good looks that had actually secured him his summer job:

202
Figure 2: Student’s ironic visualization of his personal networking activities.

Was this game-based learning? And if so, were we then “game designers”? At the time,
we would probably have said no. All we had done was to formulate an e-mail that
triggered a high level of forum activity over a certain period of time. All we had want-
ed to do was to find a way out of a situation we perceived as a barrier to a learning
experience. However, with hindsight, we can now conclude that our didactic design
did indeed have all the elements found in a game, namely:
• it involved a group of people
• these people had a set of tasks to perform or questions to answer
• a set of rules was in place
• the game took place on a “board”, in the form of a clear forum thread
• the participants could win points
• a “referee” was in place to explain the rules, make sure the game was
played fairly or help people get back on track if they got lost.

Figure 3: Online platforms generate space for social interaction—and games.

203
Today, we recognize the similarities in the concepts of “collaborative blended lear-
ning” and “game-based learning”. Indeed, we even think of “collaborative blended
learning” as a special form of “game-based learning”. However, before we go on to
discuss our experiences at a more general level, let us first look at another example of
a simple “low tech” online game.

Case B: The “Bluffing Game” or “Encyclopaedia Game” (Lexikon Spiel)

This example describes the game we used to try to stimulate a learning process with
regard to the use of “scientific terminology” (see above). “The Bluffing Game” is
based on a “real” game we used to play as children on rainy summer days called the
“Bluffing Game” or “Encyclopaedia Game”. I remembered playing the game as a
child while preparing the course material and thinking about playful ways of using
language. The rules of the game(s) are as follows:

The original “Lexikon Spiel” (Encyclopaedia Game)

Target group: 12-99 year olds, with a keen interest in words and language.
Recommended number of players: 5-10.
The game setting: Rainy day, with nothing else to do. The players sit around a
table. Each player has paper and a pencil. One encyclopaedia is used by the
whole group.
The rules: First, the players choose a word from the encyclopaedia (one none of
them know the meaning of ). Each player then writes down a (pseudo) definition
in the form of an “encyclopaedia entry” on a piece of paper, folds the piece of paper
in half so none of the others can read it and hands it to the moderator. The moder-
ator copies the correct definition from the encyclopaedia on to a piece of paper
and shuffles it with the definitions provided by the other players. Only the moder-
ator knows the “right” definition from the “wrong” ones. He/she reads the definiti-
ons aloud, one after the other and each player writes down the one he/she believes
to be correct.
How to win: Players are awarded one point for every other player who selects their
definition as the right one. Extra points are awarded to players who pick out the
correct definition from the other (pseudo) definitions. The player with the most
points at the end of the game wins

204
 Skills required: Good language skills. A good level of general knowledge and good
judgement of plausible explanations are also beneficial.

The next step was to adapt the game for use in our workshop via the Internet. The
results of this transformation are described below.

The “New” Bluffing Game

Target group: Thirty twenty-two year old third year students at the University of
Applied Sciences in Burgenland, Austria. At the time, all the students were looking
for companies to employ them for their compulsory 6 month work experience place-
ment. Following that, they would return to university to write their final year dis-
sertation.
The course setting: A soft skills workshop (“Personal Networks”) for students who
had chosen to specialize in “Knowledge Management”. The aim of the course was
to provide students with an understanding of personal networks and their relevance
both for organizational knowledge management and for the students’ own profession-
al careers. The workshop was divided into three half-day-workshops interspaced with
online phases—a typical setting for collaborative blended learning.
The targets of the use of is specific didactical design: To make the students think
about their use of language, especially with respect to their final year dissertations.
To encourage students to search for information. To relate the theoretical aspects of
personal networks with the students’ real-life situations, e.g. with the task of looking
for a job.
Adaptations required to use the game on the Internet: The encyclopaedia was
replaced with a bibliography of well-known authors and the students were provided
with literature on the topic of social networks. Instead of pencils and paper and a
table, students gave their answers via an Internet forum. Instead of playing for a
couple of hours, the game was played over several weeks. The participants were not
only permitted to but actively encouraged to research the term and refer back to the
original author’s statements and concepts.

The rules: In the first round, each student was asked to post three topic-related
statements in his/her own words (i.e. not simply by copying & pasting any infor-
mation they found) based on selected scientific literature. One of these statements

205
 had to be fake, i.e. sound highly scientific, but in fact be nonsensical. In the second
round, students were asked to classify the responses given by their peers as “correct”
or “nonsensical”. In the third and final round the cards (definitions) were put on
the table and the fake definitions revealed.
How to win: There were two winners.
1. The student who was able to identify the greatest number of fake definitions.
2. The student who was able to fool the greatest number of other players.

The live game as played in the summer of 2004 can be accessed via the
following link:
(permission for access has been given by the students, for scientific purpose only)

http://fhib5jg.factlink.net/147436.0
username: testlku, password: testen

The “game board”—a web-based discussion forum

The game took place in a discussion forum on the Internet. The game was launched
on February 3rd, with Step 1 beginning on April 18th, Step 2 on April 20th and Step
3 on April 25th. Figure 4 shows the forum thread containing the students initial
statements (Student 1, Student 2 ....). It also shows the reactions of other students,
i.e. whether they thought a particular statement was “correct” or “fake” and whether
it had any relevance for their own current situations.

Figure 4 contains a small extract of the forum thread (the names of the students have
been blanked out) and shows three of the four statements made by one student in
accordance with the e-mail instructions sent out for “Step 1”. He selected Manuel
Castells in 2001 on weak online connections (schwache on-line Verbindungen) for his
first two statements and Robert Kaplan and David Norton on Balanced Scorecards in
1997 (customer relationships—Kundenbeziehungen) for his third statement. He post-
ed all of his statements on March 20th (almost one month before the actual deadline
—making him a so-called “early bird”—yet another typical phenomenon often found
in online environments).

A month later, on April 20th, he received the first replies from his peers (in accordan-
ce with the instructions received for “Step 2”). They had been asked to assess the

206
Figure 4: Screen shot of forum activity in the “Bluffing Game”

statements provided by the other students as follows: “Is the statement relevant for
my life?” and “Is the statement correct or fake (richtig/falsch)?” As Figure 4 shows,
the students were of different opinions regarding these statements. Although they
all considered the information relevant to their own lives, two of them thought the
statement was fake, while one thought it was a correct rendition of what Kaplan &
Norton had said in their book.

A week later, on April 27th, students were asked in “Step 3” to reveal their “bad eggs”
and add up the number of points they had won. In this case, the student states in the
posting headline that he had been able to find two fake statements (“2 Mal ertappt”)
and that he had been able to “fool” two of his colleagues with his own statement (“2
Mal getäuscht”). The text body of his posting is a good example of high quality work:
the student corrects his statement (“...sind nicht auch rentable Kunden”), shows how
easy it would have been to find the right answer using an Internet search engine such
as Google and provides a link to the exact quote in the book itself.

207
Game Design
With games of this sort, the tutor assumes the role of “game designer” and is therefore
not separated from the process of game design. In most cases the “game design”
manifests itself in the e-mail or instructions sent to the participants in the game. In
our “Bluffing Game” example, an e-mail of about 40 lines in length (see below) trig-
gered and structured forum activity from 30 people for 6 weeks, with a total of over
300 forum postings. This included 90 initial statements, about 200 reactions and 30
disclosures/summaries.

Task Nr 2—Literature, recommended time: 4 H

Get inspired by the recommended readings (as handed out in the face-to-
face meeting).

Step 1 Pick out relevant, interesting new insights

Choose one (or several) book(s) or author(s) from the above list that seem
to handle a topic relevant to your current personal circumstances. If you
come across a statement, an assumption, a concept or a recommendation
for action, which you believe is particularly relevant for you and gives you
new insight, share it with your colleagues. Try to reformulate this new in-
sight in your own words (one sentence or paragraph is enough, copy&paste
not allowed) and post it into this forum. All in all, choose four such state-
ments regarding personal networks which are based on scientific work.
Make sure your quotes are exact and correct, so that your colleagues are
able to find this information (author, title, year, page, library.)

Attention: Challenge!
Now the challenge: one of the four statements should be fake and “smug-
gled” in. Ideally, your colleagues should be able to find out through simple
research on the Internet that the statement can’t be true and uncover your
manipulations. In other words, please do not use any “false statements”
that could never be uncovered or contain a negligible change such as a
slightly wrong date, etc. Your manipulations should concern the very core/
meaning of a statement. The most elegant manipulations which were able
to fool many colleagues even though it would have been easy to discover
will be awarded :-)

208
 Form: One posting per statement (again: only a sentence or paragraph,
own words, no copy&paste) Title: [Author—Year—Key word—why relevant]
Example: [Granovetter—1973—Weak Ties—Job Search]
Deadline: 18 April 24.00
Place of Delivery: here in this forum
Recommended amount of time: 2-3 hours research & formulation

Step 2: Rating of relevance / correctness

Choose 6 postings made by your colleagues. Do they seem relevant to
you? Are these correct statements?
Form: [Last Name—relevant /not relevant—correct/faked]
Place: in the respective forum thread
Deadline: 25 April 24:00
Recommended amount of time: 1 hour including research

Step 3: Disclosure of fake statements

Now disclose your phoney statement. Set it right! Ideally show us the link
to the website where we could have found it.
Format: Posting in relevant forum thread
Title e.g. [Fake—caught 3 times—fooled 0 times]
Deadline: Friday 30 April 24:00
Recommended amount of time: 1/2 hour

Once a process has started, it is very difficult to change instructions or rules. As a

result, the text containing the instructions must leave no doubt about the task, time
schedule, rules, prizes, etc.. It is a good idea to think it through step-by-step or ask
colleagues to play it on a trial run basis. The instructional text needs to be particularly
precise in cases where the tutor does not meet students face-to-face at regular inter-
vals. There are a number of books available that offer advice on how to play such
games with examples of possible designs, e.g. “101 e-Learning Seminarmethoden4”
by Hartmut Häfele and Kornelia Maier-Häfele. However, our experience shows that
games will always need to be adapted to the type/characteristics of the group and their
“cognitive style”.

4 Häfele, K. / Maier-Häfele, K. (2005):101 e-Learning Seminar Methoden. Manager Seminare Verlag

209
Critical review of the “Bluffing Game” example
What worked well in this case?
The students showed a strong interest in the game, always made their postings on
time, spared no efforts in their attempts to fool their peers and provided some very
elegant solutions when it came to the web-links leading to the correct answers. There
was a high level of activity, and the quality of some of the examples was extremely
high. The reactions and evaluations received from the students were all positive.
Students took much more time than suggested in preparing their answers.

What did not work well or could have been improved?

Students complained that there were too many postings, and that they lost the over-
view. The students were given too many books to choose from. There were no syner-
gies between the statements. Students don’t go to libraries any more, they only use
what they can find on the Internet. Intellectual activity remained more on a technical
level; the game did not make use of the social potential in the group. Students com-
plained about the amount of time they had spent on the task (an ironic complaint
given the above).

Learning effects
Obviously, another important aspect is the learning effects produced by such a game.
However, soft skills workshops of this kind often do not involve a written examin-
ation and, in this case, no research was carried out regarding its long time learning
effects. Consequently, we unfortunately can provide little in terms of measurable out-
come. However, what we do have is feedback, feedback that was even provided over a
year after the event. Some of the feedback we received on the “Bluffing Game” was as
follows:

“That was the craziest lecture I ever attended!”

“I can still remember the fake statement I posted and the
fact that it was detected by XYZ.”
“I absolutely hated playing that game, but it made me realise how easy it is to
manipulate text and therefore I think it was relevant for our lives.”

Last not least, the assessments/ratings received from students (Did you learn some-

210
thing? Did you actively participate in the seminar?, etc.) have continuously improved
since we began using collaborative blended learning methods. For us as practitioners,
this is reason enough to continue with this method of “designing” learning processes.

3.Generalisation of observations
The above examples should provide a good insight into how we actually play games
using little more than standard web-based forum technology. The following statements
draw not only on these examples but also on our many years of experience in working
in this field and reflecting this experience. They can be seen as “helpful hints and tips”
or simply as input for lecture design:

(1)Game-based learning processes can be triggered without a need for special soft-
ware. If a game can be played at a table using a pencil and paper, it can also be played
online. There are literally thousands of games that could be played in this way.

(2)When formulating the task, the rules of the game, etc. consideration should be
given both to the length of time players (students) will actually need for research,
reflection, formulating their own thoughts and giving comments and to the time they
will need for the more technical aspects, such as searches, finding their way around
the game, etc. Precise, clear, simple and easy to understand tasks and set of rules will
reduce the amount of time required for these technical aspects.

(3)The players’ (students’) focus should be directed more to the providing of relevant
answers and content than to completing the task and playing the game correctly.
The game itself is not the “content” of the exercise, it is simply the means to the
end (at least as far as the tutor is concerned). The easier the game, the greater the
role that will be played by “interesting” answers and content. Despite this, the game
should not be stupid or childish; it should be intelligent and challenging.

(4)The players should be aware of the benefits of the game, i.e. their personal benefits
or what they gain by playing. The fun of participating in the game, of competing
against one’s peers and of winning credits, etc. should not outweigh the learning
aspects and their practical benefits (use of gaming pleasure for the purpose of prom-
oting cooperative learning/working). In other words, “results” that can be expected
from the game should be clear right from the outset: the thrill of the game; credits or

211
marks for completing the task; building block towards completing the course; social
recognition in the group; self-presentation; community building; relevant, pertinent
knowledge, skills, insight; useful experience for the future.

(5)It should be clear from the design of the game/task that the assessment and the
awarding of marks by the tutor is not the ultimate purpose of the game. Instead, what
is important is the use of peer-to-peer evaluation to assess knowledge generated in the
group (Does it meet with consensus? Is it relevant? Is it of use to the individual mem-
bers of the group?) Knowledge that has been developed jointly brings the members of
the group together and they develop feelings of pride for their “baby”.

(6)Since the same game can work well in one group yet not in others, game designers
must strive to put themselves in the role/situation/conditions facing the prospective
players. There are some relatively simple tried-and-tested games that generally work
well in almost all groups (e.g. role playing games in almost all groups; or chain letter
games for predominantly male groups) and others that have to be adapted to the
needs of the respective group of students.

(7)It is advisable to offer an alternative way of contributing to the course (e.g. writing
a term paper) rather than forcing people to play the game. In this way, you not only
avoid the troublemakers, in some instances you might even encourage spoilsports to
join in after all, if they feel they have been given the choice.

(8)Even though both genders like playing games, their tastes tend to be slightly diffe-
rent: men prefer straightforward, competitive games, while women prefer games
involving social interaction and feedback loops.

(9)The greater the practical relevance of the game, the lesser the need for instructions.
Players who share a similar professional background, generally don’t need a lot of rules
to encourage interaction. What they do need is a transparent social space in which
they can share their experiences and position themselves through comments, questi-
ons, answers. Their game board—in a wider sense—is the Community or Practice
(as proposed by Lave/Wenger 1991)5 which provides ideal conditions for learning
and knowledge sharing.

5 Lave, J. / Wenger, E. (1991) Situated Learning Legitimate Peripheral Participation. Cambridge University Press.

212
4. Authors

Lotte Krisper-Ullyett
As founder and general manager of factline.com, a Vienna based company developing
innovative software for knowledge communities, Lotte got interested in the concepts
of collaborative blended learning and learning communities. Her additional occupation
as lecturer at University and continuing education programmes give her the opportunity
to experiment with innovative learning designs.

Barbara Buchegger
Despite Barbara’s academic training as an engineer at University of Natural Resources
and Applied Life Sciences in Vienna, Barbara soon discovered her fascination with the
possibilities of new internet technology and has been working in the area of virtual
communities and e-learning since 1999. Barbara is implementing collaborative e-
learning in different fields, such as education and vocational training, and has imple-
mented and developed various methods for collaborative blended learning.

Johann Ortner
After many years of conducting research and teaching Literature, Philosophy and
Social Siences at different universities in Austria, the USA and Japan, Johann worked
as General Manager for Eastern Europe in the Neurodiagnostic Instruments sector.
Johann has recently published several books and articles on knowledge management,
knowledge networks, systemic consultancy, epistemology and the philosophy of
language.

213
214
Dramaturgic E-Learning Strategy (D.E.S.)
—evaluation of a story based approach
Michael Burmester, Daniela Gerhard and Frank Thissen

Keywords
Dramaturgic e-learning, story based learning, evaluation

Abstract
The Dramaturgic E-Learning Strategy (D.E.S.) has been developed at the Stuttgart
Media University. Central assumptions of D.E.S. are that learning based on
dramaturgic stories is more engaging; facilitates intrinsic motivation and results
in positive emotions during learning. All these conditions are important to have
a positive learning experience based on e-learning technology and finally lead to
better learning results. An evaluation study was conducted with 135 students in
order to find out, if the assumptions of D.E.S. are fulfilled and to find out how
D.E.S. could be improved. No differences could be identified between D.E.S. and
the traditional approach concerning learning success and intrinsic motivation. The
general acceptance of a story-based e-learning approach was quite high. Several
possibilities for improvement of D.E.S. are identified.

1. Introduction
When implementing and conducting e-learning courses at schools, universities and
other organisations it is necessary that e-learning is accepted by all types of users, such
as learners, tutors, teachers etc. Recently the development of e-learning technology
has been the main subject of study. Increasingly the need for an effective and efficient
multimedia didactic has become clear (Beck & Sommer, 2003).

After the euphoric start of computer based and internet based education in the early
1990s, more and more disillusionment can be detected. In many cases, the promise

215
of better learning and cost reduction in training did not come true. Developing,
implementing and conducting high quality e-learning courses is often more expensive
than traditional types of training or education. The reliability of technology has not
met expectations and acceptance by learners is not very high. Reasons for the lack of
acceptance could be that learners are not experienced in self-directed learning, in self
motivation and solving technological problems on their own. Very often the learning
experience is impaired by unusable and boring e-learning applications. In reality, very
often e-learning is not a group experience; instead learners have to learn on their own.

The “Dramaturgic E-Learning Strategy” (D.E.S.) has been developed based on the
assumption that learning is an information acquisition process with emotions and
personal experiences. Emotions and immersive experiences seem to be important
in order to learn and process new learning material in a better way. D.E.S. consists
of emotional and dramaturgic elements in order to improve computer supported
learning. D.E.S. is based on scientific results of brain and emotion research (Damasio
1995, Goleman 1995, Ciompi 1997, Spitzer 1996, Ledoux 1998, Markowitsch
2002), instructional design approaches like ‘anchored instruction’ (Bransford,
Sherwood, Hasselbring, Kinzer & Williams, 1990: Cognition and Technology Group
at Vanderbilt, 1991, 1992) or goal-‘based scenarios’ (Schank, Bergmann, Macpherson,
1999) as well as modern concepts of dramaturgy (e.g. Mikunda, 2002; Breitlauch,
2003).

2. Dramaturgic E-Learning Strategy (D.E.S.)

2.1 Assumptions
Taking into account all the problems listed above, D.E.S. tries to solve some of these
problems. The starting points of D.E.S. are the following assumptions:
• Learning is an individual process of active knowledge construction. Knowledge
has to be acquired by the learner in an active way (Gerstenmaier & Mandl 1994,
Siebert, 2001).
• Multimedia does not necessarily add value. Even using multimedia all information
presented must be actively processed, e.g. in problem-based environments (Issing
& Klimsa 1995).

216
 • Learning is a social process. The communication and interaction between
learners and between learner and tutor or teacher is very important for the
learning process.
• Learning is an emotional process. Emotion and cognition are closely linked
together. Positive emotions have a positive effect on the learning process (e.g.
Csikszentmihalyi 1999) and negative emotions can hinder the learning process.
• In order to integrate new information successfully and to create a positive
emotional learning situation, stories are of central importance (digital story-
telling).
• Computers are not neutral. The medial aspect of computers causes the
fact that computers are seen as communication partner sending content-
messages and relationship-messages influencing attitude and acceptance of
users towards computers and applications (Weizenbaum 1978, Reeves & Nass
1996).

2.2 Elements of D.E.S.

2.2.1 Dramaturgic structure
The learners in a D.E.S. course can participate in stories via an e-learning platform.
This facilitates emotional participation and immersion as well as situational and problem-
based learning. The basic D.E.S. structure of drama has already been approved in
theatre and film. By dramaturgic interaction the learner can influence the story
(Breitlauch, 2003).

The development of the dramaturgical basic structure adheres to the classical Aristotelian
philosophical configuration of a three acts scheme consisting of exposition, confrontation
and solution (Aristotle, 1998). This structure is achieved through scenes and sequences
(see Figure 1).

217
 Figure 1: Classical dramaturgic structure by Gustav Freytag (2004)

When developing a dramaturgic strategy the following questions should be answered:

• What are the elements of content in order to achieve the dramaturgic
goals? E.g. love, curiosity, hedonic goals, social status, success.
• Which characters implement the main dramaturgic concept?
What are the challenges for these characters?
• Which character can be chosen for the antagonist, who will work
against the dramaturgic goal of the story?
• How can the conflict of the character and the antagonist be solved?

2.2.2 Myths
For the development of dramaturgic structures not only authentic situations can be
used, but myths, stories and narratives are of interest as well. Myths and fairy tales
may have an additional advantage for remembering information: the dramaturgic
structure is already established in the knowledge of the learner, because they have
heard them several times during their life. The story to be developed can make use
of this knowledge. An example of a well known dramaturgic structure is the myth
of David and Goliath. The dramaturgic structure is conflict between small weak David
against tall strong Goliath.

218
2.2.3 Cryptic Knowledge
The acquirement of knowledge is not solely based on collecting facts and information,
but is integrated in an active process between the educator and the learner. The “cryptic
knowledge method” structures the acquirement of knowledge in such a manner, that
the learner must gain the knowledge through learning activities. Cryptic knowledge is
an element of the dramaturgical basic structure and can for example contain new, cur-
rent and exclusive information, which the learner receives through accomplishing
a certain task. Cryptic knowledge also includes among other things rumours, incorrect
information, symbols, etc. (Mikunda, 2002).

The “cryptic knowledge method” includes the realization, that during the conveyance
of knowledge elements, so called conscious and unconscious functions such as brain
scripts or cognitive maps are activated within the learner. Brain scripts are “screenplays
in the head” of the learner, which are responsible for what the learner actually under-
stands is really happening in a story. Cognitive maps are the “internal maps of
consciousness” of humans and support orientation. Accordingly, they are developed
within a dramaturgical basic structure, which motivate the learner to interactivity,
through his/her own concern (Prensky, 2001).

2.2.4 Community
The integration of knowledge acquirement into a community conveys a contextually
related, communicative social transfer of knowledge. The community contains different
stages (levels) which the participant can reach through accquiring certain elements
of knowledge. Within the initial stage, it is the participant as learner himself. Within a
continuative course, the learner can become the tutor or coach who takes over the
tutoring or coaching of others on account of his/her experience (Kim, 2001).

2.2.5 Multimedia information and communication system

In order to complete the learning environment of D.E.S. a knowledge pool is required.
Here the learner gets background information, case studies, facts, news-ticker,
discussion forums and other media. If the learner has to solve tasks embedded in the
dramaturgic story she or he needs the knowledge pool. Figure 2 shows the integration
of knowledge pool as hypermedia and the dramaturgic structure.

219
 Figure 2: Integration of dramaturgic structure and hypermedia as knowledge pool

2.2.6 Five steps of D.E.S.

Electronic media such as film, audio, video, etc. can induce multifaceted emotional
responses, but are in themselves emotionally neutral. The control of emotional
responses therefore follows the special supportive structure of dramaturgy. The
method uses a dramaturgical concept to create electronic learning modules. The
development of a dramaturgical e-learning strategy according to the method is based
on the following implementation steps:
• Step 1: Predefinition of the target group for learning (multiplicators,
learners, etc.)
• Step 2: Ascertainment of the primary themes (basic knowledge in the
form of hypertext modules)
• Step 3: Development of a dramaturgy (goal-solution-conflict, strategic
elements such as exposition, confrontation, solution, plot-points, etc.)
• Step 4: Determination of exclusive cryptic knowledge (mythological
knowledge, rituals, proverbs, etc.)
• Step 5: Community building (levels)

220
3. Study questions
The study was necessary due to several reasons. First of all, D.E.S. is a newly developed
e-learning approach. Some pilot studies have been done, but without a systematic
empirical evaluation. Therefore, one main objective was to evaluate the D.E.S. in a
university course and to find out if D.E.S leads to the promised results and to im-
prove the D.E.S. This was clearly a formative evaluation objective.

During the development and the scientific discussion concerning D.E.S. several ques-
tions came up, which should also be answered in the planned empirical study setting.
The following questions were planed to be investigated:
1. Do learners experience more positive emotions when learning with D.E.S.?
2. Does D.E.S. facilitate intrinsic motivation?
3. Is there a general acceptance of story based e-learning seminars?
4. Is learning mor successful with D.E.S.?

4. Method
4.1 Study Design
At the Stuttgart Media University a screen design course was planed for 135 students
in winter semester 2004/2005. The topic screen design comprises content concerning
communication via internet sites, definition of target groups, layout, typography, use
of colours and forms, accessibility, screen elements (orientation, navigation, motivation,
content, interaction) as well as design of content such as text design, photographs,
animations, and graphics. The students came from the information design and business
information course. All students studied in their third semester.

The students were divided into two groups:

• Classic Group
This group learned with a traditional e-learning seminar style. They got tasks in
a weekly time schedule. They had to work on the tasks and to upload the results.
They got support by tutors and a teacher via the open source course management
system MOODLE (www.moodle.org, see Figure 3). 70 students participated in
this group

221
 • D.E.S. Group
The students of the D.E.S. group learned in the framework of a (dramatic)
story. The number of students in this group was 65.

There were several similarities between the groups. Both groups

• worked with the same hypermedia containing the screen design knowledge.
• worked with the same tasks contents. Only the presentation of the tasks was
different, e.g. embedded in a story in case of group D and formulated as an
ordinary task in the case of group C.
• had the option to ask questions and to contact a lecturer.
• were divided in working groups of two or three students.
• studied in the same semester starting at the beginning of October 2004 and
finishing at the end of January 2005.

Figure 3: Screen shot of the group C course management system presenting the weekly tasks

222
4.2 The story of the D.E.S. group
4.2.1 Story development
According to the method elements and the five steps of D.E.S. development, a
dramaturgic concept and a story was developed. A team of students together with
a professor developed the complete concept.

4.2.2 The mythological basis

The myth that the story is based on is the Cinderella story: an old story that goes
back to Egyptian times and has also some roots in 8th century China, the Romans
and in North American Indian cultures. The idea is that a very badly treated heroin
wins at the end and with the help of a good character (the fairy) the poor girl finds
the handsome prince and becomes a princess. But it’s a hard and long way.

4.2.3 The story

The story is the one of a small designer’s agency run by three people working together.
One of them—Thomas Jung—is an expert for print design and typography, his
colleague Pierre is the web site expert and the secretary Lisa runs the office. They are
not rich but they survive. Suddenly a big telecommunication company gives them a
commission for a web-site development with a very large budget. In this situation
Pierre as the screen design expert decides to work for the company on his own. So he
leaves Thomas—our hero—and opens a new office with Lisa. Thomas tries to keep
the business running—there are some customers with small commissions, but being
on his own without Pierre’s expertise and the secretary Lisa is a very hard job. And
Thomas is crazy about Lisa, which now works closely with his new enemy Pierre. In
this bad situation a mechanical engineering company asks Thomas to create a we
site for them. He needs the money, but he has no idea how to design web-sites. He
remembers a professor at Stuttgart Media University—Frank Thissen—who is an
expert in screen design and sends him an e-mail asking for help. The professor sends
the e-mail to his seminar students asking them to help Thomas Jung. The students
become Thomas Jung´s consultants. For this task they can access information from
a hypermedia system which presents the main information from the “Screen Design
Manual” (Thissen, 2003).

223
Over the following weeks Thomas Jung gets various tasks from his customers as “Our
web site has to be accessible for handicapped people. Please check it” (accessible web
sites), or “Our home page seems not to attract our customers, what can we do?” (target
groups and emotional aspects of web design), and others. With the help of the students
who do some investigation and research Thomas Jung masters his new challenges and
finally wins. Beside these struggles with his job he’s fighting for Lisa, so a love story
is running simultaneously.

The dramaturgic story was told by the following means:

• Thomas Jung´s e-mails to the students (and sometimes also e-mails from
his rival Pierre). According to the events and plot-points tasks were
embedded in the story and communicated by mail to the students.
• Company web sites produced by Thomas Jung that are changed during the
seminar.
• The online diary: all personal information about Thomas Jung was told by
using a public diary. Here, all his personal feelings are described (see Figure 4).
• Postcards that were sent from Thomas Jung to the students.

Several plot-points were integrated in the story, such as

• Thomas Jung’s love for Lise who leaves him
• A big company offers a big job to Thomas Jung, but he is not able
to do it on his own
• His most important customer is not satisfied by his work results.

The student partici-

pants were integrated
in the story as sup-
porting consultants.
All teams were asked
to act as a coach for
screen design in order
to help Thomas Jung
to be more successful.

Figure 4: Online diary of Thomas Jung (‘squares design’ is the company of Thomas Jung)

224
4.2.4 Cryptic knowledge
The concept of cryptic knowledge was realized by rumours that ran among the students
groups and gave the seminar a bit of mystery. Although we told the students at the
beginning that it was a game and the people in it are unreal, the students started
discussing the backgrounds of the story. Who wrote the online diary? Was it the
professor? Or who else was it? And were Thomas Jung, Pierre and Lisa really fictional
characters? Who sent the post cards? And weren’t the web sites real?

The online diary was not publicized, but hinted about to some students, who told
the rest of the class. (The interesting thing is that although the story in the online
diary was fictional, people from outside the university who read the diary commented
on it and gave Thomas Jung advice. So we ran into an ethical question whether it is
acceptable to tell a fictional story in a public online diary).

4.3 Research instruments

4.3.1 Semi-structured interviews
Two types of methods were used in the study. In order to learn more about the
acceptance of a story-based approach of learning and to find opportunities for
improvement a qualitative semi structured interview format (e.g. Merton & Kendall,
1979) was chosen. A question guideline with open questions was developed.
Questions were formulated in order to get information on the following themes.
1. What can be remembered of the story?
2. Is the story believable?
3. What is the opinion on different elements of the story, e.g. type of story telling
(e.mail, online diary etc.), the hero, the other characters, and dramaturgy?
4. Learning success
5. Emotional reactions
6. Hypermedia with screen design content
7. acceptance of story based e-learning seminar

For the first round of interviews six participants of the dramaturgic screen design
seminar (group D) were randomly chosen. After all interviews were conducted they
were analysed following a qualitative analysis according to grounded theory (Strauss
& Corbin, 1996) and fast text analysis (Legewie, 1994).

225
After having the first round of interviews analysed a second round again with six
participants was started. Based on the first interview results and reflections, the
interview guidelines were adapted and new interview themes were added, e.g. more
questions on the diary and the parallel running love story, new story telling elements
such as a Christmas postcard to all students from Thomas Jung and further themes
like the personal importance of screen design for the students and their course of
studies and the role of the teacher as university professor and as part of the story.
The second round of interviews were analysed the same way as the first round.

4.3.2 Questionnaire
From the interview results, theoretical assumptions as listed above, and theories
of intrinsic motivation (Intrinsic Motivation Inventory, IMI, Deci & Ryan, 2000)
statements were derived for the questionnaire. The participants were asked to rate the
statements on the basis of a Likert scale. Figure 5 shows an example of a statement
with a rating scale. The rating scale is translated from German. The German version
is an approved scale (Rohrmann, 1978, cit. from Bortz & Döring, 2002).

The approach to present seminar content as a story is very good.

absolutely true (5) quite true (4) fifty—fifty (3) is less true (2) absolutely not true (1) no answer

Figure 5: Example of a question in the questionnaire (rating scale is translated from German).

The questionnaire consisted of two parts. The first part contained questions for both
groups . The second part was just for the D.E.S.group .

The first part of the questionnaire for both groups contained 22 questions concerning
the following themes:
1. Learning success
2. Positive emotions (e.g. fun) and negative emotions (e.g. frustration)
3. Interest towards screen design
4. Group work

226
The second part of the questionnaire for the D.E.S. group contained 28 questions
for the following themes:
1. Questions concerning the story
2. Questions on the virtual seminar
3. Degree of participation and involvement
4. Acceptance of the seminar type

The main question type in both parts was statements with a rating scale. Only a few
questions were open with the possibility of free textual answers.

In order to answer the study questions 1, 2 and 3 the results were statistically analysed
by t-tests using the statistical software SPSS (2002).

4.3.3 Knowledge test

In order to get a rough assessment of the learning success a knowledge test was
developed and presented to the participants at the end of the seminar. It was the
same test for both groups.

The test consists of the two parts:

• Factual knowledge
These are questions concerning theories (e.g. design theory of Gui Bonsiepe),
methods (e.g. persona concept according to Alan Cooper), psychology (e.g.
“what is the capacity of the short term memory?”), use of colours, text writing
for internet pages, accessibility etc.
• Learning transfer
Four web-pages were presented and the students asked to write down optimization
possibilities from the perspective of screen design.

All tasks have a maximum number of points. The time to fill out the knowledge test
was 80 minutes. For all students a professor assessed the results.

The students had to participate in the knowledge test, but the result was not relevant
for their marks.

227
4.3.4 Complete study schedule
The study started at October, 15th 2004 with a preliminary investigation:
• Test of factual knowledge concerning screen design before the beginning of the
virtual seminar to make sure, that the student didn’t have already good knowledge
in screen design before participating in the seminar.
• Questions towards the interest in screen design to make sure that—in case of
good learning results—it wasn’t the students’ interest that lead to this success,
because high interest correlates with better learning results.

The seminar started on October 15th 2004 and finished January 28th 2005. The
qualitative interviews accompanied the seminar.

On January 28th 2005 the final investigation was conducted:

• knowledge test (knowledge of facts; learning transfer)
• questionnaires for formative and summative evaluation (questionnaire as
described above)
Not all data can be presented in this paper.

5. Results
5.1 Qualitative Analysis of interviews
The interviews were analysed on the basis of a qualitative analysis. On the basis of the
analysis, several themes and issues were identified. It is important to understand that
these themes and issues are a variety of possible reactions and opinions towards the
story-based e-learning seminar. Based on this analysis, questions for the quantitative
questionnaire were derived.
The following themes and issues were identified:

5.1.1 Emotional reactions

Several emotional reactions were identified. It was fun to learn in the context of a
story and an e-learning environment. Curiosity and interest were stimulated by the
story and the content screen design as well. Frustration was closely related to deficits

228
in the organisation (group forming, technological problems, and instructions) of the
seminar. Some plot-points were related with excitement.

5.1.2 Entertainment
For students who love soap operas and students without time pressure from other
courses, the story could be rather entertaining “it was my daily soap”. The seminar
had a feeling of leisure time and relaxing from other lectures.

5.1.3 Time pressure

Some students were involved in other very time consuming seminars. Their interest
in the story decreased and the e-mails generated in the framework of the story were
just analysed in order to find the task related information. The main goal for these
students was to finish the task efficiently. In this case the story was annoying.

5.1.4 Believability of the story

Students thought that it was not very believable that a web-designer running a small
company was so helplessness,
a. that he contacted a university to get help from students,
b. that the web-designer tells his very personal love live and all his troubles with
his business partner to a large group of people,
c. that his character was more like a student than a business man, and
d. that there was not enough explanation why he was so alone without any friends.

5.1.5 Reality and fiction

At the beginning of the seminar the students were informed that they would learn by
a story and this story is just fiction. In order to make it a little bit more mysterious
the story was told in a very realistic way by real media (e.g. e.mails, public diary,
paper mail). Some students thought in the first third of the seminar that it might be
real. After that they found out that it wasn’t. This was a little bit disappointing. The

229
curiosity effect of the realistic story-telling elements was quite obvious. After this first
phase the curiosity effect disappeared. Some students asked for a real project and they
said that that would be highly motivating.

5.1.6 Consistency of the story

A break in the consistency of the story was a real “show stopper”. Two reasons for
that can be identified. First, an e-mail from the professor advising the students to
work carefully and deliver the results on time was perceived as a break in the story.
Second, as soon as it was clear that the students would get marks for their work, the
story was not that interesting anymore. It was clear from that point on, that it was
not fun, it was work. Furthermore, feedback for the delivered solutions is very important
for students in order to assess their performance, but this feedback must then be part
of the story.

5.1.7 Learning success

According to the subjective assessment of the interview subjects, they asserted that they
learned interesting facts about screen design and that this type of seminar was more
interesting than a classical lecture. Because the students had to work on problems
integrated in the story, some students preferred the practical aspect of the seminar
content. But some students said, that the overall standard of the seminar was not high
enough. This must be interpreted in relation to the fact, that about one third of the
students were information design students in the third semester. They already knew
a lot about interface design and usability.

5.1.8 Elements of D.E.S.

The hypermedia system presenting the screen design content was well accepted and
seen as an important part of the seminar. The only critique was that it should be better
integrated in the story (see also “consistency of the story”). The e-mails were well
accepted as means of the story-telling. The public diary, the company web-site and
the postcards were accepted by the students who appreciated the story approach. Some
suggestions like video clips and more interactive computer game like elements were
made.

230
5.2 Quantitative Analysis of questionnaire
5.2.1 Positive Emotions
In order to answer the question whether learners experienced more positive emotions
with the D.E.S. seminar (group D) compared to the traditional virtual seminar
(group C), the participants were asked about fun. The question statement was “this
seminar was more fun than a usual seminar”. Figure 6 shows the results of the mean
ratings. The story based virtual seminar is statistically significant higher (.05 level of
significance, t-test statistic) rated than the traditional virtual seminar. The agreement
level on the five point scale (see Figure 5) for 2 is “is less true” and for 3 “fifty—fifty”.

Figure 7: mean ratings of agreement with the statement “ This seminar was more fun
than an usual seminar” or group C and D

5.2.2 Motivation
One main purpose of D.E.S. is to facilitate intrinsic motivation in order to learn via
e-learning seminars. The question is whether the intrinsic motivation is supported by
story based virtual seminars compared to traditional virtual seminars.

231
Three questions were taken from the Intrinsic Motivation Inventory (Deci & Ryan,
2000) and adapted to this study. Table 1 shows these questions. There is no significant
difference (.05 level of significance, t-test statistic) between the two groups for all
three questions. The mean rating category of agreement for question statements 1
“the seminar motivated me more for independent working than an usual seminar”
and 2 “This type of seminar encourages more intensive examination of the learning
content than an usual seminar” in both groups is “fifty—fifty”. Statement 3 “this
type of seminar arouses interest in other material” was rated as “is less true” in both
groups.

Nr. Statements on learning motivation C* D* Significance

1. The seminar motivated me more for in-
dependent working than an usual seminar 3.07 2.98 Not significant
2. This type of seminar encourages more
intensive examination of the learning
content than an usual seminar 2.97 2.61 Not significant
3. This type of seminar arouses interest
in other material 2.38 2.37 Not significant
* mean rating of the question statements
Table 1: Statements on learning motivation adapted from intrinsic motivation inventory (Deci & Ryan, 2000)

5.2.3 General Acceptance

Whether story-based e-learning seminars are accepted was another important question.
Therefore, question statements were defined to assess the acceptance of the D.E.S.
approach for virtual seminars.

Two question statements asked for a rating of general acceptance of story based
virtual seminars.

The first statement was “I think it is a good idea to present the learning content of
a seminar as a story”. The result was, that 63% of the group D participants rated
this as “absolutely true” and “quite true” (median is 4 “quite true”). Figure 7 shows
the distribution of the rating result.

232
 Figure 8: Percentage of the rating categories 1 to 5 (see Figure 5 as well) in group D for the
statement “I think it is a good idea to present the learning content of a seminar as a story“

The second question statement on acceptance was “story and learning do not fit”.
This statement is a negative formulation concerning the fit of the use of a story for
learning purposes. The result was that 66% rated this statement as “less true” and
“absolutely not true” (median is 2 “less true”). Figure 8 shows the distribution of the
ratings.

Figure 9: Percentage of the rating categories 1 to 5 (see Figure 5 as well) in group D for the
statement “Story and learning do not fit”

233
5.2.4 Acceptance related to the implementation of the story based seminar
After having tested the general acceptance of story based seminars, it was important
to check how the quality of implementation of the story-based e-learning seminar was
rated.

The questionnaire contained three statements concerning the implementation of

the story.

The first question statement was “if the story is well implemented I would enjoy a
story-based seminar”. Based on results of the qualitative interviews, we suspected
that the implementation of the story was not very good. The purpose of this statement
was to separate the actual implementation from the general acceptance. 70% of the
participants rated this statement as “absolutely true” and “quite true” (median is 4
“quite true”). The percentages of the categories according to the statement are shown
in figure 9.

Figure 10: Percentage of the rating categories 1 to 5 (see Figure 5 as well) in group D for the
statement “If the story is well implemented I would enjoy a story-based seminar”

The second statement focused on the excitement of the story. It was “the story
was exciting”. The result shows that 68% of the participants in group D rated this
statement as “less true” and “absolutely not true” (median is 2 “less true”).

234
 Figure 11: Percentage of the rating categories 1 to 5 (see Figure 5 as well) in group D for the
statement “the story was exciting”

In order to bring some mystery to the story, the story was told in a very realistic way
so that sometimes it was not clear what is real and what is fiction. At the beginning
of the seminar all participants of group D were informed, that everything in the story
was fiction. But after that, the participants of the seminar were integrated into the story
as e.g. consultants of Thomas Jung etc. Therefore, it was an important question,
whether the story is believable or not. The statement “the story was believable” was
rated as “less true” and “absolutely not true” by 72% percent of the participants of
group D (median is 2 “less true”).

Figure 12: Percentage of the rating categories 1 to 5 (see Figure 5 as well) in group D for the
statement “the story was believable”

235
5.2.5 Learning success
A first assessment of the learning success in both groups can be done on the basis of
the knowledge test described above. 64 students of the Classic group and 50 students
of the D.E.S. group participated in the knowledge test. The knowledge test scores
were compared on the basis of an average comparison (t-test statistic). The result was
that there was no significant difference between the mean knowledge test scores of
both groups (.05 level of significance, t-test statistic). Figure 6 shows the mean scores

Figure 6: mean scores of the knowledge test for Classic (C) and D.E.S (D) groups

6.0 Discussion
6.1 Positive Emotions
There was a significant difference between the two seminar groups. The story-based
seminars shows a higher score of fun than the traditional seminar. The overall value
is not very high. The agreement of the participants to the fun statement (see Figure 6)
is just “fifty—fifty”. This is not very high. Here it becomes clear that several improve-
ments of the D.E.S. method and its implementation in seminars are necessary.

236
6.2 Motivation and acceptance
On the one hand we saw that there was no significant difference between the groups
concerning the motivation statements. The conclusion is, that there is not motivational
advantage of the story-based approach compared with the traditional approach. In
both groups the level of motivation is just medium, which is not convincing as well.
On the other hand we find a clear acceptance of the story based approach (see Figure
7 and Figure 8). One reason of the small value of motivation might be that the story
was not implemented very well. This can be seen by the agreement to the statements
“if the story is well implemented I would enjoy a story-based seminar” (Figure 9) and
the disagreement with the statements “the story was exciting” (Figure 10) and “the
story was believable” (Figure 11). This is supported by the qualitative results as well,
since the curiosity died when the fictional character became obvious. Furthermore,
the story and its main character was not seen as believable. The inconstancies
between a playful story-based approach and performance rating by marks hinder
intrinsic motivation.

Also from a theoretical standpoint some explanations concerning the low intrinsic
motivation level can be found. According to the self-determination theory of Deci
& Ryan (2000) three requirements for intrinsic motivation are important:
1. Learners need to have the opportunity to perceive themselves as competent
and to feel self-efficient.
2. Learners must have the opportunity to determine their learning by themselves.
3. Learners must learn together with others (social involvement)

Social involvement was part of our setting, but regarding the two other points we can
say, that there were deficits in the story-based seminar. The deficits for requirement
one were, that there was a lack of feedback from the lecturer and there was no possi-
bility to influence the story. The students just could give advice as consultants, but
not affect on the story line. Furthermore, for requirement two of the following
deficits could be identified. First of all, the story and the tasks for students followed
a weekly time pattern. Therefore, there was not enough freedom in organizing the
work. Second, the learning goals and the content to be learned were predefined by the
tasks and the story. Students had little opportunity to follow their own interests. Third,
they worked in predefined working groups, so they could not choose their learning
partners.

237
6.3 Knowledge test
The result of the knowledge test is not very encouraging. First of all, there is no
significant difference between the mean knowledge test scores of group C and group
D. Second, the overall level of the test score is not very good. The maximum score
was 52 points and both groups reached around 18 points. A reason for the overall
level is that the knowledge test was not relevant for the student’s marks. Therefore,
the motivation to learn for the test or to work hard on the test might be quite low.

Obviously, concerning the learning success the story-based approach has no advantage
over the traditional approach. Possible explanations will be discussed in the context of
motivation and acceptance.

6.4 Conclusions
From the results the following conclusions can be drawn:
• Story-based virtual seminars are accepted on a university level.
• The story-based e-learning seminars should give the students more freedom
in organising their learning and following their own interests.
• The whole dramaturgic seminar must present all elements in a consistent
way. This means that there must be no breaks between story, content (e.g.
hypermedia system), seminar interaction (co-operation of students as well
as co-operation between students and story characters) and communication
(e.g. feedback from the teachers or tutors). Everything must be integrated in
a dramaturgic structure of a convincing story.
• The dramaturgic seminar must fit in the organizational and administrative
structure of a university, e.g. performance assessment is not really compatible
with a playful approach towards story-based learning.
• The story must be appropriate for the target group and must have a professional
dramaturgic structure.

Future studies must show, that by better implementation of the story-based e-learning
approach, the learning success is better than for other e-learning seminars.

238
7. Acknowledgement
A special thank you to all students who were involved in that study as participants of
the seminar groups C and D as well as story writers and interviewers. Thank you very
much for all the engagement and a lot of work done.

8. Authors

Michael Burmester teaches as professor for ergonomics and usability and is head of
the usability laboratory at the Stuttgart Media University, Germany. Furthermore,
he is responsible for research and innovation at the usability consultancy User Inter-
face Design GmbH in Ludwigsburg. Before that, he was researcher at University of
Regensburg, Fraunhofer Institute for Industrial Engineering (IAO) in Stuttgart and
Siemens AG Corporate Technology—User Interface Design in Munich. In 2000 he
founded the Munich office of the User Interface Design GmbH and was head of the
Usability Engineering department. His current research focus is on emotional aspects
of human computer interaction, advanced methods for user centred design, eye-tracking,
usability for e-learning applications, persuasive technology, and design of devices and
information appliances.

Daniela Gerhard is research assistant and lecturer at the University of Landau,

Department for educational science. Before this she worked at the Stuttgart Media
University on E-Learning research projects funded by the EU and at the University
of Karlsruhe.

Frank Thissen is an expert in multimedia and online learning and does research and
publications on (intercultural) e-learning. He has vast experience of undertaking a
similar exercise in the last few years, in doing online seminars with international
partners (Anadolu-University, Turkey, Salford University, Open University New

239
Delhi, Universidad de las Americas Puebla, Mexico) and in teaching, multimedia and
communication technologies and information design. Furthermore, he has large
expertise in web site usability and has carried out a research project on emotions
and online learning. Prof. Thissen also worked on usability and screen design, he
published a screen design manual at Spring

9. References

Aristotle (1998). Poetics. Chicago: University Of Chicago Press

Beck, U & Sommer, W. (Eds.) (2003). LearnTec. Tagungsband. Karlsruhe. KMK.

Bortz, J. & Döring , N. (2002). Forschungsmethoden und Evaluation: für

Human- und Sozialwissenschaftler. Berlin: Springer.

Bransford, J. D. Sherwood, R. D., Hasselbring, T. S., Kinzer, Charles K. Williams,

S. M. (1990). Anchored Instructions: Why we need it and how technology can
help. In: D. Nix & R. Spiro (Eds.), Cognition, Education and Multimedia-
Exploring ideas in high technology (pp.115-141). Hillsdale, New Jersey: Erlbaum.

Breitlauch, L. (2003): Interaktive Dramaturgie. Non published project report.

Ciompi, L. (1997). Die emotionalen Grundlagen des Denkens. Göttingen:

Vandenhoeck.

Cognition and Technology Group at Vanderbilt (1991). Technology and the Design
of Generative Learning Environments. Educational Technology, 34 – 40.

Cognition and Technology Group at Vanderbilt (1992): The Jasper series as an

example of anchored instruction: Theory, program, description, and assessment
data. In: Educational Psychologist 27 (3), 291-315.

Csikszentmihalyi, M. (1991). Flow: The Psychology of Optimal Experience. New

York: Harper Perennial

Damasio, A.R. (2005). Descartes’ Error : Emotion, Reason, and the Human Brain

240
 (Paperback). London: Penguin

Deci, E.L. & Ryan, R.M. (2000). The what and why of goal pursuits: Human needs
and self-determination of behaviaor. Psychological Inquiry, 11, S. 227- 268.

Freytag, G. (2004). Technique of the Drama: An Exposition of Dramatic

Composition and Art. Seattle, University Press of the Pacific.

Gerstenmaier, J. & Mandl, H. (1994). Wissenserwerb unter konstruktivistischer

Perspektive. Foschungsbericht Nr. 33. Ludwig-Maximilians-Universität München.

Goleman, D. (1997). Emotional Intelligence : Why It Can Matter More Than IQ.
New York: Bantam

Issing, P. & Ludwig J. (1995). Information und Lernen mit Multimedia.

Weinheim: Beltz.

Kim, A. J. (2000). Community Building on the Web : Secret Strategies for Successful
Online Communities. Berkeley: Peachpit Press

Ledoux J. (1998). The Emotional Brain: The Mysterious Underpinnings of

Emotional Life. Austin: Simon & Schuster

Legewie, H. (1994). Globalauswertung von Dokumenten. In: A. Boehm, A. Mengel

& T. Muhr (Hrsg.), Texte verstehen—Konzepte, Methoden, Werkzeuge (S. 177-182).
Konstanz: Universitätsverlag Konstanz.

Markowitsch, H. J. (2002). Dem Gedächtnis auf der Spur. Darmstadt: Wiss.

Buchgesellschaft.

Merton, R.K. & Kendall, P.L. (1979). Das fokussierte Interview. In: C. Hopf und E.
Weingarten (Hrsg.), Qualitative Sozialforschung (S. 171-204). Stuttgart: Klett.

Mikunda, C. (2004). Brand Lands, Hot Spots & Cool Spaces: Welcome to the Third
Place and the Total Marketing Experience. London: Kogan Page

Prensky, M. (2001). Digital Game-Based Learning. New York: McGraw-Hill.

Reeves, B. & Nass, C. (1996). The Media Equation. Cambridge: CSLI Publ.

241
Schank, R., Berman, T.R. & Macpherson, K.A. (1999). Learning by Doing. In: C.M.
Reigeluth (Ed.), Instructional-Design Theories and Models, Lawrence Erlbaum,
New Jersey, S. 160-181.

Siebert, H. (2001). Selbstgesteuertes Lernen und Lernberatung. Neuwied:

Luchterhand.

Spitzer, M. (1996). Geist im Netz. Heidelberg: Spektrum.

SPSS (2002). SPSS for Windows. Version 11.5.1. Computer Software. LEAD
TOOLS.

Strauss, A. L. & Corbin, J. M. (1996). Grounded Theory: Grundlagen Qualitativer

Sozialforschung. Beltz PVU.

Thissen, F.(2003). Screen Design Manual: Communicating Effectively Through

Multimedia. New York: Springer.

Weizenbaum, J. (1976). Computer Power and Human Reason: From Judgement to

Calculation. New York: W.H. Freeman & Company

242
243
244
Quality in eLearning and quality of
learning games
Claudio Dondi and Michela Moretti

Keywords
Quality, eLearning, learning games, quality stamp, criteria

Abstract
This paper is focused on the quality of learning and teaching processes by new
technology and methods such as eLearning and game-based learning. It is
structured in four main parts:
The first one is aimed at briefly presenting the results of the ongoing debate on
the quality of eLearning at European level. The second is focused on showing how
the game-based learning can answer the needs of the Knowledge Society; the third
illustrates a framework for assessing the quality of digital learning games developed
within a European funded project. The last part concludes the contribution by
presenting some relationships amongst eLearning and learning game sectors and
some final remarks.

1. Quality in eLearning in the European scenarios

Information and communication technology (ICT) offers many advantages to the
educational sector not least by providing an endless range of formats and mediums
for communication and creation/exchange of large amounts of information and
knowledge. However, despite enormous potential there is an ever present need for
quality and reliability in teaching and learning services provided by eLearning ex-
periences. This need relates to both the users’ requirements and also to ensure its
acceptance among the educational and training communities as a whole.

245
It must be remembered that distance learning (the predecessor of eLearning) was
originally conceived as a “second chance” approach to education that resulted, almost
inevitably, that it would also considered a “second choice” solution in the eyes of most
learners and teachers. The evolution of technology together with increased demand
and supply has made it abundantly clear, to even the most orthodox academics, that
eLearning is not necessarily associated with “poor” quality. There is plentiful evidence
that through individualization, internationalization, flexibility to address the needs
of users, and the ability to provide professional community building, eLearning
can produce better quality results (at least as perceived by users) than conventional
teaching, in which these characteristics are not usually found. Nevertheless quality
in eLearning is not always assured and agreed. eLearning product/service providers
find themselves in a competitive and constantly changing commercial environment
where the final judgement of the suitability, meaningfulness, relevance and quality of
the services provided is usually determined by the end user. For those involved with
eLearning in the educational/training sector, this setting differs greatly from that of
traditional education where the learners are provided with a learning experience on
which they are unable to exert any great degree of control (Scienter, 1998).

ELearning educators have now to be concerned with the learners’ specific needs and
requirements and the quality of the service provided, which ultimately determines the
success, uptake, and development of any particular programme or course. ELearners
will not choose or continue to use a specific course or programme if it fails to meet
their needs. The learners are unlikely to choose to be involved in a eLearning ex-
perience again if their first exposure has been negative. This happened in many cases
and contributed to the slow eLearning market growth and related expectation of
investors. So quality, (or better the perceived lack of quality) has been one very strong
inhibiting factor for eLearning adoption.

The increased specialization and technological complexity of today’s world presents

many challenges. In general people have the opportunity to be more informed and
to that end learners are generally more aware of the specific requirements they need.
Learners take up courses with specific objectives in mind, e.g. licensing examinations,
specific job functions, new company skills/ requirements etc. Distance educators must
attract and engage with the end user; otherwise they will fail to survive in this com-
mercially competitive and learner-driven environment.

246
With the emergence of new forms of learning in the so-called Knowledge Society and
the pervasive character of ICT in most of the education and training systems, quality,
in its broad understanding, is becoming the catalyst for excellence, openness, and in-
creased opportunities for learning in our societies. In this context, quality in eLearning
is going to be a long-standing aim for the social stakeholders and especially education
and training institutions, in their quest to sustain learning opportunities for real added
value for European citizens (Boonem & Petegem, 2005).

As it has emerged from the analysis of the European scenarios; different initiatives,
institutions, and groups in education and training environments have designed and
delivered tools, to assure or to promote a culture of the quality in eLearning (Dondi
& Moretti, 2005).

Some key elements of difference can be underlined:

Quality is a concern for all the stakeholders.

Without quality, learners may not achieve specified objectives, fail licensing examina-
tions, perform poorly on critical job functions, leave dissatisfied with their experiences
and tell others to avoid specific courses or programs. Without quality, educators may
find it difficult to attract and retain learners, experience difficulties facilitating and
managing the e-learning process, receive poor student evaluations and fail to demon-
strate an adequate return on investment. High quality programs are also necessary to
demonstrate that e-learning is a legitimate form of training and education.

The concept of quality is a complex concept.

The concept of quality applied to eLearning is a complex one and must always take
into account plural perspectives: there cannot exist a unique vision (one size fits all)
of quality that can be considered the best one. This implies that quality has a “sub-
jective”, “contextual” and “objective” component. The latter involves defining a set
of common criteria by which the quality of e-learning can be assessed, but it is certain
that the weight or significance that different stakeholders attribute to the quality criteria
will differ greatly. Therefore quality can be only derived from a process of negotiation,
mediation and agreement among the different stakeholders.

247
The way that stakeholders approach and see quality in eLearning
is different.
The difference emerges not only between education and industry sectors, but among
the stakeholders belonging to the same sector. In particular, if we look at the industry
and educational approaches we can note that: increasing quality from an industry per-
spective means reducing variance around set standards. Industry standards are defined
primarily to assure the technical quality, reusability and interoperability (or sharing)
of much smaller units of instruction called learning objects, rather than courses and
degree programs. The primary difference is that education guidelines focus on the
quality of e-learning courses and programs, whereas industry standards concentrate on
the technical quality, reusability and interoperability of learning objects. Nevertheless,
if you look at the experiences carried out with the education sector, it can be noted
that the emphasis given to the different eLearning elements is not homogeneous.
Quality seems to be in the eye of the beholder.

Quality varies tremendously within the eLearning courses and

training programs.
Despite efforts and existing quality tools and approaches the delivered and perceived
quality of the initiatives (course, materials/resources) is very different. There is a lack
of understanding of what exactly is meant by the “quality of learning” (let alone in
relation to eLearning). In this respect not all Learners have reached a stage where they
are capable of defining concisely their quality needs or requirements or providing an
explicit statement regarding learning quality. In order to be able to ask for a quality
eLearning experience, users must be in the position to recognise what constitutes
quality (or indeed the lack thereof). In other terms they need to have “quality literacy”.
Recognizing quality involves a raised awareness of what learning implies and also an
increased familiarity and knowledge of what an eLearning course should provide and
what learners should ask for SEEQUEL (2005).

The focus of these quality initiatives/approaches/framework diverge

It ranges from eLearning materials to services, from the user perspective to the institu-
tional perspective with some specific actions to the regional and societal perspective.

248
2. Game based learning and the Knowledge Society
The Lisbon European Council (March 2000) set the European Union the strategic
goal of becoming the most competitive and dynamic knowledge-based society in the
world (European Parliament, 2000). This goal was further reaffirmed at the Barcelona
European Council in (March 2002) which provided a mandate to make European
education and training a world-wide reference by 2010 (Barcelona European Council,
2002). Teachers and educators in general are facing greater changes in their job/pro-
fession due to the increase complexity and demands from the society and as well as
the changing needs of students’ and their different attitudes toward learning and
knowledge acquisition as a whole. Taking into consideration these demands, teachers
are required to promote five main pillars of education:
1) learning to know (it is intended that students discover and master
knowledge-seeking skills that will serve them for the rest of their lives),
2) learning to do (the challenge facing education today is to find methods that
will succeed in transforming school knowledge into practical competence),
3) learning to live together (it is a means of facing the challenges presented by
social, economic and political globalization of the world and requires that
citizens have the capacity to understand that their actions will have far-
reaching effects on those around them. It also means being able to under
stand that differences in culture exist and are legitimate and can represent
a source of enrichment)
4) learning to be (provide opportunities in which students can express
themselves and both their beliefs and view points),
5) learning to learn (students self discovery of the best way of learning and
joy of learning as a process from which can derived pleasure, satisfaction,
motivation and empowerment at an individual/personal level).

In addition, if we look from a different perspective such as: training for adults and
lifelong learning, the policy guidelines, principles, concerns and priority recommen-
dations presented, as example, in the Report “e-Inclusion in the framework of eEurope:
new Perspectives, New European Policy Agenda (eEuropa Advisory Group, 2005)
underline that the one of the main factors of the Digital Divide is related to “Dis-
abilities and Education and Skills” (level of education, ICT literacy, competence and
learning abilities, use of ICT in specific professions)”.

249
These policy priorities clearly highlight that there a quite large group of workers, the
majority of whom have little education qualification, IT competence, and rarely take
part in life-long learning path are at the greater risk of being marginalized from both
labour market and society at large.

The game industry is one of the fastest growing, which has lately drawn the attention
of policy makers in the EU. Especially under the Danish presidency it has been a hot
topic, and the Danish media have produced several items on the potential of computer
games. The latest development is that the EU Commission has now passed a resolution
with the intention of securing the communication of European cultural values
through interactive media, especially computer games (Egenfeldt-Nielsen, 2003).

It must be underlined that research findings have illustrated that the attributes of
computer/online and video games are coherent with the way people learn and they
can act to create effective active learning environments. In this respect game-based
learning can accommodate different learning styles: games provide players with a
continuous variety of emotional conditions or psychological stimuli. Players can learn
through performance or learning by doing, active learning, or experiential learning.
Such learning methods overcome many of the limitations associated with learning
by static learning situations, in which the knowledge gained is not often directly
transferable or applicable to the real world or to unfamiliar scenarios. The learning
experience being active, naturally promotes active discovery, feedback and stimulation.

In addition the use of games can support the process of learning without remaining
dependant methods that have not been perceived as effective and positive in meeting
learners’ aims. Game environments can offer a non-linear, immersive environment,
active performance-based setting into which there is the potential to incorporate
varying levels of complexity as well as to support the development of Strategic Thinking,
the use of Logic, Memory, Problem Solving and Critical Thinking Skills. These
conditions might include satisfaction, desire, anger, absorption, interest, excitement,
enjoyment, pride in achievement, and peer recognition to name but a few of the
complex mix of psychological conditions that influence motivation.

Games, in particular team-orientated games and those incorporating multiplayer

systems, are social environments, sometimes involving a distributed community.
Multiplayer systems and games requiring external information-inputs necessitate
collaboration, communication and the concomitant development of social skills

250
and teamwork amongst the players. Learning becomes a social and participatory pro-
cess with the potential to be carried out on a vast scale, and with high feedback levels.
Through the provision of various feedback mechanisms, games provide the player
with the opportunity of self-evaluation and assessment. Regular use of these functions
helps to develop their self-analytical skills. These skills help them to identify their
strengths and weaknesses and adopt an independent, self-directed approach to their
own personal learning and improvement (Pivec, Koubek & Dondi, 2004).

Despite the results from extensive research on the usage of game-based learning, the
educational/vocational training and lifelong learning practitioners have very different
attitude towards therefore to game-based learning. If we use for example their attitude
toward ICT, we can say that, there was and still is substantial resistance to adopting
and using ICT by the majority.

Teachers reacted to and are still reacting to the introduction and usage of game-based
learning in different ways, for example:
• The ones, who are using game-based learning as an integrative methods for
their traditional teaching. They have understood the potential of the methods
and the higher value in supporting motivation and in developing skills.
• The ones, who are using only a specific type of games (many of them do not
consider a game) like simulations (for specific subject-matter) but they are
reluctant to try other typologies.
• The ones, have never used game-based learning in their teaching and they are
reluctant to use these products since they haven’t seen any “good quality” ones,
they do not consider “game-base learning” a “serious approach” or “they are not
fully aware of the potentiality of this approach and they a reluctant to change”.

As in the eLearning sector, the issues on “quality” of learning games came up during
the last years. The broader adoption and usage of games in the formal, non formal
and informal sectors also depends on the assurance on the quality of learning games
as well as the availability of good practices and experiences of the usage of games
in teaching and learning processes.

251
3. SIG-GLUE quality stamp: an initiative to support the evaluation and
the certification of the quality of learning games

Within the SIG-GLUE project (Special Interest Group for Game-based Learning in
University and life long Learning), supported within the eLearning Initiative of the
European Commission, a framework for assessing and certifying the quality of digital
learning games has been designed and tested, and from May 2006 becomes an available
service for those who want to have the quality of their games certified.
(http://www.sig-glue.net).

Having established from the beginning that the quality of a learning experience does not
depend exclusively from the quality of teaching or learning resources but depends
also on the processes involved and on the context in which the experience occurs, the
SIG-GLUE partnership considered it important to make an effort to develop an
evaluation quality framework (criteria, evaluation tools and processes) for digital
learning games.

For learning games it means games which have an explicit “learning” purpose and
can be used, adapted and adopted for supporting, improving and fostering processes
within formal, non formal and informal learning scenarios. The quality stamp
focuses on: fully developed games already used or to be used in learning and teaching
processes. These games are eligible for obtaining the “SIG-GLUE quality stamp”.
The list of criteria presented below (Table: “Quality criteria for learning games”),
which represents the SIG_GLUE quality criteria for Learning games, has benefited
from the experiences acquired from two previous outstanding initiatives : Cerfad
Initiatives (1995-2004) and SEEQUEL Project (SEEQUEL 2004, standing for
Sustainable Environment for the Evaluation of Quality in E-Learning supported by
the European Commission—DG Education and Culture—2003-2005).

The SIG-GLUE Partnership has decided to identify different criteria in three different
area: methodological and context , content and technological. This choice does not
mean that the SIG-GLUE Partnership is not aware of the links between the different
areas, but rather that there is a need, in the present situation, to have adequate and
qualified “points of view”. The analysis and evaluation grid which is derived by the
criteria has been tested within the SIG-GLUE project life-cycle and it is now used by
the SIG-GLUE Quality Commission for assessing and certifying the quality of edu-
cational games.

252
As mentioned before the quality the criteria have been divided into three different
areas: methodological and context, content and technical. The criteria have been
articulated in different sub-criteria. The full description of the criteria and the sub-
criteria is presented in the “Guide to Quality criteria of Learning games” which will
be available on the SIG-Glue project website from May 2006 (SIG-GLUE, 2006).

CRITERIA
Pedagogical and context criteria
Target groups and prerequisites
Identification of target groups
Identification of prerequisites
Learning objectives
Clear definition of objectives
Correspondence between established objectives and the objectives that
can actually be reached by using the learning game
Context of usage
Clarity of practical instructions for the use of the learning game
Indications/suggestions on the context in which the learning game
can be used
Coherence of the game with the targeted context
Coherence between the learning game structure and the planned
training and learning context
Link between the learning game activities and the professional/working
context
Didactic strategy
Indication of the average play time
Incentives and support to motivation
Support for engagement and fun
Coherence between the game strategy and learning objectives
Quality of the game strategy with the individual player characteristics
Clarity of the game environment/ setting

253
 Organization and structure of the learning game
Clarity of the rules to be followed and decision making process
Coherence between rules and consequences
Constant focus on the player experience
Clear definition of roles (e.g. players, instructors, animators, etc.)
Coherence of the social and collaborative activity with the objectives
Communication and media
Clear and user-friendly tone and language
Quality of the interaction between the learning game and the user/player
Quality of the interaction amongst users/players/etc.
Coherence between the media used in the learning game and the contents,
the established objectives and the target group
Evaluation
Clear identification of evaluation criteria and procedures
Adequate number and distribution of evaluation activity, during
the game and at the end
Type of evaluation activity proposed
Quality of the feedback to the evaluation activity
Relevance of evaluation activity and consistency with the objectives
and/or the contents
Support to the reflexive process (e.g. players can review and rethink
their performance)

CRITERIA
Content criteria
Correct technical/scientific language and contents
Updating or obsolescence of contents
Correct and logical organisation of contents
Link between the contents and the subject area/knowledge
domain/curriculum

254
Practical contextualization of the content
Correct balance of the context in relation to the target group
Coherence of contents with the established objectives and the target group

CRITERIA
Technical criteria
Credits
Information on the producers, authors, etc.
Portability and conformance to standards
Robustness of the game
Conformance to standard
Structure and organization
Ease of installation (for offline digital games)
Modularity of the design
Modularity in use
Aesthetics and usage of the media
Quality of user/game interface
Possibility of intervention on the use of materials (stop, rewind)
Positioning of the different elements on the screen
Technical quality
Quality of image definition
Quality of image composition
Rhythm of images
Quality and definition of audio
Integration between audio and image elements
Synchronization between audio and image elements
Quality of typographic characteristics and clarity of texts
Quality of typographic characteristics and clarity of texts

255
 Technical quality of drawings
Technical quality of pictures
Technical quality of graphic animations
Information produced
Privacy and security of personal data
Storage of the game play
Storage of evaluation and activities results ( e.g. save progress)
Print of the information
Table 1: Quality criteria for learning games

Quality in eLearning and quality of learning games: final remarks

If ones analyses the criteria identified above and the issues presented in the whole
contribution, some general remarks can be derived:
1) Any digital resources which are employed in learning and teaching processes
should meet quality criteria related to methodological/context, content and
technical areas. The importance of each area depends on the overall design
of the learning experience and the role attributed to the resources themselves.
2) It goes without saying that, as mentioned before, the learning resources are
only one component of the experience: the processes involved and the con-
text in which the learning is taking placed, play a fundamental role in the
success of the learning.
3) The comparison between the criteria used to assess the quality of eLearning
resources and the ones specifically targeted to evaluate learning games shows
that there are some specific criteria which are strictly linked to the nature of
the game-based learning, just to mention a few: support for engagement
and fun, clarity of the rules to be followed, clarity of the game environment/
settings, etc.. These criteria are very relevant when assessing learning games
because they refer to features which are deeply embedded in the game design
and they are specific to this typology of product. Nevertheless, since we are
referring to learning resources and not simply entertainment products, the
quality of some aspects (e.g. definition of the objectives, coherence between
the game strategy and learning objectives, etc.) which are relevant to any
didactical and methodological resources should be met. In other words: a

256
 learning game should be a “good game” thorough which the player will
achieve the stated learning objectives.
4) The need for assuring quality in eLearning as well as in Game-based learning
is very high. It is probably higher in game-based learning where cultural,
psychological and social resistance can be overcame only by showing that
the games are “serious”, “reliable” and “effective” in supporting the learning
and teaching process.

Authors

Claudio Dondi is the President of SCIENTER (IT). He was born in Modena in

1958 and has a university background in industrial economicsand he is the President
of SCIENTER–Ricerca e Innovazione per la formazione (IT), —a non-profit research
organization active Europe-wide in the field of innovation of education and training
systems. In this position his main activities are the co-ordination of large national
and European projects, as well as policy advice and evaluation at regional, national
and international level. His other positions include: Member of the Board of the
MENON EEIG (enabling eLearning) in Brussels, Member of the Editorial Boards
of the British Journal of Educational Technology, Vice-President of Eifel – the
European Institute for e-learning, Vice-President of EDEN–European Distance
Education Network, President of EFQUEL (European Foundation for Quality in
eLearning).

Michela Moretti is the Head of the Evaluation and Training Design unit of Scienter
(IT). She took a degree in Educational Science (Pedagogy) at University of Bologna in
1988 and attended a two years post-graduate course in Management and Develop-
ment of Human Resources. She is involved at national and European levels in projects
focused on: assessment of ODL and eLearning materials/resources, quality of ODL/
eLearning services and systems, pedagogy of ODL and eLearning, game based learning,

257
quality assessment of innovative projects which employ ICT and eLearning in the
process of teaching and learning. Michela Moretti has been the co-coordinator of the
CERFAD Commission (Regional Commission for the Certification of ODL materials
of Emilia- Romagna Region in Italy, 1995-2004). (mmoretti@scienter.org—http://
www.scienter.org)

References
Barcelona European Council (2002). Presidency Conclusion, 15 and 16 march 2002.
Retrieved Mar 31, 2006 http://europa.eu.int/comm/barcelona_council/
index_en.html

Boonem, A. & Petegem, W. van (2005) (Eds). Mapping of approaches to quality.

Desk research Interim report. SEEQUEL project.

CERFAD Initiatives (1995-2004). Regional Commission for the Certification of

Open and Distance Learning of the Emilia-Romagna Region in Italy. Emilia
Romagna Region, Cerfad Commission “Guida alla qualità dei sistemi eLearning”
2004. Retrieved Mar 31, 2006 http://www.regione.emilia-romagna.it/cerfad

Dondi. C & Moretti M. (2005). L a calidad de e-Learning. @LI ISN. Retrieved

Mar 31, 2006 from http://www.alis-online.org

eEuropa Advisory Group (2005). WG 2, Version 1, April 13, 2005, page.11.

Egenfeldt-Nielsen S. (2003). The Art, Business and Science of computer games.

Retrieved Mar 31, 2006 http://game-research.com/art_eu_com.asp

European Parliament (2000). Lisbon European Council—Presidency Conclusion,

23-24 March 2000. Retrieved Mar 31, 2006 from http://www.europarl.eu.int/
summits/lis1_en.htm

Pivec, M., Koubek, A. & Dondi, C. (Eds.) (2004). Guidelines on Game-Based

Learning. Pabst Science Publishers.

Prensky, M. (2001). Digital game-based learning. New York: McGraw Hill.

258
Scienter (1998). Different paths towards odl legitimation and equivalence. Caliber-net
project- Socrates Programme. Retrieved Mar 31, 2006 from http://www.caliber-
net.odl.org/index.html

SEEQUEL (2004). SEEQUEL Core Quality Framework. Retrieved Mar 31, 2006
http://www.education-observatories.net/seequel/index

SEEQUEL (2005). The SEEQUEL project—Sustainable Environment for the

Evaluation of Quality in E-Learning: improving eLearning quality through
dialogue. Retrieved Mar 31, 2006 http://www.education-observatories.net/
seequel/index

SIG-GLUE (2006). Special Interest Group for Game-based Learning in Universities

and lifElong Learning. Retrieved Mar 31, 2006 http://www.sig-glue.net/

259
Tagungsband_Druck 17.07.2006 9:28 Uhr Seite 260

Authors List
Maja Pivec FH Joanneum GmbH, Graz maja.pivec@fh-joanneum.at

Anastasia Sfiri FH Joanneum GmbH, Graz anastasia.sfiri@fh-joanneum.at

Ioannis Antonellis Computer Technology Institute, Patras antonell@ceid.upatras.gr

Christos Bouras Computer Technology Institute, Patras bouras@cti.gr

Apostolos Gkamas Computer Technology Institute, Patras gkamas@cti.gr

Vassilis Poulopoulos Computer Technology Institute, Patras poulop@ceid.upatras.gr

Philip Bonanno University of Malta philip.bonanno@um.edu.mt

Fiona Littleton University of Edinburgh fiona.littleton@education.ed.ac.uk

Jeff Haywood University of Edinburgh jeff.haywood@ed.ac.uk

Hamish Macleod University of Edinburgh h.a.macleod@ed.ac.uk

Michael Begg University of Edinburgh michael.begg@ed.ac.uk

Rachel Ellaway University of Edinburgh rachel.ellaway@ed.ac.uk

David Dewhurst University of Edinburgh d.dewhurst@ed.ac.uk

Ulrike Spierling Fachhochschule Erfurt spierling@fh-erfurt.de

Anja Hoffmann Design Service Team SAP, Walldorf anja.hoffmann@sap.de

Ido A. Iurgel ZGDV Darmstadt ido.iurgel@zgdv.de

Felicitas Becker ZGDV Darmstadt felicitas.becker@zgdv.de

Fabian Kempf Vitero GmbH, Stuttgart fabian.kempf@vitero.de

Karin Hamann Fraunhofer IAO, Stuttgart karin.hamann@iao.fhg.de

Chris Brannigan Caspian Learning chrisbrannigan@caspianlearning.co.uk

Angela Owen Caspian Learning angelaowen@caspianlearning.co.uk

Lotte Krisper-Ullyett factline.com, Vienna lotte.krisper-ullyett@factline.com

Barbara Buchegger Team consult, Vienna buchegger@teamconsult.at

Johann Ortner free scientist ho.ortner@aon.at

Michael Burmester Hochschule der Medien, Stuttgart burmester@hdm-stuttgart.de

Daniela Gerhard University of Landau gerhard@uni-landau.de

Frank Thissen Hochschule der Medien, Stuttgart thissen@hdm-stuttgart.de

Claudio Dondi Scienter cdondi@scienter.org

Michaela Moretti Scienter mmoretti@scienter.org

260
Tagungsband_Druck 31.07.2006 11:31 Uhr Seite 4

Table of Contents

Editorial 5

SIG-GLUE: A Special Interest Group for Game-based 9

Learning in Universities and Lifelong Learning

Implementing and supporting a game based learning 23

related community

Gender-based Neurocognitive Propensities influencing 59

Gameplay: An Interactions-oriented approach

Influence of videogame play on a student’s approach 85

to learning?

Virtual Patients: considerations of narrative and game play 101

Learning with Digital Agents—Integration of 115

Simulations, Games, and Storytelling

Interactive Drama and Learning Experiences 149

YoungNet—a Virtual Learning Community 161

Platform for Youngsters

Games based learning or learning based games? 179

A case study

Field Report: Collaborative Blended Learning meets 199

Game-based Learning or how Inter-net Platforms
allow us to play 1000 “Low Tech Games”

Dramaturgic E-Learning Strategy (D.E.S.) 215

—evaluation of a story based approach

Quality in eLearning and quality of learning games 245

Authors List 260