1 Introduction and Theoretical Background

Embodied learning apps, which compose an emergent category of educational digital apps that integrate bodily movement into the act of learning, are argued to have the potential to revolutionize K-12 education [1]. These innovative apps that leverage the power of motion-based technologies and natural interfaces, have created new educational possibilities [2]. Embodied learning apps are usually equipped with motion tracking systems (e.g., Wii, Xbox Kinect, or Leap Motion) to enable hand gestures or body movements that are closely mapped to the educational content to be learned. The springboard for the development of these cutting-edge apps is that bodily involvement can support learning by allowing (a) multimodal interaction, (b) dynamic feedback, and (c) the creation of physical representations [3, 4].

Despite the tremendous educational opportunities related to embodied learning apps, their integration in authentic educational settings is still in slow pace [5,6,7,8,9]. According to Karakostas, Palaigeorgiou, and Kompatsiaris [10] embodied learning apps, are mainly developed for research purposes and do not follow the curriculum. In particular, existing embodied apps have usually derived from a designer-oriented approach, in which teachers are excluded from the development process [4]. As such, it is not a surprise that teachers do not have confidence in integrating embodied learning apps within their classrooms and teaching practices [10]. Quick et al. [11] have pointed out that supporting teachers in adopting embodied learning apps, requires teachers to obtain a sufficient understanding of technology-enhanced embodied learning, on a personal level.

Participatory design, as a bottom-up approach allows the co-design of educational technologies with the involvement of teachers and can contribute to the successful development of embodied learning apps. Participatory design is a common practice outside the field of education and aims to involve the users of a product in the design process, to ensure the usability, acceptability and effectiveness of the final product [12]. Re-contextualizing this practice in education, the co-design of educational technologies can allow teachers to result in more effective final products, as it allows a better understanding of the relationship between the underlying pedagogy and its instructional goals [13].

While participatory design methods have been mainly used in the past for the development of desktop computer apps [14], researchers have argued that the design of novel embodied interaction interfaces require different modalities of interaction with the technology and provide opportunities that are essentially different from traditional desktop-based interfaces [15]. However, existing research efforts on the participatory design of embodied learning apps are still fragmented and limited. These efforts have focused on the involvement of children, excluding the voices of teachers [16, 17]. At the same time, existing participatory frameworks, often neglect the context-specific requirements in which the co-designed embodied apps will be used [15], which is a crucial factor in ensuring that the produced embodied learning apps can be integrated effectively in teachers’ classrooms. In this paper we introduce a co-design approach to the development and classroom integration of embodied learning apps in order to address previous limitations concerning the quality of the outcome and possibilities of successful integration.

2 The Co-design Approach

The identified limitations in the design and introduction of embodied learning apps in teachers’ practices point out to the need for taking a critical stance on the development of embodied learning apps. We propose a co-design approach for the development of embodied learning apps which involves teachers throughout all the stages of the design process. The proposed co-design approach requires a shift from the decontextualized design of embodied learning apps to a serious consideration of the educational context in which they will be used, to ensure that the produced embodied learning apps are aligned with the educational curriculum and its underlying learning goals.

From a methodological point of view, the proposed approach borrows from: (a) participatory design, to gain useful insights in teachers’ needs, to then be integrated in the design process [18], and (b) design-based research to organize teachers’ contributions within an iterative design and assessment cycle [19]. The proposed co-design approach can be visualized as an iterative design process enacted by a protype and aiming at its transformation into a fully functional embodied learning app, aligned with the educational curriculum and teachers’ expectations. The workflow includes three-stages: Stage 1: Teachers as “Users”, Stage 2: Teachers as “Content experts”, Stage 3: Teachers as “Technology integrators” (Fig. 1). The proposed co-design approach was enacted in the context of the INTELed European project (INnovative Training via Embodied Learning and multi-sensory techniques for inclusive Education) [20].

Fig. 1.
figure 1

Diagrammatic representation of the co-design approach for the development and classroom integration of embodied learning apps

Full size image

3 The “Young Cardiologists” Embodied App

In this work, we applied the proposed co-design approach to the development of the “Young cardiologists” embodied app, to support 5th–6th graders’ conceptual understanding about the heart circulatory system. The co-design approach was enacted by a prototype of an embodied learning app of the heart circulatory system, controlled by a leap-motion sensor (Fig. 2). The leap motion sensor was initially set up so as to identify users’ hands enacting two types of sequential gestures (a) a close/open fist gesture simulating the heart’s pumping action, followed by (b) a hand flipping gesture simulating the heart valves’ open/closing movement allowing the blood transition to the rest of the body (i.e. right atrium, right ventricle, pulmonary artery, lungs, left atrium, left ventricle, aorta, rest of the body) (Fig. 3). What follows is the presentation of the three stages as well as the main aims, methods, findings and iterations, per stage.

Fig. 2.
figure 2

Screenshot of the embodied app

Full size image
Fig. 3.
figure 3

The combination of close/open fist gesture (A) and the flipping gesture (B)

Full size image

3.1 Stage 1: Teachers as “Users”

Aims

Having a prototype at hand, the first stage had the goal of eliciting teachers’ preferences and viewpoints on the modes of embodied interaction, during a co-design workshop.

Methods

The co-design workshop had a total duration of 1,5 h and was carried out at our university premises with a total of nine in-service primary education teachers. The workshop comprised of two main sessions. As part of the first session, which lasted for approx. 30 min, the teachers were initially requested to act as “Users” to test the prototype and familiarize themselves with the embodied app. During this session, the participating teachers were introduced to the embodied learning approach and the goals of the particular app as an example case. Subsequently they were allowed to experiment and use the embodied learning app. As part of the second session, which had a duration of 1 h, the teachers took part in a focus group in which they were requested to reflect upon their embodied interaction with the app.

Findings

We analysed the video-recordings of the workshop using the multi-modal framework analysis suggested by Maliverni, Shaper and Pares [21] for the co-design of embodied learning environments analysing and synthesizing: (a) teachers’ sensorimotor engagement with the prototype version, focusing on their in-situ interactions with the embodied app, (b) teachers’ verbal interactions, focusing on their spontaneous speech acts during their in-situ interaction with the prototype and (c) teachers’ productions, focusing on teachers’ written reports and discussions, produced during the post-activity focus group.

This analysis of the in-situ physical and verbal interactions as well as the following group interview revealed that the embodied interaction with the prototype version of the app was perceived as relatively complex by the teachers. As we found from the sensorimotor explorations, the teachers had difficulty in combining the flipping gesture with the open/close fist gesture repeatedly to control the app. Instead, the teachers proposed to retain a more simplified gesture, such as the open/close fist gesture, which could be more intuitive as a heart pumping action. The teachers had also highlighted, that as the heart “pumping” was depended exclusively on user’s embodied interaction with the app, the function of the heart circulatory system could be erroneously perceived by some of the children as a voluntary and intended activity, while in reality it is an involuntary physical activity which takes place continuously.

Iteration #1: Informing Design

Based on our findings, the embodied interaction was simplified by keeping just the close fist/open palm gesture, as a more intuitive type of interaction simulating the heart’s pumping action. In addition, the app was revised in order to run automatically with a heart rhythm, presenting the heart pumping and the blood circulation taking place on a continuous basis. Additionally, in the revised version, the users could intervene in the operation of the human circulatory system via their hand gestures, speeding up the heartbeat by moving their fists faster (Fig. 4).

Fig. 4.
figure 4

The automatized version of the app with a heart rhythm (A), which also allowed users to intervene with their gestures to speed up the heartbeat (B)

Full size image

3.2 Stage 2: Teachers as “Content Experts”

Aims

A second co-design workshop aimed at evaluating the educational content of the prototype, focusing on teachers’ perceived affordances and limitations of the app.

Methods

The co-design workshop had a total duration of 2 h and was carried out at the university premises with the same cohort of in-service primary education teachers (n = 9). As part of the workshop the teachers were requested to act as “Content experts” after they had the opportunity to interact with the revised app of Stage 1. This time their aim was to evaluate the emerging opportunities and limitations of the app in promoting students’ conceptual understanding on the topic. To achieve this goal, during the workshop the teachers took part in a group interview activity, which was grounded on the nominal technique [22]. According to this technique, the teachers were initially asked to write down and justify their viewpoints individually regarding their perceived learning affordances and limitations of the embodied app. As a second step, the teachers were asked to share their ideas with the group and the interview ended with a debriefing discussion.

Findings

The data were qualitatively analysed using thematic analysis [23], in order to identify the main themes discussed by the teachers. According to our findings the teachers’ discussion evolved around two main themes, as follows: (a) the level of embodiment afforded by the learning app, and (b) the circulatory system representation supported by the app (Table 1).

Table 1. Overview of the main themes discussed, as these were evaluated by the teachers.
Full size table

The embodiment category involved ideas around the embodied and motion-based affordances of the app. As part of their discussion the teachers evaluated positively the affordances of the app for students’ bodily involvement (i.e. learners’ kinaesthetic activity), gestural congruency (i.e. gestures alignment with the concepts to be learned) as well as the degree of interactivity (i.e. gesture-based interaction and experimentation with the app). However, the teachers noticed and evaluated negatively the lack of bodily congruency (i.e. misalignment between their body and the simulated heart), as the circulatory system in the app was presented via a mirrored view. According to the teachers this misalignment could provoke a confusion to the students.

Discussing about the circulatory system representation afforded by the app, the teachers evaluated positively its realism (i.e. fidelity of the circulatory system representation). However, the teachers evaluated negatively the lack of key terminology (i.e. terms indicating the structure of the circulatory system), suggesting the addition of all the key terms in the app. The teachers evaluated negatively the static nature of the representation, suggesting the addition of more dynamic features such as a macro (zoom out) to a micro (zoom-in) transition within the circulatory system, and the selective focus on the different types of blood circulation which was deemed as a topic of high difficulty for their students. Finally, the teachers noticed the lack of any sound effect (i.e. heartbeat sound) and suggested how the integration of the “heartbeat” sound in alignment with the pace of the simulated heart could result in a more immersive learning experience.

Iteration #2: Informing Design

Based on our findings, the embodied app was updated with a zoom-in and zoom out feature, in order to allow children’s transition from the human heart (micro-level), to a holistic overview of the blood circulation within the body (macro-level) (Figs. 5a–b). Likewise, a selective focus feature was added allowing children to activate the systemic or the pulmonary blood circulation, or both of them (Figs. 6a–b). The embodied app was also revised to match the users’ position of his/her heart location as opposed to a mirror view. Finally, key terminology was integrated in all the representations, while the heartbeat sound effect was added to the app.

Fig. 5.
figure 5

a–b. The zoom-in/-out

Full size image
Fig. 6.
figure 6

a–b. The selective focus feature

Full size image

3.3 Stage 3: Teachers as “Technology Integrators”

Aims

The third co-design workshop had the goal to evaluate whether the revised design was aligned with the educational curriculum, thus allowing teachers to integrate it within their science classrooms.

Methods

The co-design workshop had a total duration of 2,5 h and was carried out at the university premises with a total of 5 of the primary education teachers, who had also participated in the previous two stages. As part of this workshop the teachers were asked to take the role of “Technology integrators” in order to discuss and reflect on whether the revised version of the embodied app could be integrated successfully in their classrooms. In order to do so, the teachers were provided access to the refined from Stage 2 embodied app as well as to the national educative materials used to teach the human circulatory system.

Findings

The co-design meeting was video-recorded and fully transcribed. The emerging data corpus was analysed via an open coding approach [24], focusing on teachers’ verbal interactions around their efforts to develop a technology integration scenario using the embodied learning app in their curriculum. Teachers’ discussion in this stage, focused on whether the learning goals underlying the app were aligned with the national educational standards and current learning activities, as these were presented in their educative materials (i.e. teacher and student guidebooks).

Based on our findings, according to the teachers, the embodied app was in alignment with most of the national education standards as it could contribute to students’ understanding about: (a) the basic organs of the human circulatory system, (b) the position and size of the heart within the human body, (c) the structure of the heart, (d) the role of the heart as a pumper sending blood to all the parts of the human body, (e) the systemic and pulmonary circulation, and (f) the role of blood in the systemic and pulmonary circulation to transfer oxygen and carbon dioxide within the human body. However, according to the teachers, the embodied app was not capable of responding to the educational national standards related to students’ understanding about: (a) how the heartbeat varies and differentiates according to the pace of the physical exercise and (b) everyday customs which affect positively or negatively the human circulatory system (e.g. smoking, sedentary life, food and diet, physical exercise). As the teachers discussed, the embodied app could not support students’ understanding about the impact of various physical exercises (e.g. walking Vs running) on the pace of the heartbeat. In addition, the teachers highlighted that in its current form, the embodied app could not inform students about basic concepts such as atherosclerosis, cholesterol, heart attack, stroke or thrombosis which are related with lifestyle and bad habits.

Iteration #3: Informing Design

Based on our findings, a motion-based human figure was added on the upper right corner of the app, acting responsively to the pace of the children’s gestures, as captured by the leap motion e.g. walking, jogging, running (Fig. 7). In addition, a new section (Lifestyle & Bad habits) was added to the app, illustrating atherosclerosis with the build-up of cholesterol which can block gradually the blood vessels, thus resulting in thrombosis, stroke or heart attack (Fig. 8). As part of this section the children could encounter on the left part of their screen various bad habits (e.g. lack of exercise, junk foods, sedentary lifestyle). Subsequently, via their embodied interaction with the app the children could observe the negative impact of these habits, via the gradual accumulation of cholesterol in the blood vessels, resulting in thrombosis, stroke or a heart attack episode.

Fig. 7.
figure 7

The motion-based figure

Full size image
Fig. 8.
figure 8

The lifestyle & bad habits section

Full size image

4 The Evaluation of the “Young Cardiologists” App

The final version of the “Young cardiologists” was evaluated with children, with a focus on children’s learning experience and learning gains. The evaluation study was a pilot before classroom integration of the embodied learning apps, as the ultimate goal of the project.

4.1 Methods

The evaluation was conducted in the context of a summer club for children which took place at the university. As part of the implementation, a class group of 21 children (12 boys and 9 girls; age: M = 11.9, SD = 1.29) participated in three 45-min lessons (Figs. 9a–b). The children were randomly assigned in groups of 2–3 and were invited to use the final version of the embodied app to investigate the following guiding questions: (a) How does our heart function?, (b) How does the blood flow within the human body?, and (c) How can our everyday habits affect the human circulatory system?. A pre-post conceptual test was developed, adopting a combination of multiple-choice items and open-ended tasks from a published conceptual test on the topic [25]. The test had a maximum score of 10 marks and was allocated to investigate children’s conceptual understanding, before and after the learning intervention, on the three thematic areas addressed by the embodied app (Structure and function of the heart, Systemic and pulmonary blood circulation, Lifestyle and bad habits). After the intervention, a semi-structured interview was also conducted with 11 of the children, to discuss their impressions about their learning experience.

Fig. 9.
figure 9

a–b. Snapshots from children using the final version of the embodied learning app

Full size image

4.2 Findings

To investigate the impact of the embodied app on children’s learning we analysed the pre-post tests using the Wilcoxon signed-rank test, as the data did not follow a normal distribution. The test revealed a statistically significant increase in children’s total learning scores as well as in all the three conceptual aspects: (a) Structure and function of the heart, (b) Systemic and pulmonary blood circulation, (c) Lifestyle and bad habits (Table 2).

Table 2. Pre-post-test comparison of students’ conceptual understanding
Full size table

The qualitative analysis of children’s interviews reinforced and shed light on the quantitative findings. In particular, the children reported their perceived learning gains in all the three conceptual aspects. At the same time, the children highlighted that the embodied app was user-friendly, highly interactive, enabled visual feedback according to their movements, and promoted a playful learning experience.

5 Discussion

In this study, we have presented a co-design approach to the development and classroom integration of embodied learning apps. The approach aimed at allowing teachers’ contributions to guide the design of embodied learning apps in alignment with their educational curriculum and therefore, increase the potential for their successful integration in mainstream education. This work informed the development of the “Young Cardiologists” embodied learning app, capable of supporting students’ learning about the human circulatory system. The suggested co-design approach acknowledges the importance of engaging teachers in the design of products aligned with their expectations and their teaching practices [18]. In addition, the suggested approach responds to the call for the development of embodied learning apps aligned with the educational curriculum [10], thus addressing the lack of participatory approaches that ground the design of embodied learning apps within context-specific requirements [15].

Focusing on the structure of the proposed approach, each one of the three stages enabled access to the further development of the embodied learning app. The first stage of the co-design approach (Teachers as “Users”) focused on the improvement of the embodied interaction with the app. According to our findings, teachers’ contributions in this stage indicated how the initial embodied interaction with the learning app was deemed as complex and as such, was replaced with a more intuitive embodied interaction, simulating the heart “pumping” action. This finding is aligned with prior research supporting that while movement-based interfaces often provide a venue for more natural interactions than mouse and keyboard, such interfaces could also be criticized by the users when they do not resemble movements in real life, when not responding to users’ expectations or when there is no gestural congruency with the educational content to be learned [2, 26, 28]. The second stage of the co-design approach (Teachers as “Content experts”) focused on teachers’ input about the educational content and its delivery via the app. Teachers’ contributions in this stage were vital in revising and improving the content of the app. This finding is in agreement with our prior research, suggesting that teachers’ involvement in the co-design of educational technologies can result in final products, aligned with students’ needs and capable of supporting learning [13, 27, 28]. Finally, according to our findings, teachers’ contributions in the third stage (Teachers as “Technology integrators”) were crucial for the refinement of the app, ensuring its readiness to enter the school classroom. Overall, the suggested co-design approach can provide an alternative opposed to the top-down design approach for the development of research-oriented embodied learning apps [10], geared towards the development of embodied apps which match the curriculum goals as well as teaching and learning needs.

The evaluation of the final version of the app with children, allowed the investigation of the opportunities emerged of the app in promoting students’ learning. According to our findings, the embodied app could support learning on all the intended conceptual aspects while it was also positively perceived by the children. Our positive findings are aligned with previous studies in the field which presented participatory design efforts with the involvement of children, resulting in final products capable of engaging young students and promoting their learning [15, 29].

Overall, the suggested co-design approach moves beyond top-down and designer-driven development approaches, which depend on the intuition of researchers and designers, excluding the contribution of other stakeholders. Instead, the involvement of teachers, which is found in the core of the proposed co-design approach, is of paramount significance for the design of meaningful modes of embodied interaction, appropriate educational content and successful contextualization of the app in teachers’ educational curriculum and classrooms.

6 Limitations and Future Studies

Overall, the co-design approach appeared to be successful for the development of embodied learning apps aligned with both teachers’ needs and the educational curriculum as well as capable of supporting children’s learning. A limitation is that, by addressing the design approach to in-service teachers, we did not include the actual end-users in the design process, namely 5th–6th graders. While the final product was embraced by the children and promoted their learning, future studies should focus on the expansion of the proposed approach to include both teachers’ and their students. At the same time, the evaluation study presented in this work was a pilot in a summer school programme, before formal classroom integration of the embodied learning app could take place; the forthcoming research plans of this project are directed towards the integration and investigation of the co-designed learning app within authentic educational contexts. Last but not least, the “Young cardiologists” apps a just a specific example of an embodied gesture-based app. Future studies should investigate the co-design of a broader range of embodied learning apps.

Despite these limitations, this study has presented a successful design approach which moves beyond the designer-driven approaches, involving in-service teachers in the design process, and embracing cycles of refinements to optimize the design and contextualization of embodied learning apps in the educational curriculum. The co-design method can contribute to the design of embodied learning apps for classroom integration, taking into account the teachers’ needs and expectations as well as the context-specific requirements in which the app will be used.