Open Access

Valerie Czok

• The aim of this study was to identify the challenges and potential of game-based Augmented Reality for use in science teaching. This work investigated relevant criteria for AR used in science teaching and systematically developed and evaluated a game-based AR learning environment. An initial scan and literature review of current Augmented Reality applications for science teaching revealed an increasingly developing interest and application of using Augmented Reality in science in a multitude of domains. However, the extent of Augmented Reality’s potential to support the learning process remains inconclusive as previous approaches have undergone heterogenous ways of assessment and evaluation. Currently there is no established standardization or framework to assist criteria-led quality development. We created a framework and evaluation matrix for this purpose. This evaluation matrix was used to classify and compare current Augmented Reality application approaches reporting learning effects, such as motivation, self-efficacy, self-regulation, or knowledge acquisition. The rated Augmented Reality applications are represented by heptagonal output referring to the framework´s seven parameters. These heptagons were juxtaposed with their reported learning effects to examine any potential underlying correlation or pattern. The framework was used for the criteria-based development of the game-based Augmented Reality learning environment “Beat the Beast”. It covers a key issue in Education for Sustainable Development: plastics and microplastics, merging aspects of the fields of biology, chemistry and engineering. Special interest was put into the design of combining Augmented Reality technology with a game-based approach. To differentiate the features Augmented Reality and GAME, the intervention design was contrasted into the original intervention (A) using both features, interventions (B and C) respectively switching Augmented Reality or GAME on/off and lastly (D) with neither Augmented Reality nor GAME. The four settings of this learning environment were evaluated and investigated for the potential to lower cognitive load and improve motivation, user engagement, knowledge, self-efficacy beliefs, and technology acceptance. Thus, eight learning effect variables were applied: motivation, technology acceptance, user engagement, cognitive load, computer self-efficacy, knowledge, Education for Sustainable Development in the dimensions action and motivation. To investigate this, research was split up into nine research questions: Preface XVI Research question 1: What are relevant subject-related and media didactic parameters to classify Augmented Reality applications used in science and engineering teaching, with respect to application design and setup? Research question 2: What patterns, similarities, and discrepancies can be identified in terms of the application setup? Research question 3: Is there a relation between the setup and the dedicated learning effects of the reviewed Augmented Reality applications? Research question 4: How can the analysis of Augmented Reality applications based on the developed parameters aid to develop Augmented Reality applications? Research question 5: How do the settings (with/without Augmented Reality, with/without game) differ in terms of motivation, technology acceptance, user engagement, cognitive load, computer self-efficacy, knowledge, Education for Sustainable Development in the dimensions action and motivation? Research question 6: Does a game-based Augmented Reality learning environment lead to pre-service teachers improving knowledge, compared to equivalent interventions without Augmented Reality or GAME? Research question 7: Does a game-based Augmented Reality learning environment lead to pre-service teachers to develop stronger computer self-efficacy, compared to equivalent interventions Augmented Reality or GAME? Research question 8: How do Augmented Reality and GAME influence the knowledge development? Research question 9: How do Augmented Reality and GAME influence the computer selfefficacy development? Juxtaposition of rated Augmented Reality applications (heptagonal output) and learning effects (motivation / self-efficacy / self-regulation / knowledge acquisition) revealed no correlation and pattern to link the Augmented Reality applications setup to their reported learning effects. Evaluation of the game-based learning environment revealed no significant difference in motivation and technology acceptance among the four different intervention groups. However, Preface XVII it showed that the intervention groups using Augmented Reality led to higher levels of cognitive load. Disparities are found in the user engagement levels: Significant differences are found between groups A (AR + GAME) and B (AR+ non-GAME), B (AR + non-GAME) and C (non- AR + GAME), C (non-AR + GAME) and D (non-AR + non-GAME). No significant difference was found for A (AR + GAME) and C (non-AR + GAME) nor between A (AR + GAME) and D (non-AR + non-GAME). Furthermore, user engagement levels of non-GAME groups exceed those of GAME groups. The analyzation of longitudinal learning effects, knowledge and computer self-efficacy has found that all intervention settings significantly led to improvement. Thus, this improvement cannot be derived by group membership. In consideration of the total sample (all four groups: A, B, C, D) as well as consideration of the subgroup non-GAME (C, D) a moderating effect was found for the design feature Augmented Reality on the computer self-efficacy pre-post relationship. This indicates the inclusion or substitution of Augmented Reality in the setting positively influences computer self-efficacy increase from pre to post. Furthermore, the moderator condition non- Augmented Reality has a stronger effect than Augmented Reality, indicating substituting Augmented Reality enhances the computer self-efficacy development. Additionally, a moderating effect was found for the design feature GAME on the knowledge pre-post relationship. This was found for the subgroup using Augmented Reality (A, C), indicating that among Augmented Reality interventions. the substitution/inclusion of GAME affects knowledge acquisition. A stronger effect was shown for the condition non-GAME, suggesting the substitution of GAME led to higher knowledge improvement. In conclusion, the framework's parameters transformed into design criteria and guidelines, serves as a valuable instrument supporting the development process of AR applications for science teaching. The developed game-based AR learning environment has demonstrated a playful and low-threshold approach to introduce new technologies for teaching, suitable complex, interdisciplinary topics like Education for Sustainable Development. The integration of game-based learning did not result in increased user engagement, as initially anticipated. However, despite the augmented cognitive load experienced in AR-based interventions, the observed enhancements in knowledge acquisition and computer self-efficacy underscore the potential of AR in the realm of science education.