Johannes Konert

Interactive Multimedia Learning: Using Social Media for Peer Education in Single-Player Educational Games

Social Media, as an information and communication technology, enables users to exchange information about experiences and insights in easy ways. Such exchange can be used for peer interaction among players of educational computer games as well. The players profit from social media content as learning resources are created, edited, and then shared by peers. Therefore, social media applications and concepts can serve as a way to bring peer education concepts closer to educational games and enhance the way learners share hints, assess each others’ solutions, and give feedback in the learning and playing process. However, the intersection of serious games and social media appears to be a quite novel research field. This thesis contributes an architectural approach, addressing the challenge of intersecting educational games and social media from an interdisciplinary perspective. Major contributions include concepts for user-generated content exchange, game adaptation, and peer group formation. The first contribution, a concept for content exchange, supports peer tutoring and assessment of user-generated content. It’s implementation and evaluation in the PEDALE scenario addresses primarily the Individual Group Assessment Problem. This problem concerns the diagnosis and assessment of individual learners’ abilities, while simultaneously seeking group learning and collaboration. The evaluation study conducted through this work in a classroom scenario shows the benefits of the content and knowledge exchange for individual learners under certain conditions. The second contribution, a concept for game adaptation, is designed to support peer interaction between players of educational games and acquaintances in a social media environment. The thesis’ solution proposes interaction patterns to allow the influence to gameplay by peers. In the corresponding Genius scenario, the conducted evaluation study shows significantly stronger acceptance values by players experiencing the game adaptation, compared to those using a game without them. For the third contribution, a concept for peer group formation, a new algorithm has been designed on the basis of a mathematical optimization model. The underlying Group Formation Problem is formally defined and thoroughly analyzed. This problem refers to the necessity to simultaneously respect homogeneous as well as heterogeneous matching criteria when building learning groups. The simulative evaluation reveals that the proposed solution, GroupAL, generates significantly better group formation results than all reviewed algorithms from related work. The results from the three conducted evaluations contribute to the aim of bringing serious games and social media closer together. The created solutions and the unifying middleware architecture SoCom.KOM build a solid foundation for further research on the creation of social serious sames.