Pedagogical conversational agents (PCAs) are an innovative way to help learners improve their academic performance via intelligent dialog systems. However, PCAs have not yet reached their full potential. They often fail because users perceive conversations with them as not engaging. Enriching them with game-based approaches could contribute to mitigating this issue. One could enrich a PCA with game-based approaches by gamifying it to foster positive effects, such as fun and motivation, or by integrating it into a game-based learning (GBL) environment to promote effects such as social presence and enable individual learning support. We summarize PCAs that are combined with game-based approaches under the novel term “game-inspired PCAs”. We conducted a systematic literature review on this topic, as previous literature reviews on PCAs either have not combined the topics of PCAs and GBL or have done so to a limited extent only. We analyzed the literature regarding the existing design knowledge base, the game elements used, the thematic areas and target groups, the PCA roles and types, the extent of artificial intelligence (AI) usage, and opportunities for adaptation. We reduced the initial 3,034 records to 50 fully coded papers, from which we derived a morphological box and revealed current research streams and future research recommendations. Overall, our results show that the topic offers promising application potential but that scholars and practitioners have not yet considered it holistically. For instance, we found that researchers have rarely provided prescriptive design knowledge, have not sufficiently combined game elements, and have seldom used AI algorithms as well as intelligent possibilities of user adaptation in PCA development. Furthermore, researchers have scarcely considered certain target groups, thematic areas, and PCA roles. Consequently, our paper contributes to research and practice by addressing research gaps and structuring the existing knowledge base.
Game-inspired Pedagogical Conversational Agents: A Systematic Literature Review.
AIS Transactions on Human-Computer Interaction, 15(2), 146-192.
When commenting on articles, please be friendly, welcoming, respectful and abide by the AIS eLibrary Discussion Thread Code of Conduct posted here.