Educational applications of artificial intelligence in simulation-based learning: A systematic mapping review

The field of education has experienced a transformation as artificial intelligence (AI) becomes increasingly applicable for learning purposes. AI has the potential to transform the social interactions in educational contexts among learners, teachers, and technologies. In this systematic mapping review, we focus on mapping and framing trends for educational applications of AI in simulation-based learning. Fifty-nine studies met the inclusion and exclusion criteria. We coded and analyzed six mapped categories in this literature review: (1) the year-of-study trend, (2) methods, (3) AI technologies, (4) simulation, (5) study trends, and (6) learning principles and theories. To provide nuanced details from the included literature, we also synthesized three thematic trends: (1) AI built in virtual agents for simulation-based learning, (2) AI infused in simulation-based learning with affective computing, and (3) AI leveraged in simulation-based learning for assessments. Trend One builds on a general acknowledgement of virtual agents as a guide for situated learning. Trend Two posits the role of affective states in learning trajectories and suggests the related machine learning approaches. Trend Three discusses machine learning techniques and multimodal computing used for assessment and feedback. The paper concludes with implications and suggestions for research and practice in AI in education using simulation-based learning.