Metaverse and Generative AI in Education: Supplementary Materials

Résumé fourni par la source

The convergence of Metaverse technologies and Generative Artificial Intelligence (GenAI) is reshaping immersive digital education. This study presents a scoping review of Metaverse and GenAI-enabled education following the PRISMA Extension for Scoping Reviews (PRISMA-ScR) guidelines. A systematic search of ACM Digital Library, IEEE Xplore, Scopus, ScienceDirect, and SpringerLink identified 114 primary studies published or indexed from January 2022 up to the respective database search dates in 2026. Based on the synthesized evidence, we propose a comprehensive taxonomy that classifies existing research across application domains, Metaverse technologies, GenAI categories, target user groups, degrees of freedom (DoF), pedagogical foundations, evaluation practices, technical limitations, and future research directions. The reviewed studies show that Virtual Reality (VR) is the predominant immersive modality (61%), while OpenAI-based models represent the most widely adopted GenAI ecosystem (60%). Applications are concentrated primarily in STEM, higher education, and healthcare, where immersive learning environments increasingly incorporate intelligent tutoring, adaptive feedback, virtual agents, and automated content generation. Although 68% of the reviewed studies reported user perceptual evaluations, these predominantly focused on short-term usability and engagement, with comparatively limited attention to longitudinal learning effectiveness, knowledge retention, cognitive outcomes, AI trustworthiness, accessibility, inclusivity, and real-world deployment. This imbalance reveals a central maturity gap: technological sophistication is advancing faster than the educational evidence needed to establish sustained learning effectiveness and real-world readiness. Commonly reported challenges include limited empirical validation, system maturity, scalability, interoperability, AI trustworthiness, immersive performance, accessibility, and ethical governance. Based on these findings, we identify key priorities for future research, including longitudinal and pedagogically grounded evaluation, scalable and interoperable system architectures, trustworthy AI, inclusive and ecologically valid validation, and stronger integration of pedagogical and governance considerations. The proposed taxonomy provides an evidence-based framework to support future research, comparative analysis, and the design of Metaverse and GenAI-enabled educational systems.

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.