Collaborative Learning and Social Learning in Augmented Reality, Virtual Reality, and Metaverse Environments: A Scoping Review

Collaborative and social learning are increasingly supported by extended reality (XR) technologies, that is, augmented reality (AR), virtual reality (VR), and mixed reality (MR) and the metaverse. However, their potential to support broader social learning processes beyond structured collaboration remains underexplored. This scoping review examines how XR and the metaverse support collaborative learning and social learning and looks into cooperative learning, group learning, and peer learning, identifying their educational benefits, challenges, and research gaps. In total, 213 studies from ACM, IEEE, Scopus, and Web of Science were analyzed to address six research questions. Specifically, the study examines relevant review and experimental studies to synthesize evidence from both primary and secondary research. The findings indicate seven main topics discussed within the literature and reveal a growing research field increasingly emphasizing pedagogical and social dimensions, yet predominantly focused on collaborative learning (89.7%) rather than social learning (10.3%). The existing literature indicates improvements in engagement, motivation, collaboration, communication, and learning achievement, while studies that focused on social learning, which primarily employed VR in higher education, additionally highlight improvements in self-efficacy, critical thinking, and social skills. However, technological limitations, accessibility concerns, lack of training and educational resources, implementation costs, and insufficient pedagogical guidance remain significant challenges. Short-term interventions and limited theoretical grounding further constrain the understanding of sustained social learning processes. The findings highlight the need for longitudinal and theoretically grounded research to better understand observational learning, community participation, and identity development in immersive environments. Finally, it highlights generative artificial intelligence (AI)-driven avatars as emerging social agents that merit attention in future social learning and social-emotional learning research.