AI Insight
This study of 520 Chinese university English language learners found that virtual reality-enhanced collaborative learning environments were positively associated with students' self-reported metacognitive awareness. The relationship was partially mediated by social interaction, and this mediation effect was stronger among students with higher engagement levels. The cross-sectional design means these findings show associations rather than causal relationships.
Why it matters
The findings suggest that incorporating VR technology into collaborative language learning may support metacognitive development, particularly when combined with meaningful social interaction and high student engagement. These results could inform the design of technology-enhanced language instruction in university settings, though further research is needed to establish whether these associations hold in other cultural contexts and educational environments.
Understand the Science
This study examined the association between virtual reality (VR)-enhanced collaborative learning environments and self-reported metacognitive awareness among 520 Chinese university EFL students. Drawing on sociocultural theory and technology-enhanced learning frameworks, we proposed and tested a moderated mediation model in which social interaction mediates the VR–metacognition association, and learner engagement moderates the mediation pathway. Structural equation modeling (CB-SEM) with latent moderated structural equations (LMS) and bootstrap confidence intervals (5,000 resamples) was employed. Results indicated that VR-enhanced collaborative learning was positively associated with self-reported metacognitive awareness (β = 0.312, p < 0.001) and with social interaction (β = 0.441, p < 0.001). Social interaction was further associated with self-reported metacognitive awareness (β = 0.298, p < 0.001), conforming the indirect association (0.131, 95% CI [0.078, 0.192]). Learner engagement significantly moderated the social interaction–metacognition pathway (β = 0.178, p < 0.001), such that the indirect association was stronger among more highly engaged learners. These findings extend existing scholarship on technology-enhanced language learning and offer context-specific implications for designing VR-based collaborative learning in Chinese university EFL settings; their generalizability to other populations and instructional contexts remains to be established. Because the design was cross-sectional, the relationships reported are associational and should not be interpreted causally.