Understanding group behavior is essential for improving collaboration and productivity. While research on group behavior in virtual reality (VR) is significantly advanced, understanding group dynamics in mixed reality (MR) remains understudied. Understanding MR group dynamics will enable designing systems that optimize collaboration, enhance productivity, and improve user experiences. This work outlines how MR headsets sensory systems can effectively capture group behavior, devise algorithms to process and interpret the data and demonstrate the correlation between group behavior and task-related performance metrics. We propose a framework for group behavior analysis in MR, or GroupBeaMR for short, to capture and analyze group behavior in MR. Using the rich sensory capabilities of MR headsets, GroupBeaMR passively collects data on conversation, shared attention, and proximity. This data is processed using social network analysis techniques to identify patterns of interaction and assess group behavior. Our evaluation, involving 44 participants in 11 groups, demonstrates the effectiveness of GroupBeaMR in capturing and analyzing group behavior in collaborative MR tasks. An example of insight from GroupBeaMR is that balanced participation in different types of interaction leads to higher group cohesion. These findings enable real-time assessments of group behavior in MR that can enhance collaborative experiences.

翻译：暂无翻译