GroupBeaMR: Analyzing Collaborative Group Behavior in Mixed Reality Through Passive Sensing and Sociometry

TL;DR

GroupBeaMR introduces a passive-sensing, sociometry-based framework to analyze group behavior in mixed reality. By extracting conversation, shared attention, and proximity from MR headset sensors and encoding them as sociograms, the method applies weighted graph metrics to classify groups into cohesive, competitive, or fragmented states. A user study with 12 four-person groups validates that group behavior correlates with social interaction patterns rather than task performance, and demonstrates robust, configurable labeling with high stability under noise. The work offers a scalable foundation for adaptive MR collaboration systems that optimize presence, engagement, and coordination beyond traditional productivity metrics.

Abstract

Understanding group behavior is crucial for enhancing collaboration and productivity in mixed reality (MR). This paper introduces a framework for group behavior analysis in MR, or GroupBeaMR for short for analyzing group behavior in MR. GroupBeaMR leverages MR headsets' sensors to analyze group behavior through conversation, shared attention, and proximity, identifying cohesive, fragmented, and competitive interaction patterns. Using social network analysis, GroupBeaMR provides quantitative assessments of group dynamics, offering insights into collaboration structures. A user study with 48 participants across 12 groups validates the framework's ability to distinguish interaction patterns in MR environments. Our analyses show that group behavior is independent of task performance, emphasizing the significance of social interaction patterns. Our group-type assignments indicate that sensor-based assessments in MR can provide meaningful insights into collaborative experiences, supporting the design of systems that adapt and optimize group behaviors.

Figures (20)

• Figure 1: Illustration of the Collaborative MR Image Sorting Application. Left and Middle: First-person perspective of the application interface as experienced by two participants. Right: Third-person view showing participants interacting with the application in a shared MR environment.
• Figure 2: Multiple users collaborate using immersive MR applications. Their collaboration creates social dynamics reflected in sensor data captured by each MR device, which is communicated over a host network to manage the shared scene state and transitions. Pre- and Post-surveys are collected for each user. GroupBeaMR processing and analysis modules work offline to convert the raw sensor data into high-level proximity, conversation, and attention information to create a sociogram for analysis. Group behavior can be cohesive, competitive, and fragmented. Post-hoc analysis explores the correlation between the task performance (time to complete a task and accuracy/efficiency of the task) and the inferred group behavior.
• Figure 3: Controller gesture for image and label grabbing.
• Figure 4: User study procedure: Participants provided consent and completed a demographics survey followed by headset calibration, a tutorial, the main collaborative image sorting task, and a post-exposure questionnaire.
• Figure 5: Sociograms representing group interactions: (a) conversation, (b) proximity, and (c) shared attention. In the conversation and proximity sociograms, the edge thickness reflects the total duration of respective interactions; the shared attention sociogram indicates the cumulative shared gaze on virtual objects.