Presentation
Physiological Synchrony as a Predictor of Team Performance: A Machine Learning Approach
DescriptionEffective team performance is critical in domains such as military operations, aviation, and healthcare. As artificial intelligence (AI) increasingly joins human teams, understanding the physiological dynamics that drive successful collaboration becomes even more important. This study explores how physiological synchrony—how closely team members’ biological signals align—can predict team performance. Using machine learning models, we analyzed multimodal physiological data (e.g., heart rate variability, breathing patterns, brain oxygenation) from teams engaged in a collaborative rescue planning task. Results show that while individual physiological signals alone offer strong predictions of performance, incorporating interactional features such as breathing synchrony and brain activity coupling significantly enhances model accuracy. These findings suggest that synchronized physiological responses between teammates are a powerful indicator of collective success, beyond what individual data alone can reveal. This research highlights the importance of measuring interactional dynamics in human and human-AI teams and opens new possibilities for real-time monitoring and adaptive support systems based on team synchrony. Future work will extend these models by analyzing larger teams and incorporating dynamic team-level features like recurrence and entropy to capture emergent coordination patterns.
Event Type
Lecture
TimeThursday, October 16th11:30am - 11:50am CDT
LocationGrand Hall K
Human AI Robot Teaming (AI)