In Sync or Out of Sync? Understanding Stress and Learning Performance in Collaborative Healthcare Simulations through Physiological Synchrony and Arousal

Abstract Physiological synchrony and arousal are increasingly utilised to understand students’ affective and cognitive states, such as stress, which can influence their learning performance and satisfaction in collaborative settings. However, it remains uncertain whether these physiological indicators can meaningfully reflect students’ stress and learning performance satisfaction during embodied collaborative learning (ECL). With advancements in sensing technologies, AI, and multimodal learning analytics (MMLA), it is now possible to model learners’ affective and physiological states in such dynamic and physically active settings. This study investigates the role of physiological synchrony and arousal as indicators of stress and learning performance satisfaction in ECL using a mixed-method approach. We first developed linear mixed models using heart rate and survey data from 172 students participating in collaborative high-fidelity nursing simulations. The findings were then presented to educators to gain insights into their interpretation of the current findings on the relationship between students’ physiological responses and their learning performance satisfaction. Results indicate that physiological synchrony is a significant indicator of students’ perceived stress and collaboration performance satisfaction, while physiological arousal is a significant indicator of task performance satisfaction, even after accounting for individual and group differences. Educators confirmed that these findings align with their assumptions about the relationships between students’ physiological responses and their performance satisfaction, validating the ecological validity of the results. They also expressed interest in using these insights to refine simulation activities and enhance reflective practices. These findings provide empirical evidence to support the development of context-aware analytic tools using AI and MMLA to enhance collaborative learning.