Measuring Multidimensional Team Adaptation

Abstract

Recognizing their adaptive nature, recent research has shifted toward viewing teams as dynamical systems, in which team states emerge through ongoing interactions rather than being predefined at the individual level. Most studies, however, focus on single measurement modalities, limiting insights about system-wide adaptation. This study addresses this gap by using Collective Systems Adaptation (CSA) analysis, a multivariate time-series method designed to detect synchronized changes across multiple dimensions of team interaction. We applied CSA to three complementary measures of team interaction: Geospatial Coordination (spatial organization), Communication (information distribution), and Physiological Synchronization (pupillometry-based measure of workload distribution). These dimensions were integrated into one multidimensional model and compared to single-dimensional models to examine how adaptive processes emerge across multiple measures of team interaction. Teams completed high-fidelity simulated combat missions, and adaptation events identified through CSA were used to predict combat effectiveness. Our findings show that the multidimensional model outperformed single-dimensional frameworks, accounting for up to 55% of the variability in performance outcomes. Our results support the value of multi-modal analysis for understanding team dynamics and demonstrate the potential for real-time assessment of team adaptation for training, monitoring, and decision support in complex domains.