Abstract:

Measuring and predicting team operational readiness is paramount to establishing effective teams across the military. Despite there being a standard sequence of training for individuals and teams leading to the actual mission execution in operational environments, measuring and understanding performance at the team level to determine readiness remains a challenge. This challenge is even greater for rapidly composed teams with limited experience working together.

The goal of this study is to determine underlying objective biomarkers of high-performing teams. We present our initial results based on 1) data collection from a five-member Marine Fire Support Team (FiST) engaged in a virtual simulation-based training environment; 2) data cleaning and extraction of derived neural measures during key team behaviors; and 3) initial feature development as part of building a causal model to predict readiness for rapidly
composed military teams.

Our goal is to enhance team performance prediction by providing a scalable and interpretable framework for assessing team readiness in dynamic operational environments under high-stakes conditions. This approach is grounded in a theoretic framework based on operationalizing Team Behavioral Competency Theory (TBCT), which has been empirically tested across military, aviation, and healthcare settings. This approach maps specific time windows of team behavior to team-level competencies such as Leadership, Mutual Performance Monitoring, Back-up behavior, Team Orientation, and Adaptability and examines underlying neural, behavioral and peripheral physiologic measures. Specifically, this includes the application of novel high-resolution tripolar concentric ring electrode electroencephalogram (tEEG) combined with standard electroencephalogram (EEG), peripheral physiologic and behavioral measurements. Characterization of these multimodal signals is accomplished using derived metrics to identify a set of features to populate the edges of a graphical causal model that predicts overall team performance and readiness at the mission and subtask level. We show that changes in functional connectivity metrics across the team between pre-stimulus and targeted behavioral windows correlate with a higher subjective mission rating, predominantly in the delta-band suggesting these are key features of team synchrony that could be used to build out a
causal model for predicting team performance.

Keywords: AI;TEAM TRAINING

Menu

Nick Petroff