All Human Versus Human-Robot Teaming: Measuring Neurophysiological Synchrony, Team Performance, and Trust during Search and Rescue

Abstract

This study used a multiplayer virtual environment simulating search and rescue to examine teaming behaviors and trust networks in two teaming conditions using the Wizard of Oz approach: (1) an all-human team (three humans) and (2) a human-robot team (two humans, one simulated quadruped robot). Neural, physiological, and subjective responses of trust and workload at the individual and team (e.g., neurophysiological synchrony) levels, along with task performance, were measured. We observed better task performance in human-robot teams and associated higher neural synchrony, but lower trust in the team and the robot agent. Trust between the human dyads remained intact. These findings provide insights into comparing the neural bases of complex and social interactions in all-human versus human-autonomy teams.