Investigating the effects of robot affect and embodiment on attention and natural language of human teammates

HRI studies investigating human-robot interactions in mixed initiative teams typically only look at macro-level behaviors. Yet, an investigation of micro-level behaviors such as eye gaze fixations, attentional shifts, communicative acts, and others is often necessary in order to determine the exact influence of robot behaviors on human cognitive processes. In this paper, we report the first results from several novel analyses of micro-level behaviors obtained from video and audio recordings from previous HRI team studies. The analyses focus on the effects of both robot embodiment and affect expression in the robot's voice on the human's verbal behavior and allocation of attention. The findings show complex relationships among all factors that have to be better understood to improve team performance.