Assessing the Decision-Making Process in Human-Robot Collaboration Using a Lego-like EEG Headset

Abstract

Human-robot collaboration (HRC) has become an emerging field, where the use of a robotic agent has been shifted from a supportive machine to a decision-making collaborator. A variety of factors can influence the effectiveness of decision-making processes during HRC, including the system-related (e.g., robot capability) and human-related (e.g., individual knowledgeability) factors. As a variety of contextual factors can significantly impact the human-robot decision-making process in collaborative contexts, the present study adopts a Lego-like EEG headset to collect and examine human brain activities and utilizes multiple questionnaires to evaluate participants' cognitive perceptions toward the robot. A user study was conducted where two levels of robot capabilities (high vs. low) were manipulated to provide system recommendations. The participants were also identified into two groups based on their computational thinking (CT) ability. The EEG results revealed that different levels of CT abilities trigger different brainwaves, and the participants' trust calibration of the robot also varies the resultant brain activities.