The Influence of Perceived Gender on Robot Abuse

Abstract

A growing interest in social robotics underscores the need for comprehensive research on human-robot interaction to inform responsible robot design and application. Gender dynamics is a contentious topic in human-robot interaction, necessitating further investigation to address this gap in knowledge. The study involved 64 participants who interacted with a robot, subjecting it to punishment and reward while it performed two different tasks: equation solving and emotion recognition. Participants engaged with a robot that was made to appear either female or male. Subsequently, participants provided their impressions of the robot. The study’s findings revealed that there was no statistically significant difference in the severity of the punishment given to the male and female robots, contrary to the initial expectation that female robots might be punished more harshly. Additionally, no significant difference was found in the mind attribution of the robots of different genders. Interestingly, both male and female robots received significantly less punishment during the emotion recognition task compared to the equation solving task. The absence of a significant difference in punishment based on the robot’s gendered appearance that participants did not display pronounced gender bias when interacting with these robots. The lower level of punishment during the emotion recognition task indicates that task context plays a crucial role in how individuals perceive and treat robots, which has practical implications for designing robot tasks and applications. While this study did not find significant gender-based differences in robot abuse, the experiment with embodied robots provides a foundation upon which future studies can expand. Further research in this area will be instrumental in designing responsible and ethical human-robot interactions in the future.