Teaching Robots a Lesson: Determinants of Robot Punishment

There have been multiple incidents where humans attacked robots in a public environment (Brscic et al., in: Proceedings of the international conference on human-robot interaction, ACM/IEEE, Portland, 2015, 10.1145/2696454.2696468); Vincent, in: A drunk man was arrested for knocking over Silicon Valley's crime-fighting robot, 2017, ; Mosbergen, in: Good job, America. You killed hitchBOT. Huffpost, 2015, ; Mutlu and Forlizzi, in: Proceedings of the 3rd ACM/IEEE international conference on Human robot interaction, ACM, 2008, 10.1145/2696454.2696468; Rehm and Krogsager, in: 2013 Proceedings of IEEE RO-MAN, IEEE, 2013, 10.1145/2696454.2696468; (Salvini et al., in: 19th International symposium in robot and human interactive communication, 2010). Although the form of aggression suggests that this behaviour might be motivated by the aggressor's desire for social recognition rather than an urge for vandalism (Salvini et al. 2010; Keijsers and Bartneck, in: Proceedings of the international conference on human-robot interaction, ACM/IEEE, New York, 2018, 10.1145/2696454.2696468), very little is known about the underlying psychological mechanisms. Therefore, extending previous research, the current study investigated if human aggression towards a robot would be influenced by the aggressor's feelings of power, the perception of the threat that robots in general might pose, mind attribution to the robot, and the robot's embodiment. First, threat and power were manipulated. Subsequently, participants played a learning task with either a virtual or an embodied robot. Mind attribution was measured afterwards. Participants were asked to restrict the robot's energy supply after each wrong answer, which was taken as a measure of aggression. Results indicated that an embodied robot was punished less harshly than a virtual one, except for when people had been primed with power and threat. Being primed with power diminished the influence of mind attribution. Mind attribution increased aggression in the threat condition but was related to decreased aggression when people had not been reminded of threat.

Automated summary

Several incidents of humans attacking robots in public environments have been observed, with aggression potentially motivated by desire for social recognition. Little is known about the psychological mechanisms underlying this behavior. Research indicates that human aggression towards robots may be influenced by the aggressor's feelings of power, perception of threat posed by robots, mind attribution to the robot, and the robot's embodiment.