Rebecca Stower 1, Arvid Kappas 1, Kristyn Sommer 2 1 Department of Psychology and Methods, Jacobs University, Bremen, Germany 2 School of Psychology, the University of Queensland, Brisbane, Australia
Social robots embody many different social behaviours which distinguish them from other interactive technologies. In spite of this unique capacity, there is potential for social robots to fail in settings such as classrooms, where the social processes which surround social learning from humans may not generalise to interactions with robots. Consequently, the goal of this study was to compare a robot and a human agent in a selective trust task across different combinations of reliability (both agents reliable, only human agent reliable, or only robot agent reliable). Ninety-four children aged 3-6 participated in an online study where they viewed videos of a human and a robot agent labelling familiar and novel objects. We found that, although children consistently preferred to endorse a novel object label from the agent who previously labelled familiar objects correctly, independent of if this was a human or a robot, their social evaluations tended much more towards a general robot preference. When both the human and the robot were previously reliable, children’s preferences also indicated a robot bias. Children’s conceptualisations of agency when making a mistake also differed, such that an unreliable human was selected as doing things on purpose, but not an unreliable robot. These findings suggest that children’s perceptions of a robot’s reliability is separate from their evaluation of its desirability as a social interaction partner and its perceived agency. Further, they indicate that a robot making a mistake does not necessarily have a negative effect on children’s desire to interact with it.
