Mitigating the negative effect of telepresence robots via an empathy-eliciting robot moderator

Despite various advantages that telepresence robots offer in mediated interaction, the users of these robots have been found to be perceived as less trustworthy compared to their communication partners who are physically present (i.e., collocated participants) in multiparty interaction contexts [1]. In this study, we investigate whether an empathy-eliciting behavior of a social robot moderator can affect the way the personality of telepresence robot users is judged by their collocated partners. In a within-subject design, middle school students (N=48) were put into groups of three – two onsite and one via a telepresence robot – and were instructed to complete an escape room moderated by a social robot. In addition to facilitating the experiment session, the robot moderator asked a question which was aimed at creating empathy toward the telepresence robot operator. We collected data on the perceived personality of participants via two questionnaires, one before and one after the empathy-eliciting question. The analysis of the data shows that following the empathy-eliciting intervention, collocated participants rated telepresence robot pilots as significantly more trustworthy and less tense, irritable, shy, and faultfinding. Furthermore, remote participants rated collocated students as significantly more trusting and outgoing, and significantly less tense. Moreover, the robot moderator was regarded as more desirable after it made the intervention. The findings suggest that not only does the empathy-eliciting question improve the way the personality of telepresence robot pilots is judged by collocated participants, it also improves how remote participants judge the personality of their teammates.