Robot Comedy (is) Special: A Surprising Lack of Bias for Gendered Robotic Comedians

Abstract

Our own past work has shown that, surprisingly, the perceived gender of a robotic comedian does not influence human ratings of the robot’s humorousness, warmth, competence, comfort, or social closeness. This result differed from previous work on gendered robots in other historically gendered roles (e.g., healthcare and security), but the work was also conducted with university student participants, a population known to be relatively progressive. The present paper is a follow-up study in which we sought a more diverse (and possibly biased toward gender role congruence) participant group using Amazon Mechanical Turk (MTurk). Participants (N = 148) observed a clip of a robotic comedian with either a male or a female voice, after which we measured self-reported ratings of robot attributes. Results replicated the gender-related findings from our previous work, as well as provided insights as to how the previous and current participant populations differed in their opinions of robotic comedians. These findings confirm that gender stereotypes have less influence than expected on robot comedy performance, an insight on which designers of playful robots can build.