In this study, we examined how gendered robots could affect adults' and children's social cognition toward robots. Specifically, we investigated whether adults (Study 1) and 7- and 9-year-old children (Study 2) would anthropomorphize a humanoid robot, a dog-like robot, and a real dog, and attribute perceptual features to them when they were gendered or non-gendered. We provided participants with information on the targets' names and genders, and a brief explanation of what they could do. We analyzed adults' data and children's results solely in each Study and compared adults' and children's results in the end of Study 2. In Study 1, results revealed that adults were more likely to anthropomorphize the targets when they were not assigned a gender. This gender effect was not observed in their tendency to attribute perceptual features to the targets. In Study 2, children did not show a gender effect for both domains. When compared adults' and children's tendencies to anthropomorphize and attribute perceptual features to the targets, although both groups showed stronger tendencies to the real dog, adults' tendencies were stronger than children's. Moreover, 7-year-olds were more likely to anthropomorphize the humanoid robot than 9-year-olds and adults. The study provides new insights into gender stereotypes and robotics. Important implications are discussed based on the results, including for gender attribution to robots in educational settings. • Adults anthropomorphized robots more when no gender was assigned. • Adults most often anthropomorphized and sensed perceptual features in the real dog. • Children showed no gender effect for anthropomorphism or perceptual features. • Seven-year-olds anthropomorphized humanoid robots more than 9-year-olds. • Real dogs were attributed perceptual features more than robots by all groups.
Okanda et al. (Sun,) studied this question.