People regularly interact with synthetic speech systems that vary in vocal characteristics. Three studies explored whether anthropomorphic (i.e., humanlike) and gendered (i.e., masculine/feminine) features of voices affected fundamental judgments of warmth, competence, and discomfort. Study 1 used vocal recordings with semantic content removed. Natural voices were judged as warmer, more competent, and produced less discomfort than machine-like voices. Feminine voices were judged as warmer and elicited less discomfort than masculine voices. Study 2 used unfiltered versions of the same stimuli and found similar negative impacts of machine-like features of social judgments, but the effects of voice gender disappeared. Study 3 examined whether gendered effects returned when gender-ambiguous semantic content was labeled as stereotypically feminine (braiding hair) or stereotypically masculine (tying rope). Stereotype consistency between task label and voice gender mattered for discomfort ratings of feminine voices. In sum, semantic content and the consistency between content and stereotypical expectations affected judgments of gendered voices. Additional considerations regarding gendered voices and speech content should be explored in future research.

Accurately measuring perceptions of robots has become increasingly important as technological progress permits more frequent and extensive interaction between people and robots. Across four studies, we develop and validate a scale to measure social perception of robots. Drawing from the Godspeed Scale and from the psychological literature on social perception, we develop an 18-item scale (The Robotic Social Attribute Scale; RoSAS) to measure people's judgments of the social attributes of robots. Factor analyses reveal three underlying scale dimensions-warmth, competence, and discomfort. We then validate the RoSAS and show that the discomfort dimension does not reflect a concern with unfamiliarity. Using images of robots that systematically vary in their machineness and gender-typicality, we show that the application of these social attributes to robots varies based on their appearance.

Recent research indicates that robots' physical appearance affects inferences about their social traits, but little research has examined how features may influence evaluative responses (i.e., liking of and willingness to interact). The current studies investigated how robot appearance and function affect evaluative responses. Across four experiments, we assessed how robotic facial features (Study 1) and roles (e.g., companion vs. military; Study 2) influence trait inferences and evaluative responses, and the relationship between them. Study 3 examined the association between robot facial features and roles, and Study 4 examined both facial features and roles to see which drove social judgments. Results indicated that trait inferences and evaluative responses vary in response to both facial features and roles. Trait inferences predicted, with differing strength, evaluative responses towards robots. When pitting facial features against roles, features accounted more strongly for judgments. Implications for human-robot interaction and robot design are considered.