Activating elicited agent knowledge: How robot and user features shape the perception of social robots

A recent theoretical framework on anthropomorphism emphasizes the role of elicited agent knowledge in anthropomorphic inferences about nonhuman entities. According to the Three-Factor Model of psychological Anthropomorphism, people use anthropocentric knowledge structures when judging unfamiliar objects (e.g., robots). In the present research, our goal was to manipulate the accessibility of such elicited agent knowledge by varying features of a robot's voice: Specifically, we examined effects of vocal cues that reflected both gender of robot (i.e., a male vs. female voice) and voice type (i.e., a human-like vs. robot-like voice). This was done to test the impact of these vocal features on anthropomorphic inferences about the robot and on human-robot interaction (HRI) acceptance. Our results demonstrate that a robot's vocal cues clearly influence subsequent judgments of the robot and particularly so, when participant gender taken into account. Implications of our research for robotics will be discussed.