ABSTRACT
Current work in the field of robotics is flooded by efforts to make robots more human-like. Roboticists are designing robots with heads, faces, necks, eyes, ears, and human-like voices based on the premise that a humanoid robot is the most appropriate form for human-robot interaction (Ambrose et al., 2001; Brooks & O ’Reilly, 2002; Hashimoto, Narita, Sugano, Takanishi, Shirai, & Kobayashi, 2002; Ishiguro, 2003; Simmons & Nourbakhsh, 2002). Researchers argue that a humanoid form will ease interaction because rules for human social interaction will be invoked, and thus, humanoid robots will provide a more intuitive interface (Breazeal & Scassellati, 1999; Brooks, 2002). Brooks, for example, suggested that, “it will be useful for a large mobile appliance and a person to be able to negotiate who goes first in a tight corridor with the same natural head, eye, and hand gestures all people understand already” (p. 38). The premise that the humanoid form is the appropriate form for human-robot interaction, however, remains largely untested. Opponents of a humanoid form suggest that robots are machines and that humanoid features may generate unrealistic expectations and even fear (see Dautenhahn, 1999). Turkle (1984) observed that it is important to people that we be able to see ourselves as different from machines, asserting that the blurring of the line between people and machines can be disturbing and frightening. Brooks also suggested that the current infatuation with humanoid robots may be a phase through which we need to pass as we learn more about human-robot interaction. These considerations about humanoid robots, both positive and negative, may affect people’s response to professional service robots in the workplace, particularly with regard to their willingness to rely on robots to help them achieve their goals. Therefore, it is important to conduct empirical studies of human-like versus machine-like robots and to explore the trade-offs.In this study, we examine how the appearance (humanoid vs. machine like) of a robot might affect people’s willingness to rely on and share responsibility with their robotic partner. We choose to focus on these dependent variables because they are central to the collaboration process.
