Carnegie Mellon University HCII Work on these projects has finished and the last update for this site was in 2010. The site will be maintained as a community resource.
anthropomorphism.org

Psychology: Perceiving Humanlikeness

Under what conditions do we attribute humanlikeness to nonhumans?

A living being isn't necessarily humanlike. That is, we can be animistic without being anthropomorphic. Here are some conditions for anthropomorphism, given animism.

 

Having a human face

Infants are born with a tendency to attend to symmetric, top-down shapes:
Three small rectangles arranged so that two rectangles form the top of the shape with the third rectangle centered below the top two.

Over time, faces acquire emotional and social significance. People attend automatically to faces, and faces make objects seem more humanlike. In one of our studies, people rated the "humanlikeness" of robotic heads (see DiSalvo et al. below). Robotic heads with more facial features, and with a symmetric shape like the one shown above, were rated as more humanlike.

Which robotic head seems more humanlike to you?
Two 'robot'-style heads. The one on the left shows three antennas with small lightning bolts between them; one antenna on each side of the head and one on the top. It has two eyes and its mouth is open in a rectangle shape with vertical lines like a grill. The one on the right shows one antenna on top of its head with small lightning bolts coming out at different angles. It has a rounded shape on each side of the head like two ears, two eyes, a nose, and a similar rectangle grill mouth but with a curved line like a smile over the grill.

Reading

DiSalvo, C. F., Gemperle, F., Forlizzi, J., & Kiesler, S. (2002). All robots are not created equal: The design and perception of humanoid robot heads. In DIS ’02, Proceedings of the 4th conference on Designing Interactive Systems: Processes, Practices, Methods, and Techniques, London, UK (pp. 321-326). doi>10.1145/778712.778756. NY: ACM Press.

Turati, C. (2004). Why Faces Are Not Special to Newborns: An Alternative Account of the Face Preference. Current Directions in Psychological Science, 13(1), pp 5-8.

 

Physical appearance and social behaviorRobot with its right hand extended

Not just faces, but other humanlike features, such as having a head, arms, or legs, increases anthropomorphism, especially when accompanied by "social" movements such as turning the head toward a person.

Reading

Bruce, A., Nourbakhsh, I. & Simmons, R. (2002). The Role of Expressiveness and Attention in Human-Robot Interaction. In ICRA ’02, Proceedings of the 2002 IEEE International Conference on Robotics and Automation, Washington, DC (pp. 4138-4142).

 

Social context cues

All sorts of cues (stimuli) in the social environment can trigger anthropomorphism. Just naming an animal or object makes it seem more humanlike, especially when the name is a person's name. Many pet dogs are given people's names. Max, Wendy, and Larry are probably more likely to be anthropomorphized than Patches and Flip.

In 1944, Fritz Heider and Marianne Simmel from Smith College made a short stop motion (also known as stop frame animation) film. In the film, two triangles (one larger, one smaller) and a circle move against and around each other and a rectangle. Virtually all viewers (with the exception of children with autism) make up a social plot in which the "big triangle" is seen as an aggressor. Studies have shown that the movements of the shapes cause automatic animistic perceptions. The movement of the shapes is very social and natural looking. When we speed up or slow the animation, or run it backward, the illusion of humanlikeness disappears.

Reading

Heider, F. & Simmel, M. (1944) An experimental study of apparent behavior. American Journal of Psychology, 57(2), pp. 243-259. https://doi.org/10.2307/1416950

 

Social interaction with objects and animals

We humans form mental models of ourselves and others during social interaction. Anthropomorphism involves a mental model in which the nonhuman (animal or object) is thought to have human attributes. For example, we may blame the animal or object for a mistake, thinking it intended to act as it did.

Interaction might increase the likelihood of anthropomorphic mental models because interaction consumes cognitive resources. Our attention is a limited resource. When we are interacting with an animal or object, we can't think intellectually about what is really going on, for instance, a biological or computer program. Instead we focus on what the animal or object is doing, and we automatically make attributions as we do with people.

Reading

Malle, B. F., & Pearce, G. E. (2001). Attention to behavioral events during interaction: Two actor-observer gaps and three attempts to close them. Journal of Personality and Social Psychology, 81(2), pp. 278-294.

 

Relationships and possession

We apply special significance to many objects we possess, and to animals and people with whom we feel close. A personal relationship with something or someone involves an emotional bond, and a feeling there is less difference between them and us. This response, sometimes called "identification," sometimes called "self extension," can make us prone to anthropomorphistic explanations, and to interpreting the other's behavior as sympathetic or positive.

When we are close to people, we take their perspective and see the world as they do, or assume that their behavior was motivated the way ours would have been motivated. We see them less as cardboard characters; more complete—with an inner life, and pressures from the social situation and environment. Their traits and motivations are something like ours. Hence, we can expect that our own cherished friend, pet, or product will be excused from any foibles, and given credit for successes and benefits. This is a currently ongoing topic of study for this group.

Reading

Kiesler, T., & Kiesler, S. (2005). My pet rock and me: An experimental exploration of the self extension concept. In G. Menon and A. R. Rao (Eds), Proceedings of the Annual Conference of the Association for Consumer Research, 32, Portland, OR (pp. 365-370).

 

Personality

Studies using people's ratings of animals have shown that people perceive animals to have personality traits. Some scientists think these traits are very much like human personality traits (for example, extraverted dogs who are cheerful and who like others, that are outgoing, energetic, and so forth). Others believe the apparent traits of animals are not comparable to human traits. For example, friendly dogs may like the people they meet, but these dogs are unlikely to seek out social relationships, parties, or group memberships as an extraverted person would. Nonetheless, we often attach traits to animals and objects when their behavior matches our stereotypes of these traits in people. Doing so is probably automatic. It helps us predict behavior and form expectations.

Reading

Bargh, J. A., Chen, M. & Burrows, L. (1996). Automaticity of social behavior: Direct effects of trait construct and stereotype activation on action. Journal of Personality and Social Psychology, 71, 230-244.

Goetz, J., Kiesler, S., & Powers, A. (2003). Matching robot appearance and behavior to tasks to improve human-robot cooperation. In RO-MAN ’03, Proceedings of the 12th IEEE International Workshop on Robot and Human Interactive Communication, Milldrae, CA (pp. 55-60). Piscataway, NJ: IEEE Press.

Gosling, S. D., Kwan, V. S. Y. & John, O. P. (2003). A Dog’s Got Personality: A Cross-Species Comparative Approach to Personality Judgments in Dogs and Humans. Journal of Personality & Social Psychology, 85(6), pp. 1161-1169.


See more titles in the bibliography.


Next: Measuring Anthropomorphism

 

This website promotes understanding of anthropomorphism. Our work on these projects has finished, but the website will continue to be available as a community resource.