Research

Bodies are vital for animals (apparently). They provide important social cues. How do we perceive bodies?

Humans show the inversion effect for faces and bodies (e.g., Reed et al., 2003; Yin, 1969): we are better at recognizing faces and bodies when they are upright than inverted. This inversion effect was not found for other objects, such as houses. This suggests that we visually process bodies and other objects differently.

What does the inversion effect tell us exactly? There are featural information and relational information among each object. For example, for a human body image, the featural information includes the sizes and shapes of local features, and the relational information consists of the relative locations of each part (e.g., the head is on top of the torso). After being inverted upside down, the featural information does not change, while the relational information changes (e.g., now the head is in the bottom of the torso), and our performance to recognize bodies drops heavily after inversion. Therefore, the inversion effect for bodies suggests that we rely more on relational information when visually processing bodies. This kind of processing relying on relational information is called "configural processing". We do not show the inversion effect for houses and other objects. For these objects, we rely more on featural information.

Configural processing is considered to be more efficient in searching and recognizing objects that share the same layout (such as bodies and faces). I am interested in its evolutionary path: whether it is limited to humans. I tested chimpanzees and found that they also show the inversion effect for bodies. This suggests that the common ancestor of humans and chimpanzees may already have this specific efficient way of processing bodies.

• Gao, J., & Tomonaga, M. (2018). The body inversion effect in chimpanzees (Pan troglodytes). PLOS ONE, 13(10): e0204131. https://doi.org/10.1371/journal.pone.0204131

My current research project on body perception started with the inversion effect. I examined many properties of the body inversion effect, and I am extending my topics from here. I mainly use touch-screen and eye-tracking tasks in my study.

By testing chimpanzees with images of very strange bodies, we could further understand the property of their configural body processing and infer their knowledge about body structures.

Do they show the inversion effect to bodies with atypical body part arrangements? What about bodies with atypical body part proportions? No, and yes. Chimpanzees need a normal body arrangement to show the inversion effect, while body proportions do not interfere with their configural body processing.

• Gao, J., & Tomonaga, M. (2020). Body perception in chimpanzees (Pan troglodytes): The effect of body structure changes. Journal of Comparative Psychology, 134(2), 222–231. https://doi.org/10.1037/com0000214 (A free manuscript is on PsyArXiv: doi.org/10.31234/osf.io/um9vr)

Humans have visual body representations, and we would find atypical bodies strange. The results in the above column suggest that chimpanzees' visual system uses different strategies to process typical and scrambled bodies. Do they really spot atypical body parts? We did an eye-tracking experiment using typical and atypical bodies, as shown on the left.

Compared to the body parts on normal bodies, chimpanzees looked longer at human arms and legs (bottom right picture), and they showed a tendency to look longer at misplaced arms and legs (top left picture). This suggests that they may be able to detect strange body parts. We hope this provides insights into their body representation.

• Gao, J., Adachi, I. & Tomonaga, M. (2022). Chimpanzees (Pan troglodytes) detect strange body parts: an eye-tracking study. Animal Cognition. doi.org/10.1007/s10071-021-01593-2

Do chimpanzees also use configural processing (inversion effect) for bodies of other species? We found that they do if they have visual or embodied experience with the stimuli. For example, they show the inversion effect for human bodies in familiar postures (daily postures) but not for human bodies in unfamiliar postures (dancing postures); they also show the inversion effect to stimuli of horses and crawling humans, which they share the quadrupedal postures with, although they have never visually seen horses or crawling humans.

I also conducted experiments in pre-school children for species comparison and to examine the development of configural processing. Children showed the inversion effect to human bodies but not houses, suggesting that configural body processing already appears at the pre-school stage and is stable at this stage. They also showed the inversion effect for chimpanzee and horse bodies, suggesting the effect of experience. Compared with chimpanzees, children are more flexible in configural processing, and so are human adults.

• Gao, J., Kawakami, F., & Tomonaga, M. (2020). Body perception in chimpanzees and humans: The expert effect. Scientific Reports, 10(1): 7148. https://doi.org/10.1038/s41598-020-63876-x

• Gao, J., & Tomonaga, M. (2020). How chimpanzees and children perceive other species’ bodies: comparing the expert effect. Developmental Science: e12975. https://doi.org/10.1111/desc.12975

This was my first research project with chimpanzees. This idea occurred to me when I was thinking about chimpanzees' social structures: they have a strict linear hierarchy. Compared with chimpanzees, humans have more complex relationships, and we understand non-linear relationships, including the rock-paper-scissors game. Therefore, I tested chimpanzees to see if they could learn this circular game rule. I also tested children to know how this ability to understand a circular relationship develops.

Five out of the seven chimpanzees I tested learned the rule. They had difficulties learning the third pair (e.g., scissors-paper) after learning the first two pairs (e.g., paper-rock and rock-scissors). I tested children from 3 to 6 years of age and found that they could learn this rule after about 50 months old. Comparing the performances of the two species, I found that chimpanzees' performance was equal to that of 4-year-old children.

• Gao, J., Su, Y., Tomonaga, M., Matsuzawa, T. (2018). Learning the rules of the rock–paper–scissors game: chimpanzees versus children. Primates 59(1), 7-17. https://doi.org/10.1007/s10329-017-0620-0

I used pictures of chimpanzee hands and human hands to represent "rock" "paper" and "scissors". Chimpanzees showed different reactions to the hands of the two species, although they were in similar shapes, thanks to Photoshop. That made me wonder how chimpanzees perceive bodies. It then led me to my current project of body perception and cognition in chimpanzees!