Perception of partly occluded objects in pigeons and primates

Kazuo Fujita

Kyoto University, Japan

Objects are seldom fully visible in the environment. But we humans easily complete the occluded portions of objects. This process called perceptual completion has been demonstrated in chicks (e.g., Regolin & Vallortigara, 1995), mice (Kanizsa, et al., 1993) and primates (e.g., Sugita, 1999), but not in pigeons (e.g., Sekuler, et al., 1996). In two series of experiments, pigeons and nonhuman primates were directly compared in their ability for perceptual completion. In the first series (Fujita, in press (a)), rhesus monkeys and pigeons were trained to classify the absolute length of a black horizontal target bar located next to a large gray rectangle into 'long' or 'short'. Then the subjects were tested with all-reinforced probe trials in which the size of the gap between the target bar and the rectangle varied. Rhesus monkeys' proportion of choice of 'long' and 'short' was significantly biased toward 'long' only when the bar touched the rectangle. The subjects seem to perceive the continuation of the target bar 'behind' the rectangle to complete the 'occluded' portion. But pigeons failed to show such tendency. In the second series of experiments, a chimpanzee and pigeons were trained to match a horizontally-moving complete rod and a set of broken rods moving in concert in 0-delay matching-to-sample tasks. During this training a horizontal belt was presented either above or below the rods. Then the subjects were tested in all-reinforced probe trials with the same belt occluding the central portion of the rod. While the chimpanzee subject matched this stimulus to a complete rod, showing perception of object unity (Sato, et al., 1997), the pigeons matched this to a set of broken rods (Fujita, in press (b)). This failure to perceive object unity was demonstrated even when the occluding belt was either diagonal or bumpy, hence changing the shape and size of the rods above and below the belt. There seems a large difference in the ability for completion between pigeons and primates.

Fujita, K. (in press) (a). Perceptual completion in rhesus monkeys (Macaca mulatta) and pigeons (Columba livia). Perception & Psychophysics.

Fujita, K. (in press) (b). What you see is different from what I see: Species differences in visual perception. In T. Matsuzawa (Ed.), Primate origin of human cognition and behavior. Springer Verlag.

Kanizsa, G., Renzi, P., Conte, S., Compostela, C., & Guerani, L. (1993). Amodal completion in mouse vision. Perception, 22, 713-721.

Regolin, L., & Vallortigara, G. (1995). Perception of partly occluded objects by young chicks. Perception & Psychophysics, 57, 971-976.

Sato, A., Kanazawa, S., & Fujita, K. (1997). Perception of object unity in a chimpanzee (Pan troglodytes). Japanese Psychological Research, 39, 191-199.

Sekuler, A. B., Lee, J. A. J., & Shettleworth, S. J. (1996). Pigeons do not complete partly occluded figures. Perception, 25, 1109-1120.

Sugita, Y. (1999). Grouping of image fragments in primary visual cortex. Nature, 401, 269-272.

Fujita, Kazuo

Department of Psychology, Kyoto University

Yoshida Honmachi, Sakyo-ku, Kyoto 606-8501, Japan

fujita@psy.bun.kyoto-u.ac.jp