Abstract
Comparative cognitive studies have shown that the ability of perceptual grouping in humans is superior to other species (e.g., Imura and Tomonaga, 2013). However, it is still unknown whether other species represent overall statistical properties when multiple similar objects appeared simultaneously (e.g., Chong & Treisman, 2003). The present study examined the ability to compute the mean size of sets of objects. We presented chimpanzees and humans with two arrays of 12 white circles against a gray background. One of the arrays had larger mean size of circles than the other. There were three experimental conditions; single condition (only one circle was appeared in an array), homogeneous condition (circles in an array had the same size), and heterogeneous condition (circles in an array had different size). Chimpanzees and humans were required to touch the array containing larger circles. The results showed that the performance under the homogeneous and the heterogeneous conditions were higher than that under the single condition in chimpanzees and humans. In addition, both humans and chimpanzees showed no significant difference in accuracy between the homogeneous and the heterogeneous conditions. Thus, the results were almost consistent with those reported by Chong and Treisman (2003). Additional experiments ruled out the possibility that chimpanzees used local visual properties, such as luminance of visual patterns and/or the largest or the smallest circle in the arrays, to choose the array had larger mean size. The results of this study suggest that chimpanzees can represent overall statistical properties of multiple visual objects.
Meeting abstract presented at VSS 2016