Purchase this article with an account.
Yukiyasu Kamitani, Shinsuke Shimojo; Kanizsa square without pacmen created by selective edge adaptation. Journal of Vision 2002;2(7):354. doi: https://doi.org/10.1167/2.7.354.
Download citation file:
© ARVO (1962-2015); The Authors (2016-present)
A low-contrast disk can be made invisible, and filled in by the surround brightness/color, after adaptation to a high-contrast edge that spatially overlaps with the contour of the disk (Shimojo & Kamitani, VSS 2001). We report here that this edge adaptation technique can be applied to “part of” the edges in the Kanizsa configuration, leading to emergent percept of illusory filled-in surfaces. Low-contrast “packman” stimuli (5% brighter/darker than the gray background) that constituted the conventional Kanizsa square were presented, after adaptation to four high-contrast disks that alternated between white and black at 1 Hz for 15 s. The pacmen and the disks shared their edges except for the wedge portions. Thus the edges corresponding to the corners of the “square” were left unadapted. Subjects fixated to a dot placed in the middle of the four disks/pacmen throughout the adaptation and test periods. All subjects (9) reported that at the beginning of the test period (for 2.7–4.3 s), only a square was perceived on a homogeneous background, and that the square region was filled with a distinct brightness, which was darker (brighter) than the background when the luminance of the pacmen was higher (lower) than the background. A similar illusory square was perceived using different shapes of stimuli that were designed to leave the same edges (the corners of the “square”) unadapted, even if the test stimulus (cross- or pie-shaped figures) did not produce a vivid illusory square/surface by itself. The results indicate that the edges (local luminance contrast) at the corners were preserved, and sufficient to create the illusory square, while the other edges were masked by adaptation, and filling-in occurred across them. It suggests that visual surface representations can be formed based on local edges and filling-in process between them, rather than global luminance profiles, or spatial context of visual shapes.
This PDF is available to Subscribers Only