Abstract
Aim. Although evidence suggests that contour-shapes and texture-shapes are processed by different mechanisms, they nevertheless interact in an important way. Specifically, textures can inhibit the processing of the shapes of contours they surround; this is termed ‘texture-surround suppression of contour-shape’. How does this suppression operate and what is its spatial extent? Method. Subjects adapted to pairs of sinusoidal-shaped textures or of single contours that differed in shape frequency, and the resulting shift in the apparent shape-frequency of single-contour test pairs was measured. All contours consisted of strings of Gabor microelements that were oriented either parallel (‘snakes’) or perpendicular (‘ladders’) to the path of the contour. The texture adaptors consisted of a central contour and a surround made of a series of contours arranged in parallel. We varied (i) the number of contours in the surround-texture and (ii) the orientation of Gabors in the texture-surround relative to the central-contour. Results. We found that (i) for extended texture-surrounds, the coding of snake contour-shapes is strongly suppressed by snake surrounds, and ladder contours by ladder surrounds, but the suppression is much reduced if the center and surround contours are of opposite type. (ii) Both snake and ladder surrounds with 7 contours or less have the same suppressive effect on a ladder contour. (iii) Near ladder-surrounds suppress the coding of snake contour-shapes more than do near snake-surrounds. Conclusion. There are two components to texture-surround suppression: one operates locally, is broadband in orientation and disrupts contour-linking, the other is spatially extended and prevents the shape of the contour from being processed as a contour.
This research was supported by a Research Foundation - Flanders (Fonds Wetenschappelijk Onderzoek - Vlaanderen) fellowship given to E.G. and by a Natural Sciences and Engineering Research Council of Canada (NSERC) grant # OGP01217130 given to F.K.