Dissecting the influence of the collinear and flanking bars in White's effect

Abstract

In White’s effect equiluminant test patches placed on the black and white bars of a square-wave grating appear different in brightness/lightness. The effect is important because its direction is seemingly independent of the extent of black or white bar contacting the test patch and is therefore inconsistent with explanations in terms of either contrast or assimilation. We separated the influence of collinear and flanking bars by independently manipulating their luminance. Observers adjusted the luminance of a matching patch to match the brightness (apparent intensity) of test patches, which in a standard White stimulus would be described as located on a black bar or on a white bar of a 0.5 cpd inducing grating. Test patches were 1.0^o in width and 0.5^o or 3.0^o in height. Test patch and mean luminance were 64 cd/m². On each trial the luminance of the collinear (or flanking) bars assumed one of twenty values from 3.2 to 124.8 cd/m² while the luminance of the flanking (or collinear) bars remained white (124.8 cd/m²) or black (3.2 cd/m²). Although the magnitude of induction varied, the pattern of results was similar across six observers. Matching luminance varied inversely with collinear bar luminance for test patches with either white or black flanks. Matching functions for test patches with white flanks were shifted to lower luminances than those for test patches with black flanks. Varying flanking bar luminance also produced inverse test patch matching functions, although with shallower slopes. Matching functions for test patches on white collinear bars were shifted to lower luminances than those on black collinear bars. This difference was greater than that for matching functions with white versus black flanks. We conclude that for the present conditions White’s effect is a contrast-based effect in which collinear bar contrast is stronger than flanking bar contrast.

Meeting abstract presented at VSS 2015