Abstract
In the White effect (White, 1979) gray test patches of identical luminance placed on the black and white bars of a square-wave grating appear different in brightness. This effect has received much attention because unlike effects such as simultaneous brightness contrast (SBC), the White effect cannot be explained on the basis of extant edge integration models (Kingdom & Moulden, 1988). In the White effect the direction of the brightness change does not correlate with the amount of black or white border in contact with the gray test patch, or in its general vicinity. Recently, this has led many investigators to reject low-level filtering explanations not only for the White effect but for brightness perception in general. Instead these investigators have offered explanations based on a variety of mid-level and high-level perceptual organization schemes often involving T-junctions. Howe (2001) challenged explanations based on T-junctions with a novel variation of White's effect in which the T-junctions remain unchanged but the brightness effect is eliminated or reversed. The present experiments quantitatively measured the magnitude of brightness induction in a set of stimuli that formed a continuum ranging from the White effect (White, 1979) to SBC, including as its mid-point the Howe illusion (Howe, 2001). An additional variant introduced by Anderson (2001) and a White stimulus with multiple test patches were also examined. Empirical brightness matches from three observers were compared with predictions of the ODOG model of Blakeslee and McCourt (1999). Both the direction and relative magnitude of the brightness effects were predicted by the model. These findings argue that oriented multiscale spatial filtering that incorporates contrast normalization across orientation parsimoniously accounts for these effects, and challenge the view that higher-level mechanisms are required to explain them.
Supported by NSF IBN 0212789