Measuring double-increment illusions

The specific grey shades in a visual scene can be derived from relative luminance values only when an anchoring rule is given. The anchoring theory of lightness (Gilchrist et al, 1999 Psychological Review 106 795 - 834) assumes that the highest luminance is perceived as white, and the appearance of all the other regions depends on their relationship to such white. It is thus a crucial prediction of the theory that equal regions representing luminance increments relative to their surrounds shall be perceived as identical. Our stimuli were incremental targets on two uniform surrounds placed side by side. We used the method of adjustment: observers varied the luminance of the test patch (set on a black surround) to match the achromatic colour of the comparison patch (set on a variable surround). We found that a target on a black surround looks always lighter than the same target on a more luminant surround. Previous failures to observe double-increment illusions may have been due to the fact that the strength of the effect rests on the specific luminances of the target and of the pair of surrounds chosen for the display.