Abstract
I will describe a variety of recent experiments that demonstrate the usefulness of the concept of anchoring for different aspects of surface perception. One series of experiments tested anchoring theory predictions concerning simultaneous contrast, including double increment displays, staircase contrast displays, reverse contrast displays, the locus of error in the standard contrast display, and the effect of target luminance on the size of the illusion. In another series of experiments, objects that straddle two adjacent fields of illumination were studied. These have posed a problem for anchoring theory because, although the theory nicely predicts the matches made to the separately illuminated parts of an object, it has not been able to predict lightness matches made to the object as a whole. The results show that object matches can be derived using a concept of prevailing illumination that depends on both the highest field of illumination, and the largest. In a third series of experiments, a long-lasting hysteresis effect that occurs in the Mondrian World was studied. Lightness shows gross errors when the lightest surface in the whole visual field is only dark gray. When real whites are added to this scene, it takes more than a full minute before lightness values come to reflect the new highest luminance. Our results show that this is due to a resistance to change, not of surface lightness values, but of the anchor itself. And this resistance is directly proportional to the number of surfaces that are continuously within the visual field and do not change in luminance.
Supported by NSF grant: BCS-9906747 and PHS grant: S06GM08 223.