ABSTRACT
The status of the gestalt theory of lightness perception, particularly as expressed by Koffka, is reviewed with three general conclusions. (1) Key aspects have become established, such as the dependence of lightness on at least two luminances and the crucial role of edges or gradients. (2) The thorny problem of perceptual structure continues to be shunned. It is shown that a very popular embodiment of the edge approach, the so-called contrast theory, is not in fact consistent with gestalt theory. This is seen most clearly in the failure of contrast theories to distinguish illumination edges from reflectance edges. This weakness, tantamount to a failure to engage the gestalt issue of perceptual structure, is shown to afflict both the older adaptation-level theory of Helson as well as the more current models based on Land’s retinex approach. (3) Recent developments have moved us closer to a quantification of the prägnanz principle. Attention is drawn to the work of gestalt theorists Bergstrom, Gerbino, and Gilchrist and the potential in the concepts of common and relative components and of the retinal image as layered for launching a successful assault on the crucial puzzle of structure.
