Theory of brightness and color contrast in human vision

An experimental test of Jameson and Hurvich’s theory of brightness contrast

“…Second, it could be calculated also as the mean of the log magnitude estimates at some level of background and TF square, when TF and reference field were under the same illuminance. These are given in Table 3 Moreover, in the absence of negative bs, it is also the case that the bs in Table IB Because there seems to be no support in these data for the quintessence of Jameson and Hurvich's (1964) model, little will be served by further laboring the matter. Rather, it will be more useful to examine the data for what they indicate about brigh tness constancy and brightness functions under conditions of simultaneous brightness contrast.…”

“…Since the occurrence of negative functions is crucial to the credibility of Jameson and Hurvich's (1964) model and to the dismissal of the older theories, the failure Not only was brightness constancy not perfect, but it systematically became less perfect as the TF squares were varied from black to white. This significant main effect (F =25.7, df: 6/66, p < .01) occurred also at each background level.…”

“…For such surfaces the luminance distributions on the retina to which the retina is sensitive may be quite differen t as illumination is varied. Moreover, Jameson and Hurvich's (1964) model, tested here, assumed the homogeneity of the different interacting regions that it described. In an experiment, therefore, the different regions should be first approximations to Lambertian surfaces.…”

“…By implication their arguments also were against the older explanations from which these theories were derived [Gelb (1929) and Koffka (1935) in the case of Helson and Wallach;and Helmholtz (1925) and Katz (1935) in the case of Woodworth and Schlosberg]. In place of these theories Jameson and Hurvich (1964) proposed a model that is consistent with opponent process theory. (For a review of this model, see Flock, 1970.)…”

“…Starting with the experimental conditions Flock (1970) has shown that Jameson under which Jameson and Hurvich and Hurvich's (1964) model could yield observed a "-" function, one would simply outputs with some similarities to Jameson increase CR, decrease separation between and Hurvich's (1961) data. For one thing, surround and focal region, and decrease the under the appropriate conditions the angular size of the focal region.…”