On the Relationship between Two Types of Effects Caused by Color Adaptation: Changes in Color Appearance and Color Discriminability

Abstract

Varying the state of color adaptation not only causes changes in the appearance of color stimuli but also leads to changes in their discriminability. The effects of adaptation on color appearance can be determined experimentally by the method of cross-context matching. If cross-context matches are perfect with respect to all attributes by which color stimuli can be discriminated, it should be possible to use theories for cross-context matching in order to predict color discrimination across contexts. To characterize the empirical situation in which predictions of this type can be made, some conditions for cross-context matching and color discrimination are formulated in an axiomatic way and an elementary representation is derived from them. In order to test predictability of color discriminant across contexts empirically, an experiment was conducted with consists of two parts. In a first step, eight stimuli were matched across contexts. To the adjustments of each of the eight subjects a projective mappingfwas fitted. For a second experiment pairs of stimuli were chosen that, according tof, should correspond to each other. Subjects repeatedly had to match each of these stimuli within the original context taken from the first experiment. Discrimination ellipsoids were calculated, which can be considered as approximations of isodiscriminability contours in color space. For most subjects, corresponding discrimination ellipsoids can be predicted byfwith a sufficient degree of precision.