Purchase this article with an account.
V. Annan, A. L. Gilchrist; Updating the anchor in lightness computations. Journal of Vision 2001;1(3):466. doi: https://doi.org/10.1167/1.3.466.
Download citation file:
© ARVO (1962-2015); The Authors (2016-present)
In computing lightness, the highest luminance in a scene, that is the anchor, is assigned the value white and all other surfaces are computed relative to this value. Therefore in a miniature room with walls covered with only dark gray to black surfaces, the lightest surface, physically dark gray, looks white (scene 1). But when real white and light gray surfaces are added to the scene (scene 2), the original surfaces do not immediately re-anchor to the new highest luminance. Instead, the white surfaces initially look self-luminous and the original surfaces continue to look very light. The luminosity fades and the other surfaces darken only after more than a minute. A series of experiments were conducted to investigate this unusual delay in the re-anchoring process. Our results show that the lag in updating the anchor depends on an estimate of whether the illumination has changed across scenes, which in turn is determined by the number of surfaces that either change or maintain their luminance values from scene 1 to scene 2.
This PDF is available to Subscribers Only