Abstract
PURPOSE: Previously, grouping was reported to shift the color appearance of a target toward the appearance of other objects in the group. Appearance depended also on local chromatic induction, and we considered here how grouping and local induction independently affected color perception. METHODS: A small test square (13 min, 20cd/m2) was centered within a background (6.2 deg, 15cd/m2) composed of 29 alternating purplish and greenish horizontal stripes (6.2 deg wide by 13 min high, 2.3cpd). Six inducing bars, of the same height and chromaticity as the test, were inserted some distance away from the test on either 6 purplish or on a 6 greenish stripes. The observer's task was to adjust a separate comparison square on its own EEW background to match the test square. RESULTS & CONCLUSION: The six inducing bars were varied systematically in width (1.1, 1.9 & 2.8 deg) so that they formed an hourglass figure with the test patch at the center. The color appearance of the inducing bars depended strongly on whether they were on greenish or purplish stripes, and the appearance of the test square shifted toward the appearance of the inducing bars. Local chromatic induction in the test was varied by positioning the test on a purplish or a greenish stripe, or halfway between two stripes, or within an immediately surrounding gray region. Induction from the hourglass pattern was found regardless of local induction. Grouping was varied by extending the 6 inducing bars to full width (6.2 deg), which disrupted the hourglass figure; or by shortening the width of the inducing bars to be equal to the test patch (13 min), so they were vertically aligned. The largest color shift caused by the inducing bars was found with the hourglass configuration, and the smallest with the widest full-width bars. Induction from the remote inducing bars, therefore, depended on a perceptual group that included the inducing bars and test square rather than on the size or area covered by the inducing bars.