Three experiments examined how color and/or luminance of the occluder affect on Kanizsa amodal shrinkage illusion. In Experiment 1, the perceived width of the occluded square was measured with different colors (blue, red, green, and yellow) of the occluder. In Experiment 2, the perceived width was measured with different luminances of the occluder. In Experiment 3, the perceived width was measured with different colors with equiluminances of the occluder and occluded square. The results with nine observers for each experiment showed that: (a) the illusion occurred irrespective of the changes of color and luminance of the occluder, (b) the illusion was the largest when the color and luminance of the occluder was equal to that of the occluded square, and (c) the illusion disappeared with the equiluminances of the occluder and the occluded square. The illusion cannot be explained by only incorrect operation of the correcting mechanism. Rather, the result is consistent with Livingstone & Hubel’s suggestion that the geometrical illusions, such as Zollner illusion and Poggendorff illusion, disappear in the equiluminance condition because cues to depth are difficult to be used in that condition by the visual system.