Abstract
An abrupt change in the direction of motion of a moving stimulus array induces change blindness for the simultaneous rotation of one element in that array (Yao, VSS 2012). Yet, a change in the direction of the array did not induce change blindness when one element changed color. We developed a "camouflage" theory that "like conceals like"—similar types of change to the array as a whole can conceal changes to an element of the array. A change in the direction of motion conceals a motion-based change (i.e., rotation), but not a color-based one. The camouflage hypothesis predicts that a change in color in the display should conceal a color-based change, but not a motion-based one. Participants completed a new change detection task in which they had to detect Gabor rotations or color changes amidst concentric arrays of grayscale circles. Simultaneous with the target change, the circles all changed in luminance—some increasing, some decreasing in order to minimize global luminance changes in the display. The local luminance changes effectively induced change blindness for the color changes, but not for rotations, lending support to the camouflage hypothesis. A non-target change in a display will conceal a target change insofar as the two are perceptually similar. Taken together, these findings provide a potential unifying mechanism for the variety of stimulus-based methods used to induce change blindness in the lab.
Meeting abstract presented at VSS 2013