Abstract
When a test is briefly flashed adjacent to a moving stimulus, its perceived position is dragged in the direction of the nearby motion (Whitney & Cavanagh, 2000). Here we show that if the test was flashed on top of the moving stimulus, its perceived displacement increased by up to 400%. This large displacement was seen only if the moving stimulus oscillated back and forth and the test appeared near the moment at which the motion reversed. It was the motion following the test flash that determined the direction of illusory displacement, and the illusory displacement fell off sharply if the flash occurred a few hundred ms before or after the motion reversal. In a second condition, rather than flashing a test on top of the moving stimulus, we briefly changed the color of part of the moving stimulus (Cai & Schlag, 2001). Here as well there was a large illusory displacement of the flashed part of the stimulus but this displacement was not much affected by the timing of the flash relative to the motion and could be seen even for continuous motion. Moreover, the flashed component of the object did not seem displaced relative to the rest of the object, suggesting that the entire stimulus was extrapolated in the direction of motion, not just the flashed component. This difference between a flash on top of a moving object and a flash of part of an object may reveal the conditions that cause a brief flash to be interpreted as part of a moving object when it occurs at a moment of motion uncertainty, as opposed to an unrelated transient when it occurs after the object's motion is well established.
UCSD Sentate (SA) Chaire d'Excellence ANR France (PC).