Abstract
When asked to report the position of a moving object at the time of a flash, subjects indicate its position to be ahead of its actual position at the time. Simple explanations based on different neural processing times for the flash and the moving target, with or without motion extrapolation, have failed to explain all the properties of this apparently simple effect. Here we report that the position of a target moving in depth is also judged as ahead of its actual position in depth at the moment of a flash. An advantage of moving in depth is that retinal and object motion can be disocciated.
Subjects viewed a stereoscopic display (optical distance 45 cm) of a ring (inner rad. 4.5 , outer 7.1 ) and a concentric disc (2.9 radius). The disc travelled directly towards or away from the observer through the ring at 1.6 or 3.1 cm/s (simulated near viewing, 25 cm) or 125 or 250 cm/s (simulated far viewing, 225 cm). Retinal speeds were the same (0.47 or 0.93 /s) for both viewing distances. At some point along the disc's trajectory, the ring flashed. The subject's task was to maintain fixation on a stationary cross and indicate if the disc was in front of or behind the ring at the moment of the flash. A staircase procedure titrated the actual position of the disc when it was perceived as being coplanar with the ring at the moment of the flash.
Observers demonstrated a reliable flash-lag phenomenon in depth: when the trajectory was towards the observer, the disc appeared closer at the time of the flash than when the trajectory was away from the observer. The size of the effect seemed to be related to retinal speed rather than to the disc's metric velocity.
Thus the flash-lag phenomenon can be obtained when motion is in opposite directions in the two eyes at the same time, and appears to be based on retinal events.