Optimal smoothing in visual motion perception

When a flash is aligned with a moving object, subjects perceive the flash t o lag behind the moving object. Two different models have been proposed to explain this "flash-lag" effect. In the motion extrapolation model, the vis ual system extrapolates the location of the moving object to counteract neu ral propagation delays, whereas in the latency difference model, it is hypo thesized that moving objects are processed and perceived more quickly than flashed objects. However, recent psychophysical experiments suggest that ne ither of these interpretations is feasible (Eagleman & Sejnowski, 2000a, 20 00b, 2000c), hypothesizing instead that the visual system uses data from th e future of an event before committing to an interpretation. We formalize t his idea in terms of the statistical framework of optimal smoothing and sho w that a model based on smoothing accounts for the shape of psychometric cu rves from a flash-lag experiment involving random reversals of motion direc tion. The smoothing model demonstrates how the visual system may enhance pe rceptual accuracy by relying not only on data from the past but also on dat a collected from the immediate future of an event.