Abstract
Smooth pursuit eye movements change the retinal image motion of objects in the visual field. The visual system therefore has to take the eye movements into account to produce a veridical motion percept. According to the classical linear model of motion perception during smooth pursuit the perceived velocity depends on the sum of a retinal motion signal, estimating the retinal image velocity for a given object, and an eye movement signal that estimates the eye velocity. Errors in motion perception during smooth pursuit, such as the Filehne illusion and the Aubert-Fleischl phenomenon, can be explained in terms of the relative size of these signals. However, little attention has been paid to the temporal relationship between the two signals. If the eye velocity is not constant, differences between the latencies of the two signals will also produce perceptual errors. We therefore tested whether the signal latencies differ and what their perceptual consequences are. Participants judged the velocity of a sinusoidally moving random dot pattern, viewed during smooth pursuit of a sinusoidally moving target. In Experiment 1, the phase relationship between the dot pattern and the pursuit target was manipulated and in Experiment 2 we varied the motion amplitude of the dot pattern. In addition we examined whether positional cues affected performance by including a condition containing limited-lifetime dots. The relative signal size and phase difference of eye movement signal and retinal motion signal were estimated by fitting the classical linear model to the data. The model described the data well for most observers. The phase difference between the two signals turned out to be quite small, with perceptual errors mainly caused by differences in signal size.