Abstract
The perception of speed and the ability to show smooth-pursuit eye movements depend on stimulus contrast for stimuli well above the detection threshold (Hawken & Gegenfurtner, 2001). We studied pursuit eye movements elicited by targets with contrasts ranging from just below individual thresholds (0.8–10 times threshold) to 100% contrast. Stimuli were Gabor patches at three different speeds (1, 8, 16 deg/s) and four spatial frequencies (0, 1, 4, 8 c/deg). Targets appeared randomly to the left or right of a central fixation spot in a step-ramp paradigm and moved towards the fovea. Eye movements were measured with an EyeLink II system. At the end of each trial, subjects had to rate target direction and velocity. We used the method of constant stimuli with contrast starting values derived from a staircase procedure prior to the main experiment. We analyzed only those trials where the psychophysical judgement was correct and determined smooth-pursuit latency, steady-state gain, and position error. Results show that pursuit significantly depends on stimulus contrast: Gain increased and latency decreased with contrast. For all stimuli well above threshold, the effect was small. However, below a contrast level of two to three times the threshold, pursuit was severely impaired for medium and high target speeds: Pursuit latency was long, steady-state gain was low, and position error was high. Furthermore, latency and gain for medium and high target speeds varied with spatial frequency. For fast targets with high spatial frequency, most of the foveation was obtained by saccades. We conclude that poor signal quality at contrast levels around perceptual threshold makes it difficult for the pursuit system to reliably estimate velocity. The signals that drive perception of low contrast stimuli are not sufficient to guide the smooth-pursuit system. Therefore, a correct judgement of target direction and velocity does not necessarily imply that a stimulus can be pursued properly.
Miriam Spering was supported by the DFG graduate program “Brain and Behavior” (DFG 885/1).