Motion of second order stimuli: Smooth pursuit eye movements and perceived speed

Purpose. The perceived velocity of slowly moving second-order (drift-balanced) motion stimuli shows a very pronounced dependence on contrast. For fast moving second-order stimuli this is not the case. The integration period required for the perception of second-order motion is significantly greater than for first-order motion. We have studied the role of second-order motion stimuli in evoking smooth pursuit eye movements under a variety of conditions. Our aim is to determine the inter-relationship between perceptual and pursuit performance. Methods. A dual Purkinje-image eye tracker was used to measure smooth pursuit eye movements in human observers. Stimuli were random dot fields whose contrast was modulated by a vertically oriented 1 c/deg grating vignetted by a two-dimensional Gaussian with a space constant of 0.5 deg. For the first order stimuli, the random dots simply moved across the screen; for the second order stimuli they randomly reversed contrast while moving. Measurements were made at slow ( ≤2 deg/sec) and fast (> 4 deg/sec) target speeds as a function of contrast. Results. The eye velocity gain of smooth pursuit for both first and second-order slow moving targets increases with contrast in a similar way, however the gain for second-order targets is lower at all contrasts. Pursuit latency is longer in response to second-order stimuli, especially in the slow moving condition. Additionally, for slow movement, there is a consistent deceleration in the eye velocity profiles to second-order stimuli after the peak velocity is reached; this is not as evident in profiles to the first-order stimuli. Conclusions. Here we find there are subtle differences in the dynamics of the pursuit response between first and second order motion which may be the counterpart of the observed perceptual differences in processing.