Afferent vs Efferent Motion Evoked Potentials: Relationship to Dynamic Visual Performance

Visual evoked potentials (VEPs) were elicited by a computer-generated random-dot kinematogram in which a Landolt ring target moved from monitor centre to the right in front of a stationary background. Motion contrast was varied in terms of percentage of moving pixels within the ring (100%, 50%, 30%, 20%, 10%). Subjects indicated one of four gap locations (left, right, upper, lower) while fixating a small dot which was either screen centred (afferent task) or centred within the moving Landolt ring (efferent task). Correctly reported gap locations and reaction times were recorded as well as N2-amplitudes and N2-latencies. Decreasing contrast lead to decreased correct responses, decreased N2-amplitudes, and increased reaction times which were linearly correlated with increased N2-latencies. While these effects were obtained for both the afferent and efferent tasks, they were significantly stronger for the latter. At 10% motion contrast, efferent reaction times and peak N2-latencies were as much as 370 and 100 ms greater, respectively. To determine whether the stronger efferent effects were due to differences in retinal afference between the two conditions, a third stimulus was employed in which the retinal image motion of the efferent task was simulated in the absence of eye movements (stationary fixation point and Landolt ring on a leftward moving background). The effects on performance and N2-latencies and amplitudes obtained fell between those of the two initial conditions. Implications of these findings for the processing of afferent vs efferent dynamic visual stimuli and their relationship to VEPs are discussed.