What is the Transition Point between Static and Dynamic Motion Aftereffects?

Abstract

The motion aftereffect (MAE) measured with a dynamic test pattern (eg a counterphase-flickering grating) is distinguishable by a number of properties from the classical MAE obtained with a static test pattern. For a dynamic MAE, however, it is not sufficient simply to introduce dynamic properties into the test pattern. In two experiments we attempted to determine the transition point in the temporal-frequency domain at which a dynamic MAE becomes distinguishable from the static MAE. First, we examined the interocular transfer (IOT) of the MAE with conventional first-order (luminance) gratings. The amount of IOT increased with temporal frequency, and was almost complete at 1 Hz and above. In addition, the IOT of a dynamic MAE shows a drastic reduction in the peripheral visual field, possibly reflecting difficulties in feature tracking or the loss of involuntary attention. Second, we examined the MAE with second-order motion as the adaptation stimulus (contrast modulation of two-dimensional static noise). Under these conditions, similar results were obtained for first-order and second-order test gratings: MAE was not observed at low temporal frequencies and a substantial MAE was observed only at 1 Hz and above. The results agree with recent findings which showed a gradual loss of spatial-frequency selectivity with increasing temporal frequency of the test pattern (Mareschal et al, 1997 Vision Research37 1755 – 1759). The present results support the idea that two mechanisms underlie the different kinds of MAE: a low-level mechanism responsible for the MAE observed at low temporal frequencies, and a high-level mechanism operating predominantly at high temporal frequencies with a transition point at about 1 Hz.