Abstract
Shimojo et al. (1989) found that depth information is crucial in determining motion directions of barber pole patterns. Smagt and Storner(2002) reported similar effects of occlusion on motion aftereffect (MAE) by using barber-diamond patterns (Duncun et al, 2000). However, it is not yet clear whether the perceived directions in adapting phase or the occlude configuration in the test phase is crucial for the occlusion effect on MAE. To clarify this point, we examined MAE for barber pole patterns with two rectangular flanking occluders and manipulated occluder configuration in both adapting and test phases. In our stimuli, two rectangular occluders flanked either on the right and left (vertical occluders) or on the top and bottom (horizontal occluders). The arrangement was manipulated in both adaptation and test phase for all 4 possible combinations (2 in adaptation × 2 in test phase). The motion aperture where moving gratings were presented was square. The grating pattern was oriented diagonally either 45 or 225 deg. When the gratings were moved, the subjects more frequently perceived motion either right/left or up/down depending on occluder arrangements, but less frequently in diagonal (orthogonal to grating orientations) directions. Five subjects participated in the experiments. Subjects' task was to judge the direction of MAE. It was found that occluder arrangements in the test phase affected more on the MAE direction than perceived directions in the adaptation phase. Similar results were obtained with a gabor patch as adapter, which produces perceived motion in diagonal directions (orthogonal to grating orientations). These results indicate that the adaptation itself is not affected by occluders. The adaptation involved here is a low-level adaptation. Therefore, the occlusion effect for MAE is most likely produced when the internal motion representations as an outcome of adaptation is translated to subjective motion impression of MAE.