Abstract
Intuition suggests that increasing either the size or contrast of a moving stimulus should improve the detection and discrimination of motion. Here we report a result contrary to this intuition. Observers required longer exposure durations to discriminate motion direction of a large high contrast Gabor (∼60ms), as compared that of a small Gabor patch (∼15ms).
We measured the threshold exposure duration required for human observers to discriminate motion direction of a foveally presented Gabor patch (sigma = 20–150arcmin, contrast = 3–92%, SF = 1c/deg, TF = 2Hz, vertically oriented). For small Gabors, as expected, the duration thresholds first decreased as contrast increased and then remained constant for contrasts higher then 10%. For large Gabors, however, duration thresholds INCREASED about five-fold as the contrast increased from 3 to 92%. Furthermore, at high contrast, increasing the size of the Gabor patch also resulted in substantial increases in duration thresholds. This pattern of results was replicated with Gabors moving at 8Hz, and with Gabors whose spatial frequency was scaled with size (i.e. fixed spatial bandwidth).
These results resemble the known properties of antagonistic center-surround receptive fields (RF) in area MT. This counterintuitive result occurs only for Gabors that are sufficiently large to stimulate both the centers and the antagonistic surrounds of foveal RFs in MT. The large Gabor patches will activate the antagonistic motion surrounds, and the resulting inhibition may be responsible for the observed decrease in motion sensitivity. The threshold increase with increasing size is less pronounced at low contrast, in agreement with the lower contrast sensitivity of the antagonistic surrounds. Additional experiments also indicate that “the critical size” required for this effect increases with eccentricity, paralleling the corresponding increase in RF size.