Abstract
We compared the spatial extent of lateral masking for first- versus second-order images and investigated spatial interactions between these two types of images. To do so we measured the apparent contrast of a target Gabor at fixation, in the presence versus the absence of horizontally flanking Gabors. The Gabors' grating was vertical, had a peak spatial frequency of 3 cpd, and was either added to (first-order) or multiplied with (second-order) binary 2-D noise. Apparent “contrast” (i.e., the perceived difference between the high and low luminance regions of the first-order stimulus, or between the high and low contrast regions of the second-order stimulus) was measured with a contrast-matching paradigm. Using a temporal 2 AFC and the method of constant stimuli, subjects indicated in which interval the central Gabor had the higher “contrast”. For each subject, the first- and second-order Gabors were equated for apparent contrast without the flankers. Two of the authors and two naive observers participated in this study.
When first-order flankers abutted a first-order target the apparent contrast of the target was reduced by 29%, and remained reduced up to an element separation of 6 wavelengths. When second-order flankers abutted a second-order target the apparent contrast of the target was reduced by 23%, and remained reduced up to an element separation of 3 wavelengths. The spatial frequency and orientation tuning of the suppression effect was broader for second- than first-order stimuli. Second-order flankers did not reduce the apparent contrast of the first-order target; however, in three subjects, first-order flankers reduced the apparent contrast of the second-order target. This effect was tuned for spatial frequency.
Therefore, we find that lateral interactions operate over shorter distances for second- than for first-order information, and that these two types of information interact in an asymmetrical fashion.