Abstract
Abstract: :
Purpose: To measure the effect of surround modulation on the ability to make fine orientation discriminations between two very similar center grating patches, and to compare these effects with previous results obtained with spatial frequency or contrast discriminations. Methods: A signal detection rating paradigm was used to measure the ability to discriminate between two vertical 4 cpd grating patterns on the basis of small differences in orientation. Orientation differences for a 40' test disk were fixed individually under control conditions for each observer to yield a d' of about 1.2, and remained constant in all conditions. Contrast of the test was held constant at 0.10. Each day, a control condition (no surround) and 5 test conditions with 4-deg surrounds of contrast 0.025-0.50 were run in separate sessions by each observer. On alternate days, surrounds were in-phase with the center patch or were 180 deg out of phase. On each experimental trial, one of the two patterns to be discriminated was presented. Observers rated whether it was stimulus A or stimulus B, using a six-point scale. Each stimulus was presented 40 times per session to permit estimation of ROC curves and daily d' calculations. Mean d' values averaged over 5-8 replications of each condition were calculated individually for each observer. Results: When center and surround were out of phase, adding a modulated surround had no effect at any contrast. When center and surround were in phase,performance in conditions with surrounds that differed in luminance from the center also had no effect on performance relative to control values. When the contrast of test and surround were equal, performance fell dramatically. These results differed considerably from those obtained when the task was either a spatial frequency or a contrast discrimination. Conclusion: Orientation discriminations are unaffected by surrounds of different contrast, but disrupted by surrounds of the same contrast. This pattern qualitatively differs from those found when the task is a spatial frequency or contrast judgment, suggesting that different mechanisms are needed to account for orientation versus contrast and spatial frequency. The current results suggest that local luminance steps or mechanisms of object segregation can stop lateral interactions in an orientation-processing pathway.
Keywords: 586 spatial vision • 579 shape, form, contour, object perception • 510 perceptual organization