Spatial interactions are different at threshold and suprathreshold contrasts

The perception of a central stimulus can be affected by its surrounding. Such spatial interactions have been extensively studied in term of contrast perception. The surround can cause suppression or facilitation in contrast detection thresholds. The suppression can also decrease or increase the perceived contrast of the central stimulus. We examined spatial interactions at suprathreshold contrasts by measuring the surround effect with a contrast-matching task. Our results showed that spatial interactions at suprathreshold contrasts are intrinsically different from what has been reported at the threshold contrast. While a number of threshold detection experiments have demonstrated that collinear alignment is an important factor in center-surround interactions, we found that neither the contrast suppression nor the enhancement depended on the relative phase of the central and surround gratings. Furthermore, we found that: 1) surround effects did not depend on whether the surround was restricted to the sides or the ends of the central grating, and 2) contrast enhancement did not depend on orientation (parallel and orthogonal surrounds produced roughly equal amounts of enhancement). Together these results indicated no effect of collinearity on suprathreshold perception. Therefore, some mechanisms operating near threshold, like the second-order filtering models by Polat (1999) by Solomon and Morgan (2000) do not appear to contribute to suprathreshold pattern appearance.