Abstract
Purpose. Previous research has shown that the perception of interpolated shape depends on the filling in of achromatic surface color. Here, we examine the opposite relation – whether interpolating shape modulates the lightness of nearby surfaces. Method. Twenty observers discriminated fat and thin rectangles, the tops and bottoms of which were connected by illusory contours. Inducers were either all black (to produce a lightened illusory surface) or all white (to produce a darkened surface). In all trials, a small target briefly appeared at the same location – either just within the surface of a fat shape or just outside the surface of a thin shape. Target Weber contrast ranged from ±19%. Observers first decided whether the shape was fat or thin and then whether the target was lighter or darker than the background. For all correct shape response trials, the percentage of “light” responses was recorded for each target contrast value. A cumulative Gaussian function was fitted to the resulting data for each inducer polarity and for each kind of shape (fat/thin). Results. There were two main results. First, there was an interaction between shape and inducer polarity for the points of subjective equality. Specifically, when inducers were light, targets inside the shapes were perceived darker than targets appearing outside the shape; when inducers were dark, there was no such dependence. An analysis of the sensitivity parameters showed the same interaction. When inducers were light, targets appearing inside the shapes were discriminated less precisely than targets appearing outside, and when inducers were dark, there was no such dependence. Conclusions. These results suggest that 1) the precision and accuracy of achromatic color judgments depends on interpolated shape; and 2) this effect is much more apparent when the filled-in surface color is dark. These findings further highlight the complex surface-contour interactions that eventuate in the perception of coherent shapes.