Global form perception: interactions between luminance and texture information.

Abstract

This experiment reports the independence of first- and second-order processing mechanisms in form perception. Symmetrical dot patterns were created using either luminance-increment dots (luminance above background), or texture-defined dots (average luminance equal to background). The proportion of luminance increment or texture dots defining each pattern was varied among fields of noise dots of the same type to determine symmetry detection thresholds. Differences in detection thresholds were found between luminance- and texture-defined patterns. Further, symmetry detection thresholds for luminance-increment dot patterns were resistant to noise defined by dots of opposite contrast polarity (luminance-decrement dots) or texture, while texture-defined patterns were resistant to neither texture nor luminance-decrement noise. These data suggest that symmetry perception, along with other types of form perception, use both first- and second-order processing mechanisms. The data are compatible with a second-order system that includes a negative half-wave rectifying non-linearity.