Item

Abstract

We measured spatial and temporal characteristics of mechanisms that bridge gaps between line segments. The presentation time was measured which was necessary for localization and identification of a triangular shape made up of (a) pacmen, (b) pacmen with lines, (c) lines, (d) line segments (corners) or (e) pacmen with circles (amodal completion). The triangle was embedded in a field of distractors of the same components at random orientations. Subjects had to indicate whether the triangle was to the left or to the right of the midline (localization) and whether it was pointing upward or downward (identification). Poststimulus masks consisted of pinwheels (a, b, e) or randomly oriented lines (c, d). Stimuli were presented on a gray background and defined by luminance or isoluminant contrast. Thresholds were fastest when the triangle was defined by real contours (b: 98 msec; c: 106 msec), slightly slower for corners and pacmen (d:129 msec; a: 157 msec), and much slower for the amodally completed pacmen (e: 355 msec). For all pattern types localization was about 20 msec faster than identification. Compared to low contrast luminance stimuli, processing of isoluminant stimuli was equally fast for targets defined by real contours (c), but much slower for illusory contours (a). We conclude that speed of contour integration depends greatly on the spatial configuration of the stimulus, but not directly on the formation of illusory contours. The contour integration process is impaired under conditions of isoluminance, but not the perception of contours per se.