Abstract
Figural status is determined in part by convexity and symmetry. Convex or symmetric regions are more likely to be seen as figures than adjacent equal-area concave or asymmetric regions, respectively. We investigated whether the strength of these cues increased with horizontal or vertical repetition. Horizontal repetition was manipulated by varying the number of adjacent black and white regions in a display. Displays were comprised of 3, 5, or 7 black and white regions. In a given display, either the black or the white regions were cued to be figure by convexity or symmetry. (Across displays half of the cued regions were black and half were white.) Within each of the 3 horizontal repetition conditions, vertical repetition was manipulated across 5 levels by varying the number of parts (delimited by successive concave cusps along the contour) from 1 – 16. 180 stimuli were displayed in each horizontal repetition condition. 60 displays tested the strength of symmetry alone, 60 tested convexity alone, and 60 tested the relative strength of symmetry and convexity. Displays were shown on a gray background for 100-ms. Observers reported whether the black or the white region appeared to stand out as figure against a background of the other color. (Responses based on color alone would not differ from chance.) Cues effectively determined figural status in all repetition conditions. In the competition condition, convexity dominated symmetry, ps < .002. Most importantly, the strength of the configural cues increased with the number of horizontal repetitions. Both symmetry and convexity determined figural status more effectively in the largest repetition condition than in the other repetition conditions, ps < .001. Repetitions in the vertical direction had no effect, however. These results indicate that global context modulates the strength of the configural cues. Additional experiments investigated the effects of vertical region repetition and horizontal part repetition.
Supported by NSF BCS 9906063.