Separation Discrimination in the Presence of Flanking Lines

Abstract

The presence of flanking stimuli decreases sensitivity in many visual tasks. Earlier we showed that size perception depends on the distance between objects (Bondarko and Danilova, 1996 Perception25 Supplement, 94). Now we suggest that flanks not only decrease sensitivity, but also lead to misperception of size. To test this hypothesis we studied the perception of spatial interval for a set of separations from 6 to 33 min−1 in the presence of two vertical flanking lines. The method of constant stimuli was used in a temporal 2AFC paradigm. Reference stimuli were two vertical lines with separation S, either with or without two symmetrical external flanking lines. The distance to the flanking lines varied from 1/4 S to 3 S in steps of 1/4 S. The test stimuli were just two lines. The first stimulus in the pair could be either the test or the reference. Observers had to report whether the distance between the two central lines in the first stimulus were smaller or larger than the distance in the second stimulus. Probit analysis was used to estimate the threshold and the point of subjective equality (PSE). We found that the PSE implied overestimation by up to 10% of the spatial intervals for the smallest distances to the flanking lines. Increasing distances to the flanking lines finally led to slight underestimation. The strength of the illusion depended on the order of presentation of the reference and test stimuli. Thresholds increased for the smallest distances to flanking lines. Our results confirmed the hypothesis that flanks can induce distortions in size perception. We suggest that our experimental situation could be considered as a one-dimensional case of geometrical visual illusions such as the Ebbinghaus illusion.