Purchase this article with an account.
Sing-Hang Cheung, Gordon E. Legge, Susana T. L. Chung, Bosco S. Tjan; Target-flanker binding releases crowding. Journal of Vision 2006;6(6):807. doi: 10.1167/6.6.807.
Download citation file:
© ARVO (1962-2015); The Authors (2016-present)
Purpose. The interfering effect of letters on the recognition of adjacent letters is known as crowding. According to an Attention explanation, crowding occurs when visual attention cannot individualize the target. Alternatively, according to a Feature-integration explanation, the bottom-up interference among features of the target and the flankers results in crowding. Both explanations predict a release from crowding when the target and flankers touch, and are perceived as a single object. We tested this counter-intuitive prediction with target-flanker-touching and non-touching conditions. Methods. Three normally-sighted observers performed an orientation-judgment task with a 2-deg “T”, presented at 8 deg in the right visual field, flanked by four similar T's. The flankers were oriented with the leg of the T pointing away from the target. The target was always darker than the background. Flankers were either darker than the background in the same-polarity condition or brighter in the opposite-polarity condition. Center-to-center distances between the target and the flankers were 2 deg (touching) or 2.5 deg (non-touching). Task accuracy was assessed with 200 trials at each of the seven contrast levels from 3% to 100%. Results. Compared with the non-touching condition, the touching condition yielded better accuracy (mean difference: 4.19 ± 1.84%) in the same-polarity condition, but worse accuracy (mean difference: −6.71 ± 0.84%) in the opposite-polarity condition. Conclusions. Reduced crowding was found in the touching condition, only when both target and flankers have the same contrast polarity. Release from crowding occurs when the visual system interprets the target and flankers as a single object.
This PDF is available to Subscribers Only