Purchase this article with an account.
AK Nugent, RL Woods, E Peli; Flanker Size Affects Visual Lateral Interactions . Invest. Ophthalmol. Vis. Sci. 2002;43(13):4718.
Download citation file:
© ARVO (1962-2015); The Authors (2016-present)
Purpose: Polat and Sagi (1993, 1994) showed that laterally displaced masks (flankers) could facilitate contrast detection of gabor patches, independent of the size of the patches. However, when spatial scaling of these flanker effects was demonstrated, stimulus bandwidth was confounded with spatial frequency. In a previous study (Woods et al., in press), we varied the bandwidth (Gabor patch size from σ = 0.5λ to 1.5λ) of flanker and test stimulus simultaneously and showed that bandwidth affected facilitation. However, since we altered stimulus and flanker bandwidth simultaneously, it is possible that the effects were not a consequence of bandwidth alone. The results may have been influenced by the greater overlap of the test stimulus and flankers when the patches were larger (masking). To examine this confound we varied the vertical and horizontal size, separately, of the flankers while keeping the size of the test stimulus fixed. Methods: In the first experiment, the height of the flankers was varied from σy = 0.5λ to 3λ. In the second experiment, the width of the flankers was varied from σx = 0.125λ to 2λ. In both experiments, test to flanker distance was 3λ. All conditions were tested at 2, 4, and 8 cycles/degree. Three subjects in each experiment were tested in a 2AFC paradigm. Results: With flanker heights σy = 0.5λ to 1.5λ (no overlap with test patch) facilitation was about equal. As the flankers began to overlap the test stimulus (σy = 2λ to 3λ) detection contrast thresholds increased dramatically. When flanker width was varied, maximum facilitation was found at σx = 0.75λ and facilitation decreased as flanker width increased. All these effects were more pronounced at the lower spatial frequency (2 cycles/degree). Conclusion: Masking was a major component of the effects found in our previous study, as illustrated by the effects of flanker height. However, there was a bandwidth component, such that a smaller bandwidth (i.e. a wider flanker) decreased facilitation. This is not consistent with simple additive rules that could be derived from the results of previous studies. These results need to be reconciled with proposed models of lateral interactions.
This PDF is available to Subscribers Only