December 2011
Volume 11, Issue 15
Free
OSA Fall Vision Meeting Abstract  |   December 2011
Image Correlates of Peripheral Contour Discrimination in Natural Scenes
Author Affiliations
  • Thomas Wallis
    Schepens Eye Research Institute, Ophthalmology
  • Peter Bex
    Schepens Eye Research Institute, Ophthalmology & Harvard Medical School
Journal of Vision December 2011, Vol.11, 62. doi:10.1167/11.15.62
  • Views
  • Share
  • Tools
    • Alerts
      ×
      This feature is available to authenticated users only.
      Sign In or Create an Account ×
    • Get Citation

      Thomas Wallis, Peter Bex; Image Correlates of Peripheral Contour Discrimination in Natural Scenes. Journal of Vision 2011;11(15):62. doi: 10.1167/11.15.62.

      Download citation file:


      © ARVO (1962-2015); The Authors (2016-present)

      ×
  • Supplements
Abstract

Natural visual environments are replete with contours, and the phenomenon of crowding demonstrates that contours are difficult to discriminate when viewed peripherally. Here we examined the discrimination of spatial structure in natural scenes by asking observers to locate a circular patch of artificial contours superimposed on a greyscale natural image. These contours were “dead leaves”: ellipses of random size and aspect ratio matched in luminance and contrast to the image segment they replaced. Three observers identified the location of the dead leaves patch relative to fixation (N, S, E or W) at three eccentricities (2, 4 and 8°), with the size of the patch varied by an adaptive staircase. A logistic GLM was used to model task performance as a function of both the manipulated task parameters of patch size, eccentricity and visual field location, as well as the stochastically varying local image statistics of the background image. Hierarchical model selection suggests that, after considering task parameters, the local RMS contrast in an image segment and correlated statistics such as edge density and phase congruence are the most significant predictors of task performance. These models give principled predictions of the likelihood of crowding in a given natural image segment.

×
×

This PDF is available to Subscribers Only

Sign in or purchase a subscription to access this content. ×

You must be signed into an individual account to use this feature.

×