August 2012
Volume 12, Issue 9
Vision Sciences Society Annual Meeting Abstract  |   August 2012
Decoding global contour perception in the human visual cortex
Author Affiliations
  • Shu-Guang Kuai
    School of Psychology, University of Birmingham, UK
  • Alan Meeson
    School of Psychology, University of Birmingham, UK
  • Zoe Kourtzi
    School of Psychology, University of Birmingham, UK\nLaboratory for Neuro- and Psychophysiology, K. U. Leuven, Belgium
Journal of Vision August 2012, Vol.12, 877. doi:
  • Views
  • Share
  • Tools
    • Alerts
      This feature is available to authenticated users only.
      Sign In or Create an Account ×
    • Get Citation

      Shu-Guang Kuai, Alan Meeson, Zoe Kourtzi; Decoding global contour perception in the human visual cortex. Journal of Vision 2012;12(9):877.

      Download citation file:

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

  • Supplements

Successful recognition of objects in cluttered scenes entails the integration of discrete elements to global contours. Previous studies have implicated a range of visual areas in contour integration, from primary visual cortex to higher occipitotemporal regions. However, the precise contribution of each of these regions to the perception of global contours remains unclear. Here, we compared behavioral and fMRI measurements to test for visual areas that mediate global contour perception. We presented observers with contours comprising of collinear Gabor elements that were embedded in a background of randomly oriented Gabors. We manipulated parametrically contour salience by adding local orientation jitter (±0º, ±10º, ±15º, ±20º, ±30º, or ±45º) to the contour elements. In an event-related fMRI design, we asked observers to judge whether the stimulus presented in each trial contained contours or not. The behavioral results showed that observers’ performance decreased sharply for local jitter between 15° and 30° rather than linearly with increasing jitter. We then compared these psychometric functions with fMR-metric functions computed using multi-voxel pattern analysis of fMRI signals in independently localized visual areas. In particular, we trained a linear SVM (support vector machine) classifier to predict whether the stimulus on each trial contained contours or not. We plotted the performance of the classifier across local orientation jitter for each visual area. Our findings showed close correspondence in human and classifier performance for higher occipitotemporal regions (V3B/KO, LO). In contrast, for early visual areas, classifier performance increased linearly with decreasing local orientation jitter. Our findings suggest that early visual areas may signal contour configurations based on local orientation distribution, while higher visual areas mediate the perception of global shape contours.

Meeting abstract presented at VSS 2012


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.