September 2019
Volume 19, Issue 10
Open Access
Vision Sciences Society Annual Meeting Abstract  |   September 2019
The neural basis of local contour symmetry in scene perception
Author Affiliations & Notes
  • John D Wilder
    Department of Psychology, University of Toronto
  • Morteza Rezanejad
    School of Computer Science, Centre for Intelligent Machines, McGill University
  • Kaleem Siddiqi
    School of Computer Science, Centre for Intelligent Machines, McGill University
  • Allan Jepson
    Department of Computer Science, University of Toronto
  • Sven Dickinson
    Department of Computer Science, University of Toronto
  • Dirk B Walther
    Department of Psychology, University of Toronto
Journal of Vision September 2019, Vol.19, 189a. doi:https://doi.org/10.1167/19.10.189a
  • Views
  • Share
  • Tools
    • Alerts
      ×
      This feature is available to authenticated users only.
      Sign In or Create an Account ×
    • Get Citation

      John D Wilder, Morteza Rezanejad, Kaleem Siddiqi, Allan Jepson, Sven Dickinson, Dirk B Walther; The neural basis of local contour symmetry in scene perception. Journal of Vision 2019;19(10):189a. doi: https://doi.org/10.1167/19.10.189a.

      Download citation file:


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

      ×
  • Supplements
Abstract

The visual system is tasked with turning light impinging onto our retinas into meaningful percepts. Providing a strong cue to the presence of objects and surfaces, symmetry is one of the Gestalt principles underlying the grouping of visual information in mid-level vision. In fact, it has recently been demonstrated that there is a perceptual advantage for symmetry for the perception of complex real-world scenes. Here we uncover the neural basis of the perceptual advantage of symmetry. Specifically, we show how local contour symmetry modulates neural activity patterns elicited by complex scenes throughout visual cortex. Participants were shown full, intact line drawings of scenes as well as line drawings containing the most symmetric/asymmetric half of the contour pixels. We decoded scene categories from the voxel activity in early visual cortex as well as high-level, scene-selective brain regions. Decoding of scene categories was more accurate in the symmetric than the asymmetric condition in PPA, OPA, and LOC. In areas V1–4, on the other hand, scene categories were more accurately decoded in the asymmetric than the symmetric condition. A searchlight analysis yielded consistent results, showing that mid- and high-level areas show higher decoding accuracy in the symmetric condition, while the opposite is true for early visual cortex. Our results indicate a crucial role for symmetry in grouping visual information into meaningful units for high-level processing. In fact, our findings in high-level regions mirror the perceptual advantage of symmetry in behavioral categorization performance. Early visual cortex, on the other hand, does not appear to benefit from local symmetry as a grouping cue. On the contrary, the information redundancy inherent to symmetry leads to lower fidelity of the category-specific signal in V1–4. In summary, we here demonstrate the emergence of symmetry as a mid-level grouping cue for real-world scenes along the visual processing hierarchy.

Acknowledgement: Sony, Samsung, SSHIRC, NSERC 
×
×

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.

×