August 2014
Volume 14, Issue 10
Vision Sciences Society Annual Meeting Abstract  |   August 2014
Effects of spatial frequency filtering in natural scenes: Evidence from eye movements and computational modeling
Author Affiliations
  • Anke Cajar
    Cognitive Sciences, University of Potsdam
  • Jochen Laubrock
    Cognitive Sciences, University of Potsdam
  • Ralf Engbert
    Experimental and Biological Psychology, University of Potsdam
Journal of Vision August 2014, Vol.14, 1207. doi:
  • Views
  • Share
  • Tools
    • Alerts
      This feature is available to authenticated users only.
      Sign In or Create an Account ×
    • Get Citation

      Anke Cajar, Jochen Laubrock, Ralf Engbert; Effects of spatial frequency filtering in natural scenes: Evidence from eye movements and computational modeling. Journal of Vision 2014;14(10):1207.

      Download citation file:

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

  • Supplements

When looking at a scene, the foveal and peripheral visual field serve different functions. The fovea is most sensitive to high spatial frequencies and ideally suited for object identification and the analysis of details, while the peripheral field is sensitive to low spatial frequencies for saccade target selection and processing transients. How do spatial frequencies affect eye movements during scene viewing? We investigated this question in several experiments by selectively attenuating parts of the spatial frequency spectrum in the foveal or peripheral visual field. This was achieved by applying gaze-contingent high-pass or low-pass filters to natural scenes either in the foveal or the peripheral field of the viewer, thus simulating a foveal scotoma or tunnel vision. Compared to an unfiltered control condition, fixation durations increased with filter conditions where processing was predicted to be easier (foveal high-pass and peripheral low-pass filtering), while fixation durations were less affected when processing was predicted to be more difficult (foveal low-pass and peripheral high-pass filtering). These counterintuitive results challenge current theories of eye-movement control, which expect a positive correlation of fixation durations and (foveal) processing difficulty. We therefore implemented a computational model with two spatial compartments where foveal and peripheral processing interact to control fixation durations. The model reproduced experimental distributions and mean values of fixation durations by varying few parameters that were affected by the specific filtering conditions. The modeling results suggest that (1) peripheral information is critical for the control of fixation durations, and (2) foveal and peripheral information processing evolve in parallel and interact.

Meeting abstract presented at VSS 2014


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.