September 2011
Volume 11, Issue 11
Free
Vision Sciences Society Annual Meeting Abstract  |   September 2011
Asymmetry and similarity phenomena in backwards masking experiments suggest reentrant processing
Author Affiliations
  • Tsvi Achler
    Synthetic Cognition Group, Los Alamos National Labs, USA
  • Mike Ham
    Synthetic Cognition Group, Los Alamos National Labs, USA
  • Shawn Barr
    Synthetic Cognition Group, Los Alamos National Labs, USA
  • John George
    Synthetic Cognition Group, Los Alamos National Labs, USA
  • Jason McCarley
    Cognitive Psychology, University of Illinois Urbana-Champaign, USA
  • Garrett Kenyon
    Synthetic Cognition Group, Los Alamos National Labs, USA
  • Luis Bettencourt
    Synthetic Cognition Group, Los Alamos National Labs, USA
Journal of Vision September 2011, Vol.11, 1230. doi:https://doi.org/10.1167/11.11.1230
  • Views
  • Share
  • Tools
    • Alerts
      ×
      This feature is available to authenticated users only.
      Sign In or Create an Account ×
    • Get Citation

      Tsvi Achler, Mike Ham, Shawn Barr, John George, Jason McCarley, Garrett Kenyon, Luis Bettencourt; Asymmetry and similarity phenomena in backwards masking experiments suggest reentrant processing. Journal of Vision 2011;11(11):1230. https://doi.org/10.1167/11.11.1230.

      Download citation file:


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

      ×
  • Supplements
Abstract

Non-linear effects can be observed in fast backwards-masking psychophysics experiments. Such experiments show that: 1) when targets and masks are similar they require longer processing times 2) patterns with unique features are processed quicker, displaying an asymmetry 3) these properties can also determine whether masking functions follow monotonic or U-shape curves. Hypotheses regarding the structure of neural processing involved in rapid image presentations range from pure feedforward models to models with lateral connections and top-down reentrant feedback. Although the prevailing interpretation has been that for very fast image presentations under effective masking, a feedforward model is adequate, there is now growing evidence against this possibility. It remains unclear if the earliest signals of object detection in the brain are connected with object identification or more general statistical properties of the signal and 150 ms may be enough time for limited feedback processing. Furthermore it is possible that simultaneous processing of both the target and mask continues until the person responds. We construct and test a set of hypotheses for simultaneous and serial processing of the target and mask while evaluating the consequences of different cortical dynamics on speed of sight psychophysics experiments. We demonstrate that models whose dynamics are essentially feedforward HMAX-like cannot easily account for effects of differential masking and asymmetries in image sequences, because they are not well-suited for simultaneous processing. We then show that a simple model with top-down feedback naturally accounts for these properties, suggesting that even for very short presentation times reentrant connections play an important role in visual perception in humans.

×
×

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.

×