June 2006
Volume 6, Issue 6
Vision Sciences Society Annual Meeting Abstract  |   June 2006
Feature attention in motion perception - a computational account
Author Affiliations
  • Pierre Bayerl
    Dept. of Neural Information Processing, University of Ulm, Germany
  • Heiko Neumann
    Dept. of Neural Information Processing, University of Ulm, Germany
Journal of Vision June 2006, Vol.6, 513. doi:https://doi.org/10.1167/6.6.513
  • Views
  • Share
  • Tools
    • Alerts
      This feature is available to authenticated users only.
      Sign In or Create an Account ×
    • Get Citation

      Pierre Bayerl, Heiko Neumann; Feature attention in motion perception - a computational account. Journal of Vision 2006;6(6):513. https://doi.org/10.1167/6.6.513.

      Download citation file:

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

  • Supplements

Introduction. Experimental studies on feature attention in visual motion perception revealed effects at the neural and behavioral level of observation. However, it remains unclear how such results from different studies relate. In this work we build upon a neural model of cortical motion perception (Bayerl, Neumann. Neural Computation, 2004) to explain distinct experimental observations within a single theoretical framework. Model. The model consists of a layered architecture simulating the function of primate areas V1 and MT. Model neurons in each area are coupled via feed-forward integration, lateral competition and feedback modulation. To model feature attention the model dynamics is influenced, or biased, by an excitatory top-down attention signal indicating attended motion features. We use our model to link a physiological study of feature attention in cortical motion processing (Martinez-Trujillo, Treue. Current Biology, 2004) to a psychophysical experiment of motion perception (Felisberti, Zanker. Vision Research, 2005). Results. Via computer simulations, the model accounts for physiological data on feature attention and generates behavioral data in a decision experiment that is consistent with psychophysical observations. Furthermore, our investigations predict a decreased performance in motion detection tasks when the wrong direction of motion in attended. In sum, the model explains experimental findings from behavioral as well as physiological investigations. The key underlying function relates to the biased competition framework by utilizing soft-gating modulatory feedback combined with shunting competition for activity normalization. The model itself is able to account for experimental findings as well as to process real-world sequences.

Bayerl, P. Neumann, H. (2006). Feature attention in motion perception - a computational account [Abstract]. Journal of Vision, 6(6):513, 513a, http://journalofvision.org/6/6/513/, doi:10.1167/6.6.513. [CrossRef]

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.