August 2023
Volume 23, Issue 9
Open Access
Vision Sciences Society Annual Meeting Abstract  |   August 2023
Probabilistic attentional selection during continuous visual search
Author Affiliations & Notes
  • Jennifer Magerl Fuller
    University of Iceland
  • Vladislav Khvostov
    University of Dayton
  • Árni Kristjánsson
  • Árni Gunnar Ásgeirsson
    Shroff Charitable Eye Hospital
  • Footnotes
    Acknowledgements  This project was supported by grant #228366-051 from the Icelandic Research Fund.
Journal of Vision August 2023, Vol.23, 5663. doi:
  • Views
  • Share
  • Tools
    • Alerts
      This feature is available to authenticated users only.
      Sign In or Create an Account ×
    • Get Citation

      Jennifer Magerl Fuller, Vladislav Khvostov, Árni Kristjánsson, Árni Gunnar Ásgeirsson; Probabilistic attentional selection during continuous visual search. Journal of Vision 2023;23(9):5663.

      Download citation file:

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

  • Supplements

To successfully orient ourselves within noisy visual environments, we must be able to focus our attention on items of importance, whilst ignoring sources of distraction. A common assumption has been that this selective visual attention is facilitated by templates, tuned towards our current goals. However, in real-world scenes, colour and luminance values vary greatly due to perceptual interpretation and environmental factors, such as lighting changes and occlusion. Therefore, tuning attentional templates probabilistically might be a more efficient strategy than tuning them to precise values. This seems particularly important during continuous tasks such as when we have to keep track of an object, or select multiple objects that share properties. Studies of attention have not determined whether templates are tuned to exact colour values or an underlying probability distribution of feature values. The current experiment addresses this, by investigating the effects of variability in target and distractor identity, using a novel foraging task. Participants (N=11) were presented with 40 target objects to continuously select, until total depletion, during a single trial. Targets were drawn from a Gaussian distribution, sampled from a linearized colour space of 48 isoluminant hues. The 40 distractor objects were sampled from a uniform distribution, at the opposite end of the colour wheel. We report the absolute distance of each target from the mean, and the impact of previous target colour selection on current colour selection by consecutive selection episodes. We also report the speed of learning, and the impact of spatial proximity on target selection. Our preliminary results suggest a selection priority pattern favouring colour values closer to the mean of the distribution.


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.