August 2016
Volume 16, Issue 12
Open Access
Vision Sciences Society Annual Meeting Abstract  |   September 2016
Visual Search for Multiple Targets in Probabilistic Environments
Author Affiliations
  • Yelda Semizer
    Rutgers University
  • Kimele Persaud
    Rutgers University
  • Xiaoli He
    Rutgers University
  • Nicholas Kleene
    Rutgers University
  • Omer Tanrikulu
    Rutgers University
Journal of Vision September 2016, Vol.16, 988. doi:
  • Views
  • Share
  • Tools
    • Alerts
      This feature is available to authenticated users only.
      Sign In or Create an Account ×
    • Get Citation

      Yelda Semizer, Kimele Persaud, Xiaoli He, Nicholas Kleene, Omer Tanrikulu; Visual Search for Multiple Targets in Probabilistic Environments . Journal of Vision 2016;16(12):988. doi:

      Download citation file:

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

  • Supplements

Active visual search requires finding targets in environments where target presence in local regions is probabilistic and search is costly in time or energy. Efficient strategies take both probabilities and costs into account (Hutchinson et al. 2008; Wolfe, 2013). To understand search strategies, we studied active search for hidden targets (revealed by mouse clicks) when location probabilities and costs varied. Three virtual "rooms" were displayed, each with 4 possible locations (colored squares) that might contain a hidden target. The probability of a target being present in any location was chosen independently and indicated by the color of the square. Probabilities were chosen such that the overall probability of finding targets in any given room was different. Points were rewarded for targets found, and deducted for each location searched. Participants navigated between rooms using a mouse, attempting to accumulate as many points as possible. Across blocks we manipulated search costs: points deducted/search, the time to travel to a room, and the total time/trial. Participants were tested over 8 days, 1120 trials. Subjects' search strategies favored rooms with higher overall probabilities, particularly with the stricter time limits. Time penalties (points deducted/location searched; travel time between rooms) were less influential. Strategies were biased by the success of recent outcomes, despite the fact that the target probabilities were independent. In most cases, observers under-performed relative to models that took into account both probability and time costs. These results suggest that human observers were biased to overweigh cues about probabilities and recent search experience, and underweight the role of time costs.

Meeting abstract presented at VSS 2016


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.