September 2015
Volume 15, Issue 12
Vision Sciences Society Annual Meeting Abstract  |   September 2015
Investigating Confirmation Bias in Overt Visual Selection
Author Affiliations
  • Jason Rajsic
    Psychology, University of Toronto
  • Daryl Wilson
    Psychology, Queen's University
  • Jay Pratt
    Psychology, University of Toronto
Journal of Vision September 2015, Vol.15, 1356. doi:
  • Views
  • Share
  • Tools
    • Alerts
      This feature is available to authenticated users only.
      Sign In or Create an Account ×
    • Get Citation

      Jason Rajsic, Daryl Wilson, Jay Pratt; Investigating Confirmation Bias in Overt Visual Selection. Journal of Vision 2015;15(12):1356.

      Download citation file:

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

  • Supplements

In visual search, attention is biased towards stimuli that confirm a target’s presence. Although advantageous in present/absent target searches, we have shown that search can be arbitrarily biased towards one of two target stimuli, indicating a confirmation bias for the target conjunction framed as the search template. This conclusion, however, assumes that participants performed a simple, serial visual search in our task. To test this assumption, we measured behavioral and oculomotor performance in standard and gaze-contingent versions of our dual-target search task. Searches consisted of eight colored circles, seven containing distractor letters, and one with a target letter. Participants were instructed to press one key if the target letter appeared on a circle with a specified color, and to press another key if the target letter appeared on a circle in the non-specified color. We varied the number of circles that matched the specified color and the color that the target letter appeared on. We predicted that circles matching the specified color would be attended first, as these circles would confirm the presence of the target conjunction. In the standard condition, eye positions were simply recorded to measure spontaneous search patterns. In the gaze contingent condition, letters were only shown when circles were fixated. This allowed us to assess search strategy when searches were necessarily serial. The results of the standard search demonstrated confirmation bias in oculomotor selection; both search time and number of fixations increased as more circles in the specified color were present on a trial. However, the gaze contingent condition did not exhibit confirmation bias, either in total fixations or search time. We conclude that gaze contingent search leads to different selection dynamics than search when all stimuli are visible, weakening the bias towards confirmatory stimuli.

Meeting abstract presented at VSS 2015


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.