August 2016
Volume 16, Issue 12
Open Access
Vision Sciences Society Annual Meeting Abstract  |   September 2016
Learning to search for two targets with unequal occurrence rates: The role of short-term versus long-term learning
Author Affiliations
  • Sha Li
    Department of Psychology, University of Minnesota
  • Yuhong V. Jiang
    Department of Psychology, University of Minnesota
Journal of Vision September 2016, Vol.16, 18. doi:
  • Views
  • Share
  • Tools
    • Alerts
      This feature is available to authenticated users only.
      Sign In or Create an Account ×
    • Get Citation

      Sha Li, Yuhong V. Jiang; Learning to search for two targets with unequal occurrence rates: The role of short-term versus long-term learning. Journal of Vision 2016;16(12):18. doi:

      Download citation file:

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

  • Supplements

Selective attention is strongly influenced by the location or feature probability of a search target. Previous visual search studies showed that both long-term statistical learning and short-term inter-trial priming contributed to attentional learning of a target's location probability. Yet the relative contribution of long-term and short-term mechanisms in attention learning of a target's feature probability remains unclear. Here we investigated how people searched for two potential targets that appeared with unequal occurrence rates. We examined time course, persistence and primacy effects of feature probability learning. Participants searched for two pre-specified target colors, one of which appeared on each trial, and reported the target's orientation. The two targets appeared with unequal probability, requiring participants to learn to adjust their search priority. In three experiments, we showed that participants rapidly acquired an attentional preference for the more probable target. Unlike location probability learning and other forms of visual statistical learning, target probability learning quickly extinguished when the two targets became equally probable. In addition, when the two targets' probability reversed in the experiment, participants also rapidly adjusted to the new probability structure. Learning and the adjustment of learning were unrelated to explicit awareness about the targets' probability. These results indicate that, unlike location probability learning, target probability learning reflects short-term learning. Long-term learning was absent or trivial relative to the influence of short-term search history. Short-term repetition priming provided a reasonable account for the rapid acquisition and rapid extinction of the feature probability effect. We propose that people optimize the search template by considering very recent search history.

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.