August 2012
Volume 12, Issue 9
Free
Vision Sciences Society Annual Meeting Abstract  |   August 2012
Using decision models to study the time course of visual recognition
Author Affiliations
  • Imri Sofer
    Brown University
  • Thomas Serre
    Brown University
Journal of Vision August 2012, Vol.12, 160. doi:https://doi.org/10.1167/12.9.160
  • Views
  • Share
  • Tools
    • Alerts
      ×
      This feature is available to authenticated users only.
      Sign In or Create an Account ×
    • Get Citation

      Imri Sofer, Thomas Serre; Using decision models to study the time course of visual recognition. Journal of Vision 2012;12(9):160. https://doi.org/10.1167/12.9.160.

      Download citation file:


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

      ×
  • Supplements
Abstract

Primates’ ability to recognize objects in natural scenes is remarkable. As exemplified by rapid stimulus presentation paradigms, the visual system is both fast and accurate. Computational models have been proposed that predict the level of performance of human participants and how recognition may be affected by visual properties of images. However, these computational models do not make any predictions about reaction times, and cannot explain the time course of information accumulation in the visual cortex. Here we present an initial attempt to fill this gap using a decision model that allows for analysis of both behavioral responses and decision times within a unified framework.

Participants performed an object recognition task in natural scenes using a backward masking paradigm with varying stimulus onset asynchrony (SOA) conditions. We estimated decision-related parameters for the task using a hierarchical drift diffusion model, an extension of the most widely used decision model. We examined how the drift rate, a parameter associated with task difficulty, can be used to explain the results, and show that changes in the drift rate alone does not seem to account for the distribution of reaction times under different masking conditions. Interestingly we find that both the SOA and image properties affect the variance of the drift rate, and that this change does not seem to simply reflect variability in the stimulus properties across conditions. We speculate that it may reflect multiple processing strategies employed by the visual system to process information. Our results suggest that decision models may constitute a promising tool for understanding the brain mechanisms underlying object recognition.

Meeting abstract presented at VSS 2012

×
×

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.

×