September 2019
Volume 19, Issue 10
Open Access
Vision Sciences Society Annual Meeting Abstract  |   September 2019
The Coding of Color, Shape, and their Conjunction Across the Human Ventral Visual System
Author Affiliations & Notes
  • JohnMark E Taylor
    Department of Psychology, Harvard University
  • Yaoda Xu
    Department of Psychology, Yale University
Journal of Vision September 2019, Vol.19, 89a. doi:
  • Views
  • Share
  • Tools
    • Alerts
      This feature is available to authenticated users only.
      Sign In or Create an Account ×
    • Get Citation

      JohnMark E Taylor, Yaoda Xu; The Coding of Color, Shape, and their Conjunction Across the Human Ventral Visual System. Journal of Vision 2019;19(10):89a.

      Download citation file:

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

  • Supplements

How does the brain combine color and shape information into integrated representations of whole objects? Some evidence suggests that color and shape are initially processed independently, requiring a subsequent binding step; this evidence includes the existence of illusory conjunctions, slower visual search for feature conjunctions, and neuropsychological dissociations between color and shape processing. Other psychophysical and neural results, however, suggest that color and shape information is integrated at very early stages of visual processing, seemingly dissolving the binding problem. To reconcile these two seemingly disparate sets of findings, in two fMRI MVPA experiments using the methodology developed by Seymour et al., (2009, Cerebral Cortex), we examined the coding of color, shape, and their conjunction in a set of early visual areas and higher ventral visual regions known to be involved in object shape or color processing. We examined neural coding both for the conjunction of colors with simple shapes (spirals) and with more complex shapes, testing whether brain regions could decode single features, and whether they could discriminate two pairs of stimuli containing the same four features conjoined in different ways—a direct test for conjunctive coding. We found evidence of simple color/shape conjunction coding in early visual regions, but not in any of the higher-level visual regions examined. By contrast, we found no evidence for conjunctive coding of complex color/shape combinations. Meanwhile, we found that color and shape information was decodable across all of early visual cortex and much of the ventral visual pathway, even in higher ventral regions defined by their selectivity to shape or color. Taken together, these results are consistent with the hypothesis that the visual system can directly encode simple color/shape conjunctions with dedicated “conjunction detectors”, but would need employ a different strategy for more complex features to avoid the problem combinatorial explosion.

Acknowledgement: National Science Foundation Graduate Research Fellowship 

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.