Purchase this article with an account.
Rosemary Cowell, John Serences; Feature-coding transitions to conjunction-coding with progression through visual cortex. Journal of Vision 2016;16(12):755. doi: 10.1167/16.12.755.
Download citation file:
© 2017 Association for Research in Vision and Ophthalmology.
Evidence from electrophysiological studies in animals suggests that the visual object processing pathway in cortex analyzes incoming information in a staged, hierarchical manner. Neurons in early stages of the pathway are tuned to simple visual features (e.g., a line of a particular orientation) whereas neurons in later stages are selective for increasingly complex stimulus attributes (e.g., a collection of lines corresponding to a complex shape). It is widely assumed that feature-coding dominates in early visual cortex whereas later visual cortices employ conjunction-coding in which whole object representations are different from the sum of their simple-feature parts. However, most electrophysiological and neuroimaging studies have measured only a small span of the cortical hierarchy or manipulated stimulus properties at only one level of visual complexity. No study in humans has simultaneously demonstrated that putative object-codes in higher visual cortex cannot be accounted for by feature-coding and that putative feature-coding in early visual cortex is not equally well characterized as an object-code. We present a novel method that employs multivariate analysis of functional brain imaging data to measure feature-coding and conjunction-coding directly and pit them against each other throughout visual cortex. The results provide the first direct demonstration of a continuous gradient from feature-coding in primary visual cortex to conjunction-coding in inferior temporal and posterior parietal cortices. This novel method enables the use of classifier analyses along with experimentally controlled visual stimuli to investigate population-level feature- and conjunction-codes throughout human cortex.
Meeting abstract presented at VSS 2016
This PDF is available to Subscribers Only