August 2016
Volume 16, Issue 12
Open Access
Vision Sciences Society Annual Meeting Abstract  |   September 2016
Dynamic flow of Face Categorization Task Information in an MEG Network.
Author Affiliations
  • N. Rijsbergen
    Institute of Neuroscience and Psychology, University of Glasgow
  • R. Ince
    Institute of Neuroscience and Psychology, University of Glasgow
  • G. Rousselet
    Institute of Neuroscience and Psychology, University of Glasgow
  • J. Gross
    Institute of Neuroscience and Psychology, University of Glasgow
  • P. Schyns
    Institute of Neuroscience and Psychology, University of Glasgow
Journal of Vision September 2016, Vol.16, 1235. doi:10.1167/16.12.1235
  • Views
  • Share
  • Tools
    • Alerts
      ×
      This feature is available to authenticated users only.
      Sign In or Create an Account ×
    • Get Citation

      N. Rijsbergen, R. Ince, G. Rousselet, J. Gross, P. Schyns; Dynamic flow of Face Categorization Task Information in an MEG Network. . Journal of Vision 2016;16(12):1235. doi: 10.1167/16.12.1235.

      Download citation file:


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

      ×
  • Supplements
Abstract

To categorize faces by their gender and expression, the brain combines task specific visual information with conceptual information about the target categories. In the hierarchical architecture of the early visual brain, the input face undergoes a complex integration of information gathered from independently processed contralateral visual fields. How does information integration interact with the demands of multiple categorization tasks? Here, three observers categorized by gender (2-AFC) and expression (7-AFC, expressions of emotion plus 'neutral') the same male and female expressive faces while we measured observers' single-trial MEG responses. On each of the 20,000 trials per task, with Bubbles (Gosselin & Schyns, 2001) we randomly sampled pixels from the original faces using Gaussian apertures distributed across 5 one-octave Spatial Frequency bands. With Information Theoretic measures (Mutual Information) applied to uni- and multi-variate reverse correlation analyses (Ince et al. 2015), we first computed the different face pixels associated with behavior in each categorization task. Then, applying the same analyses to each voxel of MEG time series, we reconstructed the dynamic processing flow of task-relevant face pixels in the brain (Task Information, S1, one observer) and we further analyzed the flow of the task-specific features in the MEG voxels (Left vs. Right Face Information, S1). In the reconstructed information categorization task and feature information flows, we reveal a common process of visual hemifield integration whereby left and right occipito-temporal regions initially (100-150ms post stimulus) code face features from the contra-lateral visual hemifield, followed, primarily in one hemisphere (left in S1), by a period of bilateral face coding when face features are categorization dependent and integrated. Following integration, visual and task information spatially and temporally dissociate. The occipito-temporal regions keep coding of task-related face features whereas the categorization task itself shows a secondary parietal peak at 250-300ms.

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.

×