August 2023
Volume 23, Issue 9
Open Access
Vision Sciences Society Annual Meeting Abstract  |   August 2023
What does it mean to be a scene: evidence from full-field fMRI
Author Affiliations & Notes
  • Jeongho Park
    Harvard University
  • Talia Konkle
    Harvard University
  • Footnotes
    Acknowledgements  R21EY031867
Journal of Vision August 2023, Vol.23, 4968. doi:https://doi.org/10.1167/jov.23.9.4968
  • Views
  • Share
  • Tools
    • Alerts
      ×
      This feature is available to authenticated users only.
      Sign In or Create an Account ×
    • Get Citation

      Jeongho Park, Talia Konkle; What does it mean to be a scene: evidence from full-field fMRI. Journal of Vision 2023;23(9):4968. https://doi.org/10.1167/jov.23.9.4968.

      Download citation file:


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

      ×
  • Supplements
Abstract

What is a “scene”? Content-based hypotheses emphasize in particular the importance of spatial layout features and other image statistics, dissociating scenes from objects. According to a more ego-centric hypothesis, however, any information in the far-periphery could reasonably be considered part of the current “scene” in view. Indeed, classic scene-selective regions are known to have response preferences for both scene content (relative to faces and objects), as well as for general peripheral stimulation (regardless of what kind of content is depicted). However, these findings have been limited to tests over central 10°-20° visual field, given constraints of standard functional neuroimaging projection setups. Here, we leveraged our new method for ultra-wide angle projection (>175°) to investigate how far-peripheral stimulation drives scene-selective regions. We scanned participants (n=12) viewing images of full-field scenes, full-field faces-arrays, and full-field object-arrays, presented with varying degrees of scotoma—masking the center of the image with a diameter of 0°, 30°, 60°, 90°, or 140°. We found that all scene-selective regions maintained their preference for scene content over other kinds of image content, at each scotoma level (except OPA at the farthest periphery). Strikingly, we also found that the fusiform face-selective region maintained its selectivity even when faces were presented only in the far-periphery. Only the far-peripheral early visual areas (V1-V3) responded similarly across different visual content. Taken together, our results demonstrate that not just any information in the far-periphery is automatically integrated into the computations of scene-selective regions. Instead, the results indicate a neural distinction between a “scene” and a “view,” where the scene regions primarily encode particular higher-order image information present in scenes and not objects, while far-peripheral retinotopic cortex encodes information about the current view.

×
×

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.

×