EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Generalizing biological surround suppression based on center surround similarity via deep neural network models

Xu Pan iD, Annie DeForge, Odelia Schwartz iD

PLOS Computational Biology · 22.09.2023 · 19 (9) · e1011486

DOI: 10.1371/journal.pcbi.1011486 ↗

Kurz gefasst

Fachartikel in der Zeitschrift PLOS Computational Biology, online erschienen am 22.09.2023. Verfasst von Xu Pan und 2 weiteren Autor:innen. Beteiligte Einrichtung: Department of Computer Science, University of Miami, Coral Gables, FL, United States of America. Hauptthemen laut MEDLINE: Visual Fields, Visual Cortex. Die Arbeit wurde bisher 4-mal zitiert (Crossref) und nennt 87 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: University of Miami, National Science Foundation.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Sensory perception is dramatically influenced by the context. Models of contextual neural surround effects in vision have mostly accounted for Primary Visual Cortex (V1) data, via nonlinear computations such as divisive normalization. However, surround effects are not well understood within a hierarchy, for neurons with more complex stimulus selectivity beyond V1. We utilized feedforward deep convolutional neural networks and developed a gradient-based technique to visualize the most suppressive and excitatory surround. We found that deep neural networks exhibited a key signature of surround effects in V1, highlighting center stimuli that visually stand out from the surround and suppressing responses when the surround stimulus is similar to the center. We found that in some neurons, especially in late layers, when the center stimulus was altered, the most suppressive surround surprisingly can follow the change. Through the visualization approach, we generalized previous understanding of surround effects to more complex stimuli, in ways that have not been revealed in visual cortices. In contrast, the suppression based on center surround similarity was not observed in an untrained network. We identified further successes and mismatches of the feedforward CNNs to the biology. Our results provide a testable hypothesis of surround effects in higher visual cortices, and the visualization approach could be adopted in future biological experimental designs.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Visual FieldsVisual CortexPhotic StimulationNeural Networks, ComputerNeuronsVisual Perception

Autor:innen und Institutionen

  • Department of Computer Science, University of Miami, Coral Gables, FL, United States of America.
  • School of Information, University of California, Berkeley, CA, United States of America.
  • Bentley University, Waltham, MA, United States of America.

ORCID-Profile: Xu Pan, Odelia Schwartz

Förderung

  • University of Miami
  • National Science Foundation

Zitieren

APA

Pan, X., DeForge, A., & Schwartz, O. (2023). Generalizing biological surround suppression based on center surround similarity via deep neural network models. PLOS Computational Biology, 19(9), e1011486. https://doi.org/10.1371/journal.pcbi.1011486

Vancouver

Pan X, DeForge A, Schwartz O. Generalizing biological surround suppression based on center surround similarity via deep neural network models. PLOS Computational Biology. 2023;19(9):e1011486. doi:10.1371/journal.pcbi.1011486

BibTeX
@article{Pan2023,
  author = {Pan, Xu and DeForge, Annie and Schwartz, Odelia},
  title = {Generalizing biological surround suppression based on center surround similarity via deep neural network models},
  journal = {PLOS Computational Biology},
  year = {2023},
  volume = {19},
  number = {9},
  pages = {e1011486},
  doi = {10.1371/journal.pcbi.1011486}
}
Kostenlos starten →