Signatures of natural image statistics in cortical simple cell receptive fields: Computational Neuroscience: Trends in Research 2001

Norbert Mayer, J.Michael Herrmann, Theo Geisel

Research output: Contribution to journalArticlepeer-review

Abstract

Receptive fields of simple cells in visual cortex in cats are known to be shaped like Gabor functions with respect to retinal coordinates. As shown by Miller (J. Neurosci. 14 (1994) 409), oriented receptive fields may emerge prenatally from spontaneous and unspecific retinal activity. On the other hand, a striking resemblance of components of natural images to simple cell receptive field has been reported by several authors. We demonstrate that it is possible to differentiate between the spontaneous pattern formation scenario and postnatal stimulus-driven self-organization of receptive fields. Analyzing Gabor-fits of computationally generated receptive fields of simple model neurons we conclude that natural-scene stimuli are essential for the explanation of certain features of experimentally measured receptive fields.
Original languageEnglish
Pages (from-to)279-284
Number of pages6
JournalNeurocomputing
Volume38–40
DOIs
Publication statusPublished - Jun 2001

Keywords / Materials (for Non-textual outputs)

  • Visual cortex
  • Receptive field
  • Retinal filter
  • Natural images

Fingerprint

Dive into the research topics of 'Signatures of natural image statistics in cortical simple cell receptive fields: Computational Neuroscience: Trends in Research 2001'. Together they form a unique fingerprint.

Cite this