A pulsed neural network model of spectro-temporal receptivefields and population coding in auditory cortex

  • Authors:
  • Markus Volkmer

  • Affiliations:
  • Technical University Hamburg-Harburg, Department of Computer Engineering VI, D-21073 Hamburg, Germany (E-mail: markus.volkmer@tu-harburg.de)

  • Venue:
  • Natural Computing: an international journal
  • Year:
  • 2004

Quantified Score

Hi-index 0.00

Visualization

Abstract

The existence of spectro-temporal receptive fields and evidence forpopulation coding in auditory cortex motivate the development ofsuch models, that explicitly operate in the time-frequencydomain and are based on a pulsed neural network. In presenting such amodel, a formal connection of the fields of Time Frequency Analysisand Pulsed Neural Networks is established. Theresulting neural time-frequency signal representation is shown to berepresentable as a signal-dependent overcompletedictionary. It is derived from neural population coding. Signaldecomposition and filtering effects are presented, indicating obvioustechnical applications of the proposed model.