Linear and Non-Linear Responses to Dynamic Broad-Band Spectra in Primary Auditory Cortex
Abstract
Natural sounds have a broad and dynamic spectral content. We have characterized primary auditory cortical neuron's responses to dynamic spectra by both linear and non-linear response measures. The linear measure -- the spectro-temporal response field (STRF) -- is expressed and measure in terms of sounds within sinusoidal dynamic spectra, i.e. moving ripples, into which any dynamic spectrum can be decomposed (via two-dimensional Fourier analysis). We can quickly measure the STRF using special combinations of moving ripples for stimuli, and hence have a linear predictor of how the cell should respond to arbitrary broad-band sounds. The non-linear response measures, which are often smaller in magnitude, describe the systematic components in the response which are not accounted for by the STRF. These effects can include rectification, saturation, and multiplicative interaction between frequency bands. They are measured using the non-linear terms a Wiener-like functional expansion of the stimulus-response relationship, but are rather more intuitively represented as dynamic, frequency dependent rate-level functions. We discuss the implications of these results, data analysis, and prediction procedures, and show examples of predicted and measured responses to speech-like ripple combinations in ferret (Mustela putorius). [Kein, D. et al. 1999, Assoc. Res. Otolaryngol. Abs: 146.]
Document Details
- Document Type
- Technical Report
- Publication Date
- Feb 14, 1999
- Accession Number
- ADA484246
Entities
People
- D. J. Klein
- Didier A. Depireux
- J. Z. Simon
- S. A. Shamma
Organizations
- University of Maryland