Neural mass differences in gustation

Publisher Website
Google Scholar
Abstract
Correlational techniques have been developed in the chemical senses for the measurement of neural and psychophysical information; in addition to correlations per se , these have included multidimensional scaling and cluster analysis, and also may include factor analysis. Although very powerful, these methods are insensitive to numbers of neurons involved and amounts of evoked activity: particularly at low levels of activity, low reliability may be interpreted as information by these methods. A variant of correlations is discussed as a measure of neural information, which includes numbers of neurons and amounts of evoked activity in its computation, and is not unduly influenced by poor reliability at low response levels. As a neural metric, this ‘neural mass differences’ method is equally applicable to labeled-line or across-fiber-pattem models of neural coding. The data analyzed were the responses of 40 single taste neurons in the parabrachial nuclei of the hamster to 32 taste stimuli.

This publication has 0 references indexed in Scilit: