Current technology can simultaneously monitor the firing of approximately 150 neurons. The corresponding cross correlation matrix has well over 20,0000 independent entries, generally distributed in a seemingly random fashion because of the complex interconnections within typical neural tissue. The trend in experimental neurophysiology is clearly towards even larger sets of data. We propose to develop software for multineuron extracellular electrophysiology recording that will provide experimentalists with a near real-time display of the correlation in firing within such large groups of neurons. The device will be placed in the typical laboratory equipment chain after signal amplification, conditioning, and spike sorting. It will calculate (in near real-time) the correlation between all pairs of input spike trains, and then present a frequently updated display of the cross correlation matrix in such a way that highly correlated sub-sets of neurons within the recorded group will be displayed prominently. The device will be used to study changes in composition and correlation levels in these subsets as experimental conditions are changed, as an experimental animal is trained, and to help guide the placement of the recording electrode.
NOT AVAILABLE
