Phased array processing for spike discrimination

We present a novel approach for the detection, discrimination, and identification of superimposed neuronal action potentials from multineuronal, multichannel extracellular nerve recordings with low signal-to-noise ratios. The approach uses phased-array processing techniques to identify the spikes from different neurons on the basis of their unique propagation velocities. We evaluated this new approach using simulated electrophysiological data, under conditions that are known to limit the effectiveness of existing spike discrimination techniques. This approach enabled discrimination of simulated spikes from multiple simultaneously active neurons with a high degree of reliability and robustness within the expected range of experimental recording conditions, even in situations where there was a high degree of spike waveform superposition on the recording channels. Moreover, the technique enables the reliable detection and discrimination of spikes recorded with signal-to-noise ratios less than 1.