OPTIMAL NEURAL SPIKE CLASSIFICATION.

Summary form only given. Extracellular microelectrode recording of neural signals is a useful way for studying small populations of nerve cells. The recorded signal usually contains spikes from several neurons adjacent to the electrode. The authors consider the problem of assigning the spikes, on the basis of their shapes, to the different neurons. A method is proposed which results in minimum probability of error, even in case of the occurrence of overlapping spikes. Various degrees of a priori knowledge could be taken into account by this method.