Spike sorting

Figure 1. Extracellular recordings and spike sorting. (A) The extracellular activity of neurons is recorded with microwires inserted in the brain. The spikes of neurons close-by the electrode tip (in red and blue) can be separated via spike sorting; neurons further away (in dark grey) can be detected but not sorted, thus generating the multiunit activity; more distant neurons (in light gray) contribute to the background noise. (B) Main step of spike-sorting algorithms (adapted from Quian Quiroga et al. (2004)). What is spike sorting? A gold standard in neuroscience is to record extracellularly the activity of single neurons with thin electrodes implanted in the brain (Figure 1A). Extracellular recordings pick up the spikes of neurons nearby the electrode tip and the job of the experimenter is to determine which spike corresponds to which neuron. This identification is done based on the shape of the spikes, given that, in principle, each neuron fires spikes of a particular shape, depending on the morphology of its dendritic tree and the distance and orientation relative to the recording site, among other factors. Spike sorting is the grouping of the detected spikes into clusters based on the similarity of their shapes. The resulting clusters of spikes correspond to the activity of different putative neurons.