In in-vivo animal models, neuroscience experiments in electrophysiology are commonly performed with the use of extracellular electrodes implanted in the cell layer of specific brain regions of interest. These electrodes record voltage traces and capture the occurrence of putative action potentials when there is a “spike” in the voltage trace. Each spike’s source (a putative, single neuron) is unknown to the experimenter at the time of the recording, so “spike sorting” is an analysis performed to group spikes together based on some set of shared characteristics. Each grouping (or cluster) of spikes is assumed to come from the same neuron. This tutorial begins with a quick introduction to common electrophysiology recording setups. It then presents an overview of the spike sorting workflow, from filtering the raw voltage traces to assessing the quality of the final spike clusters, and ends with performing spike sorting in MATLAB.
Taught by: Jai Bhagat and Caroline Moore-Kochlacs, MIT
- An introduction to spike sorting (1:54:30)