Extracellular single and multiple units recording with microelectrodes

Introduction: the origins of extracallular spike potentials

Action potentials across the membranes of neurons generate electric current flow in the tissue around the neurons. These currents in turn generate voltages in the tissue, and these voltages can be detected as extracellular spikes. Extracellular spikes from mammalian neurons in the central nervous system have duration of between 0.2 and 20 ms, depending on the type of neuron and the bandwidth of the spike recording system. Amplitudes vary from the noise level of the recording electrode which may be as little as 2v RMS up to several mill volts: The great advantage of extracellular recording for the neurophysiologist is that you can record the activity of neurons without having to impale and thereby damage them. Impalement may be inadvisable because it damages the cell, it may be extremely difficult because the cell is small, or it may be unstable because the tissue is moving due to mechanical pulsations from blood pressure or respiratory fluctuations. For these reasons most neuronal recording in anaesthetized animals, as opposed to brain slices or cell cultures, is done with extracellular recording.

The great disadvantage of extracellular recording is that it is all too easy to record from two or more units at the same time; this is technically multi-unit recording. This, of itself, is not a problem if you can discriminate between the spikes from the two or more cells. This however, may be difficult. It is rather like having a microphone between two people talking together in a noisy environment. Which voice is active any one moment may be determined simply by the loudness of the voice; however, if the voices are of a similar loudness other characteristics such as the pitch of the voice and perhaps the speed of speaking have to be taken into account.