Local field potentials: Biophysical origin and analysis

Abstract

The previous chapter focused on the biophysics of generation of spikes, the extracellular signature of action potentials, representing the output of the computation done by individual neurons. In this chapter, we focus on the generation of local field potentials (LFPs), which in contrast appears to predominantly reflect the synaptic inputs to the neurons. The measurement of LFPs, which we here define as the low-frequency part of the electrical potential recorded by an extracellular electrode inserted into the brain, is among the oldest techniques for studying neural activity. Already in 1875, Richard Caton could report that he had measured electrical signals from the cortical surface of living animals (Caton 1875), 50 years before the first EEG was recorded (Berger 1929). As the LFP stems from activity in populations of neurons around the electrode, the signal is not straightforward to interpret. Early investigators thus focused on the high-frequency part of the signal, the multi-unit activity (MUA), which contains information about the firing of action potentials of a handful of the surrounding neurons (Adrian and Moruzzi 1939). The interest in the LFP signal revived in the 1950s with the introduction of current-source density (CSD) analysis of multiple LFP recordings across well-organized layered neural structures such as cortex, cerebellum, and hippocampus (Pitts 1952). As the CSD is a more local measure of neural activity than the LFP (Nicholson and Freeman 1975, Pettersen et al. 2006), it is easier to interpret in terms of the activity in the underlying neural circuits. A comprehensive review of the findings from such CSD analysis in the ensuing three decades can be found in Mitzdorf (1985).