Extracellular action potential
The extracellular action potential (EAP, or extracellular spike) is the small voltage transient — tens of microvolts to a few millivolts — recorded near a firing neuron, arising because the transmembrane currents of the action potential drive current through the resistive extracellular medium. Under the quasi-static approximation of Maxwell's equations, the potential at a point is a weighted sum of the transmembrane current density over the whole cell, and the membrane current itself is proportional to the second spatial derivative of the intracellular potential along the neurite (the cable equation). Because sodium influx during the spike acts as a current sink while the capacitive and potassium return currents act as sources, the waveform seen by a point electrode is typically biphasic or triphasic rather than a copy of the intracellular spike.
The recorded amplitude falls steeply with distance — roughly as 1/r for an isolated monopole-like contribution but faster (1/r^2 or more) for the dipolar field that dominates at realistic distances — so a microelectrode effectively hears only neurons within about 50 to 140 microns. Spike shape depends on cell morphology, the position of the electrode relative to soma versus dendrites, and the low-pass filtering of tissue and electrode, which is why the same neuron looks different on different contacts and why spike sorting is hard.
On a Utah array or Neuropixels contact a well-isolated pyramidal cell produces a roughly 0.5 to 1 ms triphasic waveform of 50 to 200 microvolts; units below about 50 microvolts sink into the noise and become multiunit hash.
The extracellular spike is a spatially filtered echo of transmembrane current, not a copy of the intracellular action potential.
A common misconception is that the extracellular spike is a scaled-down intracellular action potential; it is closer to a spatially weighted temporal derivative of the transmembrane current, which is why its peak does not align in time with the intracellular peak.