Extracellular microelectrode sampling bias.

Abstract The antidromic response latencies of 640 pyramidal tract neurons, isolated in pericruciate cortex of the cat, were measured after stimulation of the medullary pyramids. The expected time distribution of antidromic response latencies was then calculated from published values for axon diameters in silver-stained preparations; compensation for shrinkage was not attempted. With conduction velocity assumed to be a linear function of fiber diameter, the optimum fit between the observed and expected antidromic latency distribution was calculated. The constant of proportionality had a value of 4.72, and raising each diameter value to the 2.52 power was found necessary for an optimum fit. Extracellular microelectrode sampling is proposed to be akin to “shooting fish in a barrel,” in the sense that the probability of a successful isolation is largely a function of the cross-sectional area of the axon. Other factors, such as microelectrode resistance and investigator experience and vigilance, play a lesser role. A correction vector, based on the exponent of 2.52 and on histological measurements, was derived for postcruciate tissue, and applied to a sample of 1109 orthodromically activated neurons. This application showed the effect of electrode sampling bias to be significant.