COMPUTER-SIMULATION OF POST-SPIKE FACILITATION IN SPIKE-TRIGGERED AVERAGES OF RECTIFIED EMG

When the spikes of a motor cortical cell are used to compile a spike-t riggered average (STA) of rectified electromyographic (EMG) activity, a post-spike facilitation (PSF) is sometimes seen. This is generally t hought to be. indicative of direct corticomotoneuronal (CM) connection s. However, it has been claimed that a PSF could be caused by synchron ization between CM and non-CM cells. This study investigates the gener ation of PSF using a computer model. A population of cortical cells wa s simulated, some of which made CM connections to a pool of 103 motone urons. Motoneurons were simulated using a biophysically realistic mode l. A subpopulation of the cortical cells was synchronized together. Af ter a motoneuron discharge, a motor unit action potential was generate d; these were summed to produce an EMG output. Realistic values were u sed for the corticospinal and peripheral nerve conduction velocity dis tribution, for slowing of impulse conduction in CM terminal axons, and for the amount of cortical synchrony. STA of the rectified EMG from a ll cortical neurons showed PSF; however, these were qualitatively diff erent for CM versus non-CM cells. Using an epoch analysis to determine reliability in a quantitative manner, it was shown that the onset lat ency of PSF did not distinguish the two classes of cells after 10,000 spikes because of high noise in the averages. The time of the PSF peak and the peak width at half-maximum (PWHM) could separate CM from sync hrony effects. However, only PWHM was robust against changes in motor unit action-potential shape and duration and against changes in the wi dth of cortical synchrony. The amplitude of PSF from a CM cell could b e doubled by the presence of synchrony. It is proposed that, if a PSF has PWHM <7 ms, this reliably indicates that the trigger is a CM cell projecting to the muscle whose EMG is averaged. In an analysis of expe rimental data where macaque motor cortical cells facilitated hand and forearm muscle EMG, 74% of PSFs fulfilled this criterion. The PWHM cri terion could be applied to other STA studies in which it is important to exclude the effects of synchrony.