USE OF SPIKE TRIGGERED AVERAGING OF MUSCLE-ACTIVITY TO QUANTIFY INPUTS TO MOTONEURON POOLS

1. The goal of this study was to determine the extent to which postspike facilitation (PSpF) of electromyograms (EMGs) could be used to estimate the inputs to separate motoneuron pools, under conditions where there was wide variability in the parameters of muscle activity. These parameters included cancellation of motor unit action potentials (MUAPs), variations in EMG noise, and changes in MUAP amplitude and duration. A systematic series of computer simulations with increasing complexity were used to achieve this goal. The initial simulations (model I) included a premotoneuronal (PreM) cell connected to a single postsynaptic motoneuron (Mn), which in turn projected to a muscle. The next simulations (model II) included other target motoneurons with their efferents each projecting to separate muscles. The last simulations(model III) included more than one postsynaptic motoneuron per Mn-pool, as is the case in mammalian neuromuscular systems. 2. A sample simulation (model I) was performed to determine if the PreM-evoked effects were within physiologically observed values. A cross-correlogram (XC) calculated from a PreM cell and its target Mn, receiving a PreM-evoked excitatory postsynaptic potential (EPSP) of 0.5 mV, produced a XC peak area of 0.04 Mn-spikes/PreM-trigger. The PSpF of EMG activity evoked by this PreM cell had a mean percent increase of 4.6% (MPI = mean bin amplitude of PSpF above baseline/mean baseline level X 100). These XC and PSpF values were within the range of values previously obtained from animal experiments. 3. The magnitude of MUAP cancellation in the EMG was tested by calculating two spike-triggered averages (SpTAs) of EMGs from Mn-triggers (not PreM-triggers as in the other SpTAs): one using typical bipolar MUAPs and another using their rectified counterpart of only positive polarity to eliminate the possibility of MUAP cancellation. The PSpF calculated from bipolar spikes was 24.8% smaller than the one calculated using unipolar spikes. This cancellation could be greater or smaller depending on the state of parameters, such as the shape and number of MUAPs, that determine the probability of overlap between MUAP components of opposite polarity. All subsequent computer simulations used typical bipolar MUAPs. 4. A series of increasing motoneuron EPSP amplitudes were used to determine the relationship between PreM-Mn connection strength and PSpF area. A nearly perfect linear relationship between EPSP amplitude and PSpF area was obtained for SpTAs of rectified EMGs (r = 0.99). An equally linear relationship was obtained when averaging nonrectified EMGs (r = 0.99), but the smaller EPSPs or weaker synaptic connections were not detected. Thus PSpF area reliably reflected PreM-Mn connection strength when motoneuron EPSP amplitude was the only variable. However, there are usually other variables influencing the neuromuscular output. 5. A PreM cell with fixed excitatory postsynaptic effect on its target motor unit was simulated under conditions where the muscle showed variations in background EMG noise, MUAP amplitude, or MUAP duration. Increasing background EMG noise had little effect on PSpF area, but the pretrigger baseline of the SpTA was markedly increased. In contrast, MUAP height and duration had important effects on PSpF area. Besides contributing to a larger PSpF, the taller and wider spikes also contributed to a small increase in the mean baseline level of the SpTA. Thus variations in MUAP height and duration could influence PSpF area in the absence of changes in PreM-Mn connection strength. all contribution to the PSpF. Consequently, PSpF of EMGs would be a good estimate of inputs to motoneuron pools.