Theoretical analysis of motoneuron models predicts that the cross-correlogram between motoneuron discharges and stimuli should reflect the derivative of the post-synaptic potential (PSP) in the absence of synaptic noise. The cross-correlogram peaks become wider if the amplitude of the noise is significant. To deal with this case, the linear combination of the PSP and its first derivative has been proposed as an approximation. The synaptic noise was regarded as a Gaussian stochastic process (its diffusion approximation), and the estimation of the cross-correlation function for this case is presented. The results were found to be in agreement with reported experimental data. A comparison was made with the linear combination approximation and with the reported computer simulation.