Single P100 Visual evoked potential analyses in man

Standards evoked potential averaging leads to a loss of information inherent in the moment to moment variability of the amplitude and latency of single evoked potentials. In order to extract this information, attempts were made to develop procedures which would allow recording of single visual evoked potentials in at least a selected group of subjects. In 11 out of 25 non-selected subjects, 43% to 86% of all stimuli (reversal checkerboard pattern elicited single visual evoked potentials (VEP) which were recognized as such by a group of independent observers. The mean amplitude and latency obtained by directly measuring the peak to peak amplitudes and peak latencies of these single VEPs (arithmetic averaging) were compared to those obtained following conventional time-locked averaging of the same data. During long-term continuous stimulation, the arithmetic averaged VEPs increased in amplitude to a steady state while the time-locked averages of the same sets of responses decreased in amplitude. This reduction was found to be closely related to the latency jitter. It may provide a better understanding of the phenomenon of habituation. This finding was confirmed in model single VEPs obtained by summing on-going pre-stimulus EEG activity with a time-locked average VEP stationary wave form. The variability of the true single VEPs was found to be less than the variability of the model single VEPs. The latency and amplitude parameters of the true single VEPs were strongly correlated with each other while those of the model single VEPs were not. These findings show that single VEPs have an inherent variability which may reflect brain processing.