Abstract
A new algorithm for averaging of visual evoked potentials (VEPs) is introduced and compared with the conventional method. The root mean square (RMS) values of the alpha and theta band EEG activities at the vertex 1-s preceding stimulus presentation were computed: stimuli were blocked during phases of high prestimulus activity. Stimulation was applied selectively during periods of low activity. Light stimuli of 1-s duration were given to twelve healthy, voluntary subjects. The main results were: (1) The averaged VEPs at the vertex showed an increase in the N1-P2 amplitude of about 35% in comparison to experiments with conventional stimulation. This increase was in the range of 25–30% when an ISI correction was taken into account. Frontal, temporal and parietal locations showed also increases of the N1-P2 amplitude in evoked potentials. (2) VEPs were also affected qualitatively depending on the frequency content of the prestimulus activity.