Diminished neuronal response to repeated sensory input is a sensory-gating phenomenon that has been found to be deficient in schizophrenic patients. For example, schizophrenic patients fail to decrease the amplitude of the P50 wave of the auditory evoked potential to the second of paired click stimuli. In some studies, however, normal subjects have also failed to decrease their P50 responses. The aim of this study was to determine if accommodation to the recording situation over time would affect the gating of the P50 response. The gating of the P50 wave is measured as the ratio of the amplitude of the second response to the amplitude of the first. Three successive auditory evoked potentials were compiled, each from trains of 32 pairs of stimuli. Twelve normal subjects and 12 schizophrenic patients were studied. Unconjugated catecholamine metabolites were measured from venous samples drawn before and after the electrophysiological recording. Between the first and third trials, the normal subjects significantly increased their gating of P50. This increase in gating of P50 was related to decreased levels of the noradrenergic metabolite 3-methoxy-4-hydroxyphenylglycol. No similar phenomenon was observed in the schizophrenic patients, a number of whom had a further decrease in P50 gating over the three trials. Transient failure to observe gating of P50 in normal subjects may be related to increased state-dependent noradrenergic activity, which is known to disrupt sensory gating. This mechanism does not seem to account for the more persistent failure of sensory gating in schizophrenia.