Sensory gating in rats can be measured with a double click paradigm. The diminished response towards the second click is a physiological manifestation of reduced sensory input. This physiological process seems to be disturbed in human psychoses. It is thought that gating, as measured with this paradigm, is a preattentive, involuntary phenomenon which is not modulated by attention. If this is indeed the case, than it is hypothesized that gating should not be modulated by non-REM sleep. In the present experiment pairs of clicks (500 ms interval) were presented during wakefulness, non-REM as well as REM sleep and cortical auditory evoked potentials (AEP's) were recorded in chronically implanted rats. Rather similar AEP's were found after the first and second stimulus. However, the amplitudes of the various components of the second AEP were smaller than those of the first AEP, suggesting a gated response. This was the case during all three levels of vigilance. The amplitudes of both AEP's showed the more often reported changes in amplitude during sleep and REM sleep. Clear differences were seen in gating: compared to wakefulness a decrease in gating was found during REM sleep while gating was unchanged during non-REM sleep. The latter outcome seems to confirm that gating in rats is indeed a preattentive process. Finally, results were discussed in terms of neuronal properties of thalamic relay cells and it is suggested that firing properties of thalamic relay cells are not involved in this type of sensory gating.