Evoked potential studies provide an objective measure of the neural pathways involved with perception. The effects of cognitive factors, such as anticipation or awareness, on evoked potentials are not known. The aim was to compare the evoked potential response to oesophageal stimulation with the cortical activity associated with anticipation of the same stimulus. In 12 healthy men (23.5 +/- 4 years), oesophageal electrical stimulation (15 mA, 0.2 Hz, 0.2 msec) was applied, and the evoked potentials recorded using scalp electrodes. A computerized model of randomly skipped stimuli (4:1 ratio) was used to separately record the evoked potentials associated with stimulation and those associated with an anticipated stimulus. The electrical stimulus represented the nontarget stimulus and the skipped impulse the target (anticipatory) stimulus. This anticipatory evoked potential was also compared to auditory P300 evoked potentials. Reproducible evoked potentials and auditory P300 responses were elicited in all subjects. Anticipatory evoked potentials (peak latency 282.1 +/- 7.9 msec, amplitude 8.2 +/- 0.7 microV, P < 0.05 vs auditory P300 evoked potential) were obtained with the skipped stimulus. This anticipatory evoked potential was located frontocentrally, while the auditory P300 potential was located in the centro-parietal cortex. The anticipatory evoked potential associated with expectation of an oesophageal stimulus, although of similar latency to that of the auditory P300 evoked response, originates from a different cortical location. The recording of cognitive evoked potentials to an expected oesophageal stimulus depends on attention to, and awareness of, the actual stimulus. Anticipatory evoked potentials to GI stimuli may provide an objective electrophysiological tool for the assessment of the cognitive factors associated with visceral perception.