The aim of the study was to investigate whether the perception of intranasal chemosensory stimuli changes in relation to the respiratory cycle. We investigated 40 healthy subjects with normal olfactory function who participated in four sessions. The first session was used to adapt subjects to the experimental conditions, and, specifically, to train a certain breathing technique (velopharyngeal closure) which prevents intranasal respiratory air-flow. In each of the following three sessions one of three stimulants was tested, namely phenyl ethyl alcohol (PEA), hydrogen sulfide (H(2)S), or the trigeminal stimulant carbon dioxide (CO(2)). The sequence of testing the three stimulants was randomized across all participants. Sessions were separated by at least 1 day. Chemosensory event-related potentials (ERP) were recorded in response to 80 stimuli each. Following each stimulus subjects rated its intensity using a computerized visual analogue scale. Respiration was recorded using a probe in front of the subjects' mouth. While presentation of chemosensory stimuli was performed independent of the respiratory cycle, responses were averaged off-line according to the subjects' respiratory phase when the stimuli had been presented. Intensity of olfactory or trigeminal stimuli did not differ significantly in relation to the respiratory cycle. Olfactory ERP to phenylethyl alcohol were larger when stimuli were presented during inspiration. Similarly, responses to H(2)S tended to be larger when stimuli were presented during inspiratory phases. In addition, responses to CO(2) were larger when stimuli were presented during inspiration. Differences in relation to the respiratory cycle were found specifically for early ERP components. It is important to note that the changes of chemosensory information processing were found in the absence of changes of intranasal airflow. These data indicate on an electrophysiological level that there is priming of both olfactory and trigeminally mediated sensations in relation to the respiratory cycle.