Bisensory stimulation: Inferring decision-related processes from the P300 component.

Three experiments were conducted to evaluate the P300 component of the human evoked response as an index of bisensory information processing. On different blocks of trials, subjects were presented with auditory stimuli alone, visual stimuli alone, or with audiovisual compounds. In each series there were two possible stimuli, one of which was presented less frequently than the other; the subjects' task was to count the infrequent stimuli. In the first two experiments the information in the two modalities was redundant, whereas in the third the modalities provided nonredundant information. With redundant information, the P300 latency indicated bisensory facilitation when the unimodal P300 latencies were similar; when the unimodal latencies were dissimilar, the bisensory P300 occurred at the latency of the earlier unimodal P300. Reaction times paralleled P300 latency. When the information in the two modalities was nonredundant, both P300 amplitude and reaction-time data indicated interference between the two modalities, regardless of which modality was task relevant. P300 latency and reaction time did not covary in this situation. These data suggest that P300 latency and amplitude do reflect bisensory interactions and that the P300 promises to be a valuable tool for assessing brain processes during complex decision making.