Auditory and visual P300s in temporal lobectomy patients: evidence for modality-dependent generators

The P300 component of the event-related brain potential (ERP) was studied in epileptic patients with unilateral resections of medial temporal lobe areas of the brain. The patients and controls were tested in an oddball paradigm in two conditions: counting and reaction time. Both auditory and visual stimuli were used to elicit ERP activity in different blocks. Despite the reported presence of locally-generated auditory and visual P300-like potentials in these areas, no evidence was found of any surgically-related hemispheric asymmetries in the scalp distribution of the P300 or Slow Wave for stimuli in either modality. Moreover, compared to normal controls, there were no significant reductions in overall P300 amplitude in the patients. The patients did show a double dissociation in their frontal ERP activity: the left temporal lobectomy patients showed apparent decreased frontal auditory P300 amplitudes but normal visual P300 amplitudes, whereas the right temporal lobectomy patients showed the opposite pattern. These results appeared to be due to the presence of a long-duration slow wave rather than to alterations in P300 amplitude. These data do not support the presence of a significant contribution by a hippocampal/amygdala generator to the activity of the scalp-recorded P300 in the oddball paradigm. Topographic comparisons on normalized amplitudes revealed significantly different scalp distributions as a function of stimulus modality, event probability, and task for both the P300 and Slow Wave components. These data indicate that the amplitude variations associated with each experimental variable are due to the activity of a separate underlying neural source. The sources of task and probability effects on P300 and Slow Wave amplitude each appeared to be modality-independent generators. The nature of the third, modality-related generator is less clear. These results uphold the tenets of the model of P300 amplitude proposed by Johnson (1986) and argue against the idea that the P300 is a unitary phenomenon.