P300 event-related potentials were recorded in 10 subjects with neocortical lesions, and 10 control subjects, using a basic oddball paradigm. In separate tests, subjects discriminated rare and frequent tones that differed with respect to frequency, or discriminated the tones in a noise background. Subjects were required to count the number of rare stimuli that occurred during test runs. Recordings were obtained from vertex (Cz) sites referenced to linked earlobe electrodes. Control subjects exhibited P300s on all test runs. Decreasing stimulus differences relative to frequency, or adding background noise, produced significant increases in P300 latency plus decreases in amplitude. In contrast, two of the 10 lesion subjects failed to demosntrate P300 responses, although both subjects accurately counted the rare stimuli. With the remaining lesion subjects, absent or significantly delayed P300 responses occurred in 53 percent of the test runs, while accurate counts of the rare stimuli were maintained in all test runs. Absent or delayed P300s in the lesion group was not correlated with location or extent of the lesions. These results indicate that, while the P300 is susceptible to neocortical damage, it reflects cognitive processing other than simply discriminating differences between rare and frequent stimuli.
(C) 1990, The American Journal of Otology, Inc.