COGNITIVE FUNCTION AND EVENT-RELATED POTENTIALS IN PATIENTS WITH CEREBROVASCULAR DISEASES

Abstract

In order to study differences between event-related potentials generated during counting and button-pressing tasks, the auditory event-related potentials from 103 patients (62 males, 41 females) aged 43-90 years with chronic cerebrovascular diseases were recorded. All patients were evaluated for severity of cognitive deterioration with Hasegawa's Dementia Score (HDS) examination, and 44 also underwent Wechsler Adult Intelligence Scale (WAIS) examination.The target stimulus was a pure tone with a frequency of 2, 000 Hz delivered binaurally at 75 dB in 20 % of trials, and the non-target stimulus was a pure tone with a frequency of 1, 000 Hz delivered at 75 dB in 80 % of trials. On the counting task, the patient was instructed to keep a mental record of the number of target tones, and on the button-pressing task, to press the button switch as quickly and precisely as possible whenever he recognized the target tone. Responses were recorded from Cz and Pz electrode placements on the scalp (International 10 : 20 system) referred to linked mastoids. Waveforms were averaged on-line by a signal processor (7 T17 or 7 T18), which also controlled the stimulus presentation and artifact rejection.The button-pressing task was easier to perform than the counting task, and the P300 waveforms were clearer in the former than in the latter. The P300 latency of the button-pressing task was shorter than that of the counting task.The P300 latency of both tasks was found to be negatively correlated with HDS, and the button-pressing task showed higher correlation than the counting task.On the button-pressing task, the P300 latency was found to be negatively correlated with Total IQ, Verbal IQ, Performance IQ and many subtest scores, while on the counting task it showed negative correlation with Performance IQ and only a few subtest scores. Most negative correlations between the P300 latency and subtest scores were greater on the button-pressing task than on the counting task, indicating that the former is a better measure of cognitive function than the latter.Of the subtests, Object Assembly, Digit Symbol, Block Design and Picture Completion scores showed highly negative correlations with the P300 latency. We therefore presume that the P300 component strongly reflects constructional ability in cognitive function.HDS and Performance IQ showed significant negative correlations with the P300 latency of both tasks. Analysis of covariance for their regression lines showed significant differences.In the counting task, the N100 latency was negatively correlated with the score of one subtest, namely, Vocabulary. However, the N200 and P200 latencies did not correlate significantly with WAIS IQs or subtest scores on either the counting task or the button-pressing task.