An electrophysiological test of the arousal-hemispatial neglect hypothesis.

摘要

Sudden ablation of focal brain areas in stroke patients allows in vivo correlations of behavioural measures with lesion parameters in previously normal adults. Focal damage from stroke can cause alterations in arousal and attention and the neural mechanisms underlying these processes can be examined in this population. One of the most striking stroke-produced impairments is hemispatial neglect, which is associated predominantly with right hemisphere damage (RHD). Neglect is an impaired ability to report, respond, or orient to stimuli presented contralesionally. Some 50% of all RH stroke patients show neglect, with one quarter severely affected. Although neglect is less severe and long-lasting following left hemisphere damage (LHD), about one third do show associated disability. Electrophysiological measures of attention, including power spectra electroencephalography (EEG) and event-related potentials (ERPs), provide independent assessments of attentional processes. To determine whether reduced cortical arousal is associated with neglect and RHD, we examined the correlation between stroke-related hypoarousal, neglect, and changes in auditory ERPs. Subjects consisted of a consecutive series of patients admitted to Sunnybrook Health Science Centre with computed tomography-confirmed unilateral stroke (N = 50). The presence of neglect was systematically verified with paper and pencil tests (Sunnybrook Neglect Battery), while arousal was independently measured by RT scores and EEG desynchronization. Scalp EEG was recorded and stored using the QSI 9000 during a standard auditory oddball paradigm (high tone targets/low tone non-targets) for later off-line ERP analysis. We hypothesized that neglect results from a combination of impaired visuospatial attention and hypoarousal, with only neglect subjects exhibiting both symptoms. The experimental design facilitated examination of the attentional deficits underlying acute focal brain damage through a comparison of arousal differences in RHD patients with and without neglect. Controls (LHD and RHD subjects without neglect) allowed within- and across-hemisphere comparisons of arousal. We predicted enhanced slow frequency activity, a reduction in fast frequency activity, and excess diffuse ipsilateral EEG slowing would be observed in stroke patients suffering from neglect. ERPs were expected to demonstrate the delayed P3 latency, decreased N1, P3 and increased P2 amplitude associated with depressed arousal. We found EEG power spectra evidence of hypoarousal in RHD patients, and in patients with neglect; however, none of the anticipated ERP characteristics reflecting impaired attentional ability in these patients were observed. This result raises the issue of whether latency and amplitude indices are better represented in clinical populations by single trial analysis or averaging.