BEHAVIORAL AND SPECT ACTIVATION ANALYSIS OF HEMINEGLECT

We propose to analyze visual-spatial dysfunction in stroke patients with hemineglect by correlating clinical deficits with topographical activation patterns on Single Photon Emission Tomography (SPECT). A two phase study will use a specialized Principle Components Analysis developed by one of the sponsors to extract SPECT activation patterns related to specific tasks. The first phase of the study will extend our research on the component nature of hemineglect by examining the SPECT activation patterns of normal subjects engaging in line bisection and letter cancellation. These two commonly used tests for hemineglect have been shown by our laboratory to be dissociable in patients, depending on lesion location. The second phase of the study will focus on the perceptual aspects of hemineglect. In a three-part experiment we will examine a well-tested psychophysical model of visual-spatial perception that states that the left hemisphere is superior in categorical-based assessments (e.g. Is a dot above or below a line?) whereas the right hemisphere is superior in coordinate-based assessments (e.g. Is a line transected within 2 mm of its true midpoint?). Categorical versus coordinate-based tasks will be used for SPECT activation. In experiment 1 we will analyze the differences in activation pattern between the two task types in normal controls. In experiment 2 we will analyze activation patterns using the same pair of activation tasks in hemineglect patients, first in the acute phase of their illness and then 6 months later when the hemineglect has improved. We anticipate that SPECT patterns in the acute phase will reflect the patient's dysfunctional clinical state and that the activation pattern will change over time as clinical function returns. In experiment 3 we will probe visual-spatial dysfunction in hemineglect patients experimentally by systematically altering the test conditions. Consequent alterations in task performance will be correlated with changes on SPECT and compared to activation patterns in the patients who were not presented with behavioral interventions and to activation patterns the normal controls. We will attempt to show that experimental manipulations that improve clinical performance also systematically alter rCBF patterns, reflecting plasticity in the visual-spatial system. By combining experimental behavioral methods with newer imaging and data analysis techniques, we have the opportunity to add functional imaging data to models of brain function based to date on clinical observations of brain-injured patients and psychophysical data in normal subjects. Although we have chosen hemineglect as a model to study brain injury, the knowledge gained and techniques-advanced may alter the way we think about the brain's reaction to injury.