FUNCTIONAL MR MAPPING OF HIGHER COGNITIVE BRAIN FUNCTIONS

Functional magnetic resonance imaging (fMRI) offers a powerful experim ental tool for mapping activated cortical regions in man. Thereby, the paramagnetic deoxyhemoglobin in the red blood cells acts as an endoge nous susceptibility contrast agent, which allows the noninvasive detec tion of stimulus-induced transient changes in regional cerebral blood flow and volume. Fifteen normal subjects were examined on a convention al 1.5-T MR system to visualize cortical activation during the perform ance of high-level cognitive tasks. A computer-controlled videoproject or was employed to present psychometrically optimized activation parad igms. Reaction times and error rates of the volunteers were acquired o nline during stimulus presentation. The time course of cortical activa tion was measured in a series of strongly T2-weighted gradient-echo i mages from three or four adjacent slices. For anatomical correlation, picture elements showing a stimulus-related significant signal increas e were color-coded and superimposed on T1-weighted spin-echo images. A nalysis of the fMRI data revealed a subtle (range 2-5%), but statistic ally significant (P < 0.05) increase in signal intensity during the pe riods of induced cortical activation. Judgment of semantic relatedness of word pairs, for example, activated selectively cortical areas in l eft frontal and left temporal brain regions. The strength of cortex ac tivation in the semantic task decreased significantly in the course of stimulus presentation and was paralleled by a decrease in the corresp onding reaction times. With its move into the area of cognitive neuros cience, fMRI calls both for the careful design of activation schemes a nd for the acquisition of behavioral data. For example, brain regions involved in language processing could only be identified clearly when psychometrically matched activation paradigms were employed. The react ion time data correlated well with selective learning and thus helped to facilitate interpretation of the fMRI data sets.