Table of Contents:
Deficits in working memory (WM) performance and executive functions have long been known as endophenotypes of psychiatric diseases. That leads to the question how brain structure and function are associated with WM in healthy subjects. In this study morphometric measurements and diffusion-tensor-imaging (DTI) were utilized to identify neuroanatomical correlates of WM networks in healthy subject brains.
681 healthy subjects were recruited and underwent neuropsychological testing (NPT). We performed MRI-scans and used the letter-number-test (LNT) to measure working memory function. To control for WM specificity, we used digit symbol coding (DSC). Structural analysis of the T1 MRI images worked with voxel-based morphometry using the Toolbox CAT12 in SPM. We measured regional volume, cortical thickness and cortical gyrification parameters in conjunction with NPT score. Further we used tract-based spatial statistics in FSL to analyze the DTI data.
Structural analysis revealed significant negative correlations between high DSC scores and cortical thickness of frontal and parietal cortex regions, as well as in precuneal and cuneal areas of both hemispheres. In the left hemisphere, reductions in cortical gyrification of the inferior parietal lobe were positive correlated to DSC. There was no correlation between local brain volume and the WM-performance. In DTI studies, we observed a significant positive correlation between WM-performance and fractional anisotropy (FA) of the bi-hemispheric superior longitudinal fascicles and their temporal parts. There was also a positive correlation between performance speed and a significant increased FA in both hemispheres in the anterior thalamic radiation, inferior frontal occipital fasciculus, inferior longitudinal fasciculus, superior longitudinal fasciculus and its temporal part and the uncinate fasciculus. In the right hemisphere we found an association of increasing FA to the limbic system, including the cingulum, cingulate gyrus, and forceps minor.
These findings demonstrate the overlap of NPT performance and brain variation, which indicates an involvement of similar neuronal structures in both cognitive processes. We confirmed the existence of cortical and subcortical networks in frontal parietal, temporal and striatal areas by using cognitive functions.