Regional correlations between cortical thickness and surface area asymmetries: A surface-based morphometry study of 250 adults

We report on the patterns of asymmetries of various MRI-derived cortical phenotypes, namely cortical thickness (CT), cortical surface area (CSA), cortical volume (CV) and sulcal depth (SULC), as well as on their cross-relationships. A surface-based approach was designed to obtain cortical asymmetry maps unbiased for inter-hemispheric structural positional misalignment. Accurate vertex-wise asymmetries of CT, CSA, CV, and SULC were so obtained in 250 individuals including 120 left-handers that had been selected from a larger population as having a typical leftward language lateralization. We found no significant effect of handedness on CT, CSA or CV asymmetries, although a trend for a significant difference in the SULC asymmetry of the Rolandic genu was present (leftward in right-handers, rightward in left-handers). At the hemispheric level, we found rightward CT and CSA asymmetries that were not correlated. At the regional level, asymmetry patterns of CT and CSA were marked by a spatial overlap of both concordant and opposite CT and CSA asymmetries. Half of these regions of overlap presented a significant association (positive or negative) between CT and CSA asymmetries. Strikingly, the 3 regions showing positive correlations between CT and CSA asymmetries were those known to exhibit robust asymmetries across methodologies, such as the leftward asymmetrical planum temporale and the rightward asymmetrical superior temporal sulcus and cingulate cortex. This study demonstrates that regional correlations between CT and CSA asymmetries are a characteristic of brain structural organization that could be of significance in the choice of structural markers in studies dealing with the genetic basis of brain lateralization.