Determining the association between regionalisation of cortical morphology and cognition in 10,145 children

Abstract

During development individuals undergo protracted changes in cortical morphology, which coincide with changes in cognition. A number of studies have measured the association between brain structure and cognition in children, however these frequently rely on mass univariate statistics or ROI-based comparisons and do not always measure cortical morphology relative to global brain measures. After controlling for global brain measures, we can see a residual regionalisation pattern indicating the size or thickness of different regions relative to the total cortical surface area or mean thickness. Individual variability in this regionalisation may be important for predicting between subject variability in cognitive performance. Here we sought to determine whether the relative configuration of cortical architecture across the whole cortex was associated with cognition using a novel multivariate omnibus statistical test (MOSTest). This method is optimally powered to detect associations that are widely distributed across the cortex. We then used the Polyvertex score (PVS) to quantify the magnitude of the association between vertex-wise cortical morphology and cognitive performance. We have determined that the relative pattern of cortical architecture, after removing the variance associated with global brain measures, predicted unique variance associated with cognition across different modalities.

Individual Differences in Cognitive Performance Are Better Predicted by Global Rather Than Localized BOLD Activity Patterns Across the Cortex

“…Here, we found that including the covariance structure of the imaging phenotype during parameter estimation greatly improved the predictive performance of functional imaging phenotypes. Similar improvement of predictive accuracy of regional cortical morphology on cognitive outcomes was shown in children (Palmer et al 2020). Predictive performance is of great importance to the utility of biomarkers on clinical disorders or cognitive processes.…”

Towards Reproducible Brain-Wide Association Studies

“…Currently, brain-wide association studies are trapped in the initial discovery phase. Larger samples will inevitably allow multivariate models, both those currently in use and future improvements currently being developed [38][39][40][41] , to be employed by individual research groups. Smaller, investigator-initiated neuroimaging studies will continue to be just as important for human neuroscience as they were in the days when samples of N>10,000 for brainwidewide association studies (BWAS) were still an impossibility.…”

Distinct regionalization patterns of cortical morphology are associated with cognitive performance across different domains

“…In our previous study, we used a Bayesian polyvertex score (PVSB) to quantify the variance in each cognitive phenotype predicted by the vertex-wise imaging data for the regionalisation of CSA and CTH (see supplementary figure 3 for a summary of these findings 24 ). The PVSB represents a linear weighted sum of the vertex-wise associations, which are projected from a training set to an independent hold-out set within a cross validation framework.…”

“…We recently demonstrated that there were significant associations between the regionalisation of CSA and CTH (when controlling for global structural measures) and the fluid and crystallised composite scores from the NIH Toolbox within the ABCD baseline sample using the Multivariate Omnibus Statistical Test: the MOSTest 24 . This test aggregates vertex-wise associations across the cortex thereby improving our power for detecting widely distributed effects compared to the standard neuroimaging approach [24][25][26] . Given the graded nature of the biology underlying the patterning of the cortex this is a more appropriate statistical approach compared to standard neuroimaging methods.…”

“…However, previous studies have been underpowered, and did not employ multivariate tools, to assess the contribution of the whole configuration of cortical morphology to individual differences in cognition. We have recently shown that the regionalisation of cortical surface area (CSA) and apparent cortical thickness (CTH), relative to global structural measures, was significantly associated with cognitive performance in a large sample (N=10,145) of 9 and 10 year old children from the Adolescent Brain Cognitive Development (ABCD) study 24 . In our previous study, we adopted the Multivariate Omnibus Statistical Test (MOSTest) to demonstrate that the aggregated effect across all cortical vertices was significantly associated with individual differences in cognitive performance.…”