Brain Mapping of Behavioral Domains Using Multi-Scale Networks and Canonical Correlation Analysis

Abstract

Simultaneous mapping of multiple behavioral domains into brain networks remains a major challenge. Here, we shed some light on this problem by employing a combination of machine learning, structural and functional brain networks at different spatial resolutions (also known as scales), together with performance scores across multiple neurobehavioral domains, including sensation, motor skills, and cognition. Provided by the Human Connectome Project, we make use of three cohorts: 640 participants for model training, 160 subjects for validation, and 200 subjects for model performance testing thus enhancing prediction generalization. Our modeling consists of two main stages, namely dimensionality reduction in brain network features at multiple scales, followed by canonical correlation analysis, which determines an optimal linear combination of connectivity features to predict multiple behavioral performance scores. To assess the differences in the predictive power of each modality, we separately applied three different strategies: structural unimodal, functional unimodal, and multimodal, that is, structural in combination with functional features of the brain network. Our results show that the multimodal association outperforms any of the unimodal analyses. Then, to answer which human brain structures were most involved in predicting multiple behavioral scores, we simulated different synthetic scenarios in which in each case we completely deleted a brain structure or a complete resting state network, and recalculated performance in its absence. In deletions, we found critical structures to affect performance when predicting single behavioral domains, but this occurred in a lesser manner for prediction of multi-domain behavior. Overall, our results confirm that although there are synergistic contributions between brain structure and function that enhance behavioral prediction, brain networks may also be mutually redundant in predicting multidomain behavior, such that even after deletion of a structure, the connectivity of the others can compensate for its lack in predicting behavior.

Prenatal and childhood air pollution exposure, cellular immune biomarkers, and brain connectivity in early adolescents

“…Understanding how neural correlates via neuroimaging map onto human behavior is a complex field of study ( Fernandez-Iriondo et al, 2022 ), as differences in brain structure and function may exist with or without overt changes in behavior for various reasons (i.e., tempo and magnitude of brain plasticity changes detected by the spatial limits of MRI, a plethora of unmeasured neural compensatory mechanisms as the entire brain is involved in complex human behaviors, limited brain phenotypes usually studied in specified regions of interest, etc.). However, given the relative novelty of the neuroimaging markers used here and the exploratory nature of the current research, we also explored if the PLSC-identified brain features related to neurocognitive performance on the NIH Toolbox.…”

Exploring the Frontiers of Neuroimaging: A Review of Recent Advances in Understanding Brain Functioning and Disorders

“…This has allowed researchers to investigate the brain's functional networks and their relationship to cognitive and behavioral processes. Multimodal neuroimaging: Multimodal neuroimaging involves combining two or more neuroimaging techniques, such as fMRI and EEG or fMRI and PET, to obtain complementary information about brain structure and function [23,24]. This has allowed for more comprehensive investigations of the brain's neural activity.…”

The structure of anticorrelated networks in the human brain

“…The N = 1,000 functional images were pre-processed in a previous work (Fernandez-Iriondo et al, 2022) by correcting for gradient distortions and normalized to the standard MNI152 template of voxel size equal to 2 × 2 × 2 mm 3 using the HCP pipelines fMRIVolume and fMRISurface. After image normalization, we eliminated nuisances with a procedure that combines a volume censoring strategy and motion-related time course regression, along with physiological signal regression.…”