Precision mapping of individualized executive networks in youth

点击翻译按钮获取中文摘要

ABSTRACT Executive function (EF) improves dramatically during childhood and adolescence, and failures of EF are associated with both a broad range of negative outcomes and diverse mental illnesses. The brain circuits responsible for EF are spatially distributed, and include the fronto-parietal, cingulo-opercular, and salience systems. These networks have typically been studied using standardized network atlases, which assume a straightforward mapping between structural and functional neuroanatomy across individuals. However, multiple independent efforts in adults using precision functional mapping techniques have recently demonstrated that there is marked inter-individual variation in functional topography, which is defined as the spatial distribution of functional networks on the cortex. The over-arching hypothesis of this proposal is that individual variation in functional topography is a critical determinant of EF in youth. Our collaborative team recently published the first report of individualized functional networks in children using a cross-sectional sample (Cui et al., Neuron 2020). In this proposal, we will build upon this initial work by replicating and generalizing this finding using two large cross-sectional datasets with high-resolution imaging: the Healthy Brain Network (HBN; n=5,000) and Human Connectome Project: Development (HCP-D, n=1,300). Critically, we will also leverage the unprecedented resources of the Adolescent Brain and Cognitive Development Study (ABCD, n=11,572) to delineate within-subject change in personalized networks. In this proposal, we will first harmonize these massive data resources using advanced techniques originally developed for statistical genomics (Aim 1). Next, we will describe how personalized networks evolve with age (Aim 2) and predict EF (Aim 3). Finally, we will use machine learning tools to discover how the functional topography of personalized executive networks predict dimensions of psychopathology in a data-driven manner (Exploratory Aim 4). Throughout, we will adhere to best practices of open science to maximize reproducibility, and ensure that all processed data, code, and results are openly shared with the neuroscience community. Together, this research will establish that functional topography is essential for understanding EF, and will motivate trials of personalized neuromodulatory therapies.