Abstract
The human brain is composed of a multitude of areas with distinct roles in mental function; these regions connect and interact with a set of other brain regions, forming functional networks which are the basis of large-scale information processing in the brain. These functional networks can be reliably identified in human adults as well as other species, and small individual differences in connectomes may contribute to individual differences in behavior. Here, we aim to characterize these functional networks in neonates scanned within one week of life. We use unsupervised learning to uncover underlying patterns in resting-state functional connectivity data from a large cohort of neonates (N = 267 full-term infants from the Developing Human Connectome Project). First, we present the neonate networks determined by optimal solutions in terms of fit and replicability. We found symmetrical and hierarchical networks associated with sensorimotor, visual, default mode, ventral attention, and high-level sensory areas. Second, we explored the inter- and intra-subject variability of these networks and found that some networks (e.g. sensorimotor) had low inter-subject variability whereas others (e.g. dorsal attention) had high inter-subject variability while maintaining high intra-subject consistency. Third, we compared the neonate networks to those in adults (Yeo et al., 2011) and found similarities with sensorimotor, visual, dorsal and ventral attention, and default mode networks. However, frontoparietal and limbic networks found in adults were not discernible in neonates. Finally, we investigated differential gene expression, determined by the Allen Human Brain Atlas, as a potential explanation for the emergence of these distinct networks, and quantified within and between network similarity. These results suggest the basic network structure present in adults also exists at birth, but some important differences, particularly in association cortex, suggest a role for maturation and experience in developing adult-like functional brain organization.