Audiovisual integration and whole-brain networks in preterm and full-term neonates: A two-layer multiplex network perspective on structural and functional connectivity.

Audiovisual integration (AVI) is linked to the development of several cognitive abilities, rendering it a vehicle to better understand and anticipate a wide range of sequelae associated with preterm (PT) birth. In the present study, we aimed to complement the scarce literature on PT birth and AVI early in life, by investigating neonatal brain networks encompassing areas reported in the infant AVI literature. We used data from the Developing Human Connectome Project (http://www.developingconnectome.org/) to build two localized and one whole-brain connectomes from functional and structural connectivity data. Using graph analysis on a multiplex structural-functional brain network, we investigated the association between prematurity and (1) the edges present in functional and structural networks, (2) the integration and segregation properties of functional and structural networks, and (3) the inter-layer assortativity. We found substantial differences between PT and full-term (FT) neonates in the edges of the structural network at the whole-brain level and in one localized connectome. Across parcellation schemes, associations between prematurity and network efficiency were different, but similar inter-layer metrics were observed. In an exploratory analysis, we further showed lower functional connectivity strength in PT neonates. These findings are discussed in the light of perinatal brain developmental trajectories and deepen our understanding atypical AVI abilities in PT infants. More generally, the present work contributes to our understanding of whole-brain network development by investigating functional-structural coupling from a network neuroscience perspective.