Maturation of cortical connectivity in the core functional systems – infancy through childhood

1292 

Abstract Submission 

Jiaxin Tu¹, Aidan Latham¹, Jeanette Kenley¹, M. Camacho¹, Omid Kardan², Eric Feczko³, Sydney Kaplan¹, Trevor Day³, Chad Sylvester¹, Oscar Miranda Dominguez³, Lucille Moore³, Damien Fair³, Christopher Smyser¹, Jed Elison³, Timothy Laumann⁴, Evan Gordon⁴, Adam Eggebrecht⁵, Muriah Wheelock¹

¹Washington University in St. Louis, St. Louis, MO, ²University of Michigan, Ann Arbor, MI, ³University of Minnesota, Minneapolis, MN, ⁴Washington University, St. Louis, MO, ⁵Washington University School of Medicine, St. Louis, MO

Jiaxin Tu  
Washington University in St. Louis
St. Louis, MO

Aidan Latham  
Washington University in St. Louis
St. Louis, MO

Jeanette Kenley  
Washington University in St. Louis
St. Louis, MO

M. Camacho  
Washington University in St. Louis
St. Louis, MO

Omid Kardan  
University of Michigan
Ann Arbor, MI

Eric Feczko  
University of Minnesota
Minneapolis, MN

Sydney Kaplan  
Washington University in St. Louis
St. Louis, MO

Trevor Day  
University of Minnesota
Minneapolis, MN

Chad Sylvester  
Washington University in St. Louis
St. Louis, MO

Oscar Miranda Dominguez  
University of Minnesota
Minneapolis, MN

Lucille Moore  
University of Minnesota
Minneapolis, MN

Damien Fair  
University of Minnesota
Minneapolis, MN

Christopher Smyser  
Washington University in St. Louis
St. Louis, MO

Jed Elison  
University of Minnesota
Minneapolis, MN

Timothy Laumann  
Washington University
St. Louis, MO

Evan Gordon  
Washington University
St. Louis, MO

Adam Eggebrecht, PhD  
Washington University School of Medicine
St. Louis, MO

Muriah Wheelock  
Washington University in St. Louis
St. Louis, MO

Resting-state functional correlations (FC) of BOLD time-series detects brain regions that activate together at rest, revealing functional systems with distinguishable functionalities(Seitzman et al., 2019; Wig, 2017). Prior evidence suggested that the degree to which these systems are segregated – showing strong within-system FC relative to between-system FC – increases with brain development(Cao et al., 2017; Grayson & Fair, 2017; Tooley et al., 2022a), but the results were inconclusive and sometimes conflicting. Furthermore, FC within the same system are typically closer in space than those between systems(Zhi et al., 2022), yet the most significant developmental changes in connectivity are observed over longer distances(Liu et al., 2022), further complicating the interpretations.

We analyzed group-average vertex-wise FC resting-state fMRI data from three developmental windows: babies (0-3 years)(Howell et al. 2019; Kaplan et al. 2022), children (6-15 years)(Alexander et al. 2017), and young adults (19-32 years)(Power et al. 2014) , assessing the six core functional systems widely recognized for their distinct roles(Wig 2017; Uddin, Yeo, and Spreng 2019). Unlike traditional single-seed-based methods, we evaluated the average connectivity of all brain vertices within a system (Fig. 1A, 1B). We quantified the maturity of this system-average FC map using its eta-squared (η²) coefficient(Cohen et al. 2008) to that in the young adult. We additionally calculated the within and outside system FC for geodesic distance in fsLR standard mesh up to 150 mm in 10 mm bins to examine system segregation as a function of distance.

Our analysis revealed that system-average FC in full-term neonates closely resembles that in adults (η² = 0.82), with this similarity increasing throughout development in babies (0-3 year, growth rate of η² =0.019/year, P = 0.028) and children (6-15 year, growth rate of η² =0.003/year, P <0.001) (Fig. 1B). An age-group interaction showed a significant difference in growth rate (β = 0.016, P = 0.007). The maturity of connectivity in full-term neonates also varied across system (MI = 0.81±0.07 across six systems). There was a significance increase in maturity of connectivity in the default system for babies and children but not in the somatomotor system (Fig. 1B). We also obtain a matrix of within and between system FC by summing these average FC map within predefined system topography(Yeo et al., 2011) (Fig. 1C). We noted that while within-system FC generally increased from infancy to adulthood, between-system FC changes were more nuanced: the FC between default and frontoparietal systems decreased, but the FC between visual, somatomotor, dorsal attention and cingulo-opercular systems increased across development. Furthermore, we found that FC was higher within than outside systems across all distance bins (Fig. 2A) and systems (Fig. 2B) in young adults. System-specific long-range connectivity was also present in neonates, suggesting that it was established before birth, rather than developing throughout the first year of life (Fig. 2C). During development from neonate to adult, the within-system FC at >50 mm increased, and the within- and between-system FC at <50 mm decreased, leading to an overall increase in the difference between within-and between-system FC (Fig. 2C). These dynamics can be observed in all systems, with varying degrees of short-distance and long-distance system segregation across stages of development (Fig. 2D).

Our findings affirm that from birth, within-system connectivity is stronger than between-system connectivity for the core functional networks, and this disparity expands throughout development.

Normal Brain Development: Fetus to Adolescence ¹

Connectivity (eg. functional, effective, structural)

fMRI Connectivity and Network Modeling ²

Development

FUNCTIONAL MRI

Systems