Beyond the average brain: individual differences in social brain development are associated with friendship quality.

Andrik I Becht,Sarah-Jayne Blakemore,Lara M Wierenga,Anna Van Duijvenvoorde,Eveline A Crone,Rosa Meuwese,Berna Güroğlu,Kathryn L Mills

doi:10.1093/scan/nsaa166

Abstract

We tested whether adolescents differ from each other in the structural development of the social brain and whether individual differences in social brain development predicted variability in friendship quality development. Adolescents (N = 299, Mage T1 = 13.98 years) were followed across three biannual waves. We analysed self-reported friendship quality with the best friend at T1 and T3, and bilateral measures of surface area and cortical thickness of the medial prefrontal cortex (mPFC), posterior superior temporal sulcus (pSTS), temporoparietal junction (TPJ) and precuneus across all waves. At the group level, growth curve models confirmed non-linear decreases of surface area and cortical thickness in social brain regions. We identified substantial individual differences in levels and change rates of social brain regions, especially for surface area of the mPFC, pSTS and TPJ. Change rates of cortical thickness varied less between persons. Higher levels of mPFC surface area and cortical thickness predicted stronger increases in friendship quality over time. Moreover, faster cortical thinning of mPFC surface area predicted a stronger increase in friendship quality. Higher levels of TPJ cortical thickness predicted lower friendship quality. Together, our results indicate heterogeneity in social brain development and how this variability uniquely predicts friendship quality development.

Highlights

A critical assumption is that some trajectories of change may be more malleable to environmental input than others (Noble et al, 2015; Piccolo, Merz, He, Sowell, & Noble, 2016), which may be indicated by individual differences in the baseline and speed of brain maturation that define a certain window of opportunity (Crone & Elzinga, 2014)
We first tested for mean level changes in social brain regions, using latent growth curve models
For all social brain regions a quadratic model including random linear and random quadratic slopes provided the best fit to the data (Supplementary Table S.2 shows the fit indices AIC and BIC for the different models)

Summary

Introduction

A critical assumption is that some trajectories of change may be more malleable to environmental input than others (Noble et al, 2015; Piccolo, Merz, He, Sowell, & Noble, 2016), which may be indicated by individual differences in the baseline and speed of brain maturation that define a certain window of opportunity (Crone & Elzinga, 2014). Some developmental growth patterns may be more genetically influenced and show relative constant changes for all individuals over time, comparable to developmenta

Objectives

Methods

Results

Discussion

Conclusion