Psychopathology risk reshapes the hierarchical socioeconomic impact on personalized functional network topography in children

Introduction

Social environmental factors, including socioeconomic status (SES), play a critical role in shaping neurodevelopmental processes and mental health outcomes [1]. Recent studies have shown that widening socioeconomic disparities parallel a growing risk of psychopathology in children [2], underscoring the need to understand how socioeconomic factors become biologically embedded in the developing brain. Prior work has shown that socioeconomic influences on personalized functional brain networks follows a hierarchical cortical organization, with stronger effects in higher-order association cortex [3]. However, it remains unclear how psychopathology risk influences the association between SES and personalized functional network organization in children.

Methods

This study included more than 6,000 children aged 9-10 years old from the Adolescent Brain Cognitive Development (ABCD) Study. With personalized functional networks derived from resting-state functional magnetic resonance imaging data (fMRI), participants were stratified into high-risk and low-risk subgroups based on quartiles of a general psychopathology score (p-factor) [4]. Analyses were conducted across three SES indicators: household income measured by family income-to-needs ratio (INR), parental education (EDU), and neighborhood disadvantage measured by area deprivation index (ADI). Partial least square regression with repeated random two-fold cross-validation framework was used to evaluate the association between functional network topography and three SES factors [5].

Results

A total 6,001 children aged 9-10 years were included; 3003 (50.04%) were female, and the mean (SD) age was 9.96 (0.63) years. With a cross-validated machine learning approach, we found that all three SES factors could be significantly predicted from personalized functional network topography in both low-risk and high-risk subgroups. Prediction accuracy was measured by median r between predicted scores and observed SES scores across 101 repetitions (INR = 0.21-0.25, EDU = 0.24-0.25, ADI = 0.19-0.25, all P values < 0.001 based on 1,000 times permutation). However, prediction performance was modestly attenuated in the high-risk subgroup compared with the low-risk subgroup for family income and neighborhood disadvantage (median r: INR = 0.047, ADI = 0.066), indicating greater heterogeneity in relations between socioeconomic factors and personalized functional networks in children with elevated psychopathology risk. Despite these differences, spatial patterns of prediction weights in the high-risk subgroup exhibited a hierarchical reorganization, showing stronger alignment with the sensorimotor–association cortical axis (alignment is assessed by Spearman rank correlation, for all three SES factors, r = 0.29-0.36 for low risk subgroups, r = 0.41-0.46 for high risk subgroups), suggesting that higher-order association cortex shows greater sensitivity to socioeconomic-related variation in children with higher psychopathology risk.

Conclusions

Together, these findings suggest that psychopathology risk may reshape the hierarchical organization of how social environmental factors are represented in personalized functional brain networks during childhood, highlighting association cortex as a key interface linking socioeconomic context to overall psychopathology risk. Elucidating the mechanisms through which socioeconomic factors influence neurodevelopment may inform the development of targeted interventions and support the early identification of children with mental health problems, ultimately promoting more adaptive developmental trajectories.

References

1. Yu, C., W. Jiang, and D. Zhao., 2025. Association between adverse home environments and symptoms in drug-naive children with attention-deficit/hyperactivity disorder. Gen Psychiatr. 38(4): p. e102078.
2. Rakesh, D., Tsomokos, D.I., Vargas, T. et al. 2025. Macroeconomic income inequality, brain structure and function, and mental health. Nat. Mental Health 3, 1318–1330.
3. Zhao S, Su H, Cong J, et al. 2024. Hierarchical individual variation and socioeconomic impact on personalized functional network topography in children. BMC Med. 22(1):556.
4. Wen, X., Qu, D., Zhang, X. et al. 2023. The transition trajectories of self-injurious thoughts and behaviours among children from a biopsychosocial perspective. Nat. Mental Health 1, 782–791.
5. Cui Z, Pines AR, Larsen B, et al. 2022. Linking Individual Differences in Personalized Functional Network Topography to Psychopathology in Youth. Biol Psychiatry. 92(12):973-983.