Abstract
Introduction: Hippocampus, a medial temporal lobe structure, has significant implications in memory formation and learning. Although hippocampus activity is believed to be affected by socioeconomic status (SES), limited knowledge exists on which SES indicators influence hippocampus function. Purpose: This study explored the separate and combined effects of three SES indicators, namely parental education, family income, and neighborhood income, on adolescents' hippocampus activation during an N-Back memory task. As some of the effects of parental education may be through income, we also tested if the effect of parental education on hippocampus activation during our N-Back memory task is mediated by family or neighborhood income. Methods: The Adolescent Brain Cognitive Development (ABCD) study is a national multi-center investigation of American adolescents' brain development. Functional magnetic resonance imaging (fMRI) data of a total sample of 3067 9-10-year-old adolescents were used. The primary outcome was left- hippocampus activation during the N-Back memory task (mean beta weight for N-Back run 1 2 back versus 0 back contrast in left hippocampus). The independent variable was parental education. Family income and neighborhood income were two possible mediators. Age, sex, and marital status were the covariates. To test mediation, we used hierarchical linear regression models first without and then with our mediators. Full mediation was defined according to Kenny. The Sobel test was used to confirm statistical mediation. Results: In the absence of family and neighborhood income in the model, higher parental educational attainment was associated with lower level of left hippocampus activation during the N-Back memory task. This effect was significant while age, sex, and marital status were controlled. The association between parental educational attainment and hippocampus activation during the N-Back memory task was no more significant when we controlled for family and neighborhood income. Instead, family income was associated with hippocampus activation during the N-Back memory task. These findings suggested that family income fully mediates the effect of parental educational attainment on left hippocampus activation during the N-Back memory task. Conclusions: The effect of parental educational attainment on adolescents' hippocampus activation during an N-Back memory task is fully explained by family income. That means low family income is why adolescents with low-educated parents show highlighted hippocampus activation during an N-Back memory task. Given the central role of the hippocampus in learning and memory and as income is a modifiable factor by tax and economic policies, income-redistribution policies, fair taxation, and higher minimum wage may have implications for promotion of adolescent equality and social justice. There is a need to focus on family-level economic needs across all levels of neighborhood income.