Logo: to the web site of Uppsala University

Previous research and theory indicate an importance of the quality of the early caregiving environment in the development of self-regulation. However, it is unclear how attachment security and maternal sensitivity, two related but distinct aspects of the early caregiving environment, may differentially predict self-regulation at school start and whether a distinction between hot and cool executive function is informative in characterizing such predictions through mediation. In a 5-year longitudinal study (n = 108), we examined these associations using measures of maternal sensitivity and attachment security at 10–12 months, executive function at 4 years, and self-regulation at 6 years. Surprisingly, and despite methodological rigor, we found few significant bivariate associations between the study variables. We found no credible evidence of a longitudinal association between maternal sensitivity or attachment security in infancy and self-regulation at 6 years, or between executive function at 4 years and self-regulation at 6 years. The lack of bivariate longitudinal associations precluded us from building mediation models as intended. We discuss our null findings in terms of their potential theoretical implications, as well as how measurement type, reliability, and validity, may play a key role in determining longitudinal associations between early caregiving factors and later self-regulation and related abilities.

Background

Identifying predictors and mechanisms in the development of childhood internalizing (INT) and externalizing (EXT) problems is crucial for early intervention. Inhibitory control has been linked to INT and EXT, with emotion regulation (ER) potentially mediating these associations. However, specific pathways between early inhibitory control, ER, and later INT and EXT remain unclear. Additionally, regulation of distinct emotions (anger, fear, sadness, joy) may play a role.

Methods

The sample included 94 typically developing children from the EFFECT study, a longitudinal project on the development of self-regulation. At age 4, inhibitory control was measured using the Day/Night Stroop Task. At age 6, general ER, as well as regulation of specific emotions (anger, fear, sadness, and joy), were assessed using the Emotion Questionnaire (parent-report). INT and EXT at ages 9–10 were measured using the Strengths and Difficulties Questionnaire (parent-report). Correlational and path analyses were conducted.

Results

No longitudinal associations were found between inhibitory control at 4 years and either INT or EXT at ages 9–10, or with ER at age 6. Consequently, we found no evidence of mediation by ER. General ER at 6 years emerged as a predictor of both INT and EXT at 9–10 years. While not statistically significant, effect sizes linking regulation of some specific emotions (anger, fear) with subsequent INT and EXT problems warrant further research.

Conclusion

The results reflect the complexity of studying longitudinal effects of early inhibitory control. A modest sample size with attrition, and measurement constraints may have attenuated effects and limited generalizability. Meanwhile, our findings highlight ER as a target for intervention across both INT and EXT.

Background: Cognitive control develops progressively from childhood through early adulthood, supported by protracted maturation of the dorsal anterior cingulate cortex (dACC). While structural and functional development of this region is well-characterized, the neurochemical substrates underlying its maturation remain largely unexplored. Choline, a metabolite essential for membrane synthesis, myelination, and cholinergic neurotransmission, may play a critical role in dACC development.

Methods: We used proton magnetic resonance spectroscopy (.H-MRS) to measure choline concentration (tCho/tCr) in the dACC of 106 participants (23 children aged 6-9 years, 38 adolescents aged 13-17 years, 45 adults aged 30-40 years). Cognitive control performance was assessed using a composite measure derived from multiple cognitive control tasks. Bayesian linear regression models examined age-group differences in choline concentration and associations between choline and cognitive control performance.

Results: Adults showed higher dACC choline concentrations than both children and adolescents, while evidence for differences between children and adolescents was inconclusive. Critically, the relationship between choline and cognitive control performance showed a developmental shift; the association was negative in children, inconclusive in adolescents, and positive in adults, with strong evidence of a difference between children and adults.

Conclusions: These findings suggest a developmental change in the functional significance of dACC choline concentration, potentially reflecting a shift from supporting active circuit reorganization in childhood to maintaining established networks in adulthood. The cross-sectional design limits causal inference, and we emphasise the need for replication given the modest sample size. Meanwhile, this study provides the first characterization of dACC choline concentration across development, and indicates that a single biomarker may carry different functional significance at different developmental stages.