Cognitive Control Decision-Making Dynamics Across Internalizing and Externalizing Symptoms in Youth

Childhood behavioral problems are associated with significant long-term consequences, yet the underlying cognitive mechanisms remain poorly understood. In this study, we employed computational modeling alongside traditional reaction time (RT) measures to investigate cognitive control during a flanker task. We evaluated the predictive utility of these methods in explaining variance across eight transdiagnostic symptom domains in late childhood (mean age = 10.0 years; n = 10,343) from the Adolescent Brain Cognitive Development study. We compared simultaneous regression models across congruent and incongruent conditions using an RT-only model and a drift–diffusion model (DDM) that incorporated boundary separation, bias, drift rate, and non-decision time parameters. Results from the RT-only models indicated that slower reaction times across both task conditions were associated with higher scores on most symptom subscales, suggesting more behavioral problems. For both task conditions, DDM regressions accounted for more total variance across symptom domains compared to RT models. Additionally, DDM regressions demonstrated that impoverished evidence accumulation emerged as a shared feature of both internalizing and externalizing behaviors, while reductions in non-decision time, indicative of increased impulsiveness, were unique to rule-breaking and aggressive behaviors. These findings suggest that different aspects of cognitive control are associated with specific behavioral problems in children, rather than just overall response speed. Present results provide new insights into cognitive control dynamics and suggest that targeting ineffective cognitive control could be crucial for the prevention and intervention of childhood psychopathology.