ImportanceParenting is associated with brain development and long-term health outcomes, although whether these associations depend on the developmental timing of exposure remains understudied. Identifying these sensitive periods can inform when and how parenting is associated with neurodevelopment and risk for mental illness.ObjectiveTo characterize how harsh and warm parenting during early, middle, and late childhood are associated with brain architecture during adolescence and, in turn, psychiatric symptoms in early adulthood during the COVID-19 pandemic.Design, Setting, and ParticipantsThis population-based, 21-year observational, longitudinal birth cohort study of low-income youths and families from Detroit, Michigan; Toledo, Ohio; and Chicago, Illinois, used data from the Future of Families and Child Well-being Study. Data were collected from February 1998 to June 2021. Analyses were conducted from May to October 2023.ExposuresParent-reported harsh parenting (psychological aggression or physical aggression) and observer-rated warm parenting (responsiveness) at ages 3, 5, and 9 years.Main Outcomes and MeasuresThe primary outcomes were brainwide (segregation, integration, and small-worldness), circuit (prefrontal cortex [PFC]–amygdala connectivity), and regional (betweenness centrality of amygdala and PFC) architecture at age 15 years, determined using functional magnetic resonance imaging, and youth-reported anxiety and depression symptoms at age 21 years. The structured life-course modeling approach was used to disentangle timing-dependent from cumulative associations between parenting and brain architecture.ResultsA total of 173 youths (mean [SD] age, 15.88 [0.53] years; 95 female [55%]) were included. Parental psychological aggression during early childhood was positively associated with brainwide segregation (β = 0.30; 95% CI, 0.14 to 0.45) and small-worldness (β = 0.17; 95% CI, 0.03 to 0.28), whereas parental psychological aggression during late childhood was negatively associated with PFC-amygdala connectivity (β = −0.37; 95% CI, −0.55 to −0.12). Warm parenting during middle childhood was positively associated with amygdala centrality (β = 0.23; 95% CI, 0.06 to 0.38) and negatively associated with PFC centrality (β = −0.18; 95% CI, −0.31 to −0.03). Warmer parenting during middle childhood was associated with reduced anxiety (β = −0.05; 95% CI −0.10 to −0.01) and depression (β = −0.05; 95% CI −0.10 to −0.003) during early adulthood via greater adolescent amygdala centrality.Conclusions and RelevanceNeural associations with harsh parenting were widespread across the brain in early childhood but localized in late childhood. Neural associations with warm parenting were localized in middle childhood and, in turn, were associated with mental health during future stress. These developmentally contingent associations can inform the type and timing of interventions.