Objectives: To investigate the association between African and Native American genomic ancestry and long-term cognitive trajectories in admixed Brazilians. Design: Population-based longitudinal study. Setting: Bambui-Epigen (Brazil) cohort study. Participants: Adults aged 60 and older (N=1,215). Measurements: Participants were followed from January 1997 to December 2011. Cognitive function was assessed annually using the Mini-Mental State Examination (MMSE), totaling 12,208 measurements. We used linear mixed-effects pattern models to assess MMSE score trajectories. Ancestry was assessed using a genome-wide approach. Results: After adjustments for covariates, the highest quintile of African ancestry was associated with poorer baseline cognitive performance (β=–0.73, 95% confidence interval (CI)=–1.36 to –0.11) but not with cognitive trajectory. Educational level modified the baseline association between highest African ancestry and cognitive performance in that the association was observed only in those with very low (<4 years) education (β=–1.13, 95% CI=–2.02 to –0.23). No association was found between Native American ancestry and baseline cognitive function or its trajectory. Conclusion: Genomic African and Native American ancestry levels had no prognostic value for age-related cognitive decline in this admixed population.