OBJECTIVES: We studied if late preterm birth (34 weeks 0 days–36 weeks 6 days of gestation) is associated with performance on the Consortium to Establish a Registry for Alzheimer’s Disease Neuropsychological Battery (CERAD-NB) in late adulthood and if maximum attained lifetime education moderated these associations. METHODS: Participants were 919 Finnish men and women born between 1934 and 1944, who participated in the Helsinki Birth Cohort Study. They underwent the CERAD-NB at a mean age of 68.1 years. Data regarding gestational age (late preterm versus term) were extracted from hospital birth records, and educational attainment data were gathered from Statistics Finland. RESULTS: After adjustment for major confounders, those born late preterm scored lower on word list recognition (mean difference: –0.33 SD; P = .03) than those born at term. Among those who had attained a basic or upper secondary education, late preterm birth was associated with lower scores on word list recognition, constructional praxis, constructional praxis recall, clock drawing, Mini–Mental State Examination, and memory total and CERAD total 2 compound scores (mean differences: >0.40 SD; P values <.05), and had a 2.70 times higher risk of mild cognitive impairment (Mini–Mental State Examination score: <26 points) (P = .02). Among those with tertiary levels of education, late preterm birth was not associated with CERAD-NB scores. CONCLUSIONS: Our findings offer new insight into the lifelong consequences of late preterm birth, and they add late preterm birth as a novel risk factor to the list of neurocognitive impairment in late adulthood. Our findings also suggest that attained lifetime education may mitigate aging-related neurocognitive impairment, especially among those born late preterm.