Less childhood education is a potentially modifiable risk factor for developing incident dementia but it is not known if education in later life is protective. We sought to add to previous work by testing the association between adult education and brain volume as well as exploring the impact of continuing adult education versus intermittent participation. We used data from participants of the UK Biobank cohort, with no prevalent dementia who were asked about adult education participation at baseline and at follow-up. Dementia status was ascertained from self-report or electronic health records. Cox proportional hazards models were built to estimate hazard ratios (HRs) between participation in adult education and dementia risk. In 499,337 participants aged between 40 and 69 at baseline with 13.2years mean follow-up, in analyses adjusted for age, sex, education, deprivation, ethnicity, hypertension, diabetes, ethnicity, obesity, smoking, alcohol use, physical inactivity and social isolation, we replicated previous findings of a protective effect of adult education on dementia risk (HR 0.82, 95% CI 0.74-0.90, P < 0.001), and showed a trend towards protection against dementia if adult education was continued rather than intermittent. Additionally, adult education did not impact on total brain volume (coefficient - 657.4, 95% CI - 2795.1 to 1480.3, P = 0.547) but it was associated with increased hippocampal volume (coefficient 33.9, 95% CI 8.9 to 59.0, P = 0.008) indicating a potential mechanism for protection against dementia. We have added evidence indicating that continuing adult education participation may be beneficial, although numbers for this analysis were very small. Analysis of brain volume indicated that adult education may have a protective effect by preserving hippocampal size or slowing volume loss, in line with the cognitive reserve hypothesis.