The potential role of education attained after the age of 50, for example, vocational training or recreational courses, in cognitive reserve has been unexplored. We examined the cross-sectional and prospective associations between late life education (LLE) and global cognitive function in older adults.5306 participants (50+ years) in The Irish Longitudinal Study of Ageing answered questions about highest level of education completed and LLE (2010). Cognitive function was defined as the number of errors on the MoCA assessed in 2010 and 2014. The association between LLE and MoCA-errors was examined using poisson regression stratified by level of education. Sensitivity analyses were done to examine reverse causation and selection bias.In those with primary/no (n = 1312, IRR = 0.83, 95%CI = 0.70-0.99) and secondary education (n = 2208, IRR = 0.88, 95%CI = 0.80-0.97), but not tertiary education (n = 1786, IRR = 0.93, CI = 0.86-1.00), participating in LLE was associated with lower rate of MoCA errors. The prospective association between LLE and 4-year change in MoCA-errors was (borderline) statistically significant in those with primary/no education only (IRR = 0.86, CI = 0.74-1.00). Sensitivity analyses supported robustness of the findings.LLE may contribute to cognitive reserve and be a useful intervention to mitigate the increased risk of cognitive decline associated with low levels of education. This article is protected by copyright. All rights reserved.