Background: Research concerning the link between individual vascular risk factors and cognition is plentiful but few studies have investigated the role of global vascular risk. We examined the cross-time associations of several vascular risk scores with cognitive performance during aging. Methods: Using data from the French SU.VI.MAX cohort, we studied a sample of 3061 participants. Framingham coronary heart disease, cardiovascular and stroke risk profiles were computed using baseline data (1994-1996). Cognitive performance was assessed after a mean of 13 years via a battery of six validated instruments. Principal component analysis identified scores for verbal memory and working memory. Associations between risk profiles (as continuous variables and in quartiles (Q)) and subsequent poor performance (defined as cognitive score <= 10th percentile) were examined via logistic regression (odds ratios, 95% CI) and analysis of covariance. Results: All continuous-scale Framingham risk scores assessed at midlife were inversely and uniformly associated with subsequent poor global cognitive performance, and especially in terms of verbal memory. Considering risk score Q, higher Q were associated with poorer performance in verbal memory: The fully-adjusted odds ratios (95% CI), comparing Q4 versus Q1, were 2.84 (1.70, 4.75), 2.31 (1.43, 3.73) and 1.77 (1.13, 2.76) for Framingham coronary heart disease, cardiovascular and stroke risk profiles, respectively. Similar findings were observed when modeling cognitive outcomes as continuous variables using covariance analyses. Conclusion: This study supports the existence of an inverse cross-time association between midlife vascular risk profiles and subsequent poor cognitive performance, especially in the verbal memory domain. Beyond their importance as regards vascular risk, such risk scores may help primary prevention efforts in identifying and targeting middle-aged individuals at high risk of cognitive aging.