The mechanisms underlying cognitive impairment are incompletely understood but may include arterial stiffness and microvascular dysfunction. In the population-based Maastricht Study, we investigated the association between arterial stiffness and cognitive performance, and whether any such association was mediated by microvascular dysfunction. We included cross-sectional data of 2544 participants (age, 59.7 years; 51.0% men; 26.0% type 2 diabetes mellitus). We used carotid-femoral pulse wave velocity and carotid distensibility coefficient as measures of aortic and carotid stiffness, respectively. We calculated a composite score of microvascular dysfunction based on magnetic resonance imaging features of cerebral small vessel disease, flicker light-induced retinal arteriolar and venular dilation response, albuminuria, and plasma biomarkers of microvascular dysfunction (sICAM-1 [soluble intercellular adhesion molecule-1], sVCAM-1 [soluble vascular adhesion molecule-1], sE-selectin [soluble E-selectin], and vWF [von Willebrand factor]). Cognitive domains assessed were memory, processing speed, and executive function. A cognitive function score was calculated as the average of these domains. Higher aortic stiffness (per m/s) was associated with lower cognitive function (β, -0.018 SD [95% CI, -0.036 to -0.000]) independent of age, sex, education, and cardiovascular risk factors, but higher carotid stiffness was not. Higher aortic stiffness (per m/s) was associated with a higher microvascular dysfunction score (β, 0.034 SD [95% CI, 0.014 to 0.053]), and a higher microvascular dysfunction score (per SD) was associated with lower cognitive function (β, -0.089 SD [95% CI, -0.124 to -0.053]). Microvascular dysfunction significantly explained 16.2% of the total effect of aortic stiffness on cognitive function. The present study showed that aortic stiffness, but not carotid stiffness, is independently associated with worse cognitive performance, and that this association is in part explained by microvascular dysfunction.