There is an increasing recognition that vascular disease risk is associated with a greater incidence of cognitive impairment and dementia. Such link is supported by the physiological observation that cerebral metabolism heavily relies on vascular supply of oxygen and energy substrates. Cerebral hypoperfusion which results from vascular dysfunction causes a mismatch between energy demand and supply and is associated with the pathological features of dementia, including the impairments of action potential generation and protein synthesis, glutamatergic excitotoxicity, and the deposition of cerebral amyloid-β proteins. In contrast, habitual aerobic exercise is an established strategy to ameliorate the risk factors for vascular disease and is increasingly recognized in improving cognitive function. Accordingly, the primary purpose of this dissertation study was to investigate whether the exercise-related improvement in cognitive function was attributable to ameliorated vascular function and risk factors for vascular disease. In order to address this as comprehensively as we could, both cross-sectional and interventional studies were conducted. The primary findings from the present study were as follows. In the cross-sectional study, a greater cognitive performance observed in endurance-trained adults was associated with higher levels of cerebral CO2 reactivity and brachial endothelium-dependent vasodilation and lower levels of central arterial stiffness and plasma insulin. In the interventional study, a 3-month aerobic exercise training intervention did not improve cognitive function although central arterial stiffness and brachial endothelium-dependent vasodilation made favorable changes. However, we found that the improvement in memory performance after aerobic exercise training was associated with the reduction in central systolic blood pressure. Taken together, a better cognitive performance observed in endurance-trained adults may not directly be attributable to greater vascular function because there were discrepant changes in cognitive and vascular functions after a 3-month aerobic exercise intervention. The correlation between the changes in memory performance and central systolic blood pressure is interesting but needs further investigation using a larger sample size. The discrepancy in the results between the cross-sectional and interventional studies could be explained by the duration of exercise training and/or the time it takes for the effect of improved vascular function to translate into cognitive function. / text