The efficiency of higher cortical functions, such as memory and speed of complex information processing, tends to decrease with advancing age in normal healthy individuals. Recently, a high habitual intake of caffeine was found associated with better verbal memory performance and psychomotor speed in several cross-sectional population studies. We tested the hypothesis that habitual caffeine intake can reduce or postpone age-related cognitive decline in healthy adults. For this purpose, the cognitive performance of all participants in the maastricht aging study (maas), aged between 24 and 81 years, was reassessed after 6 years. Information on the intake of caffeine-containing beverages was available from the baseline questionnaire. After 6 years, 1376 (75.6%) individuals were available for reassessment. After correction for demographic characteristics, baseline performance and health status, there were small albeit significant associations between the overall estimated caffeine intake at baseline and the 6-year change in complex motor speed (motor choice reaction time). The earlier found association between caffeine intake and verbal memory performance was not apparent in this longitudinal study. These results imply that the longitudinal effect of habitual caffeine intake is limited and will not promote a substantial reduction in age-related cognitive decline at a population level.