There is interest in age-related cognitive decline and environmental risk factors for Alzheimer's disease (AD). This interest is focused on individual differences in exposure to agents that may harm or protect cognitive function. Caffeine is used as a short acting mental stimulant and may possess longer-term properties that protect against age-related decline and, possibly, AD. The current study aimed to: 1) examine current cognitive function in a narrow age range sample (n=351) without dementia (MMSE>25) who are, by reason of age, entering the period of increased risk of AD; and 2) link cognitive function to self-reported intake of caffeine and socioeconomic status (SES). Possible confounding by gender, childhood intelligence, education, and symptoms of anxiety and depression was introduced into the statistical model. There were significant differences between SES groups in caffeine intake (p< 0.05) and cognitive performance (p< 0.001). Higher quartiles of caffeine intake were associated with slower digit symbol speed (F =3.38, p< 0.02) but this finding was removed after allowing for SES. The results are discussed in terms of the withdrawal effects of caffeine during cognitive testing and strong links between SES and cognitive performance. No evidence in support of cognitive enhancing effects of caffeine was found.