In our recent study of glucose-expectancy effects in young adults, no convincing benefits of glucose were found. This study used the same cognitive tasks (delayed free-recall, spatial recognition, semantic classification and immediate free-recall) and dose of glucose (50 g) with three different age groups: mean ages 21 years (N=21), 38 years (N=19) and 69 years (N=22). Participants attended two sessions, one week apart, after fasting overnight, with the influence of glucose assessed using a cross-over design. In addition to the expected effects of task structure on performance measures, the oldest adults had lower levels of delayed and immediate free-recall, and slower reaction times in the spatial recognition tasks than the young and middle groups, which did not differ from each other. In the delayed free-recall task, a drink x age group interaction showed that glucose only improved recall for the middle-aged group. There was no corresponding effect on immediate free-recall. In the spatial recognition task, a drinkÃ—truthÃ—age group interaction showed that glucose did not influence target â€œpresentâ€� decisions, but slowed target â€œabsentâ€� decisions only in the oldest group. The results are broadly consistent with an effect of glucose on hippocampal function and suggest that the most beneficial effects of glucose may be for middle-aged rather than older adults.