Human and animal studies indicate that cell-signalling, anti-inflammatory and anti-oxidant properties of flavonoids and carotenoids in fruits and vegetables (F&V) may protect against, or slow, age-related decline in cognitive functioning. The Flavonoid University of Reading Study (FLA VURS), designed to explore the dose-response relationship between dietary F&V flavonoids and cardiovascular disease, enabled the investigation of such an association with cognitive function. FLA VURS was an 18-week parallel 3-arm randomised controlled dietary intervention trial with four time points, measured at 6-weekly intervals from baseline. Low F&V consumers were randomly assigned to a high flavonoid (HP), low flavonoid (LF), or control group. HF or LF F&V intake was increased by two daily 80g portions every 6 weeks, while controls maintained their habitual diet. Cognitive function and mood were measured at each visit. While no overall group differences in cognition or mood were found, age group sub-analyses (26- 50 and 51-70 years of age) showed differences from 0-18 weeks for younger adults in spatial working memory (SWM), with LF improving significantly more than the other two groups. Subsequent regression analysis was performed irrespective of dietary group, with F & V components as predictors. For the group as a whole, improvements in SWM were associated with increased flavonoid levels. For the group as a whole, better baseline SWM performance was associated with higher BDNF levels, and better baseline recognition and executive function performance were associated with higher flavonoid levels. For younger adults, better executive function at baseline was associated with higher carotenoid levels, whereas for older adults, it was associated with higher flavonoid levels. The overall lack of flavonoid treatment effects was surprising, although epidemiological predictions were supported. Further research is required to determine the relationship between F&V intake and cognitive function, the mechanisms by which age-dependent differences in F&V responsiveness may occur, and also to understand differences between F&V components on cognition when consumed as part of a normal healthy diet. The theoretical, practical and methodological implications of these findings are discussed.