Nutrition is an important lifestyle factor that can modify the risk of future cognitive impairment and dementia. Some, but not conclusive, evidence (mostly from observational studies and infrequently from clinical trials) exists of a protective association between certain nutrients (eg, folate, flavonoids, vitamin D, and certain lipids) or food groups (eg, seafood, vegetables, and fruits, and potentially moderate alcohol and caffeine consumption) and cognitive outcomes in older people. For some nutrients and food groups, protection might be greater in individuals with either deficiencies in certain nutrients or a genetic predisposition to cognitive impairment. Identification of potentially different associations between such subgroups should be a priority for future research. At present, evidence of an association between nutrition and cognitive outcomes is somehow stronger for healthy dietary patterns, such as the Mediterranean-type diet, than for individual nutrients and food groups, possibly because of the cumulative beneficial effects of the many ingredients in these diets. Multidomain interventions (including a nutrition component) might also hold some promise for the prevention of cognitive impairment and dementia, but their effectiveness is still uncertain. Use of advanced technologies for nutrition assessment (eg, metabolomics and innovative methods of dietary intake assessment) and recently identified biomarkers of nutrition and neurobiological outcomes will be important to achieve this goal.