Cognitive decline may lead to dementia whose most frequent cause is Alzheimer's disease (AD). Among the many potential risk factors of cognitive decline and AD, diet raises increasing interest. Most studies considered diet in the frame of a single nutrient approach with inconsistent results. A novel approach to examine the link between nutrition and cognitive function is the use of dietary patterns. The aim of the present review was to update and complete the body of knowledge about dietary patterns in relationship with various cognitive outcomes in the elderly. Two approaches can be used: a priori and a posteriori patterns. A priori patterns are defined by the adhesion to a pre-defined healthy diet using a score such as the Mediterranean diet (MeDi) score, the Healthy Eating Index, the Canadian Healthy Eating Index, the French National Nutrition and Health Programme (Programme National Nutrition Santé) Guideline Score (PNNS-GS), the Recommended Food Score (RFS) and Dietary Approaches to Stop Hypertension (DASH). MeDi score, RFS, PNNS-GS and DASH have been associated with lower risks of cognitive impairment, cognitive decline, and dementia or AD. Principal components analysis, reduced rank regression and clustering methods allow the identification of 'healthy' patterns associated with lower risk of cognitive decline. However, some studies did not report any associations with cognitive outcomes and results are discordant especially regarding MeDi and the risk of dementia. Several methodological challenges should be overcome to provide a higher level of evidence supporting the development of nutritional policies to prevent cognitive decline and AD.