The age-related cognitive decline and the increase in dementia patients are large problems in societies with growing ageing populations. No cure is present for dementia, while the available medication only focuses on alleviating symptoms. It is therefore of major importance to find risk factors that can modify the development of cognitive decline and dementia. Pre-clinical and observational studies suggest a role for nutrients. Evidence derived from randomized controlled trials (RCTs) is, however, limited and equivocal with most studies showing no effect and only a few studies showing a beneficial effect of a nutritional intervention. In the current thesis, we investigated the acute and longer-term effects of different nutrients, i.e. glucose and sucrose, protein, resistance-type exercise training with or without protein, and vitamin B12 and folic acid in order to optimize and preserve cognitive functions in non-demented elderly people. A comprehensive literature review was performed on the acute effects of glucose and sucrose on cognitive performance (<strong>Chapter 2</strong>). Glucose is the most important fuel for the brain, and as such, manipulation of the supply of glucose may affect cognitive functions. The main conclusion of our review was that a glucose load may have a short-term beneficial effect on episodic memory. Enhancing effects on other cognitive domains were less clear, partly due to the small number of studies examining these effects. Limited research was also done on the possible effects of sucrose on cognitive functions. Therefore, we investigated the acute effects of 50 g of glucose and 100 g of sucrose on a broad spectrum of cognitive functions reflecting performance on episodic memory, working memory, attention and information processing speed, and executive functions (<strong>Chapter 3</strong>). This was done by a cross-over study in 43 elderly participants who had self-reported memory complaints. In contrast to the conclusion of our review, we did not observe an effect of glucose or sucrose on episodic memory, though we showed a beneficial effect of sucrose on attention and information processing speed. Protein supplementation was the next nutritional intervention that was investigated. Several amino acids are precursors for neurotransmitters, and their supply may affect the synthesis and release of these neurotransmitters, and may consequently affect cognitive performance. A 24-week randomized placebo-controlled trial was carried out in 65 frail and pre-frail elderly people (<strong>Chapter 4</strong>). The protein supplementation included twice a day 15 grams of protein in the form of a drink. Reaction time improved more in the protein group compared to the placebo group, but the scores on the cognitive domains, i.e. episodic memory, attention and working memory, information processing speed, and executive functions, or the other single test scores, did not differ between treatment groups. In addition, we investigated the effects of 24 weeks resistance-type exercise training with and without protein supplementation in pre-frail and frail elderly people (<strong>Chapter 5</strong>). Exercise training without extra protein (n=62) improved performance on the domain attention and working memory. Exercise training together with protein supplementation (n=65) improved performance on information processing speed. Last, the role of vitamin B12 and folate on cognitive health was investigated. Low levels of these nutrients can increase homocysteine levels, which is a suggested risk factor for cognitive decline. The effect of daily supplementation with 500 µg vitamin B12 and 400 µg folic acid was investigated in 2,919 participants for two years (<strong>Chapter 6</strong>). Global cognitive function and episodic memory were assessed in the total study population, whereas extensive neuropsychological testing was done in a subpopulation (n=856). B-vitamin supplementation did not improve cognitive domain scores. Only a small, though significant, effect was observed on global cognitive performance, measured by the Mini-Mental State Examination, but this was suggested to be due to chance. Brain MRI scans were made in a subgroup (n=218) after two years of intervention to obtain volumetric measures of grey and white matter, and total brain volume (<strong>Chapter 7</strong>). We investigated the cross-sectional associations between follow-up levels of folate, homocysteine and three vitamin B12 status biomarkers, e.g. methylmalonic acid, holotranscobalamin and serum vitamin B12, and brain volumes. Fully adjusted regression models showed a borderline significant association between plasma homocysteine and total brain volume, with a stronger association in the group that received B-vitamin supplementation. Serum B12 and holotranscobalamin were not associated with brain volumes, whereas high methylmalonic acid levels were associated with lower brain volumes in the group that received B-vitamins. In contrast, higher folate levels were associated with lower total brain volumes. In addition, when comparing the group that received two years of B-vitamin supplementation and those who did not, we observed lower brain volumes in the B-vitamin group, which might be a result of a difference in age between the two groups. To conclude, the nutritional intervention studies showed little evidence for a beneficial effect on cognitive performance in relatively healthy older adults. Given the large problem of dementia, research on modifiable risk factors, including nutrition, should continue, with well thought out research methods, including large and long-term observational and intervention studies with high-sensitive study populations and early biomarkers (e.g. imaging techniques) for cognitive decline in combination with neuropsychological tests. In this way, nutrition can be added to the list of lifestyle factors that can fight dementia.