One of the many challenges society faces at present, is the rapidly increasing aging of the population and all of its implications. As a consequence of increased life expectancy, one can expect an increasing occurrence of aging-associated medical conditions like dementia. Alzheimer's disease (AD) is the most prevalent type of dementia, and frequently begins in late life, after the age of 60 years, although in rare cases the disease may begin earlier. The progression of the disease is a gradual and continuous process, and the average patients may expect to live from 8 to 10 years after symptom onset. Mild cognitive impairment (MCI) is generally considered as an intermediate state of cognitive function between the changes seen in aging and those fulfilling criteria for dementia and often AD. Most aging people are confronted with a gradual cognitive decline, typically with regard to memory. However, the decline is usually minor and does not compromise the ability to function. In MCI, the decline in cognitive function is beyond that associated with typical aging. Currently, only symptomatic treatment for AD is available in everyday clinical practice. Thus, it is of extreme importance to start pharmacological treatment as soon as possible, in order to slow down the progression of the disease and preserve quality of life as much as possible. Early identification of subjects at risk is indispensable within this framework.Functional neuroimaging encompass non-invasive in vivo methods to map human brain function. Electrical, metabolic or haemodynamic parameters reflecting brain function are measured during rest and/or stimulation conditions. Functional MRI is based on imaging of the blood-oxygen-level-dependent (BOLD) contrast and is nowadays a widely used brain mapping method.We investigated possible contributions of functional magnetic resonance imaging in the early detection of Alzheimer's disease. We focussed on MCI patients who had beginning subjective memory complaints, corroborated by an informant, as target population because early detection in this group of patients should lead to more adequate and faster medical interventions. We constructed and validated in healthy young controls an experimental memory paradigm (face-name association) to elicit bilateral anterior hippocampal activation, suitable for application in a sample of MCI patients and used the most efficient experimental design to apply to a sample of MCI patients. Additionally, we characterized the DMN activity during rest in MCI, also with a special interest for hippocampal and lateral temporal involvement. The conclusion in the design comparison study held that the blocked design displayed higher t-values, suggesting greater detection power compared to the permuted design which had a better estimation efficiency and was able to capture more anterior-situated activation in the hippocampus during the associative encoding of face-name pairs. When detection power is of lesser importance, researchers should be encouraged to use event-related designs because of its superior estimation efficiency, its ability to reduce confounds arising from stimulus order predictability, and its flexibility in design. It appeared that the permuted design was more suitable for our own specific needs, mainly because of its ability to capture more anterior memory-related hippocampal activation. We explained the difference as an experimental design-related reason. The permuted design is more sensitive to the novelty manipulation compared to the block design, making it a cognitively different task with respect to anticipation and attention. The block design displays a high stimulus order predictability that can account for a reduced novelty effect compared to the semi-randomized permuted design. And because novel stimuli are believed to activate the anterior hippocampus, this is the most likely explanation of the regional differences in hippocampal activation.Next, we developed an extensive image processing and statistical analysis pipeline for resting-state fMRI data and cross-sectionally investigated DMN activity in the MCI sample, with a special interest for hippocampal and lateral temporal involvement. In the DMN study, we have shown that there are significant differences in DMN activity between healthy older adults and MCI patients, with decreased activity for patients, particularly in the retrosplenial/calcarine region, inferior parietal cortex, MTL, and inferior frontal cortex. Increased activity for patients was detected in the medial prefrontal cortex, angular cortex, and lateral temporal cortex. These results are consistent with previous findings directly comparing MCI patients to a healthy older control sample. Decreased activity in resting-state networks is directly related to the altered cognitive processing and defective networks accompanying early neurodegeneration. Our correlational analyses in the MCI group revealed positive correlations between reduced medial prefrontal activity and cognitive decline represented in DRS scores, and to a lesser extent, positive correlations between a memory recall test and left hippocampal activity. Lateral temporal involvement in the DMN was in both the older adults control sample and the patient group detected, and suggested a slightly increased activation, more right than left, in middle temporal areas in the MCI patients, compared with healthy older adults.In the face-name associative memory encoding study, we have shown that our MCI sample exhibited decreased hippocampal activation patterns, compared to healthy older controls, in absence of significant regional volume and task performance differences. Our findings are in accordance with the hypothesis that the degree of memory-related MTL activation in MCI is related to the level of clinical impairment along the transitional continuum between normal aging and dementia. However, our data indicated that our MCI patient sample already exhibited decreased hippocampal activation levels, situating them in the already declining part of the continuum.A major clinical implication from this research is that the combination of cognitive fMRI, resting state fMRI and neuropsychological data can serve as a tool to detect subtle activity differences between healthy older adults and MCI patients, especially in the regions that are struck first by neurodegenerative pathology.