Background The increasing incidence of Alzheimer’s disease (AD) and related sleep-disorders with the aging of the population is a major issue of public health. The prevalence of sleep disturbances is about 50% in dementia patients. Sleep in dementia is mainly characterized by prolonged night-time awakenings, sometimes with longer sleep latency and/or early awakening. Sleep architecture is modified in AD: slow wave sleep (SWS) tends to disappear and Rapid Eye movement (REM) sleep tends to be fragmented. Muscle activity during REM sleep and micro-architecture of sleep (increase in the number of micro-arousals, decrease of slow wave activity and number of spindles) are also affected. In addition, chronic insomnia in AD patients is often associated with specific sleep disorders, such as obstructive sleep apnea syndrome (OSAS), periodic limb movements (PLMS) and parasomnias. Sleep disorders are accompanied by sleepiness and behavioural syndromes (agitation and apathy). The rest-activity rhythm is deteriorated and sometimes reversed. The presence of these sleep disturbances is the main reason for the institutionalization of AD patients, as well as a major cause of the deterioration of the caregivers’ sleep. It is well known that sleep is involved in memory consolidation and that sleep disorders or sleep restriction impair cognitive performance. While sleep disturbances have been thoroughly documented in the dementia stages of AD, disturbances of sleep in the prodromal stages of the disease as well as their potential impact on cognitive and psycho-affective functioning are largely unknown. Therefore, the challenge is to better understand the sleep disturbances of prodromal AD and their potential contribution to the progressive cognitive deterioration characterizing AD. Hypotheses - Changes in sleep architecture and/or appearance of sleep disorders occur early in the course of AD including the prodromal stages. - These sleep-related changes affecting prodromal AD patients could contribute to cognitive and psycho-affective disturbances. Study design The MEMENTO cohort is composed of patients with isolated memory complaints or mild cognitive impairment, at high risk to develop dementia. All voluntary patients from the CMRR of Bordeaux (MEMENTO cohort) will be tested at inclusion and followed-up 1 year later. For these two evaluations, actigraphic monitoring at home, 2-night polysomnography (PSG) monitoring at hospital including a 24-hour period of urinary melatonin dosage, neuropsychological tests battery, a virtual reality tool and questionnaires will be used. Outcome measures Sleep measures - PSG recordings: Apnea/Hypopnea index (AHI), PLMS index Excessive transient muscle activity, Sustained muscle activity Sleep structure parameters (% stage 1, 2, 3 and REM, Delta activity, Spindle rate) Sleep duration parameters (TST) Sleep consolidation parameters (WASO, Sleep efficiency) Sleep propensity parameters (Sleep Onset Latency) - 24-hour melatonin quantity rate - Actimetry: inter-daily stability, intra-daily variability, rhythm amplitude Cognitive measures - Reaction times and % of errors in the neuropsychological tests (memory, speed processing, language, visuo-spatial skills, attentional functions, executive functions) - % of recognition and recall from the memory tests in a virtual environment Psycho-affective measures - Apathy scores Scientific expected advances - To improve knowledge on sleep and sleep-disorders in MCI and pre-demented patients. - To determine the association between sleep parameters, sleep disorders, polysomnographic patterns and severity of cognitive impairment and psycho-affective disorders in patients with MCI. - To identify potential sleep markers of the progression of cognitive decline. - To develop an algorithm for automatic detection of MCI and dementia markers in sleep EEG. - To provide a prognostic-aid tool for dementia.