SUMMARY Elderly people frequently experience sleep disturbances, which contribute to age-related cognitive decline and are thought to be a core component of Alzheimer's disease (AD) and its pathophysiology. However, whether sleep disturbances cause cognitive impairment and neuropathology in AD remain unclear. Indeed human studies are unable to determine whether sleep disturbances precede or follow the development of AD pathology. Identifying the precise sequence of events linking sleep fragmentation and disease progression is a crucial step in better understanding the etiology of AD and identifying new therapeutic targets. Studies in animal models are needed to investigate this issue. Rodents are useful to identify basic mechanisms underlying the relationships between sleep and brain function, but also have limitations due to substantial differences from humans in sleep, cognitive and brain aging phenotypes. Using a more translational animal model with regards to sleep, cognitive function and neuropathology would likely provide critical new insights into the role of sleep disturbances in driving AD. The common marmoset (Callithrix jacchus) is ideally suited as such a model. This diurnal nonhuman primate exhibits monophasic sleep, shows age-related decline in several cognitive domains, and possesses two hallmarks of AD neuropathology, amyloid-β deposition and accumulation of hyperphosphorylated tau proteins. In addition, its short lifespan of about 10-12 years is ideal for longitudinal studies. Marmosets will be fitted with an actigraphy device to monitor sleep/wake patterns. The monkeys will be trained on a battery of cognitive tasks administered on touchscreens in their home cage. After baseline recording of sleep and cognitive performance, they will be randomly assigned to a sleep fragmentation (SF) or undisturbed sleep (control) group. The SF group will be chronically exposed to periods of disrupted sleep designed to mimic fragmented sleep in AD patients, whereas the control group will be kept undisturbed. Changes in physiology and cognitive function will be assessed throughout the experiment. The proposed study should validate the marmoset as a translational preclinical model for future studies focusing on the role of SF on AD neuropathology. The availability of such a model will be crucial for identifying preventative or therapeutic strategies that are clinically relevant.