Simple odor identification tests reveal reduced ability to identify odors in Alzheimer's disease (AD) and in populations at risk for AD because of the ApoE e4 allele, MCI, and a family history of AD. Neuropathology in olfactory areas in AD is well established in post-mortem tissue and in a few structural MRI studies. While impairment in olfaction is known to occur in AD and in those at risk for AD, its sensitivity and specificity, based on simple odor identification tests, has not been sufficient to render it a serious biomarker. Critical research is needed to test the sensitivity of impairment in the olfactory system to signal risk for AD. This will require careful consideration of the methods for testing olfactory system integrity that go beyond simple odor identification tests. Similarly, there is significant evidence to support a difference in retinal thinning between normally aging individuals AD; however whether this measure is predictive of development of MCI or AD in amyloid positive, cognitively normal individuals in unknown. There is a clear need to investigate whether assessment of multiple sensory modalities in conjunction with CSF and imaging markers will improve the detection of preclinical AD. This project aims to address this need using a multimodal approach. It will investigate whether olfactory or visual function or a combination better reflects the presence of CSF biomarkers or structural MRI markers of AD in preclinical AD. Building on a foundation of preliminary data, the project will investigate the underlying cortical structures and neural networks that serve odor recognition memory, remote odor memory and odor identification in individuals with CSF markers that reflect preclinical AD, using fMRI and functional connectivity. The project will develop a deeper understanding of the potential for sensory dysfunction to predict preclinical AD.