Alzheimer ’s disease (AD) affects over 5 million Americans aged 65 years and older. Neighborhood environments that promote healthy lifestyles and exposures throughout the life course may help increase cognitive reserve and thereby prevent AD or delay its onset. Neighborhood “greenness” (density of healthy vegetation) is one such neighborhood characteristic that may increase physical activity, improve mental health, reduce air pollution exposure, and increase social engagement, factors previously associated with late life brain health. Compared to late life, the midlife may be a critical time for neighborhood environments to have a more substantial effect, during the period prior to the start of AD pathology. While limited, the extant literature has found that residing in greener neighborhoods in late life is associated with lower dementia risk, lower odds of cognitive impairment and cognitive decline, and less brain atrophy, but previous studies have not investigated neighborhood greenness from midlife in relation to late life brain health. Thus, this study aims to determine whether 1) objectively measured neighborhood greenness during midlife is positively (i.e., beneficially) associated among ≥55 year olds with measures of a) global cognition, processing speed, episodic memory and executive function (at baseline and longitudinally); and b) ventricular volume, total brain volume from AD regions of interest, and white matter hyperintensities; 2) neighborhood greenness is more strongly associated with the cognitive and MRI outcomes among women, non-Whites/Hispanics, and those of lower socioeconomic status (SES); and 3) neighborhood greenness is more strongly associated with the cognitive and MRI outcomes in neighborhoods with more racial/ethnic diversity, lower socioeconomic status, and more forest and park space. Geocoded address data (retrospectively collected back to 1980) for 4,200 participants from the longitudinal, population-based Multi-Ethnic Study of Atherosclerosis (MESA) will be used to create measures of average neighborhood greenness from ages 45 to 54. Greenness will be based on the normalized difference vegetation index (NDVI), a continuous measure calculated from satellite imagery in ArcGIS. Brain health will be assessed with cognitive (global cognition, processing speed, episodic memory, executive function) and brain imaging measures (ventricular volume, total brain volume in AD regions of interest, white matter hyperintensities). Adjusted random intercept linear regression will test associations between average neighborhood NDVI in midlife and the late life cognitive and brain imaging outcomes, controlling for demographics (e.g., age, race/ethnicity and SES), site, neighborhood SES, and population density. Imaging outcomes will be divided by intracranial volume prior to regression to account for differences in brain size. Effect modification by the individual and neighborhood characteristics (Aims 2 and 3) will be tested using in...