Background: Normal aging is associated with a decline in cognitive abilities, particularly in the domains of psychomotor speed and executive functioning. However, ‘aging,' per se, is not a cause of cognitive decline but rather a variable that likely captures multiple accumulating biological changes over time that collectively affect mental abilities. Recent work has focused on the role of cerebrovascular disease as one of the biological changes. In the current study, we examined whether lobar microbleeds - magnetic resonance imaging (MRI) signal voids due to hemosiderin deposits secondary to cerebral amyloid angiopathy - are associated with cognitive decline in normal aging. Previous studies that reported a relationship between the presence of lobar microbleeds and decreased cognitive abilities have been primarily cross-sectional. Here, we used a retrospective longitudinal design to examine whether the presence of lobar microbleeds is associated with the rate of cognitive decline among non-demented older adults. Methods: Participants came from an ongoing longitudinal community-based aging study, in which subjects are evaluated at 18-24 months intervals and received a full medical, neurological, and neuropsychological examination at each of the follow-up visits. Gradient echo MRI scans were available on 197 non-demented participants (mean age: 84.15 ± 5.02 years). Microbleeds were rated visually on axial view and divided into subcortical (basal ganglia, cerebellum) and lobar (frontal, temporal, parietal, occipital lobe) regions, and confirmed with coronal and sagittal views to exclude artifacts. Cognition was assessed with a neuropsychological battery, providing summary scores for memory, language, executive, and visuospatial abilities. Using general estimating equations (GEE), we compared cognition cross-sectionally between individuals with 2 or more (n = 11) and fewer than 2 (n = 186) lobar microbleeds and examined longitudinal cognitive change beginning 9.47 ± 3.13 years before the MRI scan. Results: Subjects with 2 or more lobar microbleeds had worse executive functioning at the visit closest to the MRI scan (β = -0.044; p < 0.001)and had a faster decline in executive function over time (β = -0.072; p = 0.012) than subjects with fewer than 2 lobar microbleeds. The two groups were similar in age at scan date, education, ethnicity, sex distribution, and cognitive performance at first visit. Conclusions: Lobar microbleeds, a marker of cerebral amyloid angiopathy, are associated with an accelerated rate of executive function decline. The presence of cerebral amyloid angiopathy may be an important source of cognitive decline in aging. Future work should examine how cerebral amyloid angiopathy interacts with neurodegenerative processes, such as Alzheimer's disease.