Significant advances in the prevention and treatment of Alzheimer’s disease (AD) and related dementias in older populations have been slow to emerge. In response, researchers have turned to an examination of social-environmental factors to identify candidates for clinical intervention. Recently, social engagement (variously operationalized) has been linked to a shorter period of cognitive decline, greater independence, and higher quality of life in people with AD. However, the social and neurological mechanisms of this relationship are poorly understood. In the current study, we examine associations between personal social network characteristics, biomarkers of underlying neurodegeneration, and cognitive function using a sample of 110 older adults at high risk for AD. These participants are members of an NIH-funded cohort study of AD neuropathology, and were administered a social network protocol during their annual study visit. We find broad support for associations between network measures of bridging capital – that is, low density, larger networks, presence of weak ties, and greater number of friends relative to kin – and higher global cognitive function. Moreover, among older adults with high levels of bridging social capital, the expected adverse consequences for cognitive symptomatology of brain atrophy, or neuropathology, in regions implicated in early stage AD are nearly completely attenuated. These findings are broadly consistent with the cognitive reserve hypothesis of AD resilience, wherein access to cognitive stimuli works to establish reserve or promote greater neuroplasticity to counterbalance neurodegeneration.