Anthropogenic emissions of greenhouse gases (GHGs) over the past two centuries have resulted in global environmental change that is expected to strengthen dramatically in the future. GHG emissions are driven by dynamic interactions between physical, chemical, and biological components of the Earth system and human systems, and the resulting climate change alters the frequency, severity, and extent of extreme climate events (e.g., heat waves, drought, floods). Although not considered in Earth system models, perceived risk stemming from such extreme climate events may induce behavioral changes that alter GHG emissions. We linked the Climate Rapid Overview and Decision Support (C-ROADS) climate model to a social model of behavioral change to examine how interactions between perceived risk and emissions behavior influence projected climate change. Our coupled climate and social model resulted in a global temperature change ranging from 3.4–6.2°C by 2100 compared with 4.9°C for the C-ROADS model alone, and led to behavioral uncertainty that was of a similar magnitude to physical uncertainty (2.8°C versus 3.5°C). Model components with the largest influence on temperature were the functional form of response to extreme events, interaction of perceived behavioral control with perceived social norms, and behaviors leading to sustained emissions reductions. Our results suggest that policies emphasizing the appropriate attribution of extreme events to climate change and infrastructural mitigation may reduce climate change the most.