Online communities have emerged as important organizational forms, but there are many gaps in our understanding. In particular, researchers have mainly focused on individual-level drivers of behaviors in communities, while downplaying (formal, informal) context at various levels. We theorize that different dimensions of context (i.e. omnibus and discrete context) influence decision-making in online communities through mechanisms involving community members’ attention. Specifically, context influences which problems members perceive and which solutions they retrieve and apply, thereby shaping the process of matching solutions and problems. We derive four hypotheses about contribution behaviors in online communities and how such behaviors are influenced by context. The empirical setting for our study is the open-source software community. We find support for our hypotheses in a unique dataset that captures the behavior of 24,057 community members who used the SourceForge.net online platform from 2000 to 2002.