Engineering Serendipity: When Does Knowledge Sharing Lead to Knowledge Production?

We investigate how knowledge similarity between two individuals is systematically related to the likelihood that a serendipitous encounter results in knowledge production. We conduct a natural field experiment at a medical research symposium, where we exogenously varied opportunities for face-to-face encounters among 15,817 scientist-pairs. Our data include direct observations of interaction patterns collected using sociometric badges, and detailed, longitudinal data on the scientists’ post-symposium publication records over six years. We find that interacting scientists acquire and create more knowledge when they share some overlapping research interests, but are less likely to cite each other’s work when they are from similar fields. Our findings reveal both collaborative and competitive effects of knowledge similarity on knowledge production outcomes.