Proceedings of the Annual Meeting of the Cognitive Science Society

People spontaneously collaborate to solve a common goal.What factors affect whether teams are successful? Due tolack of large-scale naturalistic data and methods for investi-gating scientific questions on such data, previous work has ei-ther focused on very concrete cases, such as surveys of busi-ness teams, or abstract cases, such as GridWorld games, whereagents coordinate their movement so that each agent can get totheir own goal without obstructing other agents. We propose acomputational framework based on the multivariate Hawkesprocess and a novel algorithm for parameter estimation onlarge data sets. We demonstrate the potential of this methodby applying it to a large database of programming teams, pub-lic GitHub repositories. We analyze factors known to influenceteam performance, such as leader organization style and teamcognitive diversity, as well as other factors, such as the bursti-ness of effort, that are difficult to test using existing methods.Keywords: Collaborative cognition; Hawkes process; Organi-zational psychology; Bayesian nonparametrics