David Lazer: "Parallel Problem Solving: The social structure of exploration and exploitation"

Lecture to be held on Monday June 26, 2006, 16:15.
Room E 13, Seilergraben 49, ETH Zurich.

Whether as team members brainstorming, or cultures experimenting with new technologies, problem solvers communicate and share ideas. This paper examines (1) the phenomenon of “parallel problem solving”—a system where multiple agents are attempting to solve the same problem, and whose payoffs are not directly connected and (2) how the structure of these communication networks among those agents can affect system-level performance. We present an agent-based model of information sharing, where the less successful emulate the more successful. Results suggest that where agents are dealing with a complex problem, the more efficient the network at disseminating information, the better the short run and lower the long run performance of the system. The dynamic underlying this result is that an inefficient network maintains diversity in the system for a more extended period, and is thus better at exploration than an efficient network, supporting a more thorough search for solutions in the long run. The identified curvilinear relationship between connectivity and group performance can be seen in several diverse instances of organizational and social behavior.

David M. J. Lazer, Associate Professor of Public Policy, teaches courses on regulation and public management. Lazer has an overarching interest in the process by which connections emerge among actors and the consequences that the resultant network has for individuals and the system. He most recently completed a book on the use of DNA in the criminal justice system: DNA and the Criminal Justice System: The Technology of Justice. With the support of the NSF, he is also in the process of launching a Web-based forum on the use of DNA in the criminal justice system (www.dnapolicy.net). He has also coauthored a series of papers on the diffusion of information among interest groups and between interest groups and the government. He holds a PhD in political science from the University of Michigan.