I will describe behavioral experiments in which human subjects must create a network over which they will simultaneously play a coordination game, and compare the results to a setting in which the network was specified exogenously. Perhaps surprisingly, the networks created by the subjects seem fundamentally difficult for solving the coordination task.