COMPUTATIONAL SOCIAL SCIENCE

Department of Computational Social Science Seminar Abstract

Friday, October 14 - 3:00 p.m.
Center for Social Complexity Suite
Third Floor, Research Hall

Title: Against the Nash Program: Aggregate Steady-States Without Agent-Level Equilibria

Speaker: Rob Axtell, Chair, CSS

Abstract: In the application of game theoretic ideas to the social sciences it is the norm to 'solve' for agent-level fixed-points (Nash equilibria, refinements, minimax solutions, and so on). However, social systems are inherently multi-level systems, and stationarity on the macro-level does not imply (static) equilibrium at the agent level. Rather, aggregate regularities can be produced by constant flux and adaptation at the agent level if it occurs in the right proportions. Comparisons across levels risks the dual fallacies of division and composition. In this talk I will illustrate these ideas with agent-based models in which perpetual coevolution and on-going strategic behavior is observed at the micro-level while quasi-steady-states obtain at the macro-level.