complex theory on social science.pdf

PS 793, Fall 2008 Professor Robert Axelrod Email: axe@ Office Hours::Tuesdays 2 – 4 at 4116 Weill Complexity Theory in the Social Sciences Complexity theory is a new interdisciplinary approach to understanding dynamic processes involving the interaction of many actors. A primary methodology of complexity theory is agent-based modeling. Agent-based modeling involves specifying how individual agents (such as people, nations, or organizations) interact with each other and with their environment. Computer simulation is then used to discover the emergent properties of the model, and thereby gain insights into dynamic processes that would be too difficult to model with standard mathematical techniques. Agent-based modeling provides a third way of doing science in addition to the traditional methods of deduction and induction. Like deductive models, an agent-based model starts with a well-defined set of assumptions. But unlike deductive models, an agent-based framework is capable of revealing consequences through simulation that cannot be deduced with standard mathematical techniques. And like induction, the main method of finding these consequences (and perhaps new insights) is through analysis of a set of data - in this case data generated by running the computer simulation. The goal is to discover new principles about the dynamics of complex systems, especially complex adaptive systems that are typical of social processes. There is no need to assume rationality. The course will consider a wide variety of applications of agent-based models to the social sciences, including residential segregation, revolution, social influence, urban growth, war, alliances, organizational change, elections, and stock markets. Among the issues to be examined across models are: path dependence, sensitivity to initial conditions, emergence of self-organized structure, adaptation to a changing environment, co-evolution, information