Dynamic Systems in Social Psychology

This summer, at the International Association for Conflict Management meetings in Budapest, I had the pleasure of sharing a panel on complex systems with Andrzej Nowak and Robin Vallacher. Andrzej is Professor of Psychology at the University of Warsaw, Director of the Center for Complex Systems at the Institute for Social Studies at the University of Warsaw, Associate Professor of Psychology at Florida Atlantic University, and Professor of Psychology at the Warsaw School of Social Psychology. Robin is Professor of Psychology at Florida Atlantic University. Together, these guys are the fathers of dynamical systems theory in social psychology.

"The nonlinear dynamical approach provides an innovative way of understanding the real, complicated systems at the very heart of social psychology." Abraham Tesser, University of Georgia.

Andrzej Nowak and Robin R. Vallacher, Dynamical Social Psychology (1998).



In a collaboration with Peter Coleman of the International Center for Cooperation and Conflict Resolution at Teachers College, Columbia University, Andrzej and Robin have founded the new International Center for Complexity and Conflict at the Warsaw School for Social Psychology, where they are currently working on the application of complex systems theory to intractable conflict. Some excellent foundational material for their work can be found here.




Robin R. Vallacher and Andrzej Nowak, Dynamical Systems in Social Psychology (1994).



What sets these guys apart from alot of the work in dynamical systems and social science is their finely hewned expertise as experimentalists. Much of their work represents the iterative development between simulation, theory building, and laboratory experimentation that is all too rare.

The Emergence of Costly Punishment

Christoph Hauert, at the Program for Evolutionary Dynamics at Harvard, and colleagues have a new paper which addresses the "second order social dilemma" that arises from cooperators who may free-ride by failing to engage in the costly punishment of defectors in public goods games. Here, they consider four strategies, 1) nonparticipation, 2) participators who defect, 3) participators who cooperate, but fail to punish defectors, and 4) participators who cooperate and engage in costly punishment of defectors.

Noting that cooperative behavior in many joint enterprises that can be characterized as public goods games is enforced through institutions that impose sanctions on defectors, and that in the absence of such institutions, individuals are generally willing to engage in punishment of defectors, even when costly, they ask how this costly punishing behavior may have become established in the first place. It turns out that the ability to choose not to participate is key. Their findings suggest that circumstances where the engagement in the joint enterprise is voluntary are more likely to promote cooperation than those where participation is mandatory. The paper is available here.


Christoph Hauert, Arne Traulsen, Hannelore Brandt, Martin A. Nowak, and Karl Sigmund, Via Freedom to Coercion: The Emergence of Costly Punishment, 316 Science 1905-1907 (2007).

VirtualLabs in Evolutionary Game Theory

If you enjoyed C-MOL (see below), then check out Christoph Hauert's VirtualLabs in Evolutionary Game Theory. You'll find similar applets for experimentation, tutorials along with research papers associated with each of the models, and some excellent introductory material.

Computational Models of Life

Check out the Center for Models of Life (C-MOL), a center under the Danish National Research Foundation, located at the Niels Bohr Institute at the University of Copenhagen. Researchers at the Center use methods from physics to develop models dealing with computation and communication in biological systems. Particularly fun is their page of interactive Java applets which explore a number of applications of network theory and dynamical systems theory. Enjoy!

Resolving Community Structure in Complex Networks

Our friend and colleague, Carl Bergstrom, a theoretical and evolutionary biologist at the University of Washington, was recently appointed to the external faculty of the Santa Fe Institute. Congratulations, Carl!

In some recent work with Martin Rosvall, Carl describes an ingenious method for identifying community structure that can help to understand larger-scale social, biological, and technological networks. The method employs information theory to find an efficient compression of a given network's topology. This work is an outstanding example of the important contributions that can emerge from thoughtful interdisciplinary research. The paper is available here.

Martin Rosvall and Carl T. Bergstrom, An information-theoretic framework for resolving community structure in complex networks, 104 Proceedings of the National Academy of Sciences USA 7327-7331 (2007).

The Evolution of Cooperation among Other-Regarding Agents

Sam Bowles and Herb Gintis have a new review of game theoretic models of cooperation which is to appear in The New Palgrave Dictionary of Economics, to be published next year. Bowles and Gintis begin with dyadic interactions among self-regarding agents showing that cooperation can be maintained with such mechanisms as retribution, where cooperation is withheld from previously defecting agents (tit-for-tat) and reputation maintenance, where "mental models" of previous cooperation behavior are important to overcome the problem of infrequent interactions with a large number of partners. Among others, the standing model of Sugden, the indirect reciprocity model of Panchanathan and Boyd and the image scoring model of Nowak and Sigmund are considered.

Next, they examine the Folk Theorem in large groups of self-regarding agents operating under imperfect information, observing that the intuitions of the dyadic model do not extend to contexts involving larger group size and higher error rates. Finally, they explore cooperation in large groups of other-regarding agents, asking "how such altruistic behavior could have become common, given that bearing costs to support the benefits of others reduces payoffs, and both cultural and genetic updating of behaviors is likely to favor traits with higher payoffs." Citing a number of recent studies, they conclude that cooperation can be sustained in groups of substantial size where agents with social preferences are more likely than random to interact with other agents with social preferences (an idea akin to Brian Skyrms' notion of correlated association). They further conclude that the long term evolution of such social preferences is plausible.

Newcomers to game theoretic models of cooperation would do well to start with this excellent review. An advance version of the paper is available here.

Samuel Bowles and Herbert Gintis, Cooperation, in the The New Palgrave Dictionary of Economics, Eds. L. Blume and S. Durlauf. MacMillan (forthcoming, 2008).

For a look at Brian Skyrms work on correlated association, see

Brian Skyrms, Evolution of the Social Contract (1996).

Brian Skyrms, The Stag Hunt and the Evolution of Social Structure (2003).

Marketing in Heterozygous Advantage

Ray Hagtvedt and I have a new paper, Marketing in Heterozygous Advantage, forthcoming in the Journal of Business Ethics, in which we examine coordination problems and perverse incentives surrounding the development of markets in a particular form of human genetic engineering in which heterozygotes are fitter than both homozygotes, a condition known as heterozygous advantage. We first present a generalized model of the condition, illuminated by the application to sickle-cell anemia. Next, we propose a typology of related markets, some of which are currently functioning with available products and services, and others that are widely viewed as imminent. We suggest the manner in which perverse incentives may arise for firms that market genetic intervention in circumstances where heterozygous advantage is possible. Finally, we propose that this misalignment of incentives with social welfare has arisen from both ill-conceived market intervention where markets are capable of achieving efficient outcomes and the lack of market intervention where markets have failed. We offer specific legal and regulatory approaches for reform. An advance version of the paper is available here.

Gregory Todd Jones & Reidar Hagtvedt, Marketing in Heterozygous Advantage, The Journal of Business Ethics (forthcoming, 2007).

Simple Games in a Complex World

Randy Picker, at the University of Chicago Law School, in his work investigating the adoption of norms and the relationship of these norms to institutions like the law employed agent-based models as a means of theoretical development. To date, his work remains unique among the law reviews for the manner in which his research was distributed - the article was published with color diagrams and a CD that contained videos of typical simulation runs. The paper, along with the CD contents, can now be found here. It would be nice to see some more of this type of scholarly work. The ubiquitousness of the Internet makes this even easier today. Interested law reviews might look at journals like the Journal of Artificial Societies and Social Simulation for models.

Randal Picker, Simple Games in a Complex World: A Generative Approach to the Adoption of Norms, 64 U. Chi. L. Rev. 1225 (1997). (paper and CD-ROM publication).

For a couple of important theoretical foundations for Picker's work, and work on social norms more generally, see

R. H. Coase, The Firm, the Market, and the Law (1990) (with a reprint of The Problem of Social Cost).

Robert Ellickson, Order without Law: How Neighbors Settle Disputes (1991).

Programming Collective Intelligence

Toby Segaran, Programming Collective Intelligence: Building Smart Web 2.0 Applications (2007).

Gentlemen, start your engines. I guess describing Programming Collective Intelligence as a real page turner will give away the extent to which I am truly a geek's geek, but this is the first book in a while that I literally could not put down until I reached the end. Segaran has created an outstanding recipe book for collective intelligence algorithms that spans the gamit from Pearson's correlation and Euclidean distance measures to optimization, Bayesian filtering, neural networks, genetic algorithms, and support vector machines. Non-negative matrix factorization is thrown in to keep most of us humble. What makes the book unique is that each algorithm is described intuitively and mathematically AND implemented in practical Python applications. While a fair amount of Python sophistication is assumed, programmers of Java or C++ will have little trouble picking it up. As I read, application ideas kept popping into my head. Now its time to sit down and work through the book carefully, implementing the provided code and trying out some of my own ideas. Bravo.

Can Interdisciplinarity Really Exist?

My brother, Ronald Jones, Professor of Interdisciplinary Studies at Konstfack in Stockholm, Sweden forwards Chunglin Kwa's excellent review of Peter Weingart and Nico Stehr's edited volume Practising Interdisciplinarity at The European Association for the Study of Science and Technology. Kwa's discussion, even beyond the review, provides a nice compact series of talking points for organizing discussions about the future of interdisciplinary work in academia. http://www.easst.net/review/march2002/kwa