The Dark Side of the Force: When computer simulations lead us astray and "model think" narrows our imagination. (revised version, October 2006)

Eckhart Arnold

1 Introduction
2 Different aims of computer simulations in science
3 Criteria for explanatory simulations
4 Examples of Failure: Axelrod style simulations of the “evolution of cooperation”
    4.1 Typical features of Axelrod style simulations
    4.2 How Axelrod style simulations work
    4.3 The explanatory irrelevance of Axelrod style simulations in social sciences
    4.4 Do Axelrod style simulations do any better in biology?
5 Conclusion
Bibliography

4 Examples of Failure: Axelrod style simulations of the “evolution of cooperation”

In what follows I will to discuss a few examples of computer simulations that were designed by its authors to explain certain empirical phenomena but ultimately fail to do so. What I want to show is that these failures result from the violation of one or more the three criteria for explanatory simulations explained before.

The examples that I have chosen to discuss are computer simulations of the “evolution of cooperation” as they have become popular after the publication of Robert Axelrod's famous book with the same title.

Admittedly, these examples are examples of bad simulations. But this makes them good examples. Because the failures are just the more obvious in these examples they help us understand what to avoid.

t g+ f @