The Explanatory Power of Models and Simulations: A Philosophical Exploration

Some computer simulations have explanatory goals. In a typical case, the computer simulation exemplifies a model generating a phenomenon under investigation. For the simulation to be explanatory, the model has to be explanatory. Models that fully explain phenomena are rare, but a model may offer a partial explanation of a phenomenon. It does this if an isomorphism holds between certain features of the model and certain features of a natural system realizing the phenomenon. The first two sections elaborate this account of a simulation's explanatory power. The third section illustrates the account by applying it to Brian Skyrms's (2004) simulations of the evolution of cooperation. The final section uses the account to suggest ways of increasing those simulations' explanatory power.