Effects of the different choice rules on the simulations.

The different choice rules produced differing median asymmetry values, and also differed in the relationship between forager number and asymmetry and the spread of asymmetry values produced. We determined the greatest difference between asymmetry and the null by comparing the highest observed asymmetry value with the corresponding value in the null model. All relationships between asymmetry and forager number marked with a star are significant (Linear Regressions, all P<0.0001), and all differences between the highest asymmetry and the null were significant (Wilcoxon Rank-Sum tests, all P<0.0001).