Figure 2.

Simulation results. (a) Proportion of plurals marked using m₁. Each line shows an individual chain, for 20 simulation runs. In single-person chains, we see greater divergence between chains, with some chains converging on always or seldom using m₁; by contrast, in multiple-person chains, particularly for larger S, the initial level of variability is retained longer. (b) Entropy of plural marking, averaged over 100 runs, error bars indicate 95% CIs. This overall measure of variability shows the trend visible in (a) more clearly: while variability is gradually lost over generations in all conditions (as indicated by reducing H(Marker)), this loss of variability is slower in multiple-person chains, particularly with larger S. (c) Conditional entropy of plural marking given the noun being marked, averaged over the same 100 runs. While we reliably see the emergence of conditioned variation in single-person chains, as indicated by reducing H(Marker|Noun), this process is slowed in multiple-person chains, particularly for larger S.