Figure 3. The evolution of strategies in the alternating Prisoner's Dilemma.
In all panels, the simulations start from a randomly chosensome random point in the -simplex. In the cases of
,
, and
, the evolutionary trajectories converge to a cooperative alliance of many strategies dominated by the strategy Forgiver. In the case of
, the evolutionary trajectories converge to a mixed equilibrium of Forgiver, TFT, and Grim. The error rate
is set to
%, the number of rounds per game is
, and the mutation rate is
.