In this paper I take a dynamic point of view and give an example of a population game and of a locally improving stochastic learning process such that the quantal response equilibrium assigns to the different strategies the probabilities that are non-monotone in the payoffs. Moreover, if the...