Switching Behavior: An Experimental Approach to Equilibrium Selection

The aim of this thesis is to investigate experimentally the reliability of the predictions of evolutionary game theory concerning equilibrium selection. Particularly, I analyze how an adjustment of the initial conditions, which were stated to be one of the essential factors in determining long-run stochastic equilibrium, may change the outcome of the game. The current work studies equilibrium selection in the framework of technology adoption in the presence of an established convention. It consists of three chapters. The first provides an extensive survey of theoretical and experimental literature on equilibrium selection, technology adoption and the emergence of conventions. The second chapter presents an experiment that investigates whether a new technology, represented by an introduction of either a risk-dominant or a payoff dominant strategy, is capable to break a conventional equilibrium and provoke the adoption of another one. In the third chapter I present an experiment that studies whether adding a dominated strategy to a coordination game facilitates transition from one equilibrium to another by changing their basins of attraction.