Experiments and simulations on day-to-day route choice-behaviour

The paper reports laboratory experiments on a day-to-day route choice game with two routes. Subjects had to choose between a main road M and a side road S. The capacity was greater for the main road. 18 subjects participated in each session. In equilibrium the number of subjects is 12 on M and 6 on S.<br>Two treatments with 6 sessions each were run at the Laboratory of Experimental Economics at Bonn University using RatImage. Feedback was given in treatment I only about own travel time and in treatment II on travel time for M and S. Money payoffs increase with decreasing time. The main results are as follows. <br>1. Mean numbers on M and S are very near to the equilibrium. <br>2. Fluctuations persist until the end of the sessions in both treatments. <br>3. Fluctuations are smaller under treatment II .The effect is small but significant. <br>4. The total number of changes is significantly greater in treatment I. <br>5. Subjects road changes and payoffs are negatively correlated in all sessions.<br>6. A direct response mode reacts with more changes for bad payoffs whereas a contrary response mode shows opposite reactions. Both response modes can be observed.<br> 7. The simulation of an extended payoff sum learning model closely fits the main results of the statistical evaluation of the data.