Experiment on Destination Choice Game

摘要

A fundamental issue in traffic science is to understand the behavior of traffic participants. Ether to theoretically model the traffic phenomenon or to analyze the complex operation process of the actual traffic system, the underlying mechanism of individual traffic decision-making behavior has always been the focus of relevant scholars. Many researches on traffic decision-making have been done, but the research on destination choice which is the most basic motivation for individual travel is still lacking. In this paper, we establish a simplified laboratory experiment to study individuals' destination choice behaviors. Considering that increasingly diverse traffic information is provided to residents through the Internet, two treatments are set up to explore the influence of feedback information on human behaviors. In a lot of real scenarios, the individual is most likely to take the degree of congestion into account, so we set the payoff of each destination to a form negatively related to that degree. Experimental results demonstrate that aggregate behavior would achieve user equilibrium rapidly no matter whether there is feedback information or not. Moreover, the feedback information has a certain effect on individual choice behaviors, which is a socially significant discovery. We believe that the results can be combined with other models for comprehensive modeling, which will be helpful to the traffic planning and management in the future.