Traveler decision making under conditions of knowledge limitations and information provision

This paper presents a discrete-choice based model of traveler response to information. It contributes to existing approaches in three ways: firstly, instead of focusing on either the acquisition of information or the effect of received information on travel choices, this model describes the full sequence of possibly multiple information acquisitions, followed by a travel choice. In doing so, the model provides a more accurate account of actual travel behavior under conditions of knowledge limitations and information provision. Secondly, instead of modeling the utility of information acquisition in an ad-hoc fashion (e.g. as a linear function of attributes of the information service and the trip context), we conceive the utility of information acquisition in terms of the anticipated utility of the travel set after having received the information. Thirdly, instead of focusing on a specific type of knowledge limitation and a specific type of information, we describe response to a variety of information types, resulting from a variety of knowledge limitations. This model of travel information response is then estimated on data collected in a multimodal travel simulator-experiment with travel information provision. Estimation results provide face validity to the proposed modeling approach, as a substantial share of the variation in choices for a variety of information options and travel alternatives appears to be captured by the model. Furthermore, a number of new insights are gained with respect to the role of travel information in multimodal travel choice making. For example, it is found that travelers hold intrinsic preferences of some information types (i.e. the generation of new travel alternatives) over others (e.g. the assessment of already known alternatives) that are beyond straight economic explanations.