It is often assumed that habits constitute an important component in human behavior. However, since the beginning of the century many sociologists have overlooked their role in explaining behavior. In this article we are testing empirically the completeness of an economic model made by Stigler and Becker to explain the effect of habit on behavior in a new context. According to the model habits are economically efficient in many cases. However, when an individual is faced with a permanent change in the environment, behavior is not determined by habits, but by the amount invested in information on the best options. We collected data from a field experiment on travel-mode choice with an intervention program to change the behavior of people moving to a new town. The intervention included information on available public transportation. Habit was measured by the subject's judgments of the likelihood that using public transport will take place in different kinds of situations. Theory was confirmed by the empirical data: there was no effect of habits on behavior in the new context. People in the intervention group used public transportation more often. However, search for information prior to the move had no effect. Some socio-economic variables had an additional effect on behavior in contrast to theory. People with a car at their disposal and those with higher education used the car more often.