**5. Conclusions**

This study empirically investigated the effect of dynamic travel time information on day-to-day commuter route choice behavior by designing and running a real world experiment on 20 participants over 11 weekdays for 5 O-D pairs considering two routes for each O-D pair. Consequently, the test entailed a total of 1100 route choice decisions (20 × 11 × 5). The experiment confirms some of the results obtained from previous simulation studies, demonstrating that, in general, real-time information significantly enhances behavioral rationality especially when drivers lack long-term experience. Simultaneously, inertial choice rates decrease with information provision, demonstrating that drivers are more willing to risk switching to faster routes when they have more information about these routes. Nonetheless, the positive role of information is, to a large extent, dependent upon the individual's age, preferences, and route characteristics. The results demonstrate that travel time information may not have positive impacts on driver route choice behavior if they value other factors in making their decisions, such as route scenery, habit, number of intersections and traffic signals. The results also reveal that the effect of information on driver behavior is less evident for elder drivers, which is consistent with [17]. In addition to personal traits, route characteristics are found to be another important factor influencing the effectiveness of information. Specifically, information may not add value if one route is significantly better than the other given that drivers would be able to identify the optimum route on their own through their experiences.

The effect of the type of route information provided to the travelers on their route choice behavior was also studied. The conclusions are consistent with the results of simulation studies, demonstrating that, when drivers have limited experiences, information on expected travel times is in general more effective than information on travel time variability in enhancing rational behavior. After drivers gain sufficient knowledge of the alternative routes, however, the benefit of providing strict information appears to diminish. The results also demonstrate that drivers prefer to take the faster less reliable route as opposed to the slower more reliable route when they lack historical experience. However, as drivers accumulate experience, they become more willing to take the more reliable route, demonstrating that they become less risk seeking in the gain domain at higher uncertainty once experience is gained. In addition, the effect of information types on route choice behavior significantly differs from person to person. Which type of information is most effective to what group of travelers remains to be investigated in future research.

The experiment also demonstrates that, regardless of being informed or not being informed, the drivers' inertial behavior does not reduce in day-to-day variation, which is different from the results obtained in an earlier simulation study [16]. This may be attributed to habitual behavior or the fact that more decision considerations are accounted for in actual driving.

Finally, it should be noted that given the small participant sample size, these conclusions serve as a first attempt at understanding driver route choice behavior empirically. Further research is needed to validate these findings on a bigger sample of drivers and for different confounding factors.

**Author Contributions:** The authors confirm contribution to the paper as follows; study conception and design: H.R. and J.W.; data collection: J.W.; analysis and interpretation of results: J.W. and H.R.; draft manuscript preparation: J.W. and H.R. All authors reviewed the results and approved the final version of the manuscript.

**Funding:** This effort was jointly funded by the Mid-Atlantic University Transportation Center (MAUTC), the University Mobility and Equity Center (UMEC), and through the Office of Energy Efficiency and Renewable Energy (EERE), Vehicle Technologies Office, Energy Efficient Mobility Systems Program under award number DE-EE0008209.

**Acknowledgments:** The authors acknowledge all personnel who assisted with the data collection.

**Conflicts of Interest:** The authors declare no conflict of interest. The funders had no role in the design of the study; in the collection, analyses, or interpretation of data; in the writing of the manuscript, or in the decision to publish the results.
