Although much research into driver attitudes toward and responses to expressway-based variable message signs (VMSs) has been conducted, little research has explored motorist responses to VMSs on arterial sur- face streets. Arterial VMSs, located near main expressway entrance points, may prove to be more effective at inducing motorists to divert during congestion or incidents than their highway counterparts because they will allow drivers to have many more routing options before they commit to the expressway system. The results of a study conducted in Milwaukee, Wisconsin, to discern motorist attitudes toward arterial VMSs and the diversion behaviors induced by those signs are presented. A revealed-preference survey conducted in the immediate vicinity of the arterial VMS found that nearly two-thirds of the respondents obtained traffic information from the signs more than once per week. Further- more, 66% of those surveyed changed their route at least once per month because of the information received from the arterial VMS. These results demonstrate that motorists are responsive to messages on arterial VMSs; arterial VMSs have an important impact on motorist’s travel behavior and are thus a key component in a comprehensive traffic management system. More specifically, an ordered logit model revealed that the propensity to divert was correlated to the frequency that a driver encoun- ters an arterial VMS, motorists’ perception of the VMS information as useful, and motorists’ trust in the accuracy of the VMS information. The reasons for this diversion response are primarily to save travel time and to avoid expressway incidents. Variable message sign (VMS) systems are among the most widely used of all advanced traveler information systems (ATIS) technolo- gies; they have been installed on roadways around the globe. Indeed, such systems are widely regarded as a singularly effective means of managing transportation networks (1). Consequently, VMSs have emerged as a key component of intelligent transportation systems implementations over the past decade and a half. LITERATURE REVIEW Several studies on the potential effectiveness of VMSs in managing expressway traffic have been conducted. In the best cases, VMSs can enable a significant proportion of motorists to divert onto other routes to avoid congestion and the sites of incidents, and thereby alleviate highway network pressure and prevent the escalation of congestion and the elongation of delays. As noted in a widely cited study of VMS effectiveness in Paris, France, “VMSs alone can significantly affect vehicle diversion” (1). This is particularly true for the poten- tial effects of VMSs during peak travel times or during severe inci- dents. Under these conditions, even a relatively modest diversion rate can have large effects on traffic flow in the transportation network system. Surveys show that drivers like the idea of VMSs, whether they use them or not, and seem to want to be able to find such sys- tems useful in their particular circumstances. Most drivers make routing choices while under way (as opposed to before their trips), and drivers want to minimize the time spent in congestion (2). Many VMSs have been installed on limited-access expressway systems, usually at key route decision-making junctures (3). The placement of VMSs is essential to maximizing their effectiveness in giving motorists the information and the opportunities that they need to make efficient routing choices (4, 5). The information presented on VMSs, in other words, is not location neutral. Nevertheless, little research has been conducted to determine methods to discern the best placements for VMSs (5) and to differentiate the impacts of VMSs at different locations. Most studies of VMSs focus on signs located on limited-access expressway systems. Expressways, however, offer few possibilities for diversion, particularly in the event of an incident or high levels of congestion. Because motorists on arterial roads have many routing options before they enter the expressway system, it is possible that VMSs on arterial surface roads may be more effective in inducing diversions than VMSs on expressways. The present study addresses motorist attitudes toward VMSs placed on arterial roads, particularly those in the immediate vicinity of an expressway entry point, to determine whether drivers are dis- posed to make routing decisions on the basis of the information pre- sented on the VMSs. The results of this study are important for two reasons. First, studies have revealed that the location of a VMS is important for its effectiveness, particularly in relation to incidents or areas that are typically congested (6 ). Second, system costs are associated with diversion, particularly when many motorists divert onto the same route. As noted by McDonald et al. in a study of VMS effectiveness in Southampton, England, “the benefits of VMS are unevenly distributed. In general, those drivers who would pass a VMS on a route through [an] incident benefit, while those on diver- sionary routes disbenefit. In many situations, the net disbenefits are greater than the net benefits” (6 ). The system costs incurred when many drivers divert onto a single alternate route could be signifi- cantly diluted if motorists were made aware of incidents or conges- tion and never entered the expressway system, and thus spread the traffic load over multiple surface routes. Few studies focus on the locational issues surrounding the place- ment of VMSs (5), and only two studies that mention arterial mes- sage signs, either implicitly or explicitly, were found: Yim and Ygnace investigated diversion rates prompted by an arterial VMS in Paris (1), and Benson studied motorist attitudes toward VMSs in Motorist Response to Arterial Variable Message Signs Zhong-Ren Peng, Nathan Guequierre, and Joseph C. Blakeman Transportation Research Record: Journal of the Transportation Research Board, No. 1899, TRB, National Research Council, Washington, D.C., 2004, pp. 55–63. Department of Urban Planning, University of Wisconsin–Milwaukee, P.O. Box 413, Milwaukee, WI 53201-0413. 55