Congestion on road networks causes severe problems in and around large cities. Consequences of congestion include an increase in travel time and travel costs, environmental costs, economic costs, increased energy use and decreased economic growth, reduced travel time reliability, and reduced quality of life. Therefore, mitigation of congestion is deemed necessary. Within the Dutch program “Verkeer en Vervoer” (Traffic and Transport) funded by Connekt/NWO, strategies to mitigate congestion are examined based on a combination of behavioural change, efficient use of existing infrastructure, and efficient land use. In this thesis the focus is on the provision of travel information that leads to a more efficient use of the existing infrastructure. Information should be provided to the traveller about which route and which travel mode is optimal for a particular travel network and subsequently for the traveller in that specific situation. More specifically, this thesis focuses on research into a behavioural change as a result of travel information. This change may include a change in route or a change in travel mode. In 2002 during the start of the project that resulted in this thesis in-car travel information was not widely available. At that time the state of the art in travel information was a roadside Dynamic Route Information Panel (DRIP) that showed travel times for travelling by car. Some DRIPs also showed travel times for public transport. Around that time experiments were conducted on the use of Graphical Route Information Panels (GRIP). These panels provided dynamic traffic information graphically and offered the opportunity to display a static simplified map of the road network, including the position of traffic congestion that was shown on a part of the travel network. Short after the next-generation of GRIPs, the Full Colour Information Panel (FCIP) was introduced. On an FCIP the presentation of the map of the road network can be adapted at any time. It allows a display of any kind of text, graphical representation and combinations thereof. These features offer the opportunity to display route maps, text combined with symbols, the position of traffic congestion (colour coded according to severity), or any other representation the designer may wish. In this way, complex information can be provided in a way that makes it easier to process, understand, and remember. While an FCIP has more graphical possibilities than other road-side information devices, there is a risk that more information may be provided than drivers can actually process. The designers of travel information must tread a fine line between providing as much useful information as possible while minimising the risk of information overload. Information overload may result in safety issues and impairment of information processing, particularly, in situations that impose considerable demands on driver's attention, such as high-density traffic. Therefore, it was examined how to effectively design complex dynamic travel information so that it can be optimally used by travellers. Particularly, the effects of designs that combine a route map with colours to indicate the position and severity of congestion are explored. The focus is on the design of the information itself, and also on how travellers interpret and deal with the information that is provided. Method In this thesis, a working model was developed that helps explain why most information panel designs that are currently in use are generally not very effective at encouraging drivers to change their travel plans, while simultaneously indicating what is needed to improve the effect of travel information on behavioural change. The working model is built around the information processing theory of McGuire (1968). This theory indicates that changing attitude and behaviour through information provision is only possible if all of six information processing steps are completed successfully. These steps include: presentation of the information, attention to the information, comprehension of the arguments and of the position the information advocates, yielding to the message contents and its conclusions, retention of the changed attitude, and behaviour based on the changed attitude. This model was extended under the assumption that ergonomic design and social psychological factors in the decision making process are related to information processing. Cognitive and ergonomic guidelines were reviewed that facilitate travellers to read and comprehend the information provided. From a social psychology point of view reasons were reviewed why - in spite of the provision of well-designed information - travellers may or may not take a given advice on board, and why travellers sometimes choose a suboptimal travel option. An improvement of the information by using better ergonomic designs and by taking into account the role of social psychological principles may increase the probability that each step in McGuire’s model is completed successfully. As a result, the effectiveness of traveller information on route and mode choice may be enhanced. Therefore, a basic design of travel information was developed that took into account the effect of uncertainty in route choice. A basic design of a road map was used where both routes end at the same position on the sign. Consequently, the road map had a “closed form” showing that two routes lead to the same location. Within this design experiments were conducted on: how a route map should be displayed how colours should be used to indicate location and severity of congestion how travel times are interpreted what amount of information can safely and effectively be provided. The chosen research method involved an initial stage in which preliminary versions of traffic information designs were tested in a laboratory setting. In subsequent experiments, continually improved designs were tested in settings that more closely approached reality. When an FCIP that was able to provide reliable travel time information became available, a final experiment was conducted in a real-life setting. The aim of this experiment was to determine the extent to which results from the laboratory studies could be applied to the design of the FCIP, thereby validating its usability. Results Displaying a road map In Chapter 3 we reported research that examined the possibilities of displaying a route map. Designs were used that could be presented on several types of devices. The focus was on the use of information on a Full Colour Information Panel (FCIP). Preferences of travellers concerning the shape of the route map were examined by means of questionnaires. Route maps shaped as is in reality were compared with more abstract representations of the route. Furthermore, the effect of additional information on route choice was examined, such as the type of road or travel time. Notwithstanding the fact that there was no consensus among participants concerning preferences for the shape of a route map it was showed that travel information appearance has an effect on route choice. Even more, evidence was found that these effects were different concerning the presentation of the road map and the kind of additional information. Furthermore, it was shown that simple designs were preferred over designs that were more sophisticated and showed more (ir)relevant information. It may be concluded that it is important to exclude information that is not directly relevant for a route or mode choice but might have an effect on route choice. The use of colour to indicate location and severity of congestion Although the use of colour to indicate congestion has benefits concerning route choice there is a concern that the use of colour could increase the amount of information and, as a result, this could increase information processing time for the road user. The use of colour is seen as redundant information as travel time itself also indicates to some extent the severity of congestion. In Chapter 4, we reported two studies in which we examined the effect of the use of colour to indicate congestion on information processing times and preferences for a method to display congestion. In a lab experiment, the effect of adding colours to a road map to indicate congestion was measured. Participants were shown graphical route information and had to indicate as quickly as possible which of the routes had the shortest travel time. We found that additional traffic signs that did not indicate travel time (for example a motorway sign) led to a significantly higher information processing time. Furthermore, additional colours to indicate congestion did not lead to significantly longer reaction times. Therefore, it seems likely that additional colours have no negative effect on information processing time. We found that experts in Human Technology Interaction advise against the use of colour, whereas regular road users indicate that the use of colour was an improvement regarding information provision on congestion. We advice to use colours to indicate location and severity of congestion. The use of colour helps indicate the location and severity of congestion without the risk of extra information processing time and it also does not increase the risk of information overload. For colour blind travellers, travel times or other indicators of congestion will still be available. Our experiments show that there is no reason not to include information concerning congestion on route information, as there are no disadvantages of displaying information about congestion it easily justifies the advantage of extra traffic information. Interpretation of travel times To make clear to travellers what the presented travel time for public transport means is still an unresolved issue. Most travellers indicate that the travel time on an information panel indicates the travel time for the public transport option from station to station. However, in most situations the travel time for public transport is shown to enable a comparison with the travel time by car. Therefore, the travel time for public transport indicates the travel time from the current position to the destination station shown on the information panel. In Chapter 5 research is reported that indicates that 36% of the participants interpreted the meaning of travel time incorrectly in our study. Although this percentage is too high, it is positive compared to other studies where more than 60% interprets the travel time incorrectly. The working model introduced in Chapter 2 can provide an explanation for this finding. That is, travellers who are uncertain of the meaning of the travel time are unlikely to change their behaviour. The information should clearly indicate for which trajectory the information is applicable, whether the travel time is actual or dynamic or static. More research is needed on this issue. It is one of the most important improvements that are needed to increase the effect of multi modal travel information on travel mode choice. As long as travellers are uncertain about the meaning of the displayed travel time for public transport options, they are unlikely to choose public transport. Moreover, travellers that are certain about the meaning of the travel time but mistakenly think that it indicates the travel time from station to station will add travel time for travelling to the station. The comparison between travelling by car and travelling by public transport will in that case almost always be in favour of travelling by car. Amount of information A paradox in the presentation of travel information is that complete information will lead to a higher probability of changing travel plans. However, being complete in information provision includes adding information that may result in information overload. As a result, information is missed (because a driver has limited time to consult information) or information is ignored. The result is a lower probability of drivers changing travel plans. In Chapter 5 the effect of the amount of information provided on information processing was examined. In a traffic simulator we showed information on panels that helped to make a route choice or mode choice. We added information that we thought that travellers need to reduce uncertainty about the meaning of travel information and that would increase the attitude towards Park and Ride use. The results of this experiment suggest that the design that contained the most information elements was almost ignored completely. Probably the respondents decided that it would be impossible to read all information in the short time they had available and therefore they completely ignored the information. Our results and results of other studies imply that the amount of information that will be processed effectively is limited. Therefore, it is crucial to decide what information will be provided at what location, by means of which type of information device and at what moment. Route choice: travellers do not change plans during their trip In Chapter 6 a field experiment was reported that was held at a public event in Scheveningen (a fireworks show). This study showed that road users hardly change their travel mode during their trip. Visitors of the festival that parked on the Park and Ride facility were asked about their intention in the pre-trip stage. They either made this decision at home or didn’t make a decision at all concerning the trip towards Scheveningen. Travellers that had a travel plan hardly changed their route. This result means that: Travellers hardly change their travel plan during the on-trip stage. Only travelers without a travel plan may change to the Park and Ride location. Decisions concerning the use of Park and Ride are made during the pre-trip phase The road-side information concerning the Park and Ride facility is not used to decide to use the Park and Ride. This decision is made pre-trip based on other information (for example a website with information concerning visiting the event). Studies in Chapter 5 indicate that in-car information is more effective at encouraging change of travel mode to a park and Ride facility compared to road side information. The possibility to consult the information more than once, or more importantly, to read the information within a larger time-frame, may result in a deeper elaboration of the information. Consequently, travellers may choose Park and Ride more often. Furthermore, travellers relying on their in-car information system generally do not make travel plans. They follow the instructions of their in-car system. Effects of travel information on traffic safety In chapter 5, we presented the results of studies on the effect of different types of in-car travel information on route choice and traffic safety. In the first driving simulator study, three systems for in-car information were compared with respect to route and mode choice and traffic safety. A design with auditory information, a PDA with pictograms and a mobile phone with text messages were compared against each other. It was found that text messages on a mobile phone had a stronger negative influence on traffic safety. The disadvantage of text messaging is that the information does not fit on the screen and that the use of pictograms is not possible. As a result, the driver has to switch screens while driving. Smartphones currently in use in theory offer many functions that may help provide safe and well designed travel information. In a second driver simulator study reported in Chapter 5 in-car information on a PDA was compared to road side information, both containing the same information elements. No negative effects on traffic safety were found for those who used the PDA compared to in-car information. Mental effort to consult the information was equal for in-car and road-side information. Furthermore, travellers consulting in-car information decreased their speed more than travellers that consulted road-side information. Fixations to information should not last more than two seconds per fixation with a maximum of four fixations. At least the maximum total amount of fixations was violated for all conditions, indicating that the total amount of information on the designs is still too much to be able to process during traffic conditions. More research is needed to decrease the amount of information. Guidelines Based on the results of the research reported in this thesis, the following guidelines are proposed for the provision of information on road side panels, in-car systems and personal devices. Travellers are not easily persuaded to change their travel mode during their trip. It would therefore be advisable to focus on changing perceptions of alternatives to driving by car during the on-trip stage and to provide information on how the system works in the pre-trip stage. A change in travel behaviour should be encouraged in the pre-trip phase. Therefore, information that contains arguments in favour of travel behaviour change should be provided. If a change to public transport is possible on a route and multimodal travel information is available then the traveller should be informed during the pre-trip phase that on a certain location additional travel information is provided. The traveller is then able to plan a trip and explore the public transport option. In case the travel times for public transport are shorter than those of the car (e.g. in case of an accident) the traveller might more easily decide to change travel mode because in the pre-trip phase uncertainty about how to travel by public transport is reduced. In order to optimise the probability that travellers change their travel mode or route plans, information should be available on the internet that guides the traveller during the trip. For example, information that is available for smart phones. Uncertainty during the whole trip should be decreased. Road side travel information should inform road users about current or expected traffic situations in the travel network. The information should be designed so that elaboration on the information is hardly necessary. For example, by providing an advice instead of only information. The information should prompt the driver to change behaviour while making sure they are more or less in line with decisions that were made earlier in the pre-trip phase. Try to use the least amount of information possible. Dynamic information should be provided on a DRIP or GRIP. However, static information like parking fees, or that a parking place is secured etc., should be provided on static signs following the dynamic information. The presentation of a road structure may have the side-effect that travellers interpret a road that is displayed as longer also has a longer travel time. Do not include information that in itself would not have an effect on the route or mode choice. The road structure should not be presented as it is in reality as it then probably contains a lot of information that is not important (for example the location of bends). In-car information is more effective in encouraging changes in route or mode choice, but possibly less safe compared to road side information. Therefore, in-car travel information should be tailored information (only providing information that is relevant for the driver and the trip made); leave out information that is known by the driver. designed such that information fits on one screen designed such that with the least amount of elements the most important information is provided only provided at times when the driver can pay attention to the information Use colour coding to indicate stagnant traffic and slow moving traffic. Make the provision of travel information uniform for all information panels. The traveller should know what a travel time indicated means. During special events, travellers who are unfamiliar with the environment will change their route more easily. In such situations, travel information may be very effective particularly for this group of travellers. Information for travelling to an event (such as a festival) should be included on the website of this event. Before providing road side travel information, it is important to get answers into the following questions: for how many travellers is the information relevant? what is the origin and destination of the travellers on a specific road? what possibilities do travellers have to travel differently? during which situations does the information offer a real alternative to the travellers? what is necessary besides presenting the information panel to make the information effective?