From Signs to Minds: Wayfinding Design and Mental Maps

Michaela Skiles


When following directional signs through a new area, how much do people actually learn about the environment around them? How could you design directional signs to help people learn more? This study examines how the design of directional signs influences spatial learning, by presenting information in different spatial perspectives.

Three sign types were evaluated: Separate (directional arrows, with roads and towns on different signs), Combined (simple arrow diagrams of the intersection, with roads and towns on one sign), and Cartographic (a highly simplified map). Participants viewed a sequence of signs as if driving through a fictional environment, making turn choices according to assigned goals, and then completed a mapping task. After a second sign viewing, this time without turn decisions, participants repeated the mapping task.

For the first mapping task, participants who viewed the Cartographic signs produced more accurate maps than those viewing the Separate or Combined signs. These results suggest that guide signs with simple maps can help people incidentally learn about the spatial configuration of the environment. There was no significant difference between groups for the second mapping task, which suggests that when people are aware that they will be tested, sign type does not affect how much they can learn.

This study not only has implications for the design of directional signs, but is also an example of linking research in spatial cognition with wayfinding as a design discipline. Carried out as an undergraduate thesis, this study is evidence of an effective interdisciplinary approach to design education.


Sign Type Design

Based on background research, I created three types of directional signs to test the impact of perspective on spatial learning: Separate, Combined, and Cartographic. I used guidelines for U.S. highway signage, the Manual on Uniform Traffic Control Devices (MUTCD), as a foundation, paired with examples of existing guide signs around the world. To guide users’ turn decisions, these traditional signs provide simple directional arrows next to town names or under road shields. The signs followed the standard sequence according to the MUTCD standards: warn driver of an upcoming decision point, present destination options and then present route options, provide route confirmation, and then provide distance signs to upcoming destinations on the route.

The Separate sign type serves as the control group, as it most closely resembles the regulations and recommendations in the MUTCD. The town and road options are shown on signs that are separated in space and time. The viewer first sees the set of town-directions and then the set of road directions before making their turn decision. The confirmation information presented to the viewer after the turn decision also presents road information and town/ distance information separately in space and time. Thus, it may be challenging to use these signs to learn spatial relationships between roads and towns because of the separation.

The Cartographic signs were designed with the goal of helping users learn the spatial configuration of the landscape during travel. My approach to this design problem was to incorporate simple maps into the guide sides to not only point people in the right direction, but also provide them a simple map of the adjacent towns and roads. The Cartographic signs presented town and road information on a single sign, and showed topological relationships between towns and roads pictorially.

The Combined sign type was designed as an intermediate control group. Like the Cartographic signs, this type removed the temporal separation of road and town information by placing both kinds of information on one sign. These signs did not, however, show topological relationships between towns and roads pictorially. Instead, the signs showed a schematic representation of only the junction, while the road number, road direction, and town information were presented as a list clustered beyond each road arrow. By including this intermediate sign type, I hoped to understand whether combining road and town information on one sign or presenting it pictorially has a greater effect on spatial learning.

Experimental Design

To evaluate the impact of the three sign types on spatial learning, I developed an experiment to measure how well people could learn the layout of an environment from viewing directional signs alone. Participants viewed one of the three sign types, with approximately equal distribution between sign type groups.

Within the presentation, participants viewed a series of signs as if driving through a fictional environment. They were instructed to travel to particular goals, consisting of a target town and a specific road to take there. At each junction, the presentation displayed the turn options available (e.g. ‘left’ and ‘right,’ with an arrow for each), paused to allow participants to circle a turn choice on a paper form, and then resumed with a click of the mouse. Participants’ turn decisions served as a measure of the functional equivalency of the three sign types— whether the signs support immediate wayfinding. Additionally, this task kept participants focused on choosing the correct turn direction, thus distracting them from thinking about the overall spatial configuration of the environment.

Participants then completed an unanticipated mapping task to show their understanding of the layout of roads and towns. Provided with an 11×14- inch whiteboard, a dry-erase marker, and magnetic labels of all roads and towns, they had 6 minutes to construct a map of the environment they had experienced in the presentation. Participants were told that there was no need to use all of the town and road labels provided—if they were uncertain about any features, they could leave them off the map.

After completing the first mapping task, participants repeated the same sign viewing. This time, however, participants were not required to make turn decisions. The presentation did not pause at junctions, though it still included goals and the appropriate turn direction at each junction. Without the distraction of making turn choices, and because they were now aware of the means of evaluation, participants were able to focus their attention on learning the spatial configuration of the environment during the presentation. Participants then completed the mapping task a second time.

Because the second mapping task was not unexpected, it measured the intentional learning that is possible with the different sign types. The first mapping task, in contrast, measured incidental learning, because participants weren’t actively trying to construct a mental map of the area in anticipation of being evaluated.


Because highway driving is constrained to a network of roads and towns, I scored participants’ maps primarily based on their topological accuracy. A composite score was calculated for each map, based on the percentage of correct road-road, town-town, and road-town connections.

In the first mapping task, participants viewing the Cartographic signs showed a significantly better understanding of the connectivity of roads and towns than those viewing the other two sign types [Separate: t(32)=-2.768, P=0.009; Combined: t(31)=-2.246, P=0.032]. There was no significant difference between the Separate and Combined groups for the first mapping task [t(29)=-0.526, P=0.603]. The histogram shows the distribution of participants’ topological accuracy scores by sign type viewed. These results suggest that simple maps on signs can help people learn about the layout of their environment incidentally during travel. Combining road and town information on a single sign showed no significant impact on spatial learning unless the information was presented in the survey perspective of a map.

In the second mapping task, there were no significant differences in map accuracy between any of the sign type groups. In other words, when people are actively trying to construct a mental map, the sign type doesn’t significantly impact performance. In practice, however, intentional learning from directional signs is much less common than incidental learning. So while the results of the second mapping task are interesting to note, they are less relevant to the practice of designing wayfinding signage.


Context in Spatial Cognition

The premise of my study is rooted in cognitive research on the different perspectives we use to store and communicate spatial information: route and survey. Route perspective takes you from one point to another along a linear series of turns at decision- making points. Survey perspective is from a birds’- eye view, like a map, and gives you a more complex understanding of the layout of the area (Appleyard, 1970; Siegel and White, 1975; Taylor et al., 1999). With survey knowledge of an area, you’re better equipped to find to new destinations and take shortcuts or detours along your journey. Route perspective is the most common way of communicating information in directional signage, but survey perspective is also possible. Survey knowledge has been considered a higher level of spatial knowledge in part, because it allows for more flexible future route planning, including the ability to find new routes and shortcuts through the environment (Golledge, 1999). Survey knowledge is often associated with route knowledge based on a progression of learning, where knowledge of individual routes through an area becomes interconnected and eventually helps construct a survey understanding of the area (Golledge, 1999; MacEachren, 1992). Yet, there is also evidence suggesting alternatives to this progression—that survey information, including directions and distances, can in fact be acquired early on in a person’s exposure to an environment (Ishikawa and Montello, 2006; Montello, 1998).

Navigational aids may convey spatial information in a route or survey perspective, or some combination of the two. These differences may affect the amount of cognitive effort required to wayfind. Traditional directional signs or verbal directions convey information in a route perspective and directly instruct the user’s turn actions to reach a destination (Freksa, 1999; Hölscher et al., 2005; Raubal, 2001). In contrast, traditional maps generally require the user to translate the given information from a survey to a route perspective—from spatial configuration to individual turn actions (Freksa, 1999; Münzer et al., 2006; Shelton and McNamara, 2004).

Contributions of the Current Study

My results suggest that information design can influence people’s spatial learning. All of the sign designs helped participants choose correct turns while navigating a network. The Cartographic signs, which put simple maps on the signs to present town and road information pictorially, appear to have improved participants’ ability to passively learn about spatial configurations when compared with participants who used signs that either separated spatial information in time or presented spatial information together in time but not pictorially. While I found that the impact of design fades as experience with a place and attention to learning increase, my research highlights the potential role of design to improve incidental learning of spatial configurations by people guided by signs through an unfamiliar environment.

The relative differences in learning between the three sign type groups also offer some insight regarding design decisions that facilitate spatial learning. The Cartographic sign expressed two decisions: (1) incorporate road and town information on a single sign to minimize the split attention effect and (2) present information about topological relationships pictorially. The Combined signs only did the former, and the Separate signs did neither. I found a greater difference in topological accuracy between the Combined and Cartographic groups than between the Separate and Combined groups. This suggests that presenting road and town information on the same sign may not facilitate spatial learning unless the sign pictorially shows topological relationships in the form of a map.


Although there is already a solid body of research that informs the design of wayfinding guidance (e.g., evaluating typeface legibility), this study is an example of how cognitive research methods can be used to help create more effective signage.

Recent research has revealed that people don’t onstruct a mental map of their environment when using turn-by-turn navigational guidance like in-car GPS and smartphones. Wayfinding designers have the ability to counteract this trend and support spatial awareness, and cognitive research methods can help to identify effective design patterns towards this aim. One such pattern, as demonstrated in this study, is to provide travelers with quick glimpses of the layout of the area, in the form of simple maps on directional signs.

In particular, a cognitive approach to wayfinding design can help encourage a longer-term approach to navigational guidance. Signs can go beyond simply guiding travelers from one place to the next every time they need to take a certain route. If designers can create signs that actually teach users about the layout of their environment, eventually people will be able to travel that route more independently, and learn shortcuts and new routes within the area.


Although I used U.S. highway signage as the foundation for my signs, the practice of incorporating maps into guide signs applies to many other contexts beyond the highway. In fact, it is much more feasible to design wayfinding signs with maps for pedestrians, because sign viewing time is more flexible, allowing for more complex information. From urban bicycle routes to public transportation, the possibilities are endless. The uniting theme is a long-term, holistic approach to wayfinding guidance—not just getting people to their destinations, but also helping them learn about the environment along the way.

There are indeed many examples of map-like elements already being used. However, relative to the presence of verbal signs with simple directional arrows, cartographic signs are few and far between. Thus, while verbal signs have been highly evaluated and regulated in terms of organization, color, complexity, etc., there is much research yet to be done on the practice of incorporating simple maps on signs. As a result, there is still little guidance on how complex of a map could be presented for different modes of travel so as to not overwhelm the user. Further experimentation, both in practice and in research, will help to develop design guidelines for the most effective cartographic signs in various contexts.


Finally, based on my first-hand experience with this project, I can strongly advocate for this interdisciplinary approach to independent academic work. By connecting the practice of wayfinding design with cognitive research, I was able to identify a potentially effective design pattern and then carry out an experiment to test my predictions. This comprehensive approach to design education is unusual and challenging, but far more rewarding in the long run. I was forced to look beyond aesthetics and legibility in signage design, to better understand both how people perceive spatial information as well as how to most effectively present spatial information.


Appleyard, D. (1970). Styles and methods of structuring a city, Environment and Behavior, 2, pp. 100–117.

Freksa, C. (1999). Spatial aspects of task-specific wayfinding maps, in Visual and Spatial Reasoning in Design, ed. by Gero, J. S. and Tversky, B., pp. 15–32, University of Sydney, Sydney.

Golledge, R. G. (1999). Human wayfinding and cognitive maps, in Wayfinding Behavior: Cognitive Mapping and Other Spatial Processes, ed. by Golledge, R. G., Johns Hopkins University Press, Baltimore.

Hölscher, C., Büchner, S. J., Bröosamle, M., Meilinger, T. and Strube, G. (2005). Signs and maps: cognitive economy in the use of external aids for indoor navigation, Proceedings of the 29th Annual Conference of the Cognitive Science Society, pp. 377–382, Nashville, TN, Aug 1–4.

Ishikawa, T. and Montello, D. R. (2006). Spatial knowledge acquisition from direct experience
in the environment: Individual differences in
the development of metric knowledge and the integration of separately learned places. Cognitive Psychology, 52, pp. 93–129.

MacEachren, A. (1992). Application of environmental learning theory to spatial knowledge acquisition from maps, Journal of the Association of American Geographers, 82(2), pp. 245–274.

Montello, D. R. (1998). A new framework for understanding the acquisition of spatial knowledge in large-scale environments, in Spatial and Temporal Reasoning in Geographic Information Systems, ed. by Egenhofer, M. J. and Golledge, R. G., pp. 143–154, Oxford University Press, New York.

Münzer, S., Zimmer, H. D., Schwalm, M., Baus, J. and Aslan, I. (2006). Computer-assisted navigation and the acquisition of route and survey knowledge, Journal of Environmental Psychology, 26, pp. 300–308.

Raubal, M. (2001). Human wayfinding in unfamiliar buildings: A simulation with a cognizing agent, Cognitive Processing, 2(3), pp. 363–388.

Shelton, A. L. and McNamara, T. P. (2004). Orientation and perspective dependence in route and survey learning, Journal of Experimental Psychology: Learning, Memory, and Cognition, 30(1), pp. 158–170.

Siegel, A. W. and White, S. H. (1975). The development of spatial representations of large-scale environments in Advances in Child Development and Behavior, Vol. 10, ed. by Reese, H. W., pp. 9–55, Academic Press, New York.

Taylor, H. A., Naylor, S. J. and Chechile, N. A. (1999). Goal-specific influences on the representation of spatial perspective, Memory & Cognition, 27(2), pp. 309–319.

PDF of full article