“Turn right at the Traffic Lights”
The Requirement for Landmarks in Vehicle Navigation Systems
Dr Gary Burnett
HUSAT Research Institute, Loughborough University

ABSTRACT
This paper argues for the use of landmarks (e.g. traffic lights, churches, petrol stations)
within the turn-by-turn visual and voice directions given by in-vehicle navigation
systems. Such prominent features of the driving environment are consistent with basic
human navigational strategies, are valued by drivers, and have been shown significantly
to improve the usability of electronic in-car navigational aids. For future systems actively
to include such information, it is critical that a) only ‘good’ landmarks are used, b) such
landmarks are presented to the driver in the most appropriate way, and c) the practical
needs of industry are fully accounted for.

Key words: Vehicle Navigation, Human Factors, Design

1. INTRODUCTION
Navigating in unfamiliar road environments is perhaps the most common and demanding
cognitive activity that drivers undertake. There is a wealth of literature that indicates that
many people experience difficulties in both planning and following efficient routes to
their destinations (e.g. King, 1986; Wierwille, Antin, Dingus & Hulse, 1989). The
resulting navigational uncertainty manifests itself in a number of ways, either for the
individual driver (e.g. stress and frustration), their relationship with other road users (e.g.
misleading use of indicators, sudden braking), or for the traffic system as a whole (e.g.
poor route choices).

Route Guidance and Navigation systems for vehicles "provide information on community
and/or individual user optimum route options for specified destinations" (ISO TC204/
WG1, 1996). For the majority of current systems, positioning information is received via
GPS, an on-board compass, and wheel sensors, and this data is matched with a CD-ROM
navigable map database holding information on roads, road features, priorities, etc. With
respect to the driver-system interface, many of the systems that are presently available can
be classified as being “turn-by-turn”. The driver is given instructions (using symbols
and/or text and often voice messages) relating to the location and direction of each
manoeuvre. Current voice messages tend to emphasise distance-to-turn information,
using either absolute values (e.g. “left turn in 300 metres”), or non-absolute, time-based
terms (e.g. “left turn soon”). Visual displays also tend to stress distance-to-turn
information, and may use simple arrow symbols, or a junction-specific representation
(See Figure 1):-



It has been claimed that the widespread implementation of vehicle navigation systems
will:

It is now possible for drivers in many countries to purchase first generation vehicle
navigation systems. It has been predicted that such systems will be commonplace in
vehicles throughout the developed nations in the near future (Zhao, 1997). Their
popularity can already be seen in Japan, where it has been estimated that in excess of 3.5
million vehicles currently have route guidance systems installed (Rowell, 1999).

2. THE USABILITY OF CURRENT SYSTEMS
The human factors approach places the needs, abilities, preferences, etc. of the intended
users at the centre of the design process for a system. The usability of a system refers to
the “quality of interaction between a user and other parts of the system overall” (ISO
9241 – part 11, 1997, p.2), and has been rated as one of the most important aspects of
design for a vehicle navigation system (Barrow, 1991; French, 1997). Potentially, the use
of such sophisticated systems while driving could adversely affect the ability of drivers to
control their vehicles safely and respond to potential hazards. Lack of attention and
distraction are already major contributory factors in many road accidents (Wierwille,
1995). Therefore, any system which might add to this problem must be carefully
designed. In this respect, it is critical that vehicle navigation systems provide appropriate
information when and where needed in a form that is easily digested by the driver.

Previous research has shown the benefits that well designed turn-by-turn systems can
have over the use of paper maps. In particular, there is empirical evidence that the use of
turn-by-turn systems compared to maps leads to less navigational errors and shorter
journey times, reduced mental demands and increased confidence in navigating (e.g.
Streeter, Vitello and Wonsiewicz, 1985; Walker, Alicandri, Sedney and Roberts, 1991).
There are two main reasons why turn-by-turn systems offer such advantages over what is
arguably the worst case scenario. First, information is only presented which is relevant to
the oncoming manoeuvre, i.e. the system filters information, rather than the driver (Ross
et al., 1995). Second, the use of turn-by-turn information which is largely verbal in nature
conflicts less with the predominately visual-spatial task of driving (Wetherell, 1979).

In contrast with the research comparing turn-by-turn navigation systems with paper maps,
there has been other research which has examined driver behaviour and performance
when using turn-by-turn systems versus that attained from using instructions given by the
passenger (e.g. Fastenmeier et al., 1994; Burnett and Joyner, 1997; Zaidel and Noy,
1997). It is argued by these authors that the passenger with detailed route knowledge who
provides clear, timely instructions reflects the ideal situation. In these studies, when using
a turn-by-turn navigation system versus instructions given by the passenger, drivers have
been found to:

Clearly, the performance of drivers using current turn-by-turn vehicle navigation systems
does not attain the same level as that which is achieved when using the ideal navigator.
Indeed, there is now some evidence from Japan that accidents are arising from the use of
the vehicle navigation systems that are presently on offer. In a press release, the Japanese
Ministry of Transport declared that, in the first six months of 1998, ninety-three people
were injured and one person was killed in accidents linked to the use of in-car navigation
systems on the move (Ito, 1998). If next generation systems are to approach the ‘ideal’
standard, then it is imperative that more intelligent and naturalistic interfaces are
developed.

4. WHY SHOULD SYSTEMS PRESENT LANDMARKS?
Current vehicle navigation systems present few landmarks (traffic lights, churches,
telephone boxes, bridges, petrol stations, etc.) within their turn-by-turn (visual and voice)
directions. However, as will be seen in the following sections (4.1 to 4.3), there is
considerable evidence that such prominent features of the environment are essential for
optimal human navigation. Three basic arguments can be made as to why landmarks
should be an integral part of future vehicle navigation systems:

4.1 Landmarks are consistent with basic human navigational strategies
Studies conducted within the environmental psychology and human geography disciplines
have indicated that landmarks are core components of peoples’ mental representations of
large-scale space (commonly termed “cognitive maps”), and play an important part in the
environmental learning process (e.g. Evans et al., 1984). As a result, they are among the
most well known features of the environment and can act as powerful cues for navigation.
Thus, it is not surprising to find that they are widely used in traditional way-finding
strategies, e.g. as part of directions that people provide for others (Alm, 1990). Indeed, all
of the ‘ideal’ passenger instructions used in the studies described above in section 3 were
reported to include a number of landmarks.

The fundamental need of people to use prominent landmarks for navigation is best
illustrated by reference to environments where landmarks are not apparent. Mazes and
labyrinths are inherently difficult to navigate within, and it is their essential lack of
distinctive features that causes people to become disoriented and confused (Arthur and
Passini, 1992). People need to be able to differentiate between the scenes present at
intersections if they are to gain a cognitive understanding of an environment (Sorrows
and Hirtle, 1999). For drivers, such understanding may enable individual turning
decisions to be made, indicate that the correct route is being followed, and/or enable a
sense of location in relation to the surroundings (Golledge, 1999).

4.2 Landmarks are valued by drivers
Not only does it appear that landmarks are central to the navigational directions we
provide for others, but they are also a significant feature of directions that we request or
value from others. A survey of 1158 UK drivers has shown the importance that people
attach to landmark information. Participants were asked what information they would
want from the passenger to help them navigate, and rated landmarks as their second most
popular information type (after left-right directions) (Burns, 1997) – see Figure 2. There
have been several other studies which confirm such findings (e.g. Streeter and Vitello,
1986; Burnett, 1998).

The perception of landmark value arises because of their role in direction giving (as
discussed above), but also largely due to their individuality. There are both objective and
subjective aspects relating to what constitutes a landmark for a person. In addition to their
purely sensory visual characteristics, the majority of landmarks have functions, and thus
are associated with particular goals for an individual (e.g. restaurants, public houses,
parks) (Kaplan, 1976). Furthermore, many landmarks have a symbolic, cultural and/or
historical role within a community (e.g. monuments, town halls) (Sorrows and Hirtle,
1999).

4.3 Landmarks can improve the usability of vehicle navigation systems
The most direct evidence for the benefits to be gained from landmarks within vehicle
navigation systems comes from the human factors literature. There have been several
studies which have investigated the usability of systems which include landmarks
compared with those which do not. Usability can be defined as a function of
effectiveness, efficiency and satisfaction (ISO 9241 – part 11, 1997). At this point, it is
useful to employ these criteria to show how key research studies within the literature have
found the usability of ‘landmark’ systems to be superior to ‘non-landmark’ systems for all
three aspects of system usability.

4.3.1 Effectiveness
Effectiveness concerns the achievement, or otherwise of the goals of the user. Commonly,
a driver’s navigational goals will be to reach a destination whilst taking no wrong
turnings.

Bengler, Haller and Zimmer (1994) conducted a simulator-based experiment which
demonstrated how landmarks can improve the effectiveness of a vehicle navigation
system.

Twenty-four participants viewed a series of videotaped routes whilst carrying out a
simple tracking task. Half of the participants were provided with visual only navigation
information - a simplified representation of the junction with no other information. The
remaining participants were also provided with landmark information at junctions.
Participants were instructed to use the guidance information to make navigational
decisions, and to register their judgements by turning the steering wheel and employing
the indicators. These parameters therefore were being used to indicate navigational, rather
than driving, performance. Navigation information that included landmarks was found
significantly to reduce the number of incorrect uses of the indicators (see Table 1). There
was also a trend for reduced steering errors with landmark information (approximately
30% fewer errors), but this difference was not significant.



4.3.2 Efficiency
Efficiency largely concerns the resources that are expended by a user to achieve their
goals. For the use of a vehicle navigation system, such resources may be time-related (e.g.
number/length of glances made to in-vehicle display, journey time), cognitive (e.g. mental
workload) or relate to primary driving task performance (e.g. steering wheel variability,
use of brakes and indicators, lane changing, traffic violations). Clearly, this category
includes those measures which have the strongest links with system safety.

In some recent road-based work, the efficiency-related benefits of using landmarks for
navigation were shown (Burnett, 1998). Twenty participants drove unfamiliar routes
within an urban area using either a vehicle navigation system (turn-by-turn directions on
both a display and voice) that emphasised landmarks or one that stressed distances for the
purposes of locating manoeuvres (see Table 2). When using the landmark system,
relatively few glances were made towards the display (on average 1.6 on the approach to
a turning), and workload was perceived to be lower (mean 26.8 on a 1-100 scale, where
1=low and 100=high), in comparison with the figures attained for a distance-to-turn
oriented interface (mean number of glances: 5.0; mean workload 40.6). Furthermore, in
contrast with previous assessments of vehicle navigation system interfaces (e.g.
Wierwille et al., 1989; Burnett and Joyner, 1997), the durations of glances towards the
landmark display were low (mean 0.66 secs).

4.3.3 Satisfaction
Satisfaction is an important element of usability, and techniques such as questionnaires
and interviews have been commonly used in the literature to measure it. It is this
component of system usability for which there is greatest evidence for the benefits of
landmarks, and two significant studies are reported here.

Alm, Nilsson, Järmark, Savelid and Hennings (1992) conducted a road trial in which
twenty drivers used a simulated vehicle navigation system that provided simultaneous
visual and aural directions. In the control group ten participants were presented with only
very simple left/right/straight on information, whereas in the experimental group, the
remaining drivers received the same information plus information regarding landmarks
along the route. Participants in the landmark condition felt significantly more confident as
to where to turn, and generally more satisfied with the content of visual information (nonsignificant trend).

Green, Hoekstra, Williams, Wen and George (1993) conducted a simulator-based study in
which 48 drivers followed routes using four different types of navigation system
interface: visual navigation information only, visual with landmarks, auditory navigation
information only, auditory with landmarks. Participants were instructed to press one of
three buttons when they could see the junction referred to by the system (left/
right/straight on). In this study, drivers strongly preferred the interfaces that contained
landmarks over those without. Indeed, observation of Table 3 reveals that the
inclusion/exclusion of landmarks factor had a greater influence on driver preference than
did the factor of interface modality.



5. HOW CAN SYSTEMS INCLUDE LANDMARKS?
It is apparent from the theoretical and empirical research described above that there are
considerable benefits to be gained from the use of landmarks by vehicle navigation
systems. Despite this potential, there are a number of fundamental research and practical
challenges which must be met to ensure these strong visual cues become an integral part
of future systems.

5.1 Predicting good landmarks
The quality of landmarks for navigation purposes can vary considerably throughout the
environment. This issue is critical in the driving context, since the use of ‘poor’
landmarks, for instance, those which are difficult to find or uniquely identify, may lead to
driver confusion, increase workload, and reduce driving safety. In the human factors
literature there have been some attempts to generate lists or examples of ‘good’
landmarks. Studies have been conducted where lists of those landmarks most commonly
reported by drivers have been drawn up (e.g. Alm, 1990; Akamatsu et al., 1997), or a
limited number of landmarks have been evaluated within a prototype navigation system
(e.g. Alm et al., 1992; Bengler et al., 1994; Green, Levison, Paelke and Serafin, 1995).
The central problem with such work has been that they lead to results which are wholly
environment, country and study specific.

In addressing this concern, it would be important to establish the specific characteristics
or attributes of a landmark which will influence the ease with which it can be processed
and remembered. Alm (1990), Akamatsu et al. (1997) and Green et al. (1995) have all
commented on this issue. Alm suggests that people consider some landmarks to be more
useful than others for navigation purposes primarily because of their commonality across
urban areas. Furthermore, he states that popular landmarks tend to be visible in most
conditions, and are easy to differentiate and learn. In agreement to a certain extent,
Akamatsu et al. feel that the landmarks commonly referred to by subjects in their study
were visible from a distance, unique in appearance, and were close to or part of the road
infrastructure. Green et al. have also stressed similar characteristics of 'good' landmarks.
They believe the best landmarks are those which can be seen at a great distance (at all
times), are close to the road, near intersections, and are relatively permanent.

It is apparent that the 'common sense' observations made by all of these authors are rather
casual. No human factors research to date has addressed exactly which characteristics or
attributes of an object within the physical environment result in it being used or
considered useful as a landmark for navigational purposes. Work conducted outside of
the human factors discipline has little to offer on this issue. Geographers and urban
planners have aimed to identify the attributes of landmarks that make them distinctive or
memorable (e.g. Appleyard, 1969; Hirtle and Sorrow, 1999), whereas psychologists have
focused on the role of landmarks within peoples' cognitive maps (e.g. Sadalla, Burroughs
and Staplin, 1980; Tversky, 1993). What is needed is a structured and generic approach to choosing appropriate navigational landmarks, which can ultimately be used in any
environment (i.e. across different countries or regions).

5.2 Understanding context
To inform the choice of landmarks for navigation, there is a need to understand the
relevant contextual factors. A number of individual (i.e. person-related) and
environmental factors which may be of significance can be hypothesised.

5.2.1 Individual differences
An individual factor of potential relevance to the use of landmarks for wayfinding is
perceptual style, commonly referred to as field dependence/independence. Field-independent people are better at distinguishing relevant cues from irrelevant cues in their
environment than those who are field-dependent. Several studies have produced evidence
that field dependent drivers are more likely to have accidents than field independent
drivers (e.g. Harano, 1970). Goodenough (1976) believes the reasons for this, among
others, are that field dependent drivers do not quickly recognise developing hazards, and
are slower in responding to embedded road signs (those surrounded by many other
stimuli). It may be hypothesised that these reasons would influence an individual's
preference for, and use of, potentially embedded information within the environment,
such as landmarks (Gould, 1989).

In addition, certain drivers may prefer particular landmarks based on lifestyle
considerations. As discussed in 4.2, there are strong subjective components to what
constitutes a ‘landmark’ for an individual, based largely on its functional characteristics.
As a result, when giving navigational directions, many people use landmarks that are
appealing to them. A typical example can be seen in the person who provides directions
based mainly on particular types of shops.

5.2.2 Environmental factors
Undoubtedly, it is critical that drivers are able to see landmarks clearly from a distance.
The prevailing weather and time of day are critical environmental factors that influence
landmark visibility. All previous research concerning the interface for vehicle navigation
systems has been conducted during daytime hours in generally clear conditions. Yet, it is
likely that the strategies adopted by people when navigating under degraded visual
conditions (e.g. travelling at night-time on unlit roads) vary considerably from those used
in optimal situations.

5.3 Identifying appropriate presentation methods
It will be extremely important for the design of a vehicle navigation system to establish
exactly how to present landmark information to the driver. A large number of different
landmark types could potentially be presented by a system, and a poor visual and/or aural
representation of a landmark will have implications for the effectiveness, efficiency and
degree of satisfaction of a system.

A road-based study aimed specifically to establish effective ways of visually presenting
landmarks within a vehicle navigation system (Pauzié, Daimon and Bruyas, 1997). Two
approaches were examined: a generic presentation (e.g. the same visual icon for all
churches); or a specific presentation (e.g. a representation of a given church). In an urban
driving environment, ten participants negotiated a route using a simulated vehicle
navigation system in which turn-by-turn directions (visual only) were provided. It was
apparent from driver feedback that the familiarity of the landmark representation was the
most important factor determining whether drivers considered the specific or generic
design to be more useful for navigation. For instance, specific presentations that included
a well-known logo or name (e.g. MacDonalds, Natwest) were preferred to their generic
equivalent (e.g. a symbolic representation of a burger, coins and notes). In contrast, in
situations where the generic design was familiar (e.g. a church icon), the more detailed
specific representation was generally rated less favourably.

Such results appear to be of particular relevance to the choice of landmarks, although this
is not mentioned by the authors. Many landmarks will frequently change their labelling
(e.g. a BP petrol station becoming Shell), whilst maintaining their basic function. As a
consequence, landmarks that lend themselves to a generic presentation may be preferable
for use within a vehicle navigation system.

In addition to this work, there have been several studies which have assessed a particular
representation of a landmark as part of an overall vehicle navigation system evaluation
(e.g. Alm et al., 1992; Bengler et al., 1994). Of these, Green et al. (1995) are the only
authors to use the results of their evaluations to make some points regarding efficient
means of representing landmarks. In their paper outlining some preliminary guidelines for
designers, they state that landmarks should be provided both visually (as graphics, rather
than text) and aurally. Furthermore, they specify that traffic light and stop sign graphics
should be placed in the centre of the intersection representation. Although such a
recommendation would appear to constitute good human factors practice, it should be
noted that the evaluations were conducted in the state of Michigan, USA which has a
predominantly grid-based road layout. It is possible that this particular guideline would be
more difficult to achieve in cities which have more complex junction layouts.

Despite the initial progress made, there still remains a need for research studies which
compare alternative iconic and verbal representations for a wide range of landmarks. It
will be particularly important to develop standard iconic representations of landmarks for
use in visual displays. Many drivers will use different cars and navigation systems on a
frequent basis (e.g. habitual hired car users), and it will be necessary to ensure an
appropriate transfer of learning.

5.4 Accounting for industry’s requirements
To develop a given vehicle navigation system product, a number of design stages take
place, involving several different organisations (e.g. map database providers, navigation
system suppliers, vehicle manufacturers). There has been no research which has
investigated landmarks in the context of this design process. In practical terms, such
knowledge is essential if landmarks are to be actively used within future vehicle
navigation systems. Issues regarding the underlying navigable map database would seem
of particular importance, for instance:

(a) improving the up-dating process, e.g. a Help Desk for users to report corrections -Temes (1996)
(b) distributing maps more efficiently, e.g. the use of kiosks at car dealerships for providing map databases (Gupta and Angerman, 1996), or centralised storage of maps which are transmitted via wireless communications (Hakula, Vehviläinen and Ojala, 1996).

6. CONCLUSIONS
Electronic GPS-based vehicle navigation systems are now available to drivers in Europe
and North America, and these will soon be affordable by the mass market. Optimum
design of the driver-system interfaces is critical for the safe and effective deployment of
the technology. The level of usability of typical vehicle navigation systems can be said to
lie between that of existing paper maps and that of instructions given by the informed
passenger. Studies within the human factors discipline and elsewhere have shown that the
safety, effectiveness and acceptability of these in-vehicle systems could be significantly
enhanced, if the interface reflected basic human navigational strategies, and incorporated
landmarks within turn-by-turn directions.

However, it is not a simple task to include landmarks within vehicle navigation systems.
There is a vast number of different types of landmark that could potentially be presented
to drivers to support them in the navigation task, and designers need guidance on which
are appropriate in a given context, and how these should best be presented. Future
research must establish the salient attributes of landmarks that make them effective cues
for navigation purposes, and ascertain how the navigational effectiveness of landmarks
can be predicted. The optimum means of representing landmarks must also be
investigated. Finally, there is a practical need to identify how effective landmarks can be
incorporated within navigable map databases for presentation by future vehicle navigation
systems.

ACKNOWLEDGEMENTS
This paper was written under the funding provided by the REGIONAL (RoutE GuIdance
systems: Optimal Navigation via the use of Landmarks) project. REGIONAL is a
collaborative research project funded by the UK government (EPSRC) under the LINK
Inland Surface Transport Programme. The partners are the HUSAT Research Institute at
Loughborough University, Alpine Electronics of UK Ltd, Navigation Technologies
(NavTech), Jaguar Cars Ltd, Motor Industry Research Association (MIRA) and RAC
Motoring Services.

REFERENCES
Akamatsu, M., Yoshioka, M., Imacho, N., Daimon, T., & Kawashima, H. (1997).
Analysis of driving a car with a navigation system in an urban area. In Y. Ian Noy
(Ed.), Ergonomics and safety of intelligent driver interfaces (pp. 85-96). Mahwah, NJ:
Lawrence Erlbaum Associates.

Alm, H. (1990). Drivers' cognitive models of routes. In W. van Winsum, H. Alm, J.M.
Schraggen & J.A. Rothengatter (Eds.), Laboratory and field studies on route
representation and drivers’ cognitive models of routes (DRIVE II V1041 GIDS,
Deliverable GIDS/NAV2, pp. 35-48). Groningen, The Netherlands: University of
Groningen, Traffic Research Centre.

Alm, H., Nilsson, L., Jármark, S., Savelid, J., & Hennings, U. (1992). The effects of
landmark presentation on driver performance and uncertainty in a navigation task - A
field study (Swedish Prometheus, Tech. Rep. No. S/IT-4). Linköping, Sweden: VTI.

Appleyard, D. (1969). Why buildings are known - A predictive tool for architects and
planners. Environment and Behaviour (December), 131-156.

Arthur, P., & Passini, R. (1992). Wayfinding: people, signs, and architecture. New York:
McGraw Hill.

Barrow, K. (1991). Human factors issues surrounding the implementation of in-vehicle
navigation and information systems (SAE Tech. Paper Series 910870, pp. 1243-
1257). Warrendale, PA: Society of Automobile Engineers.

Bengler, K., Haller, R., & Zimmer, A. (1994). Experimental optimisation of route
guidance information using context information. In Proceedings of the First World
Congress on Applications of Transport and Intelligent Vehicle Highway Systems:
Vol. 4. (pp. 1758-1765). Boston: Artech House.

Burnett, G.E. (1998). "Turn right at the King’s Head": Drivers’ requirements for route
guidance information. Unpublished PhD dissertation, Loughborough University, UK.

Burnett, G.E., & Joyner, S.M. (1997). An assessment of moving map and symbol-based
route guidance systems. In Y. Ian Noy (Ed.), Ergonomics and safety of intelligent
driver interfaces (pp. 115-136). Mahwah, NJ: Lawrence Erlbaum Associates.

Burns, P.C. (1997). Navigation and the older driver. Unpublished PhD dissertation,
Loughborough University, UK.

Evans, G.W., Skorpanich, A.A., Gärling, T., Bryant, K.J., & Bresolin, B. (1984). The
effects of pathway configuration, landmarks and stress on environmental cognition.
Journal of Environmental Psychology 4 (4), 323-335.

Fastenmeier, W., Haller, R., & Lerner, G. (1994). A preliminary safety evaluation of route
guidance comparing different MMI concepts. In Proceedings of First World Congress
on Applications of Transport Telematics and Intelligent Vehicle Highway Systems:
Vol. 4. (pp. 1750-1756). Boston: Artech House.

French, R.L. (1997). Land vehicle navigation – A worldwide perspective. The Journal of
Navigation, 50 (3), 411-415.

Golledge, R.G. (1999). Human wayfinding and cognitive maps. In R.G. Golledge (Ed.),
Wayfinding behaviour: Cognitive mapping and other spatial processes (pp. 5-45).
Baltimore, MD: John Hopkins Press.

Goodenough, D.R. (1976). A review of individual differences in field dependence as a
factor in auto safety. Human Factors 18 (1), 53-62.

Gould, M.D. (1989). Considering individual cognitive ability in the provision of usable
navigation assistance. In D.H.M. Reekie, E.R. Case & J. Tsai (Eds.), Proceedings of
Vehicle Navigation and Information Systems Conference (pp. 443-447). Piscataway,
NJ: Institute of Electrical and Electronics Engineers.

Green, P., Hoekstra, E., Williams, M., Wen, C., & George, K. (1993). Examination of a
videotape-based method to evaluate the usability of route guidance and traffic
information systems (Tech. Rep. No. UMTRI-93-31). Ann Arbor, MI: University of
Michigan Transportation Research Institute.

Green, P., Levison, W., Paelke, G., & Serafin, C. (1995). Preliminary human factors
design guidelines for driver information systems (Tech. Rep. No. FHWA-RD-94-
087). Washington, DC: U.S. Government Printing Office.

Gupta, S., & Angerman, J. (1996). The advantages of Kiosk-based navigable map
database distribution: utilising lessons from the automotive logistics model. In
Proceedings of 3rd World Congress on ITS [CD-ROM]. Tokyo: VERTIS.

Hakula, H., Vehviläinen, K., & Ojala, T. (1996). Experiences with centralised route
guidance and a comparison to stand-alone approaches. In Proceedings of 3rd World
Congress on ITS [CD-ROM]. Tokyo: VERTIS.

Harano, R.M. (1970). Relationship of field dependence and motor-vehicle accident
involvement. Perceptual and Motor Skills, 31, 372-374.

ISO (1996). Transport Information and Control Systems: Fundamental TICS Services,
TC204/WG1 Technical Report, May 1996.

ISO (1997). Ergonomics requirements for office work with Visual Display Terminals
(VDTs) (ISO 9241: Part 11 ISO/DIS 9241-11: Guidance on Usability). Geneva,
Switzerland: International Standards Organisation.

Ito, H., 1998. Member of ISO TC22/SC13/WG8, personal communication, December,
1998.

Kaplan, S. (1976). Adaption, structure and knowledge. In G. Moore & R. Golledge
(Eds.), Environmental knowing: Theories, research and methods (pp. 32-45).
Stroudsburg, PA: Dowden, Hutchinson and Ross.

King, G.F. (1986). Driver performance in highway navigation tasks. Transportation
Research Board, 1093, 1-10.

OECD - Organisation for Economic Co-operation and Development. (1988). Route
guidance and in-car communication systems (A report prepared by an OECD
scientific experts group). Paris: OECD.

Pauzié, A., Daimon, T., & Bruyas, M. (1997). How to design landmarks for guidance
systems. In Proceedings of 4th World Congress on ITS [CD-ROM]. Brussels,
Belgium: ERTICO.

Ross, T., Vaughan, G., Engert, A., Peters, H., Burnett, G.E., & May, A.J. (1995). Human
factors guidelines for information presentation by route guidance and navigation
systems (DRIVE II V2008 HARDIE, Deliverable 19). Loughborough, UK: HUSAT
Research Institute.

Rowell, M. (1999). Intelligent transportation, map databases, and standards. In
Proceedings of Seminar on Integrated Solutions for Land Vehicle Navigation (pp. 1-
8). Nuneaton, UK: Motor Industry Research Association (MIRA).

Sadalla, E.K., Burroughs, W.J., & Staplin, L.J. (1980). Reference points in spatial
cognition. Journal of Experimental Psychology: Human Learning and Memory, 6 (5),
516-528.

Sorrows, M.E., & Hirtle, S.C. (1999). The nature of landmarks for real and electronic
spaces. In C. Freska and D.M. Mark (Eds.), Spatial Information Theory – Cognitive
and Computational Foundations of Geographic Information Science (COSIT ’99):
Vol. 1661 (pp. 37-50). Berlin Heidelberg: Springer-Verlag.

Streeter, L.A., & Vitello, D. (1986). A profile of drivers' map-reading abilities. Human
Factors, 28 (2), 223-239.

Streeter, L.A., Vitello, D., & Wonsiewicz, S.A. (1985). How to tell people where to go:
Comparing navigational aids. International Journal of Man-Machine Studies, 22, 549-
562.

Temes, I. (1996). Responding to customers: Timely maintenance for digital map
databases. In Proceedings of 3rd World Congress on ITS [CD-ROM]. Tokyo:
VERTIS.

Tversky, B. (1992). Distortions in cognitive maps. Geoforum, 23 (2), 131-138.

Van Duren, S., & Lydon, T.A. (1997). The task of building a European Point Of Interest
database. In Proceedings of 4th World Congress on ITS [CD-ROM]. Brussels,
Belgium: ERTICO.

Walker, J., Alicandri, E., Sedney, C., & Roberts, K. (1991). In-vehicle navigation
devices: effects on the safety of driver performance. In Proceedings of Vehicle
Navigation and Information Systems Conference (pp. 499-525). Piscataway, NJ:
Institute of Electrical and Electronics Engineers.

Wetherell, A. (1979). Short-term memory for verbal and graphic route information. In
Proceedings of the Human Factors Society 23rd Annual Meeting (pp. 464-468). Santa
Monica, CA: Human Factors and Ergonomics Society.

Wierwille, W.W. (1995). Development of an initial model relating driver in-vehicle
visual demands to accident rate. Third Annual Mid-Atlantic Human Factors
Conference Proceedings, Blacksburg, VA: Virginia Polytechnic Institute and State
University.

Wierwille, W.W., Antin, J.F., Dingus, T.A., & Hulse, M.C. (1989). Visual attentional
demand of an in-car navigation display system. In A.G. Gale (Ed.), Vision in vehicles
II (pp. 307-316). London: Elsevier Science.

Zaidel, D.M., & Noy, Y.I. (1997). Automatic versus interactive vehicle navigation aids.
In Y. Ian Noy (Ed.), Ergonomics and safety of intelligent driver interfaces (pp. 287-
307). Mawaha, NJ: Lawrence Erlbaum Associates.

Zhao, Y. (1997). Vehicle location and navigation systems. Boston: Artech House.