ISO/TR 16352:2005

TECHNICAL ISO/TR
REPORT 16352
First edition
2005-12-01
Road vehicles — Ergonomic aspects
of in-vehicle presentation for transport
information and control systems —
Warning systems
Véhicules routiers — Aspects ergonomiques de la présentation des
systèmes de commande et d'information des transports à l'intérieur des
véhicules — Systèmes avertisseurs

Reference number
©
ISO 2005
PDF disclaimer
This PDF file may contain embedded typefaces. In accordance with Adobe's licensing policy, this file may be printed or viewed but
shall not be edited unless the typefaces which are embedded are licensed to and installed on the computer performing the editing. In
downloading this file, parties accept therein the responsibility of not infringing Adobe's licensing policy. The ISO Central Secretariat
accepts no liability in this area.
Adobe is a trademark of Adobe Systems Incorporated.
Details of the software products used to create this PDF file can be found in the General Info relative to the file; the PDF-creation
parameters were optimized for printing. Every care has been taken to ensure that the file is suitable for use by ISO member bodies. In
the unlikely event that a problem relating to it is found, please inform the Central Secretariat at the address given below.

©  ISO 2005
All rights reserved. Unless otherwise specified, no part of this publication may be reproduced or utilized in any form or by any means,
electronic or mechanical, including photocopying and microfilm, without permission in writing from either ISO at the address below or
ISO's member body in the country of the requester.
ISO copyright office
Case postale 56 • CH-1211 Geneva 20
Tel. + 41 22 749 01 11
Fax + 41 22 749 09 47
E-mail copyright@iso.org
Web www.iso.org
Published in Switzerland
ii © ISO 2005 – All rights reserved

Contents Page
Foreword. v
Introduction . vi
1 Scope. 1
2 Warning signals. 2
2.1 Criteria of warning effects . 2
2.2 Categorization of warning signal failure . 4
2.3 Urgency mapping. 6
2.4 Alarm theories. 8
2.5 Design recommendations. 10
3 Psychological and physiological aspects. 10
3.1 Human processing of warnings . 10
3.2 Workload. 12
3.3 Expectancy. 13
3.4 Further human factors, individual differences . 13
3.5 Recommendations. 14
4 Sensorial modality. 15
5 Visual warning signals . 17
5.1 Psychological/physiological bases. 18
5.2 Types of visual displays . 19
5.3 Design parameters. 20
5.3.1 Sensorial-related parameters. 20
5.3.2 Coding parameters. 24
5.3.3 Organizational parameters. 37
6 Auditory warnings . 39
6.1 Psychological/physiological bases. 40
6.2 Advantages of auditory presentation . 41
6.3 Tonal signals, auditory icons . 42
6.3.1 Advantages of tonal signals . 42
6.3.2 Standards. 42
6.3.3 Attributes. 42
6.3.4 Sensorial parameters. 44
6.3.5 Coding parameters. 47
6.3.6 Organizational parameters. 54
6.4 Speech output. 55
6.4.1 Advantages of speech output . 56
6.4.2 Sensorial-related parameters. 56
6.4.3 Coding parameters. 60
6.4.4 Organizational parameters. 63
6.4.5 Warning applications of speech output . 65
6.5 Comparison of tonal signals and speech output . 66
7 Tactile warnings. 68
7.1 Advantages of tactile presentation. 68
7.2 Design parameters. 70
7.2.1 Sensorial-related parameters. 70
7.2.2 Coding parameters. 71
8 Redundancy of message presentation. 71
8.1 Visual/auditory combination. 72
8.2 Visual/auditory qualities for in-vehicle displays . 73
8.3 Visual/auditory indications for displays . 74
8.4 Visual/auditory/tactile combination . 74
8.5 Master alerting. 75
8.6 Other concepts. 75
9 Comparison of warning types, codes and modalities. 77
9.1 Visual/auditory presentation of non-verbally-coded objects . 77
9.2 Visual/auditory presentation of verbally-coded objects/abstract information. 78
9.3 Visual/auditory presentation of verbally-/non-verbally-coded spatial information. 79
9.4 Visual presentation of non-verbally-coded information/auditory presentation of verbally-
coded information. 81
9.5 Visual/tactile presentation of non-verbally-coded objects/spatial information. 82
9.6 Auditory/tactile presentation of non-verbally-coded objects/spatial information . 85
9.7 Visual/auditory/tactile presentation of verbally-coded objects/abstract information. 86
9.8 Recommendations for warning systems. 86
10 Warnings in assistance systems. 88
10.1 Distance warning systems. 91
10.2 Collision warning systems. 92
10.3 Side-obstacle warning systems. 98
10.4 Lane-departure warning systems. 100
10.5 Manoeuvring aids for low speed operation. 102
10.6 Usability of intelligent-transport-systems information for drivers . 104
10.7 Other assistance systems. 104
11 Warnings in other applications. 105
11.1 Aircraft. 105
11.2 Intensive care unit. 106
11.3 Industrial plants . 106
12 Discussion. 108
13 Summary. 110
13.1 Introduction . 110
13.2 Warning signals. 110
13.3 Psychological and physiological aspects, sensorial modality . 111
13.4 Visual warning signals . 111
13.5 Auditory warnings. 113
13.6 Tactile warnings. 115
13.7 Redundancy of message presentation . 116
13.8 Comparison of warning types, codes and modalities. 116
13.9 Warnings in assistance systems. 117
13.10 Warning in other applications. 118
Bibliography . 119

iv © ISO 2005 – All rights reserved

Foreword
ISO (the International Organization for Standardization) is a worldwide federation of national standards bodies
(ISO member bodies). The work of preparing International Standards is normally carried out through ISO
technical committees. Each member body interested in a subject for which a technical committee has been
established has the right to be represented on that committee. International organizations, governmental and
non-governmental, in liaison with ISO, also take part in the work. ISO collaborates closely with the
International Electrotechnical Commission (IEC) on all matters of electrotechnical standardization.
International Standards are drafted in accordance with the rules given in the ISO/IEC Directives, Part 2.
The main task of technical committees is to prepare International Standards. Draft International Standards
adopted by the technical committees are circulated to the member bodies for voting. Publication as an
International Standard requires approval by at least 75 % of the member bodies casting a vote.
In exceptional circumstances, when a technical committee has collected data of a different kind from that
which is normally published as an International Standard (“state of the art”, for example), it may decide by a
simple majority vote of its participating members to publish a Technical Report. A Technical Report is entirely
informative in nature and does not have to be reviewed until the data it provides are considered to be no
longer valid or useful.
Attention is drawn to the possibility that some of the elements of this document may be the subject of patent
rights. ISO shall not be held responsible for identifying any or all such patent rights.
ISO/TR 16352 was prepared by Technical Committee ISO/TC 22, Road vehicles, Subcommittee SC 13,
Ergonomics applicable to road vehicles.
Introduction
From a task/function analytic perspective, the task of driving is composed of three major interlinked categories
of activity (Hancock and Parasuraman, 1992):
a) vehicle control;
b) navigation;
c) collision avoidance.
Each of these functions contribute to the overall workload imposed on the driver. Even under routine, low-
traffic conditions, the driver must co-ordinate several tasks together and, generally, can do so quite efficiently.
Many of these task components become highly automatized with practice, so that under normal driving
conditions the demands of divided attention on the drivers will generally be within the limits of their attentional
capacity. However, during more demanding traffic situations, for example, when traffic density increases or at
intersections or traffic roundabouts, divided attention demands may sometimes exceed a driver's capabilities.
The driver has to deal with a lot of information which has different situation-dependent priorities and which is
more or less expected by the driver. Highly demanding situations are characterized by high time and spatial
density or by an extended spatial range of information. Parts of the information are natural and parts are
coded within or outside the vehicle. While receiving, processing and reacting to the information, the driver can
be overtaxed, which results in critical driving situations with increased accident probability.
This is the motivation to support the driver with assistance systems. The degree of assistance available
seems likely to increase considerably over the coming years. Assistance systems can, for example, control
speed and distance between vehicles and vehicle position in relation to the road. They not only aim to
optimize driver strain and increase driving safety, but also to achieve maximum driver acceptance. For
example, the S.A.N.T.O.S system is a (adaptive) driver-assistance system which integrates systems like
active cruise control (ACC), heading control (HC), navigation, telephone and radio (Weiße et al., 2002).
Most of these assistance systems announce any abnormal or dangerous state of the car or the driving
environment to the driver and require a relatively quick reaction by the driver. These systems warn the driver
and convey an appropriate message to the driver. So, with an increasing number of assistance systems, more
respective warnings are expected. These warnings need to be designed individually and with respect to their
interrelation.
vi © ISO 2005 – All rights reserved

TECHNICAL REPORT ISO/TR 16352:2005(E)

Road vehicles — Ergonomic aspects of in-vehicle presentation
for transport information and control systems — Warning
systems
1 Scope
This Technical Report provides a literature survey about the human-machine interface of warning systems in
vehicles, including studies of ISO/TC 22/SC 13/WG 8 and ISO/TC 204/WG 14. It covers the experimental
experiences about the efficiency and acceptance of different modalities and combinations of warnings, and
the design of the sensorial, code and organizational parameters of visual, auditory and tactile warnings (as
well as concluding recommendations). The survey should initialize standardizing activities of ISO working
groups, e.g. ISO/TC 22/SC 13/WG 8.
This literature survey comprises the human-machine interface issues of warning systems in automobiles. The
discussion of warning signals in general is dealt with in Clause 2 and concerns the definition of warning
signals, their failure and urgency aspects. Alarm theories are briefly dealt with here. The basic psychological
and physiological aspects of warnings in vehicles are the subject of Clause 3. Some issues of human
behaviour, which are relevant to handling warnings, are described.
Due to their importance, the sensorial modalities are introduced separately in Clause 4. Auditory and tactile
presentations are becoming more and more important, which is reflected in the structure of the next three
Clauses 5, 6 and 7. The specific psychological and physiological bases, benefits and types of displays for
each sensory modality are presented in these clauses. Clause 5 is dedicated to visual warning displays with a
few examples of the sensorial-related parameters. Symbols, icons and text are discussed extensively. Other
coding and organizational features are handled as far as warning signals are affected (colour, blinking,
structures, etc.).
Clause 6 is dedicated to auditory warning displays. The basic differences and the respective benefits of tonal
signals, auditory icons and speech output are explained. This is the largest clause because of its significance
for oncoming information and warning systems in cars. The display parameters, which are particularly relevant
for auditory warning signals, are presented in more detail, i.e. startling effect, temporal and spatial
characteristics. The new auditory icons are elaborated more in detail because of their relevance for collision
warning systems. The sensorial, coding and organizational parameters of speech output are described in a
comprehensive manner.
Clause 7 is dedicated to tactile warning displays. Although the potential of tactile warnings has been clearly
demonstrated, data for their design is very scarce.
The redundant presentation of warnings is described in Clause 8. The experimental results of different
visual/auditory combinations are presented, as well as visual/auditory/tactile combinations. The possible
transfer of master alertings from the avionic environment into the automobile environment is discussed. Other
concepts like the graded sequence of warnings are included.
The experimental results with different warning signals and their combinations are presented in Clause 9 with
respect to type, code and modality of the warnings. The benefits of visual, auditory and tactile warnings
depend on whether objects, spatial relations or abstract information are transmitted verbally or non-verbally. A
series of field experiments with symbolic, written, tonal, spoken and tactile warnings are reported.
Clause 10 includes some of the assistance systems that have just been introduced, such as distance warning
systems, or that are about to be introduced, such as side-obstacle warning systems. All of these are relatively
time-critical and need carefully designed warnings with a particular emphasis on auditory and tactile displays.
The recent experimental results are cited.
In Clause 11, warning systems in other domains, especially avionics, are described. The extensive
experiences with the problems of several time-critical alarms in aircrafts as well as the flood of alarms in
power plants will be exemplified.
Clause 12 is dedicated to the discussion of the previous clauses and their relevance for warnings in vehicles.
Drivers are assisted in highly demanding driving situations by technical systems. There will be
more assistance systems in the near future with appropriate warnings for the driver. Not all
warnings will be a priori appropriate. Guidance from this study will help ensure they are
“appropriate”. The scope of this Technical Report is to survey the literature about the human-
machine interface of warning systems. It includes papers about the efficiency and acceptance of
different modalities and combinations of warnings and the design of the sensorial, coding and
organizational parameters of visual, auditory and tactile warnings.
2 Warning signals
2.1 Criteria of warning effects
The word “warning” implies a range of levels from simple situation indications to more imperative warnings
and commands directed toward the driver to perform a certain task (ISO/TC 204/WG 14, Komoda and Goudy,
1995).
There are several technical processing stages of warnings (Kopf, 1998):
⎯ detection of object, reading of sensor data, filtering;
⎯ recognition of situation;
⎯ evaluation of situation;
⎯ output of warning.
Warnings are designed to provide someone, exposed to that product or situation, with information in addition
to that which that person could reasonably be expected to possess. The designer is trying in some way to
influence the behaviour of the recipient of the warning. This could mean preventing someone from doing
something that he or she otherwise might have done, or it could mean getting him or her to do something that
might otherwise have been omitted. The receiver of the warning then has the task of deciding whether the
advantages in complying with the warning outweigh the costs of doing so.
An emergency signal paradigm is usually one where two components are operating in tandem. The first
component consists of a mechanical device that uses sensor logic to determine if and when to trigger a signal
(Getty et al., 1995). It involves proper setting of the sensor decision threshold. If the criterion is set too strictly,
false signals will be minimized, but there is the possibility that dangerous situations will go unsignalled. If the
criterion is set too leniently, fewer dangerous situations will go unsignalled (missed signals), but the false
signal rate will rise. The solution to this dilemma requires designing the physical components of the system to
optimize the trade-off between minimized false signals and maximized sensitivity.
The second component of an emergency signal response paradigm is the human operator, who is responsible
for detecting, evaluating and responding (or not responding) to the signal that is generated by the sensor-
based signalling system. Consideration of the second component is necessarily a more complex process than
manipulating the first component, due to the cognitive and perceptual processes of the human operator.
One has to differentiate between behaviour that occurs naturally in the relevant situation without a warning
necessarily being present, and the ‘added value’ that the warning might bring. The particular effect the
warning will have has to be known, so that the relative effects of different warning variables on compliance
can be assessed. The distinction between amount of compliance with and without the warning is crucial.
2 © ISO 2005 – All rights reserved

Warnings are artefacts. They are representations of the situations to which they refer. Most warnings serve
two functions: the alerting function and the informing function. The alerting function is somewhat abstract,
being emotive or motivational or both. The informing function is more explicit. For example, an auditory
warning may be overwhelmingly alerting, but contain no information at all beyond the fact that something has
gone wrong. Vice versa, a warning text may contain a minimal alerting effect, but may contain lots of
information.
Also relevant is the knowledge of the situation in which the warning occurs. Together the factors that have an
alerting function can be seen as the iconic aspects of warning. Such aspects act almost instantaneously and
require little conscious information processing. Generally one of the aims of a warning is to produce a rapid
alerting response which is appropriate to the product or situation. The alerting function results from more than
just, for example, the signal word, but results from the entire warning-in-context.
A warning is rated information (Kopf, 1998). A good warning should include:
⎯ an element which attracts the attention;
⎯ a reason for the warning;
⎯ the consequences if the warning is not observed;
⎯ instruction for actions.
There are different false warnings (Kopf, 1998, see 2.2):
⎯ time-dependent false warning: too early, too late;
⎯ logical false warning: no warning in critical situation and vice versa;
⎯ qualitative false warning: too many, too few, too strong, too weak.
Figure 1 shows the remaining time as a function of time when the warning is successful, not necessary or too
late, which results in an accident, depending on the moment of the driver's reaction.
To test the efficiency and impairment of warning systems, the following aspects have to be considered (Breuer
et al., 1994):
⎯ impairment (e.g. startling effect);
⎯ reliable detection and identification (conspicuousness, clearness);
⎯ transformation in safe behaviour.

Key
X time
Y remaining time
1 time of warning
2 driver's reaction
3 unnecessary warning
4 successful warning
5 accident
Figure 1 — Time aspects of warning (Kopf, 1998)
2.2 Categorization of warning signal failure
Pritchett (1997) investigated the pilot's non-conformance to alerting systems (see 11.1). Pilot´s non-
conformance changes the final behaviour of the system and therefore may reduce actual performance from
that anticipated.
The pilots' perceived need to confirm the alerting system's commands may involve several factors, including
the following.
⎯ The pilot may be concerned that the alerting system will fail to act as it should.
⎯ The pilot may feel the alerting system cannot consider relevant information or has different objectives.
⎯ The pilot may place greater confidence in his own decisions than in the decisions of the alerting system.
This pilot´s non-conformance can be associated with the following reasons for warning signal failures.
⎯ False signals: In theory, most design and training for emergency signals is based on the assumption that,
when presented, the signal is authentic and thus heeded. However, false signals may result as the
product of an over-sensitive sensor system (conservative decision criteria) (Getty et al., 1995). In many
cases, a given signal may be correctly generated based on a threshold violation, but may be invalid or
insignificant given the specifics of the operational situation. Such inappropriate signals may create a
nuisance that diverts operator attention. Elimination of all false signals is ideal, but attempts to achieve
that goal by altering sensor detection decision criteria can lead to overly strict detection systems that fail
to signal true emergencies. Instead, it is the responsibility of the human being to make the appropriate
response decision.
4 © ISO 2005 – All rights reserved

When the alerting system is designed to prevent catastrophic events in the avionics, variance in the
sensor measurements and unpredictability in the system dynamics requires its reasoning to be
conservative (Pritchett, 1997). While a conservative design helps ensure prompt, adequate reactions to
dangerous situations, it also increases the frequency of false alarms and excessive commands from the
alerting system. Although the alerting system is performing to specifications, false alarms may appear to
the pilot as failures of the system.
⎯ Missing signals: Failures of signalling systems may take another form. Instead of generating spurious
signals, they may fail to inform about legitimate danger. In many of these cases, the problem may be
related to the first component of the signalling system: the mechanical sensor (Usher 1994). If the
sensor's decision criterion (tolerance level) is set too strictly, then the sensor may fail to signal developing
crises, or it may wait too long before warning the operator. The deactivation of a signalling system (and
accompanying missed signals) are often the result of operator mistrust, caused by frequent false alarms.
The direct effects of these failures in the avionics can lead to very high costs; for example, in the case of
a collision-avoidance system, this type of failure can have catastrophic results (Pritchett, 1997). First, if
the pilots are not confident that the alerting system will generate an alert when required, they may feel
compelled to assess the situation regularly independent of the alerting system. Second, if the pilots feel
the commanded resolution to the hazard is insufficient, they may feel compelled to make their own
decisions about a resolution to the hazard, or they may execute a more severe version of the
commanded resolution.
⎯ Multiple signals: A third problem associated with signalling systems is the generation of multiple signals
that require prioritization, or worse, that contradict each other (Bliss and Gilson, 1998). Arrays of multiple
alarms can be problematic, because operators are typically not trained to prioritize them in any given
manner. This problem can be addressed by utilising an urgency mapping technique (Hellier et al., 1993).
This technique involves manipulating aspects of an alarm stimulus to increase the perceived urgency of
the signal.
⎯ Different situation perception: The pilots’ desire to confirm alerting system commands is a perception
that, while the alerting system is functioning to its specifications, these specifications do not include all
relevant information or have the same objectives as the pilots. For example, pilots indicated in a survey
that they sometimes do not follow Traffic Collision-Avoidance System (TCAS) commands — or turn them
off — in conditions where they have visual contact with the other aircraft or have knowledge of the other
aircraft's intentions through Air Traffic Communication (ATC) (Pritchett, 1997). When pilots have a high
confidence in their own reasoning and a low confidence in the alerting system‘s reasoning, they are more
likely to act upon their own reasoning and to confirm automatic commands.
So, one of the most likely reasons why users do not comply is that the perceived benefits are not outweighed
by the perceived costs of compliance. Warnings are usually used where there are risks, and in such situations
there will be both benefits and costs involved in complying with the warning. The situation in which the
warning occurs will be assessed using
⎯ previous knowledge,
⎯ natural cues from the situation or product, and
⎯ information from the warning.
It could also be influenced by the personality or mood of the recipient.
Information should be provided to the driver whenever a warning situation occurs. The driver should not have
to directly request information from the system, i.e. query the system (NHTSA, 1996).
The effects of warning signal failure may take many forms. False, missing and conflicting signals
may undermine confidence in system accuracy and reduce subsequent reliance and adherence.
Different situation perception by the driver can result in disregarding the warning signal. So, prior
to designing the warning output in a sophisticated manner, the mechanical warning device has to
be designed elaborately. Well chosen warning criteria are possibly more important than the ultimate
choice of specific details of the warning signal.
2.3 Urgency mapping
ANSI standards have made the following signal words standard for communicating hazard intensities (Laux
and Mayer, 1993; see 6.3.5.2):
⎯ DANGER: immediate hazard which will result in severe injury or death;
⎯ WARNING: hazard or unsafe practice which could result in severe injury or property damage;
⎯ CAUTION: hazard or unsafe practice which could result in minor injury or property damage.
This can be used as a general classification of signals in the car, which try to attract the driver's attention to
any hazardous state inside or outside the car. The communication function of a danger, warning, or caution
signal (subsumed here as “warning signals”) is to alert users to the presence of a latent hazard, to let them
know how hazardous it is, and to tell them what to do to avoid the hazard and what will happen if they do not
act appropriately. The statement of the hazard can be in speech, text format or in pictorial/symbolic form.
In the meanwhile, the alert signal has also been defined and classified in other standards. Table 1 shows the
definition and classification in MIL (Military) standard, and Table 2 and Table 3 show the one in ISO/TC 159.
The definition and classification are based on the criticality, urgency of the situation and the action to be taken.
Table 1 — Examples of definitions of alert signals in Military Standard (Aircrew Station Alerting
Systems)
Source Definitions of Alert signals
MIL-STD-411E Audio warning: Audio caution:
⎯ Indicates the existence of a
(1 March 1991) ⎯ Indicates the existence of a
particular impending
particular hazardous
dangerous condition,
condition, requiring
requiring attention, but not
immediate corrective action
necessarily immediate
action
Warning visual signal: Caution visual signal: Advisory visual signal:
⎯ Indicates the existence of a ⎯ Indicates the existence of a ⎯ Indicates a safe or
hazardous condition, condition, requiring normal configuration,
requiring immediate action immediate attention but not condition of performance,
to prevent loss of life, immediate action or operation of essential
equipment damage, or equipment or attracts
abortion of the mission attention and imparts
information for routine
action purpose
Table 2 — Examples of definitions of alert signal in ISO/TC159 (Ergonomics) (1)
Source Message categories
ISO 11429: 1996, Announcement/
Danger Caution Command All clear
information
Ergonomics —
System of auditory
Urgent action for Act when Need for Public instruction Danger past
and visual danger
rescue or necessary mandatory
and information
protection action
signals
6 © ISO 2005 – All rights reserved

Table 3 — Examples of definition of alert signal in ISO/TC159 (Ergonomics) (2)
Source Message categories
Visual danger signal Visual warning signal Visual emergency signal
ISO 11428:1996,
Ergonomics —
Visual signal indicating Visual signal indicating the Visual signal indicating the
Visual danger
imminent onset or actual imminent onset of a dangerous beginning or the actual
signals —General
occurrence of a dangerous situation requiring appropriate occurrence of a dangerous
requirements,
situation, involving risk of measure for the elimination or situation requiring immediate
design and testing
personal injury or equipment control of the danger action
disaster, and requiring some
human response to eliminate
or control the danger or
requiring other immediate
action
For warning signals, it is often difficult to differentiate between the iconic and the informational components
(see 2.1). In the case of auditory verbal warnings, this differentiation is usually clearer in that the sound has an
alerting function and also precise meaning, which may be known to the recipient. The urgency, as one
particular iconic feature of auditory warnings, should relate in some systematic way to the hazardousness or
risk of the referent. Warnings can be said to be appropriately mapped when the rank ordering of the urgencies
of the warnings is positively correlated with the rank order of the urgencies or importance of their associated
referents. Then, the recipients of the warning would know how quickly they should attend to the problem
signalled.
Studies with speech output have shown, that speeding up a stimulus makes it more urgent, as does raising its
pitch or making it louder (Momtahan, 1990, see 6.4.4.2). Warnings can be created which can be reliably
differentiated from one another in terms of their urgency.
Urgency mapping is also achievable with the iconic parts of a visual warning. For example, some colours have
stronger effects on our assessment of the likely level of risk and hazard involved (see 5.3.2.3).
In the following Figure 2, there are two aspects which require mapping. The first is the relation between the
iconic aspects of the warning and its perceived urgency. This is primarily a function of properties of the
warning itself. The second is the relationship between the objective risk and the subjective perception of that
risk. This will be affected both by prior knowledge of the procedure or situation and also by the informational
properties of the warning.
For crash-avoidance warnings Lerner et al. (1996) recommend that multiple imminent warnings should be
automatically prioritized in terms of severity and urgency (see 10.2). All warnings should be presented
simultaneously by means of a visual display. Only the highest priority warning in effect should be presented by
means of an acoustic or tactile display. A clearly distinguishable cue should be provided to the driver between
the termination of the highest priority imminent warning and initiation of the next highest priority warning. In the
case of directional warnings, the directional nature of the warning indication is sufficient to provide this cue.
The cited papers show the necessity of designing multiple warning systems with some sort of
urgency mapping. The iconic cues of visual and auditory warnings have to represent the level of
hazard with respect to other warnings presented at the same time. The sensorial modality has to be
carefully chosen to represent the urgency correctly.
Application of the management procedure, based on the prioritization of information contents and assignment
of suitable physical properties for information display, could improve the acquisition of presented information
especially if multiple information were given from ITS subsystems. For example, Uno et al. (2001) examined
the effects of information management in the situation when warning, route guidance and multimedia
information were simultaneously supplied. The results revealed that the management procedure assured
successful avoidance of a potentially dangerous event (rush out vehicle at blind intersection), though fewer
drivers could avoid collisions when the warning was presented without an applied management procedure.
Figure 2 — Components of warning compliance
2.4 Alarm theories
There are different alarm theories which help to understand the reaction of users to warnings.
⎯ Classical Signal Detection Theory, SDT: The Classical Signal Detection Theory (Green and Swets,
1966) has been utilized to examine the response to an alarm system. The theory states that detection of
a signal is dependent upon two general factors. The first factor (d') is a consideration of the physical
ability of the human being to detect a signal that is embedded in noise. The second factor (ß) takes into
consideration person-specific qualities, such as motivation and past experience with the signal detection
paradigm, that may affect the propensity of the human being to detect the signal.
In a signal detection paradigm, the above factors, d‘ and ß, work in tandem so that one of four possible
events may occur: a signal may be presented and detected (a hit); a signal may be presented, but not
detected (a miss); the detector may respond even though no signal is present (false alarm); and the
detector may refrain from responding when no signal is present (correct rejection). To determine detector
performance efficacy, receiver-operating characteristics (ROCs) can be computed. This metric analyses
hit rate (rate of correct detection) versus false alarm rate (rate of incorrect detection) for different criterion
levels (typically altered by varying pay-offs).
Although Signal Detection Theory provides an explanation of elements that influence detection of a signal,
it does not adequately account for the cognitive ones.
8 © ISO 2005 – All rights reserved
...