See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/289849041

Introduction: A human-centered design approach

Article · January 2011

CITATIONS READS

21 16,374

1 author:

Guy André Boy

CentraleSupélec
265 PUBLICATIONS 2,227 CITATIONS

SEE PROFILE

All content following this page was uploaded by Guy André Boy on 27 November 2016.

The user has requested enhancement of the downloaded file.

 Introduction  to  the  Handbook  of  Human-­‐Machine  Interaction.  G.A.  Boy  (Ed.),  Ashgate,  UK.  2011.  

Introduction
to Human-Machine Interaction:
A Human-Centered Design Approach

Guy A. Boy

Rationale
Nobody questions the use of the clock today: the main function of a clock is to provide the
time to its user. A modern watch uses several resources that include a battery, internal
mechanisms and the ability of its user to (re-)adjust time when necessary or change the
battery when it is no longer working. You interact with your watch as you would with
someone who will tell you “hey, it is time to go to the next meeting!” This automaton can be
programmed and consequently act as an agent that supports many time-related facets of your
life. More generally, automation brought up and consolidated the concept of human-machine
interaction (HMI).
HMI, as a field of investigation, is quite recent even if people have used machines for a long
time. HMI attempts to rationalize relevant attributes and categories that emerge from the use
of (computerized) machines. Four main principles, i.e., safety, performance, comfort and
esthetics, drive this rationalization along four human factors lines of investigation: physical
(i.e., physiological and bio-mechanical), cognitive, social or emotional.
Physically, the sailor interacts with his/her boat by pulling sail ropes for example.
Cognitively, I interact with my computer writing the introduction of this book. Of course I
type on a keyboard and this is physical, but the main task is cognitive in the sense that I need
to control the syntax and the semantics of my writing, as well as spelling feedback provided
by my text processing application. Software makes it more cognitive. You may say that the
sailor needs to know when and how to pull the ropes, and this is a cognitive activity. Indeed,
learning is required to optimize workload among other human factors. Socially, it happens
that my colleague and I wrote this text for a community of people. Any production, which is
targeted to a wider audience than its producer could anticipate, becomes a social production
that will need to be socially accepted. This is true for an engineering production, but also for a
legal act or an artistic production. Emotionally, the artist uses his/her pen or computer to
express his/her emotions. But, emotions may come from situations also where adrenalin is
required to handle risky decisions and actions. More generally, esthetics involves empathy in
the human-machine relation (Boy & Morel, 2004).
For the last three decades, cognition was central to the study of human-machine interaction.
This is because automation and software mediates most tasks. Hollnagel and Woods (2005),
talking about the growing complexity of interaction with increasingly computerized systems,
introduced this concept of changing balance between doing and thinking. But, what do we
mean by “doing” today when we permanently use computers for most of our everyday tasks.
Doing is interacting… with software! HMI has become human-computer interaction (HCI).
 Introduction  to  the  Handbook  of  Human-­‐Machine  Interaction.  G.A.  Boy  (Ed.),  Ashgate,  UK.  2011.  

However, in this book, HMI is more that HCI even if it includes it. Driving a car, flying an
airplane or controlling a chemical plant is obviously not the same as text processing. In this
sense, HMI is not the same as HCI (Hewett et al., 1992; Card et al., 1983; Myers, 1998; Sears
& Jacko, 2007).
HMI has become a mandatory field of research and engineering in the design and
development of nowadays systems. Why? As already said, this is because what is commonly
called a user interface is currently made of software… and this user interface has become
deeper and deeper! We now interact with machines through software. The main issue is to
develop systems that enable appropriate task execution through them. We often think that we
simplify tasks by piling layers of software, but it happens that resulting interaction is
sometimes more complicated and complex. Indeed, even if HCI strongly contributes to
decrease interaction complexity, the distance between people and the physical world increases
so much that situation awareness becomes a crucial issue, i.e., we must not loose the sense of
reality.
Therefore, what should we do? Should we design simpler systems that people would use
without extensive learning and performance support? To what extent should we accept the
fact that new systems require adaptation from their users? Where should we draw the line? An
important distinction is between evolution and revolution. For example, cars are getting more
computerized, e.g., in addition to the radio, new systems were introduced such as the global
positioning system (GPS) that supports navigation, an autopilot in the form of a speed control
system and a line keeping system, an onboard computer that support energy consumption, a
collision avoidance system, an hand-free kit that enables the driver to communicate with
people located outside of the vehicle, and so on.
Even if the initial goal was to improve safety, performance, comfort and esthetics, the result
today is that there are far too many onboard systems that increase driver’s workload and
induce new types of incidents and accidents. On one side, software technology attempts to
help people, and on the other side, it induces new types of life-critical problems that any
designer or engineer should take into account in order to develop appropriate solutions. A
simple alarm provided by the central software of a modern car may end up in a very
complicated situation because neither the driver nor a regular mechanic will be able to
understand what is really going on; a specialized test machine is required together with the
appropriate specialized person who knows how to use it.
This handbook proposes approaches, methods and tools to handle HMI at design time. For
that matter, it proposes a human-centered design (HCD) approach. Of course, it must be
considered as a starter toward a deeper search into the growing bulk of HMI and HCD
literature and field. It is based on contemporary knowledge and know-how on human-
machine interaction and human-centered design. It is targeted at a diverse audience including
academia and industry professionals. In particular, it should serve as a useful resource for
scholars and students of engineering, design and human factors, whether practitioners or
scientists, as well as members of the general public with an interest in cognitive engineering
(Norman, 1982, 1986), cognitive system engineering (Hollnagel & Woods, 1983, 2005) and
human-centered design (Boy, 1998). Above all, the volume is designed as a research guide
that will both inform readers on the basics of human-machine interaction, and provide a look
ahead at the means through which cognitive engineers and human factors specialists will
attempt to keep developing the field of HMI.
 Introduction  to  the  Handbook  of  Human-­‐Machine  Interaction.  G.A.  Boy  (Ed.),  Ashgate,  UK.  2011.  

Human-Centered Automation
Current machines heavily rely on the cognitive skills of their users, who acquire and process
data, make decisions and act in real-time. In particular, we will focus on the use of automated
machines and various kinds of innovative artifacts developed to overcome limitations of
humans facing the growing complexity of their overall environment. Automation will never
stop1. This is currently the case of the evolution of air traffic management (ATM) where
controllers have to face the difficult challenge of managing a growth of 4.5% per year average
for the last twenty years; this growth is very likely to remain the same during the next
decades. Machines are becoming more complex even if the goal of the designers is to
facilitate their use during normal operations, problems happens in routine as well as abnormal
contexts. This is why human reliability needs to be taken carefully from tow points of view:
(1) humans have limitations; (2) humans are unique problem-solvers in unanticipated
situations.
Cognitive engineering should better benefit from operational experience by proposing
appropriate models of cognition that would rationalize this experience. Rasmussen’s model
has been extensively used over the last two decades to explain the behavior of a human
operator controlling a complex dynamic system (Rasmussen, 1986). This model is organized
into three levels of behavior: skill, rule and knowledge (Figure 1).
Historically, automation of complex dynamic systems, aircraft in particular, have led to the
transfer of human operators’ skills (e.g., performing a tracking task) to the machine.
Autopilots have been in charge of simple tracking tasks since the 1930s. This kind of
automation was made possible using concepts and tools from electrical engineering,
mechanical engineering and control theories, such as mechanical regulators, proportional-
integral-derivative controllers (PID), Laplace functions and stochastic filtering. Autopilots
were deeply refined and rationalized during the sixties and the seventies. Human skill models
were based on quasi-linear models’ functions and optimal control models. Human engineering
specialists have developed quantitative models to describe and predict human control
performance and workload at Rasmussen's skill level. They have been successfully applied to
study a wide range of problems in the aviation domain such as handling qualities, display and
control system design and analysis, and simulator design and use.
The second automation revolution took place when the rule-based level was transferred to the
machine. In aviation, a second layer was put on top of the autopilot to take care of navigation.
The flight management system (FMS) was designed and implemented during the eighties to
provide set points for the autopilot. A database is now available onboard with a large variety

1
Bernard Ziegler, a former Vice-President of Airbus Industrie, made the following observations and
requirements from his experience as a test pilot and distinguished engineer: “the machine that we will be
handling will become increasingly automated; we must therefore learn to work as a team with automation; a
robot is not a leader, in the strategic sense of the term, but a remarkable operator; humans will never be perfect
operators, even if they indisputably have the capabilities to be leaders; strategy is in the pilot’s domain, but not
necessarily tactics; the pilot must understand why the automaton does something, and the necessary details of
how; it must be possible for the pilot to immediately replace the automaton, but only if he has the capability and
can do better; whenever humans take control, the robot must be eliminated; the pilot must be able to trust
automation; acknowledge that it is not human nature to fly; it follows that a thinking process is required to
situate oneself, and in the end, as humiliating as it may be, the only way to insure safety is to use protective
barriers.” (Ziegler, 1996).
 Introduction  to  the  Handbook  of  Human-­‐Machine  Interaction.  G.A.  Boy  (Ed.),  Ashgate,  UK.  2011.  

of routes that cover most of the flights in a specific geographical sector. Pilots need to
program the FMS by recalling routes from the database and eventually customize them for a
specific flight. Once they have programmed the FMS, the aircraft is “capable of flying by
itself” under certain conditions, i.e., the FMS is in charge of the navigation task in pre-
programmed situations.

Figure 1. Rasmussen’s model, automation evolution and contributing discipline emergence.

Today, human operators mostly work at Rasmussen’s knowledge-based level where
interpretation has become an important work process. Basic operations are delegated to the
machine, and humans progressively become managers of (networked) cognitive systems.
Humans need to identify a situation when there is no pattern matching (situation recognition)
at the rule-based level, to decide according to specified (or sometimes unspecified) goals, and
to plan a series of tasks. These are typical strategic activities. Some people are good at
strategic activities, others prefer to execute what they are told to do. In any case, the control of
cognitive systems requires strategic training. For example, using the Web has totally
transferred the shopping task to Rasmussen's knowledge-based level, i.e., the selection of food
items is made using virtual objects, and delivery is planned with respect to the customer’s
schedule and the nature of the food.
Technology has always contributed to shape the way people interact with the world.
Conversely, interacting with the world has direct impact on how technology evolves.
Rationalization of experience feedback influences the development of theories that make new
artifacts emerge. In a technology-driven society, this goes the other way around, i.e., the use
of artifacts induces new practices, and new jobs emerge, as film technology induced the art of
film making for example. The twentieth century was rich in technology innovation and
development. The speed of evolution of technology and resulting practices is very sensitive to
economical impacts. In some cases, when economical benefits were not obvious a priori, but
the evolution of human kind was at stake, technological advances were decided at the political
level, such as designing and developing a technology that enables a man to walk on the moon.
Today following these grandiose projects, we realize that human-centered automation, and
 Introduction  to  the  Handbook  of  Human-­‐Machine  Interaction.  G.A.  Boy  (Ed.),  Ashgate,  UK.  2011.  

more generally human-centered design and engineering, is not enough effectively taken into
account at the political level yet. A new paradigm needs to be found to better understand the
optimal allocation between human and machine cognition together with the evolution of
organizations.
The term “human-centered automation” (HCA) was coined by Billings (1991) in the aviation
domain, and, among a large variety of research efforts, further developed (Boy et al., 1995).
Human-centeredness requires that we focus on some distinctions. When it was conceived,
HCA differed from human-centered design (HCD) in the sense that automation is something
added to an existing system. Since software technology is dominant in the systems that we
develop today, HCA becomes necessarily HCD. But, I think that there is an even more
important distinction between HCD and human-centered engineering (HCE). HCD is the
mandatory upstream process that enables a design team to incorporate human requirements
into the design of a system. Usually, HCD is scenario-based and prototype-based. It consists
in gathering human factors issues from an appropriate community of users or, more generally,
actors who are anticipated to act on the system being designed. These actors may be direct
end-users but also maintainers who will have to repair the system in case of failure for
example. In this case, it is not only design for usability, but also design for maintainability. At
this stage, we need to investigate possible scenarios that make actors requirements as explicit
as possible. In the same way, and as architects do for the design of buildings, mock-ups need
to be developed in order to incrementally validate actors requirements (this is formative
evaluation). HCE follows HCD. Human factors engineers are required to check the various
steps of the production of the system. If HCD is creativity-driven, HCE is a systematic
process that is based on a body of rules that need to be applied. In aeronautics, HCE is now a
common practice and is formalized by official national and international regulatory
institutions, e.g., ISO2, ICAO3, IATA4 and EASA5. Examples of such rules are provided in
EASA CS.25-1302 (2004) and ISO 13407 (1999). In this book, we will mainly focus on
HCD, even if some of the chapters treat parts of currently-practiced HCE, and insist on the
fact that end-user expertise and experience should be used during the whole life cycle of any
artifact.

The AUTOS Pyramid

The AUTOS pyramid is a framework that helps rationalize human-centered design and
engineering. It was first introduced in the HCD domain as the AUTO tetrahedron (Boy, 1998)
to help relate four entities: Artifact (i.e. system), User, Task and Organizational environment.
Artifacts may be aircraft or consumer electronics systems, devices and parts for example.
Users may be novices, experienced personnel or experts, coming from and evolving in
various cultures. They may be tired, stressed, making errors, old or young, as well as in very
good shape and mood. Tasks vary from handling quality control, flight management,
managing a passenger cabin, repairing, designing, supplying or managing a team or an
organization. Each task involves one or several cognitive functions that related users must

2
International Standard Organization.
3
International Civil Aviation Organization.
4
International Air Transport Association.
5
European Aviation Safety Agency.
 Introduction  to  the  Handbook  of  Human-­‐Machine  Interaction.  G.A.  Boy  (Ed.),  Ashgate,  UK.  2011.  

learn and use. The AUT triangle (Figure 2) enables the explanation of three edges: task and
activity analysis (U-T); information requirements and technological limitations (T-A);
ergonomics and training (procedures) (T-U).

Figure 2. The AUT triangle.

The organizational environment includes all team players who/that will be called “agents”,
whether humans or machines, interacting with the user who performs the task using the
artifact (Figure 3). It introduces three additional edges: social issues (U-O); role and job
analyses (T-O); emergence and evolution (A-O).

Figure 3. The AUTO tetrahedron.

The AUTOS framework (Figure 4) is an extension of the AUTO tetrahedron that introduces a
new dimension, the “Situation”, which was implicitly included in the “Organizational
environment”. The three new edges are: usability/usefulness (A-S); situation awareness (U-
S); situated actions (T-S); cooperation/coordination (O-S).
HMI could be presented by describing human factors, machine factors and interaction factors.
Using AUTOS, human factors are user factors, machine factors are artifact factors, and
interaction factors combine task factors, organizational factors and situational factors. Of
course, there are many other ways to present this discipline, we choose the five AUTOS
dimensions because they have been proved to be very useful to drive human-centered design
and categorize HMI complexity into relevant and appropriate issues. Therefore, I use them to
structure this introduction of HMI. These aspects include design methods, techniques and
tools.
 Introduction  to  the  Handbook  of  Human-­‐Machine  Interaction.  G.A.  Boy  (Ed.),  Ashgate,  UK.  2011.  

Figure 4. The AUTOS pyramid.

Machine factors (the A of AUTOS)

Since this book is devoted to support designers and engineers in the design and development
of human-machine systems, technological aspects are important concrete bricks that will
enable them to perform their jobs. In this handbook, machine factors will not be developed
from an engineering viewpoint, but a usage viewpoint.

Hardware factors
Today, people at work typically face computer screens of various forms and use a variety of
control devices. We usually refer to the user interface, i.e., a system in between a user (or
human operator) and a process to be controlled or managed. The terms “human operator” and
“user” can be equally used. The former come from process control and the human-system
engineering community. The latter comes from the human-computer interaction (HCI)
community. In these two communities, automation took various forms and contents. In
process control, automation was driven by control theories where feedback is the dominant
concept. In HCI, (office) automation was driven by the desktop metaphor for a long time to
the point out that usability often refers to the ability to use a graphical interface that includes
menus, buttons, windows and so on. It is interesting to note that the process control discipline
took care of real-time continuous processes such as nuclear, aerospace or medical systems
where time and dynamics are crucial issues together with safety-critical issues. Conversely,
HCI developed the interaction comfort side. HCI specialists got interested into learnability,
efficiency, easy access to data, direct manipulation of metaphoric software objects and
pleasurable user experience. Human operators are typically experts because industrial
processes that they control are complex and safety-critical. Users, in the HCI sense, can be
anybody.
For all these reasons, hardware factors are different if we talk about process control or
“classical” HCI. The very nature of processes to be controlled needs to be analyzed and
understood well enough to determine what kind of hardware would be suitable for the safest,
most efficient and comfortable human-machine interaction. However, HCI strongly
influenced our lives during the last decade to the point that usability has become a standard
even in process control interface design and development. People are now very familiar with
menus and windows. This is fine, but this also assumes that these interaction styles cover the
various constraints of the process being controlled.
 Introduction  to  the  Handbook  of  Human-­‐Machine  Interaction.  G.A.  Boy  (Ed.),  Ashgate,  UK.  2011.  

I recently heard the term “interactive cockpit”. What does that mean? I always thought that an
aircraft cockpit was the ultimate interactive interface with a dynamic safety-critical process,
and therefore interactive. But “interactive” means something else today. It does not mean
continuous interaction through a steering wheel or a handle physically connected to rudders, it
means interacting with a piece of software… typically through a computer screen with a
pointing device and a keyboard! This (r)evolution started with the glass cockpits in the mid-
eighties; we were talking about “fly-by-wire”. Today, the car industry is talking about “drive-
by-wire”, but the meaning has also changed following this technological evolution where
software is the most important component.
There are hardware factors that incrementally emerge from the design of these new machines.
Among them, the following are important: force feedback, loudspeakers, screens, signals,
buttons, keyboard, joystick, mouse, trackball, microphone, 3D mouse, data glove, data suit (or
interactive seat), metaphor for interaction, visual rendering, 3D sound rendering, 3D
geometrical model and so on.
From the time of the first flying machines at the end of the nineteen-century to the Concorde,
the number of instruments in an aircraft cockpit grew up to 600. At the beginning of the
eighties, the introduction of cathode ray tubes (CRT) and digital technology in cockpits
contributed to drastically decrease this number. Today, the number of displays in the A380 is
considerably reduced. This does not mean that the number of functions and lines of software
code is reduced. As a matter of fact, software size keeps increasing tremendously.

Software factors
Software is very easy to modify; consequently we modify it all the time! Interaction is not
only a matter of end product; it is also a matter of development process. End-users are not the
only ones to interact with a delivered product; designers and engineers also interact with the
product in order to fix it up toward maturity… even after its delivery. One of the reasons is
that there are software tests that cannot be done without a real-world exposure. It is very
difficult to anticipate what end-users would do in the infinite number of situations where they
will be evolving. We will see in this book that scenario-based design is mandatory with
respect to various dimensions including understandability (situation awareness), complexity,
reliability, maturity and induced organizational constraints (rigidity versus flexibility).
What should we understand when we use a product? How does it work? How should it be
used? At what level of depth should we go inside the product to use it appropriately? In the
early ages of the car industry, most car drivers were also mechanics because when they had a
problem they needed to fix it by themselves; the technology was too new to have specialized
people. These drivers were highly skilled engineers both generalists and specialists on cars.
Today, things have drastically changed; drivers are no longer knowledgeable and skilled to fix
cars; there are specialists that do this job because software is far too complex to understand
without appropriate help. Recent evolution transformed the job of mechanics into system
engineers who know how to use specialized software that enables to diagnose failures and fix
them. They do not have to fully understand what is going on inside the engine, a software
program does it for them and explain problems to them; when the overall system is well-
designed of course. This would be the ideal case; in practice, most problems come from
organizational factors induced by the use of such technology, e.g., appropriate people may not
be available at the right time to fix problems when they arise.
 Introduction  to  the  Handbook  of  Human-­‐Machine  Interaction.  G.A.  Boy  (Ed.),  Ashgate,  UK.  2011.  

Software complexity can be split into internal complexity (or system complexity) and
interface complexity. Internal complexity is related to the degree of explanation required to
the user to understand what is going on when necessary. Concepts related to system
complexity are: flexibility (both system flexibility and flexibility of use); system maturity
(before getting mature, a system is an accumulation of functions —the “another function
syndrome”— and it becomes mature through a series of articulations and integrations);
automation (linked to the level of operational assistance, authority delegation and automation
culture); and operational documentation. Technical documentation complexity is very
interesting to be tested because it is directly linked to the explanation of artifact complexity.
The harder a system is to use, the more related technical documentation or performance
support are required in order to provide appropriate assistance at the right time in the right
format.
In any case, software should be reliable at any time in order to support safe, efficient and
comfortable work. There are many ways to test software reliability (Lyu, 1995; Rook, 1990).
In this handbook, what we try to promote is not only system reliability, but also human-
machine reliability. We know that there is a co-adaptation of people and machines (via
designers and engineers). Human operators may accept some unreliable situations where the
machine may fail as long as safety, efficiency and comfort costs are not too high. However,
when these costs become high enough for them, the machine is just rejected. Again this poses
the problem of product maturity (Boy, 2005); the conventional capacity maturity model for
software development (Paulk et al., 1995), systematically used in most industries, does not
guarantee product maturity, but process maturity. Product maturity requires continuous
investment of end-users in design and development processes. At the very beginning, they
must be involved with domain specialists to set up high-level requirements right; this is an
important role of participatory design. During the design and development phase, formative
evaluations should be performed involving appropriate potential end-users in order to
“invent” the most appropriate future use of the product.
Interface complexity is characterized by content management, information density and
ergonomics rules. Content management is, in particular, linked to information relevance,
alarm management, and display content management. Information density is linked to
decluttering, information modality, diversity, and information-limited attractors, i.e., objects
on the instrument or display that are poorly informative for the execution of the task but
nevertheless attract user’s attention. The “PC screen do-it all syndrome” is a good indicator of
information density (elicited improvement-factors were screen size and zooming).
Redundancy is always a good rule whether it repeats information for crosschecking,
confirmation or comfort, or by explaining the “how”, “where”, and “when” an action can be
performed. Ergonomics rules formalize user friendliness, i.e., consistency, customization,
human reliability, affordances, feedback, visibility and appropriateness of the cognitive
functions involved. Human reliability involves human error tolerance (therefore the need for
recovery means) and human error resistance (therefore the existence of risk to resist to). To
summarize, A-factors deal with the level of necessary interface simplicity, explanation,
redundancy and situation awareness that the artifact is required to offer to users.
 Introduction  to  the  Handbook  of  Human-­‐Machine  Interaction.  G.A.  Boy  (Ed.),  Ashgate,  UK.  2011.  

Human factors (the U of AUTOS)

Human factors have been heavily studied during the last five decades in the HMI context.
After the Second World War, human factors specialists were mainly physicians, medical
doctors, who were taking care of both physiological and biomechanical aspects of humans at
work, i.e., ergonomics6. Work psychology, and later on cognitive psychology, progressively
emerged in the seventies to become the leading discipline in human factors in the eighties.
The main reason of this emergence is the introduction of computers and software in work
places and, more generally, everyday life. All these approaches were essentially based on the
Newell and Simon’s information processing model that is typically a single agent model
(Newell & Simon, 1972). The development of computerized systems and more specifically
networked systems promoted the need for studying social and organizational factors. We have
then moved into the field of multi-agent models.
HMI involves automation, i.e., machines that were controlled manually are now managed
through a piece of software that mediates user intentions and provides appropriate feedback.
Automation introduces constraints, and therefore rigidity. Since end-users do not have the
final action, they need to plan more than in the past. As already said, work becomes more
cognitive and (artificially) social, i.e., there are new social activities that need to be performed
in order for the other relevant actors to do their jobs appropriately. This even becomes more
obvious when cognition is distributed among many human and machine agents. Computer-
supported cooperative work, for example, introduced new types of work practices that are
mandatory to learn and use, otherwise overall performance may rapidly become a disaster.

Human body-related and physiological factors

Work performed by people can be strongly constrained by their physiological and
biomechanical possibilities and limitations. Understanding these possibilities and limitations
tremendously facilitated the evolution of civil and military aviation. Cockpits were
incrementally shaped to human anthropometrical requirements in order to ease manipulation
of the various instruments. This of course is always strongly related to technology limitations
also.
Anthropometry developed its own language and methods. It is now actively used in design to
define workspaces according to human factors such as accommodation, compatibility,
operability, and maintainability by the user population. Workspaces are generally designed
for 90% to 95% coverage of the user population. Anthropometric databases are constantly
maintained to provide appropriate information to designers and engineers. Nevertheless,
designers and engineers need to be guided to use these databases in order to make appropriate
choices.
Work organization is also a matter of trouble for professionals. Fatigue is a major concern.
Therefore, it is important to know about circadian rhythms and the way people adapt to shift
work and long work hours for example. Consequences are intimately associated with health

6
Professor Grandjean declared founded the International Ergonomics Association (IEA) on April 6, 1959, at a
meeting in Oxford, England. Today, IEA is a hyper-organization that has 42 federated societies, 1 affiliated
society, 11 sustaining member organizations, 6 individual sustaining members and 2 networks. It also has 22
Technical Committees (http://www.iea.cc). IEA includes all forms of human factors at work.
 Introduction  to  the  Handbook  of  Human-­‐Machine  Interaction.  G.A.  Boy  (Ed.),  Ashgate,  UK.  2011.  

and safety risks. Fatigue studies provide more knowledge and knowhow on how to proceed
with work time schedules, appropriate training, systematic checks, and health indicators
following. Of course, this needs to be integrated in regulatory procedures.

Cognitive factors
Cognitive factors start with workload assessment. This statement may seem to be restrictive
and old fashion, but the reader should think twice about workload before starting any work in
human factors. On one side, workload is a concept that is very difficult to define. It is both an
output of human performance and a necessary input to optimize performance, i.e., we produce
workload to perform better, up to a point where we need to change our work strategy. But on
the other side, we need to figure out a model that would quantify a degree of load produced
by a human being while working. Of course, this model should be based on real
measurements performed on the human being. Many models of workload have been proposed
and used (Bainbridge, 1978; Hart, 1982; Boy & Tessier, 1985). Workload also deals with the
complexity of the task being performed. In particular, people can do several things at the
same time, in parallel; this involves the use of several different peripheral resources
simultaneously (Wickens, 1984). Sperandio (1972) studied the way air traffic controllers
handle several aircraft at the same time, and showed that the time spent on radio increased
with the number of aircraft being controlled: 18% of their time spent in radio communication
for one controlled aircraft whereas 87% for nine aircraft controlled in parallel. In other words,
task complexity tends to increase human operator efficiency.
Human-machine interaction moves into human-machine cooperation when the machine
becomes highly automated. In this case, it is more appropriate to talk about agent-agent
cooperation. Hoc and Lemoine studied dynamic task allocation (DTA) of conflict resolution
between aircraft in air-traffic control on a large-scale simulator. “It included three cognitive
agents: the radar controller (RC), in charge of conflict detection and resolution; the planning
controller (PC), in charge of entry-exit coordination and workload regulation; and a conflict
resolution computer system. Comparisons were made on the basis of a detailed cognitive
analysis of verbal protocols. The more the assistance, the more anticipative the mode of
operation in controllers and the easier the human-human cooperation (HHC). These positive
effects of the computer support are interpreted in terms of decreased workload and increased
shared information space. In addition, the more the controllers felt responsible for task
allocation, the more they criticized the machine operation” (Hoc & Lemoine, 1998).
Situation awareness (SA) is another concept that is useful to introduce here, especially as a
potential indicator for safety in highly automated human-machine systems. During the last
decade, lots of efforts have been carried out to assess SA such as the Situation Awareness
Global Assessment Technique (SAGAT) (Endsley, 1987, 1996). Several efforts have been
developed to assess SA in the aeronautics domain (Mogford, 1997; Stephane & Boy, 2005);
the main problem is the characterization of the influence of action on situation awareness.
Indeed, human operator’s actions are always situated, especially in life-critical environments,
and SA does not mean the same when actions are deliberate as when they are reactive. In
human-machine interaction, this is a very important issue since actions are always both
intentional (deliberative) and reactive because they are mainly performed in a close loop.
Obviously, there are many other concepts and processes that need to be taken into account to
investigate cognitive factors. I would like to insist on the fact that cognition must be thought
 Introduction  to  the  Handbook  of  Human-­‐Machine  Interaction.  G.A.  Boy  (Ed.),  Ashgate,  UK.  2011.  

in a multi-agent perspective where human-machine interaction is in fact a dynamic network of

interactions among human and machine agents. Consequently, cognition should be taken
more broadly than in cognitive psychology and extend it to a social and organizational
perspective.

Social factors
There are two fields of research that grew independently for the last three decades: crew
resource management (CRM) in aviation, and computer-supported cooperative (CSCW) work
in HCI. The former was motivated by social micro-world of aircraft cockpits where pilots
need to cooperate and coordinate to fly safely and efficiently. CRM started during a workshop
on resource management on the flight deck sponsored by NASA in 1979 (Cooper, White, &
Lauber, 1980). At that time, the motivation was the correlation between air crashes and
human errors as failures of interpersonal communications, decision-making, and leadership
(Helmreich et al., 1999). CRM training developed within airlines in order to change attitudes
and behavior of flight crews. CRM deals with personalities of the various human agents
involved in work situations, and is mainly focused on teaching, i.e., each agent learns to better
understand his or her personality in order to improve the overall cooperation and coordination
of the working group.
Douglas Engelbart is certainly the most influential contributor to the development of the
technology that supports collaborative processes today. He invented the mouse and worked on
the ARPANET project in 1960s. He was among the first researchers who developed hypertext
technology and computer networks to augment intellectual capacities of people. The term
“computer-supported cooperative work” (CSCW) was coined in 1984 by Paul Cashman and
Irene Grief to describe a mutli-disciplinary approach focused on how people work and how
technology could support them. CSCW scientific conferences were first organized in the USA
within the ACM-SIGCHI7 community. Conferences on the topic immediately followed in
Europe and Asia. Related work and serious interest already existed in European Nordic
countries. During the late 1970s and even more during the 1980s, office automation was born
from the emergence of new practices using minicomputers. Minicomputers and
microcomputers were integrated in many places such as travel agencies, administrations,
banks and so on, to support groups and organizations. People started to use them interactively,
as opposed to using them in a batch mode. Single user applications such as text processors
and spreadsheets were developed to support basic office tasks. Several researchers started to
investigate the way people were using this new technology. Computer science is originally the
science of internal functions of computers (how computers work). With the massive use of
computers and their incremental integration in our lives, computer science has also become
the science of external functions of computers (how to use computers and what they are for).
We, computer and cognitive scientists, needed to investigate and better understand more how
people appropriate computers individually and collectively to support collaborative work.
Multi-disciplinary research developed involving psychologists, sociologists, education and
organization specialists, managers and engineers.
In parallel with these two fields of research, two others developed: human reliability (Reason,
1990) and distributed cognition (Hutchins, 1995). The former led to a very interesting

7
Association for Computing Machinery – Special Interest Group on Computer Human Interaction.
 Introduction  to  the  Handbook  of  Human-­‐Machine  Interaction.  G.A.  Boy  (Ed.),  Ashgate,  UK.  2011.  

distinction between two approaches of human reliability whether the focus is on the person or
the system. Each approach induces a quite different philosophy of error management from the
other. Reason developed what he called the Swiss cheese model (Reason, 1997). He stated
that we cannot change human limitations and capabilities, but the conditions in which humans
perform their tasks can be changed. Therefore, these conditions, which can be viewed as
technological and organizational constraints, should be clearly identified in order to create
defensive barriers against the progression of an unsafe act.
Distributed cognition was first developed to take into account the sharing of meaningful
concepts among various agents. Extending the phenomenological school of thought, agents
are considered as subjects and not objects. They have different subjectivity, and therefore they
need to adapt among each other in order to develop a reasonable level of empathy, consensus
and common sense sharing; this is what intersubjectivity is about: “The sharing of subjective
states by two or more individuals” (Scheff 2006). This line of research cannot avoid taking
into account intercultural specificities and differences. It is not surprising that most leaders of
such a field come from anthropology and ethnology. Obviously, the best way to better
understand interactions between cognitive agents is to be integrated in the community of these
agents. In the framework of human-machine systems, we extend the concept of distributed
cognition to humans and machines. The extension of the intersubjectivity concept to humans
and machines requires that we take into account end-users and designers in a participatory
way. To summarize, U-factors mainly deal with user’s knowledge, skills and expertise on the
new artifact and its integration.

Interaction factors (the TOS of AUTOS)

Task factors
Human-machine interaction is always motivated by the execution of a task. Therefore, the
way the task is organized and supported by the machine (prescribed task), and executed by the
human user (effective task) is crucial. Obviously the effective task, that is often called
“activity”, is different from the prescribed task.
Activity analysis could be defined as the “identification and description of activities in an
organization, and evaluation of their impact on its operations. Activity analysis determines (1)
what activities are executed, (2) how many people perform the activities, (3) how much time
they spend on them, (4) how much and which resources are consumed, (5) what operational
data best reflects the performance of activities, and (6) of what value the activities are to the
organization. This analysis is accomplished through direct observation, interviews,
questionnaires, and review of the work records. See also job analysis, performance analysis
and task analysis.” (Business Dictionary, 2009)
Task complexity involves procedure adequacy, appropriate multi-agent cooperation (e.g., air-
ground coupling in the aerospace domain) and rapid prototyping (i.e., task complexity cannot
be properly understood if the resulting activity of agents involved in it is not observable).
Task complexity is linked to the number of sub-tasks, task difficulty, induced risk,
consistency (lexical, syntactic, semantic and pragmatic) and the temporal dimension
(perception-action frequency and time pressure in particular). Task complexity is due to
operations maturity, delegation and mode management. Mode management is related to role
 Introduction  to  the  Handbook  of  Human-­‐Machine  Interaction.  G.A.  Boy  (Ed.),  Ashgate,  UK.  2011.  

analysis. To summarize, T-factors mainly deal with task difficulty according to a spectrum
from best practice to well-identified categories of tasks.

Organizational factors
Interaction is also influenced by the organizational environment that is itself organized around
human(s) and machine(s) in the overall human-machine system (HMS). More explicitly, an
HMS could be someone facing his/her laptop writing a paper; it could also be someone
driving a car with passengers; it could be an air traffic management system that includes
pilots, controllers and various kinds of aviation systems. People are now able to interact with
computerized systems or with other people via computerized systems. We recently put to the
front authority as a major concept in human-centered automation. When a system or other
parties do the job, or part of the job, for someone, there is delegation. What is delegated? Is it
the task? It is the authority in the execution of this task? By authority, we mean accountability
(responsibility) and control.
Organization complexity is linked to social cognition, agent-network complexity, and more
generally multi-agent management issues. There are four principles for multi-agent
management: agent activity (i.e., what the other agent is doing now and for how long); agent
activity history (i.e., what the other agent has done); agent activity rationale (i.e., why the
other agent is doing what it does); and agent activity intention (i.e., what the other agent is
going to do next and when). Multi-agent management needs to be understood through a role
(and job) analysis. To summarize, O-factors mainly deal with the required level of coupling
between the various purposeful agents to handle the new artifact.

Situational factors
Interaction depends on the situation where it takes place. Situations could be normal or
abnormal. They could even be emergencies. This is why we will emphasize the scenario-
based approach to design and engineering. Resulting methods are based on descriptions of
people using technology in order to better understand how this technology is, or could be,
used to redefine their activities. Scenarios can be created very early during the design process
and incrementally modified to support product construction and refinement.
Scenarios are good to identify functions at design time and operations time. They tend to
rationalize the way the various agents interact among each other. They enable the definition of
organizational configurations and time-wise chronologies.
Situation complexity is often caused by interruptions and more generally disturbances. It
involves safety and high workload situations. It is commonly analyzed by decomposing
contexts into sub-contexts. Within each sub-context, the situation is characterized by
uncertainty, unpredictability and various kinds of abnormalities. To summarize, situational
factors deal with the predictability and appropriate completeness (scenario
representativeness) of the various situations in which the new artifact will be used.

Overview of the handbook

Of course, hard decisions needed to be made on the main topics that are developed in this
handbook. In addition to this introduction and the conclusion, the book is organized into three
 Introduction  to  the  Handbook  of  Human-­‐Machine  Interaction.  G.A.  Boy  (Ed.),  Ashgate,  UK.  2011.  

parts (analysis; design and engineering; and evaluation) and twenty chapters. These chapters
include transversal perspectives on human-machine interaction, methods and tools for human-
centered design and engineering, and continuity and change in human-machine systems.
A handbook on human-machine interaction cannot avoid human operator modeling. Thierry
Bellet presents an account on analysis, modeling and simulation of human operator’s mental
activities. Even if he limits his illustrations to car driving, the various descriptions and
methods are applicable to other safety-critical systems. Most readers know what car driving is
in practice, therefore examples will be better understood.
Following up on human factors, situation awareness became a major topic over the last two
decades mainly because the sophistication of technology tends to increase the distance
between human operators and the actual work. Anke Popken and Josef Krems present the
relation between automation and situation awareness, and illustrative examples in the
automotive domain.
There are several aspects of human factors such as psychophysiology and performance that
are certainly very important to take into account at design time. Anil Raj, Margery Doyle and
Joshua Cameron present an approach and results that can be used in human-centered design.
They focus on the relationships between workload, situation awareness and decision
effectiveness.
For the last three decades, human reliability was a hot topic, mainly in aviation, and more
generally in life-critical systems. Christopher Johnson presents a very comprehensive
approach to human error in the context of human-machine interaction, and more specifically
in human-centered design.
Complex systems cannot be operated without operational support. Barbara Burian and Lynne
Martin present a very experienced account on operating documents that change in real-time.
Software enables the design of interactive documents, electronic flight bags, integrated
navigational maps, and electronic checklists for example. New technology enables the
integration of information from different sources and ease the manipulation of resulting data
in real-time.
Human-machine interaction was thought as a human operator interacting with a machine.
Today, the human-machine social environment changes toward multi-agent interactions. Guy
Boy and Gudela Grote describe this evolution and the mandatory concepts that emerge form
this evolution such as authority sharing and organizational automation.
Designing new systems involves the participation of several actors and requires purposeful
and socially acceptable scenarios. Scenarios are coming from stories that are incrementally
categorized. They are necessary for strategic design thinking. John Carroll presents the
scenario-based design approach that support human-centered design of complex systems.
Design is or must be a socially anchored activity. Complex socio-technical systems have to be
developed in a participative way, i.e., realistic stakeholders have to be involved at an early
stage of design, by developing the new system in actual contexts of use. Saadi Lahlou
presents a series of socio-cognitive issues that a design team should take into account to
design things for the real world.
Design is certainly a creative activity, but it has to be incrementally rationalized. Following
up his 1998 book, Guy Boy presents a new version of the cognitive function analysis of
human-machine multi-agent systems. He introduces general properties such as emerging
 Introduction  to  the  Handbook  of  Human-­‐Machine  Interaction.  G.A.  Boy  (Ed.),  Ashgate,  UK.  2011.  

cognitive functions, complexity analysis, socio-cognitive stability analysis, and flexibility

analysis. He insists on the fact that taking into account experience is a key factor in design,
and maturity is a matter of product testing, as well as practice evolution and emergence
identification.
Automated processes involve cooperation between humans and machines. Patrick Millot,
Frédéric Vanderhagen and Serge Debernard present several dimensions of human-machine
cooperation such as degree of automation, system complexity, and the richness and
complexity of the human component. They insist on the need for a common frame of
reference in cooperative activities, and on the importance of the authority concept.
David Navarre, Philippe Palanque, Célia Martinie, Marco Winckler and Sandra Steere
provide an approach to human-centered design in the light of the evolution of human-
computer interaction toward safety-critical systems. They emphasize the non-reliability of
interactive software and its repercussions on usability engineering. They introduce the
“Generic Integrated Modeling Framework” (GIMF) that includes techniques, methods and
tools for model-based design of interactive real-word systems while taking into account
human and system-related erroneous behavior.
Most of the chapters in this handbook deal with automation. It was important to address the
various properties of human-automation interaction that support human-centered design. Amy
Pritchett and Michael Feary present several useful concepts, which include authority (a
recurrent issue in our modern software intensive world), representations, interface
mechanisms, automation behavior and interface error states.
As already said, software is everywhere in human-machine systems nowadays. Jeffrey
Bradshaw, Paul Feltovich and Matthew Johnson present a variation of human-automation
interaction where automation is represented by artificial agents. The concept of artificial agent
emerged from the development of semi-autonomous software in artificial intelligence. It is
consistent with the concept of cognitive functions already presented.
Evaluation was for a long time the main asset of the human factors and ergonomics discipline.
Jean-Marc Robert and Annemarie Lesage propose, in two chapters, a new approach of
evaluation going from usability testing to the capture of user experience with interactive
systems. Interaction design has become one of the main issue, and consequently activity, in
the development of modern technology. Traditional design is typically done locally and
integration happens after. Using a user experience approach involves holistic design, i.e., the
product is taken globally from the start. This is what I call the evolution from the traditional
inside-out engineering approach to the new outside-in design approach.
An example of technique and tool for measuring human factors during design is eye tracking
(ET). Lucas Stephane presents the core ET research in the field of human-machine
interaction, as well as the human visual system to better understand ET techniques. Real-
world examples in the aeronautical domain are taken to illustrate these techniques.
Among the many factors useful to assess human-machine interaction, fatigue is certainly
mostly hidden because we tend to continue working even when we are tired, and extremely
tired. Philippa Gander, Curt Graeber and Greg Belenky show how the dynamics of fatigue
accumulation and recovery need to be integrated into human-centered design, more
specifically by introducing appropriate defenses. When operator fatigue can be expected to
have an impact on safety, systems being design should be conceived as resilient to human
operator fatigue in order to maintain acceptable human-machine interaction.
 Introduction  to  the  Handbook  of  Human-­‐Machine  Interaction.  G.A.  Boy  (Ed.),  Ashgate,  UK.  2011.  

People performance changes with respect to their age. Anabela Simões, Marta Pereira and
Mary Panou address older people’s characteristics and design requests to accommodate their
needs in order to ensure efficient, safe and comfortable human-machine interactions with
respect to context. They present the issue of safe mobility for older people and technological
solutions with the related advantages and inherent risks.
The various effects of culture and organization influence human-machine interaction. Don
Harris and Wen-Chin Lee present the influence of these effects on human errors. People do
not use systems in the same way when they come from different cultures. This pragmatic
aspect of human-machine interaction needs to be taken into account seriously in life-critical
systems in particular.
The Francophone school of ergonomics makes the difference between prescribed task and
activity (i.e., the effective task). Sometimes we believe that task analysis as a normative
approach will guaranty a straight human-centered way of designing systems. This assumes
that system boundaries are well-defined. In the real world this is not the case. Systems are
loosely coupled in the environment where they are working. They need to absorb change and
disturbance and still maintain effective relationships with the environment. Promoting this
kind of reflexion, Erik Hollnagel present the diminishing relevance of human-machine
interaction.
The conclusion of this handbook focuses on the shift from automation to interaction design as
a new discipline that integrates human and social sciences, human-computer interaction and
collaborative system engineering. For that matter, we need to have appropriate models of
interaction, context and function allocation.

References
Bainbridge, L. (1978). Forgotten alternatives in skill and workload. Ergonomics, 21, pp. 169-
185.
Bainbridge, L. (1987). Ironies of automation. In: J. Rasmussen, K. Duncan and J. Leplat,
Editors, New Technology and Human Error, Wiley, London.
Beaudouin-Lafon, M. (2004). AVI '04, May 25-28, 2004, Gallipoli (LE), Italy.
Billings, C.E. (1991). Human-centered aircraft automation philosophy. NASA Technical
Memorandum 103885, NASA Ames Research Center, Moffett Field, CA, USA.
Boy, G.A. & C. Tessier (1985). Cockpit Analysis and Assessment by the MESSAGE
Methodology. Proceedings of the 2nd IFAC/IFIP/IFORS/IEA Conf. on Analysis, Design and
Evaluation of Man-Machine Systems, Villa-Ponti, Italy, September 10-12. Pergamon Press,
Oxford, pp. 73-79.
Boy, G.A., Hollnagel, E., Sheridan, T.B., Wiener, E.L. & Woods, D.D. (1995). International
Summer School on Human-Centered Automation. EURISCO Proceedings. Saint-Lary,
Pyrénées, France.
Boy, G.A. (1998). Cognitive Function Analysis: Ablex Publishing, distributed by Greenwood
Publishing Group, Westport, CT, USA.
Boy, G.A. & Morel, M. (2004). Interface affordances and esthetics. (in French) Revue
Alliage, Edition du Seuil, Paris, France.
Bradshaw, J. (1997). Software Agents. MIT/AAAI Press, Cambridge, MA, USA.
 Introduction  to  the  Handbook  of  Human-­‐Machine  Interaction.  G.A.  Boy  (Ed.),  Ashgate,  UK.  2011.  

Business Dictionary (2009). http://www.businessdictionary.com/definition/activity-

analysis.html
Card, S.K., Moran, T.P. & Newell, A. (1983). The Psychology of Human–Computer
Interaction. Erlbaum, Hillsdale, ISBN 0-89859-243-7.
Degani, A. & Wiener, E. Procedures in complex systems: The airline cockpit. IEEE
Transactions on Systems, Man, and Cybernetics, 27, 3 (1997), 302-312.
EASA CS.25 1302 (2004). http://www.easa.eu.int/doc/Rulemaking/NPA/NPA_15_2004.pdf
Endsley, M.R. (1987). SAGAT: A methodology for the measurement of situation awareness
(NOR DOC 87-83), Hawthorne, CA, Northrop Corporation.
Endsley, M.R. (1996). Situation awareness measurement in test and evaluation. In: T.G.
O'Brien and S.G. Charlton, Editors, Handbook of Human Factors Testing and Evaluation,
Lawrence Erlbaum Associates, Mahwah, NJ, pp. 159–178.
Hart, S.G. (1982). Theoretical Basis for Workload Assessment. TM ADP001150, NASA-
Ames Research Center, Moffett Field, CA, USA.
Helmreich, R.L., Merritt, A.C., & Wilhelm, J.A. (1999). The evolution of Crew Resource
Management training in commercial aviation. International Journal of Aviation Psychology,
9(1), pp. 19-32.
Heylighen, F. (1996). The Growth of Structural and Functional Complexity during
Evolution. In The Evolution of Complexity, F. Heylighen & D. Aerts (eds.). Kluwer
Academic Publishers.
Hewett, T. T., Baecker, R., Card, S., Carey, T., Gasen, J., Mantei, M., Perlman, G., Strong,
G., and Verplank, W. (1992). ACM SIGCHI Curricula for Human-Computer Interaction.
New York: The Association for Computing Machinery. (ACM Order Number: S 608920),
http://sigchi.org/cdg/cdg2.html#2_1.
Hoc, J.M. (1988). Cognitive psychology of planning. London: Academic Press.
Hoc, J.M. & Lemoine, M.P. (1998). Cognitive Evaluation of Human-Human and Human-
Machine Cooperation Modes in Air Traffic Control. International Journal of Aviation
Psychology, Volume 8, Issue 1 January, pp. 1-32.
Hollnagel, E. (1993). Reliability of cognition: Foundations of human reliability analysis.
London: Academic Press.
Hollnagel, E. & Woods, D.D. (1983). Cognitive systems engineering: New wine in new
bottles. International Journal of Man-Machine Studies, 18, pp. 583-600.
Hollnagel, E. & Woods, D. D. (2005). Joint cognitive systems: Foundations of cognitive
systems engineering. Boca Raton, FL: CRC Press / Taylor & Francis.
Hutchins, E. (1995). How a Cockpit Remembers its Speeds. Cognitive Science, 19, 265-288.
ISO 13407 (1999). Human centered design process for interactive systems, TC 159/SC 4.
Lyu , M.R. (1995). Handbook of Software Reliability Engineering. McGraw-Hill publishing,
1995, ISBN 0-07-039400-8.
Mogford, R.H. (1997). Mental models and situation awareness in air traffic control. The
International J. Aviation Psychology 7 4, pp. 331–342.
Myers, B.A. (1998). A brief history of human–computer interaction technology. Interactions
5(2):44–54, 1998, ISSN 1072–5520 ACM Press .
 Introduction  to  the  Handbook  of  Human-­‐Machine  Interaction.  G.A.  Boy  (Ed.),  Ashgate,  UK.  2011.  

Newell, A., & Simon, H. A. (1972). Human problem solving. Englewood Cliffs, NJ: Prentice
Hall.
Nielsen, J. (1993). Usability engineering. Academic Press. London.
Norman, D.A. (1982). Steps toward a cognitive engineering: Design rules based on analyses
of human error. Proceedings of the Conference on Human Factors in Computing Systems
(CHI’82), Gaithersburg, Maryland, USA, pp. 378-382.
Norman, D. (1986). Cognitive engineering. In Norman, D., and Draper, S. (Eds.), User
centered system design. Lawrence Erlbaum Associates, Inc., Hillsdale, NJ.
Paulk, M.C, Weber, C.V., Curtis, B. & Chrissis, M.B. (1995). Capability Maturity Model,
The: Guidelines for Improving the Software Process. The SEI Series in Software Engineering,
Addison Wesley Professional.
Rasmussen, J. (1986). Information Processing and Human-Machine Interaction - An
Approach to Cognitive Engineering. North Holland Series in System Science and
Engineering, A.P. Sage, Ed.
Reason, J. (1990). Human Error. New York: Cambridge University Press.
Reason, J. (1997). Managing the risks of organizational accidents. Ashgate, London.
Rook, P. (Ed.) (1990). Software Reliability Handbook. Centre for Software Reliability, City
University, London, U.K.
Scheff, T.J. (2006). Goffman Unbound!: A New Paradigm for Social Science. Paradigm
Publishers, ISBN 978-1594511967.
Sears, A. & Jacko, J.A. (Eds.). (2007). Handbook for Human Computer Interaction (2nd
Edition). CRC Press, ISBN 0-8058-5870-9.
Sperandio, J.C. (1972). Charge de travail et régulation des processus opératoires. Le Travail
Humain, 35, 86 -98. English summary in Ergonomics, 1971, 14, pp. 571-577.
Stephane, L. & Boy, G.A. (2005). A Cross-Fertilized Evaluation Method based on Visual
Patterns and Cognitive Functions Modeling. Proceedings of HCI International 2005, Las
Vegas, USA.
Suchman, L.A. (1987). Plans and situated actions. The problem of human-machine
communication. Cambridge, England: Cambridge University Press.
Wickens, C.D. (1984). Processing resources in attention. In R. Parasuraman and D.R. Davies
(Eds.), Varieties of attention. London : Academic Press.
Ziegler, B. (1996). The flight control loop. Invited speech. Final RoHMI Network meeting.
EURISCO, Toulouse, France, September 28-30

View publication stats