Hollnagel, E. & Woods, D. D. (2005). Joint cognitive systems: Foundations of cognitive systems engineering. Boca Raton, FL: CRC Press / Taylor & Francis.

Preface

Deciding what to write in a preface is not as easy as it seems. If there is anything of substance to say about the subject matter, it should clearly be in the main text of the book rather than in the preface. And if there is nothing of substance to say, then why say it? Prefaces can therefore easily become the place where the authors can write their more personal views on matters large and small, express their gratitude to various people, and lament about the state of the world in general. Prefaces can also become the authors’ encomium for their own work. Publishers welcome that, since it can be put on the back cover or on the website, but common decency usually intervenes in time.

After this beginning there are pretty few options left for us to do, except to briefly reminisce on how this book came about. The collaboration between the two authors began around 1979 and soon led to the ideas that became formulated as cognitive systems engineering (CSE), first in an internal report in 1982, and later as a paper in the International Journal of Man-Machine Studies. About ten years later, in the beginning of the 1990s, we started talking about writing a book on CSE. One motivation was that the idea seemed to have caught on; another was that we by that time individually and together had produced a number of writings that both had developed the original ideas further and illustrated how they could be applied in practice.

The route from word to deed is, however, usually longer than initially hoped for. While the intention to write a book together was never abandoned, progress was painfully slow, partly because we both had entangled ourselves in too much interesting work, and partly because the much coveted opportunity to sit down together for a couple of weeks on a desert island never quite materialised.

In the end the book became a reality because we adopted a pragmatic solution – in good according with the ethos of CSE. Instead of writing one book together, the project was split into two parts, leading to two books that complement each other. While we appear as joint authors of either book, the order differs to reflect the fact that the first author in each case is the main responsible for the writing. This also helps to solve the practical problem that a single book would have been rather large. Furthermore, we had over the years developed different styles of working and writing, one being more contemplative, the other being more practical.

The end result is therefore not one but two books of which this is the first. Both books are entitled Joint Cognitive Systems with the subtitles indicating the specific focus – and therefore also what distinguishes them from each other. The subtitle of the present book is ‘Foundations of Cognitive Systems Engineering’, while the subtitle of the second book is ‘Patterns in Cognitive Systems Engineering’. The intention is that either book can be read independently, but that they also will complement each other by emphasising the theoretical and the practical aspects respectively. It is no requirement that they are read in a specific order. Yet if people after reading either would find it necessary to read also its counterpart, we would feel we had achieved our purpose.

Erik Hollnagel                                                                                 David D. Woods

 

Table of Contents

Preface
1 – THE DRIVING FORCES

The focus of CSE is how humans can cope with and master the complexity of processes and technological environments, initially in work contexts but increasingly also in every other aspect of daily life. The complexity of the current technological environment is not only something that must be mastered, but paradoxically also provides the basis for the ability to do so. This entangling of goals and means is mirrored in the very concepts and theories by which we try to understand the situation of humans at work. To set the context, this chapter gives an overview of the scientific developments of the 20th Century that have shaped our thinking about humans and machines.

Introduction
On Terminology
Computerisation and Growing complexity
Self-Reinforcing Complexity Cycle
Complexity and Unpredictability
Conspicuousness of the Human Factor
The Constraining Paradigm
Input-Output Models
The Shannon-Weaver Model Communication 
Model
Prototypical Information Processing
From Human-Machine Interaction to Joint Systems
The Cognitive Viewpoint
The Classical human-machine view
The Disintegrated View
Changing The Paradigm
Definition of a cognitive system
The Scope of CSE

2 – EVOLUTION OF WORK

The human use of technology has a long history. The uptake of technology in work, and the consequent transformation of work, can be described as going through a number of stages. An understanding of the major technological innovations and solutions in the 20th Century is helpful to understand how the nature of work has undergone radical changes, and to give an idea of the possibilities and problems that may lie ahead.

Technological Systems As Amplification
Examples of Amplification
Amplification of Control
Effects of Amplification
Amplification and interpretation
Tools and Prostheses
A Short History Of Human-Machine Interaction
The Conspicuousness of Cognition
Changing Balance between Doing and Thinking
Loss of Work-specific Information
The Law of requisite variety
Models of the Human as Controller
The Joint Cognitive System (JCS)
Control and Cognition
Disjoint and Joint Systems
Amplifying the Ability to Control

3 – THE BASICS OF A SCIENCE

It is a paramount characteristic of a scientific study that it is systematic. It must be based on a model, from which can be derived a classification scheme and a set of methods. For measurements, it is important to avoid that they are based on loosely formulated folk models. This chapter describes the basic principles behind CSE, as the study of how cognitive systems perform rather than of cognition as a mental process, and introduces the three main threads: Coping with complexity, use of artefacts, and joint cognitive systems.

Model – Classification – method
Requirements to measurements
Theory-driven Measurements
Theory-begging Measurements
The Meaning of Measurements
The Elusiveness Of Cognition
Cognition in the Mind
Cognition in the Wild
The Focus of CSE
Cognition and Context
Cognition and Control
Modelling Cognition and Context
Sequentiality in Reality and in Models
The Threads Of CSE
Coping With Complexity
Use of Artefacts
Joint Cognitive Systems
Merging the Threads

4 – COPING WITH COMPLEXITY

The basic issue for CSE is how to maintain control of a process or an environment. Both processes and environments are dynamic and therefore complex, and joint cognitive systems are striving to cope with this complexity. The coping takes place at the both the individual and organisational levels, the latter as in the design of work environments, of social structures, and of technological artefacts.

Introduction
About Coping
Sources of complexity
Losing Control
Lack of Time
Lack of Knowledge
Lack of Readiness or Preparedness (Competence)
Lack of Resources
Coping strategies
Information Input Overload
Information Input Underload
Designing For Simplicity
Simplicity-Complexity Trade-Off
Information Structuring
The Right Information …
… In the Right Form …
… At the Right Time
How Should the Interaction Be designed?
Designing For Complexity
Support for Coping
Time
Predictability
Summary

5 – USE OF ARTEFACTS

Humans have used artefacts since the invention of cuneiform writing, but the complexity of present-day computerised artefacts often create more problems than they solve. Artefacts can be either tools or prostheses, and the chapter develops these distinctions. While artefacts (technical or social) ostensibly are introduced to make life easier for people, the effect sometimes turn out to be the exact opposite.

.Introduction
Phenomenology of Coagency
Tools and Prostheses
Artefacts IN CSE
Range of Artefacts
Cognitive Artefacts
The Substitution Myth
Consequences of technology change
Traffic Safety
Typical User Responses to New Artefacts
Failure Modes of an Artefact
The Accidental User
User Models and Accidental Users

6 – JOINT COGNITIVE SYSTEMS

Humans and artefacts have traditionally been considered as separate systems that perforce must interact. CSE proposes that the human-artefact ensemble is seen as a joint cognitive system in its own right, and that design and analysis starts from this level. Joint cognitive systems can be defined recursively, and accentuates the necessity to provide clear definitions of the boundaries and the capabilities of the systems.

Introduction
On the Nature of System Boundaries
Automation and Joint cognitive systems
Degrees of Human-Machine Dependency
Automation Philosophies
The Left-Over Principle
The Compensatory Principle
The Complementarity Principle and Function Congruence
Summary
Ironies of Automation
Lessons of Automation
Function analysis and goal achievement
Goals and Means
Simple Test-Operate-Test-Exit (TOTE)
Recursive Goals-Means Descriptions

7 – CONTROL AND COGNITION

The main thrust of CSE is how control can be improved. This chapter discusses the essential issues in control, such as closed- and open-loop control, and how automation and technological support fit into that. CSE applies a basic cyclical model, which can be extended to account for multiple simultaneous layers of control. This is necessary to give a realistic description of how a joint cognitive system functions.

Introduction
Feedback and Control
Feedforward and Control
The Substance of Modelling cognition
Cognition without Context
Cognition in Context
Mental Models and the Law of Requisite Variety
COCOM Contextual Control Model
Model Constituents
Control Modes
ECOM
Extended Control Model
Tracking
Regulating
Monitoring
Targeting
ECOM Structure and Parameters
Interaction between Loops
Modelling the Loss of Control

8 – TIME AND CONTROL

Since CSE is about the control of dynamic processes and environments, time is of fundamental importance. Time plays a role in coping with complexity as well as in handling data and information. The chapter describes how time is an integral part of the CSE models, as well as the various techniques and strategies that can be used to overcome a possible shortage of time.

Orthodoxy in Modelling
Model Minimization
Two Neglected Issues
Modelling Control
Predictability
Available Time
The Modelling Of Time
Representation of Time in COCOM
Time and Control Modes
How To Enhance Control
Technological and Organisational Solutions
Time to Evaluate Events (TE)
Time to Select Actions (TS)
Time to Perform an Action (TP)
Available Time
Human Solutions
Time to Evaluate Events (TE)
Time to Perform an Action (TP)
Available Time
Conclusions

9 – CSE AND ITS APPLICATIONS

This chapter provides a brief overview of the actual and potential applications of CSE. These include work and task design, control room design, decision support, etc. The problems are, of course, well known, since they stem from the practical problems of working life and of human-machine systems. CSE provides a consistent approach to address these problems, as well as a set of methods with a common conceptual basis.

What should CSE Be About?
Extensions to Human Factors / Ergonomics
Design of work (Cognitive task design)
Making Work Easy
Making Work Safe
Problems in the Design of Work
The Forced Automaton Analogy
Control rooms
Adaptation
Decision support
The Last Words

BIBLIOGRAPHY

AUTHOR INDEX

SUBJECT INDEX

Back