User Behavior and Technology Development: Shaping Sustainable Relations Between Consumers and Technologies (Eco-Efficiency in Industry and Science)

USER BEHAVIOR AND TECHNOLOGY DEVELOPMENT

ECO-EFFICIENCY IN INDUSTRY AND SCIENCE VOLUME 20 Series Editor: Arnold Tukker, TNO-STB, Delft, The Netherlands Editorial Advisory Board: Martin Charter, Centre for Sustainable Design, The Surrey Institute of Art & Design, Farnham, United Kingdom John Ehrenfeld, International Society for Industrial Ecology, New Haven, U.S.A. Gjalt Huppes, Centre of Environmental Science, Leiden University, Leiden, The Netherlands Reid Lifset, Yale University School of Forestry and Environmental Studies, New Haven, U.S.A. Theo de Bruijn, Center for Clean Technology and Environmental Policy (CSTM), University of Twente, Enschede, The Netherlands

The titles published in this series are listed at the end of this volume.

User Behavior and Technology Development Shaping Sustainable Relations Between Consumers and Technologies

Edited by

Peter-Paul Verbeek University of Twente, Department of Philosophy, Enschede, The Netherlands and

Adriaan Slob TNO Built Environment and Geosciences, Delft, The Netherlands

In association with the sub-editors J.C. Brezet (part 3), W.A. Hafkamp (part 4), W.J.M. Heijs (part 1), and C.J.H. Midden (part 2)

A C.I.P. Catalogue record for this book is available from the Library of Congress.

ISBN-10 ISBN-13 ISBN-10 ISBN-13

1-4020-4433-X (HB) 978-1-4020-4433-5 (HB) 1-4020-5196-4 (e-book) 978-1-4020-5196-8 (e-book)

Published by Springer, P.O. Box 17, 3300 AA Dordrecht, The Netherlands. www.springer.com

Printed on acid-free paper

Cover photographs © Infinite Jeff / Jeff Berman All Rights Reserved © 2006 Springer No part of this work may be reproduced, stored in a retrieval system, or transmitted in any form or by any means, electronic, mechanical, photocopying, microfilming, recording or otherwise, without written permission from the Publisher, with the exception of any material supplied specifically for the purpose of being entered and executed on a computer system, for exclusive use by the purchaser of the work.

CONTENTS

PREFACE ............................................................................................... xv PART 1 Conceptual Frameworks for Analyzing Technology and Behavior Chapter 1 Adriaan Slob and Peter-Paul Verbeek TECHNOLOGY AND USER BEHAVIOR: An Introduction ...........................................................................................3 1. The intriguing interactions between technology and behavior ..................3 2. The need for an integrated approach ........................................................5 3. Theoretical perspectives ..........................................................................8 4. Research questions and outline of the book............................................ 11 References ................................................................................................12 Chapter 2 Albert G. Arnold and Petra Mettau ACTION FACILITATION AND DESIRED BEHAVIOR....................13 1. Introduction ..........................................................................................13 2. Model of technology-human interaction ................................................14 3. The concepts of ‘action facilitation’ and ‘affordance of an artifact’.........................................................................................15 4. Action theory ........................................................................................16 5. Optimization of action efficiency ..........................................................19 References ................................................................................................20

vi

Contents

Chapter 3 Adrienne van den Bogaard and Paul Swuste SAFETY: Technology and Behavior?................................. ......................................21 1. Introduction ..........................................................................................21 2. A short characterization of Safety Science .............................................22 3. Four causal safety factors ......................................................................24 4. Interaction between technology and behavior ........................................28 5. Conclusions ..........................................................................................30 References ................................................................................................31 Chapter 4 J.P. Groot-Marcus, P.M.J. Terpstra, L.P.A. Steenbekkers and C.A.A. Butijn TECHNOLOGY AND HOUSEHOLD ACTIVITIES ...........................33 1. Introduction ..........................................................................................33 2. Developments in Household Sciences....................................................34 3. Consumer-Technology Interaction Model..............................................35 4. Technological influences and responses.................................................37 5. Discussion and recommendations ..........................................................39 References ................................................................................................41 Chapter 5 Wim J.M. Heijs TECHNOLOGY AND BEHAVIOR: Contributions from Environmental Psychology..........................................43 1. Introduction ..........................................................................................43 2. Information processing..........................................................................44 3. Capita selecta ........................................................................................48 4. Conclusion............................................................................................51 References ................................................................................................52 Chapter 6 Peter-Paul Verbeek ACTING ARTIFACTS: The Technological Mediation of Action .....................................................53 1. Introduction ..........................................................................................53 2. Technology and human-world relationships...........................................53 3. Mediation of perception ........................................................................56 4. Mediation of action ...............................................................................57 5. Conclusion: a vocabulary for technological mediation ...........................60 References ................................................................................................60

Contents

vii

Chapter 7 Jaap Jelsma TECHNOLOGY AND BEHAVIOR: A View from STS....................................................................................... 61 1. Introduction ..........................................................................................61 2. Science studies: social construction of facts...........................................62 3. Technology studies: social construction of technology...........................63 4. From social construction of technology to configuration of users ..................................................................................................64 5. Last step: total symmetry between humans and things ...........................65 6. What have we gained?...........................................................................67 References ................................................................................................69 Chapter 8 Philip Brey THE SOCIAL AGENCY OF TECHNOLOGICAL ARTIFACTS: A Typology................................................................................................71 1. Introduction ..........................................................................................71 2. Behavioral affordances and constraints..................................................73 3. User-profile affordances and constraints................................................74 4. Material and infrastructural affordances and constraints.........................74 5. Social affordances and constraints .........................................................75 6. Cultural affordances and constraints ......................................................78 7. Conclusion............................................................................................79 References ................................................................................................79 Chapter 9 Wim J.M. Heijs and Peter-Paul Verbeek TECHNOLOGY AND USERS: A Conceptual Map ....................................................................................81 1. Introduction ..........................................................................................81 2. A conceptual map..................................................................................82 3. Districts, landmarks and edges ..............................................................85 4. Nodes and paths ....................................................................................87 5. Conclusions ..........................................................................................91 Reference ................................................................................................. 92 PART 2 Technology, Behavior and Sociotechnical Practices Chapter 10 Laurie Hendrickx and Anton J.M. Schoot Uiterkamp TECHNOLOGY AND BEHAVIOR: The Case of Passenger Transport ..............................................................95 1. Introduction ..........................................................................................95

viii

Contents

2. Conceptual model .................................................................................96 3. Car type choice .....................................................................................98 4. Transport mode choice ........................................................................ 100 5. Driving style ....................................................................................... 101 6. Conclusions ........................................................................................ 102 7. A research agenda ............................................................................... 103 References .............................................................................................. 104 Chapter 11 Gert Spaargaren, Susan Martens and Theo A.M. Beckers SUSTAINABLE TECHNOLOGIES AND EVERYDAY LIFE .......... 107 1. Introduction ........................................................................................ 107 2. The Social Practices Approach ............................................................ 108 3. Putting the concepts to work: two case studies..................................... 112 4. Discussion........................................................................................... 116 References .............................................................................................. 117 Chapter 12 Erica Derijcke and Jan Uitzinger RESIDENTIAL BEHAVIOR IN SUSTAINABLE HOUSES.............. 119 1. Introduction ........................................................................................ 119 2. Case study 1........................................................................................ 120 3. Case study 2........................................................................................ 122 4. General discussion and conclusions..................................................... 124 References .............................................................................................. 126 Chapter 13 L.T. McCalley and Cees J.H. Midden MAKING ENERGY FEEDBACK WORK: Goal-Setting and the Roles of Attention and Minimal Justification........... 127 1. Introduction ........................................................................................ 127 2. Feedback Intervention Theory ............................................................. 129 3. Minimal Justification........................................................................... 129 4. Method ................................................................................................ 131 5. Results .................................................................................................133 6. Conclusions ........................................................................................ 134 References .............................................................................................. 135 Chapter 14 Trijntje Völlink and Ree M. Meertens TECHNOLOGICAL INNOVATIONS AND THE PROMOTION OF ENERGY CONSERVATION: The Case of Goal-Setting and Feedback .................................................. 139 1. Introduction ........................................................................................ 139 2. Study 1: electronic feedback through information pages to reduce energy and water consumption ................................................. 140

Contents

ix

3. Study 2: the effect of a prepayment meter on residential gas consumption........................................................................................ 143 4. Discussion and conclusion................................................................... 146 References .............................................................................................. 147 Chapter 15 Wim J.M. Heijs HOUSEHOLD ENERGY CONSUMPTION: Habitual Behavior and Technology ......................................................... 149 1. Introduction ........................................................................................ 149 2. Theoretical framework ........................................................................ 150 3. Intervention and prevention................................................................. 153 4. The role of technology ........................................................................ 155 References .............................................................................................. 156 Chapter 16 Nicole van Kesteren and Ree M. Meertens MARKETING OF TECHNOLOGICAL PRODUCTS: Theory and Methods................................................................................ 159 1. Introduction ........................................................................................ 159 2. Advertising tactics............................................................................... 160 3. The FCB model of Vaughn.................................................................. 161 4. The Rossister and Percy grid ............................................................... 163 5. Brief history of advertising research: contribution and criticism ....................................................................................... 164 6. Means-end chain theory and laddering................................................. 166 7. Brief history of laddering research: contribution and criticism ............. 168 8. Discussion........................................................................................... 169 References .............................................................................................. 170 Chapter 17 Trijntje Völlink, Ree M. Meertens and Cees J.H. Midden DIFFUSION OF TECHNOLOGICAL INNOVATIONS: Promoting the Large-Scale Use of Technology ........................................ 173 1. Introduction ........................................................................................ 173 2. Methods.............................................................................................. 175 3. Results ................................................................................................. 178 4. Discussion and Conclusions ................................................................ 179 References .............................................................................................. 180 Chapter 18 Nicole van Kesteren, Ree M. Meertens and Mirjam Fransen TECHNOLOGICAL INNOVATIONS AND ENERGY CONSERVATION: Satisfaction With and Effectiveness of an In-Business Control System..................................................................................................... 181 1. Introduction ........................................................................................ 181

x

Contents

2. Methodology....................................................................................... 184 3. Results ................................................................................................. 185 4. Summary of results and discussion ...................................................... 187 References .............................................................................................. 188 Chapter 19 Cees J.H. Midden SUSTAINABLE TECHNOLOGY OR SUSTAINABLE USERS?...... 191 1. Introduction ........................................................................................ 191 2. Who is in charge?................................................................................ 192 3. The user as moderator of technological effects .................................... 194 4. User impacts through the sociotechnical environment.......................... 194 5. Persuading the user through technology mediated feedback................. 196 6. The ‘automated’ user of technology..................................................... 197 7. Adopting technological products ......................................................... 198 8. Conclusion.......................................................................................... 199 References .............................................................................................. 200 PART 3 Designing Technology-Behavior Interactions Chapter 20 Wybo Houkes and Pieter E. Vermaas PLANNING BEHAVIOR: Technical Design as Design of Use Plans................................................ 203 1. Introduction ........................................................................................ 203 2. Designing using by plans..................................................................... 204 3. Designing use plans............................................................................. 205 4. Communicating plans.......................................................................... 207 5. Planning behavior ............................................................................... 208 References .............................................................................................. 209 Chapter 21 Harro van Lente EXPECTED BEHAVIOR: Anticipation of Use in Technological Development.................................. 211 1. Introduction ........................................................................................ 211 2. Two types of anticipations................................................................... 212 3. The force of radical anticipations......................................................... 213 4. Obsolete technology............................................................................ 214 5. The logic of progress........................................................................... 215 6. Anticipations and positions.................................................................. 216 7. Conclusion: anticipations are not innocent........................................... 218 References .............................................................................................. 218

Contents

xi

Chapter 22 Jaap Jelsma DESIGNING ‘MORALIZED’ PRODUCTS: Theory and Practice................................................................................ 221 1. Introduction ........................................................................................ 221 2. Conceptual framework: script language ............................................... 223 3. Design methodology in eight steps ...................................................... 226 4. Pilot study: redesigning the dishwasher ............................................... 229 5. Lessons ................................................................................................ 230 References .............................................................................................. 230 Chapter 23 G.W. Wolters and L.P.A. Steenbekkers THE SCENARIO METHOD TO GAIN INSIGHT INTO USER ACTIONS.............................................................................................. 233 1. Introduction ........................................................................................ 233 2. Literature ............................................................................................ 234 3. The scenario method ........................................................................... 236 4. The application of the method in the design process ............................ 238 References .............................................................................................. 240 Chapter 24 Remke Klapwijk, Marjolijn Knot, Jaco Quist and Philip J. Vergragt USING DESIGN ORIENTING SCENARIOS TO ANALYZE THE INTERACTION BETWEEN TECHNOLOGY, BEHAVIOR AND ENVIRONMENT IN THE SUSHOUSE PROJECT .................. 241 1. Introduction ........................................................................................ 241 2. Design Orienting Scenarios in the SusHouse project............................ 242 3. Studying interactions between technology, behavior and environment ........................................................................................ 245 4. A classification of interactions applied to clothing care scenarios ............................................................................................. 246 5. Conclusions ........................................................................................ 249 References .............................................................................................. 251 Chapter 25 Valerie Frissen and Marc van Lieshout ICT IN EVERYDAY LIFE: The Role of the User................................................................................ 253 1. Introduction ........................................................................................ 253 2. Background: a ‘mutual shaping’ perspective........................................ 254 3. The domestication of ICT in everyday life........................................... 255 4. Configuring users................................................................................ 256 5. The appropriation of ICT..................................................................... 257 6. The design-domestication interface and the usefulness of experimentation ............................................................................. 260 References .............................................................................................. 261

xii

Contents

Chapter 26 Marjolijn Knot and Helma Luiten USER INVOLVEMENT IN THE DEVELOPMENT OF SUSTAINABLE PRODUCT-SERVICE SYSTEMS: The Case of the Personal Mobility System ‘Mitka’................................... 263 1. Sustainable product-service systems: technology and behavior ............ 263 2. The Mitka project................................................................................ 265 3. Conclusions and discussion ................................................................. 273 References .............................................................................................. 275 Chapter 27 Henk Muis ETERNALLY YOURS: Some Theory and Practice on Cultural Sustainable Product Development ........................................................................................... 277 1. Introduction ........................................................................................ 277 2. Product life extension.......................................................................... 278 3. Psychological life span........................................................................ 283 4. Practices.............................................................................................. 290 References .............................................................................................. 292 Chapter 28 Han Brezet DESIGNING TECHNOLOGY-BEHAVIOR INTERACTIONS ........295 1. Introduction ........................................................................................ 295 2. The need for a broader scope and approach ......................................... 297 3. Consequences for industrial designers ................................................. 302 4. Nine golden rules for responsible TBI design ...................................... 304 References .............................................................................................. 305 PART 4 Implications for Policy Chapter 29 Bas van Vliet CITIZEN-CONSUMER ROLES IN ENVIRONMENTAL MANAGEMENT OF LARGE TECHNOLOGICAL SYSTEMS ....... 309 1. Introduction ........................................................................................ 309 2. Large technical systems and the roles of their users ............................. 310 3. Intermezzo: Innovation and differentiation in utility systems ............... 312 4. Environmental innovation and differentiation in the water sector ................................................................................................... 314 5. Environmental innovation and differentiation in electricity provision............................................................................................. 315 6. New consumer roles in environmental innovation in water and electricity systems......................................................................... 317 References .............................................................................................. 317

Contents

xiii

Chapter 30 Markus Popkema and Ingrid van Schagen MODIFYING BEHAVIOR BY SMART DESIGN: The Example of the Dutch Sustainable-Safe Road System ........................ 319 1. Introduction ........................................................................................ 319 2. Human error in traffic ......................................................................... 320 3. A sustainable-safe traffic system: the theory ........................................ 323 4. Sustainable-safe road design: from theory to practice .......................... 325 5. Conclusions ........................................................................................ 327 References .............................................................................................. 328 Chapter 31 Boelie Elzen COMBINING TECHNICAL AND BEHAVIORAL CHANGE: The Role of Experimental Projects as a Step Stone Towards Sustainable Mobility .................................................................................................... 331 1. Introduction ........................................................................................ 331 2. Socio-technical learning in niches....................................................... 332 3. Brief examples..................................................................................... 333 4. Relevance for pathways towards sustainable mobility ........................ 335 5. Improving learning in experiments...................................................... 336 6. Conclusion .......................................................................................... 338 References .............................................................................................. 339 Chapter 32 David Laws THE PRACTICE OF INNOVATION: Institutions, Policy and Technology Development.................................... 341 1. Introduction ........................................................................................ 341 2. An examplar of new practice ............................................................... 343 3. The landscape and logic of development ............................................. 346 4. A new logic of practice ....................................................................... 348 5. Policy implications.............................................................................. 352 References .............................................................................................. 356 Chapter 33 Philip Brey ETHICAL ASPECTS OF BEHAVIOR-STEERING TECHNOLOGY ................................................................................... 357 1. Introduction ........................................................................................ 357 2. The freedom issue ............................................................................... 358 3. The technocracy versus democracy issue ............................................. 361 4. The responsibility issue ....................................................................... 363 5. Conclusion: dealing with the moral issues in behavior-steering technology .......................................................................................... 363 References .............................................................................................. 364

xiv

Contents

Chapter 34 Sytse Strijbos A NORMATIVE SYSTEMS APPROACH FOR MANAGING TECHNOLOGY AND COLLECTIVE HUMAN ACTION................ 365 1. Introduction ........................................................................................ 365 2. The context of social practices............................................................. 366 3. Context of socio-technical systems ...................................................... 369 4. Normative types of technology............................................................ 369 References .............................................................................................. 373 Chapter 35 Wim Hafkamp SHAPING TECHNOLOGY-BEHAVIOR INTERACTIONS: Lessons for Policy Making ...................................................................... 375 1. Introduction ........................................................................................ 375 2. The technology-behavior dilemma in environmental policy making ................................................................................................ 376 3. The Score in Policy Sciences............................................................... 377 4. Interactive processes ........................................................................... 379 5. Implications for policy ........................................................................ 381 References .............................................................................................. 383 Chapter 36 Peter-Paul Verbeek and Adriaan Slob ANALYZING THE RELATIONS BETWEEN TECHNOLOGIES AND USER BEHAVIOR: Towards a Conceptual Framework.......................................................... 385 1. The need for a sociotechnical approach ............................................... 385 2. A vocabulary to describe relations between technological products and user behavior .................................................................. 386 3. Implications for technology design and policy-making ........................ 392 4. Conclusion.......................................................................................... 398 Reference ................................................................................................ 399 LIST OF AUTHORS ............................................................................ 401

PREFACE

Adriaan Slob and Peter-Paul Verbeek

This book is the result of an interdisciplinary project that was granted by the Netherlands Ministry of Housing, Physical Planning and the Environment, and AVV, the Transport Research Centre, which is related to the Ministry of Transport, Public Works and Water Management. TNO and the University of Twente also supported the project. We are grateful for these grants. Without them, this book would not have appeared. The project was coordinated by Novem, the Netherlands Agency for Energy and the Environment. We are grateful to the respective project managers of Novem, and especially to Jessica Dirks, for their efforts to keep the project on track. The assistance of Steven Dorrestijn and Mario Willems, with respect to organizing sessions and the contents of this book proved to be indispensable. We owe a lot to Jean-Marie Buijs who helped us with compiling all contributions to the book. Marian Nieuwenhout took care of the lay-out of this book in a careful and skilful way. We would like to thank Jeff Berman for his kind permission to reproduce the photographs on the book cover. The contributions to this book are the result of an intensive interaction process between researchers from a variety of disciplines, which are all somehow connected to the domain of technology and behavior. The inspiration that sparkled from the sessions provided the energy to make this book. We would like to thank all contributors for that. The meetings with our co-editors Han Brezet, Wim Hafkamp, Wim Heijs, and Cees Midden were always pleasant and fruitful. It was a pleasure to cooperate with so many people and exchange different ideas and concepts. We hope that this book xv

xvi

Preface

may be a first step in the direction of a school of ‘technology and behavior’ scientists. Adriaan Slob, Peter-Paul Verbeek July 2005

PART 1 CONCEPTUAL FRAMEWORKS FOR ANALYZING TECHNOLOGY AND BEHAVIOR

Chapter 1 TECHNOLOGY AND USER BEHAVIOR: An Introduction

Adriaan Slob and Peter-Paul Verbeek

1.

THE INTRIGUING INTERACTIONS BETWEEN TECHNOLOGY AND BEHAVIOR

An elderly woman who is an acquaintance of one of us has a small wooden dwelling on an allotment, much like so many in the Netherlands. Electricity is not available. With the decline in prices of Photovoltaic solar energy, it became worthwhile for her to buy a PV unit, so her sons advised her to buy a set. A set of two PV collectors and a battery for storage of the electricity ʊ so that the electricity generated during the day could be consumed in the evening ʊ were installed. But things appeared to be more complicated than her sons had thought. After only two years, the battery was not functioning any more as it should; it did not store as much electricity as it used to do. The lighting in the evening could only be run until 22.30 h ʊ which was much earlier than at first. The woman complained about this to the manufacturer, but he replied that the one to blame was not him, but the woman herself. She was used to keeping the lights low and dim in order to be as frugal as possible when using electricity ʊ a habit going back to World War II. But this saving behavior appeared to ruin the battery. To function properly, the battery needs to discharge by giving a steady and strong current of electricity. The woman had unintentionally destroyed the battery. Old habits are very persistent and sometimes do not fit to new technologies. In the eighties, electricity-saving light bulbs (eslb’s) were introduced in the Netherlands. At first, consumers did not appreciate them. The bulbs were 3 P.-P. Verbeek and A. Slob (eds.), User Behavior and Technology Development: Shaping Sustainable Relations Between Consumers and Technologies, 3-12. © 2006 Springer.

4

Chapter 1

quite heavy, too big to fit into some appliances, and very expensive in comparison to conventional light bulbs. In the mid-eighties, specific actions were set up by the electricity companies to speed up the introduction of the electricity-saving bulbs. The electricity companies gave their customers the opportunity to buy eslb’s with a 20% discount, which resulted in rising sales. But it was only when smaller bulbs that fit most appliances became available that sales went up really high. At that time, a new problem occurred, however. The electricity-saving effects that were expected did not occur. It appeared that most people not only replaced existing bulbs with the new light-saving ones, but also used the new bulbs to illuminate places where there was no light before, such as the garden or the garage. Users appeared to be very creative in appropriating the new technology. They “invented” new applications that had not been foreseen or intended by the designers. New technologies appear to be able to evoke new and unexpected forms of behavior. This book is about the intriguing interactions between technology and behavior. By integrating knowledge from several disciplines ʊ such as psychology, sociology, philosophy of technology, economics, science and technology studies, and innovation and business sciences ʊ we aim to investigate the complex interactions between technology and behavior, and thereby make the results of this investigation fruitful for technology design and policy making. How can we describe and understand these interactions? What kind of scientific disciplines can add to our understanding of them? What kind of empirical research has been done on this subject? How can we deal with these interactions in the design of products, and how can we benefit from knowledge about these interactions for policy design? At the basis of this book lie a number of seminars with a group of scientists from all the disciplines mentioned above. The book explores the boundaries and the possible linkages between these disciplines with regard to their knowledge about technology and behavior interactions. The nature of these interactions demands a multidisciplinary point of view. The questions and problems we will meet transgress the traditional disciplinary boundaries, fully in line with the statement of Gibbons et al., (1994) about “mode 2 science.” On the one hand, the exploration in this book is expected to result in new and interdisciplinary ways of conceptualizing the relationships between technology and behavior. On the other hand, our expectation is that it will result in applications of the insights thereby gained on the practice of designers and policy makers.

Technology and user behavior

2.

5

THE NEED FOR AN INTEGRATED APPROACH

In policy domains such as environmental policy or traffic safety, technology and behavior are usually treated as separate entities. In environmental policy, attention is focused either on the development and promotion of clean technologies ʊ i.e. technologies that cause as little environmental damage as possible ʊ or on stimulating environmentallyfriendly behavior, with the help of information campaigns directed at changing attitudes and behavior of the public. In traffic safety policy, a lot of attention has been given to economic stimuli (fines) and attitude changes (public information campaigns) in order to bring about a change in the behavior of car drivers and to decrease the number of casualties. At the same time, the automobile industry has put a lot of effort into making cars more robust and, therefore, safer. Most cars nowadays possess airbags, anti-lock brake systems, and special safety features in the doors. This two-track approach, in our view, has reached its limits. First of all, if arrangements to stimulate the development of new technologies fail to take the behavior of users into account, unexpected and unintended side effects can occur. The changes in behavior under the influence of the new technology might cause the intended effects of the new technology to “leak away” ʊ as was the case with the energy-saving light bulbs mentioned earlier. The abundant availability of safety devices in cars and the strong reduction of engine noise, for instance, create a very safe and comfortable environment that invites the driver to drive too fast, eliminating the effect of its technological safety measures. On the other hand, the ‘communication and information approach’ has limitations as well. Changing people’s attitudes without taking the behaviorsteering aspects of technology into account does not automatically lead to behavior change. This approach runs the risk of creating its own bias, and of being a form of ‘preaching to the converted,’ since people who are not acting in an environmentally friendly way or driving safely are often less susceptible to these forms of persuasive communication. In such cases, technological devices can be more effective instruments for changing human behavior. Speed bumps are more successful in making people drive less fast than information campaigns about the risks of driving too fast. The need for a new and integrated approach of technology and behavior can be illustrated by taking a look at some examples in the field of Dutch environmental policy. An interesting example of this policy in the past decades is the introduction of the so-called ‘groenbak’ (‘green bin’). The

6

Chapter 1

‘groenbak’ is a separate bin, placed outside the house, in which only organic waste (food and garden related) is collected with the help of a small bin inside the house. The introduction of the ‘groenbak’ was accompanied by an information campaign in which Dutch households were urged to separate organic from non-organic waste. But one of the main problems of the little bin in the house appeared to be the high speed of the decaying process during the summer months, causing a very bad smell and making the emptying of the little bin in the ‘groenbak’ quite a distasteful job. This discouraged people from separating their waste. This problem was taken away, however, when a new product entered the practice of waste separation; a small biodegradable paper bag which can be placed inside the little bin and which makes it much easier to empty and clean it. In some cases, material artifacts are able to effect changes in behavior for which information campaigns are not strong enough. A second example concerns the emission of carbon dioxide. Dutch environmental policy has been quite successful during the last decades. Many problems, like water pollution, air pollution and acid rain, have been dealt with quite well. However, according to the ‘Milieubalans RIVM’ (2001), some persistent environmental problems remain, and are deteriorating rather than improving. One of these problems is the continually rising level of emissions of carbon dioxide (CO2). Despite numerous efforts, it appears to be very hard to bend down the rising curve of CO2. The most important explanation of this phenomenon is steadily rising energy consumption, which eliminates all efficiency gains of energy-using devices. As was the case with the energy-saving light bulb, the development of many new energy-saving appliances has led to an increase rather than a decrease in energy consumption. This phenomenon is often called the ‘rebound effect’. In some cases, the introduction of new technologies to solve specific problems appears to be counterproductive and to result in precisely the opposite of what was intended. The washing machine is a good example in this context. It has become increasingly efficient in a technological sense, but has been evoking less efficient behavior. Washing machines use less energy than some time ago, but at the same time people have started to use their machines more often for small quantities of laundry (Slob et al., 1996). Figure 1-1, derived from an analysis of the use of energy-saving technologies in households in the period 1980 until 1990, shows the trend in washing practices in households in The Netherlands. The figure shows that washing machine technology has been further enhanced in that period, resulting in an electricity saving effect of about 10%. Water consumption, which has a direct relationship to electricity consumption for heating, has declined very quickly in that period. In 1990,

Technology and user behavior

7

the average washing machine used 60 liters of water per wash against 120 liters in 1980. Electricity consumption was reduced also by changing washing habits. In the period 1980-1990, people increasingly washed at lower temperatures and made less use of pre-wash programs. This has resulted in a reduction of electricity use of about 15%. At the same time, however, the figure shows a sharp increase in washing frequency. In 1990, people used their washing machines about 20-25% more often than in 1980. Because of this unexpected change in behavior, the net effect of the technological innovations is about 10%, where it could have been more than 20%.

Effect on power consumption (1980 = 100%)

130

NO. OF WASHES / WEEK

120

110

100

TECHNOLOGY 90

NET EFFECT 80

WASH TEMPERATURE & PREWASH 70

1980

1985

1990

YEAR

Figure 1-1. Trends in washing practices (Slob et al., 1996)

This type of rebound effect is not the only one, however. The unexpected effects of introducing a new technology do not always lead to a sharp increase in its use, but it can also consist in bypassing the technology or not using it at all, or in using the technology in a different way than its designers and policy makers intended. The second type of rebound effect (bypassing) is often seen in the use of control systems like movement detectors for lighting, heating, etc. Consumers devise ways to escape the control of the system, since they prefer to be in control themselves (see Van Kesteren, chapter 18). The third type of rebound effect (unintended use) is closely related to this. Jelsma (chapter 22) shows how users of dishwashers rinse their dishes under running hot water before loading them into the machine,

8

Chapter 1

whereas the machine starts its cycle by rinsing as well. Another example in this context is formed by newly built energy-efficient houses equipped with new insulation materials, triple glazing, sophisticated ventilation systems, monitoring and control systems, low temperature heating systems, et cetera. In many cases, the inhabitants of these houses still open their windows for fresh air. It can be concluded that a strictly functional approach to technology pays far too little attention to the human beings who work with devices on a daily basis and live among them day and night. In order to deal in an adequate way with societal problems regarding traffic safety and environmental pollution, an integrated approach to technology and behavior is required. Technology and behavior are always intertwined. Technologies influence human behavior, and, conversely, existing patterns in human behavior influence the use and even the resulting functionality of technologies. To be sure, the influence of technology on human behavior is not always as compelling as the example of the speed bump might suggest. The influence of many products and environments is ‘softer’ than this. Intelligent products and environments, for instance, facilitate specific forms of behavior and take over some of the actions of users, like movement indicators that switch the lights on when somebody is in the room and switch them off again when the room is empty. Moreover, such products can also influence behavior by giving feedback and providing information, leaving to the user the choice of how to deal with this information. For instance, some technologies give advice about what to do, like washing machines that indicate how full it must be loaded or how much washing powder should be used in relation to the washing load. In all these cases, technologies do not influence human behavior in a determinist way, since their effect on human behavior also depends on the ways in which the technology is appropriated.

3.

THEORETICAL PERSPECTIVES

In the Netherlands, the interest of policymakers in the interactions between technology and user behavior was aroused by the plea of philosopher Hans Achterhuis for a ‘moralization of technology’ (Achterhuis 1995). Achterhuis observed that public concern about environmental problems was usually translated into a ‘moralization’ of people. Public debates usually focus upon morally desirable behavior: people should not shower too long, buy ecological products, use public transport instead of the car, etcetera. According to Achterhuis, this permanent moralization of people cannot be the ultimate solution, however. If it is applied in all

Technology and user behavior

9

domains of our daily lives, it leads to a permanent reflection on our behavior, which would have a paralyzing effect and not succeed in stimulating people to change their behavior. According to Achterhuis, therefore, we should not moralize people, but technology. For this, he elaborated the idea of the French philosopher and anthropologist Bruno Latour that artifacts can help to shape human behavior. According to Latour, artifacts contain ‘scripts’ ʊ prescriptions about how to act, just like the script of a theater play or a movie tells actors what to do and how to do it (Latour 1992). A speed bump in a road, for instance, contains a clear script: slow down when you approach me, or damage your shock absorbers. This script can be seen as a form of moralization, because it embodies and reinforces the moral decision that it is not good to drive fast at places where the bump is constructed. Morality is about the question how to act, and Latour shows that not only people are able to answer this question, but artifacts as well (cf. Verbeek, 2005). Achterhuis proposed taking Latour’s analysis as a starting point when designing technology. From an ecological perspective, this implies that technologies should not only be designed to produce as little pollution as possible, but also to steer the behavior of their users in a desired direction. Rather than demanding from consumers a continual reflection on one’s behavior, some moral decisions could be delegated to the technologies they use. The decision how much water to use when showering can be delegated to a water-saving showerhead; the decision how fast to drive on the highway can be delegated to a speed limiter. According to Achterhuis, attempts to influence human behavior in a desirable direction need to be integrated with attempts to design sustainable or safe technologies. This integration can be accomplished when the behavior-steering role of technological artifacts is taken seriously. Achterhuis and Latour were not the first, however, to seek attention regarding the close relationship between technology and behavior. They formulated a new philosophical perspective regarding a domain that had been investigated before in other disciplines. After World War II, the steep progress of complex technology in, for instance, military applications, aviation and nuclear power, instigated the importance of correct user behavior regarding, among other things, safety, control and comfort. Insights from cognitive, experimental, and (later) environmental psychology were used to certify a proper physiological and mental interaction between users and technology. Theories, concepts and methods from these fields, such as mental models, affordances and system approaches, have been adapted to the

10

Chapter 1

investigation of interaction and the formulation of design guidelines. A number of partially overlapping subfields emerged, such as ergonomics, human engineering, applied cognitive psychology and human factors. From the 1960s onwards, the scope of research on human-technology interaction has gradually extended beyond these complex systems to include parts of the larger environment, such as buildings or even cities and virtual reality, as well as smaller technological aspects of every day life, like consumer products and computer interfaces. Subjects include safety, usability and acceptance on the side of the user in conjunction with various technological features. Additional subfields carry names such as design psychology or user/human centered design. For an adequate understanding of the complex relationships between technology and behavior, it is not sufficient to simply explore the mutual influences between technologies and human actions. The context of use in which technologies are introduced, like existing consumer practices, habitual and conscious patterns of use, and decisions whether to buy a product or not, plays a crucial role as well. This context is important for the ways that users ‘interpret’ and ‘appropriate’ products, which in turn has implications for the possible influences of these products on human behavior. A hammer, for example, affords both driving a nail into the wall and assaulting other people; its context of use plays a decisive role in the eventual technologybehavior interaction that will occur. Despite the available expertise mentioned above, however, many design decisions are taken without considering or making use of it. This is partly due to the rapid pace in which technology evolves and the time and money requirements that have to be met, thereby restricting the opportunities to perform proper research. But, above all, it is due to the dispersion of available knowledge over so many subfields, the lack of sharing it, and a shortage of translating results into data that can be used by designers and policy makers. This brings us back to the purpose of this book. The theoretical aim of this book is to draw together theoretical and applied insights from scientific disciplines that investigate the various aspects of the interaction between technology and behavior. Disciplines such as psychology, ergonomics, science and technology studies, household and consumer science, safety studies, marketing research, and the philosophy of technology, all bring to light specific aspects of the ways in which technology and behavior are interwoven. In this book, we intend to show that much can be gained by bringing these fields of research into contact

Technology and user behavior

11

with each other, and by trying to relate the various theoretical frameworks and concepts to each other. Besides this analytical ambition, the book also has a synthetic ambition. Our aim is not only to inventory and link existing conceptual frameworks, but also to lay bare how anticipations of technology-behavior interactions can inform the design of new technologies, as well as policy-making processes. In order to do so, methodologies and conceptualizations of the design process and of policy-making processes will be investigated from the perspective of the question as to what role technology-behavior interactions play, or could play in it.

4.

RESEARCH QUESTIONS AND OUTLINE OF THE BOOK

This book consists of four separate sections. Each section contains short chapters where researchers from diverse disciplines present relevant conceptual frameworks, empirical research, design approaches and policy implications regarding technology-behavior interactions. Each section ends with a concluding chapter by one of the sub-editors of this book. In order to find a starting point for analyzing the relationships between technology and user behavior, in the first section an inventory is made of the conceptual frameworks that have been developed in several disciplines that study these relationships. The disciplines involved are environmental psychology, cognitive ergonomics, safety studies, domestic and consumer sciences, science and technology studies, and the philosophy of technology. The section results in a ‘conceptual map’ in which all relevant concepts are positioned as landmarks and related to each other. By using the metaphor of a map, it becomes possible to identify districts of related concepts, the edges of their applicability, and nodes and paths as relationships and possible connections between concepts.

The second section addresses the question as to what kind of empirical research has been done in several disciplines to contribute to our understanding of the interactions between technology and user behavior. This part of the book will inventory approaches to human-technology interactions in applied scientific research. It categorizes these approaches on the basis of the role users and technologies can each play in their mutual interaction. The last two sections of the book apply the insights gathered in the first two sections. Section three discusses the relevance of an integrated

12

Chapter 1

perspective on technology and behavior for the design of technologies. The main question here is how the insights in technology-behavior interactions can be applied to the development and design of products. In the development of new technologies, interactions between technology and behavior are created implicitly, and this could happen more explicitly. The section theorizes about ways to anticipate technology-behavior relationships during the process of designing technologies, and considers methodologies for putting this anticipation into practice, including examples of design projects where the anticipation of technology-behavior relationships has played a central role. After clarifying the dynamics of technology-behavior interactions and the possibilities of explicitly designing such interactions, policy issues come to the fore. These provide the focus of the last section of the book, guided by questions about how we can apply the insights in technology-behavior interactions to policy design and development, and what this implies for the organization of the policy-making process. This section pays attention to the points of application for policy-making that result from the analyses in the other sections of the book, and to the implications for the layout of the policy-making process itself. Separate attention is paid to issues of legitimization, since the acceptance of deliberately steering human behavior with the help of technologies cannot be taken as self-evident.

REFERENCES Achterhuis, H. (1995). ‘De moralisering van de apparaten’. In: Socialisme en Democratie 52 nr. 1, 3-12. Gibbons, M., C. Limoges, H. Nowotny, S. Schwartzman, P. Scott and M. Trow (1994). The New Production of Knowledge: The Dynamics of Science and Research in Contemporary Societies. Londen: Sage Publications. Latour, B. (1992). ‘Where are the Missing Masses? - The Sociology of a Few Mundane Artifacts’. In: W.E. Bijker and J. Law, Shaping Technology/Building Society. Cambridge: MIT Press. RIVM. (2001). Milieubalans. Bilthoven: RIVM. Slob, A.F.L., M.J. Bouman, M. de Haan, K. Blok, K. Vringer (1996). Consumption and the Environment: Analysis of trends, TNO-STB, University of Utrecht, CBS, Ministry of Housing, Spatial Planning and Environment, The Hague. Verbeek, P.P. (2005). What Things Do: Philosophical Reflections on Technology, Agency, and Design. University Park, PA: Pennsylvania State University Press.

Chapter 2 ACTION FACILITATION AND DESIRED BEHAVIOR

Albert G. Arnold and Petra Mettau

1.

INTRODUCTION

The objective of this chapter is to provide a theoretical model and concepts that give insights in the relation between technology and desired user behavior. This knowledge is relevant for the design of adequate technological artifacts. An artifact is said to be adequate when its characteristics facilitate the user’s actions in work situations. The insights are of a cognitive psychological nature and lie at the level of the individual user. The model and concepts presented in this chapter are based on so-called human action theory (Hacker, 1986 and Rasmussen, 1986). Human action theory offers a framework for describing a number of relevant perceptual, cognitive and motor mechanisms that play a role during human work activities. Human action theory positions the interaction between artifact and user in the context of work situations. This means that the user (or worker) interacts with the artifact in order to achieve certain goals. Desired user behavior may be defined as human activities resulting in a desired output for the environment and the user. For the environment, this means that the user’s organization and the society at large positively value the results of the interaction. At the level of the user, the assumption is that desired user behavior occurs when the user experiences the benefits of interacting with the artifact. 13 P.-P. Verbeek and A. Slob (eds.), User Behavior and Technology Development: Shaping Sustainable Relations Between Consumers and Technologies, 13-20. © 2006 Springer.

14

Chapter 2

In other words, when there is a good fit between characteristics of the artifact, the user and the user’s task. The artifact should possess characteristics that invite the user to (re-)use the artifact in such a way that optimization of his or her work behavior occurs over time. The inviting characteristics of an artifact are related to the ‘affordance of an artifact’ (Norman, 1988), and the optimization of the work behavior to the concept of ‘action facilitation’ (Roe, 1984; 1988). Both concepts are elaborated in the contribution on the basis of the model of the interaction between artifacts and humans. Furthermore, relevant issues of action theory are presented, i.e. the way people prepare, execute and regulate their actions. Finally, the importance of active optimization is stressed.

2.

MODEL OF TECHNOLOGY-HUMAN INTERACTION

The model must be seen as a heuristic model. It can be characterized as an ‘input-throughput-output’ model. At the input side of the model, three factors are presented, viz. the task characteristics (e.g. complexity, degree of structure, and degree of openness), the characteristics of the artifact, and the user characteristics (e.g. age, gender and intellectual capabilities). The characteristics of the artifact need special attention, as they may have a substantial impact on the ultimate behavior of the user. For example: • The look and feel, stability and reliability, and the performance of an artifact determine to a certain extent the frequency and intensity of use. • The degree of conformance to general operation conventions (e.g. conventions for pushbuttons, switches and turn knobs) greatly determines the degree of intuitiveness of an artifact interface, and thereby the speed of the user’s learning process. • The navigation and support functions of the artifact can stimulate the user to optimize his or her working strategies. It should be noted that there is a reciprocal relationship between the characteristics of an artifact and the user’s task. The input factors determine the kind of user interaction that will occur. On the one hand, the interaction will result in a certain level of performance (in terms of effectiveness, efficiency and quality). On the other hand, the interaction will also result in individual outcomes such as satisfaction, motivation and fatigue.

Action facilitation and desired behavior

15

The user evaluates the interaction and its results. This evaluation will have an effect on the later interactions and on the input factors. When the evaluation is positive, then a positive attitude towards the use of the artifact may occur and the artifact will be seen as adequate.

Task characteristics

Artifact characteristics

Performance

Artifact-user interaction

User characteristics

Individual outcomes

Environmental characteristics

Figure 2-1. Model of artifact-user interaction (adapted from Arnold, 1998)

The artifact-user interaction takes place in an environment with specific characteristics, such as degree of social support, organizational culture and physical aspects.

3.

THE CONCEPTS OF ‘ACTION FACILITATION’ AND ‘AFFORDANCE OF AN ARTIFACT’

As said before, central to the design of adequate artifacts are the concepts of ‘action facilitation’ and ‘affordance of an artifact’. A user action is facilitated by an artifact when proper cognitive, perceptive and motor processes are initiated and accommodated during the interaction with the artifact (Arnold, 1998). Proper, in turn, refers to the outcome of an action in terms of user’s effectiveness and efficiency. The characteristics or possibilities of an artifact may directly or indirectly become clear to the user. The possibilities may be indirectly clarified by

16

Chapter 2

special help functions or formal training sessions. However, the direct articulation of possibilities is much more powerful, because the user is invited to interact in an effective and efficient way without wasting any learning effort. In this context, the concept of ‘affordance of an artifact’ might be introduced. The concept of ‘affordance’ refers to the perceived and actual properties of an artifact, primarily those fundamental properties that determine just how the artifact could possibly be used (Norman, 1988). In order to design adequate tools, insights into the nature of the pertinent human activity are rather indispensable. It should be noted that the direct articulation of possibilities and affordances of an artifact are not always the same, for example the knob of a remote control invites the user to push, however, the icon on the knob may articulate its function.

4.

ACTION THEORY

4.1 Action preparation and execution The starting point of human action theory is the goal-oriented nature of human action. In work situations people interact with artifacts to achieve predefined goals. In order to be successful, workers or users need to translate their goals into actions and operations that serve as instructions for the artifact. Hacker (1986) uses a hierarchical classification scheme of human work activity. He discriminates between acts, actions and operations. Acts are motivated and regulated by intentions, or higher order goals, and are realized through actions. An action is defined as the smallest independent unit of cognitive and sensory-motor processes that is oriented towards conscious goals (Arnold & Roe, 1987). Operations are action components, and do not have independent goals as counterparts. The next illustration might be helpful to clarify the distinction between acts, actions, and operations. In order to organize a conference, one of the organizers is driving a car to the meeting place of the program committee. The person in the car is an experienced driver and is executing the driving task without much attention. Organizing a conference can be considered as the act, going to the meeting place as the action, and the activities necessary for driving the car as the operations. An implication of the above distinction among acts, actions and operations for the design of an adequate artifact may be found in the structure of its interface. For example, it might be considered that structuring the artifact’s functions on the basis of its users’ acts, actions and operations

Action facilitation and desired behavior

17

will make the artifact’s functions directly recognizable. In this way, the interaction with the artifact might become much more logical to the user. Human activity is based upon action programs, which are mentally represented and goal-oriented plans of future action. Action programs are organized in a hierarchical-sequential fashion. Intentions and goals occupy a crucial position in action programs. They serve to initiate and regulate action execution. To translate an intention into an actual action, an action program has to be formed or an existing program has to be retrieved from memory. This formation of an action program happens before the action itself and is called ‘action preparation’. The following steps in the preparation of an action program can be distinguished: • Goal-setting. • In this stage a personal goal is chosen. It is decided what needs to be done. In general this personal goal is a translation of the expectations of one’s management. • Orientation. • In this stage, an orientation takes place to conditions within the environment, such as the availability of information, persons, tools, etc.; and to personal conditions, such as knowledge, skills, capabilities, limitations, etc. • Design of action program. • On the basis of an analysis of the goal and relevant conditions, variants of action programs are designed, or a program designed and mentally stored earlier is reproduced. • Choice of a variant. • Within given degrees of freedom, an action program is chosen from the available variants. Or, again an existing action program is activated. An implication of the notion of action programs for the design of an adequate artifact may be found in the user support functions. If the user is doing the task for the first time, the artifact might invite him to follow a proven and efficient action program of a more experienced colleague. After action preparation the execution can take place. The action program is executed in a stepwise manner. During execution, feedback is obtained in order to guide and control the action. During an unfamiliar action, constant feedback takes place. This means that the action is executed with full attention. In the case of a familiar action, there is no constant feedback, control is taken over by specialized functional schemata (Reason, 1986). The anticipated action flow and the results are represented in these schemata. Once such a scheme has been triggered, execution is unconscious.

18

Chapter 2

Depending on the amount of practice, there is a tendency to execute more frequently at the unconscious levels. In this way, new action steps can be anticipated and planned during the unconscious execution. In order to facilitate the user, the artifact might keep track of the execution of an action program. In case of a wrong operation or an omission, the artifact might warn the user.

4.2 Action regulation An action can be triggered and controlled by various regulation mechanisms. Three levels of action regulation can be distinguished: • knowledge-based; • rule-based; and • skill-based. The highest level of regulation is the knowledge-based level. This level is concerned with programming or redesigning action programs and is characterized by its problem-solving types of processes. After a diagnosis of the environmental conditions has been made, several solution strategies are developed on the basis of hypotheses on the creation of desired future results. These strategies are mentally evaluated or tested most of the time, and an optimal strategy is chosen. Next, this strategy is translated into new or modified procedures; de facto, an action program has been developed. The actual execution of these procedures takes place by delegation to the lower levels of regulation. Another function of this knowledge-based level is guidance of actions by anticipation and supervision. Only at points in the action flow where conscious orientation is needed are results fed back to this level, i.e. under changed or new external and internal conditions. Experienced persons will switch back to this level less frequently than those who are inexperienced. At the rule-based level, earlier stored rules or procedures control the actions. The incoming stimuli from the outside world can be compared with traffic signs, i.e. as having some specific meaning that can trigger very complex actions. The incoming signs are classified and, subsequently, a choice of a suitable procedure is made. On the basis of this choice, the procedure is recalled from memory and executed by delegation to the skillbased level after checking against the condition stored also in memory. At this level, feedback of the results only takes place when the next procedure has to be started during action program execution.

Action facilitation and desired behavior

19

The skill-based level is the one at which unconscious action program execution takes place. The routines are initiated and triggered by cues or signals from the outside world that are not consciously perceived. The incoming signals are matched with the sensory-motor representations stored in memory. When a signal fits such a routine, the routine will be triggered. When no fit is possible, the higher levels of regulation are activated. When it is allowed by the work conditions, then the artifact should support working at the skill-based level by avoiding unnecessary disturbances. Since working at the skill-based level requires less effort, every deviance of the normal interaction forces the user to the knowledgebased level.

5.

OPTIMIZATION OF ACTION EFFICIENCY

When an action is frequently executed, a process of so-called ‘streamlining’ will occur. The result of this process of streamlining will be an increase of the action efficiency. There is a human tendency to strive for optimization of action efficiency. In other words, people try to learn to do a task as well, fast and easily as possible. The first step in the process of streamlining is the improvement of action programs. Initial programs are made more sophisticated by frequent execution, for example, unnecessary action steps are removed from the program, and shortcuts are found. The second step is lowering the regulation level during action execution. Through practice, action programs are transferred into more or less fixed procedures that are executed through ever-larger sensory-motor routines. When optimization of action efficiency occurs, the action will very likely be repeated, because one gets a feeling of success. The task is done fast and with ease, and the results of the execution meet certain quality criteria. In a negative way, the chance of errors increases when actions are regulated at the lower levels, because the actions are no longer under direct conscious control. Artifacts can profoundly support the tendency of optimization of action efficiency by stimulating and guiding the user in the process of optimization of action efficiency; for example, the artifact could provide the user with advice in the formation of his work strategies. And, at the same time, the artifact should prevent user errors by means, e.g. of intelligent, on time warnings.

20

Chapter 2

REFERENCES Arnold, A.G. & R.A. Roe (1987). User Errors in Human-Computer Interaction. In: M. Frese, E. Ulich & W. Dzida (Eds.) Psychological Issues of Human-Computer Interaction in the Work Place. Amsterdam : Elsevier Science Publishers B.V., pp. 203-220. Arnold, A.G. (1998). Action Facilitation and Interface Evaluation. A work psychological approach to the development of usable software. Delft: Delft University Press. Arnold, A.G.L. Zwep & W. Dzida (1994). Quality assurance as a process of improvement. In: G.E. Bradley & H.W. Hendrck (Eds.) Human Factors in Organizational Design and Management - IV. Amsterdam: North-Holland, pp. 367-373. Hacker, W. (1986). Allgemeine Arbeits- und Ingenieurpsychologie, Psychische Struktur und Regulation von Arbeitstätigkeiten. Bern: Verlag Hans Huber. Norman, D.A. (1988). Designing everyday things. New York: Currency-Doubleday. Rasmussen, J. (1986). Information Processing and Human-Computer Interaction. Amsterdam: North-Holland. Roe, R.A. (1984). Taakstructuur en software-ontwerp. In: R.P. van Wijk van Brievingh & F.G. Peuscher (Eds.) Methodisch ontwerpen van medisch-technologische apparatuur. Delft Progress Report. Delft: Delft University of Technology. Roe, R.A. (1988). Acting Systems Design - An Action Theoretical Approach to the Design of Man-Computer Systems. In: V. de Keyser, T. Qvale, B. Wilpert & S.A. Ruiz Quintanilla (Eds.) The Meaning of Work and Technical Options. Chichester: John Wiley & Sons, pp. 179-195.

Chapter 3 SAFETY: Technology and Behavior?

Adrienne van den Bogaard and Paul Swuste

1.

INTRODUCTION ‘A young child standing near to the wheel of a horse mill was by some mishap come within the sweep or compass of the cogwheel and therewith was torn in pieces and killed. Upon inquisition taken, it was found that the wheel was the cause of the child’s death, whereupon the mill was forthwith defaced and pulled down.’ (Manor roll, 1540) ‘She one day got entangled in the machinery until all her clothes were torn off the back, and the overlooker was not at hand, but we got the mill stopped, and when she was taken out she was much abused for her neglect in letting herself in.’ (Citation from a worker in a UK factory inspector’s report on an accident of a young girl in a flax workshop, 1831).

These citations represent two visions on causes of accidents, which are both mono-causal. In the first citation the cause of the accident is a technical one, the mill. The second citation is an example of rule violation, or human behavior. The tendency to reduce causes of accidents to only one single cause is widespread, but destructive for improving safety. This chapter focuses on the Safety Science approach towards risks and hazards in terms of technology and behavior, and the interaction between them. The distinction between technology and behavior is a problematic one. Technology and behavior are interwoven. Safety or the lack of safety are not determined by either technology alone or only by people working with or 21 P.-P. Verbeek and A. Slob (eds.), User Behavior and Technology Development: Shaping Sustainable Relations Between Consumers and Technologies, 21-31. © 2006 Springer.

22

Chapter 3

using the technology, but is the result of many causes operating together in a system. Automation has often been proposed as a fruitful strategy to improve levels of safety. Automation increases the distance between hazard and victim. Besides, automation takes over complex human tasks, and systems become resistant to human errors. The focus on interaction between technology and behavior will highlight serious shortcomings, as this chapter will show.

2.

A SHORT CHARACTERIZATION OF SAFETY SCIENCE

Within the field of research of Safety Science a system approach is common. This system consists of technological and human elements and their mutual relations. A system may be an installation, a factory, a sector, or an infrastructure. An accident doesn’t happen just because somebody does something stupid, as the accident prone (“blaming the victim”) theory would argue, but is the result of a gradual process of loss of control in the system, as is shown in figure 3-1 (based on Hale and Glendon, 1987). The model presented in Figure 3-1 consists of three columns. Time flows downward in the central column. This central column represents the damage process in the system, which starts with deviations and passes through ‘loss of control’. The ‘damage process’ may include material loss, but also accidents with minor or major injuries, or even disasters. The left column represents the measures that can be taken to stop the sequence of events from losing control. The right column represents the vital recovery and learning loops of a safety management system. From a design point of view, safety and its control measures can be built into the technology. This is presented by the first loop at the top of the figure. This means that the technology itself controls certain deviations. Examples are regulators, which control engines, or thermostats that control temperatures, or signs and traffic lights that control traffic flows. After a design is finished, the normal situation or Default State is defined. If deviations occur, they must be detected, and if so, hazard control measures can recover the system and bring it back to the normal state. However, if the deviations are not detected, or no adequate recovery has been found, further loss of control may happen. Then, escape measures come into the picture. And so on. These are the other loops in Figure 3-1. It is clear that deviations may occur very frequently, while a real damage process only happens

Safety

23

occasionally. Luckily, major disasters are rather rare events. An actual question in Safety Science is to characterize these deviations and distinguish between scenarios that only lead to minor accidents and those leading to disasters (Hale, 2002). Choise and design of preventive measures

Choise and design of (sub)system

Redesign

Elimination of hazard Normal situation with inbuilt hazards (default state)

Learning

Recovery

Control measures Deviations from normal situation

Detection, recovery

Reporting Loss of control (release of energy + exposure)

Escape

Transmission

Secondary safety Damage process

Rescue, damage limitation, treatment Stabilisation

Figure 3-1. Accident deviation process

Chapter 3

24

Accident scenario is the central concept in safety studies. A scenario consists of a set of potential or actual deviations and their generic causes, as well as the loss of control that follows these deviations. Accident scenarios may be used as a tool to analyze accidents and disasters that have already occurred, or as a tool to predict these outcomes. In the description of a scenario, the deviations and loss of control do not only contain technical aspects of the accident process. A quote from the report of the Three Miles Island investigation emphasizes the complex relations between technological, behavioral and organizational factors, including their interactions: While there is no question that operators erred we believe there were a number of important factors not within the operator’s control that contributed to this human failure. These included inadequate training, poor operator procedures, lack of diagnostic skills on the part of the entire management site group, misleading instrumentation, plant deficiencies and poor control room design (Nuclear Regulatory Commission Investigation into the Three Miles Island accident, 1980) Accidents or hazards can be analyzed in terms of four causal factors: technology, behavior, and the context in which the interaction between technology and behavior takes place. This context can be analyzed in terms of the safety management system and the safety culture. In Safety Science, accidents are analyzed in terms of multi-causality of these four factors. Unwillingly, this approach neglects the focus on the interaction between technology and behavior as a separate factor of importance.

3.

FOUR CAUSAL SAFETY FACTORS

3.1 Technology The socio-technological design of a system, or organization, is the starting point. Techniques are used to establish or predict all possible hazards and system failures, leading to accident scenarios. These can range from purely technical aspects (e.g. a failing valve) to human failures based on unrealistic presumptions of people, or alarm signals that are difficult to interpret.

Safety

25

From a technical point of view, various types of control measures can be applied. Next to the elimination of the hazard or danger (e.g. complete removal of asbestos-containing materials), the danger or hazard can be shielded. For instance, the thickening of the wall of boiler vessels reduced the incidence of boiler explosions dramatically at the beginning of the 20th century. Related to shielding is the strategy of increasing the distance between a hazard or danger and people, by introducing remote-controlled or automated functions (Swuste, 1996), or using zoning by-laws to separate dangerous industries from residential areas. Industries are not dependent on other actors when they pursue the strategy of automation, in contrast with zoning. Besides, the presumed driver for automation is the increase in efficiency of the system, apart from its effect on safety. So, automation is an often-pursued strategy for industries. Once the potential ‘deviations and accident scenarios’ from the Default State are known, the speed by which a deviation can lead to loss of control is important. The extent to which deviations produce a rapid loss of control is dependent on two variables: the complexity of the system and the degree of coupling (Perrow, 1984). Complexity is related to the (bio)chemical or physical processes involved. Nuclear power plants, high altitude aerospace, biotechnology, etc., are examples of processes that are not fully understood and which need series of feedback loops to monitor the Default State of the system. Complex systems always require complicated control systems, as in the case of nuclear power plants. Also traffic-jam waves can be considered a complex process because consequences of unexpected behavior are unpredictable. The degree of coupling of a system is a measure for the time available to correct deviations, before it enters a loss of control stage. When one takes a look again at Figure 3-1, it becomes clear that coupling refers to the central column, i.e. the time it takes to enter the ‘Damage process’. ‘Coupling’ refers to the interaction between the technology involved and human behavior. More specifically, it relates to the degree of freedom of action of operators to respond to deviations occurring. Tightly-coupled processes are time-dependent, like many remote-controlled production lines in industry. The time to intervene is limited and the effect of process deviations is visible by immediate heaps of lost materials somewhere else in the process. Loosely-coupled processes are more time redundant. Manual handled processes are examples of loosely-coupled systems. Time and production pressure can change a loosely-coupled system into a tightly-coupled one. This means that when a decision is made to automate manually-handed

26

Chapter 3

functions, for example, to increase its efficiency, the system is made more tightly-coupled at the same time. In tightly-coupled and complex systems, accidents and disasters are inevitable. Disasters like Three-Miles Island and Chernobyl were the effects of such tightly-coupled complex systems.

3.2 Behavior ‘Accidents or disasters are the result of rule violations’. This is a widespread vision about the lack of safety. Mostly this view is expressed after an accident or disaster has happened, when rules, norms and procedures are being adapted and refined. An example is the focus on rules and procedures after the Volendam and Enschede disasters. From a legal point of view, rule violation might be relevant to answer questions about the responsibility of disasters. Safety is only partially captured by rules and procedures. There is a substantial amount of literature on the functioning of safety rules and their limitations (e.g. Elling, 1991; Hale and Swuste, 1998). And there is substantial evidence that disasters also happen in certified companies such as the Exxon Valdez and Esso Longford (Hopkins, 2000). To understand the limitations of rules and procedures, a closer look at the behavior of people is necessary. In Safety Science, the theory of cognitive psychology is used to propose safety measures that are in harmony with human behavior. On this basis, a distinction is made between skill-, rule- and knowledge-based behavior (Rasmussen, 1993; Reason, 1990). The problem solving capacity in case of a loss of control of a system depends on the type of behavior of the people who are directly involved. Skill-based behavior denotes an automatic response to a certain event. It is a very quick response, because it is an automatic one. This kind of behavior needs refresher training because the ‘automatic pilot’ may become a bit sloppy about detecting deviations. One method to help people reflect on their own skill-based behavior is to show them videos of their own work. Rule-based behavior is based on consciously applying rules one has learned to specific situations. It takes more time than skill-based behavior, due to the conscious basis of the response. Crucial to rule-based behavior is making a diagnosis of a deviation. Making the right diagnosis is fundamental. Quite often, people make a diagnosis on the basis of expectations of hazards from the past, instead of being open to the signals revealed to them.

Safety

27

Knowledge-based behavior may involve creative thinking to solve a problem. In this state, problem-solving capacity is at its maximum, but this type of behavior is also very time-consuming. It applies to situations, which haven’t happened before. It is clear that time must be available for finding the right solutions. Problem-solving behavior is important in dealing with deviations. In tightly-coupled systems there is little time to take the right control measures. This means that skill- or rule-based behavior must be suitable to the right control measures. This is a design question: if system designers want to design automated functions, they must co-design skill-based behavior as well. Loosely-coupled systems allow for knowledge-based behavior when deviations occur. Shifts between skill- and rule-based behavior, and between rule- and knowledge-based behavior will depend on the expected scenarios of those involved in the problem-solving. Also, the environment in which people work can influence these shifts. This environment can be differentiated into the safety management system and the organizational culture.

3.3 Safety management Management of safety in an organization does not differ from the management of quality of processes, or products, or from financial management. An adequate safety management follows the design steps and then the right column of Figure 3-1, intervening through the instruments in the left column. Relevant indicators, such as system failures, deviations, or near misses, are monitored, analyzed and evaluated. And if appropriate, this leads either to an adaptation of control measures or the redesign of the system (Kjellén, 2000). These activities are part of the ‘reporting’ and ‘learning’ box of the figure, and follow up to the top loop of the figure. Insight in scenarios is fundamental to interpret the reported deviations. Scenarios enable the organization to characterize deviations either in terms of potential disasters, which need immediate attention, or in terms of consequences that are within the problem-solving capacity of the organization, and thus need no further attention. Unfortunately, many organizations do register, but limit their registration to the endpoint of the accident process, namely, accidents with minor or major injuries. The main driver behind such registration is bureaucratic, because external parties force them to do so. Without any insight into accident scenarios, these organizations thereby fail to enter the ‘learning’

28

Chapter 3

box in Figure 3-1, so the same accident sequence process can occur repeatedly without an adequate organizational response. This became visible, for instance, in the British rail transport system: although the lack of maintenance was a factor in some of the recent dramatic accidents, this wasn’t noticed at all, and the level of maintenance was not improved.

3.4 Safety culture The attention to organizational culture or, more specifically, the safety culture of an organization is fairly recent. It originates from the notion that industrial companies with similar safety management systems in place can vary substantially in their safety performance, as measured by their accident rates. These differences are not explainable from a difference in hazards and accident scenarios of their production processes. The cultural aspect of safety is defined as the unwritten rules of an organization, how they solve their problems and set their priorities of action. This implies that aspects other than just a formal safety management system are also essential. These aspects are called the ‘basic assumptions’ of an organization, and originate from its strategy to survive under hostile market conditions, leading to unwritten rules such as ‘this is the way we solve our problems here’ (Guldenmund, 2000; Guldenmund and Swuste, 2001). For instance, in a company with a dominant ‘blaming the victim’ culture, employees are hardly encouraged to report near misses or accidents because reporting can have consequences for their careers or their status. A similar attitude can be expected by companies focused primarily on production and production output. Here, safety and safety-related reporting systems are considered to be a waste of time, not contributing to higher production (Swuste et al., 1999, 2002). These basic assumptions can be researched by analyzing the level and extent of problem-solving in the company, by observation, and by studying behavioral patterns.

4.

INTERACTION BETWEEN TECHNOLOGY AND BEHAVIOR

We have argued that in Safety Science scenario-analysis may lead to automation as a solution to potential hazards because automation increases the distance between hazards and potential victims. However, we have also argued that automation makes the system more tightly-coupled. This implies that a new type of behavior must be co-designed, namely skill- and rulebased behavior. We will show this by analyzing automation in the cockpit.

Safety

29

One area where automation has expanded enormously is air transport. The cockpit has become an operating control system, and the pilot has turned into a system manager. However, in contrast with the prediction of scenario analysis, the application of automation has not increased safety. We argue that the neglect of the interaction between coupling and concomitant behavior by the system designers will explain this surprising outcome. System engineers have programmed all kinds of systemic reactions to possible loss of control (see Figure 3-1, “design preventive measures”) If the pilot, for example, switches off an engine which is not a good thing to do, the system will react with an automatic response. These systems thus have automatic protection against operations that could endanger the flight. However, the system is a black box to the pilots: they have poor representations of the architecture of the system and its functioning. One such an example is the manual over-riding of some protection systems in order to bypass restrictions in the flight envelope. This was one of the causes of the Airbus crash in France in 1988. Pilots seem to feel quickly at ease in the new and automated cockpits, but this ease can mask deficiencies in deep knowledge, and they tend to take risks without calculating consequences. Designers did not foresee this pilot behavior. There are more sources of errors than the lack of insight into the automated systems. For instance, there are errors in parameters and mode selection. These errors are called ‘finger errors’, that is, errors in parameter setting that may result in wrong goals being set for computers, with dramatic consequences if the pilot does not detect the deviation. Also, errors occur in crew-co-ordination and cockpit-air traffic controller communication. Communication can be verbal, non-verbal, or by means of computers. But it is the paradox of automation that communication is harder to perform in modern cockpits. Non-verbal communication, which used to consist of visual feedback between the pilot and the co-pilot, is now greatly reduced, as pilots are usually looking at computer displays. Whatever the source of the error, the system is very tightly-coupled, which means that there is little time to take the right control measure, although the deviation may be very complex in its effects. However, in contrast with the complexity of the right control measures, automation can seriously reduce the problem-solving capacity of pilots. Most activities can be done with skill-based behavior. Pilots tend to lose their energy. Should problems occur, they are much more likely to be associated with under-load and under-arousal, leading to reduced vigilance

30

Chapter 3

and even complacency, which both represent a serious deterioration in a pilot’s performance. So, the automation that was applied implied contrasting kinds of behavior. On the one hand, the kind of automation asked for knowledgebased behavior in the case of a certain type of deviation. However, at the same time, the kind of automation asked for skill-based behavior due to the fact that tasks have become easier and because the system has become more tightly-coupled. This explains why safety didn’t increase. New concepts in pilot assistance therefore must solve this problem of implied contrasting behaviors. Considerable efforts are directed to the design of the next generation of cockpits in order to achieve a more human-centered philosophy of automation. (Amalberti, 1993). The focus is that the automated system must be designed in such a way that the pilot’s knowledge-based behavior is asked for during the whole flight. The capacity to preserve onboard intelligence and creativity must be designed. The crew is allowed the opportunity to operate the equipment. In short, the system must allow crews to fly the way they want to fly by giving them flexibility in operating devices. Only when an unacceptable condition has occurred will the system respond with warning and alert signals. The limits of the room to maneuver, according to this philosophy, can be detected by an ‘electronic cocoon’, which is like a multidimensional ‘shell’ around the aircraft and the crew. As long as the flight stays within the cocoon, the system leaves the crew alone to fly as they want. So the nature of automation changes, considering that feedback to manual control is a better way of applying automation than taking over manual control completely.

5.

CONCLUSIONS

We have argued that scenario analysis suggests three options for increasing safety. The first is to remove behavior, which entails automation as a solution. The second is to remove the danger, which entails inherent safety analysis. The third is a more hybrid suggestion: it means that the technology must be developed to support people adequately during complex operations. Interaction between technology and behavior thus lies at the heart of this third option. The insights gained from the flight operation example regarding the interaction between technology and behavior of the flight operation, indicate that technology must leave more freedom to the pilot and its crew. This

Safety

31

conclusion is a bit paradoxical because it has often been argued that automation increases safety tout court. This means that more research on the interaction between technology and behavior in safety analysis might lead to new insights about how to develop technology to support complex operations in the best way.

REFERENCES Amalberti, R. (1993). Safety in flight operations. In: Reliability and safety in hazardous work systems: Approaches to analysis and design. Wilpert, B. and Qvale, T. (eds). Lawrence Erlbaum Associates, Publishers, Hove, UK. Elling, M. (1991). Safety rules in industry. Ph.D. thesis (in Dutch), faculty of Philosophy and Social Sciences, Technical University Twente. Guldenmund, F. (2000). The nature of safety culture. A review of theory and research. Safety Science 34 pp. 215-257. Guldenmund, F. en Swuste, P. (2001). Safety culture: magic word or object of research? (in Dutch) Tijdschrift voor toegepaste Arbowetenschap 14 nr. 4, pp. 2-8. Hale, A., Gendon, A. (1987). Individual behavior in the control of danger. Elsevier, Amsterdam. Hale, A., Wilpert, B., Freitag, M. (eds) (1997). After the event – from accident to organizational learning. Elsevier Sciences Ltd, Oxford. Hale, A., Swuste, P. (1998). Safety rules: procedural freedom or action constraint? Safety Science 29 pp. 163-177. Hale, A. (2002). Conditions of occurrence of major and minor accidents. Urban myths, deviations and accident scenarios. Tijdschrift voor toegepaste Arbowetenschap (in press). Hopkins, A. (2000). Lessons from Longford: the Esso Gas Plant Explosion. Sydney. CCH Australia. Kjellén, U. (2000). Prevention of accidents through experience feedback. Taylor and Francis, London. Leplat, J. (1998). About implementation of safety rules. Safety Science 29 pp. 189-204. Masson, M. (2002). Learning from incident reporting. What are the challenges? Proceedings of the congress on Occupational Risk Prevention. Gran Canaria. Perrow, C. (1984). Normal accidents: Living with High-Risk Technologies. Princeton University Press, New Jersey. Rasmussen, J. (1993). Learning from experience? How? Some research issues in industrial risk management. In: Wilpert, B., and Quale, T. (eds.), Reliability and safety in hazardous work systems. Lawrence Erlbaum Associates, Hove, East Sussex. Reason, J. (1990). Human Error. Cambridge University Press. Swuste, P. (1996). Occupational hazards, risks and solutions. Ph.D. thesis. Delft University Press, Delft Technical University. Swuste, P., Hale, A., Guldenmund, F. (1999). Change in a steel works: learning from failures and partial successes. Paper presented at the NeTWork workshop on ‘Achieving successful safety interventions’ Bad Homburg, June 1999. Swuste, P., Guldenmund, F., Hale, A. (2002). Organizational culture and safety in a heavy industry (in Dutch) Tijdschrift voor toegepaste Arbowetenschap 15 nr. 1, pp. 7-14. Vaughan D. (1996). The Challenger launch decision: risk technology, culture, and deviance at NASA. University of Chicago Press. Chicago.

Chapter 4 TECHNOLOGY AND HOUSEHOLD ACTIVITIES

J.P. Groot-Marcus, P.M.J. Terpstra, L.P.A. Steenbekkers and C.A.A. Butijn

1.

INTRODUCTION

The modern household is full of activities in which technology plays an important part. During the 20th century, a variety of different technologies has been introduced in households that have improved the quality of life of those who perform household tasks, as well as of other members of the household. They have enabled households to enjoy a higher standard of living, as eased the work, provided comfort, and created opportunities to meet the greater demands of a more complex living standard (Visser, 1969; Groot-Marcus, 1984). Household and Consumer Science focuses on households and household activities from a sociological, economical, and technological point of view. Its sub-department of Consumer Technology has to do with the technological approach. They study the relationships between technology, behavior, and environmental effects. The way in which household activities and relationships between technology and behavior are conceptualized in Household and Consumer Science is presented in this chapter. In this context, household activities are the behavioral component of the household. Attention will be paid to the opportunities and constraints that are encountered when technological interventions are introduced in a household system.

33 P.-P. Verbeek and A. Slob (eds.), User Behavior and Technology Development: Shaping Sustainable Relations Between Consumers and Technologies, 33-42. © 2006 Springer.

34

2.

Chapter 4

DEVELOPMENTS IN HOUSEHOLD SCIENCES

Concepts of Household and Consumer Sciences have been developed from Home Economics. That science started in the U.S. a century ago and focused principally on households as management systems (Deacon and Firebaugh, 1968; Gross and Crandall, 1973). In Germany, the emphasis is more on economics and functionality (Egner, 1952; Von Schweitzer, 1969). The development in the Netherlands in the 1970s and 1980s was characterized by a view of the household as a social phenomenon with functional relationships (Presvelou, 1980). The basic entity is the household group1, which performs activities aimed at the satisfaction of everyday material needs of human beings and which creates material conditions for their immaterial needs (Zuidberg, 1981). The purpose is to achieve wellbeing for the members of the household group. The group normally provides its own human resources2. They form, with non-human3 resources such as money and time, appliances, and other goods, the household resources4. Together with information, facilities and services from outside organizations, they generate an output: the level of living5. This is the result of all household activities taken together. Resources are generally used for different domains, such as personal care, the provision of food, clothing, etc. This causes the activities within households to be interrelated. If more of a resource is utilized for one domain, the availability for other domains will be reduced. Household groups try to bring their level of living in line with their ultimate goals for their everyday life, the standard of living6. This is a set of 1

household group: social unit, often family-based, with a communal household, i.e. one or more persons with a number of communal activities. 2 human resources: means vested in people that can be used for attaining goals and creating events: cognitive insights, psychomotoric skills, affective attributes, health, energy, and time. 3 non-human resources: non-human means for attaining goals and creating events: natural and processed consumption goods, housing, space, household capital, physical energy, money and investments. 4 household resources: means for attaining goals and creating events generated within the household system, partly provided by the group (human resources) partly present as goods and services acquired earlier, or assets. 5 level of living: quantity and quality of goods and services consumed or available. It is the result of household activities. 6 standard of living: complex of conceptions, views, habits and wishes of the members of a household group with regard to the aim of household management and the way to employ household resources.

Technology and household activities

35

standards and values concerning daily life and representing the views and ideas of the members of the household group about the way to live (Zuidberg, 1981). The standard of living is not a rigid entity. It is partly formed by the values and norms in the outside world and it adapts itself continuously to external and internal circumstances (changes in life cycle, resources). The balance between the standard of living and the level of living works as a controlling factor. Zuidberg calls this well-being7. Spijkers Zwart (1973) uses the term ‘aspiration level’8. The description of household mechanisms as above is concentrated on the social group and on the ‘human factor’. Accordingly, this approach does not include the physical interaction with the environment. For example, it considers an inflow of resources but not their outflow. The technological approach considers households as systems where an activity, in general, has energy, material, and immaterial (mostly human) resources as input. The output consists of the desired services, undesired side effects, and waste. In the so-called Consumer-Technology Interaction Model, which has been developed as an extension of Zuidberg (1981) and Spijkers-Zwart (1973), these elements are represented as well. In the next paragraph, this model will be elaborated.

3.

CONSUMER-TECHNOLOGY INTERACTION MODEL

The Consumer-Technology Interaction Model (Figure 4-1) is based on a technological approach where households are systems in which material and immaterial elements function for the satisfaction of everyday needs in interaction with society and the environment. Here, the emphasis is on the interaction between the human and material factors. The interaction occurs when material resources or physical processes partake in household activities. Physical processes are units in which material resources (energy and material) are used, and services and emissions (waste and energy) produced. Here, techniques enter the household system, but they appear not only here. Facilities (infrastructure and services) from outside, for instance, for transport, provision of food or clothes, contain techniques, and this applies also to previously-acquired material household resources inside the system, 7

well-being: evaluation of the level of living by the members of the household in relation to their standard of living. 8 aspiration level: tension between the standard of living and the actual situation in everyday life.

36

Chapter 4

such as kitchen appliances. In the model, elements, which contain techniques, are all connected to household activities.

‘level’ of well-being Facilities services infrastructure information

Resources energy material

standard of living

norms, values in society

level of living

household resources

household activities

Emission waste material energy

boundery household system

Figure 4-1. Consumer-Technology Interaction Model (→ = feedback)

Household activities are a central element in the system. Besides the meeting place of technology and behavior, they are the link between household resources, the level of living, and the level of well-being. The term household activities is not restricted to physical processes or execution of actions. They range from purely material to purely mental processes, as each activity includes more or fewer elements of decision-making, planning, organization, and implementation. As the activities within households are interrelated, leading to a goal, resources need to be managed. The feedback mechanism plays an important role (Goldsmith, 1996). Information is generated from the level of well-being, which is fed back into the system, the standard of living and/or household activities. Because the household system is a dynamic system, it requires these feedback mechanisms to keep its stability through time. A stable system strives for a return to the equilibrium. A change in one of the elements could turn the system into an unstable state. But through (negative) feedback, stability can be restored (Roberts et al., 1983). However, households do not always plan or make decisions. The diversity and daily return of household activities, as well as the incorporation of these activities into a previously constructed material background, are the

Technology and household activities

37

reason that people do not continuously decide anew how to act and gear household resources to one another. Actions that have been learned in the past often become habitual. Household resources are partly generated within the household system and are dependent on characteristics of the household group (income, space, time, abilities, skills, etc). Most of these characteristics, labor, time, space, and money, are limited. An exception is skills, because these can be learned. For example, skills to operate a new appliance or technology are learned constantly. External facilities, such as infrastructure, services, and information, contribute to household resources. For this part, the system depends on factors that are outside their own range of influence. The availability and accessibility of these facilities are determined by the arrangements made by societal institutions. These may cause restrictions when the household system uses them in different modes of application. Also, a sudden change in facilities, like the supermarket moving to another part of town, might cause instability in the system, resulting in reallocations of resources. The relationship between humans and techniques is complex, as shown by the concepts used in domestic science. When a new technology used in household activities influences the level of living, the consequences will be evaluated against the standard of living and the feeling of wellbeing of the members. In the following paragraph, these relationships will be clarified.

4.

TECHNOLOGICAL INFLUENCES AND RESPONSES

When a new technology is introduced in a household system, the fragile balance between resources, level of living, and living standard may be disturbed, introducing a reaction. Four different types of interactions can be recognized, each with its own effect on the household system. They can be characterized by: 1. perceivability. Consumers will only react and change their performance levels if there are perceptible performance indicators. If the effects can or cannot be noticed by consumers within the frame of the household, no response is needed to maintain the usual routine and level of living. An example of this kind of influence is the substitution of phosphate by other builders in detergents. This reduces the environmental impact of domestic laundering, but does not change the processing or influence the results of the laundering process (Kim et al., 1987).

38

Chapter 4

2. action compatibility. Consumers get used to behavioral patterns in processes. A change in technological input is action-incompatible when behavioral patterns are introduced which differ from patterns executed before. This goes to the level of individual processes. If, for instance, a previously used appliance works with a system of menus, it takes less adaptation of skills when a new appliance uses the same system (Freudenthal, 1999). 3. logistic compatibility. Household resources are the backbone of logistics. If a technological change is incompatible with them, the way in which household activities are run and resources are used has to be altered. There are different kinds of logistical compatibilities: − Labor distribution compatibility − Time compatibility − Space compatibility − Financial compatibility In the case of logistical incompatibility, the usual domestic activities have to be rescheduled in order to stabilize the system (negative feedback). This may involve a complete redistribution of labor, time, space, and finances. The way in which this is achieved will greatly depend on the individual household situation. An example of this change in technology is the switch from shopping by car to shopping by bicycle. In both cases, the result is the same, but the time required is different. This implies that the time left for other activities is reduced. When, in this situation, money, for example, is not a limiting factor, it will be possible to trade time for money, acquiring time for other activities. One could, for example, buy a dishwasher, because as time-budget studies indicate, using a dishwasher saves time (Aldershoff and Baak, 1986). 4. functional compatibility. The functional performance level of a household activity is an important indicator for the level of living. When a technology is functionally incompatible, the performance level of a function is altered and may become incompatible with the living standard. In this situation, two different strategies to regain the balance in the domestic system can be recognized. The first strategy is aimed at restoring the level of living. In the system approach, this is achieved by a negative feedback mechanism linked to household activities. In that case, another combination of resources may be used. For instance, one could spend more money or put in more labor by asking someone of the family to help out. Or, in other words, a decrease in the level of living is corrected by an increase in resources. An example of the first response mechanism concerns domestic cleaning activities. In the past,

Technology and household activities

39

the process temperatures for these activities have been lowered in order to save energy. This change has affected the primary performance of the activities, the cleanliness of textiles. It has been shown that, e.g. the performance loss due to the lowering of the cleaning temperature of domestic laundering is visually perceptible to consumers (Terpstra, 2000). It is assumed that the increased washing frequency and use of pre-treaters and boosters in the past years are the consumers’ response to the previous changes. Research into the nature of this response is being conducted at Wageningen University (Roozemond, 1998). The second strategy to restore the balance is the adaptation in the standard of living for that particular service. This implies that a household adopts the new situation as its new standard. Here, positive feedback is encountered because an increase in performance (level of living) causes an increase in the standard of living, and vice versa. The automation of the washing process had such an effect. Performance levels rose and, with positive feedback, the standard for clean clothes changed in the same direction. In the early 1970s, households washed about 1.5 kg per person per week (Hoogland, 1978). At the end of the 1990s, one person produced more than 6 kg laundry per week (Uitdenbogerd et al., 1998; Groot-Marcus and Scherhorn, 1999). The expectations of time-saving when using washing machines did not work out (Groot-Marcus, 1984). Another example with positive feedback might be the introduction of the economy shower-head. Environmentally-conscious people may accept a possible decrease in comfort (lower level of living), because their standard of living has been adapted to a lower level.

5.

DISCUSSION AND RECOMMENDATIONS

The household system is sensitive to external influences, especially to the influence of introduced innovative technology. This is because technology can change the function fulfilled by processes in different ways. This leads to a disturbance of the balance between level and standard of living and to a dynamic response in the household system. Innovative technology has four different characteristics, each with its own typical effect on the household system. These characteristics are: − the perceivability of the effects by the household group, − the extent of congruence with regular household activities, − the logistic fit, and − the extent to which the function fulfillment of the new technology corresponds with the traditional technology.

40

Chapter 4

If a new technology scores high on these four characteristics, it may be implemented in the household system without substantial changes, the response of the household will be minimal. On the other hand, substantial hurdles must be overcome to implement and change the system if fewer of these characteristics are present. Some notes have to be made with respect to the previous general concepts. The domestic response mechanism does not always deliver an optimal situation for the household. In some situations, a response of the system is required, but not triggered by the technology characteristics. If, for example, the household group does not notice the change, the notions about the optimal daily care are not threatened, and revisions in household strategies will not be started. Sometimes, however, non-perceptible change might lead to undesirable situations because of this ‘invisibility’. An example is the change in hygienic quality due to low-energy processes. It has been shown that the hygienic quality deteriorates with low-energy laundering (Terpstra, 2001) and dishwashing. Due to the modification in the process, the microbial contamination is higher, but consumers do not perceive this. Since the deterioration in the hygienic level is unnoticed, in these cases, there will be no response from the household. The change may therefore increase health risks, unnoticed in the domestic setting. An interesting response of the system is the so-called rebound effect. People in general strive for more and better conditions in their daily life; they want to increase their standards. So if their actual level of living increases, there is a possibility that they will adopt this higher level as a new standard, and no correcting response occurs. As the model indicates, norms and values in society may influence their standards. Innovative technology as with environmentally-saving appliances, supported by a change towards environmentally-sound societal ideas and views, could have better results in the end. Special mechanisms show up when technological changes are linked with changes outside the system, such as altering facilities. These may influence the response to the change within the household system. The introduction of separate collection of solid domestic waste, for instance, was coupled with an integral and simultaneous overhaul of its infrastructure: providing households with tools and information and reorganizing collection services and sites. This neutralizes possible limitations in household resources.

Technology and household activities

41

The actual behavior is measured best through empirical studies at the level of implementation. At that level, it is easier to recognize the opportunities and constraints when changes are introduced in the household system. Research on the relationship between technology and behavior should, therefore, start with empirical investigations of real activities. In connection with inquiries about motives and goals for consumers’ behavior, the actual behavior gives a direct link to the information about the practical reasons for that behavior. Researchers can in advance try to map out possible behavioral consequences and include these in the inquiry. With this approach, the risk to measuring intended instead of actual behavior is reduced also. The respondents will be less inclined to give socially desired answers. When a change to a predefined aim is introduced, the characteristics of the change and the response of the household system have to be known in order to predict feasibility. Since the use of technology is related to the integral household system, designers must account for the context in which technology is implemented and processes that take place. It should be investigated whether they are compatible with the household system or not. If technological changes are predicted to influence the environmental impact of households, the response of the complete system has to be studied. Policy measures meant for households need to account for the whole system. Changes in technology might result in a new organization of society, and such changes influence the household system. The option to facilitate desired changes in housekeeping by changing the infrastructure or other facilities should be taken into account seriously.

REFERENCES Aldershoff, D.E. and W. Baak (1986). Huishoudelijke productie in verschillende huishoudtypen. Swoka Onderzoeksrapporten, nr 21, ‘s-Gravenhage. Deacon, R.E. and F.M. Firebaugh (1975). Home management, Context and Concepts. Houghton Mifflin Company, Boston. Egner, E. Der Haushalt (1952). Eine Darstellung seiner volkswirtschaftlicherGestalt. Duncker und Humblot, Berlin. Freudenthal, A. (1999). The design of home appliances for young and old consumers. PhD thesis. Serie ageing and ergonomics 2, Subfaculty of industrial design engineering, Delft University of Technology, Delft. Goldsmith, E.B. (1996). Resource management for individuals and families. West Publishing Company, Minneapolis/St. Paul. Groot-Marcus, J.P. and E. Scherhorn (1999). Ontwikkeling in energiegebruik in de huishouding. Relaties met huishoudelijk gedrag. H&C Working paper 9903, Wageningen.

42

Chapter 4

Groot-Marcus, J.P. (1984). Technologie, een oplossing voor de huishoudelijke arbeid?, Wetenschap en Samenleving, nr 2, pp. 28-37. Gross, I.H., E.W. Crandall, M.M. Knoll (1973). Management for modern families. Meredith Corporation, Appleton. Hoogland, J. (1978). De gezinswas belast het milieu op veschillende manieren. TNO-project, 610. Kim, J.O., B.F. Smith et al. (1987). Comparative study of phosphate and non-phosphate detergents. Journal of Consumer Studies and Home Economics, 11 (1987), pp. 219-235. Presvelou, C. (1980). Meer dus ook beter? Inaugurele rede, Landbouwhogeschool Wageningen, Wageningen. Roberts, N., D. Andersen, R. Deal, M. Garet, W. Shaffer (1983). Computer simulation; a system dynamics modeling approach. Addison-Wesley Publishing Company, Reading Massachusetts. Roozemond, D.A. (1998). On the relation between washing behavior and cleaning performance. In: Proceedings 38th International Detergency Conference, (Krefeld, 05-07 Mai 1998). Forschungsinstitut für Reinigungstechnologie e.V., Krefeld pp. 72-79. Schweitzer, R. von (1969). Die Haushaltanalyse. Hauswirtschaft und Wissenschaft, 17, no. 1 Spijkers-Zwart, S.I. De huishouding (1973). Een oriënterende studie naar de toepasbaarheid van concepten en theorieën in niet-westerse samenlevingen. Landbouwhogeschool, Huishoudkunde, Publicatie no. 1, Wageningen. Terpstra, P.M.J. (2000). Assessment of cleaning efficiency of domestic washing machines with artificial soiled test cloth. Institut für Landtechnik der Rheinisch Friedrich-WilhelmsUniversität Bonn.Rheinisch Friedrich-Wilhelms-Universität, Bonn. Terpstra, M.J. (2001). The correlation between sustainable development and home hygiene. American Journal of Infection Control, 29 4, pp. 211-217. Uitdenbogerd, D.E., N.M. Brouwer, J.P. Groot-Marcus (1998). Domestic energy saving potentials for food and textiles: an empirical study. Wageningen Agricultural University, Household and Consumer Studies, The Netherlands, (H&C onderzoeksrapport 3). Visser, C.W. (1969). Werk dat geen naam heeft. Diesrede Landbouwhogeschool, 10 Maart, Veenman, Wageningen. Zuidberg, A.C.L. (1981). Het verzorgingsniveau van huishoudens. SWOKA Onderzoeksrapporten, nr.4, ’s Gravenhage.

Chapter 5 TECHNOLOGY AND BEHAVIOR: Contributions from Environmental Psychology

Wim J.M. Heijs

1.

INTRODUCTION

Denominational segregation is apparent in science as well as in society. Most obvious are the boundaries between the technological and social sciences. But within the latter domain, too, disciplines are often not aware of the work done by others. This is one of the reasons for writing this book, because tackling problems related to the use of technology (like increasing energy consumption or safety issues) requires a joint effort employing each others’ expertise. Societal problems involving the interaction with the (technological) environment, along with difficulties encountered in psychological research of them since the prevailing theories did not sufficiently incorporate influences of the physical circumstances, induced the development of environmental psychology in the 1960s (Stokols and Altman, 1987).9 This discipline examines relations between man-made environments and psychological processes. Being rooted in mainstream psychology, its theory and research are mostly aimed at uncovering causality and prediction in a traditional sense (unidirectional), adding environmental elements as independent variables. In line with philosophical notions of Dewey, Bentley 9

, The roots of the discipline can be traced even further back if we consider Lewin s field theory (1936), which describes behavior as a function of person and environmental variables, and adjoining areas like engineering psychology and human factors (see Nickerson, 1999). Only the name environmental psychology was not yet used.

43 P.-P. Verbeek and A. Slob (eds.), User Behavior and Technology Development: Shaping Sustainable Relations Between Consumers and Technologies, 43-52. © 2006 Springer.

Chapter 5

44

and Pepper, this is called an interactional view (Altman and Rogoff, 1987). Two other approaches based on the same notions have won ground, and are now more or less typical of environmental psychology. The organismic view implies that an event can only be adequately understood if the holistic system of person and environment is considered (the whole is more than the sum of the parts). Research focuses on the reciprocal relations between these elements, and on the effects on the system. Usually, the functioning of a system is assumed to be governed by a teleological principle (a basic goal underlying its actuality) such as a state of balance or need reduction. The second is a transactional perspective, which stresses the change of relations in holistic systems over time (processes) without assuming a teleological rule. Presently, the organismic and transactional views are often considered more suitable to study interaction than the ‘interactional’ one. The accents on solving problems and relations between qualitatively different entities (personal and environmental) requires that research be multidisciplinary and eclectic, using theories and methods from other fields that are useful for the problem at hand. Efforts are made to integrate such insights into new frameworks, and to translate results of research into guidelines for improving environmental elements or for some new design. Although research in environmental psychology has mainly been directed at the built environment, a number of concepts can also be applied to technological applications in a narrower sense. These concepts will be discussed in the next sections. First a model of information processing is presented because it provides a frame of reference for the other concepts, and also because it can be used itself as a medium for examining the interaction between users and technology.

2.

INFORMATION PROCESSING

The reciprocity of user and environment/technology can be conceptualized as an encounter of the results of internal processes in both: behavior resulting from information processing, and output resulting from in- and throughput. Since environmental processes are very diverse, depending on the environment or the technology that is involved, it is easier to describe the encounter on the basis of a general model of information processing on the user side. The model also broadly captures influences from environmental processes. Figure 5-1 shows the components of the model and their interaction (in normal print; the concepts in italics are discussed in section 3). Given the

Technology and behavior

45

subject matter of this book, the physical environment is seen as man-made. It has a special bearing on equipment, but built structures (houses or roads) are not excluded as they too influence societal problems, such as energy use or safety. Two user categories are distinguished: individuals and groups (e.g. households). Their interaction is represented by a heuristic of information processing. In specific cases, i.e. when environmental factors are less diverse, it would be feasible to construct a model that is more organismic, dealing with both sides simultaneously by shifting the information process to the right (the user side), and by filling in the corresponding technological processes on the left, environmental side (examples are mentioned in Table 5-1). environment technology (man-made)

interaction

users (individuals groups)

visibility sensation

gestalt laws

affordances perception

feedforward and feedback aesthetics

mapping cognitive interpretation affective evaluation

lens model

mental model schemata script

attitude, habit perceived control, subjective norms intention, planning

P-E fit

constraints and rebound

behavior Figure 5-1. General model of information processing and selection of relevant concepts

Sensation is assumed to be the starting point, although this is rather arbitrary because the process is cyclic. This concept refers to the discernment of stimuli by the sensory organs (e.g. visual, auditory or olfactory). Because users are, so to speak, continuously loaded with sensory input, it is necessary to select information and then to synthesize what is left into a mental image for further processing. This is done by the process of

46

Chapter 5

perception.10 The mental image has to be recognized in order to become an impetus for behavior, and often it must also be related to an emotion to determine the direction of the behavior. In cognitive interpretation and affective evaluation, cognitive and emotional labels are assigned by crossreferencing the image to existing knowledge. Between this stage and actual behavior, various mechanisms are assumed to operate, denoted by concepts such as attitude, habit, perceived control, subjective norms, intention, and planning, which will not be elaborated here (see the chapter on habitual behavior in part two of this book). Behavior is, in its turn, accompanied by new sensations. Table 5-1. Influences from environment and user sides, and general guidelines 1. Sensation influence env.: the filter is not visible user: elderly, having trouble bending over, do not notice the filter rule elements necessary for information processing must be noticeable to the relevant senses of the target group, and those that are unnecessary or distracting should be concealed (e.g. large primary buttons on the video control and the concealing of others) 2. Perception influence

rule

3. Cognition influence

rule

env.: the filter is overlooked because it does not stand out clearly against the surface user: it does not present a self-contained mental image of something that is serviceable elements must evoke mental images that are clear (unambiguous and attracting attention) and correct (corresponding to the intended use and the principles of perception) (e.g. a switch for an energy-efficient washing program that stands out against the panel) env.: there is no instruction or clue on how to unscrew the filter user: it looks like something that is known to be intended for maintenance by experts elements must, if necessary, give information about their use or consequences (instructions, feedback), by referring to present knowledge or taking present knowledge as a starting point (e.g. symbols on a remote control with established meanings) continued

10

Perception and cognition have a long history, and their meaning has changed often. It should be stressed that the diagram is merely intended as a heuristic device, and not as a representation of the latest understanding.

Technology and behavior

47

Table 5 -1 continued 4. Affection influence env.: the filter is inconveniently placed, so it is prone to receive a negative emotional label user: laymen fear that they have insufficient knowledge to operate the filter rule elements should evoke emotional responses that match the desired direction of behavior, or be neutral if no particular behavior is expected (e.g. attractive controls if they assist sustainable behavior, and plain if not) 5. Intermediate influence

rule

6. Behavior influence rule

env.: the filter is not sufficiently controllable by the user user: it causes a negative attitude or activates an unwanted habit (see part 2 of the book) elements must carry the appropriate information and affective connotations to form stable attitudes for sustainable behavior, and convey the amount of control the user has (e.g. a room thermostat that looks easy to operate, and a boiler setting that does not) env.: unscrewing the filter requires special tools, or it is stuck user: users lack the physical ability to unscrew the filter elements must make clear that the intended behavior is feasible, and the best (or only) alternative; unwanted behavior must meet with an apparent impossibility to carry it out (e.g. design can make use more comfortable or convenient, compared to alternatives)

All stages in the process are influenced by characteristics of the environment and the user (the grey arrows moving inward). These influences can serve to analyze problems (e.g. energy-inefficient interaction) and to derive guidelines for prevention and (re)designing technology. Using the filter of a washing machine as an example (more energy is used when the filter is not cleaned regularly), Table 5-1 states the nature of possible negative influences in the different stages of information processing, and gives examples of general rules that may be inferred, together with a possible application of these rules. Most guidelines in the Table are obvious but, nevertheless, they are often disregarded. Well-known shortcomings are: too many buttons, illegible symbols, obscure instructions, parts that are hard to find, and alternatives that reduce comfort or are behaviorally or financially unattractive. In reality, of course, the matter is much more complex. Subprocesses share feed-forward and feedback loops (e.g. cognition also steers perception, because experience influences selection processes). Furthermore, the two sides contain many variables (on the user side: structural characteristics,

48

Chapter 5

knowledge, needs, and also visualization and orientation abilities). These can be more or less relevant, and may have a bearing upon each other (for example, an illness that affects information-processing).

3.

CAPITA SELECTA

Formulating rules for analysis and design (in general or for individual appliances) may form a good starting point for promoting efficient interaction. The composition of rules requires an understanding and a theoretical underpinning of the aspects concerned. A number of concepts that can help to provide this are mentioned in Figure 5-1. They are positioned according to their orientation (on the user side -U-, the environmental/technological side -T-, or the interaction between them), and also according to the stage(s) in the information process in which they are most relevant. Because of the limited space in this chapter, descriptions must inevitably be brief. Starting in the middle (interaction), a concept with special relevance is ‘affordances’ (Gibson, 1979). Generally speaking, affordances are combinations of physical properties of objects that invite a certain use. They are perceived earlier than the separate properties themselves; they steer behavior directly, without the mediation of cognition or evaluation; and they can differ between groups of users. A switch can, for instance, be perceived as ‘turn-able’ or ‘push-able’, thus inviting a particular behavior without a conscious elaboration of the properties that constitute that reaction. And this perception may depend on the knowledge and experience of the persons involved. Affordances are not subjective: they are real and invariant (not depending on temporary needs). They are not objective either (whereas the properties of the object are). Affordances exist in the presence of objects and users as a result of interaction on a perceptual level. A further understanding of affordances of appliances, and of the object properties constituting them for a particular group of users, is that they can be utilized to steer behavior at a very early stage of progress, before it becomes manifest (this was done, e.g. in the design of aviation controls in WWII). A considerable body of research and theoretical additions are available in the subfield of ecological psychology (not Barker’s (1968) theory that has the same name; see Flach et al., 1995; and Rasmussen, Pejtersen and Goodstein, 1995). ‘Mapping’ is a term used by Norman (1988), in conjunction with visibility, mental models and feedback, to describe prevalent ways in which design influences behavior, and vice versa. It refers to the correspondence

Technology and behavior

49

between possibilities to act as suggested by the design and the intentions or goals of the user. These should match to prevent unwanted use, bad experiences, or even accidents. The appearance and function of serviceable parts of appliances should, for example, conform to the expected functionality and to experience with similar devices (a lack of correspondence between the position of burners on a cooking-range and their controls might lead to burned food; or a tap looking like a volume button might be turned the wrong way). Visibility is important for sensation and perception, and knowledge about mapping for composing general rules in the perception and cognitive stages (mental models are dealt with later and feedback is discussed in part two of the book). Norman has published many ideas on how to make designs more usable, and how to foster desirable behavior, based on sound theoretical grounds. Brunswik’s ‘lens model’ (1965) involves a theory and a method to describe and evaluate judgments and behavior as a function of both environmental attributes and user variables. The various attributes of an object or situation (‘cues’) are assigned weights to express their relative objective importance for an accurate judgment (the ‘ecological validity’ of the cues). The users, on the other hand, will attach their own weights based on personal variables, such as demographic traits, experience or personality (this is labeled ‘cue utilization’). The model includes methods to determine ecological validity, cue utilization and the identification of user variables. This is helpful to assess the recognition and estimation of important cues of objects while they are in use and also beforehand (in the design process). If, for example, certain controls for sustainable energy use on a washing machine are not employed, the method may show which attributes of the object and characteristics of the user are responsible. Congruence or person-environment fit (‘PE-fit’) models describe the joint influence of user and environmental factors as well, but with a primary focus on the end results: behavior and well-being (Caplan, 1983; Kahana, 1975; Slangen, 1999). They are rooted in Lewin’s field theory (1936) and Murray’s need-press model (1938), in which behavior is described as the result of motivational forces from personal needs and the environment. Two types of fit are distinguished: between (1) personal needs/values and environmental opportunities/supplies, and (2) personal abilities/competence and environmental demands. In ideal cases, needs match supplies and abilities match demands (e.g. a washing machine must supply the programs that meet the settings preferred by the users, and users must be able to meet the demands that operating the machine poses). A ‘misfit’ causes psychological strain, which has an impact on well-being and behavior. Designs that take this dual

50

Chapter 5

fit into account will result in a stronger motivation and a better performance. Several instruments are available to measure user needs and PE-fit (Caplan, 1983), and new ones are being developed (e.g. Heijs, 2003). Other concepts in Figure 5-1 are less focused on interaction but can assist in understanding the process from the user and environment perspectives The ones on the user side are taken from cognitive psychology, and they represent resources for interpretation and behavioral decision-making. Mental models (or naive theories) contain the presumed working of systems, based on everyday experience. Although not necessarily correct, their existence is reinforced by an apparent effectiveness. An example is the ‘valve-theory’ of thermostats (Kempton, 1986). According to this mental model a thermostat works like a faucet, and using it in such a way often has the intended result. In other cases, however, behavior may be less adequate if the mental model is at odds with reality, or if the models of the user and the designer do not match (examples are given by Norman, 1988). An interesting question in this connection is whether it is more appropriate to change the user model to match reality (by means of instruction) or to adapt a device to simulate a correspondence with the user model. ‘Schemata’ are cognitive structures in which knowledge about objects or events is organized (Rumelhart, 1980). They guide interpretation and the search for new information (e.g. in instructions or displays). The concept of ‘scripts’ describes memory as an episodic series of actions related to situations (Schank and Abelson, 1977). Scripts allow the filling-in of absent information, and may therefore be defective in slightly different situations: a laundry script implying washing on Mondays may lead to more energy use if there is not enough to fill the machine.11 Other theories in cognitive psychology, less frequently adopted in environmental psychology, may also be of interest (e.g. cognitive style and memory structures in relation to interface design). On the environmental side, gestalt psychologists in the 1930s formulated laws that, among other things, clarify the perception of unity of objects and of the discrimination between an object and its background. Stimuli are grouped according to their nearness, similarity, the tendency to distinguish a ‘thing’ shape, and simplicity. Design following these laws may promote the intended behavior (e.g. by making essential controls more perceptible; see Ellis, 1938 and Wertheimer, 1959). Feedforward and feedback refer to the information provided the environment that initiates and guides interpretation 11

More recently, the concept of ‘script’ was introduced in the philosophy of technology, but with a different meaning (see chapters by Verbeek and Jelsma in this volume).

Technology and behavior

51

and behavior (see part two of this book). Aesthetical evaluation (an aspect of emotional labeling) is described in Berlyne’s arousal model (1972) and the preference model of Kaplan and Kaplan (1982). The first model states that aesthetical experience (and also interpretation) depends on environmental attributes that cause moderate psychological ‘arousal’ by inducing a certain amount of visual uncertainty and subsequent investigatory action. These attributes (‘collative properties’) are: complexity, novelty, incongruity and surprisingness. In the second model, environmental preferences are seen as depending on the degrees of coherence, legibility, complexity and mystery of the environment. These factors relate to the perceived opportunities for immediate and future activities, and to the impression that an environment is understandable. The last concepts pertain to the fact that technological interventions that will constrain user behavior risk the manifestation of rebound activities (e.g. s: safety measures provoking new risks, or energy savings leading to an increase of lighting or ventilating). Further adverse actions are avoidance (not using an eco-program on a washing machine because it interferes with existing habits) and reactance (counteracting a measure that is felt as a constraint and as limiting personal control; Brehm, 1966). In the chapter on habits in part two of this book, suggestions are made for decreasing the risk of rebound activities occurring. In the case of energy-related behavior, however, this may be difficult because earnings are likely to be spent again, thus increasing energy use in one way or another. Discovering possibilities to avoid this is a major challenge in the future.

4.

CONCLUSION

Overlooking the modest review above, the recognition that knowledge is needed of the reciprocity of environment and user to bring about desired behavior is not a new one. Existing theories, concepts and methods in environmental psychology and related fields provide many opportunities for analyzing and ameliorating human interaction with technology. But to fulfill these tasks adequately, the relevant disciplines must be willing to look beyond their own turfs, to share insights, to identify common grounds and knowledge gaps, and to work towards a joint framework.

52

Chapter 5

REFERENCES Altman, I. and Rogoff, B. (1987). World views in psychology: trait, interactional, organismic and transactional perspectives. In: D. Stokols and I. Altman (eds.), Handbook of Environmental Psychology. NY: Wiley. pp. 7-40. Barker, R. (1968). Ecological psychology: concepts and methods for studying the environment of human behavior. Stanford: Univ. Press. Berlyne, D. (1972). Esthetics and psychobiology. NY: Appleton-Century-Crofts. Brehm, J. (1966). A theory of psychological reactance. NY: Academic Press. Brunswik, E. (1965). Perception and the representative design of psychological experiments. Berkeley: Univ. of Calif. Press. Caplan, R. (1983). Person-environment fit: past present and future. In: C. Cooper. Stress Research, NY: Wiley. pp. 35-74. Ellis, W. (1938). A Source Book of Gestalt Psychology. NY: Harcourt. Flach, J., Hancock, P., Cairo, J. and Vicente, K. (1995). Global perspectives on the ecology of human-machine systems. Hillsdale: Erlbaum. Gibson, J. (1979). The ecological approach to visual perception. Hillsdale: Erlbaum. Heijs, W. (2003). Founding houses for the elderly: on housing needs or dwelling needs? In: R. Garcia Mira, J. Sabucedo en J. Romay (eds.). Culture, Environmental Action and Sustainability. Göttingen: Hogrefe & Huber. pp. 367-383. Kahana, E. (1975). A congruence model of person-environment interaction. In: P. Windley, T. Byerts and F. Ernst (eds.). Theoretical development in environments for aging. Washington: Gerontological Society. pp. 181-214. Kaplan, S. and Kaplan, R. (1982). Cognition and environment: functioning in an uncertain world. NY: Praeger. Kempton, W. (1986). Two theories of home heat control. Cognitive Science, 10, 75-90. Lewin, K. (1936). Principles of topological psychology. NY: McGraw-Hill. Murray, H. (1938). Explorations in personality. NY: Oxford University Press. Nickerson, R. (1999). Engineering psychology and ergonomics. In: P. Hancock (ed.). Human Performance and Ergonomics. San Diego: Academic Press. pp. 1-45. Norman, D. (1988). The psychology of everyday things. NY: Basic Books. Rasmussen, J., Pejtersen, M. and Goodstein, L. (1995). Cognitive systems analysis. NY: Wiley. Rumelhart, D. (1980). Schemata: The building blocks of cognition. In R. Spiro, B.Bruce and W. Brewer (eds.). Theoretical Issues in Reading and Comprehension. Hillsdale: Erlbaum. Schank, R.C. and Abelson, R. (1977). Scripts, Plans, Goals, and Understanding. Hillsdale, NJ: Earlbaum Assoc. Slangen, Y. (1999). A tale of two adaptations. Dissertation. Eindhoven: University of Technology. Stokols, D. and Altman, I. (1987). Introduction. In: D. Stokols and I. Altman (eds.), Handbook of Environmental Psychology. NY: Wiley. pp. 1-4. Wertheimer, M. (1959). Productive thinking. NY: Harper & Row.

Chapter 6 ACTING ARTIFACTS: The Technological Mediation of Action Peter-Paul Verbeek

1.

INTRODUCTION

During the past decade, the role of technological objects in society and in people’s everyday lives has become a central theme in the philosophy of technology. Many authors have made analyses of the ways in which technological artifacts help to shape the ways in which humans experience reality and live their lives. A central theme in these analyses is ‘mediation’: artifacts are conceptualized in terms of their mediating role in the relationship between human beings and their environment. In this paper, I will draw together the main lines of thought that can be found in some recent approaches. This will result in a conceptual ‘vocabulary’ for understanding technological mediation.

2.

TECHNOLOGY AND HUMAN-WORLD RELATIONSHIPS

The approach I will follow in developing a philosophical framework for understanding the influence of artifacts on people’s actions and experiences is phenomenological in nature (cf. Verbeek, 2005). This framework needs some explanation before turning to the contribution of the philosophy of technology to the analysis of the relationship between technology and behavior. I will define phenomenology broadly as the philosophical analysis of human-world relationships. The central idea in the phenomenological approach is that subject and object ʊ or: humans and their world ʊ constitute 53 P.-P. Verbeek and A. Slob (eds.), User Behavior and Technology Development: Shaping Sustainable Relations Between Consumers and Technologies, 53-60. © 2006 Springer.

54

Chapter 6

each other in the relationships that exist between them. Humans and their world are always interrelated. Human beings cannot but be directed at the world around them; they are always experiencing it, and it is the only place where they can live their lives. Conversely, the world only gets a meaning for human beings in the relationships they have with it: it needs to be perceived and interpreted in order to be meaningful. Humans and their world, therefore, determine each other in the relations and interactions that exist between them. In their interrelation, both the subjectivity of humans and the objectivity of the world are shaped. This phenomenological perspective offers a framework for analyzing the relationship between technology and behavior, which is the central topic of this book. Technological artifacts are related to human behavior, because they can play a mediating role in the very relation between human beings and their world. A good starting point for understanding this ‘technological mediation’ is the analysis of the relationships between humans and artifacts, as made by the American philosopher of technology Don Ihde. Ihde (1990) discerns several relationships human beings can have with technological artifacts. Firstly, technologies can be ‘embodied’ by their users, making it possible that a relationship comes about between humans and their world. This ‘embodiment relation’, for instance, occurs when looking through a pair of glasses; the pair of glasses is not noticed explicitly but yet it coshapes the relationship between human beings and their environment. We do not look at a pair of glasses, but through it to the world around us. In the ‘embodiment relation’, technological artifacts are ‘incorporated’, as it were; they become extensions of the human body. Secondly, technologies can be the terminus of our experience. Ihde calls this relation with technologies the ‘alterity relation’. It occurs when interacting with a device, as is the case, for example, when buying a train ticket at an automatic ticket dispenser. Thirdly, technologies can play a role at the background of our experience, creating a context for it.12 An example of this ‘background relation’ is the automatic switching on and off of the refrigerator, or the temperature condition in a room as produced by a heater or air conditioner. Such devices are not experienced directly, but shape a context within which experiences can take place. This third relationship is of less importance than the other two for understanding technological mediation, and therefore I will not discuss it further. Ihde’s conceptualization of human-technology relations is based on the analysis that was made by the German philosopher Martin Heidegger of the 12

For the sake of clarity, I leave out a fourth human-technology relationship Don Ihde discerns: the ‘hermeneutic relation’. For an analysis of this relation, see: Ihde, 1990.

Acting artifacts

55

role of tools in the everyday relation between humans and their world. According to Heidegger (1927), tools should not simply be understood as functional instruments, but as ‘connections’ or ‘linkages’ between humans and reality. The way tools-in-use are present Heidegger calls ‘readiness-tohand’. By this he means that tools that are used for doing something typically withdraw from people’s attention. The attention of a person who drives a nail into a wall, is not primarily directed at the hammer, but at the nail. Human involvement with reality takes place through the ready-to-hand artifact. Only when the artifact breaks down, it demands attention for itself again. It then becomes ‘present-at-hand’; it presents itself as the terminus of our experience and cannot facilitate a relationship between user and world any more. Table 6-1 schematizes these human-technology relationships. Table 6-1. Human-technology relationships (after Ihde, 1990) embodiment relation (human – technology) Æ world (‘readiness-to-hand’) alterity relation (‘presence-at-hand’)

human Æ technology (– world)

background relation

human (– technology – world)

The Concept of readiness-to-hand is of utmost importance for understanding the relationship between technology and behavior. Artifacts that are ready-to-hand are able to bring about a relationship between human beings and their world. By withdrawing from people’s attention, they create ‘through themselves’ a relation between user and world. Artifacts facilitate the involvement of human beings with reality, and in doing so, they help to shape how human beings can be present in their world and how their world can be present for them. Things-in-use, therefore, can be understood as mediators of human-world relationships; they form a ‘medium’ between human beings and their world. Mediation should be understood in an active sense here. Artifacts are not merely neutral ‘intermediaries’, but actively help to shape human-world relationships: human perceptions and actions, experiences and existence. The mediating role of artifacts, however, does not only occur from the ‘embodiment’ or ‘ready-to-hand’ relation. As will become clear below, in the domain of action artifacts can also play a mediating role from the ‘alterity’ or ‘present at hand’ position. The work of Don Ihde and the French philosopher and anthropologist Bruno Latour offer concepts for making a closer analysis of this mediating role of technologies. In order to link their analyses to each other, I will discern two directions of phenomenology: one that focuses on perception and one on praxis. Each of these directions approaches the human-world

56

Chapter 6

relationship from a different side. Existential or ‘praxis-oriented’ phenomenology starts from ‘the human side’. Its central question is how human beings act in their world and realize their existence. The main category here is action. Hermeneutic or perception-oriented phenomenology starts from ‘the side of world’, and directs itself at the ways in which reality can be interpreted by and be present for people. The main category here is perception. The phenomenological point of view, therefore, makes it possible to analyze technological mediation in terms of the role technological artifacts play in the interrelation between humans and their world, by helping to shape human actions and perceptions.

3.

MEDIATION OF PERCEPTION

The central hermeneutical question for a ‘philosophy from the perspective of things’ is how artifacts mediate the way reality can be present for people. As Don Ihde’s philosophy of technology shows, artifacts help to shape human experiences and interpretations (Ihde 1990). Ihde’s work focuses on the technological mediation of perception. Artifacts are able to mediate our sensory relationship with reality, and in doing so they transform what we perceive. According to Ihde, this transformation always has a structure of amplification and reduction. Specific aspects of reality are amplified while others are reduced. When looking at a tree with an infrared camera, for instance, most aspects of the tree that are visible to the naked eye get lost, but at the same time a new aspect of the tree becomes visible: one can now see whether it is healthy or not. Ihde calls this transforming capacity of technology ‘technological intentionality’. In their mediation of the relationship between humans and world, technologies have ‘intentions’; they are not neutral instruments but actively help to shape the nature of the relationship that comes about. ‘Technological intentionalities’ are not fixed properties of artifacts, however. They get shape within the relationship humans have with artifacts. Within different relationships, technologies can be interpreted differently and therefore have a different intentionality. The telephone, for instance, was originally developed as a hearing aid, and the typewriter as a writing tool for people who are suffering from weakness of vision. In their use contexts, these technologies came to be interpreted in a different way than their designers intended. Ihde calls this phenomenon ‘multistability’. A technology can be ‘stable’ in different ways, in that its ‘essence’ is not fixed but depends on the way it is embedded in a use context. Technological

Acting artifacts

57

intentionalities are always dependent on the specific ways in which technologies are interpreted and used. Ihde’s analysis of the transformation of perception has important hermeneutical implications: mediating technologies not only determine human perceptions but also our interpretations of reality. This becomes most clear when investigating the role of instruments in the production of scientific knowledge. Without these, many scientific facts and theories could not exist. Instruments make it possible for scientists to perceive aspects of reality that cannot be perceived without them, like brain activity, microorganisms, or invisible forms of radiation emitted by stars. The ‘reality’ studied here, has to be ‘translated’ by technologies into perceivable phenomena. What ‘reality’ is in such situations, is co-shaped by the instruments with which it is perceived. It has no equivalent in the visible world. Medical technologies form another example of the role of technologies in human interpretation. Ultrasound scans, for instance, can be used to test nuchal translucency, the thickness of the skin at the nape of a fetus’ neck. This test gives an indication of the risk of Down’s syndrome. If this test is done, the echoscope lets the fetus be present in a very specific way. For those who will have to make a decision about abortion on the basis of the outcomes of the test, the fetus can be present only in terms of an organism with a risk of suffering from a serious disease. And the very act of having this test done already suggests an appropriate response. Ultrasound scans fundamentally shape one’s experience of an unborn child, and even of being pregnant. On the basis of their mediating role in human perceptions and interpretations, therefore, technologies can indirectly influence human actions as well. This holds true not only for medical technologies, but also for many technological interfaces, which mediate the way in which humans perceive and interpret the functioning of a device. A washing machine that indicates that its water filter needs to be cleaned, for instance (see Heijs, chapter 15, this volume), mediates how users interact with the machine, and how much energy they use. This indirect role of technologies in human behavior is of a different nature, though, than the direct influence which will be discussed below.

4.

MEDIATION OF ACTION

Within the existential or praxis-perspective, the central question is how artifacts mediate the actions of human beings and the way they live their

58

Chapter 6

lives. From a phenomenological point of view, praxis and existence are the mirror images of perception and experience. Whereas perception consists in the way the world is present for humans, praxis can be seen as the way humans are present in their world. The work of Bruno Latour (1992; 1994) offers many interesting concepts for analyzing how artifacts mediate human praxis. Latour points out that artifacts influence actions: what humans do is often co-shaped by the things they use. Actions are not only the result of individual intentions and the social structures in which these individuals find themselves (the classical sociological agency-structure dichotomy), but also of people’s material environment. A speed bump, for instance, translates a driver’s intention from ‘driving fast, because I’m in a hurry’, or ‘driving slowly in order to behave responsibly’, to ‘driving slowly to save my shock absorbers’. And the introduction of the microwave oven has not only enabled people to heat their food in a faster way, but has also changed their eating patterns. Since a microwave oven is particularly suitable for heating one-person, deep-frozen, ready-made meals, it appears to invite people to eat individually. By doing so, it weakens the ‘culture of the table’. Latour’s concept for describing the mediation of action by artifacts is ‘script’ (Latour 1992). Like the script of a movie or a theater play, an artifact can ‘prescribe’ its users how to act when they use it. A speed bump, for instance, has the script ‘slow down when you approach me’; a plastic cup from a coffee machine says ‘throw me away after use’. When scripts are at work, things mediate action in a material way, which should be clearly distinguished from the immaterial or informational way in which signs mediate human behavior as well. A traffic sign, for instance, makes people slow down in quite a different way than a speed bump ʊ if it does so at all. And people do not discard a plastic coffee cup because its user’s manual tells them to do so, but simply because it is physically not able to survive being cleaned several times. The influence of technology on human actions is of a non-linguistic kind. Things are able to exert influence as material things, not only as carriers of meaning. According to Latour, scripts are often, though not always, the products of ‘inscriptions’ by designers. Designers anticipate how users will interact with the product they are designing and, implicitly or explicitly, build prescriptions for use into the materiality of the product. Latour describes this inscription process in terms of ‘delegation’: designers delegate specific responsibilities to artifacts. To a speed bump, for instance, the responsibility was delegated to make sure nobody drives too fast. Not all scripts are the result of deliberate inscription, though. Artifacts can have scripts without these having been explicitly intended, whereas explicitly-intended scripts

Acting artifacts

59

can work out in a different way than expected. Wheelchair users know all about this: to artifacts like revolving doors and thresholds the responsibility is delegated to keep out the draft, not to keep out people in a wheelchair as well. Their discriminating scripts were not deliberately inscribed into them by their designers ʊ but nevertheless they exert their influence in many buildings. As with perception, in the mediation of action transformations occur. Within the domain of action, these transformations can be indicated as ‘translations’. For Latour, all entities ʊ human and nonhuman ʊ possess programs of action. And according to him, artifacts bring about ‘translations’ of these programs. By entering a relationship with another entity, the program of action of the original actor is translated into a new one. When somebody’s action program is to ‘prepare meals quickly’, for instance, and this program is added to that of a microwave oven, the action program of the resulting, ‘composite actor’ might be ‘regularly eating instant meals individually’. In the translation of action, a similar structure can be discerned as in the transformation of perception. Just as in the mediation of perception some aspects of reality are amplified and others are reduced, in the mediation of action specific actions are ‘invited’, while others are ‘inhibited’. The scripts of artifacts suggest specific actions and discourage others. The nature of this invitation-inhibition structure is context-dependent, as is the amplificationreduction structure of perception. Ihde’s concept of multistability also applies within the context of the mediation of action. The telephone, for instance, has had a major influence on the separation of people’s geographical and social context, by making it possible to maintain social relationships outside their immediate living environment. But it could only have this influence because it is used as a communication technology, not as the hearing aid it was originally supposed to be. An important difference with respect to the mediation of perception, however, is the way in which the mediating artifact is present. Contrary to perception, artifacts not only mediate action from a ready-to-hand position (Ihde’s ‘embodiment relation’), but also from a present-at-hand position (Ihde’s ‘alterity relation’). A gun, to mention an unpleasant example, mediates action from a ready-to-hand position, translating ‘express my anger’ or ‘take revenge’ into ‘kill that person’. A speed bump, however, cannot be embodied or ready-to-hand; it exerts influence on human actions from a present-at-hand position.

60

5.

Chapter 6

CONCLUSION: A VOCABULARY FOR TECHNOLOGICAL MEDIATION

Within phenomenological philosophy of technology, several concepts have been developed for analyzing the influence of technologies on people’s actions and perceptions. This influence can be described in terms of mediation. Artifacts mediate action by means of ‘scripts’, which prescribe how to act when using the artifact. They mediate perception by means of technological intentionalities: the active and intentional influence of technologies. Technological mediation appears to be context-dependent, and always entails a translation of action and a transformation of perception. The translation of action has a structure of invitation and inhibition, and the transformation of actions a structure of amplification and reduction. Table 6-2 below draws together all relevant concepts. Table 6-2. A Vocabulary for Technological Mediation hermeneutical perspective

praxis perspective

mediation of perception technological intentionality transformation of perception amplification and reduction

mediation of action script translation of action invitation and inhibition

delegation: deliberate inscription of scripts and intentionalities multistability: context-dependency of scripts and intentionalities

REFERENCES Heidegger, M. (1927), Sein und Zeit. Tübingen: Max Niemeyer Verlag {1986}. Ihde, D. (1990), Technology and the Lifeworld. Bloomington/Minneapolis: Indiana University Press. Latour, B. (1992), ‘Where are the Missing Masses? – The Sociology of a Few Mundane Artifacts’. In: W.E. Bijker and J. Law (eds.), Shaping Technology/Building Society. Cambridge: MIT Press, pp. 205-224. Latour, B. (1994), ‘On Technical Mediation: Philosophy, Sociology, Genealogy’. In: Common Knowledge 3, pp. 29-64. Verbeek, P.P. (2005), What Things Do – Philosophical Reflections on Technology, Agency, and Design. Pennsylvania: Pennsylvania State University Press.

Chapter 7 TECHNOLOGY AND BEHAVIOR: A View from STS Jaap Jelsma

1.

INTRODUCTION

‘Technology and behavior’ was the flag under which the project leading to this volume was initiated. The word and between technology and behavior indicates in the first place that treating both as separated phenomena is common practice. In the second place, it suggests that technology has no behavior, and that behavior is not technical. In doing so, it reinforces the traditional split between natural and socio/political phenomena. On the other hand, this volume expresses the intention to investigate how technology and behavior might be related. Restoring the relation is not a simple task. It implies bringing together what has been taken apart long ago, in an extended process reflected in the breaking up of our understanding in separate domains such as philosophy, natural and social sciences, and engineering. For this reason, it is unavoidable that any effort to re-unite technology and behavior conceptually has to dig into the foundations of our understanding about our world and ourselves. Here we come across another ‘and’: the world and ourselves. This is a major dichotomy among the many ones you will discover as soon as you dive into the history of man’s reflection about the human condition. The philosophical separation between subject and object, between living humans in a society and dead matter in a universe, has generally been seen as a precondition for the first to study the second. This precondition was not only epistemological, but also political (Van den Daele 1978). However, underneath the fruitful dichotomy between objective knowledge and knowing subjects, the embers of doubt remained simmering (will the world coincide with the world we know?), fanned every now and 61 P.-P. Verbeek and A. Slob (eds.), User Behavior and Technology Development: Shaping Sustainable Relations Between Consumers and Technologies, 61-70. © 2006 Springer.

62

Chapter 7

then by struggles such as the one between idealism and realism. Once, phenomenology pretended to reconcile this dilemma by enthroning intentional human consciousness, but soon it was the world that was felt to be too remote (see Verbeek 2000). In social science, we recognize similar recurrent tides in casting the relationship between humans and the world (here called ‘society’). There are ongoing dialectics between the postulation of dichotomies and their subsequent mitigation or fusion by ‘faces of the same medallion’ kinds of reasoning. Here the main dilemma is to link the actor (the acting subject) conceptually with constraints in his environment (usually called ‘structure’). While in psychology (especially environmental psychology) this structure may be physical, in sociology structure is always social structure. Introducing the role of social structure in determining the behavior of humans was Durkheim’s innovation, but Giddens claims to have fused agency and structure again (Giddens 1984). Bourdieu was out on a similar mission in his attempt to overcome the cleft between structuralism and Garfinkelian social phenomenology, by postulating a dialectical relationship between mental and social structure as reflected in the duality of habitus and field (Bourdieu and Wacquant 1992). But where are the things, the products of technology? In fact, things are everywhere and the appearance of new ones grows faster and faster while living species disappear. Nevertheless, things are still difficult to integrate into our intellectual fabric. Navigating through an archipelago of knowledge we are still fishing for things. In philosophical territory, postphenomenology has found a new role for technology as a mediator between humans and their ‘life world’ (see Verbeek, this volume). In a new terrain, populated by a new brood, the question that interests us here has been posed even more explicitly and systematically: ‘Where are the missing masses?’ (Latour 1992). It is the field of science and technology studies I am referring to, and to which I will take you to see which answers have been found, and where these will bring us in relating technology and behavior.

2.

SCIENCE STUDIES: SOCIAL CONSTRUCTION OF FACTS

‘Science studies’ started at the end of the nineteen seventies as a criticism of philosophy of science and the traditional sociology of science, which took the generation of the content of science ʊ the facts about nature ʊ for granted. Science studies, however, on the basis of empirical field studies

Technology and behavior

63

instead of philosophical reasoning, claimed these facts to be constructed in the social setting of the laboratory and its professional environment. Empirical observation by the ‘constructionists’ revealed that facts start their lives as local claims made by individual scientists or research teams. But claims that exist only locally are not facts, they have to enroll a larger constituency to become true. Scientists achieve this, it turned out, by connecting their claim to already established facts, for instance by using laboratory equipment that embodies earlier knowledge. If done convincingly, this constructive activity stretches the claim’s credibility such that scientists in other laboratories accept the claim as being true. The meaning of the claim has then converged, a new fact has been established, and the history of its construction is forgotten. This constructive view has two important consequences: (i) facts only exist as far as the socio-technical network that supports their meaning is spread out, and (ii) ‘reality’ cannot be used to explain why a statement becomes a fact, since it is only after it has become a fact that the effect of reality is obtained (Latour and Woolgar 1979).

3.

TECHNOLOGY STUDIES: SOCIAL CONSTRUCTION OF TECHNOLOGY

In the ‘eighties the constructive approach was extended to include technology. Constructivists started to agitate against ‘technological determinism’13 and against the resulting linear conception of innovation processes, which interprets the success of an artifact as the explanation of its development. Reasoning in a similar way as they did with scientific facts, constructivists turned the argument upside down: success (or failure) of technology is not the immutable result of its inherent characteristics, but is constructed in a social process that can and should be analyzed empirically. In a research field that became established as Science and Technology Studies (STS), this view was elaborated within two major paradigms, ‘social construction of technology’ and ‘actor network theory’. The social constructionist route was further developed along various tracks, such as the empirical SCOT14 program (Bijker and Pinch 1987).

13

14

In the deterministic view, technology is seen as an unavoidable success story, an autonomous force following a fixed and necessary sequence through history to which societies only can adapt (cf. Staudenmaier 1985, chapter 4). For a detailed discussion of technological determinism within STS, see Berg 1998. SCOT means Social Construction of Technology.

64

Chapter 7

SCOT advocates a multidirectional instead of a linear process of innovation, in which the development and stabilization of technological artifacts is conceived as the outcome of a gradual process going through cycles of variation and selection. To describe and understand such processes adequately, the student should analyze in symmetrical ways, i.e. study how losing artifacts contribute to the success of the winning ones. In the constructivist view, there is no one best way of designing an artifact. Emerging artifacts can empirically be shown as having no fixed meaning, but being interpreted flexibly by relevant groups in society depending on their political and socio-cultural context. Nineteenth-century youngsters liked bicycles with high front wheels because they permitted fast riding, while elderly people found them unsafe. Such differences in meaning can lead to different development tracks of the artifact (i.e. to new variations) through chains of problems and solutions, until a variant is recognized by all relevant social groups as the one that meets their preferences best and solves their problems with the earlier variations adequately (e.g. the safety bicycle: fast and safe). The latter artifact stabilizes, while the others disappear. Thus, success and failure of a technology are no longer ascribed to inherent technical characteristics, but as the outcome (‘closure’) of social processes of giving meaning by users. Just like facts, artifacts stabilize by meanings that converge in expanding social networks. Further constructivist tracks within STS include ‘social shaping of technology’ (MacKenzie and Wajcman 1985, Williams and Edge 1996), focusing on social forces which give rise to particular technologies, and the ‘co-evolution of technology and society’ (Rip and Kemp 1998). The latter track explains the ongoing development of technology and society as a process of mutual influences, inspired originally by ideas from evolutionary thinking in economics (Nelson and Winter 1977, Dosi 1982).

4.

FROM SOCIAL CONSTRUCTION OF TECHNOLOGY TO CONFIGURATION OF USERS

In agreement with mainstream sociology, social constructivism grants agency to humans: humans give meaning to the technical. In the beginning of the nineties, concern about the societal effects of ICT has been one of the triggers to scrutinize this unidirectionality. By investigating the metaphor of the machine as text, Woolgar put the reified ideas about the differences between animate and inanimate entities on the research agenda of STS (Woolgar 1991). The text metaphor challenges intuitive beliefs about what

Technology and behavior

65

machines are and can do. The context of Woolgar’s investigation is the design of computer systems. Such systems have users as a text has readers. To be able to use it, the user has to ‘read’ the machine. Therefore, in the design of computers, activities come on stream to define and delimit the user’s possible actions, i.e. to guide or constrain his or her reading by specific features of the machine. Consequently Woolgar prefers to say that, by setting parameters for the user’s action, the evolving machine attempts to configure the user. Thus, in contrast to (or by complementing) SCOT, Woolgar includes the design of artifacts in his analysis, showing that not only socio-cultural contexts, but also design features of the artifact proper constrain users in giving meaning. That is, both machine and user are involved in defining what the machine is and what can be done with it. Does the notion of ‘configuring’ re-introduce the notion of technological determinism? Indeed, technology seems deliberately to be made such that users cannot escape the actions it prescribes. However, Woolgar opposes this view by rightly emphasizing that of a text more than one reading is possible. Mackay et al., take this argument one step further by arguing that designers are not almighty but are configured too, both from within their own organizations and by users (Mackay et al., 2000). Woolgar’s reflections led him towards the underlying problem of agency and attribution. What are legitimate accounts of action and behavior? To which entities are we willing to ascribe intentionality? According to Woolgar, this is a question of convention, which seems to be that only animate entities have intentionality. But this conventional boundary is not absolute: people can ‘act mechanically’, space shuttles ‘behave perfectly’. In sum, compared to the SCOT position, Woolgar’s argument implies a tentative broadening of attributing agency towards human and nonhuman actors.

5.

LAST STEP: TOTAL SYMMETRY BETWEEN HUMANS AND THINGS

The final, most radical shift in equalizing humans and nonhumans with respect to action has been made in actor network theory. In a treatise on technical mediation, in which he rebukes Heidegger for arguing that modern technology is dehumanizing, Latour advocates a total symmetry between human and nonhuman entities (Latour 1994). According to Latour, philosophy cannot come very far in understanding technology as long as it sticks to the Cartesian subject/object dualism. Subjects (humans) and objects

66

Chapter 7

(nonhumans) do not exist as individual entities, they can only function in mixed collectives in which their properties are swapped, and their skills and responsibilities redistributed. Responsibility not to drive fast can be a property of humans (located in their consciences or in the authority of policemen), but it can be swapped with nonhumans: a traffic sign, a speed bump in the road. For this reason, it does not make sense ʊ it even prevents understanding ʊ to speak of subjects and objects, of actors and things: we should better speak of actants. This argument for symmetry goes even beyond speaking of ‘technical mediation’, which maintains a bias towards humans as actors: artifacts (only) mediate what humans want to do in their ‘life worlds’. Humans who take up a gun are not humans whose behavior is technically mediated while leaving their goals unchanged. They become a new actant, a body corporate, a hybrid: man-with-gun, whose action program is the resultant of the action programs of both man and gun. Latour calls this symmetrical goal shift ‘translation’ (see Verbeek, this volume). Applied to the design/use dichotomy of Woolgar, this symmetry means that designs (artifacts) and users configure each other. On the one hand, the material features of artifacts enable and constrain the actions of users. On the other hand, by embedding artifacts in practice, users give artifacts new possibilities for action. But we must recall here that users, like machines, do not act on their own but within hybrid collectives. Terrorists do not attack, Al Quaeda Inc. does. ‘Inc.’ here means incorporation indeed: of beliefs, credit cards, passports, the internet, mobile phones, skills to fly, etcetera. By linking all these elements within a new network, ‘objects’ acquire new power to act and their meaning changes: peaceful airplanes become flying bombs, kerosene becomes ammunition, skyscrapers become vulnerable piles of victims. This change is not a mere semantical ‘change of meaning’, or a switch in ‘multistability’ (see Verbeek, this volume). To swap the properties in this way requires hard work, building new networks, i.e. the work of mobilizing, enrolling and delegating actants in new ways15. Thus, according to actor network theory, things are not what they are but what they become by fixing them in a network with other actants. Consequently, there is a role here for social theory to extend philosophy, but the practice of sociology has no place for things either. Declaring a 15

For the sake of clarity, it may be useful to observe that Latour’s ontology is incompatible with Winner’s claim of technologies having inherent politics, which, in Latour’ s view, they can never possess (see Latour 1994). One could add that Winner sticks to the dualism between a ‘technical system [that] actually requires the creation and maintenance of a particular set of social conditions’ (see Winner 1986, p. 33, emphasis added).

Technology and behavior

67

(supposedly) technical problem to be ‘social’ and vice versa does not help very much, because it remains a cry from the trenches of the social/technical dualism. Ethnomethodology might be a better starting point to link humans and things within one conceptual frame, if it was not so vulnerable to charges of locality and being ahistorical: ethnomethodology obviously focuses on micro-level phenomena. Within the body of sociology, these charges continue to support the need for claims about a wider ‘social and power structure’16. Local and individual actions as they are observed and described in ethnomethodological accounts can only take meaning from and build on a wider structural context, the mainstream sociologist would say. One cannot create order on the spot. Latour’s new paradigm claims to obviate these disputes. Every time we move our mouse to operate our word processor (a local action), we mobilize a huge network of actants beyond the country we live in, and beyond Steve Jobs down to Isaac Newton. Because this network is largely blackboxed and swallowed up by the past it is invisible, but that does not mean that is not at work; to the contrary. It is especially due to the work of things, shifting, translating, delegating, overlooked by sociologists who observe only social structure carved into human beings, that actions in a distant past and place can be present here and now to enable that local action to be productive. Without realizing it, we are configured everywhere, but not only socially.

6.

WHAT HAVE WE GAINED?

During this short tour, we have seen how traditional dichotomies (subject/object, social/technical, structure/agency, design/use, technical functionality/human intentionality, man/machine, man/world) have been deconstructed for the sake of understanding technology in terms of symmetrical relations between humans and things. To get there, we have been pretty theoretical, though. Therefore, the value added for the analysis of practical problems relating to ‘technology and behavior’ should now be considered. For instance, if we are configurors, can we deliberately create an order that is less polluting and more sustainable? Or are we entirely in the power of haphazard processes of translating and being translated? In Latour’s view, a precondition for a better world is to make nonhumans 16

For instance Goffman’s problematic promotes the study not of observable interactions of everyday life as such, but of the invisible structures that govern them (see Ritzer 1992, p.522). Bourdieu argues that those who are satisfied with video recordings take aboard a pre-structured concreteness that may overlook the principles needed for interpretation. Direct observation of interaction does not necessarily reveal the underlying power relations (Bourdieu et Wacquant 1992, p. 42)

68

Chapter 7

visible by including them in our body politic. Plants, animals, and hybrids such as carbon dioxide and ozone need political representation (Latour 1991). But what about actants yet to be created, such as man-withengineered-genes, processed embryos, climate-neutral energy carriers, photovoltaic heat/power deliverers, fuel cell cars? Will they make the world a better place? Can technology be configured in a normative way, and translate behavior of the resulting actants in less damaging directions than nowadays? Or is the only option to go on indifferently while throwing new hybrids into the melting pot of shareholder value, market forces and the whims of consumers, and just wait and see how the goals shift? But if artifacts have agency to translate, they change the moral order (cf. Woolgar 1991), and can be, or should be scrutinized for that. There is an asymmetry, however, that hampers such normative scrutiny. It can be explained by examining the ambivalent definition of ‘script’ (see Verbeek, this volume, for introducing this notion). A script can be ʊ and usually is ʊ defined from the inscription side. In that case, it carries a materialized message (‘text’) from the designer to the user. Through the scripted artifact, the designer can act a distance. He does not have to be present and hold your hand in operating a cup or a spoon. The deliberate shape, colors and signs of artifacts tell you what to do with it. To quote Woolgar’s text analogy again: ‘a text [...] will be organized in such a way that ‘its purpose’ is available as a reading to the user’. At first sight, here is an entrance for designing a better world. However, as was said before, the user may read otherwise, configured as she is by ties with other actants. The text will not always succeed in being ‘isomorphic with the concept we use to make sense of it’ (cf. Woolgar). Thus, there is some (maybe more) logic in defining script from the decoding side. In that case, a script is not the intended but the proven force that an artifact exerts on a user by translating his goals and actions. How strong this force is, and in which direction it works, can only established by empirical verification afterwards17. Because of unforeseen incongruities between inscriptions in and readings of scripts that can always occur, the influence of new artifacts on the behavior of users seems inherently unpredictable. Such discrepancy may lead to unintended effects. There seems to be some truth in the ethnomethodological credo that every story about technology is different from the others. So after all, have we made no progress? Fortunately, this is not the whole story. If designers and users are configured in networks and if this configuration constrains their inscriptions and readings systematically, at 17

As I carried out in the case of the cistern interface intended to save water but failing to do so, see Jelsma, chapter 22, this volume.

Technology and behavior

69

least some learning aboutand anticipation of incongruity must be possible. To support such anticipation, I have developed the notions of design logic and use logic. Design logic is the story that tells you why an artifact is as it is from the designers’ point of view. It can be reconstructed by interviewing designers or by recording their deliberations. User representations will usually be part of it (Jelsma 2001). At the use side, there will also be logic at work, be it often blackboxed in routines. Who knows how the fridge or the dishwasher configures users in everyday practice? When you observe users within the context, i.e. while collaborating with these machines in their households, and ask them questions, they often are able to open the black box and tell you18. By comparing design logic behind scripts with such user stories tied to the same artifact, it is possible to trace asymmetries in logic. These are indications that unintended effects may occur if the design is implemented as originally intended. To avoid or mitigate such effects, the design may subsequently be adjusted to lower the asymmetry. Using this approach, one can develop tools for product development supporting the anticipation of future translations of action programs as a result of implementation of that product. I did so to improve the scripts of existing household appliances in order to (better) support ecofriendly use practices (see Jelsma, chapter 22, this volume). The approach developed is an example of a modest but practical application that may help the new paradigm to escape charges of being only descriptive and detached, instead of normative and intervening in constructing a better world.

REFERENCES Beyer, H. and K. Holtzblatt (1998), Contextual Design, Morgan Kauffmann Publishers, San Francisco. Berg, M. (1998), The politics of technology: On bringing social theory into technical design, Science, Technology and Human Values 23/4, pp. 456-490. Bourdieu, P. and L.J.D. Wacquant (1992), Réponses: pour une anthropologie réflexive, Editions du Seuil, Paris (Dutch translation: Sua, Amsterdam). Dosi, G. (1982), Technological paradigms and technological trajectories: a suggested interpretation of the determinants and directions of technical change, Research Policy Vol. 11, pp. 147-162. Giddens, A. (1984), The constitution of society: Outline of the theory of structuration, University of California Press, Berkeley. Ihde, D. (1990), Technology and the live-world, From garden to earth, Indiana University Press, Bloomington. Jelsma, J. (2001), Smart Work package 4.2: Comparing design logics, Report C-02-008, December 2001, Energy Research Center of The Netherlands, Petten. 18

This approach is called the contextual interview, a method used in contextual design (see Beyer and Holtzblatt 1998).

70

Chapter 7

Latour, B. (1992), Where are the missing masses? The sociology of a few mundane artefacts, in: Shaping Technology, Building Society (W.E. Bijker and J. Law, eds.), MIT Press, Cambridge, Mass. Latour, B. (1993), We have never been modern, Harvard University Press, Cambridge Mass. Latour, B. (1994), On technical mediation – Philosophy, sociology, genealogy, Common Knowledge, pp. 29-64. Mackay, H., C. Carne, P. Beynon-Davies and D. Tudhope (2000), Reconfiguring the user: Using rapid application development, Social Studies of Science 30/5, pp. 737-57. MacKenzie, D. and J. Wajcman (1985), The social shaping of technology, Open University Press, Milton Keynes. Nelson, R.R. and S.G. Winter, S. (1977), In search of a useful theory of innovations, Research Policy Vol. 6, pp. 36-77. Pinch, T. and W.E. Bijker, The social construction of facts and artifacts, in: The social construction of technical systems (W.E. Bijker et al., eds.), The MIT Press, Cambridge Mass., pp. 17-51. Rip, A. and R. Kemp (1998), Technological change, in: Human choice and climate change (S. Rayner and E.L. Malone, eds.), Batelle Press, Columbus, Ohio. Ritzer, G. (1992), Sociological Theory, McGraw-Hill, New York. Staudenmaier, J.M. (1985), Technology’s storytellers, MIT Press, Cambridge, Mass. Van den Daele, W. (1978), The ambivalent legitimacy of the pursuit of knowledge, in: Science, Society and Education, Conference Proceedings, Free University, Amsterdam. Verbeek, P.P. (2000), De daadkracht der dingen, Boom, Amsterdam. Williams, R. and D. Edge (1996), The social shaping of technology, Research Policy Vol. 25, pp. 865-899. Winner, L. (1986), Do artifact have politics?, in: The whale and the reactor, The University of Chicago Press, Chicago, pp. 19-39. Woolgar, S. (1991), Configuring the user: the case of usability trials, in: A sociology of monsters (J. Law, ed.), Routledge, London, pp. 58-99.

Chapter 8 THE SOCIAL AGENCY OF TECHNOLOGICAL ARTIFACTS: A Typology Philip Brey

1.

INTRODUCTION

Authors have, in various ways, attempted to account for the fact that technological artifacts are not neutral to society but are capable of significantly influencing or transforming the social context in which they are used. Technological artifacts may influence behaviors, attitudes, cultural beliefs, and modes of social organization in ways that are often not directly related to their intended functions. That they may do so has been observed in fields ranging from philosophy to technology studies to psychology to ergonomics. In this section, an attempt will be made to arrive at a typology of ways in which artifacts have been claimed in these various fields to affect their context of use. I am only aware of one previous typology of this kind, which has been presented by Richard Sclove (1995).19 My typology will incorporate some of his distinctions, but will also move beyond it. To conceptualize the (often unintended) influence that artifacts may have on their context of use, I will employ the notions of affordance and constraint.20 Artifacts, I claim, may affect their context of use in two ways. 19

20

Sclove’s typology (1995: 11-16), in which he discerns various ways in which technologies help to structure social relations, culture, and human psychology, consists of six types of constraints. The term ‘affordance’ has been introduced by psychologist James J. Gibson to denote a condition in the perceived environment that elicits action or experience, as opposed to a constraint, which hampers

71 P.-P. Verbeek and A. Slob (eds.), User Behavior and Technology Development: Shaping Sustainable Relations Between Consumers and Technologies, 71-80. © 2006 Springer.

72

Chapter 8

They may, first of all, afford, enable, allow, induce, stimulate, cause, necessitate or require certain events or states-of-affairs. For example, in vitro fertilization techniques allowed a distinction to be made between genetic motherhood and biological motherhood, and consequently required a renegotiation of the social status of motherhood and new social definitions of the social roles that genetic and biological mothers were expected to play. I will say that IVF techniques afford these events and states-of-affairs. Secondly, artifacts may discourage, prevent, constrain, prohibit or disallow, or outlaw events or states-of-affairs. For example, Winner (1980) has claimed that certain overpasses in New York were built to prevent bus access to Long Island, thus discouraging poor blacks from frequenting its public parks. In my terminology, the overpasses constrain certain events or states-of-affairs from taking place. It should be added immediately that affordances and constraints are not objective features of artifacts. Rather, the affordances and constraints that an artifact possesses depend on the setting or context of use in which it is used. For instance, the constraint imposed on the mobility of poor blacks by the overpasses in Winner’s example depends not only on the material context of the bridge (e.g. buses that are too high to fit under them), but also on the social context (e.g. the fact that many blacks are poor and, hence, rely on buses for transportation), and on common beliefs and practices (e.g. the fact that it is not customary to hitchhike from New York to Long Island). Still, it is useful to attribute the constraint of limiting access to Long Island for poor blacks as a constraint to the aforementioned overpasses, and not to the overpasses plus the entire context in which they operate. For these overpasses are still the immediate cause that poor blacks are discouraged from visiting Long Island: it is because of them that buses do not go there. The affordances and constraints I am about to distinguish are distinguished by the type of event or state-of-affairs that they relate to: is it a behavior, a social role, an organizational form, a cultural belief, or yet something else that is afforded or constrained by an artifact? I will argue that it is useful to distinguish five basic types of affordances and constraints in artifacts. They are (1) behavioral (relating to the behaviors of users); (2) user-profile (relating to fixed attributes of users); (3) material and infrastructural (relating to the physical context required for the functioning of the artifact); social (relating to social structures afforded or constrained by the artifact) and (5) cultural (relating to cultural patterns and practices). It action or experience. In relation to technology, the term has been used with a similar meaning by authors like Norman (1993) and Latour (1992). I follow Latour in broadening the meaning to refer not just to actions and experiences, but to events generally.

The social agency of technological artifacts

73

may be noticed that some of these types relate to micro-level phenomena (1 and 2), whereas others relate to phenomena that can be either micro or macro (3, 4 and 5).

2.

BEHAVIORAL AFFORDANCES AND CONSTRAINTS

These are affordances and constraints that require or induce certain behaviors in technology users. They usually reveal themselves quickly in the interaction between these users and the artifacts in question. A behavioral affordance is a feature of an artifact that invites, stimulates or requires users to perform certain actions while using, or in order to use, an artifact. A behavioral constraint is a feature that discourages or prevents a user from behaving in certain ways while using an artifact. Behavioral affordances constraints are discussed most extensively in the literature on ergonomics, for example in the work of psychologist Donald Norman (1988; 1993), who has extensively studied how the design of all kinds of everyday artifacts results in certain behavioral affordances and constraints. Let us consider some examples. A hotel key has a heavy ball attached to it to induce the user to return the key to the front desk upon leaving his or her hotel room. The behavioral affordance of the key-with-ball is, hence, that it invites returning the key to the desk. At the same time, a behavioral constraint is issued that discourages hotel guests from taking the key outside. Automobiles try to induce drivers to wear a seat belt by flashing a warning light. In the past, some cars even refused to start if the driver did not wear a seat belt. Here, the flashing light weakly constrains driving without a seat belt, whereas the car that does not start strongly constrains this behavior. A gas pedal whose resistance increases at high speeds discourages drivers from driving too fast. Automated teller machines nowadays require users to remove their ATM card before money is dispensed. This behavioral sequence is engineered in order to prevent users from forgetting their card. These and many other examples show how technological artifacts work to stage the behavior of their users. Notice that the affordances and constraints involved may either rely on physical force (e.g. the car that will not start unless a seat belt is worn), or on signs (e.g. the flashing warning light suggesting that a seat belt is worn). Physical force and the force of signs are the two primary forces by which artifacts steer the behavior of their users.

Chapter 8

74

3.

USER-PROFILE AFFORDANCES AND CONSTRAINTS

Not every artifact can be successfully used by any user. Many artifacts require or presuppose a certain user profile. For those who do not satisfy this profile, using the artifact may be less rewarding, more difficult, or even outright impossible. For example, an ATM that makes extensive use of verbal instructions largely excludes non-literate users; an ATM that requires fast responses from its users excludes slower users, such as many elderly; and an ATM that is built high up the wall may be inaccessible to wheelchair users. An automobile built to accommodate medium-sized people is less user-friendly to very small or very tall people. An expert system that helps investors build portfolios but is geared toward wealthy investors may give poorer advice when smaller amounts of money are entered, and is, hence, less compatible with the values and interests of less wealthy investors. A watch that operates on solar energy excludes users who spend most of their time indoors. Software that requires advanced computer skills for their use and that is geared towards typically ‘male’ interests excludes novice users and discourages women from using it. User-profile affordances and constraints specify aspects of the profile that users should (ideally) have in order to use an artifact successfully. This may include required or preferred physical and psychological features, competencies, skills, knowledge, values, interests, and goals. In this way, certain classes of users are ‘selected’ while others are ‘excluded’ or ‘marginalized’. Users who do not fit the user profile of an artifact may in various ways try to make a better fit, for instance by acquiring required knowledge or skills, by modifying their interests and goals, by adapting their behavior, by using various aides (e.g. glasses, hearing aides), or by letting others use the artifact for them.

4.

MATERIAL AND INFRASTRUCTURAL AFFORDANCES AND CONSTRAINTS

These are affordances and constraints relating to aspects of the physical condition of the environment in which a technology is used. Often, they are physical conditions (material and technological arrangements) presupposed by technologies. They are part of what Sclove (1995) calls background conditions: conditions that are to be satisfied if a technology is to function properly. A television set, for example, can only function properly if a whole series of background conditions is satisfied: there has to be access to

The social agency of technological artifacts

75

electricity, there must be a production and distribution of programs, the set must be protected against inclement weather, prospective users must be knowledgeable of how the set is operated, etc. Some such background conditions concern aspects of social infrastructure, such as human organization, skills and knowledge. However, I would like to make a rough distinction between physical-technological background conditions and social-organizational background conditions, which are discussed below. Next to presupposing infrastructure, artifacts may also afford new infrastructures that are not necessary for their functioning. For example, a personal computer affords the attachment of all kinds of devices, and in this way invites all kinds of new material configurations that result in enhanced functionality, ranging from web TV to remote-controlled refrigerators.

5.

SOCIAL AFFORDANCES AND CONSTRAINTS

These are affordances and constraints that relate to social entities, that is, entities that are traditionally assigned to the core domain of sociology, such as social relationships, social roles, social status, organizational forms and social institutions. It is useful to discuss separately the impact of technology on social statuses (and concomitant roles and relations), and the role of technologies in shaping larger social structures such as organizations and institutions.

5.1

Social statuses, roles and relations

Someone’s (social) status is a social location that he or she occupies for the purpose of interacting with others. Examples of social statuses are professor, student, mother, barber, and Republican. A (social) role is the dynamic aspect of status: the things one is expected to do because of the status one occupies. E.g. students are expected to study, professors to teach, and police officers to enforce the law. Social relationships constitute the structure of links between social statuses, as specified by the roles associated with each. E.g. a social relationship exists between the statuses ‘mother’ and ‘son’ in that there is a societal agreement about the behavior of each toward the other. Many affordances and constraints of artifacts are social in that at least part of their constraining effect lies in their effect on social entities of this kind. Artifacts shape aspects of social roles and relations by requiring, fostering, enhancing, discouraging, eliminating or modifying certain social behaviors and patterns of social interaction, or by changing or fostering certain perceptions of social status or the criteria by which people agree to the assignment of certain statuses or roles.

76

Chapter 8

Let us consider some examples of the often subtle ways in which artifacts promote or modify certain statuses, roles and relationships. In 19th century typesetting (Wajcman, 1991), the Linotype typesetter (competitor of the Hattersley typesetter) had a design in which the skilled job of setting and the unskilled job of distribution (putting the letters back in their letter box) were integrated and could not be separated from each other (they could be in the Hattersley typesetter). Hence, the Linotype typesetter defined a profession of (skilled) compositor/distributor, rather than two separate professions of (skilled) compositor and (unskilled) distributor. It hence imposed a constraint on the social statuses and roles that were to be found in the typesetting business. The installation of running water and private washing machines next to the public fountains and washbasins in the village of Ibieca, Spain, also brought along social constraints (Sclove, 1995). Prior to this installation, the public fountains and washbasins collected people around them and, hence, became a focal point for social interaction in the village. These social interactions, next to the joint use of facilities, resulted in informal social relationships between villagers, resulting in social roles and statuses (e.g. ‘acquaintances,’ ‘friends,’ ‘joint users’). These social relationships, roles and statuses were eroded with the installation of more convenient running water and washing machines, as their use did not involve similar social interactions. As a final example, consider the difference between ‘ladyshaves’ for women and electric razors for men. ‘Ladyshaves’ cannot be opened for repair, in contrast with electric razors for men, hence promoting a social status for women as dependent (on repair persons) and technologically incompetent. What these examples show is that artifacts are frequently not just mere passive mediators of social relationships. Rather, artifacts often help forge social relationships and help define social roles. As Latour (1992) has argued, artifacts make up the ‘missing masses’ in society that sociologists have been looking for. The sociologist who only studies interactions of human beings and does not study interactions of human beings with artifacts (or even of artifacts with each other) misses a large part of the fabric that keeps society together, and therefore cannot adequately account for social behavior. Not all artifacts shape social roles, statuses and relations in the same way, however. Sometimes artifacts only induce the renegotiation of existing social roles and statuses, without in any way constraining the result of such renegotiations. For example, in vitro fertilization techniques require a renegotiation of the social status of motherhood in relation to genetic and

The social agency of technological artifacts

77

biological mothers. Yet they do not prescribe in any way what the social roles of genetic and biological mothers should be. This appears to depend on social choices alone. At other times, artifacts do not just induce the (re)negotiation of social roles and statuses, but also strongly suggest aspects of the overall shape of these roles and statuses. For example, the cockpit of larger airplanes usually has thee seats that correspond to pilot, co-pilot, and navigator. The design of the cockpit only supports the co-operation of the plane by three persons, and assigns specific tasks to each one. Yet, even in this example, additional social choices are required to define the full roles of pilot, co-pilot, and navigator, for example regarding the decision-making hierarchies between them, and the specific nature of the tasks for which each is responsible. So it appears that artifacts can help shape social roles, statuses, and relations, but cannot define them in full.

5.2

Organizational and institutional structure

A social structure is a persistent network of statuses and roles, the relationships among statuses, and the interactions expected within those relationships that make up the organized framework for social life. E.g. the statuses (professors, students, security guards, etc.) making up a university, plus all the relationships and interactions among those statuses, constitute the social structure of the university. A society is a social system with people living in it, usually associated with a definable geographical territory. A social system is a complex network of social agreements linking social statuses to one another. Social systems may contain various subsystems, which are organizations or institutions of various sorts. Technological artifacts and systems may affect social structures by requiring or supporting certain modes of social organization or by affecting the distribution of social goods within social structures. For example, nuclear power plants have been claimed by Winner (1980) to require, or be more compatible with, hierarchical relationships for their operation and protection, whereas solar energy does not require hierarchical control. Nuclear power plants thereby define a particular form of organization, or social structure, in which hierarchical social relations and hierarchically defined social statuses and roles exist. Solar energy systems, in contrast, are very well compatible with decentralized, egalitarian modes of control, and hence with more egalitarian social roles and relationships. In a similar way, computer networks in organizations have been claimed to support ‘flatter’ organization with fewer middle managers. Another example by Winner is the mechanical tomato harvester, built in California in the seventies, which allowed tomatoes to be harvested faster and cheaper in larger farms that use a highly

78

Chapter 8

concentrated form of tomato growing. As a result of these savings by large farms, many small farms could not compete and were eliminated. Hence this machine stimulates the existence of large farms while working against smaller farms.

6.

CULTURAL AFFORDANCES AND CONSTRAINTS

These are affordances and constraints that relate to shared cultural meanings and shared cultural practices. Shared cultural meanings are meaningful structures by which members of a society interpret reality and guide their own behavior. They include symbols, concepts, beliefs, values and inferential patterns. Such structures may be altered in various ways through the use of technology. Most importantly, artifacts may transform experience and action, and in this way alter our overall relation to the world. As Ihde (1979; 1990) has argued, technical artifacts frequently mediate and hence modify the way humans relate to their world because they frequently function as means through which we perceive or act on the world. This applies to devices like telescopes, televisions, hammers and automobiles, which are not just objects encountered in the world but objects through which we perceive the world or act on it. Such perception and action brings along new experiential categories, new perceptions of the world, and new perceptions of ourselves. For example, the world has become a ‘global village’ because of telecommunications technology, and the landscape becomes a fleeting form when driving a fast car. As technologies transform our experience of the world, they transform the very basis on which we build higher or more complex meaningful structures, such as concepts, beliefs, symbols, and thought patterns. Mass media, it has been argued, are changing our concepts of evidence, knowledge, and truth. Transportation technologies may change our conception of the world we inhabit. Virtual reality is changing our concept of reality. Artifacts also affect cultural practices: typical, culturally shared ways of behaving and acting. These include organized cultural practices such as annual festivities, but also more mundane practices such as eating habits and greeting rituals. Technology has evidently greatly affected all kinds of cultural practices, from child delivery to burial. Technologies create new cultural practices that have the technology as their center point (e.g. watching television, playing video games, auto racing). Artifacts may also change an existing cultural practice by initially playing a merely supportive role in it, but over time radically modifying the practice. For example, the

The social agency of technological artifacts

79

microwave oven has not just made it easier to prepare meals, it has also been instrumental in changing eating culture by making it possible to reheat meals quickly, and efficiently prepare different meals for different members of the household. In this way, it has helped to create a more individualized and less communal eating culture. The installation of running water and washing machines in Ibieca, Spain has not just changed washing and laundering practices, it has also eliminated various cultural practices that existed around the public washing and laundering facilities, such as gossiping and watching others do their laundry.

7.

CONCLUSION

Technological artifacts afford and constrain various types of (social) phenomena in ways that are often not directly related to their intended function. I have distinguished five principal types of phenomena that are afforded or constrained by artifacts. Only the first of these five explicitly denotes individual behavior. However, individual behavior may be involved in the other four as well. User-profile affordances and constraints affect the behavior of prospective technology users by requiring certain classes of users to adapt themselves to fit the user profile. Material and infrastructural affordances and constraints may induce prospective users to alter the physical environment so that a technology can function, and may invite them to add further elements so as to create additional functions. Social affordances and constraints may affect individual behavior by consolidating certain social roles for users or other stakeholders, or by transforming the social structures within which individual behavior is defined. Cultural affordances and constraints, finally, may affect individual behaviors by modifying the cultural meanings on which people may base individual behavior, as well as by changing cultural practices. Hence, it can be concluded that technological products and systems influence individual behavior in multiple ways, and since some of these influences are mediated by macro-level structures, micro-level analyses of the impact of technology on behavior are insufficient.

REFERENCES Ihde, D. (1990). Technology and the Lifeworld: From Garden to Earth. Bloomington: Indiana University Press. Latour, B. (1992). ‘Where are the Missing Masses? The Sociology of a Few Mundane Artifacts.’ In W. Bijker & J. Law (Eds.), Shaping Technology/Building Society: Studies in Sociotechnical Change. Cambridge, MA: MIT Press.

80

Chapter 8

Norman, D. (1988/1990). The Design of Everyday Things. New York: Currency/Doubleday. Norman, D. (1993). Things that Make us Smart: Defending Human Attributes in the Age of the Machine. Reading, MA: Addison-Wesley. Sclove, R. (1995). Democracy and Technology. Guilford Press, New York. Winner, L. (1980). Do Artifacts have politics? Daedalus 109: 121-136.

Chapter 9 TECHNOLOGY AND USERS: A Conceptual Map

Wim J.M. Heijs and Peter-Paul Verbeek

1.

INTRODUCTION

The goal of the first part of this book was to identify concepts, conceptual frameworks and methods in various disciplines and areas of application that are used to describe and measure features of the interaction between technology and users. From the previous chapters it is clear that many scientific domains contain relevant knowledge for studying this relationship. However, they also show that the information is scattered and often unknown outside of a particular field. By providing an overview, this part of the book aims to take a first step towards a deepening of the understanding of user-technology interaction. The unveiling of different frameworks to conceptualize apparently similar phenomena may stimulate reflection and hopefully even invite co-operation. In the end, this may produce design guidelines that serve an easier, safer and sustainable use of technology and the well-being of its users. The following synopsis attempts to outline the various concepts and frameworks, their differences and their mutual associations by using the metaphor of a map. This ‘conceptual map’ also serves to identify areas that are covered and fallow land. Before we draw the map, some introductory remarks are required. The first concerns the selection of concepts and frameworks in this part of the book. They emanate from scientific disciplines (environmental and social psychology, philosophy of technology, science and technology studies) and some specific areas of application (action theory, safety studies, and 81 P.-P. Verbeek and A. Slob (eds.), User Behavior and Technology Development: Shaping Sustainable Relations Between Consumers and Technologies, 81-92. © 2006 Springer.

82

Chapter 9

household and consumer science). This selection does not imply that other disciplines or fields do not have appropriate information: it is the result of a search that was extensive and, simultaneously, limited because of practical reasons (e.g. authors willing to participate) and the coherence of the book as a whole (some insights have a special bearing on a particular subject and are presented in other parts). We are confident, however, that most of the primary concepts are covered here. Secondly, we would like to stress that this chapter does not offer an integrated conceptual model, nor an elaborated view of the interaction process between users and technology itself. The matter at hand is far too complex for that purpose. The frameworks in the various chapters of this part of the book all have their own bearing, specific mechanisms and evolutionary history. We aim at relating these frameworks, not molding them together. As a consequence of leaving out the interaction process, global input variables (technology and user characteristics) and results (e.g. efficiency or pay-off on the side of technology and satisfaction or well-being on the user side) are not discussed either, although we recognize that these results are crucial for the reiteration of the process. We do hope, however, that the overview will contribute to future efforts to develop such a model.

And thirdly, the map to be drawn does not represent an overview of the determinants of interaction. We focus on concepts that are useful to describe and analyze relations between users and technology. These concepts may be determinants, but they may also be research methods or analytical terms that encompass more or less complex issues. In the next section we will describe the metaphor of the conceptual map. Sections 3 and 4 explain the arrangement of the concepts on the map, their theoretical backgrounds, and their main differences and associations. Conclusions, including the identification of fallow land, are discussed in section 5.

2.

A CONCEPTUAL MAP

Figure 9-1 depicts the proposed conceptual map. It might look rather complicated at first sight, but we expect that the following clarification will make it relatively easy to interpret. The concepts are aligned on two dimensions. The first, horizontal dimension is obvious, since it consists of the ingredients of the present inquiry: technology on the left side (with technologically oriented concepts such as constraints or system control), the

Technology and users

83

user on the right side (with user-oriented concepts like attitudes or group norms) and their interaction in between. The right side is named ‘user’ instead of ‘behavior’, because this offers a more symmetrical perspective. Technology and users can both be regarded as actors whereas behavior, depending on the point of view, is one of the aspects of the interaction or a result of it.

Figure 9-1. A conceptual map

84

Chapter 9

Besides technology, user and interaction, there are two intermediate scale values (T Æ I and I Å U). These serve to locate concepts that do not indicate central aspects of the interaction itself but still refer to it, either from the technology side or the user side. Taking a laundry machine as an example, the color of the front panel is a characteristic of the machine (T) and the income of the household a user characteristic (U). Next to characteristics such as these, the interaction (i.e. the actual usage: I) is dependent, for instance, upon the perceivability of a button to save energy (T Æ I) and existing mental models of similar machines (I Å U). The vertical dimension was chosen out of a number of possibilities: the disciplinary origins of the concepts, the inductive or deductive nature of their derivation, or the echelons that are subject to research. It appeared that the latter division ʊ in micro, meso and macro ʊ provided the best opportunity to unravel different outlooks. Concepts have not only been arranged by their location on the dimensions but also in a way that makes it possible to discern differences and associations between them. To describe these relations we use the terminology introduced by Kevin Lynch (1960), that is utilized for the analysis of cognitive maps that people draw of their environment. Typical elements in these maps are districts (regions with a common character), landmarks (noticeable objects that may be used as points of reference), edges (limiting or enclosing features), nodes (points where behavior is focused, such as cross-roads) and paths (routes for transit). In the present context these cartographic elements should be interpreted as follows. The concepts themselves serve as landmarks on the map (the suffix behind them indicates the chapter(s) in which they are named). Districts are areas with a common conceptual background; these are the frames in figure 9-1. Districts are not necessarily disciplinary areas: they may be dominated by a discipline but they can also incorporate ideas from different disciplines or from areas of application with a theoretical similarity. Edges, as indicated by the thicker lines, are relatively distinct boundaries between districts with frameworks that are more specific and not easily connectable to others. Besides these elements, which indicate differences between (sets of) concepts, nodes and paths are used to portray associations. Nodes consist of concepts that are shared between districts (i.e. used in different frameworks). And paths are connections between distinct (sets of) concepts in different districts that seem to have a useful correspondence. The most important nodes and paths are indicated by bold text and arrows, respectively.

Technology and users

3.

85

DISTRICTS, LANDMARKS AND EDGES

From the previous chapters in this first part of the book seven clusters of concepts were identified based on theoretical affinity. They constitute the following districts: 1. psychological interaction concepts (micro level) 2. mental mechanisms (micro level) 3. social psychological concepts (micro through macro levels) 4. philosophical interaction concepts (micro level) 5. multilevel affordances and constraints (micro through macro levels) 6. user-technology interaction and safety concepts (micro and meso levels) 7. household concepts (meso level). Starting at the top of the map, there are three districts that have a psychological signature (1, 2 and 3). The first contains concepts that are used to analyze the mutual influences of technology and users in the interaction process from a psychological perspective. The predominant discipline here is environmental psychology (with mapping, PE-fit and the lens model as landmarks), supplemented by concepts from action theory and cognitive psychology (action facilitation, behavioral affordances). It extends towards the T Æ I side, with concepts indicating modes in which technological properties convey information that co-shapes interaction. The second district is located nearer to the user side. It consists of concepts used to analyze and describe cognitive mental mechanisms that guide the behavior of users in the interaction process. The main background of this district is cognitive psychology, with mental models, schemata and behavioral scripts (I Å U) and basic information processing (U) as landmarks. It also holds concepts from action theory that is rooted in cognitive psychology (required levels of behavior, action programs, streamlining). The third district, also on the user side and covering micro, meso, and macro levels, comprises concepts from social psychology that occur in several chapters. Whereas the former district focuses on intrapersonal information processing in a narrower sense, this area represents concepts that extend this process by including social, interpersonal influences and by providing relations with intentional and habitual behavior. Thus, the primary landmarks are components of attitude-behavior and habit models, adverse types of conduct (reactance and rebound behavior) and behavioral determinants for larger groups. Although the foci of the three psychological areas (and therefore their position on the map) are different, there are no really sharp boundaries between them. Concepts that are developed and

86

Chapter 9

researched in cognitive and social psychology are applied within the first district. There is a sharper edge between these districts and the next two (4 and 5), which are more philosophical in nature. The district of philosophical interaction concepts (4) occupies a central position at the micro level and combines insights from science and technology studies and philosophy of technology, mainly drawn from actor-network theory and phenomenology. The concepts under T Æ I indicate the ability of technological artifacts to co-shape human actions and experiences (scripts, invitation and inhibition, technological intentionality, the design logic of artifacts, amplification and reduction). The concept of mediation refers to the interaction itself, with its specific forms of translation of action programs and transformation of perception; and so does the concept of configuring (I). The concepts of ‘use logic’, ‘multistability’ and ‘delegation’ (I Å U) indicate that user interpretations of artifacts and delegations of responsibilities from users to artifacts help to determine the ways technologies are used. Closely related is the fifth district, multilevel affordances and constraints, also based on philosophy of technology. Landmarks in this area describe the supporting or limiting effects of technology on users (T Æ I) on a micro scale (behavioral, user profile), a meso scale (social relations and organizational structure) and a macro scale (cultural meanings and practices). Behavioral affordances (the first landmark) are in district 1 because they have a similar meaning in philosophy of technology and environmental psychology; in the other landmarks (this district) affordances have a different content (see below). The edge between these philosophical districts and the psychological ones stems from their separate paths of development. Until recently, there has hardly been any exchange of knowledge or findings. As can be seen in the figure, a consequence is that the same terms (scripts and action programs in districts 2 and 4, affordances in district 1 versus those in district 5) are used for different phenomena. In cognitive psychology, scripts and action programs are types of mental representations of episodic series of actions related to certain situations, such as dealing with an artifact. They reside in memory and serve to facilitate behavior, e.g. by invoking a next action more or less automatically. In actor-network theory, scripts indicate the (inscribed) ability of artifacts to prescribe specific actions to their users; and both humans and artifacts possess action programs. In environmental and cognitive psychology, an affordance is a combination of physical properties of an object that invites a certain usage by being perceived during interaction

Technology and users

87

on a pre-cognitive level: a button is push-able or turn-able. This meaning is extended in philosophy of technology to include a wider range of effects that may be produced by an artifact, so that it is located farther left and also downwards on the meso and macro levels. To indicate these differences on the map, the adjectives ‘behavioral’ or ‘human’ and ‘artifactual’ have been added to the names of the concepts involved. In spite of the contrasts, there are theoretical reciprocities between the various concepts as well, e.g. between artifactual scripts and behavioral affordances. These will be discussed more extensively below, when nodes and paths are described. The remaining two districts (6 and 7) lodge the conceptual frameworks constructed in two areas of application: safety studies (on micro and meso levels) and household and consumer science (meso level). Concepts in safety studies concern human-technology interaction aspects, as seen from the perspective of the risk of accidents. They cover the entire range of the horizontal dimension with accident scenario (I) as the chief landmark, linking aspects of technological systems (T) to behavior, management and culture (I; partly through the concept of required levels of behavior, which is in district 2 because it is the same concept as in cognitive psychology). In households, compatibility is the landmark where technology and users meet (I). The influences of technology on the household system (action bound, logistic or functional) are met by a complex balancing system involving the level and standard of living, and influencing household activities through the perceived level of well-being. Because this framework is rather specific (a systems approach with various feedback loops and concepts that are special to the field), there is an edge between this area and the others, except from social psychology because norms, goals and habits on a group or societal level are regarded as variables affecting the household system.

4.

NODES AND PATHS

Another important goal of localizing the concepts on a map is to identify connections and similarities between them. These can be described as ‘nodes’ and ‘paths.’ As stated above, conceptual nodes are corresponding concepts or (aspects of) models in different districts. Paths consist of possible connections between divergent concepts and (aspects of) models in different districts that might be fruitful to explore in the future. A description of nodes and paths, as given below, will hopefully lead to the crossing of

88

Chapter 9

disciplinary boundaries and an enhancement of our understanding of the interaction process between users and technology.

4.1 Conceptual nodes Conceptual nodes are indicated on the map as landmarks in bold type face. They have been placed in the district with which they are presumed to have the strongest liaison, but the suffixes show that they are also mentioned in chapters that have other districts as their primary location. An important first node is the concept of ‘behavioral affordance’ (district 1). It occurs in psychological models of human-technology interaction and in action theory (districts 1 and 2) and also in philosophy of technology (district 5). Behavioral affordances have a particular significance in analyzing technology-user interactions, because they arise from the interaction itself (I) and because they offer an explanation of the less conscious (perceptual) parts of the interaction process. In philosophy of technology, behavioral affordances are used in approximately the same sense as in their original, cognitive psychological context. This meaning, however, is broader in the other, related landmarks in district 5 (see section 3 above). ‘Information processing’ (district 2) forms a second conceptual node. It plays a crucial role in cognitive psychology, environmental psychology and safety studies (districts 1, 2 and 6; in the latter case resulting in organizational learning at a meso-level). Actually, it is a conglomerate of concepts that address the various stages of the input and mental throughput of information when humans interact with technology. As such, it represents the rudimentary intrapersonal mechanisms that co-shape the interaction. Once the information has been filtered, transformed and combined into a usable form, other intra- and interpersonal mechanisms (U) come into play, which are joined in the third node: social psychological determinants and processes that form the basis of intentional and habitual behavior (district 3). References to this node (or parts of it) are also found in environmental psychology, action theory and household and consumer science (districts 1, 2 and 7). In action theory, safety studies and cognitive psychology (somewhat dispersed in the districts 1, 2 and 6) a distinction is made between ‘skill-, rule- and knowledge based behavior.’ This is the fourth node we wish to mention (in district 2). The distinction is mirrored in the different ergonomical and functional relations between technologies and users, which

Technology and users

89

are conceptualized in terms of ‘operations, actions or acts’. Therefore, identifying these levels of behavior allows for a better analysis of the ways that technology and human behavior might influence each other. This designation of nodes unavoidably incorporates a historical bias. The four concepts that were selected here as nodes originate from theoretical frameworks with a longer history, which obviously allowed them more time to be incorporated into other frameworks than the newer ones. They also emanate from a period in which human-technology interaction as such did not yet receive much thought (except for affordances). Therefore, most of these nodes are located at the user side. Despite this bias, they provide opportunities for building bridges across districts. Other and newer concepts, however, might also acquire a status as ‘node’, when connecting conceptual paths, which we will describe next, are trod more often.

4.2 Conceptual paths Conceptual paths, linking disparate concepts and (segments of) frameworks in different districts, are manifold. A number of them are already being researched on a regular basis, like connections between the psychological districts or between the two philosophical areas. Others are too evident to be highlighted here, such as the path between safety studies and cognitive psychology, since cognitive psychology is the primary conceptual basis of safety studies. The description will be limited to the less apparent and more infrequently considered avenues, that, in our opinion, deserve more attention in the near future. Previously, it was stated that, despite the contrasts between some of the philosophical concepts, environmental psychological concepts and concepts regarding mental mechanisms in the districts 1, 2 and 4, there are also theoretical affinities between them. Two types of affinities may be discerned: concepts may supplement each other because they cover adjacent or partially overlapping terrain, or they may complement each other because they involve opposite sides of the interaction (user and technology) at about the same level of analysis. ‘Conceptual paths’ of both types are shown on the map. The first path is of the supplementary type: it concerns closely related concepts in the areas of technology-user interaction (I) and of direct effects of technology on this interaction (T Æ I). The concepts that are linked by this path are ‘behavioral affordances,’ ‘artifactual scripts,’ and ‘artifactual action programs.’ They represent different, but closely related, attributes of

90

Chapter 9

artifacts influencing their use: those that are inherent and invoke a certain usage in a preconscious, perceptual way, and those that are more or less deliberately inscribed and explicitly communicate specific ways of use. The concept of ‘affordances’ in district 5 might also be added to this list, although it also indicates effects of technologies on a meso and macro level. The second path is of the first type as well. It goes between a number of concepts on the user side of interaction (IÅU) in the psychological and philosophical districts (1 through 5): ‘mental models’, ‘behavioral scripts’, ‘schemata’, ‘human action programs’, ‘design and use logic’, and ‘multistability’. These are fairly cognate concepts regarding the presence and structure of more or less conscious inclinations, knowledge and thoughts that users as well as designers bring into the interaction process. Affinities of the second, complementary type exist between several of the concepts on the technology and user sides, but we wish to highlight the relation between ‘artifactual scripts’ on the one hand, and ‘behavioral scripts’, ‘schemata’, and ‘mental models’ on the other (the third path). In actor-network theory a parallel is drawn between characteristics of an artifact prescribing its usage, and the script in a stage-play instructing the actors. Cognitive psychology uses the script concept to indicate relatively automated action sequences stored in the mind of the user, much like the representation of a stage script in the actors' memory. These two definitions of scripts complement each other: user behavior comes about in the interaction between the ‘artifactual script’ and the ‘behavioral’ or ‘mental’ script. To be sure, this is recognized in both fields of research. Actornetwork theory does not see scripts as merely intrinsic properties of artifacts, since scripts are formulated in the network that arises between user and artifact. And cognitive psychology takes into account the influences of artifacts on the formation of behavioral scripts. To arrive at a design that instigates proper and intended employment of an artifact, knowledge is necessary of both these scripts (or design and use logic, casu quo mental models of designers and users) and of their mutual relations. The last two paths are of the first, supplementary type again. The fourth path connects ‘person-environment fit’ and ‘action facilitation’ (district 1) with ‘user-profile affordances and constraints’ (district 5) and ‘compatibility’ in household systems (district 7). All of these concepts refer to the amount of fit or mis-fit between technology and (groups of) users, albeit from different perspectives. The ‘person-environment fit’ model examines the congruence between environmental resources and demands on the one hand, and user needs and abilities on the other. ‘Action facilitation’

Technology and users

91

considers the degree to which perceptual, cognitive and motor processes are accommodated by technology. ‘Compatibility’ focuses on the correspondence between technology and household processes. In these three concepts, interaction is the starting point (I). User profile affordances and constraints are located further on the technological side (TÆI), indicating the possibility that a specific design of artifacts might exclude certain groups from using it. In spite of the differences, these concepts might borrow useful insights from each other ʊ thereby crossing both main edges in the map ʊ and also methods to measure the amount of fit or mis-fit. The fifth path is between ‘social relations affordances and constraints’ (district 5) and ‘compatibility’ and ‘balancing’ in household and consumer science (district 7). The first concept refers to the possible influences of technological artifacts on social roles and relations, and this might be a relevant addendum in the study of compatibility and balancing issues in the household system. On the other hand, existing insights in the latter issues may increase the conceptual depth of social relations affordances. Many more paths can be identified, for example between ‘interfaces’, ‘artifactual scripts’ and ‘technological intentionality’ (districts 1 and 4), since the first concept may support the other two; between ‘perceived control’ and ‘system control’ (districts 3 and 6) to study their proportional effects in different circumstances; or between ‘social, organizational and cultural affordances/constraints’ and ‘group/societal norms and goals’ (district 3 and 5) to chart mutual relations. While searching for feasible paths the map proved to be very inspiring. We hope that it will invite readers to discover paths that can lead to new alliances between disciplines and stimulate new insights and research.

5.

CONCLUSIONS

All contributions to this part of the book have made one thing very clear: technological and user variables cannot be dealt with separately, but should be conceptualized through the complex interplay between them. Focusing on one of these poles (technological or behavioral determinism) denies the importance of the other. The positioning of existing concepts as landmarks on a map makes clear that viewpoints may be clustered into a number of districts, which are close in some cases and distant in others. The clarifying power of this ‘helicopter view’ resides in the visualization of relations between (parts of) those districts that remain obscure when they are approached in a mono-disciplinary way. Therefore, the most important

92

Chapter 9

outcomes of reading this conceptual map consist of precisely these relations: the edges, nodes and paths. They illustrate the necessity of an interdisciplinary approach as is laid down in this book, and also that the dispersion of relevant research across several disciplines does not need to impede co-operation. Nodes and paths provide plenty of opportunities for the exchange of knowledge and for taking joint initiatives in future theorizing and research. The map shows that some areas are subject to intensive scrutiny while others were left barren. The dense areas ʊ in particular the realm of interaction and the user side on the micro level ʊ contain concepts that are partially overlapping or that differ in minor ways. This is understandable in view of the different interests of the disciplines and areas of application that are involved. But scientific progress in the comprehension of technologyuser interaction would dictate a more concise collection of concepts that may resist Ockham’s razor. Fallow ground is apparent in the technological area and in the meso and macro fields of the map. These domains require supplementary theoretical conjectures and new methods. In particular, little is known about the interaction processes in groups (e.g. decisions in households regarding the acquisition and the use of technology) or about interactive consequences for society as a whole. Such insights are essential in order to provide adequate information for the designers of tomorrow’s products.

REFERENCE Lynch, K. (1960). The image of the city. Cambridge (MA): MIT Press.

PART 2 TECHNOLOGY, BEHAVIOR AND SOCIOTECHNICAL PRACTICES

Chapter 10 TECHNOLOGY AND BEHAVIOR: The Case of Passenger Transport

Laurie Hendrickx and Anton J.M. Schoot Uiterkamp

1.

INTRODUCTION

Different policy strategies may be used to decrease the adverse environmental effects of motorized transport. Technical innovations, such as catalytic converters, engine efficiency improvements, or low-noise road surfaces, may lower the emissions, energy use, or noise production caused by passenger transport. Behavioral measures, on the other hand, aim to change environmentally relevant decisions and behaviors of transport consumers, such as the type choice when purchasing a car, the transport mode choice, or the driving speed. Despite obvious complexities in predicting the effects of such policy measures, assessments of the environmental potential are available for a variety of improvement options and impact categories. For instance, Bouwman & Moll (2000) and Moomaw & Moreira (2001) review the energy use reduction potential of various, mainly technical improvement options in passenger transport. Cavalini, Hendrickx & Rooijers (1995, 1996) estimate the energy use and CO2 emission reduction potential of several behavioral policy measures. Dings (1996) and Nijland (1997) review the emission and noise reduction potential of various (mainly technical) measures. Such assessments are usually based on ceteris paribus assumptions regarding other relevant factors. When assessing the effects of technical innovations, behavioral variables (e.g. total transport demand) are assumed 95 P.-P. Verbeek and A. Slob (eds.), User Behavior and Technology Development: Shaping Sustainable Relations Between Consumers and Technologies, 95-106. © 2006 Springer.

Chapter 10

96

to be constant or to follow some predefined autonomous trend. In other words, technology (T) and behavior (B) are considered as independent. However, in reality technology and behavior often interact. For instance, vehicle or infrastructural innovations aimed at increasing the efficiency of the transport system may also (inadvertently) affect environmentally relevant behaviors (T Æ B interaction). Or, vice versa, policy campaigns aimed at changing the behavior of transport consumers may also influence key decisions of technology designers or producers (B Æ T interaction). This chapter focuses on the first type of interaction (T Æ B). We present a conceptual model that enables us to systematically identify and analyze possible T Æ B interactions. We use the model to analyze three environmentally relevant decisions and behaviors of transport consumers: (a) type choice when purchasing a car; (b) transport mode choice; and (c) speed choice. On the basis of recent research on the determinants of these behaviors, we analyze how these determinants may be affected by technological innovations, and we assess how this would affect the environmental impact of the innovation. The chapter ends with general conclusions and an outline of a research agenda based on our approach.

2.

CONCEPTUAL MODEL

Figure 10-1 presents a conceptual model of T Æ B interactions. Technical innovations, such as an increase in motor efficiency, a catalytic converter, or

Technology and behavior

97

a low-noise road surface, are aimed at improving the average system efficiency, i.e. to lower the negative environmental effects (the costs) per functional unit produced (the benefits). Costs may refer to various impact categories, such as energy use, the emission of harmful substances, or noise production, whereas benefits are usually expressed in passenger kilometers (pkm). Increases in system efficiency due to technical innovations will lower the overall impact of mobility, assuming ceteris paribus with regard to other relevant system parameters (e.g. other vehicle fleet characteristics), the total travel demand, and the modal split. This intended effect of technological innovations is shown in the upper route of Figure 10-1. The possibility of T Æ B interactions is represented by the lower route in the figure. The introduction of technical innovations may affect various car characteristics, such as purchase price, fuel costs per km driven, safety level, or degree of comfort when driving (left-most lower box). Such changes may influence environmentally relevant behaviors, such as type choice, travel mode, or driving style (right-most lower box). To predict whether specific changes in car characteristics will affect mobility decisions and behaviors, we need to understand the psychological processes underlying such decisions (middle lower box). Numerous models of the psychological processes underlying mobility behavior have been proposed. For instance, decision theoretical models, like Multi Attribute Utility Theory (e.g. Von Winterfeldt & Edwards, 1986) have been used to analyze car purchase decisions (cf. Van Oijen, 1996). “Means-end chain” models (e.g. Gutman, 1997), originally developed by consumer and marketing psychologists, have been used to analyze the transport mode choices of commuters (cf. Breemhaar, Van Gool, Ester & Midden, 1995). Attitude theories, like the Theory of Planned Behavior (Ajzen, 1991), and various models based on Schwartz’s (1977) Norm Activation Theory, were used to describe and explain travel behavior and travel mode choices; for a recent review of this line of studies, see Harland (2001). Travel mode choices have also been analyzed and studied from a social dilemma perspective; see, e.g. Garvill (1999), Van Lange (2000). It is unlikely that one particular type of model is appropriate for all mobility-related decisions and behaviors; the psychological processes that underlie infrequent decisions with important, long-term consequences (e.g. purchasing a new car) may differ considerably from, for instance, highly repetitive, habit-driven types of behavior (e.g. speed choice on a particular road). A detailed discussion of the various mobility behavior models proposed in the literature, and of the extent to which they are appropriate for different types of behavior, is beyond the scope of this chapter.

Chapter 10

98

A basic notion all models appear to share is that the available choice options ʊ for instance, cars one may purchase, or travel modes available for making a particular trip ʊ are evaluated in terms of a limited set of “value dimensions” (or “decision attributes”, “goals”, “needs”, “costs and benefits”, “individual outcomes”, “utilitarian consequences”, or whatever label is used to indicate a set of evaluative aspects). Empirical studies, discussed in the next sections, have shown that aspects like expected travel time, perceived safety, reliability, flexibility, and expected costs play a key role in many mobility-related decisions. Innovations that affect these factors (or are expected to do so) are likely to result in behavioral changes: for example, changes in expected costs or in perceived safety may affect car type preferences; changes in expected fuel costs or travel time may alter transport mode choices; and changes in perceived comfort may influence driving style. In turn, such behavior changes may affect relevant efficiency parameters and, consequently, the environmental impact of the innovation. For instance, changes in car fleet composition, in modal split, or in mean driving style may affect energy use, emissions, and noise. In sum, environmentally-motivated technical innovations aim to increase system efficiency parameters and to decrease the total environmental impact. In practice, however, such innovations may also have behavioral side effects that alter the intended effects. The model in Figure 10-1 constitutes a general and simplified representation of the complex processes that determine the environmental impacts of mobility. It needs to be elaborated to analyze the effects of a specific innovation. The crucial point, however, is that to recognize T Æ B interactions and to assess their effects, we need: (a) to identify environmentally relevant behaviors, (b) to understand how changes in these behaviors affect environmental impacts, (c) to know the main determinants of these behaviors, and (d) to understand if and how the technical innovation studied affects these determinants. A comprehensive description of environmentally relevant behaviors and their relation to system efficiency parameters is beyond the scope of this chapter (cf. Cavalini, Hendrickx & Rooijers, 1993). To illustrate our approach, we will focus on the three examples in the right-most lower box of Figure 10-1: car type choice, car use, and driving style.

3.

CAR TYPE CHOICE

If a technical innovation affects important determinants of car type choices, this will either accelerate (if the innovation makes a car more

Technology and behavior

99

attractive) or decelerate (if the innovation makes it less attractive) the implementation of the innovation. As a consequence, the innovation’s aggregate environmental benefits will either be reinforced (if the implementation is accelerated) or weakened (if it is delayed). Determinants of car type choice have been studied by, e.g. Van Oijen (1996) and Hagreis (1996). Based on the scientific literature, popular car magazines, advertising materials, and interviews with car dealers, Van Oijen identified 26 car characteristics that affect car type choice. In an interview study, the relative importance of these attributes for business drivers with a leased car was assessed. The 26 characteristics could be meaningfully categorized into nine main factors. The average importance of the factors, as obtained by Van Oijen, is presented in the middle column of Table 10-1. Hagreis (1996) used a similar design to study car type preferences of private drivers, i.e. people who mainly use their car for private purposes like shopping, social visits, or holidays. The right column of Table 10-1 presents mean importance scores of the nine main car characteristics, as obtained by Hagreis. Table 10-1. Relative importance of car characteristics for car type choice relative importance $

relative importance $

(business drivers)

(private drivers)

Safety

0.19

0.16

Reliability

0.18

0.20

Comfort

0.15

0.10

Performance

0.12

0.07

Costs

0.10

0.17

Appearance

0.08

0.07

Functionality

0.08

0.10

After-sales (e.g. service)

0.06

0.05

Environmentally friendly

0.04

0.08

car characteristic

* Source: Van Oijen (1996), Hagreis (1996) $

Higher score Æ characteristic more important; scores rescaled to add up to 1

Table 10-1 shows that safety, reliability, and to a lesser extent comfort, costs, and performance, appear to be the main aspects people consider in

100

Chapter 10

selecting a car. Innovations that ʊ in the consumer’s eyes ʊ decrease car safety or reliability, may encounter serious implementation problems. Environmentally-motivated innovations that may evoke safety concerns include the use of lightweight materials in cars or the introduction of alternative fuels like hydrogen. Reliability concerns may drive people away from radical technological changes, such as the introduction of fuel cell cars. Characteristics like comfort or cost may be used to accelerate the implementation of an innovation. For instance, emphasizing that cruise control and low-noise tires enhance driving comfort may be a more efficient way to promote such innovations than highlighting their environmental benefits. As often, the cost aspect forms a double-edged sword. Innovations that reduce travel costs, for instance because they lower the fuel consumption per km, may be implemented relatively quickly, but they may also induce shifts in modal split (see next section). Table 10-1 also reveals that the average driver weighs costs more heavily than environmental benefits. This suggests that people will not be very willing to pay more for innovations that only reduce environmental impacts. Temporary subsidies or tax compensations, for instance on low-sulfur fuels or on hybrid cars, may be necessary to overcome financial barriers.

4.

TRANSPORT MODE CHOICE

The environmental effects of transport mode choices are complex. The relative efficiency of different transport modes varies across impact categories, but also depends on trip-specific factors, e.g. the particular vehicle used, the occupancy rate, and the trip length. For a comparison of the energy and emissions factors associated with different modes of passenger transport in the Netherlands, see Van den Brink & Van Wee (1997). In general, traveling by car or airplane has a larger environmental impact than traveling by train, bus, or metro; non-motorized transport (cycling and walking) clearly has the lowest impact. The determinants of transport mode choice have been studied by e.g. Steg (1996) and Tertoolen, Van Kreveld & Verstraten (1998). Speed and independence (“leave any time I want, get anywhere I want”) are important determinants of transport mode choice. Other relevant factors are comfort, costs, social safety, physical safety, health, environmental friendliness, and luggage capacity. Each of these determinants may serve as a mediator of T Æ B interactions. For instance, innovations that decrease the perceived level of independence (e.g. electric cars that require frequent recharging) may be difficult to implement (see previous section). However, once

Technology and behavior

101

implemented, such innovations may bring about a shift in modal split away from the car, which augments the innovation’s environmental benefits (assuming the shift is towards transport modes with lower impacts). Technical innovations that increase automobile fuel efficiency also reduce driving costs, which may increase car use, either because the total mobility demand increases or because the modal split shifts towards car use. This effect is probably not very large, as both studies cited above indicate that for the average driver costs are not a main determinant of transport mode choice. Studies on fuel price elasticity confirm this finding: fuel price increases tend to have relatively small effects on car mileage (price elasticity of approximately –0.2/–0.3, see Pronk & Blok, 1991). Another possible mediator of T Æ B interactions is the extent to which (potential) drivers perceive car use as harmful to the environment. It is conceivable that if environmental concerns play a significant role in transport mode choice, people may curtail their car use because of such concerns. If so, innovations that make cars more “environmentally friendly” may backfire if they induce people to lift the self-imposed restrictions. Unfortunately, the extent to which environmental concerns play a role in transport mode choices is unclear. The studies cited above yield conflicting findings; other empirical studies (reviewed in Steg, 1999) show that the correlation between environmental concerns and car use tends to be rather weak. Steg (1999), for instance, obtained a correlation value of –0.11. Nevertheless, technology is sometimes used as an excuse for unwanted behavior, for instance when people justify excessive car use by pointing out that their car has a catalytic converter or runs on “clean” fuel.

5.

DRIVING STYLE

Driving behavior affects the environmental impacts of car use. Average fuel consumption, the emission of harmful substances, and noise production decrease if drivers select an adequate cruising speed, avoid extreme acceleration, and anticipate oncoming traffic situations (e.g. Ericsson, 1999; Van der Voort, 2001). On the basis of a literature review, Orlemans (1997) identified a large number of factors that possibly influence driving style. Individual characteristics, such as age and gender, are related to driving style: young drivers and/or male drivers tend to drive more aggressively than older and/or female drivers. Infrastructural characteristics, such as lane width and road surface roughness, also affect speed choice (Martens, Comte & Kaptein, 1997). Orlemans mentions various car characteristics that may be related to driving behavior (e.g. top speed, motor power, car age, car size and type, brand, comfort, fuel type, fuel use), but relevant research is lacking.

Chapter 10

102

Side effects of a technical innovation on driving style determinants may alter the innovation’s overall impact. If fuel cost considerations affect the way people drive, then increasing a car’s fuel efficiency may result in more aggressive driving. T Æ B interactions on driving style may also be mediated by driving comfort. Driving comfort tends to decrease at higher speeds, for instance because of the noise level inside the car, the effects of road surface irregularities, and the mental effort required for driving, all tend to increase with driving speed. Such comfort-related factors may be a reason for drivers to restrict their speeds. If this is the case, then technologies that increase driver comfort, such as low-noise tires, silent road surfaces, or cruise control, may evoke higher driving speeds. And if environmental concerns affect the way people drive ʊ which, to our knowledge, has not been studied yet ʊ then innovations that make cars (appear) more “environmentally friendly” may tempt people to alleviate self-imposed restrictions regarding, e.g. speed choice. However, the lack of knowledge about the determinants of driving style prohibits firm predictions about the effects of technical innovations on driving behavior.

6.

CONCLUSIONS

The conceptual model of T Æ B interactions in Figure 10-1 offers a useful methodology for identifying possible behavioral side effects of technical innovations. However, the driving style example particularly shows that, of the four steps necessary to identify T Æ B interactions (see above), identifying the behavioral determinants (step c) is the hardest one. For many behaviors, insight into the underlying mechanisms and processes is still rudimentary. With regard to future research we therefore expect that studies aimed at clarifying the determinants of environmentally relevant behaviors will have the largest added value. In theory, behavioral side effects of new technologies may increase their environmental effects. In fact, as illustrated in the section on car type choice, a behavioral side effect may be used intentionally to reinforce the primary effects of an innovation. However, our examples suggest that many T Æ B interactions will be counterproductive and will result in what economists call rebound effects (cf. Binswanger, 2001). Timely awareness of such effects may prevent unrealistic optimism and overestimation of the environmental benefits of technological progress. Particularly in the transport domain, environmental policy targets often needed to be “adjusted” because the yield of technical improvements turned out to be less than expected. Awareness of possible T Æ B interactions may also create opportunities, e.g. for car designers or policy makers, to avoid or minimize undesirable side effects.

Technology and behavior

103

A four-step procedure for identifying possible T Æ B interactions was presented in section 2. Case-specific information on if and how a new technology will affect relevant behaviors may be collected in various ways. Sometimes literature-based analyses may provide sufficient information. It may also be necessary to conduct interview studies in which relevant consumer groups are asked how the new technology would affect their behaviors. In addition, small-scale pilot projects may be conducted in which the behavior of new technology users is systematically monitored. If such studies indicate that major behavioral side effects are possible, the introduction of the innovation should be accompanied by large-scale evaluation studies, in which both the critical behavior and the targeted environmental parameters are examined. While consumer costs do constitute an important mediator for T Æ B interactions (T Æ Costs Æ B, see above), costs are by no means the only route through which T Æ B interactions may occur. Our examples suggest that perceived safety, trip speed, reliability, comfort, and environmental friendliness may also mediate T Æ B interactions. Therefore, when studying T Æ B interactions, focusing solely on financial rebound effects is insufficient. Recently, Binswanger (2001) expanded the traditional economic (price-mediated) analysis of the rebound effect to include “a rebound with respect to time …This means that the introduction of a time-saving device for the production of a service will also lead to an increase in the demand for a service” (op. cit., page 128). The T Æ B interactions presented here suggest that Binswanger’s quantitative modeling of time-related rebound effects, though highly valuable, only constitutes the first in a series of necessary model expansions.

7.

A RESEARCH AGENDA

To conclude, we will outline a research agenda based on our approach. With regard to the information necessary to identify possible T Æ B interactions (see the four-step procedure indicated above), a better understanding of the psychological determinants of various mobility behaviors is most critical. The majority of studies on mobility behavior determinants have focused on travel mode choice (for recent examples, see Steg, Vlek & Slotegraaf, 2001; Hunecke, Blöbaum, Matthies & Höger, 2001). Other behaviors, such as the decision whether or not to buy a car, type choice when purchasing a car, the routing and timing of trips, and driving style, have received much less attention. This is remarkable, as some of these behaviors are highly relevant from an environmental point of view.

104

Chapter 10

Therefore, studies addressing the psychological determinants of, say, car purchase decisions, type choice, and driving style should form a first priority. Such studies should preferably examine actual rather than reported preferences and behaviors, and they should aim at determining how these behaviors vary as a function of individual, car and traffic-system characteristics. For some behaviors it may be necessary to distinguish subgroups of transport consumers, differing in their sensitivity to specific behavior determinants. For instance, Table 10-1 shows that “costs” constitute a more important determinant of type choice for private drivers than for business drivers. This research line should aim at developing comprehensive models of the psychological factors and processes underlying various mobility behaviors. A second research line should focus on actual or foreseeable technological innovations in the transport system. First, the impacts of an innovation on actual car or infrastructural characteristics should be analyzed and assessed. Next, these “innovation effects” should be related to the sets of behavioral determinants provided by the first research line. This will result in specific hypotheses about the behavioral effects of the innovation at hand. Again, it may be necesssary to specify such effects separately for subgroups of mobility consumers differing in sensitivity to behavior determinants. Such hypotheses should then be tested, either in small-scale pilot studies, or in monitoring programs that accompany the large-scale introduction of the technology (see previous section). The primary aim of this second research line is to assess the nature, the size, and the environmental effects of T Æ B interactions, induced by specific innovations. However, as the introduction of a new technology may present unique opportunities for conducting (field) experiments, these studies may also be useful for validating more general behavior models and theories. In this way, the two research lines proposed here may fruitfully interact.

REFERENCES Ajzen, I. (1991). The theory of planned behavior. Organizational Behavior and Human Decision Processes, 50, 179-211. Binswanger, M. (2001). Technological progress and sustainable development: what about the rebound effect? Ecological Economics, 36, 119-132. Bouwman, M. E. & Moll, H. C. (2000). Energy use reduction potential of passenger transport in Europe. Transport Review, 20, 191-203. Breemhaar, B., Van Gool, W., Ester, P., & Midden, C. (1995). Life styles and domestic energy consumption: a pilot study. In S. Zwerver, R. S. A. R. Van Rompaey, M. T. J. Kok, & M. M. Berk (Eds.), Climate change research. Evaluation and policy implications (pp. 1235-1240). Amsterdam: Elsevier.

Technology and behavior

105

Cavalini, P. M., Hendrickx, L., & Rooijers, A. J. (1993). Differences among car user groups regarding CO2 emissions (Rep. No. IVEM OR-65). Groningen: Center for Energy and Environmental Studies (IVEM), University of Groningen. Cavalini, P. M., Hendrickx, L., & Rooijers, A. J. (1995). Verschillen tussen groepen automobilisten met betrekking tot milieurelevant gedrag {Differences among car user groups regarding environmentally relevant behavior}. Milieu, Tijdschrift voor Milieukunde, 10, 18-25. Cavalini, P. M., Hendrickx, L., & Rooijers, A. J. (1996). Verschillen tussen groepen automobilisten met betrekking tot hun gevoeligheid voor beleidsmaatregelen {Differences among car user groups regarding their sensitivity to policy measures}. Milieu, Tijdschrift voor Milieukunde, 11, 138-148. Dings, J. M. W. (1996). Kosten van milieu-effecten van technische maatregelen in het verkeer Delft, the Netherlands: CE, Centrum voor Energiebesparing en schone technologie. Ericsson, E. (1999). The relation between vehicular fuel consumption and exhaust emission and the characteristics of driving patterns. In L. J .Sucharov (Ed.), Urban Transport 5: Urban transport and the environment for the 21st century (pp. 137-147). Southampton: W.I.T. Press. Garvill, J. (1999). Choice of transportation mode: factors influencing drivers’ willingness to reduce personal car use and support car regulation. In M. Foddy, M. Smithson, S. Schneider, & M. Hogg (Eds.), Resolving social dilemmas: dynamic, structural, and intergroup aspects (pp. 263-279). Philadelphia (PA): Psychology Press. Gutman, J. (1997). Means-end chains as goal hierarchies. Psychology & Marketing, 14, 545-560. Hagreis, A. (1996). De autotypekeuze van privérijders (Rep. No. IVEM DV-36). Groningen, the Netherlands: Center for Energy and Environmental Studies (IVEM), University of Groningen. Harland, P. (2001). Pro-environmental behavior. Ph.D. Thesis, University of Leiden, the Netherlands. Hunecke, M., Blöbaum, A., Matthies, E., & Höger, R. (2001). Responsibility and environment. Ecological norm orientation and external factors in the domain of travel mode choice behavior. Environment and Behavior, 33, 830-852. Martens, M. H., Comte, S., & Kaptein, N. A. (1997). The effects of road design on speed behavior: a literature review (TNO report TM-97-B021). Soesterberg, the Netherlands: TNO Human Factors Research Institute. Moomaw, W. R. & Moreira, J. R. (2001). Technological and economic potential of greenhouse gas emissions reduction. In B. Metz, O. Davidson, R. Swart, & J. Pan (Eds.), Climate Change 2001: Mitigation (pp. 167-277). Cambridge (UK): Intergovernmental Panel on Climate Change, Cambridge University Press. Nijland, H. (1997). Invloed van technische maatregelen wegverkeer op geluidhinder. In J.A.Annema & R. M. M. Van den Brink (Eds.), Proceedings Colloquium “Verkeer, Milieu & Techniek”, 1997 (pp. 171-178). Bilthoven, the Netherlands: Rijksinstituut voor Volksgezondheid en Milieuhygiene, RIVM. Orlemans, F. M. (1997). Rijstijl en milieu: een literatuur- en empirisch onderzoek naar de determinanten en milieugevolgen van rijstijl (Rep. No. IVEM DV-71). Groningen, the Netherlands: Center for Energy and Environmental Studies (IVEM), University of Groningen. Pronk, M. Y. & Blok, P. M. (1991). De prijselasticiteit van energiegebruik in het wegverkeer Rotterdam, the Netherlands: Nederlands Economisch Instituut (NEI). Schwartz, S. H. (1977). Normative influences on altruism. In I. Berkowitz (Ed.), Advances in experimental social psychology, vol. 10 (pp. 221-279). New York: Academic Press.

106

Chapter 10

Steg, E. M. (1996). Gedragsverandering ter vermindering van het autogebruik: theoretische analyse en empirische study over probleembesef, verminderingsbereidheid en beoordeling van beleidsmaatregelen {Behavior change for reducing car use in The Netherlands}. Ph.D. Thesis, University of Groningen, The Netherlands. Steg, L. (1999). Verspilde energie? Wat doen en laten Nederlanders voor het milieu. Den Haag, the Netherlands: Sociaal Cultureel Planbureau (SCP). Steg, L., Vlek, Ch., & Slotegraaf, G. (2001). Instrumental-reasoned and symbolic-affective motives for using a motor car. Transportation Research Part F, 4, 151-169. Tertoolen, G., Van Kreveld, D., & Verstraten, B. (1998). Psychological resistance against attempts to reduce private car use. Transportation Research A, 32, 171-181. Van den Brink, R. M. M. & Van Wee, G. P. (1997). Energiegebruik en emissies per vervoerswijze. Bilthoven, the Netherlands: Rijksinstituut voor Volksgezondheid en Milieuhygiene, RIVM. Van der Voort, M. (2001). Design and evaluation of a new fuel-efficient support tool. Ph.D. Thesis, University of Twente, The Netherlands. Van Lange, P. A. M. (2000). Choosing between personal comfort and the environment; solutions to the transport dilemma. In M. Van Vugt, M. Snyder, T. R. Tyler, & A. Biel (Eds.), Cooperation in modern society. Promoting the welfare of communities, states and organizations; (pp. 45-63). London: Routledge. Van Oijen, R. (1996). De autotypekeuze van zakelijke automobilisten (Rep. No. IVEM DV39). Groningen, the Netherlands: Center for Energy and Environmental Studies (IVEM), University of Groningen. Von Winterfeldt, D. & Edwards, W. (1986). Decision analysis and behavioral research. Cambridge: Cambridge University Press

Chapter 11 SUSTAINABLE TECHNOLOGIES AND EVERYDAY LIFE

Gert Spaargaren, Susan Martens and Theo A.M. Beckers

1.

INTRODUCTION

This paper introduces a methodology for analysing the opportunities for a more sustainable organisation of daily routines in consumption. In this approach, which is referred to as ‘Social Practices Approach’, the interaction between technology and behavior is put at the centre of the analysis. In traditional discussions about transitions, technology-related concepts and the jargon of system dynamics are often favoured at the expense of behavioral factors. These technology-biased approaches do not acknowledge the central role citizen-consumers as ‘knowledgeable and capable actors’. Although transition theories are right in emphasising the importance of (sustainable) technologies, the role of human agency in technological change must be given systematic attention, because citizen-consumers are the ‘changeagents’ who make transitions work. While making a strong case for the role of human agency in transition-processes, we think it equally important to avoid some of the pitfalls of the socio-psychological perspectives that have often been connected with existing analyses of behavioral changes (Spaargaren, 1997). Within the socio-psychological paradigm, a strong emphasis is put on the motives, values and beliefs of an individual human being. It is assumed or believed that the only way to bring about sustainability transitions is to change the value- or belief-systems that are ‘guiding’ individual behavior. This emphasis on micro-cultural factors results in the neglect of the influence of social (technological) structure on behavior.

107 P.-P. Verbeek and A. Slob (eds.), User Behavior and Technology Development: Shaping Sustainable Relations Between Consumers and Technologies, 107-118. © 2006 Springer.

Chapter 11

108

The Social Practices Approach is situated between the technologicalsystem approach and the social-psychological approach. It combines the notion of human agents as knowledgeable and capable actors playing an active role in sustainable transitions, with an equal emphasis on the influence of the social and technological context on human behavior. It provides a common ground for technologists and sociologists to analyse and discuss transitions towards sustainability. We will first present the conceptual model and some core concepts of the Social Practices Approach (2), which we then ‘put to work’ by analysing transition processes in two domains of daily life, ‘sustainable living’ and ‘tourism mobility’ (3). We conclude with a few remarks on the future research- and policy-agenda on transitions in daily life.

2.

THE SOCIAL PRACTICES APPROACH

The Social Practices Approach ʊ derived from Giddens’ structuration theory ʊ does not start from the individual attitude or norm for predicting the environmentally (un)friendly behavior of an individual, but instead departs from the actual behavioral practices an individual shares with other human agents (Giddens, 1984; Spaargaren, 2001). These social practices are at the centre of the model. When talking about ‘transitions’ in daily routines, we refer to specific changes with respect to the ways in which groups of citizen-consumers feed themselves, inhabit their homes, or travel for holidays. The concept of transition refers in structuration theory to circumscribed or ‘defined’ processes of change. Transitions are ‘trajectories of change’ within the time-space bound reproduction of social practices. Figure 1 illustrates that, for understanding transitions within social practices, both the concepts of ‘lifestyles’ and of ‘systems of provision’ are of crucial importance.

2.1

Lifestyle(s)

The lifestyle of an individual actor is constructed from a series of building-blocks that relate to the sets of social practices an individual is involved in when enacting his or her daily life, together with the story-telling that goes along with it (Giddens, 1991). When looking at transitions from a lifestyle-perspective, two obvious questions come to the fore. First, to what extent does the transition in one’s social practices, and its corresponding lifestyle-segment, fit into the overall level of (environmental) quality an actor wants to maintain for him or herself? Second, what can be said about

Sustainable technologies and everyday life

109

the level of integration of the different lifestyle segments or social practices? When the assumption of an attempted (minimum) level of integration makes sense, it can be expected that strong articulations of environmental codes at one social practice might flow into other domains of actors’ daily life. These spill-over processes, as well as the reverse processes of purposeful isolation or insulation of certain lifestyle-segments from environmental considerations, are interesting topics for (future) research. Actors

Social Practices

Structures

Dwelling the house Individual agent

Going for a holiday

Rules & Resour ces

Shopping for Food

Figure 11-1. Social Practices and the duality of structure

2.2

Systems of provision

The concept of ‘system of provision’ refers to the contexts of action, specified in terms of the sets of rules and resources (structures) that ‘help’ agents to organize social practices. Without structure, no action is possible, as is rightly emphasized by structuration theorists. Outlining the ‘enabling’ aspect of social structures is important because in the social-psychological paradigm structures tend to figure only in their constraining role. The rules and resources that ‘govern’ social practices ʊ and by implication the various lifestyle-segments of the agent ʊ are different for the distinct social practices. The systems of provision implicated in the social practices of food, differ from the ones involved in sport, travel or dwelling. It is important to keep in mind that actors are not ‘outside’ the systems of provision. On the contrary: They (have to) make use of the systems of provision, and by

Chapter 11

110

making use of them, they help to reproduce the systems themselves. With regard to the social practices of the domestic consumption of energy, water and waste-services, Otnes described this notion of duality of structure as a process of householders “serving and being served” by collective, sociomaterial systems, such as the drinking water-system, the energy systems, etc. (Otnes, 1998).

2.3

Technology

When transitions are analyzed following the Social Practices Model, it should be possible to avoid both the voluntarist bias of socio-psychological approaches as well as the deterministic bias of technological system analyses. Before we go on to discuss the application of the model in empirical research, we will first look in more detail at the role of technology in transitions. Otnes’ hybrid concept of socio-material systems expresses the fact that technological systems (expert systems) involved in the production and consumption of energy, water and waste-services, have both a ‘material’ and ‘social’ dimension to them. The technological dimension ʊ in the terms of everyday life ʊ refers to the pipes, pumps and taps as vital parts of the system as a material system. Technology as a theoretical concept refers to the ‘allocative resources’, in distinction from the ‘authoritative resources’ that refer, for example, to the information and legislation used to regulate and control the users of the taps and the pipes. Needless to say, for understanding the (power) dynamics of technological systems, the interplay between both types of resources is the most interesting and relevant aspect, and not its materiality in itself. To understand the role of citizen-consumers in ‘sustainability transitions’ in daily life, we have to adapt and complement the primarily productionoriented schemes of ‘cleaner production technology’ as they are widely used in traditional environmental social sciences. The process of innovation and diffusion of clean technologies in industries differs from the (non-) application of more sustainable socio-technological devices within households (Shove, 2003). When talking about the role of environmental technologies in daily life, the notions of ‘end-of-pipe’ versus ‘preventive’ technologies do not ring a bell for most citizen-consumers. Therefore, we need at least two additional steps before arriving at a set of concepts that can be made operational in empirical research. First, the process of innovation and diffusion has to be specified, and second, the concept of ‘system of provision’ has to be described in more detail.

Sustainable technologies and everyday life

2.4

111

Innovation and diffusion of sustainable technologies

In order to analyze the innovation and diffusion of technologies that enable groups of citizen-consumers to organize their daily routines in a more sustainable way, we combined the Social Practices Approach with a description of the movement of socio-technical devices through the ‘chains’ of production-consumption cycles. These chains connect the ‘Providers’ (P’s) of an innovation with the end-users or Citizen-Consumers (CC's) of the innovation. Figure 11-2 shows the different modalities of interaction between Providers and Citizen-Consumers in the different phases of the production-consumption cycle. For an innovation to be implemented successfully, it has to pass through the different phases in the chain, which is possible only when Providers-logic’s match sufficiently with the (life-world) rationality’s of Citizen-Consumer at all the relevant loci within the chain (PCC). To analyze the (mis)matches between the different rationalities, we suggest employing the notion of ‘Slots’.21 mode of (end-)use C-P

mode of access

mode of

mode of

C-P distribution C-P production C-P

mode of design

Socio-technical innovations implied in the ‘greening’ of social practices Figure 11-2. Environmental innovations in production-consumption chains

2.5

Circuits as interlinked networks within chains

There does not exist one (most successful) mode of production, provision, distribution, access or use. In the alternative food circuit, for example, the distribution can be organized in a different way, as compared to the role of retailers in the ‘industrialized’ food chain. Because in all the major phases of the production-consumption cycle we are dealing with a number of modalities (Warde, 1998), the production-consumption chain has to be specified according to the different combinations or ‘routes’ that an innovation might follow. To avoid a reductionist-technocratic view on such 21

The notion of Slots as it is used in the world of ICT and air traffic, refers to the possibilities to fit in, to make a connection. At least in the Dutch language the notion of a ‘slot’ also refers to the stagnation of processes when being blocked or locked in. In both dimensions, the concept of slots represents the dynamics of ‘fitting or not fitting’, for example between system- and life-world-rationalities.

112

Chapter 11

routes through production-consumption chains, Van der Meulen (2000) introduces the concept of a ‘circuit’. A circuit can be defined as a set of interlinked networks connecting providers to consumers in a specific way. At the right side of the production-consumption-chain, the networks tend to be ‘provider dominated’ (hence the concept system of Provision is appropriate here), while at the consumer side of the chain, the networks will show a greater contribution by consumers, as illustrated in Figure 11-3.

network C

C

MODES OF USE

C

P

C

P

MODES OF ACCESS

C

P

C

P

MODES OF PROVISION

C

P

P

P

MODES OF PRODUCTION

C

P

Figure 11-3. P- and C-dominated networks in circuits of innovation and diffusion

Having developed the conceptual tools for analyzing transitions in daily life, we now move on to describe the way in which these tools can be organized for making possible empirical research on the ‘greening of everyday routines’.

3.

PUTTING THE CONCEPTS TO WORK: TWO CASE STUDIES

In analyzing the ‘greening’ of social practices by citizen-consumers, socio-technological innovations are taken as departure point, as indicated in Figure 2. Following the theory of ecological modernization (Spaargaren and Mol, 1993; Spaargaren, 2001), it is assumed that from the mid-eighties onward, socio-technological innovations have become available at least in modern industrial societies to a considerable extent. These technologies make possible ʊ both in principle and in practice ʊ a ‘switch-over' (Huber, 1982) or transition into a more sustainable modernity. Hence the question is not if there are any sustainable technologies available, but whether or not these technologies are actually adopted and applied by citizen-consumers in the context of their daily lives. This ‘adoption of innovations’ is to be

Sustainable technologies and everyday life

113

understood not in terms of the ‘conscious choices’ made by ‘early adopters’ on the basis of a well-developed ‘environmental consciousness’ or an environmentally friendly attitude, as suggested in socio-psychological models. Neither is the actual use of innovations in daily routines to be analyzed as the ‘automatic’, involuntary and mechanistic result of changes in technological systems, as is assumed in many clean-technology studies dominated by an engineering outlook. Instead, we are interested in the social conditions governing sustainability transitions within designated social practices, and in the methods used by citizen-consumers as knowledgeable and capable agents to bring about these transitions. Figure 11-4 illustrates the different phases of transition-processes in daily life.

transition Phase IV: Re-routinisation

Phase I: deroutinization Level of routinisation

Phase II:

Phase III:

development of heuristics

analysis of ‘slots’

time Figure 11-4. Transitions as de- and re-routinisation

3.1

Transitions: de- and re-routinisation

Because everyday life is routinised to a considerable extent, human agents do not consider the environmental impacts of certain consumption routines discursively or reflexively unless these routines are temporarily interrupted and set aside. And, although it is in general impossible to ‘plan’ and ‘organize’ transition-processes from the perspective of one single (governmental) agent or organization, the management of transitions always presupposes the conscious, intended input from designated groups of actors.

114

Chapter 11

In our case, the involvement of groups of citizen-consumers in the management of transitions was chosen as the central object for research, while using the social practices of dwelling and (tourist) travelling as empirical examples. With figure 11-4 in mind, the research was directed towards the following issues: 1. Making social practices operational from an environmental and a citizenconsumer’s point of view. 2. The interactive search for environmental heuristics as guiding principles for behavior. 3. The analysis of ‘circuits for innovation and diffusion’ within the productionconsumption cycle (figure 11-3), and the identification of ‘slots’ found underway. In the following, we will briefly address these issues by giving selective insights into the research undertaken in the case of ‘dwelling’ and ‘tourist mobility’, and in the preliminary results of a series of focus groups organized within these cases. 1) To make operational the social practices of ‘dwelling’ and ‘tourist travel’ in a way that enables the analysis of sustainable transitions in relation with the life-world rationality of citizen-consumers, it proved necessary to make a subdivision within these practices. The distinguished sub-practices are to be regarded as environmentally relevant by researchers and environmental policy makers at the national level, and must at the same time be recognizable for consumers as ‘normal’, daily routine-behaviors. For example, the practice of dwelling was specified as divided into the subpractices of heating, lighting, decorating, gardening, and do-it-yourself. In the Netherlands, several classifications of sets of social practices which could become the core of consumer-oriented environmental politics are being developed at the moment, and the variation in numbers and labels suggested does not seem to be unworkable or impractical from a policy perspective. 2) The search for what we call ‘environmental heuristics’. Within an environmental heuristic or heuristic device, two elements are combined: a ‘definition of the problem’ and a ‘practical guideline’ for working towards solutions to the problem. With respect to the definition of the (environmental) problem associated with the social practice under study, the primarily technical definitions offered by (governmental) experts have to be adapted to the less articulated, perhaps even fuzzy or blurred but always ‘pragmatic’ definitions employed by citizen-consumers in the context of daily routines. With respect to the general guidelines for change, the (new)

Sustainable technologies and everyday life

115

acceptance and application of more sustainable technological devices seems to be a feasible option when judged from a citizen-consumer perspective. The concept of a technological device or socio-technical innovation must be broadly defined in this respect: every product, service or method/procedure counts, provided that it enables dwellers or tourists to improve their environmental performance within existing behavioral routines. A formal distinction between technology and behavior does not seem appropriate in this respect. Socio-technological innovations always comprise both elements, in varying proportions. 3) The third research issue concerns the analysis of the ‘circuits’ along which the socio-technical devices ‘travel’ from the Provider side of the production-consumption cycle to the Citizen-Consumer side. Different types of provider networks are active in production and provision of sustainable innovations, thereby determining the trajectory an innovation takes through the cycle. Some of the relevant aspects of provider networks in this respect are: the segment of the market associated with the innovation (alternative market, niche-market, mainstream market), the involvement of ‘public’ next to ‘private’ providers and the question whether or not consumers-interests are recognized in the innovation-process. On the consumer side, differenttiation pertains to the levels of involvement (from self or co-providing citizens to captive consumer) and the levels of green choices recognized or opted for.

3.2

Focus-groups: testing and developing concepts in dialogue with citizen-consumers

The notion of citizen-consumers as knowledgeable actors has consequences for the methodology and design of the research process. One way for an active involvement of actors is the technique of focus groups. In focus group settings, consumers discussed socio-technological innovations available in the concerned (sub-)practice and the trajectory of these innovations through the production-consumption cycle. Beforehand, the researchers had distinguished three types of innovations on the basis of their position within the social practice of a dweller: products, packages and totalconcepts. Consumers reflected upon the technology-behavior composition of these innovation-types and upon the way interaction with those innovations takes place in the social practice. An example from the sub-practice of do-ityourself: Wood with an FSC-hallmark is a typical product innovation. Product innovations are largely technical and can be fitted into a social practice piecemeal without changing its intrinsic character. The employment of a package-innovation has more influence on the design of a social

Chapter 11

116

practice. The term package-innovation refers to a set of products that has to be employed in coherence or to a combination of a sustainable product and sustainable conduct. In the context of D.I.Y: The re-use of old wood in combination with the economical use of FSC-wood when new material is needed. The utilization of a total concept leads to the largest impact on organization of and behavior in a social practice. The whole D.I.Y. practice is seen through a sustainable lens and rearranged accordingly, from the purchase of materials to the handling of waste. In general, we can say that ‘products’, ‘packages’ and ‘total-concepts’ can serve as ʊ in an environmental and behavioral sense ʊ more or less far-reaching guiding concepts or ‘heuristics’ for citizen-consumers’ thinking of and striving for sustainable transitions of social practices. In relation to the innovation-types described above, (the dynamics in) provider- and consumer-involvement in the production-consumption cycle was discussed. The outcome of this is the identification by consumers of slots, or mis-fits between provider- and consumer-rationality. Slots can be located more at the provider or consumer side of the cycle, and can have a stronger technical or behavioral character. An example from the sub-practice of lighting: When the Compact Fluorescent Light-bulb (CFL) was first put in the market it produced a bright white light that Dutch dwellers ʊ attached to atmospheric lighting ʊ found very unattractive. An obvious slot or mis-match between provider- and consumer-rationality prevented a sustainable transition. The occurrence of slots can be linked to the innovation-trajectory. Another example from the practice of Lighting: several consumers mentioned mistrusting the ecological soundness of Green Electricity because there is no government hallmark for it.

4.

DISCUSSION

The Social Practices Approach is a promising concept for better understanding the interdependence between technology and behavior, and to bridge the gap between traditional technological determinism and psychological optimism. It fits very well in the increasing scientific interest in transitions and systems innovation. It forces researchers to take into account the real lives of citizen-consumers, and to define actors as agents of change. To further test, validate and generalize the Social Practices Approach more case studies are needed and quantitative methodologies have to be applied to complement the qualitative approach used so far22.

22

To fulfil these aims, the Dutch CONTRAST research program (2005 – 2009) was developed to study CONsumption TRAnsitions for suSTainability in four consumption

Sustainable technologies and everyday life

117

The approach is also a useful analytical framework and toolkit for policymakers and providers to evaluate innovations towards sustainable consumption.

REFERENCES Beck, U. (1986), Risikogesellschaft. Auf dem Weg in eine andere Moderne, Frankfurt am Main: Suhrkamp. Beck, U., A. Giddens, S. Lash (1994), Reflexive Modernization; politics, tradition and aesthetics in the modern social order. Cambridge: Polity Press. Beck, U., A. Giddens, S. Lash (1994), Reflexive Modernization; politics, tradition and aesthetics in the modern social order. Cambridge: Polity Press. Beckers, T.A.M., P. Ester en G.Spaargaren (red.), Verklaringen van duurzame consumptie; Een speurtocht naar nieuwe aanknopingspunten voor milieubeleid. Publicatiereeks Milieustrategie 1999/ 10. Den Haag: VROM. B & A. Groep (2000), Burger en Milieu. Verslag van een verkenning naar de potentie en meerwaarde van ‘burger en milieu’. Hoofdrapport. Den Haag: VROM. Callon, M., J. Law, and A. Rip, (eds) (1986), Mapping the Dynamic of Science and Technology. London: Macmillan. CEA (1999), Minder energiegebruik door een andere leefstijl? Eindrapportage Project Perspectief. Den Haag: Ministerie van VROM. Chappells, H., M. Klintman, A.L. Lindèn, E. Shove, G. Spaargaren, and Vliet, B. van (2000), Domestic consumption, utility services and the environment. Final Domus report. Wageningen: Wageningen University, p. 185. Cowan, R. Schwartz (1983), More work for mothers. New York: Basic Books, Inc. Cramer, J. en J. Schot (1990), Problemen rond innovatie en diffusie van milieutechnologie. Een onderzoeksprogrammeringsstudie verricht vanuit een technologiedynamica perspectief. Rijswijk: RMNO. CREM (2000), Domeinverkenning recreëren. Milieuanalyse recreatie en toerisme in Nederland. Amsterdam: CREM DTO visie 2040 – 1998 (1998), Technologie, sleutel tot een duurzame welvaart. Den Haag: Mark S. Storm. Fine, B. and E. Leopold (1993), The World of Consumption. London: Routledge. Giddens, A. (1984), The Constitution of Society. Cambridge: Polity Press. Giddens, A. (1991), Modernity and Self-Identity. Cambridge: Polity Press. Huber, J. (1982), Die verlorene Unschuld der Ökologie. Neue Technologien und superindustrielle Entwicklung. Frankfurt/Main: Fisher. Huber, J. (1985), Die Regenbogengesellschaft. Ökologie und Sozialpolitik, Frankfurt am Main: Fisher Verlag. Huber, J. (1991), Unternehmen Umwelt. Weichenstellungen für eine ökologische Marktwirtschaft, Frankfurt am Main: Fisher. Jelsma J. (1999), Philosophy meets design, or how the masses are missed (and revealed again) in environmental policy and ecodesign. In: Reader Summerschool on ‘Consumption, Everyday Life and Sustainability. CSS, Lancaster: Lancaster University, pp. 66-75) artikel in summerschool of winterschool reader.

domains: food, housing, daily mobility and leisure related mobility. In this program, two universities collaborate with national research institutes and several ministries. For further details please contact the first author of this chapter.

118

Chapter 11

Koppen, C.S.A. van and J.L.F. Hagelaar (1998), “Milieuzorg als strategische keuze: van bedrijfsspecifieke situatie naar milieuzorgsystematiek”, Bedrijfskunde 70: 45-51. Leroy, P. and J. van Tatenhove (2000), ‘New policy arrangements in environmental politics: the relevance of political and ecological modernisation’. In: G. Spaargaren, A.P.J. Mol and F. Buttel (eds), Environment, Sociology and Global Modernity, London etc.: Sage pp. 187-209. Mol, A.P.J., G. Spaargaren and A. Klapwijk (eds) (1991), Technologie en milieubeheer. Tussen sanering en ecologische modernisering, Den Haag: SDU. Mol, A.P.J., D.A. Sonnenfeld (2000), Ecological Modernisation Around the World: Perspectives and Critical Debates. London: Frank Cass, ISBN 07 146 50641 (also published as a Special Issue of Environmental Politics), p. 300. Otnes, P. (ed.) (1988), The Sociology of Consumption. Humanities Press Int. Pinch, T., W. Bijker and T.P. Hughes (eds) (1987), The social construction of technology. London: MIT. Rotmans J., R. Kemp et al. (2000), Transities en Transitiemanagement; onderzoekrapportage ten behoeve van NMP-4. ICIS/ Merit. Maastricht. Schot, J.W. (1992), Constructive technology assessment and technology dynamics: The case of clean technologies. Science, Technology and Human Values, 17, 1, pp. 36-56. Schot, J.W. (red.) (1994), Onderzoekprogramma Geschiedenis van de Techniek in Nederland 1890 – 1970. Eindhoven: Stichting Historie der Techniek. Shove, E. (2003), Comfort, cleanliness and convenience: the social organization of normality. Oxford: Berg. Schuttelaar and Partners (2000), Domeinverkenning Voeden. Ingrediënten voor een gezond milieu. Den Haag. Spaargaren, G. (1997), The ecological modernisation of production and consumption: essays in environmental sociology, dissertation Wageningen University. Spaargaren, G. (2000), Ecological Modernization Theory and Domestic Consumption. Journal of Environmental Policy and Planning 2: 4, p. 323-335 Spaargaren, G., A.P.J. Mol and F.H. Buttel (2000), Environment and Global Modernity. London: Sage, p. 257 Van Vliet, B., Wüstenhagen, R. and H. Chappells (2000), New Provider-Consumer Relations in Electricity Provision. Green Electricity Schemes in the UK, The Netherlands, Switzerland and Germany. Paper for the Business Strategy and the Environment Conference. September 18-19 2000, Leeds. Vringer, K. and K. Blok (1993), The direct and indirect energy requirement of households in the Netherlands. Utrecht: Vakgroep Natuurwetenschap en Samenleving. Vringer K., T. Aalbers et al. (2000), Nederlandse consumptie en energieverbruik in 2030; een verkenning op basis van twee lange termijn scenario’s. RIVM-rapport 408129015. Bilthoven: RIVM. VROM (1989), Nationaal Milieubeleidsplan (NMP). Tweede Kamer, 1988-1989, 21137, nrs. 1-2. Den Haag: SDU. (NEPP). VROM (1993), Nationaal Milieubeleidsplan 2; Milieu als Maatstaf. Tweede Kamer, 19931994, 23560, nrs. 1-2. Den Haag: SDU. (NEPP2). VROM (2000), De warme golfstroom: heroriëntatie op communicatie over milieu. Den Haag: Centrale Directie Communicatie DG.

Chapter 12 RESIDENTIAL BEHAVIOR IN SUSTAINABLE HOUSES Erica Derijcke and Jan Uitzinger

1.

INTRODUCTION

In the Netherlands, many municipalities require that houses are constructed according to national guidelines for sustainable building (“Nationaal Pakket Duurzaam Bouwen”). However, a sustainable house does not necessarily perform well, environmentally. The environmental performance also depends on the behavior of its residents. If they use the technology in a way not intended by the designer, the environmental benefits designed for in the building stage may be partially lost during use. This interaction between systems and behavior was investigated in two empirical case studies. The central question was: “How do sustainable building systems work out in practice, and how can the use of the technical systems by residents be improved?” One case study23 was conducted to determine residential behavior towards water, material and energy-saving systems in houses (Derijcke et al., 2001). The other study24 focused on a specific sustainable energy system: photovoltaic panels applied on residential roofs (Visser et al., 2001). In a social monitoring study the ideas and experiences of residents were investigated, and in a technical monitoring study the yield of the system was evaluated. Furthermore, it was investigated if demand-side management (DSM) occurred due to the solar panels. 23

Under authority of environmental education centre De Kleine Aarde.

24

Under authority of NUON and in co-operation with ECN.

119 P.-P. Verbeek and A. Slob (eds.), User Behavior and Technology Development: Shaping Sustainable Relations Between Consumers and Technologies, 119-126. © 2006 Springer.

Chapter 12

120

In this paper, only the results relevant for the subject ‘interaction between the technological system and behavior’ are described. For further information, please read the full reports (in Dutch).

2.

CASE STUDY 1

The main objective of this case study was to explore how residents used water, material and energy-saving techniques that had been implemented to enhance the sustainable use of the house.

2.1

Methodology

Six sustainable building projects were selected from a new housing estate in the city of Breda. Each project consisted of at least 45 houses in which the residents, at that time, had lived for at least one year. Face-to-face interviews were carried out with at least one resident per project to discuss the resident’s behavior and to inspect the sustainable measures applied. As a result, it was determined how the resident's behavior disagreed with the intended behavior. Based on this information, a survey was designed and run among 400 households. As the applied techniques differed among the building projects, IVAM made six different questionnaires. Respondents could win a small gift as a reward for participating in the survey. The response rate was around 38%, except for two projects comprising rental houses, where the response rate was relatively low.

2.2

Results

The most important factors for residents in the decision to buy a house were price, appearance, size and the fact that it is a new estate. The fact that the house was sustainable was much less important. The few respondents that mentioned sustainability as a factor were the ones that lived in a housing project that allowed for resident participation in the design of the houses. Apparently they were more involved. The toilets were equipped with a flush stop to save water. A reasonable share of residents did not know that their toilet had a flush stop, and therefore did not use it. Almost 25% of the other residents sometimes or never used the flush stop. In one of the projects, the three levels of the mechanical ventilation were numbered 0, 1 and 2. Over 25% of the residents believed that the ventilation was turned off at level 0, which was not the case. As a consequence those

Residential behavior in sustainable houses

121

residents tended to use the ventilation at a higher level than necessary (Ȥ2 = 2,85, p < 0,10), thereby wasting energy. In one of the projects, the windows’ inner posts were only primed in the building stage to reduce the use of materials. So the residents could choose the color they liked for the upper layer. 27% of the residents painted the inner posts with traditional alkyd-based paint (see table 12-1). In three other projects, the window inner posts were primed and painted in the building stage. The second layer consisted of paint that is relatively environmentally sound (water based, or high solid paint). 10% of the residents added a third layer of traditional paint. So, compared to the projects where the inner posts were primed and painted, more residents used traditional paint when the inner posts were primed only. Table 12-1. Type of paint on window inner posts Traditional alkyd based

Water based or high solid

paint (%)

paint (%)

Only primed (N = 22)

27

36

Primed and painted (N = 41)

10

5

2.3

Conclusions case 1

Firstly, our findings show that the environment was not a very important issue in buying or renting a house. In order to encourage residents to behave in environmentally-friendly ways, they should be informed of the proper use of the house. It was not investigated whether the environment is a sufficiently strong motivation for residents to change their behavior. Making residents paint by themselves is not necessarily positive for the environment. Consumers should be aware of environmentally ‘sound’ paints. Proper information in paint-selling shops could be an important factor to achieve this. However, an easier solution would be to improve the design of some systems in such a way that environmentally beneficial conduct is automatically induced. The flushing device, for instance, could be designed such that flushing a small amount of water is its standard. In the example of mechanical ventilation, the number “0” led to the misinterpretation of the appliance being turned off. Therefore, it would be better to number the ventilation levels 1, 2 and 3.

122

3.

Chapter 12

CASE STUDY 2

The second study3 focused on a specific sustainable energy system: photovoltaic panels applied on residential roofs. In a social monitoring study the ideas and experiences of residents were investigated, and in a technical monitoring study the yield of the system was evaluated. Furthermore, it was investigated if demand-side management (DSM) occurred due to the solar panels.

3.1

Methodology

The study was based on social and technical monitoring. Two surveys were conducted among residents of 80 houses in the city of Apeldoorn: one survey just after the house was bought, and the other survey after three years. The response rate was 69% in the survey just after buying the house, and 55% after three years. In four houses, the electric load was measured every ten minutes to see if demand-side management occurred due to the solar panels. See figure 12-1 for an example of a ten-minute electric load pattern of a household. The four households were chosen in such a way that they could show an optimal effect: at least one person was at home during daytime, and they had at least three appliances (washing machine, dryer and dish washer) with which it is possible to choose the time of use. Electric load patterns of six comparable households without solar panels were used as a control group.

Figure 12-1. A ten-minute household electric load pattern

Residential behavior in sustainable houses

123

The following appliances can be seen in the pattern: Table 12-2. Load patterns of appliances Appliance Load (Watt) Standby 80 Fridge 120 Lighting 300 Washing machine 2000 Dryer 2500

3.2

Pattern continuous spikes increased load in the evening peak load due to heater broad peak load due to heater

Results

Replicating the findings of the first case study, the environment did not seem to be a decisive factor when buying a house. The solar panels played a positive role when people subscribed for the houses, and also later, when they bought the house. However, the solar panels were of minor importance compared to aspects such as size and price of the house. Residents were interested in getting more information on the applied techniques. Just after buying the house, 75% of respondents were interested in receiving more information. Three years after buying, people still wanted information, especially on financial costs and benefits, technical aspects, and the electrical yield of the solar panels. The surplus electricity, i.e. electricity generated by the PV system but not used directly, is fed to the electricity grid. The households had three electricity meters: one for the yield of the PV system, one for the surplus electricity and one for the usual electricity consumption. From an environmental point of view, it is best to consume the generated electricity directly (without increased consumption, of course). In that case, there is no transport loss and demand-side management makes it possible to avoid extra capacity. In this case study, the payback rate was lower than the kilowatt-hour price. In other words, using one’s own generated electricity was financially attractive. As a result, changing the time of use of appliances in favor of periods of sunshine was financially beneficial for residents. Not every electric appliance is suited for load management. Electric cooking, for example, is difficult to control, because then one needs to change one’s dinnertime. Lighting is difficult to control as well. The fridge and freezer can hardly be controlled, and work more or less autonomously. For the purpose of using one’s own self-generated electricity, the washing machine, dryer and dishwasher are best suited. Of course, residents have to be at home during the daytime to be able to change the time of use of these appliances. A technical solution would be to use a time switch.

Chapter 12

124

Although it was not specifically asked for, answers on some statements showed that people change the time of use for environmental reasons. This factor appeared to have greater weight than ‘financial benefits’. Besides, residents did not know how much money they could save on a yearly basis. Compared to the control group, the four households, whose load patterns were measured, shifted 15% of their use of the washing machine, dryer and dishwasher to sunshine hours. A tentative conclusion is therefore that solar panels can be included in the Demand-Side Management approach. It appeared that residents use the electricity meters to see if the PV system generated enough electricity. Furthermore, it appeared that the electricity consumption of appliances such as fridges and freezers is covered by the PV system as long as the sun is shining. This means that in households where during the daytime nobody is present, electricity of PV systems is used automatically.

3.3

Conclusions case 2

The solar panels were not an important attractive point in buying a house. However, residents did shift their times of use of the washing machine, dryer and dishwasher to hours of sunshine to save energy. Households shifted up to 15% of their use of appliances (such as of washing machine, dryer and dishwasher) from the evening to hours of sunshine. Residents saved energy to contribute to the solution of environmental problems rather than for financial reasons. They use their electricity meters to see if the PV system generates enough electricity.

4.

GENERAL DISCUSSION AND CONCLUSIONS

The limited number of respondents hampered the interpretation of the results of both case studies. In the first case study, the new estate project consisted of many different types of houses. Therefore, some measures could only be investigated in a small number of houses. Together with an average response rate of 38%, it was difficult to obtain statistically significant results. Some of the research questions could not be answered due to the small number of respondents. In the second case study, the demand-side management was only measured in four households. In other studies, it has also been found that environmental issues do not play an important role in buying a house (e.g. Silvester & de Vries, 1999).

Residential behavior in sustainable houses

125

Our first case study indicates that environmental issues were more important for residents that participated in the building process. Likewise, Silvester and de Vries (1999) found that for 70% of the residents of sustainable projects in which they could participate in the design process, the applied sustainable techniques were of importance when buying the house. The numbering of the settings of the mechanical ventilation can be interpreted in terms of an inscription that is misunderstood by its user. The designer did not target the description for these end users. Jelsma and Popkema (1998) describe a design methodology that avoids these faulty designs. In order to avoid waste of energy or materials, in newly built houses some things are left to the residents to choose or to apply. For instance, it would be a waste of materials if the resident replaces a standard built-in kitchen for a luxurious one. Silvester and de Vries (1999) found that a reasonable share of people think that due to a free choice of kitchen and bathroom, the interior will have a longer life span. However, if residents can freely choose, they would not end up with a sustainable kitchen and bathroom. This was also found regarding painting the inner posts. In one project, the painting of the inner posts was left to the residents. The residents of this project participated in the design of their sustainable house. Nevertheless, they bought traditionally alkyd paint. Silvester and de Vries (1999) recommend that the kitchen, bathroom and do-it-yourself-branch offer more sustainable products. Moreover, consumers should be well informed about the possibilities to purchase sustainable products. The results of these two case studies show that some sustainable building measures do not work in practice, as intended. Therefore, more insights should be gained about current practices in residential behavior. This could direct how to improve the design of sustainable products and services. Sustainable systems in residences should be designed in such a way that environmentally-preferred behavior is also the most logical and easiest accomplished. Otherwise, the system should be designed in such a way that it is indifferent to residential behavior. Furthermore, understanding residents’ behavior will yield insights about the kinds of information needed to use water, material and energy systems more efficiently. Residents should be aware of the products they can buy and on how to use them. Furthermore, information or feedback can also help residents to adjust their behavior, as was shown in the PV project.

126

Chapter 12

REFERENCES Derijcke, E., J. Uitzinger and B. van Mierlo (2001) Bewonersgedrag en duurzaam bouwen maatregelen. Amsterdam: IVAM Environmental Research. Jelsma, J. and M. Popkema (1998) Gedragsbeïnvloeding door technologie. Publicatiereeks milieustrategie nr. 1998/1. Den Haag: Ministerie van VROM. Klopstra, A. and B. van Mierlo (2000) Wonen in een woning met zonnepanelen in Apeldoorn. Onderzoek naar de ervaringen van bewoners van het Woudhuis, Amsterdam: IVAM Environmental Research. Silvester, S. and G. de Vries (1999) Woonsatisfactie, bewonersgedrag en bewonerswensen bij Voorbeeldprojecten Duurzaam Bouwen. Technische Universiteit Delft, Design for Sustainability Program and V & L Consultants. Visser, E. (NUON), N.J.C.M. van den Borg (ECN), A.L.J. Klopstra (IVAM) and J. Uitzinger (IVAM) (concept) Beheer en Monitoring van 100 netgekoppelde PV-installaties op nieuwbouwwoningen te Apeldoorn.

Chapter 13 MAKING ENERGY FEEDBACK WORK: Goal-Setting and the Roles of Attention and Minimal Justification L.T. McCalley and Cees J.H. Midden

1.

INTRODUCTION

Several forms of intervention to promote household energy conservation have been tested over the years, but research has primarily focused on energy consumption feedback (Shippee, 1980). Although a number of earlier researchers reported success in reducing home energy with feedback (e.g. Hayes & Cone, 1981; McClelland & Cook, 1979; Palmer, Lloyd & Lloyd, 1977), others have reported failure (e.g. Seaver & Patterson, 1976). Even when results were positive, they often could not be adequately explained (Seligman & Darley, 1977). Some energy researchers attributed success to combining the motivational effects of goal-setting with feedback (e.g. Becker, 1978; van Houwelingen & van Raaij, 1989), but results remained variable. In 1981, Shippee noted a lack of research to identify those processes that mediate the effectiveness of consumption feedback, but, to date, they have remained largely unidentified. This is most likely due to the fact that energy feedback studies have concentrated on attempts to change particular behaviors or actions ʊ such as turning off lights (Winett, 1978), or air conditioning units when the outside temperature is cool (Becker & Seligman, 1978) ʊ without investigating the underlying relationship of feedback and action. In other words, experimental designs investigating energy feedback have been, on the whole, hit-or-miss, while feedback intervention study results in general have also remained highly variable (Kluger & DeNisi, 1996).

127 P.-P. Verbeek and A. Slob (eds.), User Behavior and Technology Development: Shaping Sustainable Relations Between Consumers and Technologies, 127-137. © 2006 Springer.

128

Chapter 13

In 1998, a study was designed to reexamine the effects of energy feedback in an applied setting (McCalley, 1999) within a cognitive framework based on attention (Kluger & DeNisi, 1996). In this manner, behavior response to feedback could be methodically explored in relationship to various potential moderating factors. The washing machine was chosen as the representative application domain due to high household penetration and, thus, high consumer familiarity. Laboratory experiments were designed using a computer simulation based on an existing washing machine (Miele Novotronic W941 Super) with a microprocessor-run control panel. The use of a simulated control panel allowed feedback to be integrated directly into the appliance interface, allowing for optimal feedback presentation in terms of specificity (e.g. specific to a single use of a single appliance and specific amounts of energy and cost) and speed. The first series of experiments tested various combinations and types of feedback, including monetary and energy (kWh) information, without further manipulations in order to establish an estimate of baselines for conservation behavior response. The results in terms of generating conservation behavior were minimal, as expected. The next series of experiments explored feedback in relation to various types of specific goals to save energy (McCalley, 2000) with the rationale, based on Feedback Intervention Theory (FIT) (Kluger & DeNisi, 1996), that setting a conservation goal would focus attention on feedback related to the washing task, making it more salient. The primary experiment compared the effect of energy (kWh) feedback on a group of subjects who were asked to set a conservation goal for themselves of between 0 and 20 percent and a control group (McCalley & Midden, 2002). The resulting response to the energy use feedback was found to be highly significant for the group receiving the goal-setting treatment. Results thus confirmed that substantial amounts of energy could be saved if the subject was asked to make a prior commitment by setting a specific energy-saving goal. The outcome was thus in accordance with FIT (Kluger & DeNisi, 1996), which attributes successful conservation behavior to goalsetting as a means of focusing attention on the washing task, thereby making the energy feedback more salient, as mentioned previously. However, the experiment was not a direct test of the theory. A further experiment was designed to specifically test the assumptions of FIT by manipulating the levels of attention focus. This design also served a second purpose of allowing a comparison of FIT with another framework that was previously used to develop means to encourage household energy conservation. The comparison framework is based on a social-psychological formulation of the minimal justification principle (see Katzev & Johnson, 1983), called the

Making energy feedback work

129

“foot-in-the-door” technique, that depends, first, on an attitude change precipitated by a small request that is then followed by the “target” or desired action request.

2.

FEEDBACK INTERVENTION THEORY

FIT (Kluger & DeNisi, 1996) is a framework developed specifically for the domain of task performance, and suggests that all performance is dependent on the interplay of several variables that direct attention vertically up or down a hierarchical arrangement of global (self-oriented, highest level) to task-specific and task-learning (lowest level) thoughts. As outlined by Kluger and DeNisi (1996), FIT has five basic arguments. Firstly, behavior is regulated by comparisons of feedback to goals or standards. Secondly, goals or standards are organized hierarchically. Thirdly, attention is limited, and therefore only feedback-standard gaps that receive attention actively participate in behavior regulation. Fourthly, attention is normally directed to a moderate level of the goal hierarchy; and fifthly, feedback interventions change the locus of attention, and therefore affect behavior. Attention is thus the underlying drive in task performance, and must therefore be directed to a specific task-oriented goal. If commitment is to a higher-level goal, attention is directed upwards to a more self-oriented level (“meta-task level”), and the desired response to task-related feedback will be attenuated. Based upon FIT, results of the goal-setting experiment described above can be interpreted as evidence that by encouraging an individual to choose a conservation goal the task-related attention focus (e.g. setting the washing temperature lower saved energy and temperature setting is part of the washing task) was enhanced, resulting in energy savings. (All prominent theories of motivation feature goals as the prime determinants of motivation; Klein, Wesson, Hollenback & Alge, 1999; Phillips, Hollenback & Ilgen, 1996). However, results might also be attributed to the foot-in-the-door technique in the manner described in the following section.

3.

MINIMAL JUSTIFICATION

The foot-in-the-door, or compliance, technique is based on the principle of minimal justification, and is one form of a minimal justification procedure. The technique is to ask individuals to comply with a small request and to later ask them to comply with a larger, more difficult (and

130

Chapter 13

generally related) request (Freedman & Fraser, 1966). This principle is derived from the observation that moderate, rather than highly attractive justifications, are often more effective in promoting various behaviors. Katzev and Johnson (1983) designed a study to test whether this technique could enhance residential energy savings. Based on previous minimal justification research in the domain of social psychology, they made the assumption that individuals would be more likely to respond to a request to conserve electricity if they had first agreed to perform a much smaller energy-related task. They then compared the electricity use of four groups of households where residents were either simply monitored as to usage (control group), given a short questionnaire about energy use, asked to reduce their electricity use by 10%, or given the questionnaire and asked to curtail electricity use by 10% (foot-in-the-door treatment). No feedback beyond the normal monthly bill was given to the participants. Katzev and Johnson (1983) found that participants in the foot-in-the-door group qualified as “conservers” during the 12-week follow-up phase. Although the actual savings of this group was not significantly different from the other groups’, their behavior was consistent with a greater desire to conserve. In brief, subjects given the foot-in-the-door treatment did not save significantly more energy than the other groups, but did express a desire to be conservers. If we view the earlier goal-setting experiment in the same terms as that of Katzev and Johnson (1983), it could be supposed that the request to answer a short personal and household information questionnaire at the beginning of the experiment served as the first, or compliance request, and the request to set a goal served as the second, or target request. If lowered energy use were indeed the result of a compliance manipulation, then the resulting decrease in energy use could have been due to a change of attitude towards energy conservation brought about by a change in self-perception. Multiple requests have been found to increase the effects of compliance (Dillard, 1991; Gorassini & Olson, 1995), and it was therefore decided to test the source of the earlier goal-setting experiment’s results by adding another request to the design. Increasing sequential requests (otherwise termed the ‘multiple request strategy’) reinforces the effect by changing the self-perception of the individual to be positive in the direction of the request. If the result is an increase in energy savings, then it would support the view that conservation behavior is related to a change in attitude or self-perception. On the other hand, FIT predicts that any manipulation that directs attention to the self will attenuate the effects of feedback, thus predicting that the addition of a compliance request will result in a situation where energy use does not change despite the goal-setting treatment. In other words, the addition of another mild request should increase energy savings in comparison to the

Making energy feedback work

131

goal-setting-only condition if conservation is a direct result of attitude change, but should decrease savings if conservation is only indirectly related to attitude and depends primarily on having a goal that matches an alreadyexisting attitude which helps focus attention on the task. The earlier experiment was thus replicated with some minor changes in order to ascertain whether response to immediate energy feedback results from focusing attention on the task ʊ at hand by providing a goal, or if it results from an underlying change in an environmentally-related attitude. In this manner, it should be possible to understand where, on the goal hierarchy, attention should be directed to best encourage conservation behavior in response to feedback. According to Katzev and Johnson (1983), directing attention to underlying attitudes will give a stable and long-lasting effect of response. However, according to Austin and Vancouver (1996), higher level goals operate on slower time scales, and are therefore not subject to consideration as often, so are thus less likely to receive attention processes. Furthermore, Austin and Vancouver (1996) claim that it is the midlevel goals that are more commonly accessed. It is therefore possible that, although attitude might be changed by a foot-in-the-door procedure, the setting of immediate task-relevant goals, which are not contrary to an already existing attitude, are the source of attention focus and energy-saving actions. Based on this assumption, it was predicted that the foot-in-the-door treatment would have an effect on attitude that would be indicated by an increase in the amount of the goal set by subjects receiving the foot-in-thedoor treatment. In turn, this would cause subjects to attend to the level of the self in the goal-setting condition, and to not save as much energy as subjects who set an energy conservation goal but who did not receive the foot-in-thedoor treatment.

4.

TESTING MINIMAL JUSTIFICATION VS. FEEDBACK INTERVENTION

An experiment was carried out to find out the dominance of the two described processes, namely, minimal justification vs. feedback intervention. This study and its findings will be described in what follows.

4.1

Method

The subjects were 120 residents of the Eindhoven area who were experienced at using a washing machine (e.g. using the washing machine one or more times per week).

132

Chapter 13

The design was a 2 (goal/no goal) x 2 (foot-in-the-door/no foot-in-thedoor). Two dependent variables were measured; total amount of energy use per wash, and the level of the goal set. The effect of a minimal justification, foot-in-the-door, treatment on goal-setting was tested using four feedback groups: two receiving a self-set goal-setting treatment and the other two receiving no goal-setting treatment. Of the two goal-setting groups, half received a foot-in-the-door treatment while the other half had not, and the same was done with the two no-goal groups. Subjects were first shown a Miele Novotronic Super washing machine, and introduced to the control panel so that they were able to associate the real washing machine control panel with the simulated version used for the experiment. They were then seated behind a computer in order to perform the experiment. Groups receiving the foot-in-the-door treatment did so before beginning. The treatment was simply a polite request asking the subject if they would be willing to answer a short questionnaire about household energy use. They were told that the questionnaire would consist of no more than six questions, and it would be mailed to them, with a stamped return envelope, within the next two weeks. Compliance by the subject was considered to be the completion of the treatment. All subjects complied. For subjects not receiving the foot-in-the-door treatment, the procedure was identical, except that they were not first asked if they would be willing to fill out the small questionnaire at a later date, and proceeded directly to the introduction to the computer experiment phase. The design of the simulated panel was a copy of the Miele Novotronic machine control panel, with the exception of the elimination of the timer display and the addition of an energy meter (in the feedback conditions), similar to the temperature meter on the real machine. Subjects proceeded to the next section of the experiment where they did ten washing trials. A trial consisted of a display of the control panel with a description of a load of wash below. Subjects chose the program and options from the control panel in order to do the wash as instructed, just as they would at home given the same load. The washes were chosen to represent the same balance of likely washing program choices as the remaining 20 trials. This was later confirmed by analyzing the data of the control groups. All four groups received the kWh average of these six washes. The two control groups (no feedback-no goal and feedback-no goal) proceeded to the final 20 washes after receiving a simple message that people could save at least 20 percent of the energy used for washing by lowering washing temperatures. The goalsetting group was asked to choose a goal from a selection of levels; 0, 5, 10, 15, and 20 percent. After the goal-setting manipulation, subjects proceeded

Making energy feedback work

133

to the 20 washing trials. After the trials, all subjects were told their average energy use per trial.

4.2

Results

The main effect of goal-setting was found to be highly significant. There was no significant main effect of the foot-in-the-door treatment. There was, however, a significant interaction between minimal justification and goalsetting. Subjects in the goal-setting condition who had first received the footin-the-door treatment did not save as much energy as the goal-setting group receiving no minimal justification treatment. The response comparison between the no-goal groups was just the opposite.

Mean Energy Use (kWh)

18.0 15.0 12.0 9.0 Foot-in-the-door

6.0 3.0

No

0.0 No

Yes Yes Goal Condition

Figure 13-1. Energy saved as a function of the foot-in-the- door treatment and goal-setting

25

To enhance accessibility for a broad readership, this results section does not contain numerical statistical information. Readers interested in the statistical details are asked to contact the authors.

Chapter 13

134

Finally, minimal justification appeared to influence the level of the goal set (0, 5, 10, 15, or 20 percent energy saving goal) within the goal-setting condition (Table 13-1). Table 13-1. Relationship between foot-in-the-door and goal-setting Goal Level Set Foot-in-the-door No 5% 1 10% 16 15% 7 20% 7

5.

Yes 1 6 5 16

CONCLUSIONS

In contrast to the Katzev and Johnson (1983) study, a significant effect of goal-setting alone was found, replicating the results of the earlier McCalley and Midden (2001) study. Product-integrated energy-use feedback, being specific and immediate, was likely more salient for the subjects in contrast to the monthly utility bill used by Katzev and Johnson (1983). This outcome suggests that frequent and specific feedback is a successful means of encouraging energy conservation when users are first encouraged to set a goal. The significant interaction between minimal justification and goal-setting indicated that subjects who were asked to set an energy conservation goal and who also received the foot-in-the-door treatment did not save as much energy as those asked to set an energy conservation goal who did not receive the foot-in-the-door treatment. It is thus apparent that the minimal justification manipulation inhibited response to the goal. In terms of FIT, the outcome can be interpreted as evidence that the minimal justification treatment directed attention to a meta-task level (self) and away from the task performance level, thereby attenuating the effects of the feedback. Additional proof for this interpretation lies in the fact that S’s set higher goals when they received the foot-in-the-door treatment. This is in accordance with a self-perception interpretation, which means that S’s were thinking at the meta-level of attitudes and not directing attention to the specific task. In sum, feedback serves as the information a user needs to assess where they stand in relation to the desired goal, and thus energy conservation results only when the user forms (or has formed) a specific goal to save

Making energy feedback work

135

energy which is matched to the energy feedback. Goal-setting prioritizes the goal to conserve energy above other goals in the case of the washing machine. It is the setting of a specific goal to save energy that directs and maintains attention to a particular energy-related task, such as washing clothes, and enables the feedback to be processed by the user. Attitude towards energy conservation can be influenced, resulting in a higher level of anticipated performance by the user, but attitude change is not necessarily translated directly into conservation behavior.26 The best explanation for the results is given by FIT, which predicts that drawing attention to a level of the goal hierarchy that is higher than the task performance level will cause feedback effects to be attenuated.

REFERENCES Austin, J. T., & Vancouver, J. B. (1996). Goal constructs in psychology: Structure, process, and content. Psychological Bulletin, 120, (3), 338-375. Balzer, W. K., Doherty, M. E., & O’Connor, R., Jr. (1989). Effects of cognitive feedback on performance. Psychological Bulletin, 106, 410-433. Becker, L. J. (1978). Joint effect of feedback and goal setting on performance: A field study of residential energy conservation. Journal of Applied Psychology, 63, 4, 428-433. Becker, L. J., & Seligman, C. (1978). Reducing air conditioning waste by signaling it is cool outside. Personality and Social Psychology Bulletin, 4, 412-415. Beurden, K. van (1982). “Indicator huishoudelijk gasverbruik” (Indicator for In-Home Gas Use), Gas, 102, February, 58-64. Brandon, G., & Lewis, A. (1999). Reducing household energy consumption. Journal of Environmental Psychology, 19, 75-85. Cameron, L. D., Brown, P. M., & Chapman, J. G. (1998). Social value orientations and decisions to take proenvironmental action. Journal of Applied Social Psychology, 28, 8, 675-697. Carver, C. S., & Scheier, M. F. (1981). Attention and self regulation: A control theory to human behavior. New York: Springer-Verlag. Costanzo, M., Archer, D., Aronson, E., & Pettigrew, T. (1986). Energy conservation Behavior: The difficult path from information to Action. American Psychologist, 41, 5, 521-528. Dillard, J. P. (1991). The current status of research on sequential-request compliance techniques. Personality and Social Psychology Bulletin, 17, 283-288.

26

The relationship of attitudes and behavior has been a topic of much research and is thus beyond the scope of this chapter. Generally, behavior has not been found to necessarily reflect attitude (e.g. Cone & Hayes, 1980; Shippee, 1980; Heberlein, 1976). In the case of the present study, attitude may indeed support higher level goals which are, in turn, necessary to support sub-goals related to conservation behavior. However, this is yet to be investigated. For a more recent study and discussion of the broader topic of attitudes and behavior, see Cameron, Brown and Chapman, 1998.

136

Chapter 13

Erez, M. (1977). Feedback: A necessary condition for the goal setting-performance relationship. Journal of Applied Psychology, 62, 624-627. Freedman, J., & Fraser, S. (1966). Compliance without pressure The foot-in-the-door technique. Journal of Personality and Social Psychology, 4, 195-202. Gorassini, D. R., & Olson, J. M. (1995). Does self-perception change explain the foot-in-thedoor effect? Journal of Personality and Social Psychology, 69, 91-105. Hayes, S. C., & Cone, J. D. (1981). Reduction of residential consumption of electricity through simple monthly feedback. Journal of Applied Behavior Analysis, 14, 81-88. Houwelingen, J. van, & Raaij, W. F. van (1989). The effect of goal setting and daily electronic feedback on in-home energy use. Journal of Consumer Research, 16, 98-105. Hutton, B., Mauser, G. A., Filiatrault, P., & Ahtola, O. T. (1986). Effects of cost-related feedback on consumer knowledge and consumption behavior: A field experimental approach. Journal of Consumer Research, 13, 327-336. Ilgen, D. R., Fisher, C. D., & Taylor, M. S. (1979). Consequences of individual feedback on behavior in organization. Journal of Applied Psychology, 64, 349-371. Katzev, R. D., & Johnson, T. R. (1983). A social-psychological analysis of residential electricity consumption: The impact of minimal justification techniques. Journal of Economic Psychology, 3, 267-284. Klein, H. J. (1991). Control theory and understanding motivated behavior: A different conclusion. Motivation and Emotion, 15, 29-44. Klein, H. J., Wesson, M. J., Hollenback, J. R., & Alge, B. J. (1999). Goal commitment and the goal-setting process: Conceptual clarification and empirical synthesis. Journal of Applied Psychology, 84, 885-896. Kluger, A. N., & DeNisi, A. (1996). The effects of feedback interventions on performance: A historical review, a meta-analysis, and a preliminary feedback intervention theory. Psychological Bulletin, 119, 2, 254-284. Locke, E. A. (1991). Goal theory vs. Control Theory: Contrasting approaches to understanding work motivation. Motivation and Emotion, 15, 1, 9-44. Locke, E. A., Shaw, K. N., Saari, L. M., & Latham, G. P. (1981). Goal setting and task performance: 1989-1980. Psychological Bulletin, 90, 125-152. Locke, E. A., & Latham, G. P. (1990). A theory of goal setting and task performance. Englewood Cliffs, NJ: Prentice Hall. McCalley, L. T. (1999). Eco-feedback I (Novem b.v.). Netherlands: Eindhoven University of Technology, Department of Human-Technology Interaction. McCalley, L. T. (2000). Product-integrated Eco-feedback (Novem b.v.). Netherlands: Eindhoven University of Technology, Department of Human-Technology Interaction. McCalley, L. T., & Midden, C. J. H. (2002). Energy conservation through product-integrated feedback: The roles of goal-setting and social orientation. Journal of Economic Psychology, 23, 5, 589-604. McClelland, L., & Cook, S. W. (1979). Energy conservation effects of continuous in-home feedback in all-electric homes. Journal of Environmental Systems, 9, 2, 169-173. Messick, D. M., & McClintock, C. G. (1968). Motivational basis of choice in experimental games. Journal of Experimental Social Psychology, 4, 1-25. Norman, D. A. (1986). Cognitive Engineering. Chapter 3 in: D. A. Norman & S. W. Draper (Eds.), User centered system design: New perspectives on human-computer interaction. Hillsdale: Erlbaum. Pallack, M. S., Cook, D. A., & Sullivan, J. J. (1980). Commitment and energy conservation. In L. Bickman (Ed.), Applied Social Psychology Annual (Vol. 1, pp. 235-253). Beverly Hills, CA: Sage.

Making energy feedback work

137

Palmer, M. H., Lloyd, M. E., & Lloyd, K. E. (1977). An experimental analysis of electricity conservation procedures. Journal of Applied Behavior Analysis, 10, 665-671. Phillips, J. M., Hollenback, J. R., & Ilgen, D. R. (1996). Prevalence and prediction of positive discrepancy creation: Examining a discrepancy between two self-regulation theories. Journal of Applied Psychology, 81, 498-511. Seligman, C., & Darley, J. M. (1977). Feedback as a means of decreasing residential energy consumption. Journal of Applied Psychology, 62, 4, 363-368. Seligman, C., Becker, L. S., & Darley, J. M., (1981). Encouraging residential energy conservation through feedback. Advances in Environmental Psychology, 3, 93-113. Shippee, G. (1980). Energy consumption and conservation psychology: A review and conceptual analysis. Environmental Management, 4, 297-314. Winett, R. A., (1978). Prompting turning-out lights in unoccupied rooms. Journal of Environmental Systems, 1, 237-241.

Chapter 14 TECHNOLOGICAL INNOVATIONS AND THE PROMOTION OF ENERGY CONSERVATION: The Case of Goal-Setting and Feedback Trijntje Völlink and Ree M. Meertens

1.

INTRODUCTION

One of the most complex, and difficult to solve, environmental problems is the increase in carbon dioxide (CO2) emissions that add to the greenhouse effect. Increasing the production of sustainable energy and taking measures to promote energy conservation are important options in contributing to the reduction of CO2 emissions. In the Netherlands, households represent an important target population. They use about 24% of the total electricity consumption and 16% of natural gas consumption (Energy Center Netherlands [ECN], 1999). Besides structural methods like insulation and more energy-efficient apparatus use, changing behavioral patterns (like lowering the thermostat setting to 15°C one hour before bedtime) is thought to be a useful tool for further reduction (Department of Economic Affairs, 1999). Previous studies revealed that frequent (weekly) feedback in combination with a realistic but challenging goal was an effective strategy for reducing total energy consumption of households by 12.3% - 25% (Becker, 1978; Van Houwelingen & Van Raaij, 1989; Winett et al., 1982). When feedback was provided less frequently (once every two weeks or monthly), the reduction decreased to 3%, which could also be achieved by self-monitoring. According to Locke and Latham (1990), challenging and specific goals lead to improved performance because (1) they are associated with higher selfefficacy, (2) they are less ambiguous about what constitutes high or good 139 P.-P. Verbeek and A. Slob (eds.), User Behavior and Technology Development: Shaping Sustainable Relations Between Consumers and Technologies, 139-148. © 2006 Springer.

140

Chapter 14

performance, (3) they lead individuals to persist longer, (4) they are more effective in directing attention and action, and (5) they motivate people to find suitable task strategies, and to utilize strategies that they have been taught. Although weekly written feedback may be too expensive for large-scale application, expectations for technical innovations like electronic feedback devices are high (Van Houwelingen & Van Raaij, 1989). In this paper, the results of two studies will be described in which the interaction between technological innovations and behavioral change was evaluated. Both studies examined the effect of feedback and goal-setting on energy (study 1 and 2) and water (study 1) consumption by households. Study 1 examined the effect of five information pages in teletext format consisting of weekly feedback in combination with a target for energy and water consumption by households. Study 2 examined the effect of prepayment gas meters as a feedback instrument on in-home gas consumption by households, as well as the level of satisfaction with the prepayment meter. With a prepayment meter (charged with a smart card), households pay for their gas as they use it.

2.

STUDY 1: ELECTRONIC FEEDBACK THROUGH INFORMATION PAGES TO REDUCE ENERGY AND WATER CONSUMPTION

For the content and layout of the information pages, three characteristics of goal-setting and feedback were taken into account. These were (1) frequency of feedback, (2) goal difficulty, and (3) type of information (percent change, cost or amount of energy). As stated above, feedback on the total energy consumption (outcome feedback) must be provided at least weekly, since interpreting outcome feedback requires retrospective assessment of past behavior and translation into specific energy-reducing activities. Goal difficulty is another important feature influencing the level of success of the goal-setting strategy. In the work of Becker (1978), every three weeks’ feedback on energy consumption was combined with a difficult 20% savings goal or an easy 2% conservation goal. Only the families who set a difficult 20% reduction goal used significant less electricity in

Technological innovations and the promotion of energy conservation

141

comparison to the control group. Finally, the best type of information (e.g. kWh or M3, percent change and/or cost) remains unclear from earlier studies. Probably different information provides consumers with different insights, which all have their relevance. So the feedback in the five information pages was provided once a week in kWh/m3, as well as in percent changes and costs. In summary, the purpose of the present study was to examine whether information pages in teletext format were an effective medium in reducing gas, electricity and water consumption of households. Exposure, comprehensibility and reliability of the information pages were the evaluation criteria of the success of teletext pages as a medium for informing households about their gas, electricity and water consumption.

2.1

Method

2.1.1

Participants and design

The design of this study was a pretest-posttest design with three groups; an experimental group (n = 29), a waiting list control group (n = 33) from the same district, and an extra, comparable control group from another district (n = 21). Subjects were assigned randomly to the experimental and the waiting-list control group. The extra control group was added to the design because we expected test effects in the waiting-list control group, who also had to fill in questionnaires on specific gas, electricity and water consumption reducing behavior. The households from the experimental and waiting-list control group were all house owners living in the same energyefficient new housing estate in a small Dutch village. Households in the extra control group were selected at random from a comparable energyefficient new housing estate, from which only monthly meter readings were collected; they received no questionnaire. An annual consumption norm for gas, electricity and water was used as a baseline for these three study groups. 2.1.2

Intervention

With a two-way communication system, the gas, electricity and water meters were read automatically, and households in return received weekly feedback through 5 information pages on channel 37 of their television sets. Their consumption levels for recent weeks were compared with a self-set target. Every week the pages were updated, and participants were able to consult the pages whenever they liked. The gas-consumption target was

142

Chapter 14

corrected for the outside temperature using the degree day method27. Since the houses were newly built, historical data about the energy and water use of the households was missing, so the target use was derived from a consumption norm calculated by the utility company. The target was the “normal” use minus the self-set conservation goal of 5%, 10% or 15%. 2.1.3

Procedure and instruments

The intervention with the information pages was started within the experimental group 4 months before the waiting list control group received the information pages. In total, both the experimental group and the waiting list control group received five questionnaires (1 pretest and 4 posttests). These questionnaires were designed to gain more insight into the effect of feedback on behavioral determinants and behavior. However, due to space restrictions we will not elaborate on the questionnaires here. Data from the meter readings were collected automatically for the experimental group and the waiting list control group. In the experimental group, the meter readings started one month before the experimental group received the information pages. At the same time, a meter reader started monthly gas meter readings in the extra control group in the comparable district. Households in the extra control group did not fill in questionnaires.

2.2

Results

2.2.1

Exposure to the information pages

Reduction effects can only be attributed to the information pages when the households were really exposed to the pages. Exposure as reported by the respondents in the experimental group (n = 26) was measured from T2 to T5 for the respondents and, when applicable, for their partner and children. In 27

The concept of ‘degree days’ is designed as a quantitative indication of the heating needed on a certain day with a certain outdoor temperature. One degree-day is the deviation of the average outside day temperature by one degree below 18°C. Since a decrease in temperature from 2°C to -3°C has a greater impact on the gas consumption level than a decrease from 18°C to 13°C, an extra correction has been applied. The degree-days of the period from November to February were multiplied by a factor of 1.1, whereas the degreedays from March to October were multiplied by a factor of 1.0, and the degree-days from April to September by a factor of 0.8.

Technological innovations and the promotion of energy conservation

143

total 77% of the respondents at T2 and 73% of the respondents at T5 reported that they consulted the information pages at least once a week. Furthermore, the majority of respondents (56% at T2 and 54% at T5) reported that their partners also consulted the pages at least once a week. Other respondents and their partners also consulted the pages, but less frequently. 2.2.2

Comprehensibility and reliability of the information pages

Respondents in the experimental group (n = 26) evaluated the content of the information pages on the level of comprehensibility at T2 using a fivepoint scale, ranging from strongly incomprehensible (1) to strongly comprehensible (5). In total 77% of the respondents judged the content of the pages as being very comprehensible, 15% comprehensible to a certain extent, whereas 8% found the content neither comprehensible nor incomprehensible. Furthermore, after two months of use (T2) the reliability of the information pages was judged by the respondents of the experimental group as being 69% for gas, 73% for electricity and 85% for water. After one year of use (T5), these percentages were, respectively, 81% for gas, 69% for electricity, and 73% for water. 2.2.3

Gas, electricity and water reduction

Comparison of the experimental group with the extra control group revealed a reduction of 18% for water consumption, 15% for electricity, and 23% for gas after 4 months. The significance of these results was tested with a MANOVA for repeated measures, which revealed a significant interaction effect for time and group F (1,45) = 11,92, p <. 001. Univariate testing for water, gas and electricity revealed that the reductions were significant. However, when the experimental group was compared with the waiting-list control group, only the water consumption was significantly more reduced in the experimental group than in the control group F (1, 45) = 11,92, p < .001. Respondents in the experimental group used 18.66 m3 less water (a reduction of 6%) than the waiting list control group during the first four months of the intervention period. No effects were found for gas and electricity reduction.

3.

STUDY 2: THE EFFECT OF A PREPAYMENT METER ON RESIDENTIAL GAS CONSUMPTION

With a prepayment meter, households pay for their gas as they use it. A display on the meter shows the amount of credit left, the cumulative amount

Chapter 14

144

of gas used, and the price of the gas per m3. The prepayment meter used in the present study was charged by means of a smart card. The prepayment meter is a modern version of the token meter, which disappeared in the Netherlands at the end of the sixties due to its susceptibility to fraud (Vermeer, 1998). The prepayment meter was not originally designed to give feedback on gas consumption. It may even provide meaningless feedback, since it provides only information about the amount of credit left, and does not indicate a reference consumption rate. This means that it is difficult for households to interpret their gas consumption in terms of high or low consumption. Knowledge about gas consumption alone may not be powerful enough to initiate action (Locke & Latham, 1990). In contrast with previous feedback and goal-setting interventions, the prepayment meter project was initiated by an umbrella organization of the utility companies, in association with a specific utility company that would be responsible for the trial. The fact that the project was an initiative by the utility companies (and not by scientists or the government) might improve the successful diffusion of this innovation, should it prove successful. As utility companies already have their reasons for introducing the prepayment meter on a wide scale, it is worthwhile to examine the success of this instrument as a feedback device, and if necessary to formulate recommendations for improvement.

3.1

Method

3.1.1

Participants and design

The design of this study was a pretest-posttest experimental design. Participating households (n = 114) were randomly assigned to an experimental group with a prepayment meter (n = 44), an experimental group with a prepayment meter and a notebook (n = 41), and a control group with normal, standard meters (n = 29). 3.1.2

Intervention

The prepayment meter was installed in the meter box at a level at which its display could be easily read. Nevertheless, the meter box is not the most accessible location, which might negatively influence the frequency of checking the meter. Participants received two smartcards that were unique to

Technological innovations and the promotion of energy conservation

145

their prepayment meter and that could be recharged at the local supermarket. The maximum amount transferable per transaction was 226.89 Euro.

3.2

The notebook

Participants in the intervention group received a notebook in which they could set a personal savings target of 5, 10 or 15%. Their gas consumption from the year before the prepayment meter project (1997/1998) was used as the baseline. Based on the savings target, a target consumption per degreeday was calculated for them, which they could find in their notebook. The aim of the notebook was to allow households to compare their target gas consumption per degree-day with their actual gas consumption per degreeday. They had to calculate their actual gas consumption themselves and contacted the utility company by phone to find out the degree-days for a particular period. 3.2.1

Procedure and instruments

Data was gathered through structured questionnaires and objective measurement of gas consumption. The intervention groups and the control group received three questionnaires during the prepayment meter project, one before the start of the project (T1), one after 9 months (T2), and one a week after the conclusion of the one-year project (T3). For the sake of brevity we will not go into the questionnaire data in this study either.

3.3

Results

3.3.1

Satisfaction with the prepayment meter

The participants had high expectations of the advantages of the prepayment meter. Overall, 75% of the households (n = 83) in groups A and B were highly satisfied with their decision to participate in the prepayment meter project after one year of use. Others were partially satisfied to some extent (18%) or neutral (7%). 3.3.2

Frequency of checking the meter

The results showed that 57% of the respondents checked their prepayment meter at least once a week, while 30% did so once every two weeks, and 13% checked their meter about once a month.

146 3.3.3

Chapter 14 Gas reduction

A statistical test of the percentage of savings in households in groups A and B (M = –1.96) with the percentage change in gas consumption in the control group (M = 2.11) showed a significant difference between the two groups. The results revealed that households with a prepayment meter reduced their annual gas consumption by 4%, whereas the notebook did not have an extra conservation effect. Fitting in the notebook may have been too complex and laborious.

4.

DISCUSSION AND CONCLUSION

The main purpose of the studies discussed in this paper was to assess the success of a innovative medium (information pages in teletext format) and an innovative product (prepayment gas meter) to reduce the energy and water consumption (study 1), and the gas consumption (study 2) of households. Study 1 showed that the information pages were effective in reducing the water consumption of households. A comparison of the experimental group with the waiting list control group revealed a reduction of 6%, and comparison with the extra control group revealed a reduction of 18% in their water consumption level. This study also showed a reduction effect for electricity (15%) and gas (23%) when the experimental group was compared with the external control group. However, no reduction effects for electricity and gas were found when the experimental group was compared with the waiting-list control group. The lack of significant effects on gas and electricity consumption in this last comparison may be attributed to test effects, i.e. repeatedly completing energy-related questionnaires. This attention to energy and water savings, as well as the knowledge that they were participants in the study, might have led them to pay more attention to energy and water conservation in their daily activities. However, we still found a reduction effect for water in spite of the test effects in the waiting list control group. The relatively low financial cost of water and the scarce attention to water conservation might have caused households to be less conscious of their water consumption. This might explain why feedback and goal-setting had a much stronger effect on water reduction in comparison to the reduction of electricity and gas.

Technological innovations and the promotion of energy conservation

147

In study 2 it was found that a prepayment meter reduced the gas consumption of a group of households by 4%. A reduction effect of 4% is comparable to that of self-monitoring of energy consumption (Hayes & Cone, 1981). However, people may not persevere with monitoring their energy consumption over the long term (Intomart, 1993). A prepayment meter forces them to persevere in monitoring to guarantee gas provision, so in the long run a prepayment meter might have a stronger effect on gas consumption. A notebook allowing a comparison between the actual gas consumption and the target consumption did not lead to greater reductions in gas use. Perhaps filling in the tables of the notebook each time before charging was too laborious, and calculating current gas consumption per degree-day was too complex. Although the methods of both studies were effective, study 1 ʊ in which feedback was combined with goal setting ʊ seemed to be more effective than the prepayment meter, which lacked an explicit gas reduction target. The feedback function of the prepayment meter might improve if households have the opportunity to compare their cumulative gas consumption level with a cumulative consumption target. However, the main conclusion of this paper is that feedback on energy consumption can contribute significantly to CO2 reduction goals.

REFERENCES Boer, J. de., & Ester, P. (1983). Consumentengedrag en energiebesparing: een veldexperimenteel onderzoek naar de effectiviteit van voorlichting, feedback en zelfregistratie. Amsterdam: V.U. Instituut voor Milieuvraagstukken. Becker, L.J. (1978). Joint effect of feedback and goal setting on performance: A field study of residential energy conservation. Journal of Applied Psychology, 63, 428-433. Department of Economic Affairs, (1999). Action program energy conservation (vos no. 13b63). Breda: Louis Vermijs. Ellerman, A.D., Jacoby, H.D., & Deceaux, A. (1998). The effects on developing countries of the Kyoto Protocol and carbon dioxide emissions trading. Washington: The World Bank, Development Research Group, Infrastructure and Environment. Energy Center The Netherlands (1999). Energie verslag Nederland [Energy report, the Netherlands]. Petten: Author. Geller, E.S., Winett, R.A., & Everett, P.B. (1982). Preserving the environment: new strategies for behavior change. New York: Pergamon Press. Hayes, S.C., & Cone, J.D. (1981). Reduction of residential consumption of electricity through simple monthly feedback. Journal of Applied Behavior Analysis, 14, 81-88. Houwelingen, van, J.H., & van Raaij, W.F. (1989). The effect of goal-setting and daily electronic feedback on in-home energy-use. Journal of Consumer Research 16, 98-105. Intomart, (1993). Actie Zuinig stoken Zuinig aan [Campaign Be Economic, Heat Economically] (4.9377/EB/jvdg). Hilversum: Intomart.

148

Chapter 14

Locke, E.A., & Latham, G.P. (1990). A theory of goal setting & task performance. Englewood Cliffs: Prentice Hall. Vermeer, B. (1998). Praktijkproef prepayment gasmeter: communiceren via de chipkaart [Field study prepayment gas meter: communicating with a smart card]. Gas Tijdschrift in de Energiemarkt, 118, 16-19. Winett, R.A., Hatcher, J.W., Fort, T.R., Leckliter, I.N., Love, S.Q., Riley, A.W., & Fishback, J.F. (1982). The effects of videotape modeling and daily feedback on residential electricity conservation, home temperature and humidity, perceived comfort, and clothing worn: winter and summer. Journal of Applied Behavior Analysis, 15, 381-402.

Chapter 15 HOUSEHOLD ENERGY CONSUMPTION: Habitual Behavior and Technology Wim J.M. Heijs

1.

INTRODUCTION

Consumers generally have a positive attitude towards energy conservation and a fair amount of environmental consciousness. Still energy use in households (and in particular electricity) continues to rise (EnergieNed, 1995; ECN, 1996). The transformation of this consciousness into action is obstructed, among other reasons, by the fact that energy use is often bound up with more comprehensive behavioral patterns and habits. Applied research on energy-related behavior shows that habits can be better predictors than variables in the predominant models of attitude-behavior relationships (Engel, Blackwell and Miniard, 1995; Karns and Khera, 1983; Macey and Brown, 1983; van Raaij and Verhallen, 1983; Ritsema, Midden and van der Heijden, 1982). Habits may resist the cognitive and financial means that are primarily based on these models (de Bruin and Siderius, 1993; Gladhart, 1977; Hoevenagel et al., 1996; Jelsma and Popkema, 1997). In this chapter, the question is raised how technology can contribute to decrease household energy use resulting from habitual behavior. As a first step in answering this question the concept of habitual behavior must be elucidated. In the applied research mentioned above many examples are given, such as clothing habits, leaving lights on, using bedrooms for other purposes, ventilation habits or doing the laundry on Mondays. Relevant habitual behavior thus covers a broad range: specific or more general, frequent as well as less frequent, and more or less automatic or subconscious activities play a role. However, an unambiguous definition and 149 P.-P. Verbeek and A. Slob (eds.), User Behavior and Technology Development: Shaping Sustainable Relations Between Consumers and Technologies, 149-157. © 2006 Springer.

150

Chapter 15

models explaining the concept are mostly absent. Definitions and frameworks can be found in fundamentally oriented research. Two groups may be identified. In the first, the variable habit is used to enhance the predictive power of models of repetitive actions. It is defined as ‘past behavior/experience’, and this behavior can be more or less conscious and frequent (e.g. Goldenhar and Connell, 1992; Hamid and Sheung-Tak, 1995; Macey and Brown, 1983). The second group limits habitual behavior to frequent, situation-specific, goal-oriented and automatic acts (e.g. Aarts, 1996; Triandis, 1977; Mittal, 1988; Ronis, Yates and Kirscht, 1989; Bargh and Gollwitzer, 1994; Ouelette and Wood, 1998). Repetition causes reasoned actions with the associated situations and goals to be stored into mental structures. These structures guide future behavior without conscious monitoring when triggered by an external cue. ‘Habit’ is measured variously as ‘frequency of past behavior’, ‘unconscious recurrence of the acts in the past’, or ‘frequency of association’ of certain situations with behavioral choices.28 The main problem posed by present literature is that the definitions or models are either not specified or seemingly incompatible with habitual activities in household energy use. Doing the laundry on Mondays, for instance, does not fit a description of an automatic act without conscious monitoring. There are also some additional problems. It is not certain whether ‘habit’ refers to a mental disposition, to the behavior, or to both. The criteria are not clear, e.g. what frequency indicates a habitual act, and when may it be called really ‘automatic’ or ‘unconscious’? Repetitive nonbehavior is not explicitly dealt with; this type of habit, though, plays an important role in energy-related activities (i.e. leaving lights on). And the situation or cue (e.g. technology) are mentioned occasionally, but without elaboration.

2.

THEORETICAL FRAMEWORK

To overcome the main problem, and also some of the additional ones, a new definition and model were constructed, while trying to maintain the possibility of using existing results. The definition separates ‘habit’ from ‘habitual behavior’. ‘Habit’ is: ‘a mental structure, composed of a situation or domain, a related goal, a behavioral disposition to reach this goal and a 28

This paper is based on an extensive literature review. Details of the research and the models referred to are described in a report (Heijs, 1999), and they will be publicized more extensively elsewhere.

Household energy consumption

151

cue (a stimulus triggering the structure and thus the behavior), that is learned through reinforced repetition of the behavior in that particular situation and in response to that particular cue’. This usually is the independent variable. An example is the habit to ventilate (disposition) the bedroom (situation) to obtain a healthy atmosphere (goal) around 10 o’clock (cue). These elements are represented and connected in the mental habit structure. ‘Situation’ refers to part of the physical or social environment, and ‘domain’ to a rounded part of life (e.g. personal hygiene). The ‘cue’ generally is an environmental stimulus, but it can also coincide with the situation as a whole or a domain (e.g. leaving the doors open when at home). Subsequently, ‘habitual behavior’ is: ‘the manifestation of a habit in repeated, overt (non)behavior'. This usually is the dependent variable. The amount of consciousness at the start, and the frequency and degree of automatism while performing, may vary. These criteria for the habitual nature of an activity are replaced by the likely existence of a mental 'habit' structure.29 The model (figure 15-1) is based on the theory of planned behavior (Ajzen, 1991; in the dotted frame). The main reasons to use this theory are that it provides a link to existing research on repetitive behavior, that it may be used to illustrate how planned behavior and habitual actions are related, and that the mental ‘habit’ structure is akin to the concept of ‘attitude’ in the theory. Both are presumed to be ‘mental guides’ of behavior. Therefore, they are placed next to each other in the new model. The situation and cue are added as variables that are conditional for the activation of a habit or of the mechanisms of planned behavior, e.g. if suitable habits are lacking (the arrows pointing to the dotted frame). Behavior may consist of habitual action, planned behavior or a combination, and it is recursively connected to the habit, representing the development and strengthening of habits through repetition. The model allows for the differentiation among three main kinds of behavior. Assuming that action is preferably initiated by mechanisms that require less cognitive effort (such as habits) instead of more demanding ones (such as attitudes and planning), the first kind is habitual behavior with an unconscious start and automatic progress (route 1 in the figure). The habit is activated by a situation and a cue, and triggers action without conscious choice or planning (i.e. automatically turning on the lights when entering). 29

Methods are proposed to verify such existence, but that exceeds the scope of this paper (see Heijs, 1999).

152

Chapter 15

This kind is often well-rehearsed and strong. The second kind (route 2) is habitual behavior with a conscious start and more or less automatic progress: activation of a habit leads to intentional choices and planning because, for example, it is less strong or it has decision rules built into it (e.g. careful sorting when doing the laundry on Mondays). In both cases, the attitude is not an active component, even though this might be ostensible in the latter. The third kind is planned behavior (route 3). This will be activated, for example, in unfamiliar situations or when cues or appropriate habits are absent. Repetition may lead to the development of a new habit. Hybrid forms are possible, such as habitual acts with an unconscious start and nonautomatic progress, e.g. when situational interruptions occur. 1,2,3

cue

1,2

situation/ domain

1

habit

2

3

attitude

subjective norm 3

perceived control

3 3

intention/ planning

2,3

habitual behavior planned behavior

3

1: habitual action: unconscious start, automatic progress 2: habitual action: conscious start, more or less automatic progress 3: planned behavior Figure 15-1. A model of energy-related habitual household behavior

Based on the mentioned literature, habitual activities not only differ in kind but also by their genesis. First, they may be learned in various ways. In operant conditioning they are strengthened or weakened through reinforcement (e.g. because the action generates comfort). They can also result from classical conditioning: a neutral stimulus becomes a cue by frequent joint occurrence with a former cue (e.g. if it is dark often when arriving at home, the lights may be turned on at arrival). And they can be formed by social learning (from the imitation of a model that is reinforced). Second, the behavior can initially be conscious (if the habit follows repeated planned behavior, such as leaving the lights on because one thinks that turning them off and on costs more energy) or unconscious (unaware of the actions or

Household energy consumption

153

goals at the time, e.g. leaving the lights on by coincidence). And thirdly, habits may spread. They can be generalized unconsciously to a new, similar situation with a similar cue (e.g. likewise turning the lights on at the office). This may also occur in less similar situations if a habit is strong. Conscious generalization is possible if, for example, the activity or the goal is valued (striving for the same comfort elsewhere). In both cases, dysfunctional behavior may result (e.g. ventilating for several hours after moving to a new house with mechanical ventilation).

3.

INTERVENTION AND PREVENTION

The model is constructed on the basis of existing theories and research, but it is not yet tested. This applies also to the following suppositions, which are logical inferences from the model but require research to prove their validity. Bearing this in mind, one can discern four general possibilities. The first is the alteration of the mental habit structure itself. Using information or de-conditioning one can try to change the elements or weaken the links between them (e.g. explain how mechanical ventilation works or apply energy taxes to modify the disposition to ventilate or the link between ventilating and a healthy atmosphere as the goal). A second way is to block the activation of the habit. According to the model, this requires a transformation of the situation or the cue, e.g. by technological means (abundant daylight in the office may prevent the habitual switching on of lights). Third, when a habit is already activated, the activity can be hindered by situational, psychological or social barriers, thus forcing a transition to planned behavior (e.g. additional buttons to push on a new washing machine to override a low temperature setting, or social control at work when leaving the lights on). And fourth, negative consequences of habitual behavior can be mitigated by technology. This seems to be the preferred strategy because behavioral modification is made superfluous (a design that makes appliances ‘fool-proof’ or evokes the desired conduct by means of affordances that are in line with a natural way of operating). Applied research indicates that any behavioral intervention should meet a number of general conditions. They must conform, among other things, to present knowledge, the feasibility of alternative action and constraints posed by existing behavioral patterns, goals, values, costs and benefits. In addition, users require an explanation of the necessity of the intervention, they should retain sufficient control and be given compensation (i.e. an increase in comfort).

154

Chapter 15

In the case of habits, however, consideration of these general conditions is probably not sufficient. A selection of methods must also be based on the kind of habitual activity and its genesis, because activities that appear to be similar can in fact be totally different. Leaving on the lights, for example, might be consciously learned and started because one is convinced that turning them on and off costs more energy. It might also be unconsciously learned and started because it is easy. In the latter case, a technological intervention (such as a sensor) can be applied, but in the former this could cause adverse reactions, as it may be perceived as a violation of personal control. In view of the many possibilities, an attempt was made to build an aid for the selection of strategies (“HABIT”: Habit Assessment and Behavior Intervention Typology). It consists of eight cells resulting from a fusion of the two main kinds of habitual action (unconscious start/automatic performance, and conscious start/more or less automatic performance) with four kinds of genesis ((un)conscious learning x degree of generalization). Each cell corresponds to a type of habit and contains the recommended techniques and their preferred order of application. An unconsciously learned and started habit of leaving lights on at home, for instance, may be changed by a sensor. Information can be used as support but not as a first step, since this type is often strong and resistant to cognitive measures. The same behavior can also be the result of a consciously learned and started habit that has generalized to the office. In this case, a technological intervention as a first step carries a risk of reactance; it may therefore be more advisable to first disprove the ideas that have caused this habit and to pay attention to both places (to prevent that the behavior in the office might generalize back again to the home). After this, technology may be used to make the change more durable. Prevention subsumes the hindering the unwanted generalization of habits to other situations or appliances and precluding the formation of unwanted habits in relation to new products. In this case, technological measures seem to be prominent because they offer an opportunity to leave out costly and extensive behavioral corrections afterwards. This requires a thorough analysis of the unwanted habits, regarding among other things the type and affordances and features of the ‘natural’ and ‘desirable’ behavior with similar applications. The preferred method conforms to the second strategy in intervention: obstructing the activation of the habit (i.e. by adjusting the situation or cue, or by avoiding unsuitable affordances). As a result the habit might cease to be. Alternatively, the habitual behavior could remain intact while the results are corrected (as in the fourth intervention strategy). In view of the many examples of unwanted habits that might be prevented this way (e.g. less energy-efficient habitual activities in operating a washing

Household energy consumption

155

machine), the incorporation of preventive analyses and measures in a new design methodology is an important objective.

4.

THE ROLE OF TECHNOLOGY

In the preceding sections, the role of technology in the intervention and prevention of habitual behavior was mentioned occasionally as a part of the rationale. In regard to the main question of the chapter, this role will now be made more explicit. A general answer to this question is that technology can contribute to the basic strategies for intervention and prevention. It may, for example, be used to change the habit structure by providing information through feedback and feed-forward (perhaps on a display or by visual signals to correct undesirable behavior). The alteration of the situation or the cue to prevent activation of a habit can be accomplished by a new design or replacement of the equipment. In the design, barriers or extra steps can be implemented to hinder habitual behavior after activation, forcing the user to initiate a process of planned behavior (an additional button to be pushed in order to proceed). Feedback and persuasive messages on a display are also functional in this respect. And the design of ‘fool-proof’ applications or the mitigation of negative results of undesirable habitual activities explicitly require technological measures in both intervention and prevention. This general answer, however, needs some fine-tuning because there are probable limits to the potential of technology regarding habitual behavior. According to HABIT, technological measures should preferably be used as the first step only in relation to some types of habit, namely those that are learned (and maybe generalized) unconsciously, started unconsciously and performed automatically. These habits are usually the stronger ones (the first kind in the model), and possibly less sensitive to other methods like information. If, on the other hand, some features are more conscious or less automatic (in the remaining types), technological measures should be used with caution because of possible adverse reactions. In these cases, technology may be supportive of information or behavioral modification, which should come first (e.g. in order to correct faulty cognitions). Finally, the general conditions concerning behavioral interventions mentioned earlier allows a preference ranking of the form that technological measures might take in case they seem liable. These conditions state, among other things, that interventions must conform to the feasibility of behavior, to

156

Chapter 15

existing behavioral patterns and to retaining sufficient user control. This will promote acceptance of the measures and decrease the risk of reactance. Therefore, the top of the list will consist of measures that do not require a change of conduct: the behavior remains the same while the consequences are corrected (e.g. a thermostat that evaluates a setting in relation to the outside temperature before signaling the furnace). Second are appliances that render the undesirable behavior superfluous but not impossible (e.g. an automatic shower thermostat, which might also be experienced as comfortable). Measures that elicit the proper conduct in a ‘natural’ fashion (using affordances of user interfaces) are third. They evoke behavioral changes but do not require an additional effort on the part of the user (such as a panel on a washing machine that makes the choice of an eco-program the natural thing to do). In fourth place are measures that try to adjust unwanted actions (giving feedback and advice, correcting the situation or the cue, e.g. by visual or auditory signals). Fifth are changes that hinder behavior through alterations in the situation or cue (such as pushing extra buttons). And interventions that explicitly obstruct unwanted actions are last (a washing machine that has no 90°C setting will be less acceptable to a number of users). In summary, technology can make a valuable contribution to decreasing household energy use that results from habitual behavior. As a first step, however, technological measures are indicated only for those types of habitual behavior that are unconsciously learned, and started and performed automatically. In other cases, technology should be supportive of information and behavioral modification. If technological measures are considered, the preferred strategy would be to choose those that do not require a behavioral change or that cause the activities to be redundant (while maintaining subjective control and possibly adding comfort). Strategies explicitly blocking habitual behavior without additional measures should be applied with caution because they might meet with adverse reactions.

REFERENCES Aarts, H. (1996). Habit and decision making: the case of travel mode choice. Dissertation. Nijmegen: KUN. Ajzen, A. (1991). The theory of planned behavior. In: E. Locke (ed.). Organizational behavior and human decision processes, 179-211. Bargh, J. and Gollwitzer, P. (1994). Environmental control of goal-directed action: autonomic and strategic contingencies between situations and behavior. Nebraska Symposium on Motivation, 41.

Household energy consumption

157

Bruin, P. de and Siderius, P. (1993). Energiebesparing bij huishoudens: een scenariostudie. Den Haag: Swoka. ECN. (1996). Energieverslag Nederland. Petten: ECN. EnergieNed. (1995). Energiedistributie in Nederland 1995. Arnhem: EnergieNed. Engel, J., Blackwell, R. and Miniard, P. (1995). Consumer behavior. Orlando: Harcourt. Gladhart, P. (1977). Energy conservation and lifestyles: an integrative approach to family decision-making. Journal of Consumer Studies and Home Economics, 1, 265-277. Goldenhar, L. and Connell, C. (1992). Understanding and predicting recycling behavior: an application of the theory of reasoned action. Journal of Environmental Systems, 22, 91-103. Hamid, P. and Sheung-Tak, Ch. (1995). Predicting antipollution behavior. Environment and Behavior, 27, 679-698. Heijs, W. (1999). Huishoudelijk energiegebruik: gewoontegedrag en interventiemogelijkheden. Rapport. Eindhoven: TUE/Novem. Hoevenagel, R., van Rijn, U., Steg, L. and de Wit, H. (1996). Milieurelevant consumentengedrag: interimrapport. Rijswijk: SCP. Jelsma, J. and Popkema, M. (1997). Gedragsbeïnvloeding door technologie. Publicatiereeks Milieustrategie. Rapport 1998/1. Den Haag: MVROM. Karns, D. and Khera, I. (1983). US consumer attitudes and home-heating conservation behavior: a multiattribute longitudinal model. Journal of Economic Psychology, 4, 57-70. Macey, S. and Brown, M. (1983). The role of past experience in repetitive household behavior. Environment and Behavior, 15, 123-141. Mittal, B. (1988). Achieving higher seat belt usage: the role of habit in bridging the attitudebehavior gap. Journal of Applied Social Psychology, 12, 993-1016. Ouelette, J. and Wood, W. (1998). Habit and intention in everyday life: the multiple processes by which past behavior predicts future behavior. Psychological Bulletin, 124, 54-74. Raaij, Th. van and Verhallen, T. (1983). A behavioral model of residential energy use. Journal of Economic Psychology, 3, 39-63. Ritsema, B., Midden, C. and van der Heijden, P. (1982). Energiebesparing in gezinshuishoudens: attitudes, normen en gedragingen. Leiden: Werkgroep E&M/ESC. Ronis, D., Yates, F. and Kirscht, J. (1989). Attitudes, decisions and habits as determinants of repeated behavior. In: A. Pratkanis, S. Breckler and A. Greenwald (Eds.). Attitude structure and function. Hillsdale (NJ): Lawrence Erlbaum Ass. Triandis, H. (1977). Interpersonal behavior. Monterey, CA: Brooks/Cole Publ. Cy.

Chapter 16 MARKETING OF TECHNOLOGICAL PRODUCTS: Theory and Methods Nicole van Kesteren and Ree M. Meertens

1.

INTRODUCTION

Whereas the previous chapters have focused on the influence of technology on human behavior, this chapter looks at how technological products can best be presented to consumers. It will be clear that a welldesigned technological product, for example an ecologically built house (see section 2.1) or an energy efficient washing machine, is less effective and will have less influence if users are not interested in buying the product in the first place. The main question addressed in this chapter is what we should communicate about the product to consumers or product users. Should we, for example, stress how environmentally friendly the washing machine is? Or should we try to alter the dull image most ecological products have? Marketing is a discipline that is primarily concerned with how promotional efforts should be designed. This chapter introduces some well-known theories and models in this area. First, we introduce two marketing models that go into further detail about the relationship between message strategy and consumer behavior theories. These models will give more insights into the emotional response to advertising and provide practical guidelines for designing communication strategies and developing the more creative part of a campaign. Second, we will discuss another important tool in marketing and advertising: image management. The notion that perceptions of reality are more important than objective reality has long been recognized as a powerful tool in influencing 159 P.-P. Verbeek and A. Slob (eds.), User Behavior and Technology Development: Shaping Sustainable Relations Between Consumers and Technologies, 159-171. © 2006 Springer.

160

Chapter 16

consumer or buying behavior. The research technique called ‘laddering’ can be used to identify and alter images of consumer products. Empirical evidence for the advertising planning models and the laddering technique are discussed.

2.

ADVERTISING TACTICS

In advertising and marketing literature several ideas have been developed about the stages that people go through before taking action. These stages can be summarized by the ‘learn-feel-do’ sequence. This sequence suggests that consumers will purchase a product only after the target group has gained knowledge (exposure, attention, understanding) and has developed a more positive attitude (attitude change) about the product or the behavior. Elaboration on the ‘learn-feel-do’ sequence has led to the development of several alternative models. Although these alternative models may differ in some respects, they are based on specific ideas about consumer involvement and brain specialization (Vaughn, 1980; 1986; Rossiter, Percy & Donovan, 1991). Consumer involvement refers to a continuum of consumer interest in products, services or behaviors. On the one hand, products are represented that are important to the consumer and have some concomitant financial, technological and social risks. In such cases, it is worth the consumer’s time and energy to pay more attention to the decision-making process and to carefully consider product alternatives. On the other hand, products are represented that are not as important to the consumer, and lack interest and associated financial, technological and social risk. In such cases, it may not be worth the consumer’s time and energy to carefully consider product alternatives. Low involvement purchases are likely to result in a limited decision-making process. The ideas about specialization of left and right cerebral hemispheres arose from research showing that the left cerebral hemisphere is focused on logic, language and analysis. The right cerebral hemisphere is focused on intuition, vision and synthesis. In short: the left cerebral hemisphere is more cognitive, whereas the right cerebral hemisphere is more affective. Vaughn states that the location of cognitive and affective processing in the brain is less important than the fact that people are capable of processing information in both a cognitive and an affective way. In the next section some of these alternative models ʊ the FCB model of Vaughn (1980; 1986) and the model developed by Rossiter and Percy ʊ will be described. In both models the ‘learn-feel-do’ sequence and the theoretical insights into consumer involvement and brain specialization are taken into account.

Marketing of technological products

3.

161

THE FCB MODEL OF VAUGHN

With the introduction of the FCB model30, Vaughn offered marketers the first helpful model for organizing strategic advertising planning. In the FCB model, two dimensions are distinguished: the thinking-feeling dimension and the high-low involvement dimension. This distinction suggests that some purchase decisions are dominated by thinking or by feeling, while other purchase decisions are characterized by higher or lower levels of involvement. Of particular concern was the thinking-feeling dimension, which takes into account a person’s ability to process information in both a cognitive and affective way. Combining these two dimensions results in the delineation of four quadrants (see Figure 16-1). Each quadrant can be taken as a starting point for the development of an advertising strategy.

High Involvement Low

Thinking

Feeling

Informative

Affective

Learn → feel → do

Feel → learn → do

Habit formation

Hedonistic

Do → learn → feel

Do → feel → learn

Figure 16-1. Vaughn’s FCB model (Vaughn, 1986, pp. 58)

The first quadrant (high involvement, thinking) is described as ‘informative’. According to Vaughn, the valid effect hierarchy in this quadrant is the classic sequence ‘learn→feel→do’. In this quadrant, the consumer will feel an urge to gather a lot of information prior to purchasing because of their involvement in the purchase and the cognitive character of the purchase. Examples of products belonging to this category are a car or a house. Technological examples in this category are a low-energy boiler or a PV-installation on the roof. Because the consumer might be viewed as a ‘thinker’, the advertising strategy should focus on specific product information and rational arguments. The use of product demonstrations is another possibility.

30

FCB is an abbreviation for ‘Foote, Cone & Belding’, an advertising agency in Los Angeles (Vaughn, 1986).

162

Chapter 16

The second quadrant (high involvement, feeling) is described as ‘affective’. For this quadrant, Vaughn suggests that the sequence ‘feel→learn→do’ is the relevant effect hierarchy. The purchase decision does matter, but specific product information is less important than the general attitude or feelings towards the product. Product information is less important because the involvement of the consumer is focused mainly on the relevance of the product to his or her self-esteem or social status. Examples of products in this category are jewellery and fashion. Many commercials by the Dutch government (e.g. ‘Postbus 51’) on mobility seem to be directed at this sequence. By creating a threatening atmosphere around environmental problems resulting from excessive car use, or a positive atmosphere around the use of alternative forms of transportation, such as public transport, carpooling and biking, they try to appeal to people’s feelings. The third sequence (low involvement, thinking) is labeled ‘habit formation’. The effect hierarchy here is ‘do→learn→feel’. The purchase is comparatively unimportant. The consumer’s behavior is to test the product first to determine whether or not it is acceptable. In these purchase decisions, habit formation could occur because habits often preclude thinking about purchases. The daily purchase decisions, such as buying food and household goods, belong to this quadrant. Cheap and ordinary technological products, e.g. batteries, will be part of this quadrant as well. Central to a successful advertising strategy in this quadrant is the creation of habit formation in buying behavior. The most important goal of the advertising strategy is to remind the consumer of the product. The distribution of coupons and free samples may be helpful in reaching this goal. The fourth quadrant (low involvement, feeling) is called ‘satisfaction’. The accompanying effect hierarchy is ‘do→feel→learn’. As in the third quadrant, the consumer will test the product or service and find out automatically whether it is satisfactory. Examples of relevant products in this category are cigarettes, candy and entertainment (e.g. movies). Central to the advertising strategy is to get attention regularly, for example by the use of billboards and newspapers. According to Vaughn, the quadrants in this model should not be seen as a strait-jacket, but rather as a way of clarifying the main decisions in advertising and marketing. There are no clear lines between the quadrants. Because involvement and thinking and feeling are dimensions rather than categories, forms other than those depicted in the matrix are conceivable.

Marketing of technological products

4.

163

THE ROSSISTER AND PERCY GRID

Rossiter and Percy (1987) elaborated on the FCB model and introduced an extensive advertising grid, which they call the Rossiter-Percy grid. Although the Rossiter-Percy grid is based on the FCB grid ʊ like Vaughn, Rossiter and Percy also present a 2 x 2 framework by combining two dimensions of consumer behavior ʊ there are some major differences. First, Rossiter, Percy and Donovan (1991) state that developing a favorable consumer attitude towards a product or brand has no use unless the consumer is reliably aware of the product or brand. As a result, creating brand awareness is a necessary communication objective for advertising prior to influencing brand attitude. Brand awareness could happen in two different ways. Suppose a customer goes to a shop to buy a new audio set. If the customer recognizes a specific brand at the moment of purchase, it is a matter of brand recognition. If the customer remembers the existence of the brand before the moment of purchase, it is a matter of brand recall. Second, Rossister and Percy introduced a more precise conceptualization of the involvement dimension. In the Rossiter-Percy model, the measurement of involvement is limited by asking consumers whether the risk of the choice for this brand or product is so low that it can be described as ‘I will try this brand/product and I’ll see’ (low involvement), or whether the risk of the choice is so high that it can be described as ‘It is worth studying the information at a more detailed level’ (high involvement). The FCB model has a broader measurement of involvement, including the importance of the decision, the level of thinking needed for the decision, and the involved risks. Rossiter et al., consider the clearer operationalization of ‘involvement’ as a big advantage for the practical applicability of the model. Moreover, they consider the unambiguous, dichotomous division as an advantage. In the FCB model it is not clear whether purchase decisions can be classified as ‘minimally involved’ or ‘highly involved’. Third, Rossiter and Percy have replaced the ‘think-feel’ dimension with a dimension called ‘informational-transformational’. This dimension departs from the idea that all products and brands are purchased in response to either an informational or a transformational motive. Informational motives are negative in origin. They stem from the buyer’s experience of an aversive stimulus or event. This aversion pushes the buyer’s emotional state or ‘drive’ below the equilibrium level, and the buyer will be motivated to remove the aversion or at least reduce it as far as possible towards the equilibrium. Transformational motives are positive in origin. They arise from the buyer’s

164

Chapter 16

positive or rewarding stimulation. In this case, the person is motivated to increase the emotional state or ‘drive’ above the current equilibrium level to a more desirable state. The negative motives should respond mainly to an informational or ‘reason why’ style of advertising. The consumer should be provided with information about benefits that will eliminate or prevent problems, meet an ideal, resolve consumption and choice conflicts, or simply provide a convenient way to re-stock the brand. The positive motives should respond better to a transformational or ‘image’ style of advertising in which the consumer perceives that by buying the brand he or she can gain sensory, intellectual or social rewards. Nevertheless, the Rossiter-Percy grid supposes there are cognitive and affective tactics for each quadrant of the model. This reflects the notion that all advertisements represent a balance between socalled ‘rational’ and ‘emotional’ stimuli. In addition, the Rossiter-Percy grid offers some leads for the communication strategy concerning low and high involvement purchases. In short, low involvement tactics should focus on just one or two benefits of the product, whereas high involvement tactics should emphasize the multiple benefits of a product, which characterizes the carefully considered comparative decision process a consumer is going through. According to Rossiter and Percy, it is relatively easy to recover the underlying category of purchase motivation (that is informational or transformational in nature) of a brand. Consumer motives can best be measured and analyzed by motivational research methods, such as in-depth interviews and projective techniques. Having identified the basic category of purchase motivation, the marketer must position the brand or product based on this motivation category. The application of this approach to the marketing of environment-conscious products could be useful. For example, is the purchase of a low-energy washing machine guided by the need to solve problems (negative in origin, e.g. the old washing machine is broken) or the need for social approval (positive in origin, e.g. low-energy washing machines are very popular)? The advertising or promotion campaign needs to emphasize the washing machine’s unique benefits. In this context Rossiter and Percy distinguish five brand attitude strategies in order to position a brand by emphasizing its product benefits (see table 16-1).

5.

BRIEF HISTORY OF ADVERTISING RESEARCH: CONTRIBUTION AND CRITICISM

The models described above are characterized by their high level of face validity. Nevertheless, they derive their acceptance more from their intuitive

Marketing of technological products

165

plausibility than from large-scale empirical support. The strength of these models is that they give insight into the issues that are relevant for communication activities. Moreover, they have contributed to more realistic expectations about the behavioral effects that can result from communication activities. Rossiter et al.’s ideas are mainly applied for the benefit of product marketing. A preliminary study indicates that advertising that followed the Rossiter-Percy model indeed led to more purchase interest than advertising that was not compatible with the model (Kover & Abruzzo, 1993). Table 16-1. Alternative strategies for influencing product or brand attitude Strategy 1: Increase the brand’s perceived delivery of a benefit. Note, however, that some benefits have a threshold beyond which further increases are meaningless (e.g. a plastic toothpaste package be ‘more plastic’) or there may be an ideal point beyond which increases would be worse (e.g. too much spearmint taste in a toothpaste). Strategy 2: Increase the importance of a benefit that the brand delivers uniquely. May be feasible in new product categories, but difficult to accomplish in established categories where benefits are well known. Strategy 3: Add a new benefit that is important and that the brand delivers uniquely. Typically, this requires a product improvement of an innovative and major nature; small improvements do not carry enough importance or weight. Strategy 4: Weaken a competitor’s perceived delivery of a benefit. Can sometimes be achieved by comparative advertising to increase the brand’s relative uniqueness. Strategy 5: Alter the composition rule so as to favor the brand. This amounts to altering benefit importance weights. As in the second strategy, this can be difficult except in relatively new product categories. Rossiter en Percy (1997, pp.151).

A model from the field of social psychology that has gathered increasing empirical evidence in relation to the process of persuasion is the Elaboration Likelihood Model (Petty & Cacioppo, 1986). This model supposes two routes by which people process information: a central route and a peripheral route. This model predicts that targets will process information primarily via the central route when motivation is high and when they have the capacity to process information thoroughly. People will process information primarily via the peripheral route if they are not motivated or capable of processing information thoroughly. Attitudes formed by the central route are, according to Petty and Cacioppo, constant in time and difficult to change, whereas attitudes formed by the peripheral route are unstable in time and easily changed by new information. The impact of several variables on the formation of attitudes has been studied extensively (for a recent review see Petty & Wegener, 1998). For example, it has been found that high-quality arguments

Chapter 16

166

are most likely to be effective if processed via the central route. Other relevant factors that determine how people process information and form attitudes include: classic source (e.g. expertise), recipient variables (e.g. mood) and contextual variables (e.g. distraction). Another model from the field of social psychology that may be especially relevant to the high involvement-thinking quadrant (or high involvementinformational quadrant) is the Theory of Planned Behavior (Ajzen, 1991). According to this theory, the persuasive message should be made relevant by, for example, adapting the message to existing beliefs, by reinforcing positive beliefs, by changing negative beliefs, or by introducing new information.

6.

MEANS-END CHAIN THEORY AND LADDERING

Particularly relevant to the study of consumer behavior is the consumer’s perception of a product’s attributes or perceived images of product categories or brands. A product can be seen as a set of product attributes, and varies in the extent to which they seem to agree with an individual’s self image. Because the majority of product classes do not differentiate from others significantly, marketers try to fulfill a specific need by creating an effective product image in the mind of a customer. The image of a product with respect to how the consumer sees him/herself is probably more important to its ultimate success than are its actual characteristics. In the 1980s Reynolds and Gutman (1984; 1988) introduced a research technique, called ‘laddering’, to systematically analyze the way consumers perceive or structure images of products. Laddering is based on the ‘Means-End Theory’ of Gutman (1982). A means-end chain is defined as the relationship between product attributes, consumer consequences and personal values. Product attributes are aspects of products or services. Consumer consequences are the result of consuming products or services. Those consequences may be desirable or undesirable. Personal values are important beliefs people hold about themselves or beliefs other people hold about them. The means-end chain perspective relies first on the distinction Rokeach (1973) made between instrumental and terminal values. An instrumental value reflects an external orientation concerning how others perceive a person. A terminal value refers to an internal orientation of how a person sees him/herself. The second theory underlying the means-end perspective is Rosenberg’s Expected-Value Theory. This theory

Marketing of technological products

167

posits that behavior is influenced strongly by personal values. Personal values can be defined as the ultimate criteria on which people choose between alternative products, behavior and other stimuli. According to Rosenberg, consumers learn to associate consumer consequences with product attributes they have reinforced through their buying behavior. Consumers will choose those products or behaviors that maximize desired outcomes and minimize undesired outcomes. The means-end chain model embodies different levels of abstraction. Physical characteristics are defined as those that are measurable in physical units, such as ‘color’ or ‘miles per gallon’. The abstracted properties represent attribute designations that are more subjective in nature, such as ‘looks good’ or ‘beautiful color’. Functional consequences are exemplified by outcomes such as ‘saves money’. Such consequences are instrumental in achieving psychosocial consequences, such as ‘having more friends’ and ‘having fun’. The instrumental-values level reflects an external orientation with regard to how we are perceived by others. For example, ‘makes me feel more important’ or ‘makes me feel accepted’. The terminal or internalvalues level relates to how we see ourselves, for example ‘self-esteem’ or ‘security’.

Final values Instrumental values Social-psychological consequences Functional consequences Abstract characteristics

Concrete characteristics

sociable and successful person ↕ social prestige ↕ quick and athletic man ↕ Athletic ↕ motor performance, petrol expenditure Ĺ red color, small wheel, low roadholding qualities

Figure 16-2. A simple fictitious example of a ladder (based on Stienstra (1998))

Reynolds and Gutman (1988) translated the means-end perspective to the development of marketing strategies in a process they called laddering. Laddering refers to an in-depth, one-on-one interviewing technique used to

168

Chapter 16

develop an understanding of how consumers translate the attributes of products into meaningful associations with respect to self. Consumers are ‘helped up the ladder’ through a series of questions, such as: “Why is that important to you?” The qualitative research results permit an understanding of consumers’ underlying personal motivations with respect to a given product class. Each unique pathway from an attribute to a value represents a possible perceptual orientation with respect to viewing the product category. An example of a simple ladder is represented in Figure 16-2. This ladder shows the steps in a laddering procedure, with one respondent considering the purchase of a car. For a low-energy car the sequence could be as follows: the attribute electric gas injection, for example, implies or leads to the consumer consequence ‘economical’, a benefit. This property may lead to satisfaction of the value environment-minded (Van Raaij, 1994). In general, the final results of a laddering procedure are based on several question rounds with more respondents (60 or 70). It is therefore both more complex and informative. Reynolds and Gutman developed a set of guidelines for interviewing and for interpreting and organizing research results. Other authors have proposed procedures that categorize respondents’ reactions in less structured ways, by eliciting people’s reasons for pursuing a certain goal or having a certain opinion (Pieters, Baumgartner & Allen, 1995; Bagozzi & Dabholkar, 2000). With such procedures, the results of the laddering process are chains of arguments rather than chains starting with concrete elements and ending with abstract elements. It is beyond the scope of this chapter to discuss these guidelines in detail (see Reynolds and Gutman, 1984; 1988). Determining which aspects in specific product categories are important to consumers, and how these aspects are related to the personal values of consumers, offers a clear starting point for the development of ideas for advertising campaigns. A simple example might be: If many respondents report the same links for the previously mentioned low-energy car, it would be reasonable to emphasize the economical and environmentally friendly character of the car in a commercial.

7.

BRIEF HISTORY OF LADDERING RESEARCH: CONTRIBUTION AND CRITICISM

Since its introduction, laddering has become a widely applied method for accessing core constructs in personal meaning systems. Although this technique has proven its validity primarily for commercial goods (see for example Gutman, 1997; Gengler & Reynolds, 1995; Seth, Newman & Gross, 1991; Walker and Olson, 1991) and services (Botschen, Thelen &

Marketing of technological products

169

Pieters, 1999), to our knowledge only two studies have been published that examined the value of laddering in relation to environmental issues and technological products. The first contribution (Bagozzi & Dabholkar, 1994) successfully used the laddering procedure to uncover beliefs behind attitudes towards recycling. This study found that both altruistic and egoistic goals played a role, but that the latter dominated the means-end structure. In the second contribution (Van Gool, Breemhaar, Ester & Midden, 1997), laddering was used to study the means-end structure behind domestic energy consumption. Several consumer groups could be identified based on different consequence-value structures in relation to car use and use of the freezer. Perhaps even more important, it was found that alternative forms of transportation (e.g. biking) and the use of the freezer (e.g. tele-shopping) were more positively valued when these alternatives were important to personal motives. Although it seems essential to link products to the consumer in a meaningful way, there has been some criticism of the laddering technique. For example, the laddering technique has been criticized for its assumption that the results of the laddering approach represent cognitive representations of internal mental processes (Bagozzi & Dabholkar, 2000). In addition, Huey (1999) has criticized the laddering approach for its assumption of linearity.

8.

DISCUSSION

This chapter has focused on the principles of marketing and technological products. We have stressed the importance of understanding the consumer’s reaction to a technological product in order to promote the product effectively. We also tried to identify some central leads for shaping products and advertisements in order to serve the consumer’s needs better. First, we described two well-known advertising grids. According to the FCB planning model, buying behavior differs greatly with the degree of consumer involvement and thinking or feeling associated with the purchase decision. Depending on the type of purchase decision, advertising should focus on specific product information, feelings towards the product, habit formation or satisfaction. The Rossiter-Percy grid elaborated on the FCB planning grid. The most important difference was the distinction between informational and transformational purchase decisions. Both advertising grids stress the importance of analyzing purchase decisions. Such information should be the basis for developing an advertising strategy.

170

Chapter 16

Image management is another tool marketers have to shape products and advertising strategies. Consumers typically choose those products that best fulfill their needs. To the extent that a product is positioned as providing a superior value compared to other products in the same brand category, it gains competitive advantage. This can be done by stressing the importance of the product to the self-image of the consumer. Laddering is a research technique that helps a marketer gather data on the specific relationship between the product attributes, consumer consequences and personal values. The product attributes that are important to the consumer should be communicated through advertising. Although several authors have pointed out the importance of marketing technological products, it remains a largely unexploited domain. Very few attempts have been made to apply advertising grids and the means-end methodology to the development of communication strategies for technological products. This may be the result of difficulties that have been identified in attempting to translate marketing principles used for commercial products to less commercial products, like the products that are at stake in the context of this book. For example, because of environmental concerns it is sometimes necessary to persuade those people who have the most negative attitudes towards the technological product ʊ the opposite of the situation most commercial marketers encounter (Kotler & Alberto, 1989). Nevertheless, it is worth the effort to investigate the importance of marketing and advertising approaches in order to promote and communicate the value of technological products effectively.

REFERENCES Ajzen, I. (1991). The theory of planned behavior. Organizational Behavior and Human Decision Processes, 50, 197-211. Bagozzi, R.P. & Dabholkar, P.A. (2000). Discursive psychology: An alternative conceptual foundation to means-end chain theory. Psychology and Marketing, 17, 535-586. Botschen, G., Thelen, E.M., & Pieters, R. (1999). Using means-end structures for benefit segmentation. An application to services. European Journal of Marketing, 33, 38-58. Gengler, C.E., & Reynolds, T.J. (1995). Consumer understanding and advertising strategy: analysis and strategic translation of laddering data. Journal of Advertising Research, July/August, 19-33. Gutman, J. (1982). A means-end chain model based on consumers’ categorization processes. Journal of Marketing, 46, 60-72. Gutman, J. (1997). Means-end chains as goal hierarchies. Psychology and Marketing, 14, 545-560. Gool, W.A.C. van, Breemhaar, B., Ester, P., & Midden, C.J.H. (1997). Leefstijlen en duurzame gedragssubstitutie. In: A.E. Bronner, P. Ester, A.J. Olivier, W.F. van Raaij, M.

Marketing of technological products

171

Wedel, & B. Wierenga. Recente ontwikkelingen in het marktonderzoek. Jaarboek 1997 van de Nederlandse Vereniging voor Marktonderzoek en Informatiemanagement. Haarlem: Uitgeverij de Vriesebroch. Huey, B. (1999). Advertising’s double helix: A proposed new process model. Journal of Advertising Research, June, 43-51. Kotler, P. & Alberto, E.L. (1989). Social marketing: strategies for changing public behavior. New York: Free Press. Kover, A.J. & Abruzzo, J. (1993). The Rossiter-Percy grid and emotional response to advertising: an initial evaluation. Journal of Advertising Research, 21-27. McGuire, W.J. (1985). Attitudes and attitude change. In: G. Lindzey & E. Aronson (Eds.). The Handbook of Social Psychology, 2 (pp. 233-346). New York: Random House. Petty, R.E., & Cacioppo, J.T. (1986). The elaboration likelihood model of persuasion. In: L. Berkowitz (Ed.). Advances in experimental and social psychology: vol. 19. London: Academic Press. Petty, R.E. & Wegener, D.T. (1998). Attitude change: Multiple roles for persuasion variables. In: Gilbert, D.T., Fiske, S.T., & Lindzey, G. (Eds.). The handbook of social psychology, 4th ed., Vol. 1, pp. 323-390. Boston: McGraw-Hill. Pieters, R., Baumgartner, H., & D. Allen, D. (1995). A means-end chain approach to consumer goal structures. International Journal of Research in Marketing, 12, 227-244. Raaij, W.F., van (1994). Consumentengedrag en milieu. In: Midden, C.J.H., & Bartels, G.C (Eds.). Consument en milieu: beoordeling van milieurisico’s en sturing van gedrag. Houten: Bohn Stafleu Van Loghum. Reynolds, T.J. & Gutman, J. (1984). Advertising is image management. Journal of Advertising Research, 24, 27-37. Reynolds, T.J. & Gutman, J. (1988) Laddering theory, method, analysis, and interpretation. Journal of Advertising Research, 28, 11-31. Rokeach, M. (1973). The Nature of Human Values. New York: The Free Press. Rossiter, J.R. & Percy, L. (1987). Advertising and Promotion Management. New York: McGraw-Hill. Rossiter, J.R., Percy, L., & Donovan, R.J. (1991). A better advertising grid. Journal of Advertising Research, 31, 11-22. Seth, J.N., Newman, B.I., & Gross, B.L. (1991). Why we buy what we buy: A theory of consumption values. Journal of Business Research, 22, 159-170. Stienstra, J.P. (1998). De auto kan best een dagje zonder u: Een betekenisstructuuranalyse van automobiliteit. In: Bartels, G., Nelissen, W., & Ruelle, H. (Reds.). De transactionele overheid. Communicatie als instrument: zes thema’s in de overheidsvoorlichting. Kluwer Bedrijfsinformatie. Vaughn, R. (1980). How advertising works: A planning model. Journal of Advertising Research, 20, 27-33. Vaughn, R. (1986). How advertising works: A planning model revisited. Journal of Advertising Research, 26, 57-66. Walker, B.A. & Olson, J.C. (1991). Means-End Chains: Connecting Products With Self. Journal of Business Research, 22, 111-118.

Chapter 17 DIFFUSION OF TECHNOLOGICAL INNOVATIONS: Promoting the Large-Scale Use of Technology Trijntje Völlink, Ree M. Meertens and Cees J.H. Midden

1.

INTRODUCTION

This paper discusses the interaction between technology and behavior from the perspective of the individual decision to adopt or reject a technical innovation. It is only when energy conservation interventions are made available to the public by policymakers that they can contribute to energy conservation. Dutch energy distribution companies have agreed to make a substantial contribution to energy conservation policy. So-called ‘coordinators of Environmental Action Plans’ have a key position in these organizations, and have to decide, for example, whether a certain energy conservation innovation will be offered to households. Thus, before the diffusion of innovations among end-users can start, an innovation has to be adopted by these coordinators of environmental action plans. Many evaluation studies on interventions or technical innovations to reduce energy use by households were conducted in the late nineteenseventies and ‘eighties, after the first oil crisis. Although the results of these evaluation studies showed that psychological strategies can contribute to the motivation to adopt energy reducing activities in households (Winett & Neale, 1979; Baum & Singer, 1981; Cook & Berrenberg, 1981; McDougall, Gordon, Claxton, Ritchie & Anderson, 1981; Seligman & Hutton, 1981; Geller, 1989; Kempton, Darley & Stern, 1992; Cherulnik, 1993; Dwyer, Leeming, Cobern, Porter, Jackson, 1993), such interventions failed to come into general use. As might have been expected, neither consumers nor the 173 P.-P. Verbeek and A. Slob (eds.), User Behavior and Technology Development: Shaping Sustainable Relations Between Consumers and Technologies, 173-180. © 2006 Springer.

174

Chapter 17

responsible policymakers seem to use or adopt successful interventions automatically. The process of up-grading interventions from small-scale to widespread use can be conceptualized as the “diffusion of an innovation.” Rogers (1995) defines an innovation as an idea, practice or object that is perceived as new by an individual or another unit of adoption. Our study used Rogers’ diffusion theory to examine the diffusion of interventions in terms of technical, educational and/or motivational aspects among policymakers. According to Rogers (1995), the decision to adopt or reject an innovation is affected by five innovation attributes: observability, relative advantage, compatibility, trialability and complexity. The first four of these attributes are positively related to the rate of adoption, while complexity is assumed to be inversely related to the adoption rate (Rogers, 1995). The observability of an innovation is the degree to which the results of an innovation are visible to potential adopters. The relative advantage of an innovation is the degree to which an innovation is perceived as being superior to current practice or to the interventions it replaces. The compatibility of an innovation is the degree to which it is perceived to be consistent with social and cultural values and beliefs, previously introduced ideas and/or perceived needs. Trialability is the degree to which an innovation can be experimented with on a limited basis. Complexity is defined by Rogers as the degree to which an innovation is difficult to use and understand. Rogers’ ideas on the diffusion of innovations have had a tremendous impact on the field, and have placed diffusion theory in the spotlight in a variety of disciplines, like marketing, psychology and communication sciences. According to Rogers, 49 to 87 percent of the variance in the adoption rate is explained by observability, relative advantage, compatibility, trialability and complexity. He assumes that the relative importance of the attributes differs with the nature of the innovation, although he remains rather vague about the decision rules behind the assessment of the five attributes. If people are indeed, as Fiske and Taylor (1991) put it, “cognitive misers”, striving towards relatively simple and heuristic methods of decision-making, it seems very plausible that not all innovation attributes are rated before making the decision to reject or adopt an innovation. For instance, although a decision to adopt might be based on an assessment of all innovation attributes, the decision to reject an innovation might be based on less information processing (e.g. why bother about trialability if the relative advantage of the innovation is not very obvious?).

Diffusion of technological innovations

175

We hypothesized that the decision to adopt an energy conserving innovation is a stepwise process, in which a conjunctive decision rule is applied. The use of a conjunctive decision rule implies that a decision-maker is supposed to define minimum cut-off points for each attribute. If an alternative does not exceed the cut-off value for one or more of the attributes, its adoption will be rejected. In this model, high values for one attribute cannot compensate for a below cut-off value for another attribute (Bettman, 1979). We assumed that not all attributes are relevant to the potential adopter at the same time, so that the conjunctive decision rule is applied to one innovation attribute at a time. The decision process was presumed to stop as soon as a relevant innovation characteristic is rated below the minimum cut-off value (decision: no adoption), or when all innovation attributes have scored at least the minimum cut-off values (decision: adoption). In view of the goal of an energy conservation innovation, it was hypothesized that coordinators are above all interested in the relative advantages (such as cost-effectiveness) of such innovations. In a second step, an energy conservation innovation would be judged on its compatibility with existing structures. Figure 1 shows the stepwise model we proposed. The first two steps of this model were examined in our study, and are described below.

2.

METHODS

2.1

Participants

It is only when energy conservation interventions are made available to the public by policymakers that they can contribute to energy conservation policy. Dutch energy distribution companies have agreed to make a substantial contribution to energy conservation. Thus, before the diffusion of innovations among end-users can start, those staff members of the energy distribution companies who are responsible for this, i.e. must first adopt any innovation the coordinators of environmental action plans. At the time of our study, there were 42 different energy distribution companies in the Netherlands. All 42 coordinators of the Environmental Action Plan were asked to participate in our study. Forty-one agreed to take part in the study (98%), and thirty-five actually participated in all parts of the study (83%).

176

2.2

Chapter 17

Selection of successful energy conservation interventions

The selection of successful behavioral energy conservation interventions was based on a literature study, which revealed 35 relevant studies in household settings and 22 relevant studies in non-residential building and retail trade settings. These studies were evaluated on the following five criteria (see figure 17-1): (1) the design of the evaluation study had to be adequate: preferably a pretest-posttest design with a control group; (2) the innovation was not to have been implemented on a large scale in the Netherlands, (3) preference was given to innovations that had proved to be effective in more than one study, (4) the utility companies (or comparable organizations) had to be able to play a key role in the diffusion of the intervention, and (5) there had to be an indication that large-scale implementation of the intervention was cost-effective. Advantage

-

Decision to reject an innovation. The evaluation process stops

Compatibility

Evaluation process for other attributes (complexity, trialability, observability) Figure 17-1. Stepwise process of assessing an innovation on its attributes

Insofar as the information was available, the interventions were compared on the basis of these criteria for each individual sector

Diffusion of technological innovations

177

(households, non-residential buildings and retail trade). Table 17-1 shows the interventions that were finally selected.

2.3

Procedure and Instruments

A combination of qualitative and quantitative methods was used to collect the data. The questionnaire (a quantitative method) was used to find possible associations between four of the five innovation attributes proposed by Rogers (1995) (relative advantage, compatibility, complexity and trialability) and the intention to adopt an energy conservation intervention. Observability of the outcomes of an innovation was not measured, because observability of innovation results (e.g. the installation of an electronic indicator) is mainly relevant at the level of the end-user. Each item measuring the innovation attributes had five response categories, with values ranging from 1 (strongly disagree) to 5 (strongly agree). For each attribute, item responses were recorded when necessary, summed up and then divided by the number of items. The intention to adopt an energy conservation intervention was measured by the question: “Do you think this energy conservation intervention will be implemented in your organization in the future?” Table 17-1. Description of the four interventions selected Interventions

Description

1. Load management for washing machines

The utility company installs a switch that automatically switches off the washing machine. The goal of the intervention is to shift electricity use to the off-peak periods. This might eventually result in more evenly spread energy use. The utility company provides households with an electronic indicator, which compares daily energy use in the house with an individual conservation target, visualized on a display. The intervention is intended to stimulate tenants to agree with insulation measures. The utility company is responsible for the initiation and implementation of this intervention.

2. Electronic indicator providing feedback on the in-home energy consumption of households 3. Community-based communication intervention aimed at stimulating the adoption of insulation and energy saving behavior 4. Personalized advice on energy saving given to the retail trade by a consultant of the utility companies

The utility company is responsible for the provision of personalized energy advice to the retail trade by a consultant. The personalized advice consists of the following: (1) determining energy use, (2) differentiating energy use applications within a company, and (3) drawing up a report showing applicable energy conservation interventions.

178

Chapter 17

A qualitative method (face-to-face interviews) was used to find further evidence for the presence of a stepwise process, as well as possible clues for further specification and/or modification of the innovation attributes. Interviews were recorded and transcribed.

3.

RESULTS31

3.1

Analysis of the innovation adoption process

Although we had tried to select only interventions that were not widely used, we found that some companies or coordinators had already implemented one of the interventions. If this was the case, this coordinator was not included in the analysis for that specific intervention. The results for load management for washing machines show that 34% of the variance in the intention to adopt was explained by compatibility and complexity. For the Electronic indicator (N = 31), 27% of the variance was explained by compatibility, and for the community intervention (N = 26), 20% was explained by compatibility. One possible explanation for the absence of a relationship between some of the innovation attributes and the intention to adopt is the proposed stepwise process of judging an innovation on its characteristics. If it is true that potential adopters do not seriously assess the next attributes if the evaluation of the previous attribute is negative, one is not likely to find a relationship between the intention to adopt and these next attributes. It was hypothesized that the correlation between compatibility and the intention to adopt is only significant if the perceived advantage is high. If the perceived advantage is minor, further evaluation of an innovation will not take place. In order to check whether advantage affects the relationship between compatibility and the intention to adopt, an interaction variable was added to the regression analysis. The analysis indicated that the interaction between compatibility and advantage was significant for “community intervention” after correction for the main effects of compatibility and advantage. The regression analysis for “electronic feedback” revealed that the interaction variable was marginally significant 31

To enhance accessibility for a broad readership, this results section does not contain numerical statistical information. Those readers who are interested in the statistical details are requested to contact the authors.

Diffusion of technological innovations

179

In order to find out whether compatibility only predicts the intention to adopt in a situation where the perceived advantage is great, separate regression analyses were carried out for respondents with a high score on advantage and for those with a low score on advantage. In the group of respondents with a high score on advantage, compatibility explained 44% (electronic indicator, M > 2.8, n = 18) and 52% (community intervention, M > 3.4, n = 17) of the difference in the intention to adopt, whereas in the group of respondents with a lower score on advantage, compatibility did not have an effect. For two of the four interventions, these results support our hypothesis of the existence of a stepwise evaluation process. The idea of a stepwise process was also confirmed by the analysis of the arguments for rejecting or adopting a particular energy conservation intervention mentioned by coordinators in the face-to-face interviews.

4.

DISCUSSION AND CONCLUSIONS

The results of the present study reveal that perceived compatibility is a general and important predictor of the intention to adopt energy conservation interventions. Except for compatibility, Rogers’ innovation attributes do not seem to be very strong predictors of the intention to adopt energy conservation interventions. We suggest that the hypothesized stepwise process is one of the reasons for this low predictive value. The results suggest that potential adopters seem mainly interested in the advantage of an innovation. Only if the advantage is considered sufficiently great will they proceed to evaluate the intervention on the basis of the other attributes. If the advantage is perceived as minor, the evaluation process is stopped. The results regarding the electronic indicator and the community intervention suggest that the coordinators continued to evaluate the innovations for compatibility only when the outcome of the evaluation on advantage was positive. Our hypothesis was not confirmed for intervention 1 (load management of washing machines) or intervention 4 (personalized advice on energy savings given to the retail trade by a consultant of the energy distribution companies). A possible explanation might be that the sequence of innovation attributes in the stepwise assessment process may be different for different types of innovation, depending on the characteristics of these innovations. Future research should address this question. The results do suggest, however, that decisions about innovations are made in a stepwise process in which advantage is the first critical attribute for continuing or discontinuing the assessment. This finding suggests that

180

Chapter 17

practitioners who try to promote the diffusion of a particular innovation should start by convincing potential adopters of the advantages of the innovation. Only when they are convinced of the advantages of a technical innovation, will they be interested in its other aspects.

REFERENCES Baum, A., & Singer, J.E. (Eds.). (1981). Advances in environmental psychology. Volume 3 Energy: psychological perspectives. Hillsdale, New Jersey: Lawrence Erlbaum Associates. Bettman, J.R. (1979). An information processing theory of consumer choice. Reading: Addison Wesley Publishing Company, pp. 173-228. Cherulnik, P.D. (1993). Applications of environment-behavior research: case studies and analysis. New York: Cambridge University. Cook, S.W., & Berrenberg, J.L. (1981). Approaches to encouraging conservation behavior: a review and conceptual framework. Journal of Social Issues, 37, 73-107. Dwyer, W.O., Leeming, F.C., Cobern, M.K., Porter, B.E., & Jackson, J.M. (1993). Critical review of behavioral interventions to preserve the environment. Environment and Behavior, 25, 275-321. Fiske, S.T., & Taylor, S.E. (1991) Social Cognition, New York; McGraw-Hill. Geller, E.S. (1989). Applied behavior analysis and social marketing: an integration for environmental preservation. Journal of Social Issues, 45, 17-36. Kempton, W., Darley, J.M., & Stern, P.C. (1992). Psychological research for the new energy problems: strategies and opportunities. American Psychologist, 47, 1213-1223. McDougall, G.H., Gordon, H., Claxton, J.D., Ritchie, J.R., & Anderson, C.D. (1981). Consumer energy research. Journal of Consumer Research, 8, 343-354. Rogers, E.M. (1995). Diffusion of innovations (3rd & 4th ed.), New York: The Free Press. Seligman, C., & Hutton, R.B. (1981). Evaluating energy conservation programs. Journal of Social Issues, 2, 51-72. Winett, R.A., & Neale, M.S. (1979). Psychological framework for energy conservation in buildings: strategies, outcomes, directions. Energy and Buildings, 2, 101-116.

Chapter 18 TECHNOLOGICAL INNOVATIONS AND ENERGY CONSERVATION: Satisfaction With and Effectiveness of an In-Business Control System Nicole van Kesteren, Ree M. Meertens and Mirjam Fransen

1.

INTRODUCTION

Government and consumers’ organizations agree that reducing energy consumption is necessary in domestic as well as in business and government sectors. The consumers in these different sectors must achieve energy savings through, for example, more energy-conscious behavior. Their behavior can be influenced by various behavioral change strategies, such as information provision, methods to increase involvement, prompts and systematic experience with behavior, followed by feedback, confirmation or rewards (Becker, 1978; Van Houwelingen & Van Raaij, 1989; Winnet & Ester, 1983; see McCalley, 1998 for an overview). However, promoting efficient energy consumption is not the only avenue for resolving energy concerns. Increasingly, solutions are sought at the technical level. This is called efficiency enlargement (Midden & Bartels, 1994). Technical solutions to promote energy savings seem to offer specific advantages over behavioral interventions. After all, technical interventions function primarily without the interference of human behavior. This does not mean that the way the technical innovation is used by the consumer has no influence on energy savings. Several studies in households indicate that a high variance in energy consumption is directly related to the consumption patterns of individual consumers in the same technical environments (Shippee, 1980). Therefore the influence of technical measures on energy 181 P.-P. Verbeek and A. Slob (eds.), User Behavior and Technology Development: Shaping Sustainable Relations Between Consumers and Technologies, 181-189. © 2006 Springer.

182

Chapter 18

consumption also appears to be dependent on consumers’ behavior. Furthermore, the effects of certain technical improvements could be undermined because individuals find ways to avoid or eliminate technical measures when these cause inconvenience or are seen to threaten individual freedom (March & Collet, 1986; Van Vugt, Meertens & Van Lange, 1995; Tertoolen, Van Kreveld & Verstraten, 1998). When a person is not free to behave as he or she wants, a motivational state called ‘psychological reactance’ arises (Brehm, 1966), leading to (sometimes rebellious) actions to restore freedom. Another way in which the effects of technical solutions could be undermined is ‘compensating behavior’. This means that people compensate for their energy saving by spending more energy in another domain (the so-called rebound effect, for example leaving PCs and radios on because people work in such energy-efficient buildings). The aim of this study was to evaluate whether an advanced in-business control system in a large office block in Den Bosch, The Netherlands, is effective in reducing energy consumption, and whether this effect is influenced by attitude, motivation and psychological reactance. The control system manages the lighting and strives for both efficient energy consumption and a comfortable indoor environment. One of the features of this control system is a regulator by which employees could establish a certain lighting level between 20 lux and 800 lux (the standard is 500 lux). The system also contains daylight-dependent regulating equipment that manages the established lighting levels: when there is more daylight the level of the artificial light is dimmed, and vice versa. Another feature is an absence detector: when the employee is absent for more than 15 minutes, the artificial light is automatically switched off. For the benefit of the study, the system was introduced in phases. Consumers had access to the advanced features of the system only in the experimental group. The consumers in the control group had to accept the conventional adjustment of the control system, as in most other business settings. Because consumers in the experimental group could not select the advanced equipment of the inbusiness control system, and because certain energy-related acts were automatically controlled by the system, a restriction of freedom of choice and loss of control of the consumers’ environment was expected. Reactance could be a consequence. Different hypotheses were considered. On the basis of the advanced technical qualities of the control system, it was expected that in the experimental group less energy related to artificial light would be used than in the control group, in spite of the ability of users in the experimental group to adjust their lighting level above the standard of 500 lux. Because of

Technological innovations and energy conservation

183

practical difficulties we were unable to use a design with a pre-test during which both the experimental group and the control group were limited to the conventional adjustment of the control system. As a consequence, in this study energy savings could be defined only in relation to the control group. Therefore, the term ‘energy savings’ refers, implicitly or explicitly, to a comparison between the experimental and the control group. Second, consumer satisfaction with the technical equipment, as well as their level of satisfaction when dealing with the system for a long period, was also investigated. There were no specific expectations regarding possible differences in satisfaction between the experimental and control group. Third, the study examined the extent to which the realized savings were connected with the consumers’ attitudes towards energy savings. Various studies have shown that attitude is a good predictor of actual energy-saving behavior if the measurement of the attitude is specific enough (Cunningham & Lopreato, 1977; Heberlein & Black, 1976; Farhar et al., 1980; Milstein, 1977; Tashchian & Slama, 1985; Seligman, 1986). Based on these findings, the following hypothesis was formulated: The more positive the consumers’ attitudes towards energy savings, the less energy will be consumed in the experimental as well as in the control group. The effect of the consumers’ motivations on energy savings was also measured. This led to the following hypothesis: The more motivated consumers are to save energy, the less the energy consumption will be in both the experimental and the control group. No hypotheses were formulated with regard to differences in attitude and motivation between the experimental and the control group. The relationship between psychological reactance and energy consumption was measured in two ways. First, we assessed the resistance of users to the in-business control system and the consequent effect of this resistance on energy conservation. More resistance was expected in the experimental group than in the control group. Second, psychological reactance was measured as a personal characteristic (Hong & Page, 1989; Hong, 1992). When consumers in the experimental group attach greater value to their personal freedom, and their motivation to restore their personal freedom is greater when this personal freedom is threatened, they can be expected to use more energy. No such effect was expected in the control group. Finally, we investigated whether a number of factors, for example factors that were related to the presence or absence of users, influenced light-related

Chapter 18

184

energy consumption. After all, the absence detector would lead to more energy savings while the consumers are absent during the day. For this reason, a correction was made for the function of the user and the number of people that share the room. Additionally, the side of the building (north, east, south or west) one is working in would be expected to influence the effect in energy savings. The energy consumption on the south side of the building could be smaller because of the greater amount of daylight.

2.

METHODOLOGY

2.1

Design, procedure and respondents

The study consisted of an experimental design with two post-tests. The 430 rooms in the building were randomly assigned to an experimental group (advanced adjustment of the in-business control system) and a control group (conventional adjustment of the in-business control system). A stratification based on function, number of people in a room, and building orientation (north, east, south, west) was made in advance. Because only nine rooms were oriented at the northern side of the building, and no energy data on these rooms was available, we could investigate the influence only of the southern, eastern and western sides of the building. During the entire study period (October 1998 – July 1999) the artificial light was continuously measured in all 430 rooms. On the basis of these measurements a reliable estimation could be made of the related energy consumption. Every room received two questionnaires. In November 1998, a first questionnaire (T1) was sent to all 430 rooms. The person who spent the most time in the room was asked to answer the questionnaire. A total of 284 questionnaires (66 percent) were returned. In April 1999, these respondents received a second questionnaire (T2), which had a response rate of 80 percent (196 questionnaires). The study population in T1 consisted of respondents belonging to the legal profession (35 percent), registrars (25 percent), administrative staff (18 percent), other employees (16 percent), and judicial employees (6 percent). The study population was predominantly male (56 percent), had an average age of 40 years, and shared the room with one other person.

Technological innovations and energy conservation

2.2

185

The questionnaire

The questionnaire measured satisfaction with the in-business control system, attitude towards energy savings, motivation to save energy, resistance to the in-business control system, and psychological reactance. Satisfaction with the in-business control system, attitude towards energy savings, and motivation of the respondents were measured on five-point scales (Cronbach’s alpha at T1 = .67, .87, .86, respectively. Cronbach’s alpha at T2 = .65, .91, .87, respectively). Respondents were asked how important they think it is to save energy, or to what extent they are motivated to save energy. To measure the resistance to the in-business control system, respondents were asked if they had taken action (and what kind of action) to mitigate the disadvantages of the system. Psychological reactance was measured by the ‘Psychological Reactance Scale’ (Hong, 1992; Hong & Page, 1989), and was translated into Dutch for the benefit of this study. Respondents were asked to mark on a five-point scale whether they agreed or disagreed with 14 statements (Cronbach’s alpha = .84). 32

3.

RESULTS

3.1

Energy consumption

To study the energy savings of the in-business control system, the consumption of artificial light was analyzed. The results indicated a significant difference in energy consumption due to lighting between the experimental group and the control group (Mexp = 175 MWh and Mcon = 277 MWh). In the experimental group, 37 percent less energy was used compared to the control group.

3.2

Satisfaction with the in-business control system

On average, users of the system were neither satisfied nor dissatisfied with the control system (see Table 18-1). Satisfaction did not differ between 32

To enhance accessibility for a broad readership, this results section does not contain numerical statistical information. Those readers who are interested in the statistical details are requested to contact the authors.

186

Chapter 18

the experimental and the control group. Users were however more satisfied with the system at T2 than at T1. Furthermore, 35.5 percent of the respondents were dissatisfied or very dissatisfied with the control system.

3.3

Resistance against the advanced in-business control system

The users in the experimental group did not resist the control system in greater numbers than users in the control group. Figure 18-1 illustrates that the majority of users did not take any action to mitigate the adverse effects of the system. There was no effect of condition, and no difference between before and after. A frequency analysis showed that 51 of the 271 users took action to moderate the disadvantages of the system. Most of them took control over the artificial light themselves and reported complaints at the service desk. Only five users mentioned more direct actions against the inbusiness control system. For example: taping over the sensor on the daylight-dependent regulating equipment so the artificial lighting level was not affected by daylight; putting a book on the button of the daylightdependent regulating equipment so the artificial light continued burning at full strength; and using a desk lamp.

3.4

Occupation, number of people in the room and building orientation

The influence of the occupation of the respondent on energy savings was examined. Results showed a significant effect of function and of condition. The interaction between occupation and condition appeared not to be significant. The effect of numbers of people in a room, and the effect of condition appeared to be significant, the effect of occupation was not significant anymore. Finally, the influence of the building orientation and condition on energy savings was examined. These factors didn’t show any effect.

3.5

Attitude, motivation, resistance and psychological reactance

To test the hypotheses that people use less energy when they have a positive attitude towards energy savings and when they are motivated to save energy, and that they use more energy when they score higher on the psychological reactance scale, a hierarchic regression analysis was performed. Condition, number of people in a room, attitude, motivation,

Technological innovations and energy conservation

187

resistance, psychological reactance, and interaction terms were considered to be predictors of energy consumption. The analysis showed that both condition as well as number of people in a room contributed to the prediction of the savings effect. Neither the main effects of attitude, condition, resistance and psychological reactance, nor the interaction terms contributed significantly to the variance.

4.

SUMMARY OF RESULTS AND DISCUSSION

The most important questions of this study were: what is the savings effect of the in-business control system, and to what extent is this effect influenced by attitude, motivation, resistance and psychological reactance? The control system is one possible technical solution to promote energy savings. Because technical solutions are partially independent of human behavior, the expectations of the system in terms of energy savings were quite high. The effect of the system on energy savings was examined in this study. The experimental group was expected to use less energy than the control group. Results show that the control system did lead to a reduction of 37 percent in energy consumption in the experimental group, as compared with the control group. Second, user satisfaction with the system was studied. The results indicated no differences in satisfaction between the experimental and the control group. On average, the users of the system were neither satisfied nor dissatisfied with the control system. The users even seemed to be more satisfied with the system at T2 than at T1. A frequency analysis showed that 35.5% of the respondents were dissatisfied or very dissatisfied with the control system. Third, the amount of actions to mitigate the disadvantages of the system was examined. The results did not support the expectation that the resistance in the experimental group would be higher than in the control group. This could be attributed to the fact that users in the experimental group did not feel they had control, possibly because they could regulate the lighting level themselves. Fourth, the extent to which energy consumption can be predicted by function, number of people in the room, and building orientation, was studied. The only significant predictor of energy savings was the number of people in the room. Finally, the extent to which energy consumption is influenced by attitude, motivation, resistance and psychological reactance was examined. The hypothesis that attitude and motivation are positively related to energy savings and energy consumption, and the hypothesis that psychological

188

Chapter 18

reactance is negatively related to energy savings, were not supported by the results. Other studies, however, have demonstrated that psychological reactance could undermine technical measures (Marsch & Collet, 1986; Tertoolen et al., 1998; Van Vugt et al., 1995). Perhaps in this study the respondents experienced the in-business control system as a convenience and not as a restriction on personal freedom. Previous studies have also demonstrated the significance of attitude in energy consumption (Cunningham & Lopreato, 1977; Heberlein & Black, 1976; Farhar et al., 1980; Milstein, 1977; Tashchian & Slama, 1985; Seligman, 1986). It should, however, be noted that these studies were usually implemented in domestic settings and were not related to the implementation of a technical solution. In this situation, there appeared to be no relationship between attitude and energy consumption in either the experimental or the control group. It can be concluded that technical solutions for energy problems in the workplace show great promise. In addition, on the basis of this study, it can be concluded that it would be beneficial from an energy point of view to implement in-business control systems on a large scale. There are no limitations in regard to building orientation, function of the building’s occupants, or their attitudes towards energy savings. According to the current study, these variables play an insignificant role in achieving energy savings.

REFERENCES Becker, L.J. (1978). Joint effect of feedback and goal setting on performance: A field study of residential energy conservation, Journal of Applied Psychology: 63, 428-433. Brehm, J.W. (1966). A Theory of Psychological Reactance. Academic Press, New York. Cunningham, W. & Lapreato, S.C. (1977). Energy Use and Conservation Incentives. New York: reager Publishers. Farhar, B.C., Unseld, C.T., Vories, R. & Crews, R. (1980). Public opinion about energy, Annual Review of Energy: 5, 141-172. Heberlein, T.A. & Black, J.S. (1976). Attitudinal specificity and the prediction of behavior in a field setting, Journal of Personality and Social Psychology: 33, 474-479. Hong, S.M. (1992). Hong’s psychological reactance scale: a further factor in analytic validation, Psychological Reports: 70, 512-514. Hong, S.M. & Page, S. (1989). A psychological reactance scale: development, factor structure and reliability, Psychological Reports: 64, 1323-1326. Houwelingen, van, J.H. & van Raaij, W.F. (1989). The effect of goal-setting and daily electronic feedback on in-home energy use, Journal of Consumer Research: 16, 98-105. Marsh, P. & Collet, P. (1986). Driving Passion: The Psychology of the Car. Jonathan Cape, London.

Technological innovations and energy conservation

189

Midden, C.J.H. & Bartels, G.C. (1994). Het milieu, een kwestie van mensenwerk. In: C.J.H. Midden & G.C. Bartels. (Eds.). Consument en Milieu. Houtem/Zaventem: Bohn Stafleu Van Loghum. Milstein, J.S. (1977). Attitudes, knowledge and behavior of American consumers regarding energy conservation with some implications of governmental action, In: W.D. Perrault, Jr. (Ed.) Advances in Consumer Research: 4, 315-321. Ann Arbor. MI: Association for Consumer Research. Seligman, C. (1986). Energy consumption, attitudes and behavior, In: M.J. Saks & L. Saxe (Eds.) Advances in applied social psychology: 3, 153-180. Hillsdale, NJ: Erlbaum. Shippee, G. (1980). Energy consumption and conservation psychology: a review and conceptual analysis, Environmental Management: 4, 297-314. Tashchian, R.O. & Slama, M.E. (1985). Survey data on attitudes and behaviors relevant to energy: Implications for policy, Marriage and Family Review: 9, 29-52. Tertoolen, G., van Kreveld, D. & Verstraten, B. (1998). Psychological resistance against attempts to reduce private car use, Transportation Research: 342, 171-181. Winett, R.A. & Ester, P. (1983). Behavioral science and energy conservation: conceptualisations, strategies, outcomes, energy policy applications. Elsevier Science Publishers B.V., 203-229. Vugt, M. van, Meertens, R.M. & van Lange, P.A.M. (1995). Car versus public transportation? The role of social value orientations in a real-live social dilemma, Journal of Applied Social Psychology: 25, 258-278. Zajonc, R.C. (1968). Attitudinal effects of mere exposure, Journal of Personality and Social Psychology: 9, 1-27.

Chapter 19 SUSTAINABLE TECHNOLOGY OR SUSTAINABLE USERS?

Cees J.H. Midden

1.

INTRODUCTION

The main question posed in this book is how to manage humantechnology interaction to optimize its various outcomes. Sustainable use and a sustainable society are focal points in this analysis. In the traditional technical approach, solutions for sustainability are looked for in technology: for example, a washing machine is made more energy-efficient when it uses less water and suffers less heat loss. These kinds of measures have been proposed repeatedly in past decades. Although these improvements certainly have merit, it has become evident that consumption levels cannot be reduced sufficiently by this approach. The simple reason is that it does not deal with the intensity and way of use. An increment in use may compensate for efficiency gains. Neither does it deal with inefficient behavior by the user. These constraints caused by users have made engineers long for full automation, assuming that by excluding the user from the operational process, the efficiency, for example of a washing machine, can be optimized. One might wonder if such a setup would really lead to more sustainable use. In this part of the book we focus on the interaction between technology and users, and show how outcomes in technology are shaped by this interaction, how the user affects the system, and what roles are being played by the system and by the user. A sustainable use of technological systems can only be accomplished if we understand better the nature of this interaction 191 P.-P. Verbeek and A. Slob (eds.), User Behavior and Technology Development: Shaping Sustainable Relations Between Consumers and Technologies, 191-200. © 2006 Springer.

192

Chapter 19

process. Consistent with the map outlined in Part 1, different modes of human-technology interaction can be distinguished. The papers in this part of the book have shed some light on these distinctions, not only at the conceptual level, but by offering empirical data as well. First of all, technological efficacy requires that the user be willing to allocate control to the system.

2.

WHO IS IN CHARGE?

Our daily life is increasingly characterized by the use of complex automated systems. Human tasks are taken over by self-regulatory systems. We see this in all domains of everyday life, such as car driving, watching TV, doing the laundry, regulating our climate systems at home, electronic shopping, etc. The same is true for many professional environments. People seem motivated to reduce personal efforts by implementing automated systems. Automation promises performance benefits, makes life easier, may reduce errors, and improve safety. However, automation is not always beneficial for the user. Automated systems make errors and often cannot completely deal with complex tasks, for example, for optimizing heating comfort. Trust is not always rightly calibrated (Lee & Moray, 1992). Trust is a mental state that allows a person to delegate a risky judgment or choice to another person, an organization or a system. The trust choice implies that the entrusted entity is expected to have the intention and capacity to achieve positive outcomes for the person, and/or to prevent negative outcomes. People may overestimate their own abilities compared to the system, leading to sub optimal choices, slowness, fatigue and other disadvantages. However, too much trust may also happen, leading to lowered attention to errors and decreased situational awareness, reducing the opportunities to intervene effectively. A reduced level of self-confidence may be a consequence. It does not seem very rational to have no trust at all, but the same is true for blind trust in a system. The question can be raised how to get at a proper balance between trust and mistrust, and which information and opportunities a system should offer to allow monitoring and eventual intervention? In a more general sense we may wonder about a proper division of tasks between user and system.

Sustainable technology or sustainable users?

193

The question of trust and control delegation is important because in many cases the user will have the choice between use or not-use. Systems that create a conflict with the needs and goals of the user will most likely fail. Users will simply tend to avoid systems that force an individual into an unwanted type of use. If the system is not forcing the user into a certain way of use, the choice will be for the user, who will only be willing to delegate control on the expectation that his goals of use will be achieved. Another phenomenon that reduces the effectiveness of the technical approach is what has been called the rebound effect. This effect describes how users change their use because of efficiency gains. The effect occurs when technological improvements increase efficiency. This efficiency might, for example, lead to lower energy consumption. However, such effects often do not occur. Efficiency gains appear to be converted into more consumption. In other words, the positive effects on sustainability are missed because consumers take more for the same costs. The enhanced use of highefficiency light bulbs clearly exemplifies this phenomenon. The effect can also cross domain boundaries, for example when lower costs for heating create budgets for more air travel. Rebound effects clearly require more attention in the user-system interaction. The aforementioned issues raise questions about how user behavior is shaped in this interaction, how the user affects the system, and what roles are played by the system and by the user. A sustainable use of technological systems can only be accomplished if we understand better the nature of this interaction process. Different modes of human-technology interaction can be distinguished. The papers in this part of the book have shed some light on these distinctions. In addition to issues of how people interact with technology in the use phase, we should address the issue of system adoption. Before even entering the phase of use, a system has to be adopted that addresses strategic decisions about purchase and use. In the following sections we will reflect on the different contributions and identify different perspectives of user-system interaction. Subsequently, we will discuss how the user can be a moderator of technological impacts on sustainability; how technology may impact on user behavior or even may enforce it, how the system may act as an information source persuading the user, and how adoption barriers may prevent system adoption at all.

Chapter 19

194

3.

THE USER AS MODERATOR OF TECHNOLOGICAL EFFECTS

Hendrickx & Schoot Uyterkamp discussed in Chapter 10 how the user makes personal choices based on the opportunities that the system and the environment allow, of course optimizing personal goals and minimizing efforts. This model shows how the user can modify the impact of technological change. The model supposes that environmental gains can be lost because the user intervenes through his or her decisions, for example when making a purchase or during use. Various models can be applied to describe the moderating role of the user, such as TRA, MAU, and Mean End Chain theory. The common notion is that behavioral options will be evaluated in terms of a set of value dimensions relevant to the user. An example is the introduction of a more efficient engine that makes less noise. This environmental improvement may also change however the comfort level of the driver, which may allow for a faster driving style with negative environmental effects. In that chapter, the authors suggest that the behavioral side effects of technical innovations should be assessed at a small-scale level before a large-scale introduction. Perhaps even more important is a need for further theorizing on the mechanisms that cause technological changes to induce behavioral side effects.

4.

USER IMPACTS THROUGH THE SOCIOTECHNICAL ENVIRONMENT

In the paper by Spaargaren, Martens and Beckers, the role of the sociotechnical environment is outlined, following Giddens’ Structuration theory (Giddens, 1984). The paper opposes a purely technological fix and also criticizes the social-psychological approach that focuses on attitudes, values and intentions. It claims a position in between the technological and the social-psychological approach, combining the notion of knowledgeable actors with the influence of the social and technological environment. The paper emphasizes rightfully that the socio-technical environment fosters constraints but also opportunities to enhance actions. It introduces the concept of ‘social practice’, which can be understood as a certain routine in a behavioral domain, such as for example feeding or having holidays. Social practices correspond to segments of person’s lifestyle. Expressions of these practices, for example with regard to the environment, may flow across

Sustainable technology or sustainable users?

195

domains. One may wonder about the nature of this process of practices spreading out. Are these different segments supported by the same underlying value structure? Do we generalize our actions from one domain to another? The model also identifies a ‘system of provision’ as a set of rules and resources that help shape and enable the social practice. Technology is part of the resources, next to the regulatory mechanisms that control the user. The system of provision can be related to the user in different phases of an innovation process, such as design, production, distribution, access and use. This suggests that a social practice may change automatically by changing the technological resources in the system of provision. The approach identifies a transitional process that starts with ‘deroutinisation’, and then goes on with the identification of heuristics, the definition of ‘slots’. It ends with (re) routinisation. The merit of the social practice model is that it articulates that the sociotechnical environment may influence a person’s behavioral choices. In addition, it specifies how this influence may occur at different stages of innovation diffusion. The chapter by Derijcke and Uitzinger offers two interesting case studies showing how technological changes may change user behaviors, because the contextual opportunities and contingencies change. These effects can occur, even if not planned in advance, by the user. On the other hand, these case studies also suggest that behavioral change may not occur, as long as the user has no awareness of technical opportunities, for example a toilet flush stop that reduces water consumption. These findings raise questions about how the linkage between the technological context and the users’ behavior operates, and which processes moderate effects. Adopting a mainly sociological perspective and a focus group methodology, the ‘social practice’ approach does not specify the psychological nature of these mechanisms. Although the knowledge of the user is emphasized, the authors assume that everyday life is routinized to a considerable extent, not discursively or not reflexively. However, the model does not explain what these routines mean specifically for the interaction. One may wonder about the functions of user knowledge; how the assessment of the ‘system of provision’ evolves; whether these routines should be seen as inescapable, one-to-one occurrences; whether users are aware of their choices and if these habits are perhaps intentional and goal-dependent and, thus, still under the control of the person.

Chapter 19

196

In spite of criticisms of the psychological paradigm that predicts behavior from beliefs and values, the social practice approach also refers to evaluative processes that occur in the product adoption process, which closely resembles an attitudinal approach, e.g. an efficient light bulb produces too white a light, which may limit its market potential. This also underlines that the interaction between user and technical systems cannot be understood from one theoretical perspective. Multiple processes are at work.

5.

PERSUADING THE USER THROUGH TECHNOLOGY MEDIATED FEEDBACK

A more precise, though partial, picture of the psychological processes mentioned in the preceding section is offered in the studies by McCalley & Midden and Vollink & Meertens. They model the interaction between system and user as a dialogue in which the system supplies the user with relevant information, which is taken into account by the user to make a decision about use. In other words, in this setup the user is approached as a conscious person making trade-offs and choices based on expected outcomes. The system functions as a change agent that can make certain outcomes more salient. It can frame outcomes differently, or encourage the user to make certain goals more explicit. This idea has correspondence with the alterity mode as suggested by Ihde (see chapter 6). The script concept, however, seems less volatile than the feedback approach presented here. The feedback approach does not exclude that certain choices will be curtailed and partially or completely routinized after a certain amount of experience. One might say that the system persuades the user in a certain sense to make more energy-efficient choices. The mainly experimental research shows that feedback information is really effective if it fits with the goals, which are active for the user. If the feedback is irrelevant, because other goals are active, it will not lead to behavioral change. If the feedback is not presented on the task level, it may inhibit task performance, because it shifts attention away from the task. This approach has some complementarity with the social practice approach. The feedback model does not specify explicitly how the socio-technical context impacts on user behavior. However, there is a link, because goals and feedback are usually rooted in the socio-technical context that defines which goals can be put forward and which effects may occur. Basically, the feedback approach supposes a user, who is able and motivated to process relevant information.

Sustainable technology or sustainable users?

197

One may question its effectiveness when behaviors are heavily routinized. It can be predicted that persons while conducting automated tasks will not be open to new information. In such situations, the feedback has to attract attention before it can be processed and converted into action. This may require special procedures. Various laboratory and field studies (for an overview, see, for example McCalley & Midden, 2002) have demonstrated that properly specified and design feedback procedures can generate substantial savings and efficiency improvements.

6.

THE ‘AUTOMATED’ USER OF TECHNOLOGY

The role of routines in human-technology interaction has been discussed in more detail in the chapter by Heijs. The idea of a habit is that, based on frequent and consistent performance in the past, a person can develop behavioral acts that are automatically initiated. The habit is not the behavior itself, but, as Heijs also formulates it, it is a mental structure in which an association exists between a situation, a goal, and a cue. This means that, in a certain context, cues may occur that trigger the behavior directly, not preceded by an elaborate choice process. The concept does not mean that intentions and goals are irrelevant. On the contrary, it has been shown that goals are needed to really initiate the action in the presence of a relevant cue (Aarts, 1996; van de Berg, 2002). It also does not imply that the behavior itself is completely automatic, as is the case with skill-based behavior (e.g. bicycling). The habit is characterized by the automatic initiation of an action. A decision-action sequence can become habitual in various ways. One possibility is that the intention to perform an act, and also the execution of the act, are automatic ʊ for example, taking a certain route to the office. Another possibility, however, is that the intention to act is conscious, but that the execution is largely automatic ʊ for example, when cleaning the car. Heijs discusses how the technological environment may play a role as the environment in which the behavior is embedded. Technology can have a cueing function, e.g. action is needed in case of a certain signal. Changing the signal may influence the activation of the habitual behavior. Changing the technological context may also prevent the habit from being executed, encouraging the user to reconsider behavioral options. Finally, technology might mitigate the effects of bad habits, e.g. by making appliances ‘foolproof’. Feedback procedures and also feed-forward procedures may change habits, provided that sufficient attention is given to the feedback. The concept of habit implies that the environment plays an important role in the

198

Chapter 19

development of behavioral patterns. However, these patterns are not automatically induced by the context. Usually, the user develops a behavioral pattern that relates to the present contingencies and promises of positive outcomes. The habit will be consolidated if in subsequent feedback loops these expectations are confirmed. Learning new behaviors may be inhibited by available habits. New intentions may be overruled at the moment that contextual cues automatically activate habitually-based intentions and actions. For example, a person decides in the morning to go to the office by bicycle, but on leaving the house, the habitual intention to take the car gets activated, suppressing the bike intention, finally making the person go to the office by car. In sum, it can be concluded that habits can operate without many of the cognitive efforts that precede reasoned action, but that on the other hand the habit is not a process occurring completely out of the consciousness and control of the person. Monitoring processes will continue and may cause the person to resist the habit and executing another intention or even change the habit itself, because it does not serve the user’s goals anymore.

7.

ADOPTING TECHNOLOGICAL PRODUCTS

So far, we have mainly discussed issues of how people interact with technology in the use phase, and barriers to change. Before even entering this phase a system has to be adopted, which refers to strategic decisions of purchase and use. This awareness is important, because it emphasizes that the user/consumer role is of central importance, even when technological means can be used to induce environmental change or even to create contexts in which people are encouraged to behave in environmentally conscious ways. The papers by Vollink, Meertens & Midden and van Kesteren & Meertens have explained some of the complexities of these adoption processes. The Vollink et al., paper focuses on the decision process of a potential buyer, showing that these decisions are not simple one-in-a-time decisions. Rather, these decisions appear to be taken step-by-step. Consumers will first evaluate some crucial attributes, in particular any advantages of the innovation, before other characteristics and considerations enter the decision process. Finally, we want to point at evaluative processes in the use phase. Van Kesteren & Meertens point to the detrimental effects of reactance which

Sustainable technology or sustainable users?

199

occur when a user feels frustrated or annoyed during use because the system behaves inadequately, or at least not according to expectations and the users’ liking. This phenomenon is also relevant to other approaches discussed here. We already suggested how reactance and consumer refusal might counteract attempts to change social practices, due to environmental changes. Also in the feedback approach reactance may play an important role, for example when the user gets irritated because the feedback is not appreciated or unpleasantly presented. Appliances with an overbearing moral dimension will annoy many people, by telling them how to behave. Designers of new systems should realize that in the end it is the user who decides about using or not using.

8.

CONCLUSION

In this discussion we have outlined, in five sections, different approaches of human-technology interaction. Each of these approaches offers a distinct perspective and shows specific abilities to analyze how interactions between humans and technological systems may influence environmental outcomes. One basic distinction that can be identified is the direction of influence. The user may moderate the effects of technological improvements. Also, technology as embedded in appliances, systems and environments, may shape the ecological impacts of human behavior. The models differ in how the relation between technology and user should be described. The social practice model assumes more environmental determinism than the feedback approach or the adoption approach. However, these differences are certainly not absolute. Social practices acknowledge the role of the knowledgeable consumer/user and their evaluative potentials. On the other hand, the feedback approach identifies effects, which are embedded and rooted in the socio-technical environment. On top of that we see that modern feedback procedures are designed for implantation in the relevant behavioral domain. For example, feedback on a variety of technology-supported tasks can be channeled through the user interface of the system, allowing for immediate feedback on specific task choices. In other words, feedback design is carefully designed for a proper fit in prevailing practices. One of the main conclusions that follow from most approaches concerns the centrality of the interaction between the system and the user. Evidently, the user cannot be excluded from the action loop, and often plays a more active role than is sometimes assumed.

200

Chapter 19

Clearly, the technological environment influences people’s behavior and its outcomes. Including the technical context in our analysis obviously adds to our understanding of human behavior. However, understanding the user as a languid product of his or her context, automatically executing routine behaviors, is certainly too gloomy a representation of human functioning and the human-technology interaction. The user still appears responsible for her or his actions, irrespective of contextual effects and habit development. In most cases, the user will be ultimately responsible for choosing or not choosing to use a system, and for allocating the control in the interaction. A system that respects the user’s freedom of choice, and that can even enlarge or increase this freedom, might be an effective change agent. Modern ‘smart homes’, for example, seem to offer many opportunities for guiding user behaviors in various domains, such as by regulating the indoor climate, cleaning, bathing, cooking, lighting and communicating. However, as we have seen, poor design is likely to fail and be counter-productive. Designing human-technology interaction requires accurate specifications, based on careful behavioral and contextual modeling and research. This interactive approach may allow for the creation of sustainable systems that are not only technically sound, but also effective and desirable.

REFERENCES Aarts, H.A.G. (1996). Habit and decision making, the case of travel mode choice. Dissertation, University of Nijmegen, Nijmegen. Berg, S.M. van de (2002). Prospective memory, from intention to action. Dissertation, Technische Universiteit Eindhoven, Eindhoven. Giddens, A. (1984). The Constitution of Society: outline of the theory of structuration, Cambridge: Polity Press, Berkeley: University of California Press. Lee, J. and Moray, N., (1992). “Trust, control strategies and allocation of function”, in McCalley, L.T. & Midden, C.J.H. (2002) Getting energy conservation feedback to work: a new understanding. In: Bartels G & Nelissen, W. (eds), Marketing for Sustainability, towards transactional policy-making. Amsterdam: IOS Press.

PART 3 DESIGNING TECHNOLOGY-BEHAVIOR INTERACTIONS

Chapter 20 PLANNING BEHAVIOR: Technical Design as Design of Use Plans

Wybo Houkes and Pieter E. Vermaas

1.

INTRODUCTION

Our department moved in 2000 to a new building. Two surprises awaited us there: strange light switches and coffee mugs placed ostentatiously on our desks. We welcomed the coffee mugs as presents. It was harder to figure out how to operate the light switches. They looked like screens, so people touched them. That did not work. Then a rumor spread that a little button was hidden underneath the screens. It had three different positions. In the first position, the lights were out, in the other two they were on. After this discovery, we went to work in our newly lit building, with fresh cups of coffee. Yet there were still teething problems: sometimes the lights turned off, but turned back on as soon as you walked towards the switch. After a week we were settled in and realized that our building was environmentally friendly. An e-mail was sent to explain that the coffee mugs were not presents but part of a rotation system: we could use them and take them to the canteen when they were dirty. The three modes of the light switch were ‘on’, ‘off’ and ‘on when detecting movement’. We had to put them in detection mode to make sure that the lights were off after we left our offices. After a month we were really settled in. Our rotating coffee mugs were at the canteen. We did not know where to get clean ones so inefficiently washed our old mugs again. The lights were permanently on. In this contribution we consider the process of technical design. This process is often described as starting with specific goals and resulting in 203 P.-P. Verbeek and A. Slob (eds.), User Behavior and Technology Development: Shaping Sustainable Relations Between Consumers and Technologies, 203-210. © 2006 Springer.

204

Chapter 20

artifacts. For instance, the goal may be to save electricity and the design results were our new light switches. In sections 2 and 3 we present an alternative account by Houkes et al. (2002) in which design is primarily the planning of user behavior. In our view, design starts with goals and ends with use plans, which involves an ordering of behavior by which users can reach these goals. Designers may also design artifacts (and if they do, be occupied with this most of the time) but designing artifacts is subordinated to designing use plans. In the above example, design results in a plan that includes a light switch with a movement-detection mode. In section 4 we describe the use of artifacts in terms of use plans and criticize the view that use plans are communicated to potential users only through the properties of artifacts. Sometimes this communication can indeed be established through the designed artifacts themselves: we did eventually figure out the use plan connected to our light switches. But this is not always the case: the coffee mugs on our desks did not reveal that they were part of a rotation system. In this last case, we needed more than the artifact to execute the use plan; this is the subject of section 5.

2.

DESIGNING USING BY PLANS

Before focusing on design, we should briefly consider use. Our starting point is that both design and use are types of action that can be described in terms of plans, intentions and practical reasoning. The intentions of an agent are usually33 characterized in action theory as requiring a combination of desires and beliefs: one intends to do something if and only if one wants to do it and believes that one has a chance of achieving it.34 A plan is an ordering of considered actions that are means for achieving a goal.35 These orderings may be linear, determining the exact order of the actions, or partial, including multiple options. Practical reasoning is the process by which an agent forms intentions and plans, based on desires and beliefs. It is often assumed that every intention and action can be reconstructed as being based on practical reasoning, even if there is no sequence of explicit decisions and belief-formations in the agent’s mind.36

33

34 35 36

The literature on action theory contains many objections and alternatives to this view, but it is perhaps the closest thing to a received view in the field. Davidson (1985). Audi (1991). This ‘inferentialist’ view is defended in Audi (1989), pp. 113-119.

Planning behavior

205

Applying these action-theoretical notions, using an artifact A — for simplicity, we consider only one artifact — can be described as executing a use plan P for achieving a goal g; the artifact’s function is the role it has in this plan. Assume, for instance, that you own a boat moored on a canal in Amsterdam. If it rains, your boat fills with water and you have to bail it out with a bucket. In terms of a use plan, you have the goal of emptying your boat and a plan to do this by taking a bucket, going to the boat, filling the bucket with water from the boat and pouring its contents into the canal; the bucket’s function, relative to this plan, is to bail out the water in the boat. A plan is subject to various standards of rationality.37 Its goals should, for instance, be realizable at the same time, and the planning agent should rely only on beliefs held to be true. Moreover, the plan should be means-end coherent: the means chosen must be considered appropriate to the goals.38 This implies that the user should believe that the use plan P he selects is appropriate for realizing his or her goal. There appear to be three often overlapping grounds for this belief about the plan. It may be based on beliefs about the physical properties of the artifact A to be used. The behavior and stories of fellow users provide another source of beliefs about plans. And finally, the user may believe that P is appropriate because he believes that P, and perhaps the artifact A, has been designed for bringing about the goal. For the boat, the second ground is most prominent. The first becomes important as soon as your bucket is broken and you look for alternative objects for bailing out the water. And the third shifts our attention to design. The first and second grounds serve to make a plan rational; the third does not just make it rational, but proper as well.39 Perhaps the bucket used to bail out the water was designed to hold mayonnaise; it is then used rationally but improperly. Of course, if an artifact is used mainly within an alternative plan, one may eventually take this rational use as proper as well.

3.

DESIGNING USE PLANS

Suppose you are tired of bailing and consult a designer. She may then start constructing a new use plan P for emptying the boat. This new plan need not involve novel artifacts — she may advise mooring the boat under a 37 38 39

Bratman (1987). Audi (1989). Houkes (2003).

206

Chapter 20

bridge — but typically it will. The development of this use plan and the possible artifacts it involves can itself be described as a plan D of the designer, aimed at contributing to the realization of the users’ goal g. If design can be described as a plan, it has to satisfy the corresponding standards of rationality. Specifically, design should be means-end coherent: the designer’s actions should be appropriate to contributing to the users’ goal g. For your boat this means that the designer cannot just give you a pump. The designer should check your ability to carry out the new use plan P (e.g. whether you can operate the pump as planned) and whether it indeed leads to an empty boat. Spelling out some details, design can be reconstructed as the following sequence of decisions and belief-formations: D.1 D.2 D.3 D.4

D.5 D.6 D.7 D.8 D.9

D.10

40 41

42

The designer wants to contribute to the users’ goal g. The designer believes that goal g′ is the closest consistent and viable approximation of g.40 (from steps 1 and 2 by the characterization of intention) The designer intends to contribute to bringing about goal g′. The designer believes that a considered user who is following an appropriate use plan P that involves the use of artifact A, will bring about g′.41 (from steps 3 and 4 by practical reasoning) The designer intends to construct a use plan P and to communicate it to considered users. (from step 5 by inclusion) The designer intends to contribute to producing artifact A by designing42 A, and acts accordingly. The designer checks whether the resulting design of A is coherent with P, and returns to either step 4 or 5 if this is not the case. The designer decides to communicate P to the considered users, and acts accordingly. The designer believes that g′ can or cannot be brought about by considered users to whom P is communicated. This belief is based on observing that some of these users go through a sequence of actions P′ and bring about g′′, and on comparing g′′ with g′. The designer decides whether or not aim to contribute to bringing about g′ has been achieved. If the latter, she may decide to repeat

This goal g′ may be identical to g. This belief is based on beliefs about the appropriateness of the means to the considered user’s goal, about the physical circumstances of considered use, and about the skills of the considered user. In practice, the designer may gradually construct P by formulating a tentative plan, checking data on consumer behavior, reformulating the plan, etc. ‘Designing’ is here used in the broad sense of describing the artifact in words, pictures and gestures, along with instructions for producing it.

Planning behavior

207

the entire sequence D, settle on another plan (return to step 4), redesign the artifact (return to step 6) or reattempt communication (return to step 8). This sequence is a considerable over-simplification. Actual design involves more agents than a single designer and user: it is done by teams, the construction, testing and communication of P may be done by different agents, and the agents that hire designers need not be the considered users (consider the coffee mug example). We also omitted the reasons why the designer may choose a goal g′ different to g, such as government regulations and the designer’s wish for safe and durable artifacts. And we ignored how the artifact A is designed. In Houkes et al. (2002), we present a more elaborate reconstruction of design that addresses a number of these simplifications. In design methodology the design of the artifact A (step 6) is often identified as central to designing. Roozenburg and Eekels (1995, p. 35), for instance, describe this step as finding ‘a suitable geometrical and physicochemical form for the product and its parts, so that the given function, or functions, can be fulfilled’. Practically speaking this step indeed dominates design; it usually takes up most of the designers’ time and energy. By our analysis this step comes conceptually after the design of the use plan: the function(s) the artifact A has to fulfill are only defined properly once the use plan P of which A is a part is laid down. Roozenburg and Eekels mention the context of use of the artifact in passing when they point out the importance of instructions for use (§4.2.5). But, unlike in our approach, developing these instructions is not explicitly taken as a conceptually separate design stage.

4.

COMMUNICATING PLANS

If our description of use and design in terms of plans is correct, designers do more than only desining artefacts. They also design use plans and influence user behavior by communicating these plans. How do they communicate use plans? A first, narrow answer is that designers communicate their developed use plans through the physical properties of the artifacts that play a part in those plans. A second, broader view is that this communication can also be established by other means. In the sociology of artifacts inspired by Akrich (1992) and Latour (1992), we find expressions of the narrow view. Akrich and Latour describe artifacts

208

Chapter 20

as texts (scripts) that are inscribed in objects by their designer. Users decide how to use the artifact on the basis of their interaction with these scripts: scripts ‘define a framework of action together with the [users] … and the space in which they are supposed to act’.43 Although this definition does not exclude a broader view of communication, the emphasis lies on the features of the used object. Jelsma (2000), for instance, writes that ‘[s]cripts are the structural features of artifacts encouraging certain user actions while counteracting others.’ If one accepts this narrow reading, then designers pass information to users by the physical properties of artifacts only. Disregarding some underlying philosophical divergences between our view and that of Akrich and Latour, we consider the latter as too narrow a way to describe how users learn to use artifacts. Surely it holds for some artifacts: such everyday artifacts as coffee machines and corkscrews have typical physical properties through which their uses can be recognized. Still, it does not hold in general. Consider, for instance, artifacts that are white odorless pastes, such as filler, condensed milk, anti-ageing and medical cremes, etc. Users need the containers of these pastes to tell them apart and to recognize their use. Those containers have explicit information in pictures or printed words about how to use the paste. Or the containers may remind users of information (demonstrations) provided in commercials or in shops. This information originates in part from the designers: they have, for instance, communicated to the marketing department how to present the paste. If one considers more complicated artifacts, it becomes even clearer that designers pass information to users through other channels than the physical properties of the artifacts. Take, for instance, a new atomic submarine or the latest MRI scanner. A crew of sailors does not simply read off the operating principles from the properties of the submarine. Instead, users are explicitly trained in the use of these artifacts. Providing information for this training is a task of the designers of these artifacts.44

5.

PLANNING BEHAVIOR

On our account, designers are primarily planning behavior by designing use plans. In other words, designers are not just technologists who solve

43 44

Akrich (1992), p. 208. At some points, Akrich (1992, p. 208) and Latour (1992, no. 5) seem to take a broader view of communication through scripts by suggesting that advice by the designer, user manuals and demonstrations are all part of the scripts.

Planning behavior

209

technical problems concerning material artifacts. They are above all social engineers who solve problems in the realm of human action. Therefore, they need to have knowledge about users as well as technical and scientific knowledge. The pleas in this book to involve prospective users in designing can be understood as pleas for providing designers with that user knowledge. To return to the example of our new building: if the designers had known that our work does not involve moving around a lot, they probably would not have installed the strange switches. But knowledge of users is not enough for successful design. Perhaps the designers of the mug-rotation plan consulted many coffee drinkers and confirmed that the plan worked according to everybody’s wishes. But because they did not let us know what the mugs were for and where to get clean ones, the plan was never executed. Hence, designers should proficiently communicate the designed use plans to their prospective users. In the case of environmentally-friendly design, it seems that designers hope that they can append their sustainable artifacts to existing users plans without any active communication to users (e.g. water-saving showerheads) or by letting the sustainable artifacts themselves do the talking (e.g. the two buttons on the water-saving toilet cisterns). But users cannot always guess the proper use of sustainable artifacts by the properties of those artifacts alone; they may be in need of extra information (in this volume, Derijcke describes a case in which users of solar panels requested information about their use). On our account, designers should actively provide such information: they produce more than just artifacts, and should therefore present more than just artifacts.

REFERENCES Akrich, M. (1992). The De-Scription of Technical Objects. In Bijker and Law (1992), pp. 205–224. Audi, R. (1989). Practical Reasoning. (Routledge: London), chapter 7. Audi, R. (1991). Intention, Cognitive Commitment, and Planning. Synthese 86, 361–378. Bijker, W.E., and J. Law (eds.) (1992). Shaping Technology/Building Society: Studies in Sociotechnical Change (Cambridge, Ma: MIT Press). Bratman, M. (1987). Intentions, Plans and Practical Reason. (Cambridge, Ma: Harvard UP). Davidson, D. (1985). Replies to Essays I-IX. In B. Vermazen and M. Hintikka (eds.) Essays on Davidson: Action and Events. (Clarendon: Oxford). Houkes, W. (2003). Het Oneigenlijk Gebruik van Artefacten. Filosofie 12, 34–38. Houkes, W., P.E. Vermaas, K. Dorst and M.J. de Vries (2002). Design and Use as Plans: An Action-Theoretical Account. Design Studies 23, 303–320.

210

Chapter 20

Jelsma, J. (2000). Design of Behavior Steering Technology. Contribution to ‘Strategies of a Sustainable Product Policy’ Summer Academy of Technology Studies, Deutschlandsberg, Austria, July 9–15, 2000. Latour, B. (1992). The Sociology of a Few Mundane Artifacts. In Bijker and Law (1992), pp. 225–258. Roozenburg, N.F.M., and J. Eekels (1995). Product Design: Fundamentals and Methods. (Chichester: John Wiley & Sons).

Chapter 21 EXPECTED BEHAVIOR: Anticipation of Use in Technological Development

Harro van Lente

1.

INTRODUCTION

Developers of technology face an intriguing task: they have to produce things that are designed for situations that, by definition, do not yet exist. Since the new technology is still non-existent, the way the technology is taken up in firms, organizations and households is unknown, too. Yet developers need information about the future use of technology to make strategic decisions about their products, systems and markets. How can they do so? What is their secret weapon? The answer is: instead of information about future use of technology, they use anticipations. The development of technology is embedded in a wide variety of projections and images of the future use. In this article I will investigate these projections and analyze the role of anticipations of use in the development of technology. This paper starts with a basic distinction between two types of anticipations of use. This leads to an investigation of how anticipations may become forceful in technological development. Two elements turn out to be important here: first, the roles for engineers, firms and users that follow from these anticipations, and second, the fertile ground for projections about future use. Finally, I will discuss challenges for sustainability.

211 P.-P. Verbeek and A. Slob (eds.), User Behavior and Technology Development: Shaping Sustainable Relations Between Consumers and Technologies, 211-219. © 2006 Springer.

212

2.

Chapter 21

TWO TYPES OF ANTICIPATIONS

A basic notion in studies of innovation is the distinction between ‘incremental’ and ‘radical’ innovations (Utterback, 1994). While incremental innovations are improvements of existing products and processes, radical innovations introduce new types of products and, as a consequence, often new markets as well. Incremental innovations relate to step-by-step improvements, due to learning by doing and to economies of scale. Radical innovations, on the other hand, are far more dramatic and lead to a ‘creative destruction’ ʊ as Schumpeter phrased it ʊ of established products, habits and firms. Likewise, we may distinguish between two types of anticipations on the future use of a technology. First, technologies will be adapted to practices that are more or less known already. The types of products and processes already exist, and can be analyzed in real life. Here we find all kinds of activities to improve the new technology vis-á-vis current behavior of consumers and other users, such as ergonomic studies, user feedback or market research (Cavaye 1995, Lin and Shao 2000).45 These activities deal with anticipations of future use that we may call ‘incremental’. In contrast, ‘radical’ anticipations relate to forms of use that do not exist; instead they are projected for the future. Projections of radical new forms are, of course, much more uncertain, as the following example shows. In 1878 Thomas Edison published an article in which he listed ten forms of use of his ‘phonograph’, which he had invented a year earlier (Basalla, 1988). He discusses a number of ways to use his invention: recording of important speeches, a ‘talking book’ for the blind, presentation in the classroom, reproduction of music, recording of family events, such as the words of a dying family member, making new sounds, such as clocks that ‘tell’ the time, or, finally, to record telephone calls. In spite of this list of possible ways to use the phonograph, Edison, in fact, only had faith in the professional use of the phonograph: his invention was a dictation device. In the end, as we all know, the recording of music ʊ the fourth item on his list ʊ turned out to be its most important application, and has made the phonograph a success that Edison did not foresee. In fact, he even resisted putting his invention to such frivolous use.

45

Akrich (1994) has analyzed these attempts as representations of users: through surveys, stories, graphs, and other assumptions “users” are constructed. Design processes, then, are adapted to these representations of users.

Expected behavior

213

Clearly, the relation between expected behavior and the development of technology is complicated. Inventors cannot predetermine the use of technologies, not even a genius like Thomas Edison. Frissen en Van Lieshout (this volume) explain that technologies need to be ‘domesticated’, i.e. they need to become a part of daily practices for households and firms. The fate of technologies, therefore, is in the hand of their users.

3.

THE FORCE OF RADICAL ANTICIPATIONS

The highly speculative nature of ‘radical’ anticipations of future use raises questions about their value and utility. Yet, and this is the thrust of this paper, they play a key role in the development of technology. While anticipations can hardly be justified (as you can justify scientific claims), they are very important in decisions that engineers and firms take. Again, an example may be helpful: The American historian of technology Edward Constant has given an elucidating example of the role of radical anticipation on future practices (Constant 1980). In the 1930s, the propulsion of airplanes changed radically: propeller-cum-piston engines were replaced by gas turbines. The projection of future needs was crucial: aircrafts with propeller engines were reaching a speed limit, while gas turbines could eventually reach much higher speeds. It was a dramatic change, which Constant terms a turbojet revolution. The term ‘revolution’ refers to Thomas Kuhn's ideas of scientific revolutions and the key role of anomalies. An anomaly is an element, say a measurement or a theoretical deduction, which does not fit into the established theoretical frame (the ‘paradigm’). It can, therefore, be taken as proof that the paradigm fails; normally, considerable efforts are then directed to saving the paradigm with all kinds of strategies. Sometimes, however, a non-fitting element leads to putting aside the frame or paradigm. In that case we speak of an anomaly causing a revolution. Constant’s goal is to apply Kuhn’s thesis to the area of technology, but by doing so he has to make a significant shift. In the case of gas turbines, it was not an actual anomaly causing the turbojet revolution, but an anomaly that was expected to occur in the future, a ‘presumptive’ anomaly. The capacity of propeller-cum-piston engines only appeared as an anomaly in the light of an extrapolation of the gradual increase of speed of airplanes. Within that imagined context that was generated by extrapolation, the piston engine failed, and a radical change seemed timely. The development of the turbojet, thus, is the result of anticipation about the future use of airplanes. It was not the current use of airplanes, but their

214

Chapter 21

projected future use that shaped the turbojet revolution. Only in terms of a future context did the option of the jet engine become important. It was not an answer to actual problems or actual needs, but to projected, future problems and future needs. The example of the turbojet shows that radical anticipations can be forceful in shaping new technology. Two steps in the shaping process are important in particular. First, the development of the turbojet engine took place on the basis of a shared expectation that it would do a better job. This expectation was, as one generally finds in technological change, both technical and social. On the one hand, it was based on the perceived difficulties in extrapolating piston engines to higher speeds, and the perceived potentialities of application of aerodynamics resulting in a superior turbojet engine; on the other hand, it was based on the expected importance of speed and efficiency for users of airplanes. Second, the expectations about the future form and use of turbojet technology provided roles and guidelines for the various players that were involved. The options that were considered feasible and promising were translated into requirements, guidelines and specifications. When the gas turbine is the future of the aviation industry, the argument goes, we must make it our task to develop it. Indeed, Constant describes how a community of technologists was formed that defined its tasks and specified their goals in terms of the non-existent turbojet engine. Some made calculations of aerodynamics while others searched for optimal turbine blades or studied materials. Once radical anticipations are shared, they demand action.

4.

OBSOLETE TECHNOLOGY

Radical anticipations on future use, thus, are important in the development of technology. How do these anticipations enter the scene? They do not fall out of the blue, but developers of technology actively pursue them. It is the task of engineers, designers and other technologists to 'follow' the developments in their research areas; that is, they have to check what has been tried elsewhere, what presents itself as a new option, and what are the implications for present technologies and for market relations. At meetings in laboratories, in technical journals, and during conferences, new options are presented and the community of technologists contemplates whether it is ‘promising’or not (Van Lente, 1993). Indeed, an important part of research and development is to scan possibilities and assess their viability. A particular technological option, then, may become more popular and emerge

Expected behavior

215

as ‘the next step’. In the same movement descriptions of future use and user behavior emerge. In this continuous search for new, promising technological options, assessments of what is ‘obsolete’ play an important role. These terms are crucial in the decisions of engineers and other developers of technologies, and some reflection on this term is justified. When technologies are so-called ‘obsolete’, they will be replaced, but not because they are dysfunctional. The black-and-white television of the 1950s and 1960s is obsolete, but still functions, as does the good old typewriter, or even the horse cart. An ‘old’ technology does not fail in the sense that a theory may be said to fail, or like a hypothesis that may turn out to be wrong: We are often misled by the cliché that technology proceeds by trial and error: abandoned technologies are in no sense erroneous (or failures or falsified). Watt steam engines, Bessemer converters, clipper ships or Fordson tractors presumably work as well as they ever did. (Hamlin 1992: 529) An old technology does not fail, but is abandoned, and this is the reason why it does not function anymore, except perhaps in museums. They are like old soldiers, who fade away because they are abandoned. An old technology is obsolete, we say, or it has been rendered out of date and is superseded. What happens instead is that old technologies fail in terms of what is considered as ‘the next step’. The failure of the obsolete technology is only visible in the light of the possible future technology. The case of the turbojet revolution is, again, a good example: the failure of the propeller technology is a malfunction or breakdown, but it appears as obsolete in terms of what might be possible in the turbojet future. And once users are getting used to the ‘next step’, the proof is complete that the ‘old generation’ is outdated: black & white television is outdated by color television, the typewriter has given up its place in favor of word processors, and the horse-cart is superseded by the electrical train.

5.

THE LOGIC OF PROGRESS

The metaphor of a ‘new generation’ superseding the ‘old’ one is as widespread as suggestive: the notion of an old generation suggests that it is natural to replace it by a new one. In many areas of technology we find the argument about a ‘new generation’ replacing the old one. The microchip

216

Chapter 21

development, where the Pentium generations are simply numbered, is one example, and the development of television (from black and white, to color, to HDTV) another (Van Lente 2000). Strictly speaking, the argument that the new generation should replace the old one is groundless, since it amounts to circularity: the old one is old because there is a new one. For technological development itself, however, the argument is of great importance since it urges actors forward (Van Lente 1993). Expectations about the future form and use of technology are not just statements that can be right or wrong, they provoke action because they position users, developers, governments and competitors. Technological developments take place within a space of expectations about the future form and use. The paramount idea that technical change will occur and that one has to be prepared, leads to a bias in the assessment of viability of technological promises. This bias has been studied for historical proposals by Steven Schnaars (1989). In his rather amusing book Megamistakes. Forecasting and the Myth of Rapid Technical Change, Schnaars gives hundreds of examples of biased evaluations. By far the majority of them were too optimistic. According to Schnaars, the standard mistake in evaluations and forecasting is an interesting shortsightedness; it is a combination of (a) the tendency to imagine the future in terms of the present, and (b) the easy acceptance of Technological Wonders, thereby assuming a radical change of the present. These two seemingly contradictory elements are combined in what Schnaars calls the ‘dominant themes of the day’. These themes are temporary: “Few of the dominant themes carry over into subsequent decades. They dissipate and are replaced by new dominant themes.” (65). In the 1950s, for instance, we had the theme of the ‘jet engine’, which in the next decade was followed by the ‘space race’ and by ‘nuclear energy’ in the 1970s. In the 1980s, ‘energy crisis’ was a dominant theme, while ʊ I would say ʊ ‘sustainability’ is an excellent candidate for our decade. Schnaars's examples show an amazing receptivity for technological promises, not only amongst firms and consultants, but also in politics and in public opinion.

6.

ANTICIPATIONS AND POSITIONS

The anticipations of future use in technological development are more than descriptions of future products and systems that can be true or not. They may change the world, since they guide the activities of developers of technology. Important here is that anticipations distribute roles for engineers, firms, governments and future users. In the projections that caused the turbojet revolution, future users, for instance, were expected to demand

Expected behavior

217

higher speeds and longer distances. Competing firms, in their turn, were expected to strive for the new generation aircraft propulsion, i.e. the turbojet. The anticipations on future use, thus, shaped the developments in the aircraft industry. In general, statements that shape reality are, what Austin (1961) in his classic study called performative statements. When you utter a promise, or an acknowledgement, for example, you do not give a factual description, but you create mutual obligations and rights. While in a descriptive statement, something is claimed about reality “out there”, in performative statement something in social reality is altered. When you say “I do” in a wedding ceremony ʊ at the right time and place ʊ a factual social condition has changed: you are married. Words can do things. Of course, performative statements do not only occur at special occasions such as weddings, but also abound in ordinary conversations. Anticipations in technological development, as a rule, are also performative (Van Lente 1993, Van Lente and Rip 1998).46 However, performative statements such as anticipations in technology are not powerful in themselves. What is needed for any effect to occur are two other elements: a position that adds credibility to the statement, and a generally accepted storyline.47 In technology, for instance, decisions are made against a background of a story line of new generations succeeding old ones. Whether future users actually accept the roles that are given them in anticipation of future use, is not certain, of course, as the example of Edison makes clear. Frissen and Van Lieshout (this volume) analyze the processes 46

47

Searle (1979) has coined the term ‘speech-act’ to capture this potential. In my study I also argued that the notion of ‘speech-act’ should be extended to statements in the press, and publicly visible acts. The act of a government giving subsidy, for instance, can be ‘read’ in multiple ways. When it is in a technological area where there are already broadly accepted expectations about future performance, such an act can easily be ‘read’ as “we, the government, think that technology X is promising”. These three ingredients, position, story line, statement, are the basic methodological framework of positioning theory. Positioning theory is relatively new, although its roots, social constructionism and discourse analysis, are in place for several decades (Harré and Van Langenhove 1999). The basic idea is that within conversations ‘positions’ are put forward that are embedded in a moral universe with rights and obligations ʊ which may be accepted or not. When Jones asks Smith “Could you please iron my shirt?”, Jones positions Smith as someone who can be asked such a thing, while the question positions him or herself as someone who has the moral right to do so. Smith might accept these positions (“Yes, I will do it in a moment”), or reject these (“Why should I do that?” “I am not your maid.”). Positioning theory can be seen as a dynamic version of role theory in social psychology, and has been applied in such domains as the study of autobiography and cultural studies.

218

Chapter 21

that determine whether or not users accept their role. The important point for this chapter is that the mere existence of these projected roles (“in the future, users demand higher speeds and longer distances”) shapes technological development.

7.

CONCLUSION: ANTICIPATIONS ARE NOT INNOCENT

Anticipation of use guides the development and design of technologies. The key phenomenon here is that technical expectations distribute roles to developers, competitors, governments and users. This provides an opening for concerns about directions of technological change, and shows possibilities for more sustainable options. Once the ‘next step’ is seen in terms of, say, more efficiency and less pollution (instead of, say, faster and bigger), developments will be shaped accordingly. The ‘next generation’, then, will be more sustainable. It is not straightforward, however, how to change the definition of the next steps. There are only a few successful examples, such as the Californian experiment to enforce ‘zero emission vehicles’. This indeed has caused car manufactures to define more sustainable cars as the next step in their industry (Pilkington, 1998). As a consequence, cars that do not live up to these new norms are increasingly seen as ‘obsolete’; highly efficient and clean cars are presented as ‘cars of the future’. Engineers and other developers are continuously very keen on ‘the next step’ in their fields of expertise, and are continuously assessing when the technology they are dealing with will be obsolete and what the next generation will be. Technologies, therefore, are shaped by anticipations. The future of a technology may be in the hand of users, the shape of a technology depends on the anticipations of use. To conclude, anticipations of use do matter. They may be uncertain and unjustified, but they are neither useless nor innocent.

REFERENCES Akrich, M. (1995). User Representation: Practises, Methods and Sociology. In A. Rip, T. Misa & J. Schot (Eds.), Managing Technology in Society - The approach of Constructive Technology Assessment London: Pinter Publishers, 167-184. Austin, J. L. (1962). How to do things with words, Cambridge, MA.: Harvard University Press.

Expected behavior

219

Basalla, George (1988). The Evolution of Technology, Cambridge: Cambridge University Press. Cavaye, A. L. M. (1995). User participation in system development revisited. Information & Management, 28, 311-323. Constant, Edward W. (1980). The Origins of the Turbojet Revolution, Baltimore: The John Hopkins Press. Harre, R. and Luk van Langenhoven (1999). Positioning Theory: Moral Contexts of Intentional Action, Blackwell Publisher. Lin, W. T., & Shao, B. B. M. (2000). The relationship between user participation and system success: a simultaneous contingency approach. Information and Management, 37, 283-295. Pilkington, A. (1998). The fit and misfit of technological capability: responses to vehicle emission regulation in the US. Technology Analysis & Strategic Management, 10, 211-224. Schnaars, Steven (1989). Megamistakes. Forecasting and the Myth of Rapid Technical Change. New York: Free Press. Searle, J. R. (1969). Speech Acts. An Essay in the Philosophy of Language, London: Cambridge University Press. Van Lente, H. (1993). Promising technology. The dynamics of expectations in technological developments’ Eburon: Delft. Van Lente, H. en A. Rip (1998), ‘The Rise of Membrane Technology. From Rhetorics to Social Reality’, Social Studies of Science, 28 (2), 1998, pp. 221-254. Van Lente, H. (2000). ‘Forceful Futures: From Promise to Requirement’, in Nik Brown, Brian Rappert and Andrew Webster (eds.), Contested Futures. A sociology of prospective techno-science, London: Ashgate Publishing Company, 43-64. Utterback, J. M. (1996). Mastering the dynamics of innovation. Boston, Massachusetts: Harvard Business School Press.

Chapter 22 DESIGNING ‘MORALIZED’ PRODUCTS: Theory and Practice

Jaap Jelsma

1.

INTRODUCTION

One paradox of our times seems to be our wish to establish a sustainable society on the basis of market principles. That is, we want to achieve a public goal by means of processes and choices taking place in the private domain. We want our economy to grow, we produce and use more and more goods, especially luxury ones, in the private sphere. At the same time, we want to be more sustainable. In the public sphere, government makes moral appeals to consumers to be prudent with the use of all these goods: take the bicycle, load your washing machine fully, do not drive too fast. That is: ‘a good environment begins with you!’48 Thus, in working toward sustainability there is a division of labor based on the way we perceive the nature of things (viz. products of technology) versus the nature of humans (viz. designers and users of technology): • technology is functional, neutral, has no morality. In neo-classical economic theory, technology is defined as exogenous to economic processes. • designers are either just technical experts (engineers) or artist-like. The former are nerdy working on technology that is functional. The latter are frivolous, wanting technology to look nice.

48

Slogan used in a campaign for the environment by the Dutch government.

221 P.-P. Verbeek and A. Slob (eds.), User Behavior and Technology Development: Shaping Sustainable Relations Between Consumers and Technologies, 221-231. © 2006 Springer.

222

Chapter 22

• consumers are common people who are blamed for damaging the environment by using all these goods. Government finds that they have to be educated. That is, they should change their values and attitudes. Their eco-awareness has to be increased, so that they will behave in more friendly ways towards the environment. In this division of labor consumers are the moral party. Only their morality is of the wrong kind, and has to be changed. They should behave (in Dutch) ‘milieubewust’. This division of labor is unfair, and based on assumptions about human behavior that are grossly mistaken. Mainstream psychologists think that the main drivers of human behavior are attitudes, values and intentions. But in everyday life practice, behavior is embedded in habits and routines (as explained by Jelsma, in Part 1 of this book). Routines are patterns of unconscious actions guided by material infrastructures acting like beacons and signs. Specific material features of the artifacts involved (e.g. those of cup, saucer and spoon in coffee drinking) support and guide the actions of the user. By realizing this, we start to perceive artifacts in a different way. We had better start seeing them as actively taking part in human action, as drivers of routine action, i.e. as actors. This means that these artifacts have a co-responsibility for the way the action develops and for what results. If we waste energy or produce waste in routine actions, such as in household practices, this has to do with the way artifacts guide us. But artifacts themselves do not invent their actions towards users. Artifacts are designed and made by another class of people, designers. So here we have another group that is acting morally. Designers are the people who inscribe a morality in the things they invent and shape. They make many things that invite us to use more resources (water, energy, materials) than needed, or than we can afford. Under current ways of policy-making, this immorality has to be corrected by moral behavior ʊ that is, conscious behavior ʊ from the side of the users of these ‘immoral’ products. This approach imposes considerable costs on users: costs in terms of time, effort, money (taxes, tariffs), and feelings of guilt and failure.49 Now we have spread the responsibility for the lack of sustainability more evenly between (products of) technology, its users (i.e. consumers) and its makers (i.e. designers). For producing sustainability, we have reached a 49

A more general argument about the moral role of artifacts can be found in ‘Where are the missing masses?’ (Latour 1992).

Designing ‘moralized’ products

223

fairer distribution of labor. With designers having their share of the moral task now, how can they make products that support behavior in a sustainable society? To do so, they need new methodologies that make explicit the active and moral role of artifacts in producing and supporting responsible behavior. In the rest of this article, the design and testing of such a methodology is described.

2.

CONCEPTUAL FRAMEWORK: SCRIPT LANGUAGE

What we need first is a concept that links the design and use of artifacts. The semiotic concept of script lends itself well to this purpose. A script is a material structure that, by its specific layout, exerts force on the actions of its user. That is, the script of an artifact invites certain user behavior while counteracting other behavior. For example, when a device stimulates men to use it but discourages women from doing so, we say that the device carries a gender script. In other words, scripts create gradients of resistance in the material landscape we live in (cf. Latour 1988) and so influence (‘translate’) the direction of human behavior. Three related notions are derived from the concept of script, i.e. inscription, prescription, and de-scription (Akrich and Latour 1992). By relating these notions, processes of design and use can be linked in a dynamic way: To be methodologically usable, the script concept has to be elaborated as follows. Properties of scripts are: • force The prescriptive force of a script can vary depending on the opportunities it leaves the user to enroll the artifact in a practice of unintended behavior. Designers can dose the script’s force by restricting the opportunities for undesired use, or strengthening the stimuli for desired use by the layout of the hardware. Anticipation of use practice will help such design choices. • scale Scripts can be identified on different levels of complexity. Here we restrict ourselves to the level of artifacts, mostly conceived as the micro level. • direction The scripts of artifacts can steer behavior in different directions, depending on how and where they create gradients of resistance for behavior in the socio-technical landscape. One vector is the direction pointed out by the costs of behavioral change. A script that increases comfort and lowers environmental burdens of behavior at the

Chapter 22

224

same time (so called win/win solutions) is easy to implement, because it works in the direction of existing gradients. • distribution A script distributes tasks, responsibilities and power between humans and nonhumans, i.e. between users and artifacts. Where the artifact is fully in control of the function, we speak of delegation of tasks to the artifact, or automation. The example below, showing two interfaces of different water-saving cisterns, illustrates the force and direction of scripts on the scale of an artifact: Fictive user

Designer Inscription DESIGN Realized design: device with a Script Prescription

Design(ers) world Description

Use(rs) world)

Real user Private use

Shared use

Figure 22-1. Script terminology connecting design and use process

2.1

User logic and script logic

The script language described above offers terms by which moral parameters of the design process that have implications for the process of use are made visible and accessible to manipulation. If we would stop here, the framework would suffer from technological determinism, however. Simply forcing behavior on users by cleverly designed artifacts is not only undesirable, but also impossible (cf. Brey, this volume, chapter 33). Design/use

Designing ‘moralized’ products

225

processes are not linear. Users give meaning to artifacts from their own socio-technical context. It is not by accident that the arrows in Figure 22-1 go both ways between users and scripts. Research outcomes at the user side indicate that users actively domesticate novel artifacts, instead of just following unequivocal messages from scripts in one inescapable way. Artifacts never determine user behavior completely. To tackle this problem and reduce unintended outcomes such as rebound effects (Tenner 1994), we introduce the concepts of user logic and script logic.

Figure 22-2. The upper system has been installed in some office buildings of Twente University. A modest study showed that most users did not understand it and/or ignored its water saving potential. The interface permitted this negligent behavior. By pressing the large lower button the water starts to flow, and only stops flowing after the upper STOP-button has been pressed. The script follows the familiar routine of flushing (press a button and leave), and the user has to carry out an extra action to save water that is not stimulated by the device. For that reason, this interface has a weak script for saving water. The script of the lower interface exerts a stronger force on the user’s behavior to save water. In this case, the script goes against the familiar routine. The script forces a shift to reasoned behavior: the user has to think about which button to use, the one for a small amount or a large amount of water. Saving water and flushing are combined into one action.

User logic (or ‘folk logic’) is the consistent whole of heterogeneous rationales that consumers mobilize in their interaction with scripts in everyday practice (for an elaborate description, see Jelsma 2005). This logic is often buried in routines, i.e. it is subconscious knowledge triggered by artifacts. To establish and support more eco-friendly consumer routines by ‘correcttly’ designed artifacts, knowledge of user logic is crucial. Because of

226

Chapter 22

the concealed character of this logic, we need a methodology for disclosing and mapping user logic before it can be taken into account in designing. In a similar way, there is a ‘script logic’ (a form of design logic, see Jelsma2005 and Part 1 of this book) comprising the ideas, values and intentions in the design context that have been inscribed in the hardware of the artifact. Comparison of user logic and script logic may reveal discrepancies, which may lead to forms of unintended use. The framework developed here has similarities and differences with the action-theoretical approach of Houkes et al. (this volume). The main similarity is the normative starting point that users matter, and designs should be checked against ‘user plans’ or ‘user logic’. The ‘user plan’ of Houkes et al., however, is a construction made by designers, and the definition of the artifact these authors use is linked to it. Thus, as soon as the user intervenes by his or her own plan and domesticates (our term) an artifact, this gives rise to ‘metaphysical absurdities’. However, by keeping the user plans of designers and of users themselves conceptually separated, we are able to compare them and learn form their discrepancies to improve designs. Finally, Houkes et al., by using action theory, restrict themselves to consciously reasoned behavior in a philosophical rational reconstruction exercise, of which the relevance for design practice still has to be proven. Our mission is interventionist, we want to change everyday consumer practice in a normative way. For that mission, we think the study of routine behavior is more urgent.

3.

DESIGN METHODOLOGY IN EIGHT STEPS

The design method based on the framework explained so far has the following key elements: A technology side: • identification of relevant scripts of the appliance and the logic underlying these scripts (script logic); and • tracing the energy-intensive actions of the appliance. A user side: • reconstruction of user logic, and comparison with the script logic; and • involving users in the (re)design process of the appliance. These elements have been elaborated into a stepwise approach for the design of ‘moralized’ artifacts (see figure 22-3). In this section, we explain only the crucial steps.50 50

An elaborated explanation will be found in J. Jelsma (1999).

Designing ‘moralized’ products

227

0. Selection of appliance(s)

1. Formation of team

2. Selection of script(s) Reconstruction of script logic

3. Selection of user(group)s

4. Data collection

5. Interpretation of data Reconstruction of user logic

6. Redesign of scripts

7. Formulation of concepts

8. Making and testing prototypes

Figure 22-3. Design methodology in steps

3.1

Step 1: Making a team

The initial team should comprise designers, (a) technician(s) and qualified observers of use practice (such as anthropologists). Later on, users will join the team.

3.2

Step 2: Selection of scripts and identifying their logic

This step is to be carried out by the designers and technicians on the team. The aim of this step is to identify the scripts that are most rewarding to be redesigned from the viewpoint of ‘moralization’. In the case of energy savings, these are the scripts supporting user actions that are (relatively) the

Chapter 22

228

most energy consuming ones (to be estimated by the technician). Scripts aiming to let the user save energy (e.g. through specific saving or ecobuttons) are also relevant to be selected, to check whether they function as intended by the underlying script logic. Carrying out this step requires a number of actions: (1) making a list of the types of interactions of users and machine, (2) identifying the hardware involved in these actions, and (3) reconstructing the script logic behind the scripts of this hardware. For instance, in loading a dishwasher, the user interacts with racks and baskets whose shapes carry the inscribed intentions of designers about this hardware’s use. The scripts of the racks (whether used as intended or otherwise) influence the efficiency of loading, and so the energy efficiency of the washing process. To reconstruct the script logic, the most reliable approach is to interview the designers of the original hardware. In practice, this might be too far-fetched in most cases, and an estimation will be made instead by the designers on the team. On the basis of the script logic an image of the intended use can be made. This image can be compared later on with the real use. Data gathering in this step is supported by a table-like format that is being filling in during research.

3.3

Steps 4 & 5: Data gathering on use practice and reconstruction of user logic

For developing this step, we borrowed methodology from contextual design (Beyer and Holtzblatt 1998). Contextual design starts at the user side, mapping use practice carefully by contextual interviews. This can be done by designers or anthropologists, preferably by both. A contextual interview is carried out in the real context of use, i.e. in the case of household appliances in the homes of users. The observer(s) watch the user while interacting with the machine in normal practice, posing questions to disclose reasons for routine actions, and taking notes (step 4). These data form the basis for (i) comparing real and intended use, and (ii) reconstruction of user logic within the team, which can then be compared with the script logic (step 5). Comparing both logics is meant to fuel discussions (guided by a checklist of questions) within the team about design suggestions that are relevant for saving energy in practice. The most promising ideas form the input for step 6.

3.4

Step 6: Redesign of scripts

A selection of users now joins the design team. The designers confront the users with ideas for redesign developed in step 6, to evoke reactions.

Designing ‘moralized’ products

229

These reactions may help designers to estimate better which of their ideas fit best with user logic, and to anticipate possible reactant behavior of users. Another option is to let users design their own machine first, and put forward redesign suggestions. Technicians can provide an initial estimate of the energy-saving potential, and designers can evaluate the ideas of users against their own. The outcome of this step is a number of concepts for a moralized appliance that can be elaborated towards prototypes in the next steps.

4.

PILOT STUDY: REDESIGNING THE DISHWASHER

Two pilot studies (dishwasher and refrigerator) were undertaken by separate design firms to test and improve the method explained above. We will only provide a few examples from the dishwasher study here (Groot et al., 2000). In step 2, clusters of use actions were categorized, such as sorting, rinsing and loading, applying detergent, choosing the program, unloading and maintenance. From among these, rinsing, loading and setting temperature (included in choosing program) were selected as the actions that were most rewarding for further research. The parts of the machines guiding these actions and their scripts were identified, and their script logic was identified. On the user side, 12 families with dishwashers were recruited (step 3) for the contextual interviews (step 4). To our surprise, having used dishes watched by strangers appeared to be a privacy-sensitive matter. Several housewives did not allow the observers to videotape the loading of the machine with dirty dishes. They only allowed observation of loading clean dishes. By comparing real use with intended use of scripts, several discrepancies with relevance to energy use arose. These could be explained by the user logic tapped. For instance, three families rinsed the dishes under running hot water before loading. The dishes went into the machine almost completely clean. However, the design logic has delegated the rinsing to the machine. Users said that they did not know this, or that they did not trust the machine in carrying out this task adequately. Thus the presence of a rinse script, invisible to, or not trusted by users triggered energy inefficient behavior. Design suggestions to repair this evil (coming out of step 6) were: (i) extending the interface, with a rinse button to be pressed to set rinsing in motion, (2) giving feedback about rinsing through a display (‘machine is rinsing’), and/or (3) enhancing the transparency of the process by making the front panel of the machine transparent, so that the user notices the start of

230

Chapter 22

rinsing, and is able to judge the result. Another remarkable finding was the script logic of the racks to come from Brussels: the EU has set standards for dishwashers prescribing options for loading of 12 covers, 12 cups and saucers, etc. This standard prevents efficient loading, for instance in cases of families with little children who use beakers instead of cups and saucers.

5.

LESSONS

The design firms’ evaluation of the general approach was that the involvement of users is not uncommon as such, but practicing it in a framework of ‘moralization’ is new, as well as carrying out user research at home. They judged the chances for implementation were rather high, since the method remains close enough to normal design practice. To make this perspective more promising, the method should be streamlined and made less ‘academic’. Furthermore, there were doubts as to whether the method can do more than only improve existing designs. My own answer to this question is that not only existing, but also new concepts have been demonstrated to gain from systematic exploration of user logic and context (Oudshoorn et al., 2004).

ACKNOWLEDGEMENT The development of the design method and the test pilots were sponsored by Dutch energy agency Novem.

REFERENCES Akrich, M. and B. Latour (1992), ‘A summary of a convenient vocabulary for the semiotics of human and nonhuman assemblies’, in: Shaping Technology, Building Society (W.E. Bijker and J. Law, eds.), The MIT Press, Cambridge Mass., pp. 259-265. Brey, Ph., this volume, chapter 33. Houkes, W., P.E. Vermaas (2002), K. Dorst and M.J. de Vries, ‘Design and use as plans: an action-theoretical account’, forthcoming in Design Studies. J. Jelsma (1999), ‘Huishoudelijk energiegebruik: Beter gedrag door beter ontwerpen’, Novem Utrecht. J. Jelsma (2005), ‘Bridging gaps between Technology and Behavior: A Heuristic Exercise in the Field of Energy Efficiency’ in Households, in: User Involvement in Innovation Processes, Strategies and Limitations from a Socio-Technical Perspective (H. Rohracher ed.), Profil, München, pp. 73-107. Latour, B (1988), The Pasteurization of France, Part Two, Irreductions, Harvard University Press, Cambridge Mass.

Designing ‘moralized’ products

231

Latour, B. (1992), ‘Where are the missing masses? The sociology of a few mundane artifacts’, in: W.E. Bijker and J. Law 1992, op. cit., pp. 225-259. Oudshoorn, N.E.J., E.W.M. Rommes and M. Stienstra (2004), Science, Technology & Human Values, Vol. 29 No. 1, pp. 30-63. Tenner, E. (1994), Why things bite back, Technology and the revenge of unintended consequences, Alfred A. Knopf, New York.

Chapter 23 THE SCENARIO METHOD TO GAIN INSIGHT INTO USER ACTIONS

G.W. Wolters and L.P.A. Steenbekkers

1.

INTRODUCTION

This research was conducted at the request of a manufacturer of domestic appliances. Designers and manufacturers of domestic appliances often receive complaints about the usability of their products. One reason for these complaints is that consumers use the appliances in other ways than the designer had meant or foreseen while designing the product. This might influence the functionality and usability of the product. Many different aspects have to be taken into account when designing products; usability is one of them. Design methods, such as the one of Roozenburg and Eekels (1996), intend to structure the design process. In this (and other) method(s) hardly any attention is given to detailed use actions. This information is, however, badly needed. Therefore, the request was to develop a method that would enable the designers and developers of new products to gain in a structural way insights into the actions of the user in the different phases of use. The developers of the company already habitually described use actions, and called that description a use scenario. In order to develop a method to write down such scenarios systematically, the concept of a ‘use scenario’

233 P.-P. Verbeek and A. Slob (eds.), User Behavior and Technology Development: Shaping Sustainable Relations Between Consumers and Technologies, 233-240. © 2006 Springer.

Chapter 23

234

must first be defined, and after that a definition of the objective and the contents of a ‘use scenario’ needs to be provided. These are required to develop the method to write a use scenario, but not without taking into account the requirements of the people who will use the method. The scenario method is a tool that can be used during all stages of the process of development of (new) appliances or even services. While theories with a more psychological background look at product use from the perspective of attitude and behavior (see, for an overview, Heijs, 2002), the scenario method approaches the matter from the perspective of a user’s actual physical actions.

2.

LITERATURE

A number of theories about the interaction between users and appliances can be found in the literature. One is described by van Aken et al. (1996). This theory describes interactions between different factors in the phase of use of a product. When the user interacts with a product, the outcome leads to desired and undesired output. For example, the desired output of a coffeemaker is coffee; noise, vibrations and waste can be seen as its undesired outputs. It is not only the user who comes in contact with the product, but other people near the product can also have contact with the product. This is why ‘contact group’ is used instead of ‘user’. The interaction between the contact group and the product takes place in a certain environment of use. This environment influences the contact group and the product, and both product and contact group influence the environment of use. Figure 23-1 shows a graph of the described theory:

environment of use contact group

operation information

product

Figure 23-1. Interaction between contact group and product within its environment of use (van Aken et al., 1996)

The scenario method to gain insight

235

Benedyk and Minister (1998) describe the BeSafe method, a method developed to estimate and reduce risks in an industrial context. By adjusting this method to one useful for consumer products, a structural safety analysis can be made for these products. When using the BeSafe method for consumer products the use of a product should be described. To make a complete description of use, information is needed about different groups of users and the physical and social environment of use. In the case of consumer products, these factors are variable and cannot be controlled by any standard (Benedyk and Minister, 1998). In Figure 23-2 the most important parts of the BeSafe method are shown. location and environment of use

assessment of product use

characteristics of user

environmental audit

task analysis

target population audit

description of product use

Figure 23-2. Part of the BeSafe method for evaluation of consumer products (Benedyk and Minister, 1998)

Dirken (1999) describes the interaction between the user and the product in the following way: information goes from the product to the user of the product, and operating influences go from the user to product. In the process of the use of a product different phases of use can be distinguished. Examples of these phases are: making ready for use, using, cleaning and storing. In each phase of use the information and the operation are different, so all different phases have to be taken into account (Dirken, 1999). It was not only van Aken et al. (1996), Benedyk and Minister (1998) and Dirken (1999) who wrote about the interaction between users and appliances. So did Weegels (1996), Green and Jordan (1999), Freudenthal

Chapter 23

236

(1999), Stanton (1995), ter Hark (1996), Jordan (1998), Kanis (2000), Norris and Wilson (1999), Säde (1999), and Norman (1990). In every theory three factors that influence the interaction between user and product are mentioned. These factors are: product, user and environment of use. Most of the time these factors are described as follows: examples of these factors are reach, length, and force, or many different features of human beings have to be taken into account, like age, body size, experience, etc. It seems difficult to give a clear description of the influencing factors. Everyone talks about ‘etceteras’, but it is remarkable that no one describes how the different factors influence product use. Different authors provide these remarks themselves. Weegels (1996) writes that the factors in literature are conflicting, and not always convincing. Ter Hark (1996) notes that it is impossible to give a complete description of use circumstances. About the features of users, Dirken (1999) writes that the relation between the chance of accidents and user features is not proven, and that research results on this subject differ.

3.

THE SCENARIO METHOD

From discussions with designers and developers of products, and in the literature, the concept ‘use scenario’ is defined as follows: it is a description of all the possible physical actions of a consumer while using a product. Every theory in the literature pays attention to different aspects of use. For the most complete view of the process of use of a product, it helps to combine the different theories. The basic theories are the interaction theory of van Aken et al. (1996), the BeSafe method of Benedyk and Minister (1998), and the notion of phases of use by Dirken (1999). Van Aken et al. (1996) indicate in their model the factors that influence product use. Most of the other models in the literature use the model of van Aken et al. (1996) as their basis. The different authors disagree on the extensions of the model, so the basic model of van Aken et al. (1996) was chosen. The BeSafe method is a method to describe product use. Dirken (1999) describes the division of use in different phases of use. The combination of these three theories is shown in figure 23-3.

The scenario method to gain insight

237

The first and third block in Figure 23- are blocks that are added to the original BeSafe method; some other blocks have changed places. In the scenario method, several steps have to be taken. In the first step, the different phases of use of the product have to be determined. This step can be linked to the theory of Dirken (1999). The second, third, and fourth step are to describe the characteristics of the environment of use, the product, and the contact group. These steps are a combination of the BeSafe method and the theory of van Aken et al. (1996). The fifth step is to perform a task analysis, which means writing down all the physical actions a person from the contact group can do within this specific phase of use. This step is taken from the BeSafe method. The last step is to combine the actions in a certain phase of use with the characteristics of the environment of use, the product, and the contact group. The combination of the characteristics and the actions leads to a list of requirements for the product.

Determination of the phases of use, to be described in the scenario

Place of use and environment Analysis of environment

Product features

Contact group characteristics

Product analysis

Contact group analysis

Determination of product use

Task analysis

Description of product use (use scenario)

Figure 23-3. The scenario method

At different places in the literature, the phases of use are raised. All the phases found in the literature are listed. This checklist has to be used to determine the different phases of use for a certain product. For each phase of use a table can be filled (figure 23-4). Characteristics of the environment of use, of the product and of the contact group that can influence the use of a

Chapter 23

238

product are found in literature as well. These characteristics are also listed, and these checklists have to be used to determine the characteristics that are important for the use actions in a certain phase of use. So, in the end, a use scenario provides a description of all possible actions of the user during each phase of use (Wolters, 2000). In Figure 23-4, an example is shown of a part of a use scenario. First, a particular phase of use and the place of use are determined. Then the characteristics of the user, the product, and the environment that are important in this particular phase of use are taken from the checklists and filled in in the upper part of the scenario ( factors). After this, all possible actions of the user in this phase of use are written down in the left column (actions of the consumer). The combination of each action with different factors leads to a list of requirements for the product (requirements). In this way, a use scenario can be made for each phase of use.

4.

THE APPLICATION OF THE METHOD IN THE DESIGN PROCESS

During the process of making the method it became clear that the members of the design team, who are responsible for the information about the user, wanted to utilize the use scenario to map and analyse the different actions of a person using a product. They brought the information from the use scenario to the team of designers and developers so that, based on this and other information, design decisions can be made. In this way, they contributed to the development of the appliances. By writing a use scenario, the situations in which consumers and products come together and interact with one another are described. The developers can consider this information during the design and development processes. For example, in the development process it is good to know where the product can or will be used, and what are the characteristics of the environment of use. Thus, in designing kitchen appliances, it is very important to know that the environment of use can be wet, sticky and shady. These factors can have a big influence on the user’s actions or the results of their actions. So utilizing a use scenario during the design and development

The scenario method to gain insight

239

processes facilitates making a product usable and safe in many circumstances, because it makes the designer/developer more aware of influencing factors on its actual use.

1

Factors that influence use (characteristics of the user, the product and the environment) Worktop height 4 Visual capacity 7 ……. 10 …….

2

Moving parts

5

Moisture

8

…….

11

…….

3

Attention

6

Eye-hand coordination

9

…….

12

…….

a b c

Actions of the consumer add filter to coffeemaker put coffee into the filter put water in the reservoir

d e f

……. ……. …….

g

…….

h

…….

a6

Requirements place for filter is easy to reach

b3

when spilling coffee due to inlack of attention, no dangerous situation might occur

b6 c1

place for filter is easy to reach opening of the reservoir must be within easy reach when spilling water due to inattention, no dangerous situations might occur …….

c3

Figure 23-4. Example of a part of a use scenario for a coffeemaker

A year after starting to utilize use scenarios, employees working with the scenario method were asked to provide remarks and comments on the usability of this method. Their first conclusion was that the scenario method is really useful to gain structure in describing product use. With the method it appears to be possible to analyse in a structured way what users do, what they might do, and how failures can be prevented. The method is also useful in the case of risk analysis. Especially because when having written a scenario it can be proved that product safety was included during the design and development of the product. This is very important because when accidents occur, the manufacturers have to prove they worked on the safety of the product. The use scenario also appeared to be a good tool to estimate a detailed call-rate. Because the existing checklists of possible influencing factors are currently very long, and it therefore takes much time to go through the lists,

240

Chapter 23

the employees suggested making different lists of relevant characteristics for the different product groups on which they are working. In the development process of completely new product concepts, the scenario method can be used as a tool to structure interviews and to do research on future users. In these cases, the scenario method is used as a kind of steppingstone. Another way to use the scenario method is to discover the knowledge gaps about the use of a certain product. In addition, the scenario is also used as a kind of database for knowledge about product use.

REFERENCES Aken, D. van, W.A.M. Hoefnagels, A. van Randen and M.H. Sonneveld (1996). Handboek ontwerpen van veilige producten. Utrecht, Uitgeverij Lemma BV. Benedyk, R. and S. Minister (1998). Applying the BeSafe method to product safety evaluation. In: Applied Ergonomics. Vol. 29, no.1, pp. 5-13. Dirken, J.M. (1999). Productergonomie; ontwerpen voor gebruikers. Delft, Delft University Press. Freudenthal, A. (1999). The design of home appliances for young and old consumers. Delft, Delft University Press. Green, W.S. and P.W. Jordan (1999). Human factors in product design. London, Taylor & Francis. Hark, T.A. ter (1996). User interfaces voor apparaten. Den Haag, Uitgeverij ten Hagen & Stam b.v. Heijs, W.J.M. (2002). Technology and behavior: contributions from environmental psychology. This book. Jordan, P.W. (1998). An introduction to usability. London, Taylor & Francis. Kanis, H (2000). Concept dictaat gebruiksonderzoek. Delft, TU Delft. Norris, B. and J.R. Wilson (1999). Ergonomics and safety in consumer product design. In: Green, W.S. and P.W. Jordan (1999). Human factors in product design. London, Taylor & Francis. Norman, D. (1990). Dictatuur van het design. Utrecht, A.W. Bruna Uitgevers. Roozenburg, N.F.M. and Eekels, J. (1998). Productontwerpen, structuur en methoden. Utrecht, Uitgeverij Lemma BV. Säde, S. (1999). Representations of Smart Product Concepts in User Interface Design. In: Green, W.S. and P.W. Jordan (1999). Human factors in product design. London, Taylor & Francis. Stanton, N. Ecological ergonomics: understanding human action in context. In: Contemporary ergonomics 1995. London/Bristol, Taylor & Francis Ltd. Weegels, M.F. (1996). Accidents involving consumer products. Delft, Proefschrift TU Delft. Wolters, G.W. (2000). De ontwikkeling van een methode voor het schrijven van gebruiksscenario’s. Wageningen, Wageningen Universiteit.

Chapter 24 USING DESIGN ORIENTING SCENARIOS TO ANALYZE THE INTERACTION BETWEEN TECHNOLOGY, BEHAVIOR AND ENVIRONMENT IN THE SUSHOUSE PROJECT

Remke Klapwijk, Marjolijn Knot, Jaco Quist and Philip J. Vergragt

1.

INTRODUCTION

To date, most government-sponsored programs for sustainable development have been aimed at either technology and product development (EcoDesign, Cleaner Production, Sustainable Technology Development Program, Economy, Ecology and Technology Program51), or at changes in behavior and attitude (Perspectief Project, Eco Team Programme52). The combination, however, of changes in technologies, behavior, and in the organizational context, seem to be necessary for larger environmental improvements on the longer term because these changes require each other and may even reinforce other. Substantial environmental change can be achieved through system innovations or transitions that affect a system as a whole and that will always combine technological, behavioral and organizational changes (VROM et al., 2001). However, when considering 51

52

Ecodesign (Te Riele and Zweers 1994), Sustainable Technology Development Program (DTO 1997 and Weaver et al., 2000) Economy, Ecology and Technology Program (EET 2000). Perspectief: Schmidt and Postma 1999. Van Luttervelt 1998.

241 P.-P. Verbeek and A. Slob (eds.), User Behavior and Technology Development: Shaping Sustainable Relations Between Consumers and Technologies, 241-252. © 2006 Springer.

Chapter 24

242

system innovations or transitions, it is important to achieve a good understanding of how technology, behavior and environment may interact and may mutually influence each other. This understanding can be used to optimize and enhance the (environmental) benefits, to identify and prevent possible rebound effects, and to increase the congruence with societal needs. This paper presents a scenario-based approach to study the interactions between technology, behavior and environment within a framework of system innovations. A design-orienting scenario is a future image of a new system that fulfils a household function such as clothing. The presented approach pays specific attention to the organizational context in which the user uses the technologies because this context changes often radically in system innovations and affects both the technologies used and the user behavior. An example of a change in organizational context is a shift from selling products to selling services. The scenario approach is illustrated with examples from a case study on Clothing Care in the Netherlands. The scenario approach is based on a methodology that was developed in a EU-funded research project called ‘Strategies towards the Sustainable Household (SusHouse)’, and tested in nine case studies in five European countries about three household functions: Clothing Care, Nutrition and Shelter (Vergragt 2000; Green and Vergragt 2000: Quist et al., 2001). In the SusHouse project scenarios were developed to explore possible future ways of sustainable function fulfillment (Manzini and Jégou 2000), and to find ways to innovate in sustainable directions. We also report on the interactions that were revealed between the organizational context on the one hand, and the artifacts and users on the other hand, and present a classification of these interactions that can be used to study interactions of system innovations in a systematic way. In section 2 to 3 we explain how the scenarios were used to study interactions in the SusHouse project, in section 4 we present a classification of interactions before discussing conclusions in Section 5.

2.

DESIGN ORIENTING SCENARIOS IN THE SUSHOUSE PROJECT

The term Design Orienting Scenario (DOS) was used in the SusHouse project for a normative future image of a new system that inspires the

Using design orienting scenarios: SusHouse project

243

fulfillment of a household function such as clothing in a sustainable way. In order to generate the necessary creative and radical ideas for factor 20 sustainability solutions, the SusHouse project did not take the current situation as a starting point, but applied a back-casting approach that started by formulating future images for the year 2050, followed by reasoning backwards to find out what concrete activities are currently required. Design Orienting Scenarios (DOS) were developed in close co-operation with relevant stakeholders from public interest groups, companies, government and research bodies for each household function (Green and Vergragt 2002; Quist et al., 2001). DOS contain a central future vision and a description of main characteristics. They also include a storyboard that depicts the lifestyle of the household members and how they use new artifacts, technologies and services, including leasing and sharing concepts. In addition, DOS-type scenarios depict an overview of the stakeholders involved, their activities, and of novel product service proposals that support the sustainable fulfillment of the household function in that scenario. Proposals often focus on new physical artifacts in combination with new services. Technologies and arrangements influence behavior of household members, and vice versa. For example, a scenario contained the proposal to develop a wardrobe that could air, deodorize, refresh and moisten clothes. This wardrobe will affect the washing behavior of consumers while consumers may influence the design of the wardrobe or use the wardrobe for other purposes than expected. In the Netherlands, three scenarios were developed for nutrition (Quist 200), and four Clothing Care scenarios (Knot 2000). A clothing care example is given below.

Figure 24-1. Example of clothing care

Chapter 24

244

A range of scenarios were developed for each function, next to Clothing Care and Outsourcing three more DOSs have been developed for Clothing Care in the Netherlands: • Eternally Yours: limited wardrobes of high quality unique made-tomeasure clothes; maintenance and laundry is serviced; • Clothing Pool: shared stocks and shared maintenance of children’s clothing at, e.g. the neighborhood level; • Chains of Users: clothing is bought new or second hand via e-commerce, and the laundry is done at home.

Example of a SusHouse DOS: Clothing and Clothing Care Outsourcing (summary) Outerwear is owned and maintained by professional clothing service organizations. Households are provided with clean clothes via different service systems. Households can subscribe to a Clothiery in their district where can get clothes on loan; they can also subscribe to a Clothing Portfolio service, which provides a person with a set of clean clothes regularly; and it is also possible to make use of a Personal Wardrobe lease service, which provides one with a personally-selected and composed wardrobe for a longer time-span. In all cases, the laundry is done in professional laundries by the service organizations, facilitated by dirt indicators that indicate whether a piece of clothing needs to be washed and how it should be cleaned. At home, local stains can be treated with stain removers. Underwear is made of biodegradable disposable materials. Product and service proposals • The Clothiery • Disposable underwear • The Clothing Portfolio • Dirt indicators Service • Stain removers • The Personal Wardrobe Lease system

Using design orienting scenarios: SusHouse project

3.

245

STUDYING INTERACTIONS BETWEEN TECHNOLOGY, BEHAVIOR AND ENVIRONMENT

During the project, the focus was originally on developing and optimizing sustainable system innovations, and ways to achieve implementation and follow-up through stakeholder alliances. We realized only later that the scenarios were at many stages of the project used to reflect on possible interactions between technology, behavior and environment. A full description of the SusHouse process is given below (Quist et al., 2001). Here we focus on the activities that were most helpful for revealing interactions: the environmental assessment, the consumer research and the second stakeholder workshop. The environmental assessment used indicators like energy usage, energy content, water usage and waste, for evaluating the environmental improvement of complete scenarios on a system level (Bras-Klapwijk and Knot 2001 and Knot and Bras-Klapwijk 2001). It forced the research teams to detail how and how often certain actions and activities would take place, thereby making assumptions about connections between technology and behavior. Of course, these assumptions and estimations influence the environmental impact calculated, but this process makes the assumptions and estimates explicit and reveals relevant technology-behavior-environment interactions. The consumer acceptance research was performed through consumer focus groups, to explore if and how consumers were willing to apply scenarios or specific scenario elements in their present daily lives (Bode 2000), and how scenarios could be improved in terms of consumer attractiveness. Storyboards provided a snapshot of a day in the life of a household living ‘in’ a scenario, and images were used to stimulate respondents to imagine themselves in a scenario. It appeared that consumers were very capable of evaluating scenarios describing situations in the future, and connecting these to their present situation and daily lives, while the articulation of likes and dislikes could be used for improving the consumer acceptability of scenarios. Stakeholders also evaluated scenarios and assessment results as part of a second stakeholder workshop. They discussed possible interactions, their possible desired and undesirable effects, and implementation proposals. An example of an undesired effect: participants expected that lower-income households would not buy Eternally Yours Clothes, as these high-quality

246

Chapter 24

clothes are very expensive, therefore they proposed a financial arrangement as part of the service, enabling lower income households to buy Eternally Yours clothes. Looking back at the SusHouse project, reflection on interactions between technology and behavior took place from various perspectives (environment and consumer-acceptance) and by various actors (stakeholders, researchers, consumers) using the DOS as a central instrument. This was not the intended use of the scenarios, and we realized only later that the scenarios can be used more intensely for studying technology-behavior interactions on a systemic level, including service and leasing concepts. We will deliberate upon this in the next section.

4.

A CLASSIFICATION OF INTERACTIONS APPLIED TO CLOTHING CARE SCENARIOS

Traditionally, technology and behavior studies focus on the interactions between users and artifacts; for an overview, see Verbeek (2000). Artifacts are material or physical objects. In the case of system innovations, ‘users’ refers not only to consumers and household members, but also to the employees in the companies providing services in their role as professional users. The concept of a user-artifact interaction is very useful in understanding the interaction between material and non-material elements. In case of system innovations, it is important to realize that the behavior of consumers, including the amount and nature of products that they use, is influenced by the organizational context, and especially by actor relations and the way these are arranged in the form of agreements, appointments. For example, to run a Clothary system, arrangements between the Clothary and its clients have to be made ʊ such as appointments about borrowing clothes. Should the client select the clothes he/she wants every week by internet, or must that person travel to the Clothary every three weeks for a set of clothes? The answer to this will influence, for example, the transport of the clothes, and probably even the amount of clothes used. To deal with the influence of this organizational context, we propose to add the concept of organizational arrangement as a third element alongside the existing concepts of users and artifacts. By organizational arrangements we mean the relations and appointments between the different actors in the system. There is some alignment of our triple concept with the classification of Brezet et al. (2000),

Using design orienting scenarios: SusHouse project

247

who distinguish artifacts, user practices, and institutional arrangements within (product-service) systems, although further comparison is necessary. The SusHouse project made clear that organizational arrangements also influence the behavior of suppliers of products and services. This observation is supported by the literature on services; see Heiskanen and Jalas (2000) for an overview. In the SusHouse scenarios, we identified many additional interactions between organizational arrangements and the other elements that influence the environmental effects in both positive and negative sense. Having defined three elements that can interact with one other leads to a possible classification of three types of interactions: user-artifact interactions, arrangement-user interactions, and arrangement-artifact interactions (of course, it must be noted that interactions are mutual). Due to fact that conceptualization and identification of relevant interactions are still preliminary, we cannot give a comprehensive and full list of all relevant interactions yet. Although examples of interactions are provided below, namely, between the arrangement and the other two elements of the proposed temporary classification, these examples focus especially on interactions that influence the environmental effects of the system innovation. First, the organizational arrangement may interact in the following way with the behavior of consumers and the artifacts they use: • An arrangement and its artifacts attract specific consumers. The consumer acceptance tests show that different consumer groups prefer different DOSs, which also influences the difference in environmental effects between the current and the new behavior (see Knot and Jelsma 2001). The environmental saving potential is lower when the consumers have a more efficient reference behavior. • The arrangement influences the type and amount of products that are used by consumers for function fulfillment. Ownership will often lead to a more frequent use of the product but it also reduces the flexibility to select the product used for function fulfillment. A consumer may rent a large car when driving with the family, and a smaller car when going alone. On the other hand, services may lead to waste when agreements on fixed amounts or packages of products have been made. When people have to pay the laundry service anyway for 5 kg of laundry, people may fill their 5 kg laundry baskets with half-dirty items (Jelsma and Knot

248

•

•

• •

•

•

Chapter 24

2001). The environmental burden of meals in restaurants is much higher than for home-prepared meals (Quist 2000). The arrangement influences the carefulness of consumer behavior (Heiskanen and Jalas 2000). Rented and leased products may be handled more carelessly, especially if the contract includes free maintenance or replacement. A new arrangement may change the efficiency of processes conducted by the consumer. When households outsource most but not all of the work, the scale of the process in the household will decrease. E.g. when laundry is outsourced consumers will occasionally wash clothes that are urgently needed. The environmental costs per kilo of laundry will be relatively high. An arrangement influences the amount of durable goods owned by consumers. When the activities are shared or done by a professional service provider, consumers need no, smaller or fewer durables. Services that are added to the basic service or product influence the time the consumers use the product. Services that help the consumer to select adequate products may extend the period that the consumer uses the product. Color and style advice will prevent early throwing away of clothes. A change in arrangement changes time and money spent on a function. As a result, households have a different amount of money and time to spend on other activities, which may in turn influence the environmental burden per household. Combining services may influence the consumers’ behavior and/or the behavior of service providers. It may attract more consumers and offer environmentally-efficient opportunities for delivery.

The organizational arrangement also influences the processes and artifacts used by the producers and suppliers of services and goods. A shift from user-owned to renting, sharing, pooling, and selling services may have the following effects: • Arrangements based on servicing often require additional processes. For leasing, servicing, pooling and sharing, additional processes are needed ʊ such as transport of the artifacts from the service-provider to the consumer. Another type of additional processes is those needed to maintain the appearance of products: a Clothary might treat clothes with chemicals when they are returned by a consumer to improve their appearance. These additional processes may decrease the environmental benefits of a shift to services.

Using design orienting scenarios: SusHouse project

249

• The arrangement influences the scale of the task or process, and thus the technologies used. Service-providers use often more eco-efficient and modern technologies because the scale of the processes increases. For washing, professional laundries use lower doses of detergents, energy and water, and they recycle rinse water. On the other hand, professional laundries use tumble dryers while a large percentage of households dry their laundry on the line. • The arrangement influences the product quality and lifetime. It is expected that service providers will choose more for higher-quality products that have a longer lifetime and need less repair than producers selling products will. Incentives for a better product are available because revenues are not gained from selling as many products as possible, but from selling as many value or services as possible. In addition, services such as repairs, upgrading and maintenance are easier to organize for service-providers, as they own entire pools of products (Heiskanen and Jalas 2000). • The arrangement influences the intensity of the use of products. Sharing, renting and pooling enable products to be used more intensively than user-owned ones, and this leads to a lower stock and faster turnover, and thus facilitates the utilization of the best available technology (Heiskanen and Jalas 2000). Above, we have mainly described the influence of organizational arrangement on the users and artifacts used, although the interactions also take place the other way around. For example, expensive and high-quality products that are only occasionally needed stimulate sharing and renting services, as well as maintenance, repairs and second-hand services. Another important aspect is the influence of consumers on the organizational arrangement. In practice, these will co-evolve. Based on consumer response, organizational arrangements can be redesigned.

5.

CONCLUSIONS

We have described how the Design Orienting Scenarios enabled stakeholders, consumers and researchers to form a mental image of possible system innovations, and improved their understanding of the possible interactions between technology, behavior and the environmental effects of a system. The stakeholder workshops, consumer focus group discussions, and

250

Chapter 24

environmental assessments together provided a rich picture of the possible interactions. The Dutch Ministry of the Environment and others want to use the mechanism of system innovations to achieve a more sustainable society. However, the presented interactions make clear that system innovations are not always beneficial for the environment; emissions and resource use may increase tremendously as well, as seen in the case of servicing, leasing and hiring. To design sustainable system innovations, it is important to achieve a good understanding of how technology and behavior interact and influence the environmental effects of a system. User behavior, including the type and amount of artifacts used, is always influenced by the larger organizational context. In the case of system innovations, this context changes, and to account for this we introduced the concept of Organizational Arrangement. A checklist or classification of interactions between organizational arrangements and the behavior of consumers and companies has been developed and illustrated with SusHouse results from the Clothing Care case study, and these have been used to illustrate the possible interactions between organizational arrangements and the behavior of consumers and companies, especially those interactions related to environmental effects. This classification can be used as a starting point for a more systematic development and evaluation of system innovations, although it needs further elaboration. Furthermore, empirical information on interactions is important, because it is impossible to foresee all the possible interactions of a transition. Reallife case studies must provide additional information on the nature and strength of the interactions and how these can be influenced towards sustainability. In business-oriented innovation projects, steps towards these scenarios might be taken by carefully-designed experiments in which new technologies and products, services, and organizational arrangements are introduced while monitoring user behavior before, during and after this introduction. In the Wash-in project, which is a spin-off of a Clothing Care scenario, these studies are under development (Jelsma and Knot 2001).

Using design orienting scenarios: SusHouse project

251

REFERENCES Bras-Klapwijk., R.M. and J.M.C. Knot (2001). Environmental Assessment for Sustainable Households in 2050; Illustrated for Clothing, Sustainable Development, Vol. 9, No. 2, pp. 109-118. Bode, Matthias (2000). Consumers’ Acceptance Analysis of Scenarios, Final Report, SusHouse Project, Hannover University, Lehrstuhl Markt und Konsum, Delft: Delft University of Technology, TBM Faculty. Brezet, J.C., S.S. Bijma, J. Ehrenfeld and S. Silvester (2001). The Design of Eco-efficient Services. Method, Tools and review of the case study base ‘Designing Eco-efficient Services Project, Delft: TU Delft, Design for Sustainability Program, June. DTO (Program Sustainable Technology Development), (1997). DTO visie 2040-1998. Technologie, sleutel tot een duurzame welvaart, Den Haag: DTO, Ten Hagen Stam. EET (Economie, Ecologie en Technologie), maart (2000). Doorbreken naar duurzaam. Innovatie, samenwerking en kennisontwikkeling aanjagers van economie en milieu, EET, Zeist: Drukkerij Kerckebosch. Green K. and Ph. Vergragt (2002). Towards Sustainable Households: a methodology for developing sustainable technological and social innovations, Futures 34 pp. 381-400. Heiskanen, E. and M. Jalas (2000). Dematerialization Through Services: A Review and Evaluation of the Debate, The Finnish Environment, Ministry of the Environment Helsinki. Knot, M. and R.M. Bras-Klapwijk (2001). Milieuanalyse van toekomstbeelden op systeemniveau. Het SusHouse project: gebruik en onderhoud van kleding als voorbeeld, Milieu, no. 4-5, pp. 194-208. Knot, J.M.C. (2000). Sustainable Clothing Care in 2050; Clothing Care Function, The Netherlands, Final Report, Rapport to the EU, Delft: TU Delft, TPM-faculty. Knot, M. and J. Jelsma (2001). Designing environmentally efficient services: a script approach, Sustainable Services and Systems Conference: Transition towards Sustainability?, October, Amsterdam. Luttervelt, P. van (1998). The Eco team programme. A new policy instrument in perspective, www.global-action-plan.nl. Manzini E. and F. Jégou (2000). The Construction of Design Orienting Scenarios, Final Report, SusHouse project, Politecnico di Milano University, Department of Industrial Design (CIR.IS-DI-Tec), Delft: Delft University of Technology, TBM Faculty. Quist, J.N. and Ph. J. Vergragt (2000). System Innovations towards Sustainability Using Stakeholder Workshops and Scenarios, POSTI Conference, London, UK, December 1-3. Quist, J.N. (2000). Towards Sustainable Shopping, Cooking and Eating in the Netherlands, Final Report, SusHouse project. Rapport to the EU, Delft: TU Delft, TPM-faculty. Quist, J., M. Knot, W. Young, K. Green and P. Vergragt (2001). Towards Sustainable Households Using Stakeholder Workshops and Scenarios. International Journal of Sustainable Development, 4 (1) : 75-89. Riele, H. te and A. Zweers (1994). Eco-design: acht voorbeelden van milieugerichte producktontwikkeling (PROMISE), TNO Produktcentrym i.s.m. TU Delft, faculteit Industrieel Ontwerpen en NOTA, Den Haag. Schmidt, T. en A.D. Postma (1999). Minder energiegebruik door een andere leefstijl? Project Perspectief, december 1995-juni 1998, Rotterdam: CEA.

252

Chapter 24

Verbeek, P. (2000). De daadkracht der dingen. Over techniek, filosofie en vormgeving. Boom: Amsterdam 2000. Vergragt, Ph. J. (2000). Strategies towards the Sustainable Household, Final Report, SusHouse project, Delft: Delft University of Technology, TPM-faculty. VROM, (2001). Nationaal Milieubeleidsplan 4. Een wereld en een wil: werken aan duurzaamheid, Den Haag: SDU.

Chapter 25 ICT IN EVERYDAY LIFE: The Role of the User

Valerie Frissen and Marc van Lieshout

1.

INTRODUCTION

The history of innovation processes related to information and communication technologies (ICT) is an interesting mix of both massive market failures and successful and even groundbreaking innovations with the potential to bring about radical shifts in everyday life. To forecast these successes or failures is seen as extremely difficult, particularly because of the assumed whimsical and unpredictable behavior of future users. However, in our view user behavior is not so much unpredictable, but badly understood. What we need is an approach to understand the dynamics of user behavior. This enables us to understand the possible contradictions or discrepancies between the perception of user needs and behavior in the early stages of technological innovation, and the actual social practices of users in which these innovations have to find their place. To illustrate this with an example: the huge success of Short Message Services (SMS), particularly among young people, came largely as a surprise to the industry, and thus cannot be understood by looking at the way user needs and behavior were perceived in the design and marketing stages of the ICT application. In these stages, the ‘ideal user’ of SMS was imagined as a businessman who used SMS rationally and instrumentally for time-saving and planning purposes. Among young people, however, SMS became extremely popular for continuous connectivity with their peers and

253 P.-P. Verbeek and A. Slob (eds.), User Behavior and Technology Development: Shaping Sustainable Relations Between Consumers and Technologies, 253-262. © 2006 Springer.

254

Chapter 25

for creating their own symbolic environment. And this in spite of the design of the interface and service, which are not particularly suitable for these purposes, at least at first sight. To understand such ‘irrational’ consumer behavior we have to understand the social and cultural practices of young people in which these applications are embedded, and the way the design of the actual artifact/application interacts with these practices. In this contribution we present an approach towards understanding the way ICT are incorporated into the social practices of everyday life, by focusing on the role of the user. Central to this approach is how users are involved in both the design and appropriation stage of new technologies. Empirical and theoretical knowledge of user involvement will enable us to understand the incorporation (or the lack of incorporation) of ICT in everyday life. The approach also considers the usefulness of experiments with user involvement in design processes to intervene with or influence the embedding of ICT into everyday life. This approach thus consists of three key elements: 1. The understanding of the configuration of users in the design and diffusion process of ICT: in other words, the perceived role of the user in ICT-related innovation processes; 2. the understanding of the appropriation and adoption of ICT by users (or the ‘domestication’ of ICT); and 3. the use of experimentation to understand (and possibly influence) the dynamic relationship between configuration and appropriation (or: the design-domestication interface). We shall illustrate the key elements of this approach by using examples from our own research about acceptance and uses of ICT within the (present and future) home environment, specifically from user-oriented design experiments (such as the ‘Media@Home’-project).53

2.

BACKGROUND: A ‘MUTUAL SHAPING’ PERSPECTIVE

Our approach is based upon the premise that technological and social changes are mutually shaping each other. In terms of ‘technology and behavior’ this implies that we consider ICT-related behavior as a process of double articulation. On the one hand, ICT has a range of characteristics that make ICT a quite powerful enabler of behavioral changes. This, however, does not imply that ICT-related innovations determine behavior. The social 53

See: Frissen & Punie, 2001.

ICT in everyday life: the role of the user

255

context in which ICT-related behavior takes shape is, in turn, a strong enabler of (or barrier to) technological innovations. Studies of adoption of ICT in everyday life have clearly shown that many innovations are simply not accepted or not used because they fail to be incorporated into the social and cultural dynamics of everyday life, which evidently implies that their effect on behavior is to be neglected. This research also shows that innovations tend to have unexpected consequences, as users tend to ‘reconstruct’ an innovation in their own terms; which means that technology is skewed towards patterns of existing behavior. However, once a technology finds its place in the routines and practices of everyday life, it does have the potential to bring about substantial changes in those routines and practices, due to the specific characteristics of ICT. This is what we mean by ‘double articulation’. As an example of this articulation process we may look at the answering machine. When introduced on the market it was assumed to enable changes in behavior, as it would make users accessible to others all the time, independent of time and place. This is what Van Lente (in this volume) refers to as a radical anticipation of the future use of a technology: the foreseen uses of the technology do not exist already and are projected into some future ‘new’ behavior. However, in their everyday practice people tended to use the answering machine not primarily to be accessible, but to filter incoming calls and even to be selectively inaccessible (Bergman et al., 1995). The use of the device thus substantially increased individual control over communication processes. We may conclude that the answering machine, and after that voicemail, GSM and e-mail, did contribute to changes in behavior (namely the individualization of behavior), but the nature of these changes was largely unexpected. A mutual shaping perspective thus does not look for causal explanations, but considers the technology-behavior relation in terms of interactions and interdependencies, which implies an openness for ambiguities and the unintended and even contradictory effects of technology on behavior.

3.

THE DOMESTICATION OF ICT IN EVERYDAY LIFE

As a further elaboration of this mutual shaping perspective, focusing on the role of users, we take the work of Silverstone and others54 as a starting 54

See, for instance, Silverstone et al., 1992; Silverstone and Mansell, 1996; Lie, M. & Sørensen, K., 1996; and Silverstone & Hartmann, 1996-1998.

256

Chapter 25

point, and particularly the concept of domestication. This concept has empirically been particularly useful to study processes of appropriation of ICT. On a conceptual level, it makes a distinction between the configuration of users in, for instance, the design stage of an innovation process and the appropriation of a technology by users in the stage of acceptance and use. The concept of domestication refers to the capacity of individuals, households or other institutions to bring new technologies and services into their own culture, to ‘make them their own’ (cf. the metaphor of ‘taming the wild’). The domestication of ICT is seen as ‘contextualized’: it can only be understood within the dynamic and pluralistic context of everyday life. Characteristic of this approach is that technological objects are not only seen as material objects, but as having a strong symbolic value as well. An innovation is materially produced but is also ‘loaded’ with all kinds of symbolic meanings by producers, designers, marketers, etc. Thus, the user is constructed in a certain sense even before the artifact has entered the user domain (some authors, such as Madeleine Akrich, refer to this as the user script within the technology). Users have to interact with these meanings when they consider buying an ICT-based artifact, or when putting it to use in their everyday life. To understand this process of both material and symbolic acceptance, we have to reconstruct this ‘struggle over meaning’; it is this interaction between configured uses and actual uses which eventually decides what place a technology will have in everyday life. Theoretically, domestication is broken down into three dimensions: commodification, appropriation and conversion (Silverstone & Haddon, 1996). Here we will only focus on the first two dimensions.

4.

CONFIGURING USERS

The first dimension. ‘commodification’, refers to the stage in which an innovation is conceived, designed and produced, or is being created as a ‘commodity’. This usually also involves the conception of an ‘idealized user’ or some more or less explicit idea about how, when and by whom the product will be used. The implicit view among producers and designers often seems to be that the inherent technological potential of an innovation determines its uses, and therefore those who are already interested in an innovation (the innovators and early adopters) are usually conceived of as the ideal users. The assumption is that diffusion or ‘trickling down’ will eventually, and more or less naturally, seduce the other adopter groups (the followers and the late majority). As Van Lente argues (in this volume), these strong anticipations of future use can be very powerful and have the potential to firmly shape innovation processes, as is illustrated by the

ICT in everyday life: the role of the user

257

example of the aircraft industry. A strong example within the ICT-field is the notion that every two years computers need to be replaced by a new generation. As Van Lente ironically remarks, the old one is old simply because there is a new one and not because the old ones have become obsolete. Particularly if we take a close look at everyday practices of use (which are often not very sophisticated) there are no evident rational arguments to replace these old technologies every two or three years. But new computers are bought anyway. However, this does not imply that these anticipations of use are powerful in themselves. They have to resonate somehow in everyday practices of use, in this case in a culturally-shared belief in what Van Lente refers to as the ‘logic of progress’. From the perspective of a domestication approach, the technological optimism that is characteristic of the diffusion approach is unrealistic, as it fails to answer to the differing needs and everyday socio-cultural contexts of user groups. In TNO research we are currently experimenting with an approach that aims to include user perceptions as much as possible in the configuration process. In the following example this approach is described in more detail. For the Media@Home project ʊ a TNO-project aimed at developing a user-oriented approach towards the design of ICT-services for the future home-environment ʊ the challenge was to break through the more or less taken-for-granted views on ‘idealized uses’ and diffusion processes. This implied dropping a one-dimensional focus on innovators and early adopters, and developing a more sophisticated user-oriented approach towards the design of ICT services. We started by doing a socio-technological trend and scenario study on developments concerning the home environment. Based on this analysis of trends and insecurities, we distinguished strategic user groups for the future home environment and viable clusters of ICT services. Based on in-depth interviews with these four strategic user groups, we constructed scenarios for future use of multimedia services, which were also translated in scenarios for a specific integrated multimedia service concept: the ‘pro-active diary’. Although the technological concept for this diary is basically the same, the design of the interface and the services integrated in the diary are different for all groups. These scenarios were discussed with the users, and based on their comments they have been further developed into first-user requirements and second demonstrators, which were again tested and evaluated by the users.

5.

THE APPROPRIATION OF ICT

The second dimension in the domestication concept is called ‘appropriation’, which refers to the incorporation of an innovation into the

258

Chapter 25

everyday culture of, for example, a household; in other words, it refers to what actually happens after the artifact has passes the threshold of the home. Based on the body of research work in this field, Punie (2000) has developed a model that consists of factors (building blocks) that enable us to describe and explain the process of appropriation of ICT in everyday life. This model is a valuable synthesis of factors coming out of both quantitative, descriptive research, and more in-depth qualitative research. He distinguishes between factors strongly correlating with ownership and use of ICT that have come up in quantitative research (e.g. SCP, 2000; Punie, 2000), completed with more qualitative factors which were particularly described in the domesticcation research, focusing on the ‘social’ and symbolic’ dynamics within the user context (in Punie’s case: the household).

Attitude

Technological culture ICT Ownership

Sociostructural position

Knowledge & capabilities

Functional & symbolic meanings of ICTs ICT Usage - use or non-use - frequency of usage - ICT content

Time-space

Daily practices

Figure 25-1. Conceptual model for the study of ICT domestication (Punie 2000: 557)

Crucial to this approach is that the outcome of the innovation process is not predetermined nor fixed, but differentiated according to the way these different variables are lived in everyday life (in ‘contexts of use’). The long grey box constitutes the central dimension of this model ʊ as a result of both quantitative and qualitative research. It demonstrates that traditional quantitative variables concerning the socio-demographic and socio-economic position of users ʊ concerning their attitudes, knowledge, capabilities and concerning time-space dimensions ʊ do not have direct, linear relations with ICT ownership and use. These variables are more closely related to the qualitative dimensions of the context of everyday life, i.e. the technological

ICT in everyday life: the role of the user

259

culture of the household, everyday practices and the functional and symbolic meanings of ICTs. It is within this dimension that domestication occurs, and that the agency of users and non-users can be observed. In our research, we use this model to collect data about particular user groups. In these studies account is taken of both the structural variables of everyday life (sociodemographic and socio-economic variables, attitudes, knowledge and time space) and the more dynamic technological culture of the household, the functional and symbolic meanings of ICTs and, last but not least, everyday routines. •

• • • •

•

•

The ‘building blocks’ of the model are the following: A distinction between adoption (ownership) and use of ICTs, as the motives for buying or using ICTs, are seldom the same as a posteriori use (and non-use). Moreover, from an analytical point of view, the usage dimension can be differentiated according to use or non-use, frequency of usage (e.g. daily, weekly, and monthly), and use of specific applications or content. Structural dispositions of individuals and households: age, gender, ethnicity, socio-economic resources, lifestyles, stage–in-the-lifecycle, and household composition. Attitudes towards ICT and knowledge about ICT (cf. cognitive resources, SCP 2000). Spatio-temporal factors: patterns of using and organizing space and time. The technological culture of the household, meaning the way people deal with technological artifacts and applications, as this reflects, sustains or disturbs the existing social relations in and the everyday culture of households. This notion can also be used to differentiate between lowtech, middle-tech and high-tech households. The technological culture is reflected in what is seen as the ‘biography’ of technological objects and applications. Functional and symbolic characteristics of ICT, i.e. the ‘public’ meanings of ICTs created by industry, marketers, salespersons, policy makers, media, etc., that are pushing users to accept or to use ICTs in a particular way. These meanings are negotiated in private life and are feed-backed in the innovation process. Everyday practices and routines, i.e. what people usually do without reflecting about it.

In the Media@Home project described above, we used this model for collecting data about the everyday life of the four strategic user groups, and we translated these findings in scenarios for future use of ICT services by these groups.

Chapter 25

260

6.

THE DESIGN-DOMESTICATION INTERFACE AND THE USEFULNESS OF EXPERIMENTATION

The study of configuration and appropriation processes can be complemented with a more pro-active approach in what we refer to as ‘Living Labs’. Living Labs can provide opportunities to study the mutual shaping and reshaping of technology and social contexts on the spot. Reallife experimental settings are interesting from two complementary perspectives. Firstly, they add to understanding the specifics of and the interaction between configuration and appropriation processes. Secondly, studying the set-up of an experimental setting itself contributes to our understanding of what might be favorable conditions for furthering domesticcation processes, and what might be detrimental approaches. Real-life experiments offer a playing ground for designers, suppliers and users to experiment with new applications and services, and to ‘learn by experimenting’. When properly organized, these real-life experiments thus offer the opportunity to monitor both the domestication process and the learningby-doing process. This increases the sensitivity to ground ICT-applications and services in situated user needs and expectations, and thus, possibly, reduces the risk of market failures. The mutual-shaping perspective requires Living Labs to be an open space, not conditioned by a priori expectations of what might be successful uses and implementations (of services, for instance), and under what conditions. The mutual-shaping perspective requires the shaping and reshaping process to be guided by ‘learning by doing’ and ‘learning by experimenting’ strategies. Users need to be able to learn for themselves and to act as co-producers of both functional and symbolic meanings of the technologies at stake. The experimental stage should not be conditioned by what is allowed in uses, nor by whom, but should be open to unforeseen and innovative practices of use. The following project can be seen as an illustration of this Living Lab approach.

‘Digital Living Moerwijk’ is a project TNO recently started that aims to investigate how ICT services may contribute to social coping strategies of the inhabitants of a ‘less favored’ neigborhood in the city of The Hague, the Netherlands. The targeted users in this case are not very experienced ICT users, which implies that it is very difficult for them to express needs for specific ICT services. The project is set up as an experiment in which these groups are offered access to ICT, help and training facilities. In turn, participants in the project are asked to use their own experiences with the technology to interest others in their social network to get acquainted with

ICT in everyday life: the role of the user

261

ICT. The proposed method for this is to organize home ICT parties, in the tradition of the famous Tupperware parties. By doing so, a living lab of 500 participants is organically built up, at the same time as the ‘social capital’ and the social networks to keep the experiment alive are strengthened. These networks are then used to experiment with the development of ICT services that are embedded in the social situation of the participants, and with the ‘social coping’ strategies that the participants use or would like to use in their everyday life. For a year, these participants will be closely monitored through a combination of quantitative and qualitative research methods. At the same time, the development process of the project itself and the actions of the differing stakeholders will be followed through the participant observation of a researcher who is a member of the project team. The project will result in a report that focuses on learning-by-doing experiences, in the form of an ‘action guide’ for projects concerning ICT and Social Quality. At the same time, we will use the project to increase our knowledge about (the interaction between) configuration and appropriation processes. This Living Lab serves as an interface between design and domestication. The type of experiments described in the illustration more or less remind us of the ‘social experiments’ as formulated and undertaken in the action research traditions of the 1980s55. However, the focus on everyday life practices, in combination with the increased awareness of the need to give users a stake in the process of design (co-design, or co-production), distinguish these experiments from the social experiments of fifteen years ago. Therefore, we prefer the notion of a ‘living lab’ to indicate both the structured aspects within the experimental setting, and the unstructured and uncontrollable aspects that are common to everyday life practices. To bring together designers and users in a particular real life situation will not only increase our insights into the interaction between configuration and domestication processes, but, in our view, in the long term will also contribute to a more sustainable information society.

REFERENCES Bergman, S., Frissen, V. & Slaa, P. (1995). Gebruik en betekenis van de telefoon in het leven van alledag. In Rathenau Instituut (red.) Toeval of noodzaak? Geschiedenis van de overheidsbemoeienis met de informatievoorziening. Amsterdam: Cramwinckel. Bergman, S. & Frissen, V. (1997). De eindgebruiker bestaat niet. De dynamiek van het gebruik van ICT in het leven van alledag, I & I, 15(2), 68-74.

55

See for instance Qvortrup, L. (1984, 1986).

262

Chapter 25

Frissen, V. (ed.) (1997). Gender, ICTs and Everyday Life. Mutual Shaping Processes. Proceedings from COST A4/GRANITE workshop. Brussels: EC/DG XII. Frissen, V. (2000a). ICTs in the Rush Hour of Life. Acceptance, use and meanings of ICTs in ‘busy’ households. In: The Information Society, 16 (1), p. 65-76. Frissen, V., Francissen, L. & Leurdijk, A. (1998). Locating the virtual community in Dutch households of the future. Research Report for NCR Financial Services. Amsterdam: ASCoR Frissen & Punie (2001): Present users, future homes. A theoretical perspective on acceptance and use of ICT in the home environment Position Paper Media @ Home – project. Delft, TNO-STB. Lie, M. & Sørensen, K. (eds.) (1996). Making technology our own. Domesticating technology into everyday life, Oslo/Stockholm/Copenhagen/Oxford/Boston: Scandinavian University Press. Mansell, R. & Silverstone, R. (1996). Introduction, in R. Mansell & R. Silverstone (eds.), Communication by design. The politics of information and communication technologies, Oxford: Oxford University Press, 1-13. Punie, Y. (2000). Domesticatie van informatie- en communicatietechnologie. Adoptie, gebruik en betekenis van media in het dagelijks leven: Continue beperking of discontinue bevrijding? Proefschrift Vrije Universiteit Brussel, Juni. Qvortrup, L. (1984). The Social Significance of Telematics. Amsterdam: John Benjamin. Qvortrup, L. (1986). Social experiments with Information Technology: Conclusions and Recommendations. In: Ancelin, C. et al. (eds.), Social experiments with Information Theory - Proceedings of the Odense, Denmark Conference, 11 - 13 January 1986. Brussels: EEG, FAST document. Rogers, E. M. (1995) Diffusion of innovations. New York: The Free Press, Fourth Edition. Silverstone, R. & E. Hirsch (eds.) (1992). Consuming technologies. Media and information in domestic spaces. London, Routledge. Silverstone, R. & Haddon, L. (1996). Design and the domestication of ICTs: technical change and everyday life. In: Mansell, R. & Silverstone, R., (eds.), Communication by Design. The politics of Information and Communication Technologies. Oxford: Oxford University Press, p. 44-74. Silverstone, R. & Hartmann, M. (eds.) (1996-1998). The European Media Technology and Everyday Life Network (EMTEL) Working Papers. Falmer Brighton: University of Sussex. Sociaal Cultureel Planbureau, (2000). Digitalisering van de leefwereld: een onderzoek naar informatie- en communicatietechnologie en sociale ongelijkheid. Den Haag : Sociaal en Cultureel Planbureau. Woolgar, S. (1991). Configuring the user. The case of usability trials. In: Law, J. (ed.), A Sociology of Monsters. pp. 57-99. London: Routledge.

Chapter 26 USER INVOLVEMENT IN THE DEVELOPMENT OF SUSTAINABLE PRODUCT-SERVICE SYSTEMS: The Case of the Personal Mobility System “Mitka” Marjolijn Knot and Helma Luiten

1.

SUSTAINABLE PRODUCT-SERVICE SYSTEMS: TECHNOLOGY AND BEHAVIOR

For a sustainable development of production and consumption, efficiency improvements of a factor 4, factor 10 or even a factor 20 are thought necessary (Von Weiszäcker et al., 1997; Hinterberger et al., 1996; Weterings en Opschoor 1992). Such large efficiency leaps cannot be achieved by technological innovations or behavioral changes alone, but require combined changes in technology, behavior, culture and (economic) organization (Weaver et al., 2000). This implies (long-term) solutions on a systems level: interrelated sets (arrangements) of products, services, rules, organizations, supporting (infra)structures and user-practices that all together fulfill a user’s need. Traditional approaches for environmental product development or ecodesign do not seem suitable for the development of such comprehensive system strategies. Instead of individual products, complex product-service systems have to be developed. This has implications for the design process, and hence for the activities, knowledge and skills of designers. Adapted and sometimes additional design and research tools may be necessary in various phases of the design process (Brezet et al., 2001a; Brezet et al., 2001b). And

263 P.-P. Verbeek and A. Slob (eds.), User Behavior and Technology Development: Shaping Sustainable Relations Between Consumers and Technologies, 263-276. © 2006 Springer.

264

Chapter 26

because of the higher complexity and cross-business character of productservice systems, much more attention is needed for vision development and partnership building, compared to the basic product development process described, for example, by Roozenburg and Eekels (1995). In practice, designners and researchers often even have to take part in the actual business development process, to facilitate continuity (by being able to take a metaposition, beyond the commercial core businesses). See, for example, Berchicci et al. (2002), for an exploration of the network building processes during systems development. The area where also substantially different approaches seem necessary, and which is central in this paper, is the area of the consumers’ research during the development process. The sustainable product service systems may radically change the way a user satisfies certain needs. Use and users’ research is therefore very important, but also complex. The “newness” of the concept, its complexity, and long-term developments, call for reconsidering the traditional approaches. Already for simple, near-future product concepts, the newness of the concepts and unfamiliarity of the user with relevant developments may make research results less valid (Schoormans and de Bont, 1995). Instead of, or besides, use research, the users should become involved in the process as actual “co-developers” of the very solutions that are to be tested. Users’ expertise concerning daily practice is indispensable for developing appropriate and valuable solutions, and, beyond that, involvement actually facilitates understanding, and hence the evaluation of proposed solutions.

Figure 26-1. The three pillars of the users research in the development of product-service systems

User involvement in sustainable product-service systems

265

Moreover, consumer behavior and environmental gain are closely related. In order to determine the environmental gain of an innovation, an estimation of the future behavior (acceptance and use) must be made. This means that the development of solutions should be an iterative process of concept development, consumer ‘research’ and environmental assessment (see Figure 26-1). This paper discusses the use and user research during the development of “the Mitka”56, an example of a sustainable product service innovation. Section 2 presents the Mitka system and the various use(r) research activities throughout the development process. Section 3 draws some lessons concerning the roles of the user in sustainable product-service systems, and the implications for innovation processes.

2.

THE MITKA PROJECT

2.1

The system

The Mitka system is being developed in The Netherlands as a potential sustainable mobility solution for individual short-distance travel, aiming to reduce the amount of car kilometers that are driven. 80% of the car trips that are made in the Netherlands are of between 5 and 20 kilometers (Flipsen 2000). For these short distances, the toxic emissions of a car are relatively large. At the beginning of the project, the focus was limited to transportation vehicles. Later, the focus shifted to a system level (see section 2.3). The Mitka system consists of a human-powered vehicle with electrical power assistance, as well as various non-material components that should make the use of this vehicle more easy and attractive. These are, for example, product extension services, such as repair and maintenance services, product use services, such as a lease construction for the Mitka, and car sharing offers for longer distances, and extras like grocery deliveries and a transport service that brings the children to their activities. Also (infra)structural components will belong to the system, like parking places, special Mitka routes, or even bicycle highways, and specific traffic regulations. See Figure 26-2.

56

Mitka is an abbreviation that (translated) stands for: ‘Mobility solution for Individual Transportation on Short Distances’.

266

Chapter 26

The Mitka system is being developed by a consortium consisting of Kathalys57 (project management, concept development, consumer research and ergonomics), Nike headquarters in Hilversum (as lead user), Gazelle (bicycle producer), Van der Veer Designers (product development & design) and Freewheel Techniek (engineering). The development started in 1998, and is currently (autumn 2002) in the phase of producing prototypes for a pilot test. Until this phase, the project ran on subsidies from the Ministry of Transport, Public Works and Water Management, and the funding of Kathalys by the Ministry of Environment, Housing and Spatial Planning, and the Ministry of Economic Affairs.

infrastructure

vehicle …

three wheels rain protection • power • assistance

repair service

…

…

financial agreements

Mitka system

Figure 26-2. The Mitka system elements

2.2

Future- and concept-exploration: ViP and future conditioning

The first use-related research activity in the Mitka project was focused on future- and concept-orientation, by means of a “Vision in Product development” process (ViP) and a “Future Conditioning” exercise. The ViP approach (Hekkert, 1997) can contribute to creativity in finding solutions. Current solutions are to be set aside, and only needs are to be focused on. Central in the ViP process is to design not the solution, but the expected or 57

Cooperation between TNO Industrial Technology and Delft University of Technology facilitates and develops sustainable (system) innovations.

User involvement in sustainable product-service systems

267

desirable future interaction between the user and the future solution. Input from trend watchers is very important in this process. In the Mitka project, this approach was applied to develop the product characteristics of a sustainable mobility concept for 2005 (Maas, 1998). The problem definition was (temporarily and rather unconsciously) narrowed down to developing a vehicle for sustainable travel for short distances. See Figure 26-3 for some product ideas that resulted from the ViP exercise in the case of the Mitkaproject.

“Pigeon”

“Squirrel”

“Ant”

Figure 26-3. Visuals for new vehicles from the ViP process (Maas, 1998)

Next, a concept test gave a first impression of the acceptance of these ideas. Abstract sketches were presented to potential future users, who were asked to choose their preferred one. However, before those drawings were shown, the respondents were ‘brought’ to the future by means of “Future Conditioning” (Urban et al., 1996). This was important because the users had to evaluate a product that would come to the market only after about 7 years. It involved a radio play with sound and visualizations in which different characters, such as an entrepreneur, a politician, and a specialist in mobility, talked about the situation in 2005. By means of advertising commercials, three concepts were shown. Future Conditioning helps users to break away from existing solutions, current norms and present impossibilities, and also facilitates an understanding of the functioning of the new concept in everyday life.

2.3

Redefining the innovation space, choosing a client, building a partnership

The first exploration process yielded several conceptual ideas for new vehicles. At this point in the process, it was however decided to leave the vehicle level and go back to the level of the mobility system. The project team thought that only a new vehicle would not yield a solution attractive and adequate enough to become a true alternative to the car. Moreover, if a

268

Chapter 26

solution is sought in new products only, these should fit existing infrastructures. The innovation space increases if the system level is enlarged. This also increases the chance for solutions with higher environmental savings potential. So the Mitka system level was extended to encompass also (infra)structures, (organizational) arrangements and services. Furthermore, it was decided to involve an innovative client as a partner in the project. This client should be the first lead user of the new system. The new system should be developed according to this client’s special needs. The Nike headquarters office in the Netherlands was interested to play this role. They expected a parking problem in the near future, and the Mitka could also support their corporate sustainability policy. In the meantime, facilitated by the commitment of this lead user, the other partners were enrolled in the project (so far, only Kathalys was involved)58.

2.4

Testing the innovation vision: questionnaire amongst Nike employees

In March 2000, an interactive questionnaire was put on the Nike intranet. This questionnaire was meant for gathering information on the current mobility situation of the employees, as well as for gathering information on acceptance of possible vehicles and services (visualizations were shown). The respondents were able to construct their own ideal product service system by combining elements, such as two or three wheels, adding a rain protection shield, choosing between various luggage elements, and among three service packages: ‘damage control’ (reparation, maintenance), ‘comfort for you’ (call-a-car service, a shopping service, etc.) and ‘comfort at work’ (showers, etc.). This approach is also known as “interactive concept testing” (Schoormans en de Bont, 1995). See Figure 26-4. The results of this questionnaire revealed a large diversity of opinion concerning the characteristics of the vehicle. About 60 percent of current car drivers preferred a two-wheeled version, 40 percent a three-wheeled version. The project team, however, decided to develop a three-wheeler, as a twowheeled version would not be innovative and challenging enough. Moreover, a two-wheel version would be problematic for designing a roof

58

The network development process was much more complicated than can be described here. See Berchicci et al. (2002) for an elaborate description.

User involvement in sustainable product-service systems

269

construction, which was seen as crucial for the added value of this new vehicle. The questionnaire also revealed that the daily distance that people think they can cover with a Mitka, is between 5 and 15 km one way (29% of carusers at Nike live within this distance).

Figure 26-4. Two computer screens from the Nike Intra-net questionnaire

2.5

Idea generation by the potential users: group discussion

So, a three-wheel vehicle was designed. In February 2001, a 1:5 scale model (figure 26-5) was shown to nine Nike employees, who had been invited for a group discussion to discuss the Mitka and the services.

Figure 26-5. 1:5 scale model

270

Chapter 26

The discussion focused on personal opinions concerning the Mitka system, the problems that Nike employees could foresee when they would actually use the Mitka system in their own situations, and possible solutions and improvement options. To facilitate the imagination and concept evaluation, the participants were asked to imagine themselves using the Mitka in daily practice, and then to describe “the ideal ride on the Mitka from home to work…” During this exercise, they were confronted with problem situations, like a flat tire or bad weather. How would that be solved in their ideal situation? The interaction between the group members contributed to opinion development and idea generation. The group discussion yielded insights on the demand for the new system, and generated a lot of ideas for valuable services and possible arrangements. Several participants were quite enthusiastic about the design, but another relevant number thought it would be ‘too new’ for them. They were afraid to become too conspicuous, and did not feel like becoming ‘the innovators’. Mainly because of the problems that were expected concerning its maneuverrability (which was seen as one of the most important advantages of a bicycle above the car), the three-wheel design was not so much appreciated. The speed of the Mitka and the power assistance were mentioned as an important condition for use (the maximum speed should be at least 35 km/h), and it was strongly suggested to equip the vehicle with a foldable but fixed rain protection. Problems with traffic jams or parking might win them over to use the Mitka, and, furthermore, financial support, no ownership (leasing), and maintenance and repair arrangements were suggested as important conditions for adoption.

2.6

Bicycle fair; quantitative concept testing

In March 2001 a 1:1 model of the Mitka vehicle was finished (Figure 266). It was especially developed to be presented on the Fiets Rai, the largest bicycle fair in the Netherlands. It was very important for the commitment of several network partners to show the newest developments of the Mitka concept in this setting. The opportunity was taken to ask visitors of this fair for their opinion about the presented system, consisting of the model, a video-impression of the technical prototype, and a graphic presentation of some service arrangements that had been proposed by the Nike employees in the group discussion.

User involvement in sustainable product-service systems

271

The results of the Rai questionnaire revealed enthusiasm: 46% of respondents (N = 142) and even 57% of the target group (car commuters between 5 and 20 km) evaluated the Mitka system as appropriate for their situation, about 48% of the of target group stated they definitely would use the Mitka, and even 61% of the target group answered positively on the question whether they would consider to buy one. However, 33% of the target group also found the Mitka system not appropriate for their situation, and a rather large part of the target group did not have clear opinions. In general, opinions diverged a lot, as concerns desired service elements. Services that did not (yet) exist, or do not directly relate to the use of the vehicle, were generally evaluated as relatively unimportant. System elements that were rated as important were special parking places at work and at home, showers, road assistance, battery service units along the road, maintenance service, and non-ownership (leasing). A not-binding test period was perceived as very important, to become acquainted with the Mitkasystem.

Figure 26-6. The 1:1 model of the Mitka vehicle, presented at the Bicycle fair and in the interviews

2.7

Feedback from the potential users; in depth interviews

As complementary to the quantitative questionnaire, in-depth interviews were organized with twelve Nike employees to find out more qualitative information: the reasons and values behind their product and service preferences. They were asked to respond to the 1:1 model and to a short movie that showed the technical prototype in use. They were also confronted with proposals for other system elements (services, organizations and infrastructures) that were suggested as important in the group discussion and the questionnaire.

272

Chapter 26

Most of the respondents really admired the design, although they often stated it should look less heavy. Again, in a proportion that equalled the opinions in the group discussion and the questionnaire, some respondents thought the vehicle was ‘too new’ for them. However, it was striking that some of theose who were rather skeptical during the first group discussion were now quite interested. A considerable part of the respondents saw the Mitka system as a possible substitute for their second car, on condition that they could make use of a car-sharing service when necessary. Cost savings appeared as an important reason for this. The general opinion was that the Mitka system should be offered in a lease arrangement. The reasons for this were the shift of responsibility for maintenance and repair to (professional) others, and the lower (financial) thresholds for stepping in and out of the system. Technical assistance or a pick-up service along the road was, although often in second place, appreciated. Another striking result was the differentiated evaluation of the rain protection shield. There appears to be a need for a variety of rain protection gear, and, what’s more, a considerable group found rain protection not so important! Again, a desire for a two-wheel design was expressed, because of the maneuverability and existing infrastructures, and because of parking problems. A special Mitka highway was often highly appreciated, making the Mitka system much more attractive to use. Special parking facilities and less parking problems at the office appeared to be not so relevant to the Nike employees, as no-one felt or expected parking problems for their car. Safe (covered) parking space for the Mitka vehicle near home and at other places, was however mentioned as very important. This was because of (expected) parking space problems (at home), but also because there was fear of damage and theft that might ensue from the conspicuous and luxurious appearance of the vehicle.

2.8

Next steps: prototype testing, pilot testing

“Real-life” testing is indispensable for new product service system innovations like the Mitka mobility system. It proved difficult for the potential users to imagine themselves using the Mitka in actual practice. This made a realistic evaluation of the vehicle and its services rather difficult. The use and value of many system elements will only become clear during a considerable period of actual use. In the project planning, ten refined prototypes would have been available in summer 2002, for “real-life” pilot testing with Nike employees. However,

User involvement in sustainable product-service systems

273

at the moment of writing (September 2002), only one prototype is available (see Figure 26-7). This suggests that the first pilot test will unfortunately be limited to only a few (ca. 4) test drivers. The tests will take place in the next months. The project subsidies have finished at this stage of the project. The project managers hope that their role will be taken over by a regional organization that is also involved in the development of a bicycle highway between Eindhoven and Helmond (“VLITS” project). It is hoped that this partner can generate new funds for the necessary further technical and organizational developments. There are contacts with a large insurance company that expressed an interest in providing organizational and service arrangements (such as leasing contracts, pick-up services etc.). A large bicycle retail organization has been found that is prepared to deliver (lease) Mitkas for company bicycle plans and to individuals. A producer however still has to be found. Another development is that two project partners decided to investigate the technical and economical viability of a two-wheel version of the Mitka vehicle. Furthermore, it is being investigated whether the Mitka can function as part of a sustainable mobility system for tourists on the island of Texel.

Figure 26-7. The Mitka prototype, Sept. 2002

3.

CONCLUSIONS AND DISCUSSION

Testing the acceptance and needs for a product-service system turned out to be difficult. Especially evaluating the non-material elements of the system is problematic, because their contents are not easily visualized or described.

274

Chapter 26

There is a need for tools that can contribute to represent the non-material system elements. One could think of story-telling, future conditioning techniques, or role plays. The Mitka case has provided the experience that it is very useful to involve users in these exercises, in order for them to form vivid mental images and hence an understanding of the proposed systems. Illustrative is the ‘road assistance’ service, to which interviewees often responded by saying that the prototype looked quite solid and maintenancefree. However, after imagining themselves actually using the new vehicle, they often added that if it would break down, it would not be easy to repair it oneself, or to take the Mitka home by hand. So on second thought, the road-assistance (pick-up and repair) was thought necessary after all. This experience also teaches us that we should be alert for wishes that are probably not that important at first sight, but may become important during actual use. Moreover, it points out the need for developing the system by means of (stepwise) implementation; designing (non-material) system elements by trying them out. Only during actual use of the new system as a whole, the users may realize the value or problems of the various system elements. Therefore, a practical test in a natural surrounding as soon as possible is recommendable. More generally, because of its complexity, the development of a “sustainable product-service system” should be seen as a mid- or longterm learning-by-doing process, stretching out far beyond the first market implementation! The actual involvement of users in the concept development (through consumer research activities) proved also important for generating support. The newness of the Mitka system deterred potential users in the beginning. In some cases, it was quite clear that the opinion was adjusted when they heard more about it over time and when they could talk about it with others. So the consumer research activities as such, as a channel of information and opinion forming, can of itself spark enthusiasm. The acceptance seems to benefit from a slow and gradual acquaintance with the new concept. Engaged (potential) users may be the best ambassadors of the new concept. It is recommendable to contact, involve and inform users on a regular basis, and from the very beginning of the project. However, this requires a clear group of lead users. In the Mitka case these indeed are present, in the form of the Nike employees. Another important issue to be aware of, when developing product-service systems, is the possibility of any rebound effects, or unintended use. These may nullify any possible environmental saving (see for more elaboration on

User involvement in sustainable product-service systems

275

this issue Jelsma and Knot, forthcoming). In case of the Mitka, current bike users can switch to the Mitka, which implies an environmental loss. This means that the development should be focused explicitly on the car users and their needs and wishes. Especially owners of two cars saw added value in the Mitka, because the second car is relatively expensive for the functions it fulfils. This offers ideas for positioning and introduction strategies. Very important is a thorough research to the current car-users’ behavior and their values and logic behind that mobility behavior. The Mitka system should fit their daily needs and routines as closely as possible, and should also prevent unintended use. Not enough information has been gathered yet on these issues; the research so far has mostly been dedicated to evaluating the Mitka system. The approach of ‘contextual design’ (Beyer and Holtzblatt 1997; Jelsma 1999) can give valuable guidelines.59 Moreover, as mentioned above, the coming pilot experiments will yield valuable information on the appropriateness of the Mitka concept for daily needs, routines and values.

REFERENCES Berchicci, L., S. Silvester, M. Knot (2002). Innovative artefacts for sustainable mobility systems - the example of the “Mitka”, Conference Proceedings of the 10th International Conference of the Greening of Industry Network, June 23-26 2002, Göteborg, Sweden. Beyer, H. and K. Holtzblatt (1997). Contextual design: A customer-centered approach to systems designs, Academic Press/Morgan Kaufmann. Brezet, J.C., A. Bijma, J. Ehrenfeld and S. Silvester (2001a). Designing Eco-efficient Services. Ministry of the Environment and Spatial Planning / Delft University of Technology. Den Haag/Delft. Brezet, J.C., Ph. Vergragt, T. van der Horst (2001b). Kathalys: Vision on sustainable product innovation, Delft: Kathalys, TNO/TUD. CSD, the Centre for Sustainable Design, Sustainable Services & Systems (2001): Transition towards Sustainability? Proceedings of the 6th International Conference Towards Sustainable Product Design, 29-30 October 2001, Amsterdam. Flipsen, S.F.J. (2000). Eindrapportage Mobiliteitsconcept voor Individueel Transport op de Korte Afstand (MITKA), TNO Industry report. Goedkoop, M.J., C.J.G. van Hale, H.R.M. te Riele, P.J.M. Rommens (1999). Product Service Systems, Ecological and Economic Basics, Report of Pi!MC, Storrm C.S. and Pré consultants, in assignment of the Dutch Ministries of Environmental and Economic affairs. Den Haag.

59

However, although the design process should be directed towards the most relevant target groups, it may be crucial to reach a critical mass of innovative users. The most relevant users (in our case, the car drivers) not always want to be the lead users, they may prefer to see themselves as followers. The Mitka system was often perceived as too new and too conspicuous.

276

Chapter 26

Hekkert, P. (1997). The VIP method; (a method he developed for industrial designers, Delft University of Technology, Faculty of Industrial Design Engineering. Delft. Hinterberger, F., F. Luks, M. Stewen (1996). Ökologische Wirtschaftspolitik zwischen Ökodiktatur und Umweltkatastrophe, Birkhäuser Verlag, Basel. Jelsma, J. (1999). Huishoudelijk energiegebruik: van beter gedrag naar beter ontwerpen. Een aanzet tot een integrale benadering. Utrecht: NOVEM Gammaprogramma. Jelsma, J. and M. Knot (forthcoming), Designing environmentally efficient services; a ‘script’ approach, accepted for publication in The Journal of Sustainable Product Design. Hippel, E. Von. (2001). Perspective: user toolkits for innovation. Journal of Product Innovation Management (18), p. 247-257. Maas, T. (1997). Master’s Thesis. Delft University of Technology, Faculty of Industrial Design Engineering. Delft. Mont, O. (1999). Product-Service Systems, Shifting corporate focus from selling products to selling product-services: a new approach to sustainable development, AFR-report nr. 288. Lund University. Lund. Roozenburg, N.F.M. and J. Eekels (1995). Product Design: Fundamentals and Methods, New York: Wiley. Schoormans, J. and Bont, C. de (1995). Consumentenonderzoek in productontwikkeling, Lemma B.V. Utrecht. Silvester, S., H. Brezet (2001). Experiments in Eco-efficient product-service systems design, Proceedings of 6th International Conference “Towards Sustainable Product Design”, 29-30 October, Amsterdam, p. 137-146. Urban, G.L., D. Weinberg, & Hauser, J.R. (1996). “Premarket Forecasting of Really-New Products” Journal of Marketing (60), p. 47-60. Weaver, P., L. Jansen, G. Van Grootveld, E. Van Spiegel and Ph. Vergragt (2000). Sustainable Technology Development. Greenleaf Publishing. Sheffield. Weiszäcker, E. von, A.B. Lovins, H.L. Lovins (1997). Factor Four. Doubling wealth – Halving resource use, The new report to the Club of Rome, Earthscan, London. Weterings, R.A.P.M., J.B. Opschoor (1992). The environmental capacity as a challenge to technology development, Advisory Council for Research on Nature & Environment (RMNO), Rijswijk.

Chapter 27 ETERNALLY YOURS: Some Theory and Practice on Cultural Sustainable Product Development Henk Muis

1.

INTRODUCTION

Solutions for problems concerning the natural environment are often sought in either a change in technology or a change in behavior. However, a society that is inherently sustainable must be based on an integration of technological innovation and cultural change. The challenge is to shape both technology and culture in such a way as to bring a sustainable society a bit closer (Oele, 1995). Products play an important role, as they materialize current practices. Product life cycles sometimes have become so short that new models replace the old ones before they even reach the consumer market. More and more products are discarded because the user gets annoyed, not because the product fails in its primary function. The consequence is an enormous waste of material, energy, time, space, and real innovative potential. Green product design or recycling do not really solve these problems. More fundamental attention has to be paid to aspects like the meaning of products, provoking more sustainable patterns of behavior through shape and construction of products, combining a prolonged life with innovation, economic aspects of longer life spans, and the design of sustainable practices. The aim of the Eternally Yours Foundation, founded in November 1995, is “…to generate and disseminate knowledge of the design, production and use of durable products, including related services, which through their

277 P.-P. Verbeek and A. Slob (eds.), User Behavior and Technology Development: Shaping Sustainable Relations Between Consumers and Technologies, 277-293. © 2006 Springer.

Chapter 27

278

quality and character, provoke eternal use”. The initiators were motivated by the growing consumption of material goods that is unlikely to be continued in the future. Quickly-changing trends, fashion and hypes, supported by seductive commercials, reinforce the habit of discarding products faster and faster. This paper examines the different aspects of product longevity, the reasons for disposal, and strategies that can be followed to increase the psychological lifespan of consumer products. Attention is paid to the interplay of objects and the behavior of individuals, groups and societies, introducing the concept of sustainable practices.

2.

PRODUCT LIFE EXTENSION

There are several ways in which the useful life of a product ends. The useful life or durability of a product is the only aspect of product quality with a time dimension. It has a direct impact on the frequency of repeat purchases by consumers and, consequently, affects producers turnover, consumers’ volume of accumulation of goods, and the consumption rate of resources. No other aspect of product quality is more economically nor more environmentally important than durability. The lifespan of a product hinges upon a continuous process where the consumer evaluates its overall quality in relation to the product itself, and in relation to potential substitutes. Durability is the ability of a product to perform its required function over a lengthy period of time under normal conditions of use, without excessive expenditure on maintenance or repair. The term durability is synonymous with technical life. More important, however, is the actual lifespan or service life of products. A product may become obsolete because of a system change, for instance, as occurred with tele-printers (Telex) and 5¼-inch ‘floppy’ disks. A product may become functionally obsolete when a new product has a far higher energy efficiency. It may also become psychologically obsolete when a new product looks better or feels nicer.

2.1

Material standards

Average lifespans strongly depend on material standards of living. In the United States, the average life span of a car is about 3 years, and of a washing machine, 5 years, whereas in developing countries a car is used for over 40 years, and a washing machine for 25 years (Pantzar, 1992). Rising standards of living are accompanied by rising quality of production processes and products, and, at the same time, diminishing periods of actual

Eternally yours

279

use for almost every single product. Three aspects can be considered when speaking about the lifespan or use of a product, or its longevity: technical life, economic life and psychological life.

2.2

Longevity

2.2.1

Technically

Technical life is determined by the quality of design, used materials, possibilities and costs of repair, and availability of spare parts. Paradoxically, technical life is determined to a large extent by non-technical factors. A long lifespan can make it difficult to have spare parts in stock, depending on the pace of innovation. A problem with technical lifespan is information asymmetry: buyers cannot judge lifespan in advance. Durables are experience goods. A potential buyer will look for signals that reveal the quality of the product, such as brand name, producer reputation, warranty, higher price. But she can be easily mistaken, or even fooled. Therefore, the role of consumer organizations is especially important for consumer durables (Table 27-1). Table 27-1. Estimated life span of durables Technical (DCU, 1995) audio equipment

Economical (DCU, 1996)

8 - 10

8 - 10

bicycle

15 - 20

5 - 10

gas stove

15 - 20

10 - 15

color television

8 - 10

8 - 10

refrigerator

10 - 15

8 - 10

PC, monitor

5 - 10

3-5

vacuum cleaner

10 - 15

5-8

washing machine

8 - 10

8 - 10

video recorder

5-7

clothing

1-4

carpet

3-5

furniture

3-8

2.2.2

Upgrade

Useful product life can be prolonged if it is designed in such a way that core elements have a long life and separate modules can be easily exchanged to enhance function and performance. Especially for electronic equipment

280

Chapter 27

this seems a promising strategy for producers, consumers and the environment (Bayley, 1995). Benefits for the producer include: customer loyalty, cost reduction (especially development and marketing), new business generation and product flexibility (possibility for customization). Consumers protect their investment, are more flexible, can more easily catch up with new developments, and experience more convenience (by keeping the product they are used to). For the environment, longer life obviously reduces depletion of natural resources and the amount of waste generated.

life span

technical

economical

psychological

fashion repair cost

corrosion use maintenance energy

Doesn’t work

repair

changed circumstances

personal development

interest

sell, give away

innovation

Doesn’t pay

dispose off

social mobility

Doesn’t please

upgrade

renovate

Figure 27-1. End of life scenario’s

2.2.3

Repair

Lifespan of vulnerable devices extends through better design. This is in the interests of both user and environment. Electrical tooth brushes and cordless vacuum cleaners, for instance, are disposed off only too often, because the rechargeable, but not replaceable, batteries are depleted. Retail prices of new products decline while cost of labor is rising. So more and more durable consumer goods are disposed of at the moment of their first defect, quite often before the age of 5. Repair is getting more complex

Eternally yours

281

because devices themselves and the way they are constructed both resist repair. Repair often turns out to involve replacing entire components, or even the entire product. 2.2.4

Economically

The economical life of a product is the number of years in which the product is depreciated. The end of economic life span is reached when the cost of use, maintenance and/or repair are considered too high, compared to disposal of the old one and purchasing a new product. So economic life will increase when disposal involves a high price and when new products are more costly. In reality, disposal is cheap and new products are usually cheaper than the old ones. 2.2.5

Psychologically

The end of the psychological life span of a product is reached when the user does not want to use or keep the product any longer, although it still functions properly. In this case, one could argue that, while the primary function of the product is intact, the user is no longer interested in the primary function, or the secondary functions no longer correspond with the user’s preferences. Generally speaking, there are three possibilities: the product has changed (colors fade, wear and tear, dirty, etc.), the user has changed (older, other preferences, etc.) or the circumstances have changed (new technology, other social group, fashion change, values change, etc.).

2.3

Reasons for disposal

Now, what are the reasons for disposing of actual products, once they have been purchased and used (Blonk, 1993)? People get rid of drilling machines, vacuum cleaners and washing machines because they don’t function any more and when they are too expensive to repair. On the other hand, stereos are disposed of in most cases (60-90%) because one wants more functionality. Stoves are replaced when the kitchen is being refurbished (75%). Telephones and computers are replaced because of a need for technical upgrading. Disposal is more likely if the user has difficulty finding out what the trouble is in case of malfunction (Antonides, 1988). In roughly half the cases, people get rid of the product while it is still functioning properly. In general, there are several reasons why products that are still functioning will be disposed of.

282

Chapter 27

In roughly another quarter of the cases, products could be repaired. The problem is that repair services are mostly poor, and people don’t trust the quality of repaired products any more, or are they not prepared to pay for it. In the United Kingdom, the average age of discarded household appliances ranges from 4 (mobile phones) to 12 years (fridge, cooker), and one third of them were still functioning (Cooper and Mayer, 2000). Almost half of the owners rarely or never get their products repaired, and about half of the population is of the opinion that products do not last as long as they would like.

washing machine drilling machine vacuum cleaner television raizor still functioning video not functioning properly anymore

stereo

not functioning telephone stove computer 0%

20%

40%

60%

80%

100%

Figure 27-2. Reasons for disposal

Table 27-2. Reasons for disposing of functional products Technical change: • new technology creates obsolescence of old technology (slide ruler vs. electronic calculator) • enhancements in existing technology (e.g. cathode ray tube in television) • technology diffusion (new technology is incorporated in old technology, e.g. increasing computerization) Changing consumer needs: • change in personal life (e.g. getting married, children outgrowing their clothes, toys) • old product didn’t fit in with changing environment (wrong style, colour, etc.) • old product no longer corresponds with one’s preferences or self-image External factors: • receiving a replacement as a gift • changes in complementary products

Eternally yours 2.3.1

283

Conclusion

For a majority of durable consumer products, the psychological lifespan seems to be the most important factor in discarding products.

3.

PSYCHOLOGICAL LIFE SPAN

3.1

Ageing with dignity

In The Garden of Objects (Manzini et al., 1992), Ezio Manzini speaks of human artifacts much as one would of plants, with a life of their own, and attractive and varied. Objects win our affection for their essence and for their products, like a fruit tree. Objects provide a service, and that, in turn, requires care. In this garden, technology is cultivated, as in a garden of plants nature is cultivated. According to Ezio, “We live in a world of products that require very little effort and only minimal attention, a world of disposable products, made up of objects and images that slip right past us without leaving the slightest lasting impression on our memories. Objects that clearly minimize the effort and attention demanded, but which, at the same time, produce minimal levels of quality and an enormous mass of waste and detritus. In opposition to this impoverished and wasteful type of relationship with objects, we may propose a shift in objectives. We might suggest that we begin to work to achieve, no longer a minimum of effort but a maximum of quality. A quality of relationship that requires attention and care.”

Figure 27-3. Surface is more than you see, tactility is important

284

Chapter 27

In dealing with products, we might distinguish among three different approaches: the hard product itself, its form, color, material and shape; the cultural meaning, stories and implicit directions for use; and the functions of the product in an economic environment, such as sales, customer loyalty, after sales service. The three approaches will be briefly discussed. 3.1.1

Shape and surface

Looking at a product, its shape and surface characteristics can be easily observed. In fact, a lot of products are especially designed or styled in such a way as to attract buyers while on display in the shop window (‘Kaufreiz aesthetik’). The surface of a product can also be designed to appeal to other senses: e.g. to feel soft, smooth, cold, polished, etceteras. Clino Trini Castelli may be called the pioneer of “soft” design, of the sensory experience (Giachetti, 1992). In the ‘seventies, he embarked on an investigation of the immaterial qualities of space: color, light, smell, sounds and surface, in short: “design primario”’. Castelli wants to design more rich, elite products. “We must add so much quality to the object that it cannot be used in the traditional manner and thrown away”. An important aspect of products is aesthetics. Our aesthetic judgment is based both on what we see and what we know about the perceived object, so these judgments change and evolve over time (Walker, 1995). If products are to have a long life, their aesthetic qualities must also have a long life. This may imply graceful, well-proportioned simplicity (i.e. ‘elegance’). Visual simplicity and visual comprehension facilitate the ease of disassembly, maintenance and repair.

Figure 27-4. Visual comprehension or transparency in this printer by Donald Carr

Eternally yours

285

Figure 27-5. Getting better through wear

Another aspect is the capacity of a product to absorb abuse and everyday wear and tear without detracting from its appearance. This partly depends on the ability of the surface to be maintained, cleaned, repaired, or to age gracefully. There is also the capacity of the design to maintain the user’s sense of value for the product, related to questions like the real usefulness of the product concept, fashionability, and the symbolic messages within the design. Interesting surfaces should perhaps display macro-simplicity, when viewed from a distance, and micro-complexity at short range: texture, variation in texture and color, irregularities in contours, diversity in finishes from glossy to matte, intentional ‘irregularities’. 3.1.2

Signs and scripts

It has to be kept in mind that the skin of a product is not just surface: skin has a certain thickness, it may be neat or fluffy. The shape or form of an object can refer to older models of the same kind, to its function or to style periods in the past. It can be archetypal, modern, post-modern. Characteristics of shape and surface contribute to the attachment of people to their belongings. Simple traditional products can often be used in different ways, sophisticated modern products are designed to serve just one purpose, or consist of components too complicated to be understood by the consumer (Oberascher, 1992). At present, the functional value or usefulness of a product tends to become gradually less important to the consumer. Industrial products more and more fulfill the role of cultural symbols. Color is a powerful tool to alter the symbolic value of a product. Trend colors meet the consumer’s need for change and for collective individualization.

286

Chapter 27

Figure 27-6. Archetypal vase of silicon rubber by Hella Jongerius

Most durable products are mass-produced. They hardly display individual characteristics, nor are they imbedded in daily rituals. They form no part of an extensive system of mutual references (Kockelkoren, 1993/1994). When a new technology is introduced in our world, at first it stands between us and the material reality. After a while, it may be incorporated: the same technology is then used as a means to relate to reality. If, as Latour (Latour, 1994) states, all man-made objects, both material and immaterial, are actors, like ourselves, in our own stories, maybe we should make the interaction and communication between us as rich as possible. See the contributions of Jaap Jelsma and PeterPaul Verbeek on conceptual frameworks regarding behavior and technology. To prolong the life of our objects, the cultural arena seems the most important place for action (Koskijoki, 1996). Professional tools and instruments quite often are long-lasting products, developed through close co-operation between producer and consumer (user). This is an argument for close customer relations and a human perspective in the design and planning of consumer products for long life. Another aspect in this cultural arena is the trickling-down effect. The environmental effect of the diffusion of, for instance, the unwanted habit of premature discarding of products has a far greater impact than any technological improvements. This is a vital mechanism, both in the social strata of a specific society as at a global scale. The phenomenon of ‘neophilism’ is fairly new to the world of commodities.

Eternally yours

287

Before the 17th century, the value of household furnishing increased with age. Patina was regarded a tribute to family history. It was not until the advent of the bourgeoisie and new engineering skills that newness in clothing, furnishing and homes came to be regarded as having intrinsic value.

Figure 27-7. Not just a vehicle...

Figure 27-8. Multifunctionality improves product life (Moniek Gerner)

Users can have a durable relation with their products (Verbeek and Kockelkoren, 1998). For this, objects need to be transparent, not only in

288

Chapter 27

order to repair them and thus prolong their physical life, but also in order to re-establish a relationship with their owner. Product transparency also helps to engage us in their functioning, products should not just be a machinry delivering a result, a commodity, but the device should involve our participation. Things that require our attention are called focal things. Compare, for instance, the hearth (fire place) with a central heating system. The present course of technology and product development is towards a retreat of the machinery that tends to become invisible. In fact, the trend towards more comfort and convenience is separating us more and more from the material world. An interesting approach may be to design products that have to be assembled by the consumer. It increases the users’ understanding of the construction of the product, and may also reduce the costs for the producer and facilitate upgrading, as Papanek suggests (Papanek, 1985).

Figure 27-9. Concept for electric heater (Sven Adolph)

3.1.3

Sales and services

In our industrial market society the economic function of production and consumption often seems predominant. Companies tend to think that a short lifespan is good for business. But according to Deutsch (Deutsch, 1994) marketing has four good reasons to sell products with an increased life. Firstly, an increasing share of consumers are getting tired of always catching up with the latest developments: they wait and see. Other consumers ignore the fast-changing fashion and trends, and start looking for their own thing, their own personal taste and individuality: nice, expensive, and with a longer life. Also, a large group of consumers appears to exist that values solidity

Eternally yours

289

and quality, aspects that have been ignored in the last few decades. And, last but not least: the growing concern for the environment induces consumers to look for longer life spans of their products. The challenge for designers and marketers is to provide these products without getting boring.

Figure 27-10. Hiring, not selling

As business changes towards services or add services to their products, interesting business opportunities arise. MEWA textile-service is renting cleaning cloth that remain their property and is cleaned and reused all the time. Collected oil is used for generating electricity, cloth that is really worn out is reused as new cloth. Their clients save money on waste disposal and paper work. A strong customer relation is another positive result. In many other cases, renting instead of possessing is smarter, cheaper and easier. For instance renting a pair of skis at the resort saves the trouble of buying, transporting, storing, maintaining, repairing and disposing of them, while one can always utilize the latest equipment. For the producer, a major advantage is getting a good insight into the quality and weak spots of the product. And long life can become a selling point in a market where all products tend to look alike. Offering high-quality repair and upgrading possibilities and a long period of guarantee helps to take away feelings of insecurity of the consumer. Competing on a global scale cannot be done successfully on price alone. Western industrialized countries have no advantage in that field. But they can have an advantage when it comes to core competence, knowledge, creativity and vision. Long life and innovation can be combined when products are made such that worn out parts can be easily exchanged for better ones.

290

4.

Chapter 27

PRACTICES

Objects play a role in practices, and in doing so become meaningful. Examples of practices or regimes are cooking, eating, sleeping, transportation. Unlike the discussion on product/service combinations (‘is a cash dispenser a product or a service?’) practices have more levels of interpretation. A freezer may be viewed as a cupboard where a constant temperature of 18 degrees below zero is maintained, or as a provider of freezing services, but in fact it constitutes a private supermarket in your own kitchen that is open 24 hours a day, containing a diversity of ingredients or precooked dishes (and even becoming an à-la-minute specialty restaurant in combination with a magnetron in case unexpected guests arrive…).

Figure 27-11. Advertising can sustain life span

In this latter view (Shove, 1998) a refrigerator only forms the tip of an enormous iceberg of social-economic organization: the contracting growers of deep-freeze green peas, the food industry, deep-freeze transportation with special lorries, deep-freeze islands in supermarkets, shopping with large intervals (by car because of time savings and large quantities), methods of preparation, menus independent of the season. The concept of actual

Eternally yours

291

practices provides a framework for a meaningful discussion of sustainability issues and offers a better starting point for the development of policy aimed at sustainability (Beckers et al., 2000).

4.1

Designing practices

Design or redesign of actual practices should take into account three levels of operation: individual habits, social customs and cultural tradition. The individual level is where most designers operate. On a social level, the territory of sociologists, lifestyles, social arrangements, policy arrangements, and diffusion of innovations are at stake. At the cultural or societal level, the playing field of anthropologists and philosophers, issues like focal practices, societal and cultural embedding of technology, and dominant value systems all play a part. Alternative practices must be appealing, acceptable and attainable in order to make some impact, and they should fit with shared values. Artifacts that surround us play a dominant role in this respect. Through their use we get used and attached to these objects, and we get acquainted with the world through its habits and customs (Arendt, 1958). Therefore, wherever possible, objects, but the same holds for services, infrastructures and other organizational relations, must encourage a sustainable way of living. In other words: sustainable practices must be embedded in material or organizational structures, such as concrete products, services and/or arrangements, in order to survive for a long time. The intensive cooperation of behavior scientists, industrial designers, policy makers and commercial parties is essential. If the actual practice is the central issue in the institutionalized societal system of the generation of products and services, then the distinction between the latter two disappears. Designers, entrepreneurs, service providers, health care institutions ʊ all contribute to a diversity of practices. In the business-to-business market, this is common routine, just like craftsmen in the third world. They provide what the customer needs: not a product per se, but they keep a particular industrial process going or carry out a particular activity. Actual practices can be sustained by creating or making available (through selling or hiring or leasing or lending) products, services, examples, recipes, space, opportunities, chances.

4.2

Actors

To get a practice introduced, different routes are available: the traditional product route, where the product forms a vessel for the new practice. After all, every product presupposes a number of habits, ways of behavior, infrastructure.

292

Chapter 27

Another route is through providing services. But also authorities, pressure groups, trade organizations or other non-governmental organizations can introduce or encourage new practices.

supplier

environmental organizations

government

consumer organizations

manufacturer

retail

consumer

designer

Figure 27-12. Actors in the production-consumption chain

Apart from traditional actors (designer, manufacturer, trade and institutions like environmental organizations, consumer union and government) other actors, known from societal marketing theories, play a decisive role: innovators and opinion leaders. Innovators pioneer new things or practices. Opinion leaders have an important social position, and serve as advocates of new practices (Rogers, 1995). They set the example. Because habits develop from childhood on, education is important. Socially preferred behavior is also transferred in corporate life. Its principals become aware of the societal and cultural role they play, as expressed in mission statements where corporations profile themselves as socially involved actors that do not just provide goods or services. ‘Caring capitalism’ becomes popular. Finally, the media are important in transferring knowledge, concepts and opinions on how to live, eat or move about.

REFERENCES Antonides, G. (1988). Scrapping a Durable Consumption Good, thesis, Rotterdam. Arendt, Hannah (1958). The Human Condition. Bayley, Neil (1995). Making the most of life: upgradeability, UK Centre for Economic and environmental development, August. Beckers, Theo (2000). Gert Spaargaren en Bertine Bargeman, Van gedragspraktijk naar beleidspraktijk, Tilburg, September.

Eternally yours

293

Blonk, drs. T.J. (1993). Afdankgedrag t.a.v. wit- en bruingoedprodukten, samenvatting van NRO enquêtes, conceptrapport augustus. Cooper, Tim and Kieren Mayer (2000). Prospects for household appliances. Deutsch, C. (1994). Abschied vom Wegwerfprinzip; Die Wende zur Langlebigkeit in der industriellen Produktion, Schäffer/Poeschel Verlag, Stuttgart. Dutch Consumer Union, Consumentengids juni 1995, p. 404-407. Dutch Consumer Union, Consumentengids mei 1996, p. 309. Giachetti, Mario (1992). Minder met meer, Industrieel Ontwerpen, vol. 8, februari-maart, p. 34-36. Kockelkoren, Petran (1993/1994). Naar een technische intimiteit met de dingen, Wijsgerig Perspectief 34 (1993/1994), no. 6, p. 187-193. Koskijoki, Maria and Mika Pantzar (1996). Consumers as players, consumers as scientists and consumers as artists - looking for a valid consumption critique, Seminar paper, Helsinki, 3 September. Latour, Bruno (1994). Wij zijn nooit modern geweest, pleidooi voor een symmetrische antropologie, Van Gennep, Amsterdam. Manzini, Ezio, et al. (1992). The Garden of Objects, in: Life Between Artifact and Nature: Design and the Environmental Challenge, Triennale di Milano 18 February - May, p. 126-167. Oberascher, Leonhard (1992). Ecological psychology and design - examples in architecture, furniture, texture and colour, öko-psy, Salzburg, 1st European Conference design for the environment, 21-23 September Nunspeet, workshop 2.3.1. Oele, A.P. (1995). Beleid en politiek in de ban van de techniek, Coutinho, Bussum. Pantzar, Mika (1992). The growth of product variety: a myth?, Journal of Consumer Studies and Home Economics, 16, 345-362. Papanek, V. (1985). Design for the real world: human ecology and social change, Thames and Hudson, London. Rogers (1995). Diffusion of Innovations. Shove, Elizabeth and Dale Southerton (1998). Lancaster University, Frozen in Time: convenience and the environment, paper for workshop March. Verbeek, Peter-Paul and Petran Kockelkoren (1998). De woorden of de dingen, over het platonisme in de industriële vormgeving, to be published. Walker, S. (1995). Superficial Evidence - The Environment, Product Aesthetics and Surface, DESIGN ISSUES - Volume 11, number 3, Pittsburgh, USA, Fall.

Chapter 28 DESIGNING TECHNOLOGY-BEHAVIOR INTERACTIONS Han Brezet

1.

INTRODUCTION

By profession, industrial designers are key players in the development of technology-behavior interactions. As creators of products ʊ and services ʊ for mass consumption and professional use, they are supposed to be the experts in combining the requirements of the market, on the one hand, and the opportunities of technology on the other hand. According to Roozenburg & Eekels (1995), product development is the development of a new business activity around a new product, demanding a multidisciplinary approach from marketing, innovation management, design (form-giving), ergonomics and engineering design (technology). In the product development process, a distinction is usually made between the first, more strategic phase of formulating the product-market scope and product planning, versus the second phase of strict design, in which not only the product drawings, models and/or prototypes are realized, but also the production and marketing plan. While in general this basic approach to product development and innovation still is valuable, the dynamics of technological and societal developments require a continuous update of design insights, methodology (methods and tools) and business practice. For instance, on the market side, the trend of “individualized mass consumption” requires the creation of goods that give the individual user a

295 P.-P. Verbeek and A. Slob (eds.), User Behavior and Technology Development: Shaping Sustainable Relations Between Consumers and Technologies, 295-306. © 2006 Springer.

296

Chapter 28

feeling of a special and perfect fit, whereas these goods have to be designed for mass production and distribution. Market trends and consumer preferences are very dynamic nowadays, leading to a relatively shortened product lifetime on the market. As a consequence, next to quality and price, “time-to-market” has become a decisive factor in the market success of new products. New emerging technologies such as ICT (information and communication technologies) have to be integrated more and more into products’ concepts or into their functioning within the product life-cycle. As a consequence, designers need to acquire the capacity to understand the consequences and opportunities of these technologies for new products. They will have to learn to cooperate with experts in different areas, such as computer interfaces, ICT, nano-materials, fuel cells, photo-voltaics etc. With globally operating businesses and developing markets, the industrial designer also has to learn to work in international teams located in various countries (collaborative design), and to approach the production of the new products from an international perspective (collaborative engineering). Extra design constraints are being put forward from the perspective of sustainable development. This requires the redesign of existing products and the design of completely new ones, with relatively less energy use, less toxic emissions, and improved reusability over their product life-cycle, given the same or enlarged functionality. Given the increased complexity of today’s products and design tasks, several methods, tools and practices have been suggested to assist the product developer or product developing team in the daily work. Some of them are more generally oriented, while others specifically aim at the improvement of the environmental quality of product designs. Particularly, in addition to Parts 1 and 2, in this section attention is given to the following aspects: • User Involvement; • Scenario Building; and • Intelligent Products and Services. Although not of crucial importance, it should be mentioned that the articles presented in this section have a strong focus on “sustainability” as their focal area of study and cases.

Designing technology-behavior interactions

2.

THE NEED FOR A BROADER SCOPE AND APPROACH

2.1

User involvement

297

In the view of Houkes and Vermaas, the activities of technical designers could be more efficient and could lead to bigger successes on the market, if the design of artifacts would be accepted as a subordinate part of the design of user plans, that is, orderings of actions by which users can reach goals. This requires another vision of and an adaptation of the basic approach on industrial design of Roozenburg and Eekels (1995): before the strict design phase, the development of a user plan is required! Such a plan not only helps the user to deal with the product’s physical characteristics, but anticipates extra communication needs, training, etc., related to its use and use context. Houkes and Vermaas consider designers not to be just technologists, but social engineers that also involve prospective users in the creation process, via communication and discussion of user plans. In contrast to other authors in this part of the book, such as Jelsma, they are less convinced that intelligent, behavior-steering designs are sufficient for the reduction of products’ environmental impacts: users in general will need additional information to behave more sustainably, and designers should provide this information. In their contribution, we saw Luiten and Knot bring the concept of user involvement into practice in the case study on the “MITKA”, an advanced vehicle for commuting. The MITKA can be considered a combination of a bicycle and a small car, using on the one hand human power for propulsion and, on the other hand, car ergonomics, comfort and electrical assistance as points of departure for the design. In this case, several new approaches have been applied to involve the user in the prototype design. The ViP (Vision in Product development) process helped to create a vision of the future use context and on the desirable interaction between user and unknown artifact, deepening the product vision. When assessing the first product concepts with potential users, the “Future Conditioning” tool was applied, facilitating the product in its future context. Furthermore, by means of a web-based interactive questionnaire, the employees of Nike, the lead user company in the MITKA project, have been actively involved in the design process. Via the intranet tool and additional tools, such as group discussions and in-depth interviews, they were able to create their own ideal commuter concept, including services for rain protection, luggage, road assistance, and comfort at work.

298

Chapter 28

Although the strong user involvement approach, in the case of Luiten and Knot, resulted in a successful MITKA prototype, they indicate that the actual test of the complete product-service system in its natural surrounding has proven to be vital for revealing unexpected and crucial extra users’ demands. Both contributions plead for a stronger focus on user involvement in product design and suggest new tools for that. In the case of MITKA, these tools also have been tested in practice and have proven to be of great value for creating a vision on the potential future users of new artifacts and the context in which these artifacts might operate. Only after assessing the future use context, and the (also environmentally) desirable actual use of the product, the strict development phase of the innovation process can start. It should be realized, of course, that the application of these new tools by designers is envisaged to paramount on top of a thorough understanding of strategic marketing in the first place and user behavior in particular. This also includes the need for the designer to include, in his/her toolbox, knowledge of models from psychology, such as those presented by Heijs in Chapter 5, or insights in modern feedback intervention theory, such as those presented by McCalley & Midden in Chapter 13. The demand for a strong design focus on potential future user behavior seems in line with the developments in the area of Product Life Cycle Management (PLM) of complex technical products (Euroforum, 2002). Here, management systems, supported by software tools, are helping design teams to communicate and collaborate worldwide with all stakeholders involved in the product life-cycle, such as the future users, suppliers, producers, marketers, recyclers, etc. PLM not only gives insight into technical data, such as programs of requirements and physical characteristics, but also assesses the consequences of design change proposals in terms of value creation for the involved actors. 2.2

Scenario building

It is not only important to create a vision about the future end-user, but also about all other relevant stakeholders in the technology-behaviorinteraction design process. In this respect, van Lente illustrates and stresses the importance of anticipation of use for guiding the development and design of technologies and products. Although inventors cannot predetermine the use of new technologies, and the future of products is in the hands of their users, technical expectations distribute roles to product developers and other stakeholders, such as businesses and governments, as well as users. As part

Designing technology-behavior interactions

299

of the product-system, industrial designers and other engineers are constantly aware of the opportunities and threats of emerging technologies for their field of concern. By anticipation, they are influencing the actual shaping of these technologies. According to van Lente this phenomenon could provide an opening for concerns about technology change, and could offer possibilities for more sustainable designs. Once the future is undoubtedly seen in terms of more efficiency and less pollution, rather than in terms of “faster and bigger”, the projection of future stakeholder roles could offer new options for sustainability. In this way, design and designers can and will shape technologies and products by anticipation. Scenario Building creates possibilities for anticipation, according to both Wolters and Steenbekkers and Bras-Klapwijk et al. Wolters and Steenbekkers propose “use scenarios” as a systematic method to include in the product development process the improvement of such products’ usability. A use scenario is a description of all possible physical actions of a consumer while using a product. First, the different phases of use are determined. Then, the characteristics of the context of use, of the product group, and of the contact group are described, followed by a task group description. Finally, by combining the characteristics and the actions, a list of requirements for the product is formulated. According to experiments of Wolters and Steenbekkers in design practice, the scenario method proves to be relevant for product-developing teams, particularly for improving the safety aspects of products. In future, the method could be used as a database on use characteristics, and to help to explore knowledge gaps with respect to new products. In addition, in their approach, Bras-Klapwijk et al. focus on the application of Design Oriented Scenarios (DOS) for future sustainable household functions. Here, scenarios are being applied to explore and understand the possibilities of substantial environmental changes via system innovations, combining technological, behavioral and organizational changes. This understanding is not only aimed at optimizing and enhancing environmental and societal benefits, but also at identification and prevention of possible rebound effects. Within the “Sushouse” program, scenarios have been developed for nutrition and clothing care. From the project it emerged that the future organizational arrangement interacts in various ways with the behavior of consumers and artifacts. For instance, different consumer groups prefer different DOSs, and the proposed arrangement influences the type and amount

Chapter 28

300

of products used by consumers as well as the processes conducted by the consumers, etc. Based upon the experiences in the Sushouse project, a first classification of interactions between organizational arrangements and the behavior of consumers and companies could be developed. The scenario-building methods and tools indicated above show promising potential for anticipation about future use contexts and user behavior, also applicable for issues of an environmental nature. Research on sustainable systems innovation from the Dutch Kathalys program (Brezet, Vergragt, van der Horst et al., 2001) underscores this. Poelman and Van den Hoed (2002) state that industrial design engineers have far from comprehensive and accurate knowledge on emerging new technologies. On the contrary, due to their prior history and experience with routine solutions, many technological opportunities for superior product design are not even explored or considered by industrial designers. Work of Kruijssen (1999) and Brezet and Silvester (2004) support Poelman’s research findings for several application areas: renewable energy for mobile products, nano-technology, and fuel cells. There is a lack of appropriate, systematic approaches that explore and assess the potentialities of these new technologies for the user needs of today and of the future. This pleads for a strengthening of the responsibility of designers not only from the perspective of understanding the future user, but also in exploring and assessing the potentialities of the new product-technologies of, say, the next 30 years. 2.3

Intelligent products and services

Another important topic for the design of in the end sustainable technology-behavior interactions is the integration of intelligence into products and practices. Jelsma introduces the concept of “scripts” to realize this. A script is a material structure, purposely created by the designer, which exerts force on the action of its user: the script of an artifact invites certain user behavior while preventing other behavior. Script-based design of products requires a substantial new approach, from identification of the underlying logic of the appliance “to be moralized” and its environmentally relevant use aspects, to reconstruction of user logic to be compared with script logic. Although in pilot projects the user involvement of this design approach has not been assessed as uncommon, practicing it in a framework of “moralization” was perceived as new. In the

Designing technology-behavior interactions

301

vision of Jelsma, the script approach breaks away from the classical separation in product design between functionality (related to things) and intentions (located in humans). The article of Muis relates to the design of intelligent services rather than products. In this case, intelligence refers to the generation and dissemination of knowledge on the design, production and use of durable products, which provoke life-long use vs. disposable products. Nowadays consumers discard roughly 50% of their products for purely psychological reasons; (these are products that are still functioning properly). Therefore, Muis proposes a design practice that takes into account signs and scripts for provoking “ageing with dignity” of products, services, and the systems they are functioning in.

Frissen and van Lieshout study the process of domestication of ICT for everyday life use in households. The assumption in their approach is that users will reshape an (ICT-based) artifact to establish an appropriate functional and symbolic meaning within daily practices. This reshaping requires the artifact to be open to change and presupposes experimental settings in which an appropriation process for the users can be accommodated. In such a “living lab” setting, the possibility of for instance energyefficient designs might be a result, by its proper integration into a smart systems design process. All three of the abovementioned approaches include the design of intelligence into the interaction between the user and the product or service. This intelligence can be imbedded into the artifact, teaching the user the “optimal” behavior from a certain perspective or goal. But it can also be a two-way approach, where the artifact is designed in such a way that it can learn, in good communication with the user, to adapt to the user behavior characteristics, and vice versa, for the benefit of the joint behavior-productsystem performance. It is clear that these and future intelligent technology behavior interactions cannot be designed without the ongoing progress in basic technologies: it is the interplay of micro-system technology, biotechnology, nanotechnology, etc., that leads to superior product-managing possibilities and human-product interfaces, allowing new interactions in the first place. In this respect, as assessed earlier in this chapter, the role of the “technology push” influence in creating new technology behavior interactions is generally too often overlooked, underestimated and/or profited from by many actors active in “social engineering”.

Chapter 28

302

A good illustration here is the case of Aasman (2002), who foresees that in the period from today to 2010, four ICT-based trends will strongly invade the household: (1) in-house computer networks; (2) broadband contents distribution (infotainment etc.); (3) domotica and telemetrics (in-house automation of lights, heating, safety devices etc.); and (4) ambient intelligence, the integration of intelligent devices into in-house automation. All technologies offer opportunities for optimal future TBI interaction on various household activities: use of energy, leisure, entertainment, shopping, working at home, etc. However, it is not a question whether these ICT-based products or services will conquer our households and the rest of our society, but rather what product-intelligence they will incorporate, and how and when industrial designers can create the opportunity to express and bring in the needs of the future user and other actors/interests of society, such as sustainability? Given a global market that is highly dominated by competition and technological progress, complex theories or soft ideas on changing user behavior might turn out not very helpful for the few forces working inside industry capable of representing both the user and the sustainability perspective: the design engineers. Maybe some simplification c.q. streamlining of – too abstract – scientific theories and models could lead to design tools and matrices that are applicable in the daily life struggle of design engineers?

3.

CONSEQUENCES FOR INDUSTRIAL DESIGNERS

In this part of the book it has been argued that product design, as an activity mediating between user behavior and technology, is becoming a more complex task in a world of growing user demands and global competition. New skills have to be learned and new disciplines have to be integrated into the industrial design profession via a continuous learning process, going beyond basic methodologies such as those by Cross (2000) and Roozenburg and Eekels (1995). Therefore, both in educational curricula and business practice, industrial designers should broaden their scope – or specialize –, taking into account new approaches and tools: 1. user involvement; 2. scenario building; and 3. intelligent products and services.

Designing technology-behavior interactions

303

Furthermore, some other strong trends in the world of industrial design engineering are taking place, which should be shortly addressed here, due to their consequences for future designs of technology-behavior interactions.

3.1

From big industry to new ventures

Industrial design engineers experience more and more the limitations of the existing big product formulating companies, when designing the more radical, even breakthrough product and service innovations (Tischner, 2000; Brezet et al., 2001). This, for instance, is the case in the transition towards more sustainable product-service systems that should use a factor X (4-20) less materials and energy over their life-cycle, while maintaining their business and socialeconomic value. The interests and routines of most existing companies prohibit such a transition. Therefore, industrial design engineering students, but also the curricula educating and that train them, show a growing interest in creating their own business ventures, aiming at developing breakthrough innovations. Along this line, Lordkipanidze (2002) stresses the importance of linking the design of new, more sustainable systems innovation with entrepreneurship. This entrepreneurship is partly a matter of inherited characteristics, but systems designers and policy developers can create conditions and a regional climate in which “born” entrepreneurs are being more frequently confronted with innovative ideas on sustainable products, and facilitated in their work towards more sustainable businesses. Likewise, Stanford University has developed a special program that combines business sciences and product engineering as a new path to radical innovation (Feland, 2003). Therefore, it is to be expected that mastering the entrepreneurship factor will be an important element in future technology-behavior interactions design.

3.2

The factor sustainability

According to environmental scientists, products and the context in which they function (the product systems), should use a factor of 4 to 20 times fewer inputs (energy, materials, space etc.) in the year 2040 compared with today’s situation. Designers contributing to more sustainable products and services should acquire knowledge and know-how capable of the development of technology-behavior interactions contributing to these targets. In that respect,

304

Chapter 28

it is no longer sufficient to know the basic “eco-redesign or dematerialization” approaches, such as from Van Hemel (1998), that are oriented to the individual product and/or service level. In the vision of Ehrenfeld & Brezet (2001), industrial designers should focus on “sustainable satisfaction-delivery systems” as the target of their activities. Human beings seek satisfaction (completion or perfection) by pursuing intentional goals, in which artifacts (products) are involved directly or indirectly. According to these authors, industrial designers should focus on gains in the naturalistic dimension (dematerialization) simultaneously with positive results in the humanistic domain. This last includes a learning approach, and offers a set of design criteria fitting with the potential of people to “flourish”, satisfying the human striving for authenticity.

3.3

Other trends in industrial design

Finally, a few other trends in industrial design should be mentioned here. First of all, recent research from Desmet (2002) suggests a stronger focus on the design of emotions with respect to products in order to improve the adoption of new products by consumers. For this target he developed successfully a “product and emotion” navigator as an additional tool for industrial designers. Secondly, at the so-called “bottom of the pyramid” a new global market is emerging. The upcoming low-income markets of newly-industrializing countries present a prodigious opportunity for global industry ʊ to seek to create special products and services for 4 billion people in China, India, Latin America, etc., who have begun earning about 1,500 dollars/year/capita. According to Prahalad et al. (2000), the opportunity here only can be realized when sustainable novel products and services can be developed that include consumer education, the tailoring of local solutions, and new business models that must not disrupt local cultures and lifestyles.

4.

NINE GOLDEN RULES FOR RESPONSIBLE TBI DESIGN

Based upon the findings of this Chapter, I suggest the following list of guidelines for responsible industrial design engineers to take into account when developing or contributing to the realization of future technology and behavior interactions (TBIs).

Designing technology-behavior interactions

305

1. Designers of TBIs should have comprehensive knowledge of the latest user behavior models (Chapter 1 and 2 of this book). 2. Designers of TBIs should take into account the involvement of the – future – TBI users, moving from standard market studies to more advanced methods and tools in user involvement planning and simulation (Chapter 3). 3. Designers of TBIs should be ready to build scenarios and use other future-oriented vision approaches to simulate the future contexttechnology-product-service-user interaction and create a product-service vision (Chapter 3). 4. Designers of TBIs should anticipate and make optimal use of the potentialities of information and communication technologies for the realization of intelligent product-user systems (Chapter 3). 5. Designers of TBIs should anticipate and explore the potentialities of emerging product-technologies, such as photo-voltaics, fuel cells, human power and nanotechnology, as potential contributors to the sustainability of future TBI-based product-systems. 6. Designers of TBIs should develop the skills and attitudes to contribute to an optimal societal match between future user needs and the potentialities of new technologies, in the time frame from today to 30 years. 7. Designers of TBIs should not restrict their work activities inside or with the big industries, but consider a new business or venture approach as an alternative when striving to achieve radical TBIs or product innovations. 8. Designers of TBIs should integrate new insights on emotional design into their practice, understanding these new insights from the user behavior theory and models presented in this book. 9. Designers of TBIs should adopt a global view of the world: in the next decades most new TBIs will be developed for people belonging to the “bottom of the pyramid” market in newly-industrializing countries.

REFERENCES Aasman, J. (2002). Energy Efficiency Transitions in the Household. Contribution to the Transition Program of the Netherlands’ Ministry of Economic Affairs. The Hague. Brezet, J.C., Vergragt, Ph., T. Van der Horst (2001). Vision on Sustainable Product Innovation. Kathalys Program. Delft, September. Brezet, J.C. & S. Silvester (2004). Responsible Industrial Design Engineering. Proceedings of the TMCE 2004. April 12-16, Lausanne, Switzerland. Edited by Horvath and Xirouchakis. 2004 Millpress, Rotterdam, The Netherlands. Cross, N. (2000). Engineering Design Methods: Strategies for Product Design. John Wiley & Sons. Chichester. Desmet, P. (2002). Designing Emotions. Dissertation. Delft University of Technology.

306

Chapter 28

Euroforum (2002). Symposium Product Life Cycle Management 2002: Samenwerken aan efficientere productontwikkeling. WTC, Rotterdam. 12 September. Ehrenfeld, J. & J.C. Brezet (2001). Towards a new Theory and Practice of Sustainable Product-Service Systems. Proceedings of the 7th European Roundtable on Cleaner Production. IIIEE – University of Lund, May. Feland III, J.M. et al. (2003). Comprehensive Design Engineering: a new Path to Innovation. International Conference on Engineering Design ICED 03. August 19-21. Stockholm, Sweden. Hemel, C.G. van (1998). Ecodesign empirically explored. Dissertation. Delft University of Technology. Kruijsen, J. (1999). Photovoltaic Technology Diffusion. Dissertation. Delft University of Technology. Lordkipanidze, M. (2002). Enhancing Entrepreneurship in Rural Tourism for Sustainable Regional Development. Thesis. IIIEE Institute, Lund University, Sweden. Meijkamp, R. (2000). Changing Consumer Behavior trough Eco-efficient Services. Dissertation. Delft University of Technology. Poelman, W.A. & R. van den Hoed (2002). Technology Diffusion Through Product Design. In: TMCE 2002, Horvath et al. (eds.). Wuhan, China. Prahalaad, C.K. & S.L. Hart. The Fortune at the Bottom of the Pyramid. Working Paper Strategy + Business. Issue 26. www.hbsp.harvard.edu/hbr/index.html Roozenburg, N.F.M. & J. Eekels (1995). Product Design: Fundamentals and Methods. John Wiley & Sons, Chichester. Tischner, U. et al. (2000). How to do Ecodesign? Verlag form. Franfurt am Main, February.

PART 4 IMPLICATIONS FOR POLICY

Chapter 29 CITIZEN-CONSUMER ROLES IN ENVIRONMENTAL MANAGEMENT OF LARGE TECHNOLOGICAL SYSTEMS Bas van Vliet

1.

INTRODUCTION

The fourth National Environmental Policy Plan states that contemporary environmental policy-making is occupied with the management of long-term societal and technological transitions in energy and transportation systems, agriculture and our dealing with natural resources. Redesigning large technological systems requires a thorough understanding of the social processes that lie behind such dynamics. In this chapter, I wish to emphasise the interaction between energy and water systems on the one hand, and citizen-consumers as eventual users on the other. Especially in water and energy systems, consumers have so far been treated as non-descriptive ‘captive consumers’ or end-users. Due to processes of liberalisation of utility markets and ecological modernisation of production and consumption, this captive consumer role is changing into a variety of roles for consumers in the operation and environment-induced change of systems of water energy provision. I wish to emphasise not so much the role of consumers in the processes of technological change (which is done more extensively elsewhere in this volume), but rather how consumers relate to large technical systems in daily operation and management of these systems. Therefore, I will briefly explore

309 P.-P. Verbeek and A. Slob (eds.), User Behavior and Technology Development: Shaping Sustainable Relations Between Consumers and Technologies, 309-318. © 2006 Springer.

310

Chapter 29

some theories on (environmental) management of water and electricity infrastructures. By making use of the relevant elements of these approaches, and by adding empirical examples of environmental innovations in the water and electricity sectors, this chapter concludes by outlining the different roles that citizen-consumers may occupy in their dealings with energy and water systems and environmental change. It will show that environmental innovation requires the recognition of new roles for consumers: from the traditional role as ‘captive consumers’ of water and energy systems, to more participatory or co-providing roles within the technological systems that support the daily consumption of water and energy.

2.

LARGE TECHNICAL SYSTEMS AND THE ROLES OF THEIR USERS

In the debates about technology-behavior interactions, there is an overemphasis on development, while daily use, maintenance and operation only gets little attention. Therefore, I want to address the role of social actors and citizen-consumers in the daily operation of technical systems. The various management approaches of technical infrastructures that have been developed by economists and engineers take the unique technical features of network-bound systems as a starting point. Engineers focus on the technical aspects of the relationships between physical infrastructures and dependent functions: how can we design and maintain a water system that provides safe drinking water for all at any time of the day? Economists deal with allocation issues associated with the establishment and openness of infrastructure. Both engineers and economists have in common that they refer to a phenomenon on which other functions (engineering) or activities (economics) depend (Firth et al., 1999). Hence, most economic and engineering definitions of infrastructures just specify the functions they serve: facilities and related institutions that provide water or electricity. In most economic and engineering approaches, the management of systems seems to be restricted to the big actors, like managers, regulators, NGOs and the like, while citizen-consumers are the subjects of change, qualified as end-users, consumers or simply ‘the demand side’. An example of an approach that has become popular among both engineers and economists is that of Demand Side Management (DSM). DSM approaches attempt to avoid environmentally- and economically-expensive supply investment, such as a new water reservoir or power station, by

Citizen-consumer roles in environmental management

311

managing both the level and timing of demand placed on networks through the implementation of energy and water efficiency measures. Traditionally, DSM has included strategies such as load-management programmes, whereby utilities attempt to even out peak loads to improve the efficiency of their generation and transmission networks, or the promotion of energyefficiency measures to curb demand in certain ‘stressed’ areas of the network (Gellings, 1996). Connected to new resource ceilings (regulations on supplybuilding and the ‘energy crisis’), DSM has been used to describe initiatives mainly undertaken by energy utilities, to manage their resources more economically and with increased environmental efficiency, and to develop more ‘consumer-focused’ relationships (Gellings, 1996; Siohansi, 1996). Such demand-oriented approaches have provided an ideal framework for engineers for whom consumers look like kilowatt meters. The claim is that utilities have been able to develop a more integrated approach that looks at “technical options, customer needs and utility requirements” (Gellings, 1996, p. 285). Moving beyond the meter, utilities can now view consumers as a series of half-hourly or night-and-day loads who are switched on and off to match certain network capacities, or as a series of hot and cold spots on infrastructure networks (Guy and Marvin, 1996). Although ‘demand-side’ implies a more active role for consumers in utility system management ʊ compared to supply-side options ʊ little is known about what these precise roles might be. Consumer roles seem to be still tightly structured, as seen in the profusion of technical fix solutions (water-efficient toilet cisterns, energy-efficient light bulbs) or providercontrollable storage and monitoring devices (heaters and meters). Networkbound system management from a DSM perspective envisions compartmentalized consumers who operate in a relatively unchallenging utility world of neatly arranged and largely provider-controlled activities (Van Vliet Chappells and Shove, 2005). A more sociological approach towards the management of networkbound infrastructure would rather focus on the social relations that come along in the development and operation of infrastructures ʊ as Spaargaren (1997 based on Otnes, 1988) did in defining utility networks (among others) as collective socio-material systems. These relations are collective, as they connect ʊ via their networks ʊ many actors to each other, and they are social (apart from material) because only management and use by social actors legitimize systems and make them work. Such a concept is attractive as it describes network-bound systems from a dual perspective of actors and structure, rather than just describing its hardware or its functions. Installing,

312

Chapter 29

operating, using and maintaining network-bound systems are all collectively shared social practices involving competent, knowledgeable actors, including end-users and system managers. Spaargaren’s perspective on network-bound systems opposes the idea that ‘captive consumers’ have become trapped in a ‘technotope’ in which their social actions have become more and more determined by large technical systems. Rather, these systems have become the ‘public underpinnings of private life’, while the relations between consumers and providers are reciprocal: household consumption is merely the ‘serving of and being served by’ collective socio-material systems. The technical and economic bias within approaches towards infrastructure management can be resolved by taking on board such a reciprocal actor-structure perspective of social action. With the help of such a perspective, the roles of consumers in adopting and using environmental innovation in electricity and water systems can be assessed. I will do so by discussing cases of recent environmental innovation in water and electricity systems, respectively. However, first a brief theoretical intermezzo is needed to show how environmental innovation and differentiation in utility systems may proceed.

3.

INTERMEZZO: INNOVATION AND DIFFERENTIATION IN UTILITY SYSTEMS

Utility systems, such as water and electricity systems, may be simplified as large technological systems linking natural resources, providers and consumers (Van Vliet, 2002).

Cs Captive consumers

T

P Monopolist Provider

T

Rs Nonrenewable resources

Provision sphere

Consumption sphere Figure 29-1. Basic system of utility provision

Citizen-consumer roles in environmental management

313

The horizontal spine in figure 29-1 shows the relations between consumers (C), providers (P) and resources (R). Providers are the intermediaries between consumers and natural resources. Collective sociomaterial systems have emerged to mediate the provision of these resources to consumers: these include the electricity grid and water works. Located within these utility systems are a series of mediating technologies (Ts), including distribution, storage, efficiency and monitoring devices. Different combinations of these devices assist providers or consumers in managing resource flows in time and space. The role of consumers in this type of provision is rather passive: their influence on which and how energy and water are provided is almost negligible. There is no choice between technologies, providers or resources. The physically-determined cleavage between the spheres of consumption and provision is that of the utility meter. As I have more extensively argued elsewhere (Van Vliet, 2002), environmental innovation as well as liberalization of utility markets have led to a subsequent differentiation of resources to draw on (including renewables), a differentiation of providers (large and small, ‘green’ and ‘grey’), intermediate technologies (smart meters, PV-panels, rain water systems), and, consequently, a differentiation of consumer roles vis-à-vis utility systems. The picture presented in figure 29-2 can be transformed into the following:

Figure 29-2. Utility system of provision - differentiated systems of utility provision.

The barrier between spheres of consumption and provision have partially been dissolved, with the differentiation of consumer roles from captive consumers into more active modes of involvement in how utility services are delivered to them. I have proposed to distinguish four ideal types of consumer roles in this matter (Van Vliet, 2002; 2003): apart from captive consumers, environmental innovation and liberalization of utility services have enabled the roles of Customers, Citizen-consumers and Co-providers of utility services (fig. 29-3).

314

Chapter 29 Captive

Customer Cs

Citizen-Consumer

Co-provider

Figure 29-3. Differentiated Consumer roles

In the following section, these different roles are illustrated through examples of environmental innovation and differentiation in water and electricity sectors.

4.

ENVIRONMENTAL INNOVATION AND DIFFERENTIATION IN THE WATER SECTOR

Since its widespread implementation, drinking water provision to Dutch household consumers has been based on uniform water quality, provided by state-owned companies through large-scale technical networks. Water consumers have been considered ‘captive consumers’, as they are connected to a single network without having a choice between different water qualities, services, let alone providers. Although this picture is still valid for the larger part of drinking water provision and consumption, there are nowadays technological developments that could imply a change of wellestablished consumer-provider relations and consumer roles. Here, two examples will be shown. The example of household water is presented to show that not every environmental innovation automatically implies a radical change in relations between consumers and providers of networkbound services. Rainwater systems are discussed to show how these relations can change. Household water systems Since the end of the 1990s, a number of water companies have started with the construction of a second piped water system, beside normal drinking water systems, to eliminate the use of valuable drinking water for minor domestic uses (flushing, washing, gardening). A social monitoring study in one of the sites where household water was applied consisted of a survey among future users of household water and a survey among the same

Citizen-consumer roles in environmental management

315

sample of respondents after half a year of usage (Van Vliet, 2000). A great majority was just happy with having the system installed. However, some more possibilities to participate in the design of the project and on the specifications of the system would have been appreciated. Also, the information that residents received before moving to their new homes was beneath expectations. A considerable number of residents only found out about the system after moving into their new homes. So, although a differentiation into two water qualities is unique in a standardized world of water provision, it has not changed much in the relations between ‘captive’ consumers and ‘monopolist’ providers60. Rain water systems Before household water systems were introduced, there were ʊ and there are ʊ many consumers using rainwater storage systems for the same purposes as household water is meant for. The systems range from small tanks placed in the attic to huge underground tanks from which rainwater is distributed by an electric pump. What is interesting here is the social meaning of using and managing rainwater systems, compared to using and managing household water systems. While in the case of household water, water companies still provide for all water demand, the use of rainwater systems implies a partial disconnection from the water mains. Consequently, consumers become their own providers of part of their water demand. Thus, in case of rain water systems, consumers are not ‘captive water consumers’ any more, but co-providers of water.

5.

ENVIRONMENTAL INNOVATION AND DIFFERENTIATION IN ELECTRICITY PROVISION

The emergence of ‘green electricity’ is a useful phenomenon to understand newly-emerging relations between electricity consumers and providers in the operation of electricity systems. With the opening of the green electricity market for Dutch household consumers in July 2001 60 After an evaluation by the Ministry of Environment in 2003 all household water experiments came to an end. The environmental benefits of household water projects compared to single water supply were not as large as expected and appeared to be expensive in comparison with conventional single-water supply systems. Decisive factor to the cancellation of the projects was however the health risks. Due to misconnection between the drinking-water supply system and the pipes of the household water system in two experiments, people were drinking lower quality household water for quite some time.

316

Chapter 29

ʊ meaning that consumers may choose their own green electricity provider ʊ old and new electricity providers raised their advertising efforts. For the first time, consumers in one region were asked to consider a switch to another electricity provider in another region, just because the one is supposed to be ‘greener’ than the other. The environmental differentiation of sources for electricity production (that started with the introduction of green electricity in 1995) was thereby extended through a differentiation of providers. Besides, if current services do not satisfy, consumers are free to switch to another provider, which means the end of the role of captive consumer, and the start to green electricity customers. New providers range from commercial, internet-based energy retailers to idealistic associations for wind or solar energy provision. The consumer role that fits the latter kinds of green electricity provision and consumption is that of a citizen-consumer, where consumers participate in environmentally-sound systems of provision just to make electricity provision in the Netherlands a bit greener than they would be without their participation. Yet another form of green electricity provisioning is supplied by the Sunpower system. It consists of a set of PhotoVoltaic (PV) panels that can be directly installed on one’s roof and connected to the domestic electricity system by using a normal socket, making the electricity meter run backwards if more electricity is produced at a given time by the solar array than is being used in the house. The initiative of Sunpower shows that energy companies are actively trying to broaden their business: apart from being electricity distributor, they wish to develop into ‘energy service providers’. The role of consumers in such form of electricity supply is, as in the case of the mentioned rainwater systems, that of co-provider of electricity. Although this selection of green electricity schemes is not exhaustive, it provides a comprehensive picture of the current variety of green electricity resources and modes of provision. In addition, it gives us a clearer idea of the diverse relations being worked out between green electricity providers and consumers. Processes of diversification of scales are likely to continue in the electricity sector, as electricity is a flexible, easily transportable form of energy provision. With co-generators like PV panels on consumers’ roofs, the electricity grid is partly turned into a storage system of redundant electricity that is produced in the household. Instead of a one-way road, the main grid has been opened up in both directions. Green electricity schemes thus reflect changing relations between electricity consumers and providers. Depending on the precise configuration

Citizen-consumer roles in environmental management

317

of technical and organizational arrangements, consumers’ roles may range from customers, citizen-consumers or co-providers in (green) electricity service provision. In general, green electricity schemes imply more involvement among many consumers and much more involvement among a few coproviders and self-generators. This is, however, subject to the level of transparency in the ‘greenness’ of electricity, the arrangements concerning grid access for new providers, and the willingness of consumers to consider electricity consumption as a tool to ‘green up’ their lifestyles.

6.

NEW CONSUMER ROLES IN ENVIRONMENTAL INNOVATION IN WATER AND ELECTRICITY SYSTEMS

The examples of environmental innovation in water and electricity systems show that there is currently a differentiation in resources for water and electricity provision, as well as in providers of water and electricity. Besides that, consumer roles towards electricity and water provision have also been differentiated. A perspective that recognizes the reciprocity between consumers and providers of water and electricity helps us to get beyond black-box qualifications like ‘the demand side’ for citizen consumers. Apart from the ‘captive consumer’, environmental innovations in water and electricity systems have led to the emergence of new ideal-type roles: customers, co-providers and citizen-consumers. Identifying these new roles and its combinations helps us to specify what is and could be the involvement of consumers in environmental innovation processes in large technical systems. The ultimate use of innovations depends not just on the question of whether or not consumers will adopt a new technology. The way new technologies fit into daily household routines and the new power balances between providers and consumers should be well taken into account. The division between theories of technological change and of management and operation of large-scale technological systems should therefore be dissolved: successful environmental innovation presupposes an eventual successful daily operation.

REFERENCES Firth, L. K. Boersma and B. Melody (1999). Infrastructure Concepts and Classifications – A Framework for Scenario Analysis of Infrastructures in an Economic Perspective. In:

318

Chapter 29

Weijnen and Ten Heuvelhof (eds.) The Infrastructure Playing Field in 2030. Proceedings of the First Annual Symposium. Noordwijk, November 19, 1998. Gellings, C.W. (1996). Then and Now ʊ The perspective of the man who coined the term ‘DSM’. Energy Policy, 24:4, pp. 285-288. Guy, S. and S.J. Marvin (1996). Transforming Urban Infrastructure Provision ʊ The emerging logic of demand side management. Policy Studies, 17:2, pp. 137-147. Otnes, P. (1988). The Sociology of Consumption. New Jersey: Humanities Press International. Siohansi, F.P. (1996). Editors Introduction - DSM in Transition: From Mandates to Markets. Energy Policy, 24:4, pp. 283-284. Spaargaren, G. (1997). The Ecological Modernisation of Production and Consumption. Essays in Environmental Sociology. Wageningen: WAU (PhD thesis). Van Vliet, B., Chappells, H. and E. Shove (2005). Infrastructures of Consumption. Environmental Innovation in the Utility Industries. London: Earthscan. Van Vliet, B. (2000), Huishoudwater in Wageningen Noord-West. Wageningen: Wageningen Universiteit, LSG Milieubeleid. Van Vliet, B. (2002). Greening the Grid. The Ecological Modernisation of Network-bound Systems. Wageningen: Wageningen University (PhD dissertation) Van Vliet, B. (2003). Differentiation and Ecological Modernisation in Water and Electricity Provision and Consumption Innovation 16:1, pp. 29-49.

Chapter 30 MODIFYING BEHAVIOR BY SMART DESIGN: The Example of the Dutch Sustainable-Safe Road System Marcus Popkema and Ingrid van Schagen

1.

INTRODUCTION

Recent work in Science & Technology Studies emphasizes the social and societal component in the development of technology. Many studies in this field of work show that the actual shape of a particular technology can be interpreted as the result of negotiations in a social setting (Bijker et al., 1987, Latour, 1997). It can even be that technologies function as the materiallization of normative positions, and as such have distinctive effects when they are in use (Popkema, Pieters & Harbers, 1997). It was Achterhuis (1995) who suggested that these insights could be made productive by turning the perspective around: if technologies are (socially) constructed in an interaction between several actors, then technologies can also be shaped with the explicit aim to serve societal goals. Technologies should explicitly become the bearers of morality, as Achterhuis stated, since human beings fall short in behaving in the ‘right’ way. Only then our safety, environment, etc., can effectively be taken care of. This perspective stimulated researchers to find ways to moralize about the apparatuses. As might be clear, the content of this volume can be read as an active way to develop this line of thought. When aiming for such a deliberate shaping of technology for societal goals, it needs to be clarified how designers could reason when they want to come to an actual new arrangement of technology-in-use. The scope of the

319 P.-P. Verbeek and A. Slob (eds.), User Behavior and Technology Development: Shaping Sustainable Relations Between Consumers and Technologies, 319-329. © 2006 Springer.

Chapter 30

320

designer needs some broadening, since the aim is no longer solely productfocused. In this new perspective, the goal is to design a new practice of product-user interaction where the possible use is pre-structured in such a way that desired behavior is induced. The most important issue then is how behavior can be influenced by smart design. This requires insight into how the influence of technology on behavior can be anticipated during the designing process. Thus far, knowledge on this topic is underdeveloped. In this paper, we try to contribute to this field of work by presenting an example of an explicit effort to alter behavior by design. It concerns the development of a sustainable-safe road system in the Netherlands. Road safety is one of the elements of Dutch traffic and transport policy. In order to stimulate and structure road safety measures, the Dutch central government introduced quantitative road safety targets: minus 50 percent fatalities and minus 40 percent serious injuries in 2010, as compared to 1986. In the early ‘nineties, it became clear that strong, innovative measures were required to bring these targets within reach. That was the immediate reason why the concept of sustainable safety was developed, aiming at a traffic system that is inherently safe (Koornstra et al., 1992). Within the sustainable safety concept, one of the basic ideas is to convey to road users ‘by design’ what kind of behavior is expected. The main aim is to prevent unintentional errors, supported by physical measures to prevent intentional violations and, as such, to minimize the chance of an accident. By following the steps devised in developing the sustainable-safe design criteria, we will explicate which line of thought can be followed if a technology designer aims to influence the users’ actual behavior when interacting with an artifact or system. We start off with a short discussion on road user behavior, and sources of error. This results in what can be called a ‘script’ for a sustainable-safe traffic system. The script for sustainable road safety can be defined as a set of assumptions that road safety experts hold on the origins of human behavior and human error in traffic. A number of examples will illustrate how the theoretical principles are being translated into practical road design guidelines. We will conclude by discussing the type of reasoning that is followed within the sustainable safety road system, in order to uncover how technology might be shaped during the design process with the aim of influencing behavior for societal purposes.

2.

HUMAN ERROR IN TRAFFIC

In a sustainable-safe traffic environment, the starting point is the road user with his physical and cognitive capabilities and limitations. The idea is

Modifying behavior by smart design

321

that all elements of the traffic system must be maximally tuned to these capabilities and limitations. In this approach, not only the capabilities and limitations of an ‘average’ road user would need to be taken into account (e.g. the 40-year-old car driver), but also those of, for example, children, the elderly, the handicapped, cyclists, pedestrians, etc. In other words, the road system has to be designed such that the capabilities of different types of road users are made productive, that the chance of error is reduced, and that the effect of error is mitigated. The human factor is an important factor in the origin of road accidents. It is estimated that in around 60 percent of accidents the main contributing factor is the road user itself, and in around 25 percent of accidents it is a combination of road user factors and road factors (Rumar, 1985). In the traditional safety approach, stemming from the industrial safety culture (Embrey, 1994; in Davidse, 2003), human factors were considered as the main cause of an accident. A lack of motivation, lack of discipline and a lack of knowledge and skills cause errors and unsafe behavior. Measures to prevent errors were based on the principle of “fitting the person to the job”, and consisted of selection, training, publicity, punishment, and reward systems. Many road safety measures were and still are based on this traditional approach, involving road safety education, driver training and licensing, road safety publicity campaigns, and police enforcement. In the human factors approach, the ergonomic approach, and, still later, the cognitive approach, the human being has a central place: “fitting the job to the person”. In this approach, accidents are considered to be likelier when the task requirements are not attuned well enough to the human abilities. Consequently, accidents can be prevented when the task requirements are designed is such a way that they take account of the physical and cognitive characteristics of the task performer. It is this latter approach that forms the basis of the sustainable road safety philosophy. In the road traffic task, three levels of tasks can be distinguished. These are the strategic level, the tactical level, and the operational level (see Table 1). The operational level contains the elementary vehicle control tasks, such as steering, shifting gears, accelerating, and decelerating, and keeping course. These types of tasks can be characterized as being skill-based and, as such, they are performed in an automatic way, requiring little cognitive effort and having a time span of a few seconds. They are not very sensitive to error. At the tactical level, decisions are being made on specific maneuvers and behavior, such as overtaking, speed choice, giving priority, and position on the road. These tasks can be described as rule based; they have a time span of some minutes and require more cognitive effort related

322

Chapter 30

to perception, information processing, and decision-making. For this reason, these types of tasks are more sensitive to error. The highest level of the road traffic task is the strategic level at which decisions are being made about the trip; for instance, about the traffic mode, the route, and the departure time. These types of tasks are knowledge-based, they have a relatively long time span, up to several days, and require conscious decisions. At this level, it is unusual to speak about errors. Decisions at this level, however, do affect safety. Table 30-1. Characteristics of different levels of the road traffic task Levels of the road traffic task Examples or related tasks Characteristics Operational level Elementary vehicle control Skill-based tasks, e.g. Short time span - steering and course keeping Automatic - shifting gears Little cognitive effort - decelerating, accelerating Tactical level Decisions about maneuvers, Rule-based e.g. Intermediate time span - overtaking or not More conscious - speed choice More cognitive effort - giving priority or not Strategic level Decisions about trip charac- Knowledge-based teristics, e.g. Long time span - route choice Conscious - modal choice Much cognitive effort - departure time

Based on this type of information, the ‘script’ (Akrich, 1992) of a sustainable-safe road design can be formulated. The script is defined as the set of assumptions held within the sustainable safety system about the behavior of road users, and the role of the road environment. The script can thus be formulated as: Road users are equipped with capabilities on the one hand, but make errors on the other. Both apply to the operational level, the tactical level and the strategic level. Skill-based tasks on the operational level can be internalized, although a certain level of errors will always occur. Rule-based tasks at the tactical level can be taught, although a certain degree of errors will always take place. Knowledge-based skills at the strategic level can properly be considered, although a certain level of mistakes will always occur. However, the characteristics of the road and the road environment can guide road users to a considerable extent, so that the chance of error is reduced and the consequences of remaining errors are mitigated. Hence, the job can largely be fitted to the road user.

Modifying behavior by smart design

3.

323

A SUSTAINABLE-SAFE TRAFFIC SYSTEM: THE THEORY

As indicated in the sustainable safety script, the overall starting point of sustainable safety is that road users will always make errors, but that the road system and the road layout may help to reduce the chance of errors and in other ways may help to be forgiving to such errors, so that they do not result in accidents or injuries: ‘fitting the road to the road user’. Three key safety principles were defined which are supposed to help meet these objectives: functionality, homogeneity and predictability. Functionality refers to the use of the road network, aiming to avoid unintentional use of the road network. The road network has to consist of a small number of road types or road categories, with each category having its own and exclusive function and its own and exclusive requirements regarding use and behavior, and these must be easy to understand and to comply with by all types of road users. In a sustainable-safe traffic system three traffic functions are distinguished: • the flow function: a through road for long distance travel at high speeds and, generally, for high volumes; • the distributor function: serving districts and regions containing scattered destinations; and • the access function: enabling direct access to properties alongside a road or street. Contrary to the common practice today, in a sustainable-safe traffic system a road is mono-functional and not multifunctional (see Table 2). The safety principle of functionality mainly aims to prevent potentially unsafe decisions at the strategic level of the road user’s task relative to route choice. Homogeneity refers to the prevention of large differences in speed, mass and direction. For example, roads with a flow function enabling high speeds for motorized traffic are closed for agricultural vehicles, since speed differences are too large. They are also closed for bicycles, since both speed and mass differences are too large. Opposing traffic streams are physically separated in order to avoid accidents by vehicles coming from opposing directions. If different traffic modes have to mix, the speed of motorized traffic has to be low. The homogeneity principle reduces the need and possibility for complex maneuvers at high speeds, and thereby errors at the tactical level.

324

Chapter 30

Table 30-2. Common practice and sustainable-safe practice of categorizing roads and streets (from: Wegman & Mulder, 1998; see also Van Schagen & Janssen, 2000) Common practice of today Existing types of Traffic function roads Motorway Increasing Through and Decreasing Access Motor road Main distributor

Sustainable-safe practice Traffic function Sustainable-safe types of roads Through Ia. Motorway

Ib. Motor road or

Local distributor

Distributor

IIa. Distributor road (rural) IIb. Distributor road (urban)

or District artery

Decreasing Through and Increasing Access

IIIa. Access road (rural)

Neighborhood artery Access Residential street

IIIb. Access road (urban)

Woonerf Residential function

Residential function

Predictability, the third key principle of a sustainable-safe traffic system is more general in nature. It is directly related to road users and the way they perceive the road system. The road network and individual roads in the network must be clear and unambiguous, and prevent uncertainties for road users. Road users immediately recognize the type of road they are traveling on; they know its function, they know what other types of road users and the type of behavior they may expect, and how they themselves should behave. The prevention of uncertainty also refers to the consistency of design along a particular stretch or road, avoiding, for example, unexpectedly narrow bends and an unexpected road-narrowing. The latter is a well-known principle in designing for safety (OECD, 1999).

Modifying behavior by smart design

4.

325

SUSTAINABLE-SAFE ROAD DESIGN: FROM THEORY TO PRACTICE

The three key safety principles of sustainable safety contain valid points to improve road safety in a structural, sustainable way. However, the main challenge was and still is to convert largely theoretical notions into guidelines for actual road design. To structure this process, based on the principles of functionality, homogeneity and predictability, 12 functional requirements for a sustainable-safe traffic system were developed. From these, as a next step, operational requirements were deduced, and these in their turn, would give direction to criteria for actual road and network design (CROW, 1997). Both empirical knowledge and ‘tacit’, expert knowledge of road engineers and behavioral scientists formed the basis for this translation from theory to design practice. A few examples may clarify the process. Three of the functional requirements stemming from the homogeneity principle are ‘avoid conflicts with oncoming traffic’, ‘avoid conflicts between transport modes of different speed and mass’ and ‘reduce speeds where conflicts can occur’. The related operational requirements are relatively straightforward. For example: • barriers, a central reserve or an extended double centre line: to separate opposing traffic streams at high speed road sections; • bicycle lanes or parallel roads for agricultural vehicles: to avoid large differences in speed and/or mass in the same physical space; • reduced speed limits, supported by physical measures, e.g. speed humps, road narrowings, roundabouts: to make road users reduce their speed at intersections and where different traffic modes have to mix ʊ such as on minor rural roads or residential streets in built-up areas. Lastly, the design criteria specify a range of technical details and optimal and minimal values of road characteristics; for example, lane width, number of lanes, distance between lanes, place of cyclists and priority regulation at roundabouts, etc. In the examples above, the majority of measures physically restrict road users’ choices about how to behave. Nevertheless, even in a fully sustainable-safe traffic system individual road users will have many degrees of freedom to make behavioral choices. The predictability principle is meant to guide road users by means of road design in making the correct and safe decisions in these situations. Two examples of functional requirements that relate to the predictability principle are ‘make road categories recognizable’ and ‘reduce the number of design solutions for similar traffic situations’. The

326

Chapter 30

operational requirements are less straightforward than those for the functional requirements for homogeneity, but include elements such as: • the same speed limit for the same road category; • the same type of road markings for the same road category; • the same type of intersection layout for the same type of road categories. Another aspect of predictability is that a road must be self-explanatory, so that road users more or less automatically behave in a correct and safe way. The issue of speed choice may clarify this issue. Speed is one of the most important aspects of road safety. Lower speeds and smaller speed differences result in fewer accidents and less severe injuries, in cases where an accident cannot be prevented. Therefore, speed management is a crucial element of sustainable safety. Posted speed limits, in combination with physical measures such as roundabouts, road humps as well as police enforcement, are all means to convince or force drivers to drive at a speed which is considered suitable for a particular road. However, it is not feasible to apply these means everywhere and always. Therefore, road design and road environments should be such that drivers automatically choose the correct driving speed. The idea of using the characteristics of the road and its environment explicitly as a means to guide road users and to influence their behavior is relatively new. Much more empirical knowledge is needed to get an overall picture of the possibilities and limitations. From the area of speed choice we know that there are indeed many opportunities. From various sources it is known that road width, surface quality, and road marking affect people’s speed choice. Thus, the approach proposed has effects. The following empirical data illustrate this: • Pavement width is a major cue for speed choice of car drivers. The average speed at a road with a pavement width of approximately 6 meters is about 80 km/h; the average speed at a road of 8 meters wide is about 90 to 100 km/h (Martens et al., 1997). • Road and pavement markings serve as a cue to show the correct path to follow and, as such, they have a beneficial effect on safety, in particular on curves. Good visual guidance also results in an increase in driving speed, at least partly overriding, in particular on rural roads, the initially positive effects (OECD, 1999).

Modifying behavior by smart design

327

Not only the characteristics of the road itself, but also the direct surroundings of the road have a large effect on the expectations of road users with respect to the desirable driving speed. For example, at a road with houses alongside it, the desirable speed was considered to be much lower than at exactly the same road, but with trees and green alongside it (Brouwer et al., 2000). This type of information explains, for example, why car drivers are not inclined to reduce their speed at 50 km/h if the sign is posted far from the houses and buildings, in the middle of the countryside, although this location may be the official border of a town. Another example is the speedreducing effect of vertical elements alongside the road, such as trees or high buildings due to the increased stimulation of peripheral vision (ETSC, 1995).

5.

CONCLUSIONS

The sustainable road safety concept was developed in the early ‘nineties in the Netherlands, with the aim to give a new impulse to road safety work, to bring the road safety targets for 2010 within reach. It took several years and much discussion among road safety experts with both engineering and behavioral backgrounds, to convert the largely theoretical notions of an inherently safe traffic system into practical guidelines for road design. The formulation of functional and operational requirements of the road system, based on existing knowledge of desired traffic behavior on the one hand, and the physical and cognitive capabilities and limitations of road users on the other, were essential intermediate steps in this process. Clearly, the resulting design criteria will always have a ‘preliminary’ status, since more knowledge and new insights on the relationship between road characteristics and road user behavior will emerge that may shed new light on the current guidelines. In this paper, the case of sustainable safety was used as an example to illustrate that, in essence, it is possible to modify (road user) behavior by smart (road) design. It shows that there are opportunities for designers to develop a technology-in-use in which the artifact induces the desired behavior in its users. Formulation of a ‘script’ may help to articulate the assumptions that could guide the design process. The example also illustrates that the inclusion of influencing behavior into the design process is not really a straightforward exercise. When it is decided which behavior is desired, and what options in conduct could lead to it, then still, at least in this example, some steps have to be made to come to an actual design. Moreover, a certain level of empirical knowledge is required about how users might interact in the new practice. Either to

328

Chapter 30

evaluate whether the desired behavior is induced, or to help the designer with developing the new technology-in-use. This paper should be read as a first exercise to unravel the design process that aims at modifying user behavior. It would be too ambitious to advance a model based on one case study. Discussing other examples might help with improving insight in this activity. Maybe then a model can be developed about how designers work in these cases. Such a model could be of great help for stimulating designers to shape technologies for societal purposes.

REFERENCES Achterhuis, H.J. (1995). De moralisering van apparaten, Socialisme en Democratie, 52(1), 9-18. Akrich, M. (1992). The De-Scription of Technical Objects. In: W.E. Bijker & J. Law (eds.), Shaping Technology/Building Society: Studies in Sociotechnical Change. Cambridge, Massachusetts: MIT Press. Brouwer, R.F.T., Janssen, W.H., & Muermans, R.C. (2000). Duurzaam veilige wegcategorieën en wegkenmerken: De invloed van de omgeving op de categorisatie van wegbeelden. Rapport TM-00-C012. Soesterberg: TNO Technische Menskunde Bijker, W., Hughes, T.P. & Pinch, T. (eds.) (1987). The Social Construction of Technological Systems. London: MIT Press. CROW (1997). Handboek categorisering wegen op duurzaam veilige basis: (voorlopige) functionele en operationele eisen. Ede, Centrum voor Regelgeving en Onderzoek in de Grond-, Water-en Wegenbouw en de Verkeerstechniek. Davidse, R.J. (2003). Op zoek naar oorzaken van ongevallen: lessen uit diverse veiligheidsdisciplines. Leidschendam: SWOV Institute for Road Safety Research. ETSC (1995). Reducing traffic injuries resulting from excess and inappropriate speed. Brussels: European Transport Safety Council. Koornstra, M.J., Mathijssen, M.P.M., Mulder, J.A.G., Roszbach, R. & Wegman, F.C.M. (1992). Naar een duurzaam veilig wegverkeer. Leidschendam: SWOV Institute for Road Safety Research. Latour, B. (1997). De Berlijnse sleutel. Amsterdam: Van Gennep. Martens, M., Comte, S. & Kaptein, N. (1997). The effect of road design on speed-behavior: a literature review. Deliverable D1 EU-project Master. Helsinki: VTT Communities & Infrastructure. OECD (1999). Safety strategies for rural roads. Paris: Organisation for Economic Cooperation and Development. Popkema, M., Pieters, T. & Harbers, H. (1997). Technologie en zwangerschap: de politiek van een prenatale screeningstest. Kennis en methode, 21(2), 97-123. Rumar, K. (1985). The role of perceptual and cognitive filters in observed behavior. In: L. Evans & R.C. Schwing (Eds.). Human behavior and traffic safety. New York: Plenum Press, 151-170. Schagen, I. & Janssen, T. (2000) managing road transport risks - sustainable safety in the Netherlands. IATSS Research, 24(2), 18-27.

Modifying behavior by smart design

329

SETRA/CETUR (1992). Sécurité des routes et des rues. Bagneux: le Services d’Etudes Techniques de Routes et Autoroutes, Centre d’Etudes des Transports Urbains. Wegman, F.C.M. & Mulder, J.A.G. (1998). Sustainable safety in the Netherlands. Report A98-10. Leidschendam: SWOV Institute for Road Safety Research.

Chapter 31 COMBINING TECHNICAL AND BEHAVIORAL CHANGE: The Role of Experimental Projects as a Step stone towards Sustainable Mobility

Boelie Elzen

1.

INTRODUCTION

A well functioning traffic and transport system is a necessity for an industrialized and complex society. The way in which the existing transport system is shaped and functions, however, causes major societal problems. Emissions of pollutants from vehicles create health hazards to humans and other living species. CO2-emissions contribute to global warming. The continuously increasing numbers of vehicles cause congestion, worsen accessibility of many destinations and threaten the livability of cities and living quarters. Since the 1960s these problems have been on the agenda of public authorities. Typically, they are split into behavioral and technological problems. Emissions are seen as a technical issue related to the engine technologies and fuels used. Congestion is primarily a behavioral issue caused by people traveling (too) much and/or choosing an inefficient mode of travel. The split strongly determines the search for solutions. Emissions are primarily targeted as the problem of the vehicle industry that is asked or forced through legislation to develop cleaner cars and other vehicles. Congestion is primarily tackled by making an appeal to people’s responsibility to society via awareness campaigns, asking them to travel less or make more use of public transport.

331 P.-P. Verbeek and A. Slob (eds.), User Behavior and Technology Development: Shaping Sustainable Relations Between Consumers and Technologies, 331-339. © 2006 Springer.

332

Chapter 31

The success of the latter approaches has been minimal. As a result, policy makers as well as many others have become increasingly skeptical towards possibilities to influence people’s travel behavior. A shift has taken place towards almost exclusively relying on technical solutions to tackle mobility problems. This shift does not remain uncontested, however. A variety of actors, especially public interest groups concerned with the environment and the livability of cities, continue to emphasize that only a change of behavior of travelers can lead to fundamental (or sustainable) solutions to traffic and transport problems. This split between technical and behavioral problems and approaches is not only inappropriate but also counter productive. More often than not, technical change and societal or behavioral change go hand-in-hand. New technologies rarely just replace existing ones. They also have new characteristics and new qualities that influence the behavior of users and other relevant actors. The personal computer, for instance, was not developed as a typewriter but once on the market it has made the typewriter obsolete in just a few years, at the same time drastically transforming office work. Socio-technical transformation, implying changes of technologies as well as behavior, in itself is nothing new. The current traffic and transport regime, for instance, differs drastically from that of 50 years ago. The question is whether it is possible to induce such a transformation and guide it in a direction with less societal problems. To answer that question in full is beyond the scope of this paper but I will provide part of the answer by emphasizing the need to learn in sociotechnical experiments. On the basis of some experiences with electric vehicle projects I will argue that there is more room to change people’s mobility behavior than is typically assumed and I will indicate how further experimentation could provide valuable information on possibilities to realize sustainable mobility.

2.

SOCIO-TECHNICAL LEARNING IN NICHES

The prevailing expectation that it is difficult if not impossible to change people’s mobility behavior does have good grounds. Such a change would require a quite drastic transformation of the present situation which is difficult to realise because a large set of interrelated barriers impede radical change. These may include:

Combining Technical and Behavioral Change • • • • • • •

333

technological factors government policy cultural and psychological factors market factors production factors infrastructure and maintenance possible undesirable societal and environmental effects of new technologies.

Despite these barriers, history shows that radical change may still take place because new technologies, when they are not (yet) ready or able to compete with existing technologies, are initially developed and experimented with in ‘protected spaces’. The new technology is protected by various actors who believe in its long-term prospects and who are willing to invest time, money, and effort in ‘making it work’, both in the technological and in the behavioral sense. Such protected spaces are called technological niches1 or just niches. These niches are experimental situations characterized by a approach of ‘learning by doing’. Especially for ‘radical technologies’ that do not simply fit an existing regime2 (e.g. because of a lacking infrastructure, misfit with existing userpreferences, etc.) development processes in niches imply a mutual tuning of technical and behavioral characteristics. Such processes thus combine technical and behavioral change already at a very early stage of development.

3.

BRIEF EXAMPLES

Some of the most illuminating examples in the transport domain deal with electric vehicles (EVs). EVs fit the current traffic and transport regime poorly. Energy is stored in large batteries that weigh of the order of several hundred kilograms which only give the vehicle a range of 100-200 km. 1

The concept of a ‘technological niche’ should not be mixed up with a ‘market niche’. The latter refers to a subsection of a larger economic market with specific characteristics, like the market for advanced sports cars. A technological niche represents a specific phase in an innovation process, preceding market development, whereas a market niche represents a specific type of market. Cf. Elzen et al. 1999. 2 The phrases ‘regime’ (to indicate a domain of socio-technical development like transport) and ‘niche’ as a breeding ground for (radical) innovation are part of the so-called multi-level perspective on innovation. Cf. Kemp et al. 2001. On niches specifically see also Hoogma 2000.

334

Chapter 31

Recharging the battery typically takes 6-8 hours although with expensive equipment this can be speeded up to the order of half an hour. The large batteries are expensive, costing several thousand euro for a typical vehicle and, depending on the battery technology, may have to be replaced one or more times during the vehicle’s lifetime. This list of negatives has made skeptics argue that EVs have been, are and always will be the technology of the future. So why bother? The major advantage of EVs is that they produce no emissions while driving. That’s why they are also referred to as ‘zero-emission vehicles’ (ZEVs). Emissions do take place at the power plant producing electricity but there they are less harmful to humans than the emissions from the vehicles running in cities. Even when counting power plant emissions the overall emissions from EVs can easily be 50-80% better than that of conventional vehicles, partly depending on the type of power plant. Across the world, a wide variety of demonstration projects have been and are carried out. Interestingly, the ‘technical drawbacks’ of EVs led to innovative thinking concerning user behavior in connection with these vehicles. Various experiments have been developed where EVs were used in a way different from conventional vehicles. In many such projects, EVs were rented for a short period for relatively short trips, either by residents of a city or people coming from elsewhere by train. Examples are the Elettra Park project in Turin, Italy, the City Car project in Martigny, Switzerland, various projects with so-called station cars in the U.S., a project by the name of Praxitèle near Paris. Several types of EVs have been used in these projects. Some were converted conventional vehicles which lead to heavy vehicles with large battery packs as in the Praxitèle and Elettra Park case. Others used specially built light-weight EVs which were relatively small (some two-seaters) and also had a much smaller battery pack3. Concerning the topic of this paper, the most striking feature of all of these project was that the users of these schemes made a significant change 3

The author has analysed a wide variety of such transportation projects within the framework of various international research projects sponsored by the EU. The most important of these were UTOPIA (Urban Transport: Options for Propulsion Systems and Instruments for Analysis; cf. http://www.utopia-eu.com/) and INTEPOL (Towards an INteractive TEchnology POLicy; cf. Elzen et al. 2001a and b). Analyses of case-studies from UTOPIA can be found in Zwaneveld (2000) and Ricci (2000).

Combining Technical and Behavioral Change

335

in their travel behavior. Whereas they used to take a car from door-to-door before, in these projects they used a combination of modes, e.g. a train with the EV. Many of these users were specifically attracted by specific features of the EV compared to a conventional vehicle, for instance because they considered it clean, silent or ‘futuristic’. Such changes of behavior were more or less designed into these projects. They can also be an unintended consequence, however, as is illustrated by developments in Switzerland in connection with lightweight electric vehicles (LEVs). In ex ante surveys, the envisaged users indicated that they saw the new option as an addition to their existing options, next to continuation of using their conventional car. As most trips in a household are short trips, the LEV could be used quite often. For a variety of reasons (quietness, environmental friendliness, novelty) many users subsequently started to prefer driving the EV to their conventional vehicle and they tried to avoid using the latter by planning trips more carefully. They tended to avoid long and energy-intensive trips and always sought the shortest way to reach a certain destination. Thus, unexpectedly even for themselves, they tried to make as many trips by LEV as possible and changed their behavior accordingly.

4.

RELEVANCE FOR PATHWAYS TOWARDS SUSTAINABLE MOBILITY

Most actors in the traffic and transport domain are skeptical about possibilities for behavioral change as they take the attachment of people to their car as a starting point. They support their view on the basis of stated preference surveys, the most widely used instrument to determine people’s transportation needs and to assess the potential of possible innovations. Such surveys, however, give a poor indication of the potential for change as people tend to think in terms of their current needs and the technologies they are familiar with. The LEV example from Switzerland illustrates that ‘hands-on’ experience in experiments may make them do things they did not anticipate. Indeed, in the present situation there is no alternative that has sufficient attraction to make the majority of travelers change their habits. However, that is not the way processes of socio-technical change work. Especially more radical innovations that also induce behavioral change do not immediately attract mass market but start by attracting ‘early adopters’

336

Chapter 31

(Rogers 1983). In some cases they may gradually attract larger groups, either because of new features that are attractive to these users, or because of increasing problems in the existing regime (or a combination of both). In this respect, a first hurdle in the experiments described above has already been taken which is to demonstrate that such a group of ‘early adopters’ can be found for a multi-modal scheme with EVs which at least illustrates that there is more room for change in travel behavior than is typically assumed. An interesting next step would be to take a closer look at these first users and to try to draw general conclusions on what stimulated them to change their attitude and from this infer barriers and opportunities to attract wider groups of users. On the basis of such an analysis hypotheses might be developed to be tested in further experiments.

5.

IMPROVING LEARNING IN EXPERIMENTS

This means that experiments like the ones above can be important step stones towards socio-technical transformations. This raises the questions on how to organize and set up such experiments such that they render the most useful results. Improving the design of experiments should increase the yield in terms of lessons learned about the potential and feasibility of the technology, the world in which it has to function and the measures that need to be taken to mutually adjust the technology and the social environment in which it has to be produced and used. Especially the behavioral side deserves more attention. One of the most ‘wasteful’ characteristics of the current mobility regime is the habit of the individual to use a relatively large vehicle from door-to-door. Furthermore, these vehicles on average stand still more than 95% of the time, just using precious urban space. New technologies can be a way to create enthusiasm among ‘early adopters’ to change their behavior as this is a way to distinguish themselves. This effect can be strengthened by not only inviting them to experiment with some ready-made alternative but by inviting them to actually engage in the design process. The Mitka-case (Luiten and Knot, this volume, chapter 26) provides a nice illustration of this. This vehicle is not really used yet, though, and further experimentation will have to show whether the enthusiasm will uphold in practice. This underscores the need that learning in experiments should be organized so that ‘early adopters’ get a chance to develop new ideas and try

Combining Technical and Behavioral Change

337

them out in practice. Experiments can thus help to explore the room for socio-technical change by gaining experience with new transport concepts that do not fit the current regime but that have attractive features from the sustainability standpoint. That is especially the case for ‘radically new’ technologies for which substantial behavioral change is needed to ‘make them work’. In such cases it is not clear initially whether and, if so, how a tuning can be achieved between the technical and the behavioral aspects. Experiments can then be a tool to develop such a tuning and to try and explicitly learn on a variety of dimensions, both behavioral and technical. Experiments can then broaden the knowledge base on the practical value of transport innovations which can help to inform further steps. Looking at past and ongoing experiments (usually called demonstration or pilot projects), these issues often do not get the attention they need. Most experiments are either seen as the final step towards implementation, even when there are many unknown issues, or they are a one-shot affair without a clear vision on how to use the results for next steps. Because of the focus on the short term and direct implementation of the set-up most projects also have a strong emphasis on economic aspects. This makes little sense, however, for more radical innovations because transport concepts, user behavior, vehicle characteristics, etc. will only become clear in a longer iterative process of socio-technical change. The optimal vehicle in a Praxitèle like scheme, for instance, is probably not the heavy type of EV used in the experiment but, more likely, a very small, relatively lightweight vehicle with a rather limited range that could be much cheaper than current vehicles on the market. Assessing economic aspects on the basis of the current experiment for Praxitèle is therefore highly unrealistic. This implies that learning and evaluation need to get a broader meaning than in current practice. On the one hand, learning should be more precise and be described as specifically as possible. A general conclusion like “electric vans cannot replace diesel vans” (which is a conclusion from various EV projects) does not mean a lot because it will be based on a variety of assumptions concerning present habits and preferences. To attain sustainable mobility it is more appropriate to seek new ways to ‘question’ these habits and preferences and ‘open them up’. The small nitty-gritty lessons from various projects in themselves may not seem to lead to straightforward conclusions but combining lessons across projects may reveal the ‘contours of a promising new direction’. This implies that it is useful to follow a deliberate strategy to learn across a variety of projects within a niche (like the EV niche) and, as a next step, to use this knowledge to define further experiments attempting to integrate

338

Chapter 31

findings. This requires the co-ordination of the activities of a wide range of actors. The (policy) approach targeting this co-ordination is called strategic niche management (SNM) (Weber et al. 1999, Hoogma et al. 2002). SNM emphasizes that learning needs to be specific but it also needs to be ‘open’. This type of learning is qualitatively different from learning in the situation where objectives are rather specific and fixed. This is called first order learning. If the learning also takes place in relation to the definition of the objectives themselves, implying that various actors become reflexive in relation to their own starting points and assumptions, this is called second order learning. This second order learning is essential if the overall target is to achieve technical as well as behavioral change. So many things will have to change that it is not possible to assess upfront which directions seem most promising. To be critical of one’s own assumptions and preferences is essential to find out what may have practical value, especially since it is required to create a minimal degree of consensus between actors in a situation where a large degree of dissensus is the starting point.

6.

CONCLUSION

The EV projects briefly indicated above are a clear example that new transport concepts can be fruitfully explored in a process that allows both behavioral and technical change. In these cases, some specific characteristics of EVs triggered changes of behavior while the experiences in the projects subsequently helped to define the technology further. This, of course, does not prove that such a change can be reproduced on a large scale but it does give an indication that there is more room for change than is commonly assumed and suggests it is useful to conduct a strategy to explore this possibility further via experiments. The general point to be made here is not that these EV-based systems in their current form should be a central element of a sustainable traffic and transport system. The knowledge base to decide this upon is still far too small. The main lesson at this point is that users are willing to and do change their behavior when they are confronted with specific new transport options in experiments. Stated preference surveys, the most widely used instrument to assess user requirements, are a poor indicator of this and can never match learning from experience. The challenge, therefore, is to design a sufficient range of experiments to be able to draw more general conclusions on what is possible under which conditions.

Combining Technical and Behavioral Change

339

This could render a range of options that have been proven in practice and may point to a variety of ‘promising’ options that are too easily dismissed in current transport policy. How to actually stimulate implementation of the most sustainable options is a follow-on question that is beyond the scope of this paper but by broadening the range of alternatives the ‘socio-technical’ strategy sketched above can make an important contribution en route to a sustainable traffic and transport regime.

REFERENCES Elzen et al. 1999. Boelie Elzen, Robert Ayers, Elizabeth Couzineau, Wolfram Krewitt and Andrea Massini, Inventory of Market Acceptance Factors, Enschede: University of Twente. Deliverable 5 of EU-funded UTOPIA project. Elzen et al. 2001a. Boelie Elzen, Ulrik Jørgensen, Knut H. Sørensen en Øyvind Thomassen, Towards an Interactive Technology Policy - Implications from the social shaping of mobility and transport policies for a new technology policy paradigm, Enschede: University of Twente. Final report from the EU-funded INTEPOL project. Elzen et al. 2001b. Boelie Elzen, Ulrik Jørgensen, Knut H. Sørensen en Øyvind Thomassen, Tackling Transportation Problems around the World - Case-studies used in the INTEPOL project, Enschede: University of Twente. Annex to final report from the EU-funded INTEPOL project. Hoogma 2000. Remco Hoogma, Exploiting Technological Niches, Dissertation University of twente, Enschede: Twente University Press. Hoogma et al. 2002. R. Hoogma, R. Kemp, J. Schot and B. Truffer, Experimenting for Sustainable Transport - The approach of Strategic Niche Management, London: Spon Press. Kemp et al. 2001. René Kemp, Arie Rip and Johan Schot, “Constructing Transition Paths through the Management of Niches”, in Raghu Garud and Peter Karnøe (eds.), Path Dependence and Creation, London: Lawrence Erlbaum Associates, Publishers, pp.269299. Ricci 2000. Stefano Ricci et al., Final validation of the evaluation framework and methodology, Rome: University of Rome, DITS. Deliverable 15 of EU-funded UTOPIA project. Rogers 1983. E.M. Rogers, Diffusion of innovrations, New York: The Free Press. Weber et al. 1999. Matthias Weber, Remco Hoogma, Ben Lane and Johan Schot, Experimenting with Sustainable Transport Innovations – A workbook for Strategic Niche Management, Enschede: University of Twente. Prepared for the European Commission, DG XII, Contract No. ENV4 – CT96 0275 (January 1999). Zwaneveld 2000: Peter Zwaneveld et.al., Analysis of demonstration projects with new transport and propulsion systems, Delft: TNO-INRO. Deliverable 6 of EU-funded UTOPIA project.

Chapter 32 THE PRACTICE OF INNOVATION: Institutions, Policy and Technology Development

David Laws

1.

INTRODUCTION

We know that institutions matter. They set the rules of the game. They provide the staging and context for interaction and communication. They shape our identities and sense of what is fair and appropriate. They provide the means to inquire into facts. They ground thought and action. For all these reasons we want this ground to be stable. There is increasing evidence, however, that it is shifting under our feet. Old sources of stability recede before we have the language to grasp new ones. Disagreements lose their vitality as old tradeoffs lose their grip on our imaginations. Take, for example, policy disputes over whether markets or administrative rules are the best way to advance technological development. These disputes embed an agreement that the choice between market and hierarchy is meaningful and provides stable ground for debate. If institutional arrangements began to crop up that did not fit this dichotomy and, moreover, looked attractive, then the tug of war between these poles would lose its interest and vitality. In this chapter I trace some prominent features of what I believe is a new institutional landscape and discuss their implications for efforts to promote innovation through direct action and policy design. I start with what amounts to stipulating two general features. The first is a breakdown of the distinction

341 P.-P. Verbeek and A. Slob (eds.), User Behavior and Technology Development: Shaping Sustainable Relations Between Consumers and Technologies, 341-356. © 2006 Springer.

342

Chapter 32

between markets and hierarchies hinted at above. We see signs of this breakdown when observers in fields like business begin to try out new vocabularies to describe the landscape they see. In a recent issue of the Sloan Management review, a group of authors commented, for instance, that “[m]odern societies are not market economies; they are organizational economies in which companies are the chief actors in creating value and advancing economic progress.”61 Others have highlighted changes in the character of the relationships through which coordination is achieved as the feature that distinguishes the new institutional terrain. Sabel notes that efforts to capture these patterns of practice in the vocabulary of market and hierarchy… …overlook ...the possibility that the new firms operate by principles of decentralized coordination so different from those of the preceding epoch of large-scale organizations, and so disruptive of the institutional connections to the administrative state, the ‘market economy’ and ‘representative democracy’ of the coming decades will look as different from the timeless victors of today as the latter now seem from their mid-nineteenth century, small scale predecessors.62 The first feature is this shift away from a choice between market and hierarchy to patterns of coordination based on organizational relationships that operate by new “principles of decentralized coordination.” The second is often described in terms of a blurring of the boundaries between public and private and a concomitant growth in the size, role, and resources of the civic sector. Again, it is the practical significance of new arrangements that has attracted attention.63 In a variety of policy arenas, observers regularly note the prevalence and success of arrangements that do not fit with established notions of government. The relationships and patterns of action in these settings: ...indicate ...a shift away from well-established notions of politics and bring ...in new sites, new actors and new themes. There is a move from the familiar topography of formal political institutions to the edges of organizational activity, negotiations between sovereign bodies, and 61 62 63

Ghoshal, Bartlett, and Moran, (1999) p. 9. Sabel (1995) p. 1. “A new range of political practices has emerged between institutional layers of the state and between state institutions and societal organizations. The new language is rooted in an appreciation of the importance of these new political practices. Authors as varied as James Rosenau, Judith Innes, and John Dryzek have pointed out that it is these often transient and informal arrangement that produce solutions, not conventions among states, directives, or authoritative decisions Hajer and Wagenaar (2003) p. 1.

The practice of innovation

343

inter-organizational networks that challenge the established distinction between public and private. The disparate actors who populate these networks find nascent points of solidarity in the joint realization that they need one another to craft effective political agreements. Their efforts to find solutions acceptable to all who are involved (and to expand the circle of involvement) nibbles and gnaws on the constitutional system of territorially based representative democracy. Notions of politics itself change as new themes occupy centre stage. It is probably no coincidence that these practices are more developed in ‘new’ spheres of politics such as the environment and the ’life politics’ of food and technology.64 There is not space here to make good on these assertions empirically. The anecdotal evidence provided by direct experience probably provides the best support that this is not idle chatter and something is afoot, even if we cannot fully grasp it. Public-private partnerships and practices like “social entrepreneurship” draw our attention even as they resist understanding. Success in complex ventures increasingly hinges more on sustaining interactions among a diverse group of organizations than on accumulating authority and resources. Any particular case is likely to look like an ad hoc adjustment to particular circumstances. If we squint, however, and blur the details, we can begin to see the outlines of a new institutional landscape.

2.

AN EXAMPLAR OF NEW PRACTICE

To evoke the prominent features of this landscape, I turn to an example. SunLine Transit in Thousand Palms, California, is noted for its success in technological innovation, in renegotiating institutional rules, and in working across public and private boundaries. The features that distinguish SunLine’s story are most apparent through a historical retracing of the development of the organization, the transportation technology it employs, and the relationships that have been central in its development. The story starts in an unlikely place — failure. In the late 1980s and early 1990s, SunLine experienced a steady decline of service that eventually became unacceptable, even in light of diminished expectations. SunLine was continuously trying to make do and keep a fleet of second-hand buses on the road. George Earl, the former Director of Maintenance, described the atmosphere: “In the summertime, I’d have half the fleet just sitting dead in 64

Hajer and Wagenaar (2003). p. 3.

344

Chapter 32

the road because they couldn’t operate in this environment. We never had a maintenance program. We were too busy putting out fires all the time.”65 This pattern came to a crisis on a summer day in 1993, when two-thirds of SunLine’s fleet went out of service. SunLine’s Board of Directors realized that the agency could not survive, much less provide a reasonable level of service, with second-hand buses. They needed wholesale change; they needed new buses. Dick Kelly, the Chairman of SunLine’s Board at the time, approached General Manager Dick Cromwell. As Cromwell recounts, Kelly said, “Dick, we need to get new buses, so let’s go get new buses. Everybody else has new buses, why don’t we have new buses?” The story that follows was triggered when Kelly added, “Oh, by the way, make them alternative fuel.”66 The Board’s commitment gave Cromwell and his staff the opportunity and incentive to explore technical options and created a context in which to discuss development. Cromwell used the commitment to learn and to build agreement on which technological path to take. This effort initiated a chain of events that have pulled the organization into unexpected areas of development. The first problem in this chain was how to choose an alternative fuel from the many that were available. Cromwell and his staff explored electricity, propane, fuel cells, and liquefied natural gas (LNG) before settling on compressed natural gas (CNG) as the technology that best fit their needs and circumstances. The internal process of learning and developing support for this choice was tied to a process of discussion with external 65

All of the quotes by George Earl were gathered in an interview conducted on August 3, 2000.

66

All of the quotes by Dick Cromwell in this report were gathered in an interview conducted on August 3, 2000. This account of the origin appears, after multiple interviews, to be true. There is disagreement about the origins of the commitment. Cromwell suggests “ignorance was a partner” and the board’s early commitment to alternative fuels as made “somewhat out of blindness” and with a sense that there was “nothing to lose.” Some board members agreed with this explanation. They said they didn’t fully comprehend the ramifications of the commitment they were making. Kelly, on the other hand, maintains that there was sufficient experience to understand what the commitment implied. He cites his experience bringing electric vehicles into Palm Desert and the knowledge he and other board members had about alternative fuels were important factors. For example, he and Cromwell had seen a CNG bus at an American Public Transit Association (APTA) meeting in Toronto, and marvelled at how clean the emissions were. He believed fuel cell buses were ideal, but realized the technology was not ready. Based on this cumulative experience, Kelly felt they were simply taking a well-calculated chance.

The practice of innovation

345

stakeholders that built support for the transition. The ‘end in view’ (Dewey) that drove this discursive process was the need to finance the new buses. SunLine’s Board, which is made up of elected officials from each of the local municipalities and the county, gained the support of other local and state leaders for the project. In Cromwell’s words, “I had senators and congressmen, I had every elected official. I had every school board member. I had everybody.” This proved to be an effective strategy. SunLine eventually got $2.5 million from Congress and an additional $10 million from the U.S. Department of Transportation. Getting funding was just the beginning of the story, however. Cromwell and his staff immediately faced a new problem. They had committed to a technology they did not know how to take care of and for which no infrastructure existed. Solving the maintenance problem and building infrastructure framed a new horizon for action, and that would extend SunLine’s network of partners. Maintenance was interpreted as a demand for training that drew SunLine into relationships with the local College of the Desert, which brought expertise in training, and the original equipment manufacturers (OEMs), who brought technical expertise, around the problem of how to design and deliver a training program. These efforts built ties with the U.S. Department of Energy (DOE). The program these organizations developed eventually became the model for the DOE’s technician certification. Training also made SunLine a central node in a network of technicians who consult one another on maintenance issues.67 These ties have become permanent features of the organization. They have made SunLine a worldwide destination for transit organizations interested in alternative fuels, and built an ongoing role for SunLine in beta-testing technological developments. The latter brings them into substantive interaction with OEM’s on a regular basis and keeps the staff involved with new technology. Out of these interactions, SunLine has been able to experiment with fuel cell buses from three different manufacturers. The effort that broke the chicken-and-egg impasse of infrastructure development followed a similar pattern. A partnership between SunLine and the Gas Company that produced the first refueling facility has been extended to a joint venture with (formerly) Pickens Fuel. These organizations now 67

The program was introduced throughout California through the community college system, and also led to the establishment of the Energy Technology Training Center at the College of the Desert and the statewide Advanced Transportation Technologies Initiative. SunLine technicians now participate in the ongoing redesign of the training program, and the last eight hours of the 40-hour basic course are conducted in SunLine’s shop.

346

Chapter 32

own and develop refueling stations together. SunLine worked with municipal governments to bring CNG technology into their fleets and the fleets of their service providers. A recent joint venture with Ford brought CNG cabs into service in the Valley, using funding from the U.S. Department of Transportation (DOT). Together, these efforts have increased the density of demand and stretched the reach of infrastructure until it approaches the goal of being able to support a drive from Los Angeles to Phoenix. Development has occurred through the sequence of problems that has pulled SunLine and its partners forward. This pull developed as the resolution of each problem led to the articulation of a new one that extended the horizon of action. The successful adjustment to CNG technology led the organization to ask, “What would it mean to be a zero emissions transit company?” This question has driven the interest in fuel cells and intermediate technologies such as hythane. The habits of innovation developed in this process have pulled SunLine well beyond anything contemplated in the original commitment to “make [the buses] alternative fuel.” A sustained “sociability” has transformed pragmatic evaluation of means into occasions to also deliberate about ends. These conversations have fostered a sense of community in the Coachella Valley that suggest a potential for democratic engagement in practical conversations about technology. The comments of Julie Bornstien, Representative to the California State Assembly, suggest the tie between technological innovation and democratic community in the Valley and the broader potential of these interactions. “I think for the Valley, it just puts us way out ahead of innovations throughout the country. We pride ourselves here on our natural environment, and by switching from diesel fuel to natural gas, we’re going to preserve that environment not only for our children, but for many generations to come.”

3.

THE LANDSCAPE AND LOGIC OF DEVELOPMENT

Sociability was sustained in a pattern interaction among organizations — a network — that crossed boundaries between public and private. At the heart of the networks in the SunLine story are the nodes at which interaction

The practice of innovation

347

occurs.68 These are the points at — and moments in — which actors meet and through their interactions recognize and explore interdependence. Through the exploration of interests and problems at these nodes SunLine and its partners have been able to repeatedly “…turn ...amidst the flux of economic [political, and social] life the pragmatic trick of simultaneously defining a collective-action problem and a collective actor with a natural interest in addressing it.69 By turning this trick, the SunLine and its partners have made the development of insight about why and how to act coincident with the development of the capacity to act. The nodes in this story are not the fixed institutional sites featured in many accounts of networks.70 They are ad hoc and fluid and constituted in efforts to address common problems. The need to find financing, solve training problems, and address mismatches between standards and practices each created a context for interaction that highlighted interdependence. No single organization controlled the resources or authority to unilaterally solve the problems that surfaced. The curious thing is that these limits on agency were not debilitating. Shared problems provided a focus and basis for an interaction that allowed a diverse set of organizations to further their distinct instrumental interests as they built formal and informal ties (i.e. a network) 68

69

70

Castells (2000) describes a network as “a set of interconnected nodes. A node is the point at which a curve intersects itself.” p. 501. Sabel (1994) p. 272. It helps to think of the example of a start-up company. The organizational horizon of action is always being reframed as priorities change and experience accumulates. Whether organizational attention is fixed by the need to raise an initial round of financing, to meet a technological milestone, or to convince investors that a new interpretation of the market is needed, the effort to address these problems creates a context for interaction among the firm and other parties who share interests, commitments, or who have resources that are needed. Progress invariably involves a renegotiation of these relationships that is at once a reflection on and reframing of the problem horizon and a re-elaboration of the network. “What a node is, concretely speaking, depends on the kind of concrete networks of which we speak. They are stock exchange markets and their ancillary advanced services, in the network of global financial flows. They are the national councils of ministers and European Commissioners in the political network that governs the European Union. They are coca fields and poppy fields, clandestine laboratories, secret landing strips, street gangs, and money-laundering financial institutions in the network of drug traffic that penetrates economies, societies, and states throughout the world. They are television systems, entertainment studios computer graphics milieu, new teams, and movie devices generating, transmitting, ad receiving signals in the global network of the new media at the roots of cultural expression and public opinion in the Information Age.” Castells (2000), p. 501.

348

Chapter 32

with one another. This network is the capacity for collective action and the device for developing a shared understanding of the problem, of one another, and of common goals. The “pragmatic trick” of addressing these demands simultaneously elicited the potential for learning and development implicit in the interactions. Of course recognition of the need for inter-organizational coordination does not guarantee success in technological development any more than in other domains of organizational life. Such demands more often highlight disparate interests, conflicting imperatives, divergent time frames, and distinct agendas, goals, and vocabularies, than a commitment to engage one another and develop understanding through collective action. To many, the depth of interdependence highlighted in the account above would constitute grounds for caution, not optimism. This raises the question of how it is that these interactions produced the outcomes described above, rather than the skepticism, misunderstanding, and conflict that ʊ which are more often associated with demands for inter-organizational coordination. What, in other words, creates the capacity for meaningful, informed, effective action in the new institutional landscape?

4.

A NEW LOGIC OF PRACTICE

I suggest that we can get guidance by turning from an analysis of when and where collaboration occurred to how it occurred. The key to transforming a chain of exchanges into the kind of conversation in which development can occur is the character of interaction that goes on within and among organizations.71 Take the maintenance problem. It was a question of knowledge and skills, but not in the sense of learning a few new tricks of the trade. The shift from a stable, well-understood practice — diesel mechanics — to an openended one demanded more substantial reflection and re-evaluation. Mechanics would have to move from matching known problems to known 71

For example: “ ...the rules of ...unbalanced growth transform ...a chain of exchanges ...into a continuous discussion of joint possibilities and goals, where the parties’ historical relation defines their mutual expectations. Just as in a discussion, the parties suppose their understanding of their situation is limited. Therefore they jointly specify what they believe they understand so as to expose and begin exploring the limits of that understanding. Just as in a discussion, they must accept the possibility that their views of themselves, of the work, and the interests arising from both—their identities, in short — will be changed unexpectedly by those explorations.” Sabel (1994) pp. 247-248.

The practice of innovation

349

solutions to figuring out what the problems were and designing ways to deal with them. No manual was available for the problems that would crop up in service, particularly in a demanding desert climate. It was no secret that such a change would trigger individual and organizational resistance. As George Earl, the former director of maintenance at SunLine put it: “[F]or the most part, people don’t like change. If I had 18 years of being a diesel mechanic, why would I all of a sudden have to change my whole way of thinking and the way I do things?” It is, of course, impossible to say just what made the transition work. One feature stands out in SunLine’s approach, however. Earl and his staff were engaged throughout the process in defining and solving problems. They played a role in choosing the technology. The mechanics went through and then became part of the training program that they had helped to design and, in the process, became technicians with a new future. Moreover, maintenance practices were always up for discussion as the staff wrote and rewrote the manual together. This was carefully organized so that it did not lapse into “anything goes” (at any point in time there was an agreed-upon approach), but the commitment to these recipes was always provisional and open to be overturned by fresh insights. The practical value of this commitment was made good through respect for norms of conversation that was demonstrated by listening, by treating others’ views as understandable, and by giving reasons for our own that open them to scrutiny. George Earl’s description captures the close tie between the character of the interaction and the development of practice. “We’re always coming at each other with, ‘This ain’t working,’ or ‘This is working,’ or ‘Can we do it this way?’ And then…among the supervisory staff we’ll get together and say, ‘Well, what do you think?’ And they’ll go out to their shift and they’ll talk about it out there and then we’ll come back and meet again. Once we decide [on a solution], we create a paper that goes in each mechanic’s thing that [describes] the new procedure…. Everything is…up for grabs and [can] be revised at any time, as long as we can prove to ourselves that we found an easier and better way of doing something, and we’re accomplishing the same thing or making it better.” This approach extended to SunLine’s interactions with other organizations. Construction of the CNG refueling station brought the fire department and other groups that had responsibilities for public safety into the story. These groups were nervous about the new station, and the possibility for crisis and controversy it created. Rather than reassure them, Cromwell and his staff invited them into the conversations about design and

350

Chapter 32

implementation. Through their participation the groups began to understand each other’s concerns and to see how actions addressed them. Over time, trust developed out of these open interactions. “[W]e told them what we were going to do [regarding the fueling station], because there were no rules, there were no safety rules, there wasn’t anything there. This was 7 years ago. [We were] the first people in the United States to do this. Well you know, all of a sudden you find out the fire guys didn’t show up at the meeting anymore. You call them and say, ‘You weren’t at the meeting.’ They say, ‘Oh, it’s all right, we feel comfortable with what you’re doing.’ Then the safety guys are gone. And then all parties that would have an issue are gone because they felt comfortable.” The benefits of this approach were realized when SunLine had an accident. A flare of gas ignited and burned near the ceiling of the shop. The fire was put out and the problem that caused it was fixed. But a fire burning in the air is just the kind of thing to excite imaginations, raise anxieties, and derail development. The preceding conversations provided a context to collectively interpret the incident and, in this case, avoid disrupting the bus program. SunLine used the accident as an occasion to extend the commitment to open conversation. Instead of seeking to control publicity or spin it, they published an account of the accident. Hydrogen raises more serious concerns about safety. SunLine has taken a similar approach as they have begun to experiment with hydrogen fuel. “We’re developing right now a basic 8th grade curriculum [on hydrogen technologies] that I’ll put everyone in this agency through, from the fuel hustlers to the receptionists…. Everyone wants to talk about the Hindenburg, [so] let’s talk about the Hindenburg. Let’s get that over with. Let’s talk about the volatility of hydrogen, let’s talk about whether it’s dangerous or not dangerous, let’s talk about what hydrogen’s all about.” (Dick Cromwell) Bill Clapper, Executive Director of SunLine Services Group, described the learning process that has occurred through these discussions. “[The maintenance staff ] were a little concerned last year when we started talking about [hydrogen]…. But that’s because of what they didn’t know. As we got smarter, we started giving them information and showing them things. What I think really sold them is last November we had the fuel cell bus…. We popped open the trunk where the engine is, and [they said] ‘Oh, yeah, there’s the air compressor and there’s the camshaft, and I

The practice of innovation

351

recognize that. Oh, by the way what are those two black boxes up there?’ ‘Those are the fuel cells.’ ‘Oh, well what do you do with those?’ ‘You just remove and replace.’ ‘Oh, OK. Well it’s got brakes, it’s got lights….’” This conversational approach to practice extends to another hallmark of SunLine’s method — the persistent effort to find a way to make things work in light of the diverse interests of its partners. Dick Cromwell captured what this means in practical terms: At the end of the day, we find a way to make public entrepreneurship work, because it has to work for all the parties, just like in real business. And so you try to identify what is your issue and what is my issue and is there a way I can make that work within this box I have to work within? And then if the box get too confining, what can I do to make the box over a little bit . . . together? We’ve had some pretty good success with that. A critical moment in the development of the fuel cell bus program illustrates the importance of this approach. Cromwell and his staff were eager to move ahead and get hands-on experience with fuel cell technology. After negotiations were fairly advanced, an impasse developed between the Department of Energy (which was subsidizing the project) and XCELLCIS, the bus manufacturer. DOE wanted to collect and share performance data during the trial period. XCELLCIS was concerned this would compromise its proprietary technology. The conflict threatened the initiative. SunLine staff understood the interests of both parties and facilitated a negotiation in which XCELLSIS and the DOE reflected on their interests and the importance of the trial, and this led to an agreement on terms. SunLine’s ability to broker this understanding kept the program on track, and secured access to the benefits that have accrued. The common element in these examples is that organizations that shared an interest in a problem have been able to cooperate and respond effectively to an uncertain future because of the way they talk to each other. Talk has been the “institutional device” for modulating the interplay between shared concerns and divergent perspectives and interests. Development hinged on the ongoing modulation of the (potential) partners’ views of the problem, their stake in it, and, their relationships with one another through conversation. The stakes were sufficient to generate the commitment to talk in an open way, to reflect on self, other, and the future in light of these exchanges, and, thereby, to sustain the possibility of talking in a way that allows the networks to develop that secure and sustain a capacity to perceive and to act.

352

5.

Chapter 32

POLICY IMPLICATIONS

The preceding sections have not been an extended introduction to a discussion of policy implications. They describe a process of policy development in and through the interaction of organizations involved in a process of technological innovation. The broad features of this account should be familiar: interdependence, inter-sectoral networks, and collective action facilitated by deliberative interaction. Development and innovation were simultaneously and inextricably substantive and institutional. In working out substantive problems, the parties learned to work with each other from their distinct organizational starting points, across public-private boundaries, and in the face of persistent uncertainty about the future. In the process, they rewrote the policy manual together. Some of the changes were explicit — the Department of Transportation changed its funding rules to make them more conducive to the kind of partnerships SunLine and other organizations were becoming involved in — others were implicit in the habits and practices that developed through working together. In this account, policy is less an object — a carrot or a stick — than a way of shaping relationships and of acting within the context of the relationships we create. In this concluding section I would like to try to draw out more explicitly some of the policy implications of the preceding analysis. To help make this transition, I introduce a distinction made by Norton Long: For certain purposes the individual is a useful way of looking at people; for many others the role-playing member of a particular group is more helpful ...If we know the game being played is baseball and that X is the third baseman, by knowing his position and the game being played we can tell more about X’s activities on the field than we cold if we examined X as a psychologist or a psychiatrist... The behavior of X is not some disembodied rationality but, rather, behavior within an organized group activity that has goals, norms, strategies, and roles that give the very field and grounds for rationality. Baseball structures the situation. 72 In other words, there is intelligence in behavior, if we can grasp it, which can give us insight into the way situations are structured — in the current vocabulary, to the institutional landscape. The role, or better set of roles, that players take is what helps us make sense of their behavior and get access to 72

Long (1958) p. 252.

The practice of innovation

353

the way the game structures their interaction. This kind of analysis is continuous, with intuitive questions one might ask a colleague after an ambiguous meeting, “Was she buying or selling?” The articulation of roles — buyer and seller — interprets the situation and helps us analyze it. Misinterpretation puts us on the wrong track. It should be noted that in Long’s account structure is not a fixed thing, but the kind of dynamic set of relationships that is expressed by the idea of an ecology — here an ecology of roles or games. In an analysis of the SunLine case and similar experiences, my partners and I have described a set of five roles that we believe reveal something about the character of institutions that “structure the situation” in cases of technological innovation. These roles also offer a guide for action. It is possible to reject our exact specification of roles and still accept the notion that historical developments and experiences with innovation both demand reflection on roles, and that such reflection can guide the development of policy. The role set we have described stresses facilitative rather than managerial activities, and tends to parallel the range of entrepreneurial roles that private sector actors have played in the dynamic economic sectors, such as information technology, over the past few decades. We have described the roles as: • Pioneers who recognize opportunity, seize initiative, and catalyze action by making commitments. • Public Venture Capitalists who understand and embrace risk and package financial, social, and human capital to meet project driven needs. • Superintendents who provide an environment in which innovation can flourish by fostering the development of relationships that are sustained through formal and informal networks. • Mediators who build consensus on goals and direction, and bring directed problem-solving to bear on conflicts that threaten to stall or derail the development of ventures. • Stewards of the common good who focus attention on the common good, maintain standards for responsible behavior, and facilitate the coalescence of democratic community around programs of action. One way that these role descriptions can influence policy is by providing practitioners with a basis for reflection that gives them leverage on the complexity of their experience. A role set helps us ask, “What is it about our actions that makes sense?” and diagnose mismatches between intent and

354

Chapter 32

effect. This kind of reflection is particularly important in a period of transition when the intelligence of practice may outstrip the account we can provide. During such periods, it becomes important to be able to reflect on, and get access to, the insights implicit in action. An apt role set can also facilitate reflection at a system level that might direct policy to structural deficits that consistently undermine development. For instance, it might become possible to recognize that a program packages resources in a way that is at odds with the development needs of the very initiatives on whose success the policy depends. It might help us diagnose the small conflicts that derail innovation efforts, and build a capacity for mediation that could help resolve these conflicts. System level reflection on roles might call attention to the need to build ties among organizations and a capacity to talk about differences before they are needed. Each of these diagnoses draws on role descriptions to interpret the complex and ambiguous information we often have about practice. The specific insights that develop out of reflection on the particulars of an organization and its environment are likely to be more valuable than these general observations. More specifically, it is possible to interpret treat roles as a rough guide for action. Public agencies may look at for different ways they can make commitments that galvanize action — e.g. through their own purchasing practices — and recognize the degree to which they are dependent on sources of commitment outside the agency. They may come to see that sources of commitment should not be taken for granted, but cultivated and nurtured. The other roles provide some outlines for good husbandry in this regard. Public funding programs are often at odds with the ad hoc character of initiatives. Requiring these initiatives to meet standardized categories and guidelines either creates burdens at difficult times or makes funding so unwieldy that it is not pursued, even when it would be invaluable. This means that public agencies must find ways to meet the demands they face for accountability in allocating public funds that are more in step with the initiatives they want to prosper. Here, a closer understanding of how accountability is provided in practices like venture capital, and a practical appreciation of the potential for direct democracy that inheres in the interactions around technology, may provide new insights into how to provide accountability by tying evaluation of action to deliberation about shared expectations and the common good. Simple steps like organizing a service to coordinate access to funding programs offered by different agencies, can go along way to turning public funding into the kind of venture capital that is often in short supply.

The practice of innovation

355

Persistent efforts to cultivate and renew consensus on strategic direction can enhance coordination by interactively shaping expectations. The value of such efforts will be limited if they are not tied to the development of focused problem solving capacity, however. The conflict of interests that threatened SunLine’s fuel cell program is one example where development hinged on the ability to mediate a dispute that arose in the development process. Innovation efforts that develop along entrepreneurial lines draw on acquaintances and other thin ties that thicken as they become active, and become active as they become relevant. Relevance is judged in terms of an unpredictable horizon of action. Effective initiatives are, thus, often those that can draw on a rich background of relationships. Leaving these ties to chance creates a fragile system. Policy-makers and public officials can contribute to the resilience of initiatives and sectors by providing forums that attract a diverse group of organizations and allow individuals who represent them to meet one another and learn about their common and complementary interests and competencies. Such forums can also provide a means for engaging in public discussion of the sources of risk associated with new technologies and the means available to address them. These brief signposts show how the roles can guide the development of policy in a fairly straightforward way. Reflecting on the roles that we play as mediators, superintendents, public venture capitalists, and stewards of the common good, sheds light that reveals new contours in practice. Taken together, these roles point to a need to reframe practice in terms that are more consistent with our historical circumstances and with the distinctive demands of innovation. Any reframing that is substantial enough to meet these demands will generate organizational tensions. A role set offers a way to grasp these tensions and open them to reflection by facilitating a comparison between old and new. If these tensions between the demands generated by these role sets do not receive attention, they will undermine reform efforts and insulate the interference from reflection. In this account, the boundaries between policy, institutions, and innovation are blurred. Innovation draws actors into a process of institutional development, and policy is reshaped in the process. Policy is as much (or more) about establishing the vocabulary, grammar, and ground for interaction as it is about strategic intervention. The interplay between institutions, innovation, and policy can be understood at three levels. In day-to-day practice it creates a need to reflect on the intelligence of action and to

356

Chapter 32

reframe the categories through which we make sense of the world and decide how to act. Such reflection can also inform action at a system level, where policy shapes the ground on which interaction occurs and can balance the ecology of roles organizations take. Finally, policy is a process of managing the tensions raised at the organizational level by these revisions in our sense of what is problematic and what constitutes intelligent action.

REFERENCES Castells, Manuel (2000). The Rise of the Network Society. Oxford; Malden, MA: Blackwell Publishers. Ghoshal, Sumantra, Christopher A. Bartlett, Peter Moran (1999). “A new manifesto for management” Sloan Management Review, (Special Issue: In Search of Strategy) Spring. pp. 9-20. Hajer, Maarten and Hendrik Wagenaar (eds.) 2003. Deliberative Policy Analysis. Cambridge: Cambridge University Press. Long, Norton (1958). “The Local Community as an Ecology of Games,” American Journal of Sociology. pp. 251-261. Sabel, Charles 1995. “Design, Deliberation, and Democracy: On the New Pragmatism of Firms and Public Institutions.” Paper presented to the Conference on Liberal Institutions, Economic Constitutional Rights, and the Role of Organizations, European University Institute, Florence. December 15-16. Sabel, Charles 1994. “Learning by Monitoring: The Institutions of Economic Development,” in L. Rodwin and D.A. Schon (eds.), Rethinking the Development Experience. Washington D.C.: Brookings and Cambridge, MA: The Lincoln Institute. ” pp. 231-274.

Chapter 33 ETHICAL ASPECTS OF BEHAVIOR-STEERING TECHNOLOGY Philip Brey

1.

INTRODUCTION

For a short period of time in the mid-1970s, a federal law in the United States mandated cars to be designed not to start if seatbelts were not worn. Cars produced during that brief period of history had an electric link between the seats, seat belts and starter. In the seats, there were weightsensing elements that registered whether a person was in the seat. If so, the seatbelt for that seat would have to be fastened or the starter would not work (and a buzzer would sound). This mechanism is an example of behaviorsteering technology, which is a technology in which one of its main functions is to make its users behave in a way that is not necessarily desired by the user but that is desired by some other party in control of the technology. Usually, such behavior-steering functions are designed to be part of the technology by order of a government or other standard-setting body, or by an organization that commissioned the technology. Some other examples of behavior-steering technologies (BSTs) are gas pedals that increase their resistance to the foot of the driver when the car is going above a certain speed limit so as to encourage a more economical use of energy, or the heavy weights on hotel keys that induce hotel guests to drop off the key at the receptionist before going outside. The question discussed in this essay is under what circumstances, if any, the use of behavior-steering technology is justified from a moral point of

357 P.-P. Verbeek and A. Slob (eds.), User Behavior and Technology Development: Shaping Sustainable Relations Between Consumers and Technologies, 357-364. © 2006 Springer.

358

Chapter 33

view, and under what circumstances its use may become morally problematic. The U.S. seatbelt case illustrates that behavior-steering technologies are sometimes controversial. U.S. car drivers did not appreciate being mechanically forced to wear their seat belts, and many drivers had the mechanism illegally removed. Some people even mounted a court challenge: they felt that the coercive mechanism went against their civil liberties. As a result of these protests, the law was repealed, and wearing seat belts became again something that was mandatory but no longer mechanically forced. So when are behavior-steering functions morally permissible? When is the moral price that has to be paid for the perceived benefits of behaviorsteering seen to be too high? In what follows, I will discuss three moral issues that need to be considered in the use of behavior-steering technologies (BSTs): the freedom issue, the technocracy versus democracy issue, and the responsibility issue.

2.

THE FREEDOM ISSUE

The seatbelt example suggests that there may be circumstances in which the use of BSTs infringes on freedoms or rights. Even if the seatbelt mechanism at issue can be shown not to infringe on basic freedom rights, it is easy to think up examples of BSTs that do. One could, for example, imagine a car with an on-board navigation system that puts certain areas of town off-limits whenever the city government believes this to be a good idea. Or one could imagine automobiles in fundamentalist countries that only start after an intelligent camera on board has determined the driver to be male, or to be wearing the right state-prescribed clothing. Or one could imagine a car with a mechanical arm inside that hits the driver when he is going over the legal speed limit or swerves on the road. There are also examples of BSTs that clearly do not infringe on freedom rights. The hotel key with the heavy ball, for example, can hardly be argued to be a sinister coercive instrument by hotel owners that wrongly restricts the freedom of hotel guests. So, under what circumstances do BSTs restrict human behavior to the extent that basic freedoms or rights are violated? To answer this question, I will first briefly discuss the classical distinction between two conceptions of freedom, or liberty, that has been proposed by philosopher Isaiah Berlin in his famous essay Two concepts of liberty. Berlin argues that, historically, the concept of liberty has two quite different meanings. In the first, ‘negative’

Ethical aspects of behavior-steering technology

359

sense, a person is free to the extent to which his actions are not obstructed or interfered with by others. Your amount of personal freedom, in this sense, is given by the area within which you can act unobstructed by others. In the second, ‘positive’ sense, a person is free to the extent to which he is his own master, whose life and decisions depend upon oneself and not upon external forces of any kind. Such a person is autonomous, or self-determining, and is able to think freely, bear responsibility for her own choices, and able to explain them by reference to her own ideas and purposes. He is his own master, and a slave to no one. To better understand this contrast, imagine a ruler or king who has unlimited freedom to do whatever he pleases ʊ none of his subjects dare disobey him or get in his way. Yet, all his actions turn out to be strongly conditioned by his father, to whom he is psychologically dependent to a large extent: all his choices and actions in life are in fact mere attempts to please his father rather than the reasoned, autonomous choices of a free agent. Such a ruler, then, enjoys a large amount of ‘negative’ freedom but little or no ‘positive’ freedom. In contrast, a Nelson Mandela, when still locked in his cell during the apartheid regime in South Africa, had very little ‘negative’ freedom, but his independent, unbroken spirit conferred on him a great amount of ‘positive’ freedom.

‘Negative’ freedom, or freedom from interference This distinction suggests that BSTs may restrict freedom in two ways: by restricting negative liberty and by restricting positive liberty. Let us first turn to restrictions of negative liberty. These occur when BSTs interfere with the activity of technology users. But, as the hotel key example shows, not every such interference constitutes an infringement on human rights. So when do interferences by BSTs go too far? A first possible line of response to this question would be to argue that interference goes too far when the majority of technology users believes that the BST interferes with their actions to an unreasonable extent, as happened with the seatbelt mechanism. Yet, this response is not wholly satisfactory: that a majority believes that a particular technology is unacceptable does not necessarily mean that it is, in fact, unacceptable. Usually, BSTs are used to promote some socially desirable end, such as sustainability, safety, efficiency, or equity. The price that is sometimes paid for these ends is that limits are imposed on the scope of individual behavior, thus constraining individual liberties. In trading off the positive and negative

360

Chapter 33

impacts of BSTs, the importance of promoting positive ends must be weighed against the cost to individual liberty. A BST may then turn out to be justified, even if there is resistance to it. This is the case when the cost of not using the BST can be shown to be greater than the cost to individual liberty when it is used. For instance, a careful assessment may show that meeting environmental goals requires more efficient use of the automobile, and that this requirement can only be met through certain coercive measures, such as speed delimiters built into automobiles, because less coercive alternatives have failed. When it is determined, in the democratic political arena, that sustainability is a greater good than the right of drivers to drive at high speeds, then such coercive methods may turn out to be justified. At the same time, there may be some basic freedoms or rights that arguably should not be violated under any circumstances: rights such as those included in the Declaration of Human Rights, such as the right to freedom of movement, freedom of speech, and freedom of assembly. These are ones that are essential to realizing one’s individual life plans.

‘Positive’ freedom or autonomy Let us now turn to ‘positive’ freedom or autonomy. Autonomy is not threatened by a few BSTs that restrict freedom of movement. Autonomy becomes threatened when essential parts of our lives are conditioned by BSTs, to such an extent that they do not merely interfere with our actions, but go as far as to shape and condition our plans and goals. Suppose, for example, that automobiles do not just require us to wear a seat belt, but in fact do the driving for us: they are programmed to select routes for us, select driving speeds, stop at stop signs, regulate the internal temperature, and so on. Driving in such ‘intelligent’ automobiles is not an autonomous activity, but is governed by the dictates of the automobile; the human driver merely follows its choices. Suppose, now, that other machines like our personal computer, toaster, laundry machine, refrigerator and personal digital assistant, are programmed similarly to make our choices for us ʊ which they increasingly are. Living in such a technological culture in which machines decide for us, we are no longer autonomous decision-makers, but leave it to machines to tell us what is good for us, and what we should therefore do. The programmed choices of these machines may sometimes coincide with choices that we would have made ourselves, but at other times they may reflect interests of a government or of a company one works for, or they may simply make wrong guesses about our own preferences.

Ethical aspects of behavior-steering technology

361

A transfer of intelligent thought and decision-making from persons to machines is not necessarily a bad thing; there are many things that we would rather have machines decide on: how hard to suck in air when vacuuming different surfaces, or when to inflate an air bag in an automobile. It is only when machines try to determine more overarching goals of activities that autonomy is significantly eroded: when an automobile tells us where we want to go or how fast we want to get there, or when word-processing software changes spelling, grammar and style of our writings without giving us the opportunity to overrule its choices.

3.

THE TECHNOCRACY VERSUS DEMOCRACY ISSUE

BSTs have been accused of being a technocratic method for shaping society and implementing policy. A technocracy is a society in which political power rests to a significant extent in the hands of scientists, engineers and other experts who use scientific principles and technological means to attain political ends. A technocratic solution to a policy problem is, therefore, a solution in which technical solutions to problems have taken the place of political decisions. This is customarily held to be undesirable for at least two reasons: because technocratic solutions are not democratic, and because they depend on the false idea that social problems can be solved by means of a technological ‘fix’. The first of these criticisms, that technocratic solutions are not democratic, has been held by critics of technocracy such as Jürgen Habermas, Helmut Schelsky, Jacques Ellul and Max Horkheimer. They argue that science and technology have an internal logic or rationality of their own that presents itself in an objective form that cannot easily be criticized in a democratic political arena. If scientists and engineers have determined that solving world hunger requires the genetic engineering of crops, and that solving traffic jams requires satellite monitoring of traffic, how can politicians dispute these ideas if they lack the proper expertise? Scientific-technological decision-making and (democratic) political decisionmaking, hence, seem to belong to separate spheres; yet, in a technocracy the first increasingly takes the place of the second, and hence politics becomes less democratic. The image of science and technology on which this criticism rests has since been shown to be problematic, however. In contemporary science and

362

Chapter 33

technology studies (e.g. Mackenzie and Wajcman, 199; Bijker, Pinch and Hughes, 1987), it is held that science and technology are socially shaped: they are not the result of objective, rational principles but are contingent outcomes of confrontations between humans and nature that involve valueand interest laden choices. Therefore, there is no principled dichotomy between science and technology on the one hand, and politics on the other. This is most obviously so for technology: technological artifacts and systems can be developed to reflect the narrow interests of designers or corporations, but can also be developed in a more democratic manner, to reflect the values and interests of a wide range of people (e.g. Sclove, 1995). Therefore, technological solutions to social problems are not necessarily less democratic than social solutions such as laws and protocols. Nevertheless, there is still a danger that BSTs become undemocratic technologies because their basic functionality may be largely decided by engineers, rather than by democratic representatives. A democratic design of a BST requires that a procedure is followed that is similar to the development of a new law: a careful consideration of alternatives, and a democratic procedure in which parties with different interests can let their voices be heard. The functional properties of the BST, if not the precise underlying mechanism, must be sufficiently clear to non-experts; this requires an effort by both the experts and laypersons to find a common vocabulary to discuss the political consequences of the technical choices that can be made. The idea that social problems can be solved by means of a technological ‘fix’ is perhaps more problematic in this context. The seatbelt case shows that engineers and policy makers are often overoptimistic about the usefulness of technologies in solving social problems. As pointed out by Norman (1988), the American seatbelt mechanism that required car drivers and passengers to wear seat belt had many problems. For example, it could not distinguish legitimate cases in which the seatbelt should not be buckled from illegitimate ones. If you wanted to carry a heavy package, for example, you were forced to buckle it. And the mechanisms were not reliable, so they often failed, wrongly buzzing and stopping the engine. Moreover, the mechanism was easy to circumvent by simply buckling the belts and stuffing them under the seat. This example shows that BSTs require careful assessment to make sure they are reliable, to work right, and to distinguish legitimate violations from illegitimate ones. This is very hard to achieve for any technological solution to a social problem.

Ethical aspects of behavior-steering technology

4.

363

THE RESPONSIBILITY ISSUE

The final moral issue relating to BSTs is that they may create a responsibility vacuum: a situation in which it is fundamentally unclear who is responsible for actions and their consequences. Many BSTs take away some amount of responsibility from their users. For example, there are automobiles that take over the brakes when a car goes into a skid. But this means that, in a resulting fatal accident not only the driver’s behavior can be blamed but also the car’s behavior itself, and the company that programmed the brakes must respond in a certain manner. In society, individuals are expected to take moral responsibility for their actions, which means that they can be held accountable for them, and can receive praise or blame for them. The use of BSTs complicates the assignment of responsibility, because users may claim that the responsibility did not rest with them.

5.

CONCLUSION: DEALING WITH THE MORAL ISSUES IN BEHAVIOR-STEERING TECHNOLOGY

When developing a BST, three moral issues must be considered: (1) what are its negative consequences for the individual freedoms of users, and can such consequences be morally defended? (2) Is there sufficient democratic input during the process of design and implementation of the technology? And (3) what implications does the BST have for the distribution of responsibility between the user on the one hand and the technology and its developers on the other? Regarding the first issue, I have suggested that BSTs should refrain from violating basic freedom rights, but may constrain human action if there is a greater good that is served by this, and I have suggested that BSTs should not erode autonomy by taking away decision-making power over the basic ways in which individuals live their lives. Regarding the second issue, I have argued that BSTs can be designed in more democratic and more technocratic ways, and for BSTs that are intended to provide policy solutions an effort should be made for the development process to be as transparent and democratic as possible. Also, it is desirable to anticipate the side-effects of the use of BSTs in real settings, so as to avoid BSTs from being poorly thought-out ‘technological fixes’. Regarding the final, responsibility issue, it is desirable that the question of how responsibility is distributed in the use of a BST is answered before a BST is put into use: involved actors (notably, the

364

Chapter 34

manufacturer and the user) should be clear about who is responsible when the use of the BST results in harmful consequences.

REFERENCES Berlin, I. (1979). “Two Concepts of Liberty,” Four Essays on Liberty. Oxford: Oxford University Press, p. 118-172. Bijker, W., Pinch, T., and Hughes, T., (eds.) (1987). The Social Construction of Technological Systems: New Directions in the Sociology and History of Technology. Cambridge, MA: MIT Press,. MacKenzie, D., and Wajcman, J., (eds.) (1999). The Social Shaping of Technology, 2nd ed. Open University Press. Norman, D. (1988). The Design of Everyday Things. New York: Currency. Sclove, R. (1995). Democracy and Technology. New York: The Guilford Press.

Chapter 34 A NORMATIVE SYSTEMS APPROACH FOR MANAGING TECHNOLOGY AND COLLECTIVE HUMAN ACTION

Sytse Strijbos

1.

INTRODUCTION

To most people technology means material artifacts, the ‘things’ we use for various purposes in our everyday lives. We take the car or we go by public transport to get somewhere. We sit down at our personal computer to send an e-mail or surf for information on the Internet. We pick up a telephone to congratulate someone on his or her birthday. And so forth. To most of us most of the time, technology means ‘things’ — and that approach to technology is central even in this book. The question of the relation between technology and behavior is, in that case, one of how we use things and how the things we use affect our actions.73 Yet technology is more than the set of material components we find in our surroundings. Technology also has fundamentally altered those very surroundings. Our everyday lives with technological things occur within a technological environment. Any approach made exclusively in terms of ‘things’ can therefore not help us to understand adequately the relation between technology and behavior. 73

The terms ‘behavior’ (gedrag) and ‘action’ (handelen) are used interchangeably here. I mean human acts based on a conscious decision of the will. These may have a routine character or be stylized by the forming of habits, to be sure, but they are not purely instinctive.

365 P.-P. Verbeek and A. Slob (eds.), User Behavior and Technology Development: Shaping Sustainable Relations Between Consumers and Technologies, 365-373. © 2006 Springer.

366

Chapter 34

That contextual factors play a role in the interaction between technology and behavior may be clear from a simple example. During our coffee break at work we go to the canteen and drink coffee from a little plastic cup. At the end of our break, we put the cup into a refuse bin especially designed for it. We know that such cups are meant to be used only once. They invite consumerist behavior. Yet this interaction between technology and behavior is only possible within a specific context. The little plastic cups we so nonchalantly discard fit into a collective consumer practice, and presuppose certain provisions for collecting and processing the refuse. In short, the relation between technology and behavior encompasses more than individual people using particular things. The use of technology of all sorts is embedded in a socio-technical context. This is true for something as simple as a little plastic cup, but even more true for complex artifacts such as automobiles. The latter are able to fulfill their intended function thanks to a system of highways, gas stations, road signs, toll booths, and a variety of other provisions. Implicit in the design of the automobile is the fact that it forms part of a more comprehensive system. This contribution to this book does not focus primarily on the level of the particular artifacts that individuals use. What follows is meant rather to shed light on the contextuality of technology and on the relation between technology and human behavior.74 Thus, I do not proceed from the things to a subsequent examination of the context. I look at matters from the other direction. Our point of departure is located in the context within which technological things can fulfill their function.

2.

THE CONTEXT OF SOCIAL PRACTICES

To apprehend technology in context one may begin by noticing that people make use of technological artifacts across a broad spectrum of practices. A researcher in a laboratory employs a microscope to examine the crystalline structure of this or that material. A dentist photographs a patient’s teeth. A cashier at the supermarket uses a scanning apparatus to read the prices of the products a customer takes from a shopping cart and places on a conveyor belt. And so forth. Technology and behavior are inseparable; they appear in their mutual interconnectedness in concrete situations and practices. Such concrete practices, or ways of doing things, steer in important measure how technology is put to use. 74

Elsewhere I have discussed the de-contextualization of technology and its implications for the problem of development and underdevelopment in a global society (Strijbos 1998).

A normative systems approach for managing technology

367

By situating technology and behavior in various practices I expressly distance myself from notions of technology as a neutral means or thing that people may dispose of in order to realize some purely subjectively determined goal. People do not act and live in total isolation from each other, but are connected with each other in ties and relationships that have grown up historically and that are culturally conditioned. One’s actions and ways of using technology are conditioned by the context and steered by common ‘practices’ that take on a definite form in social interaction between people. To attain a more systematic grasp of the relation between technology and behavior it is therefore necessary to have a closer look at the concept ‘practice’.75

2.1

Structure and direction 76

‘Practice’ is a plural concept, as we have already seen. Human actions occur within a diversity of practices the uniqueness of which we generally recognize without effort in everyday life. Thus, there are educational practices, commercial practices, social assistance practices, legal practices, and the like. And upon closer examination these practices represent various types that are susceptible to further differentiation. For example, there is education for the professions but also an educational practice intended to instill in people the scientific attitude of thought required in a particular discipline. Sometimes we are critical, whereupon we may say, for example: “this is no longer scientific education.” In short, we have the capacity to recognize something as a particular practice, and we have implicit or explicit knowledge of the norms for that practice. We clearly recognize or identify a practice from a complex of norms that together ‘bring to expression’ its normative structure. Moreover, this normative structure determines in important measure the interaction between technology and behavior in a particular practice. It makes, for example, quite some difference indeed whether a rifle is used as a weapon in a situation of war or is intended instead for hunting. The rules for good use of a rifle are given, in part at least, with the context of these different practices. I have just spoken deliberately of the ‘bringing to expression’ of the normative structure of a particular practice. Different practices or ways of doing things arise and develop in the interaction between people and are thus 75

76

An analysis of the concept ‘practice’ focused on medical practice may be found elsewhere (Polder et al. 1997, p. 413; see also Jochemsen, Glas and Hoogland, 1997, pp. 65ff.). A broader discussion of the distinction between ‘structure’ and ‘direction’ in the sense meant here is presented by Wolters (1986).

368

Chapter 34

variable as such. In other words, the form of a particular practice is not fixed for all places and forever. There is a certain cultural variation. Furthermore, this variation is related to a choice of direction in human action that is conditioned by a world or life view. While ‘structure’ pertains to the obtaining complex of norms for a particular practice, ‘direction’ indicates an intellectual or spiritual dimension in human behavior. People receive orientation from the particular understanding they have of an activity, and act out of a spiritual motivation, ethical attitude or ethos. The various notions people may have consciously or unconsciously of a particular practice belong to the ‘direction’ that underlies the practice. These notions steer people’s behavior and the way in which they work out the ‘structure’ of a practice in a concrete situation, including the role of technology. I shall return to this in section 4. We shall see that the way people view a practice is translated into the technology of the practice.

2.2

‘What and ‘how’, qualifying and founding aspects

The distinction between structure and direction does not exhaust an analysis of the concept ‘practice’. In the complex of norms that together determine the normative structure of a particular practice, one may make further distinctions. Consider for a moment the diversity of practices of a society and culture, from weaving garments, plowing the ground, and fashioning wood to providing education, managing an enterprise, or organizing a celebration. In all these different cultural practices two structure-determining components may be distinguished. There are norms that pertain to the ‘what’ of a practice. These norms are connected with what one may call the qualifying aspect of a practice. The ‘what’ of something is qualifying for the practice — a health and welfare practice, for instance — and makes it structurally different from other practices — from the commercial practice of an entrepreneur, for instance (cf. Strijbos 1999, p. 19-30). However, there are also norms connected with ‘how’ something is done, with what has been called the technology of a particular activity, with the “ways of doing something” (Franklin 1990, p. 15). The norms that give expression to the ‘how’ of something refer to what one may name the founding or technical side of a practice. ‘What’ and ‘how’ are, thus, two sides of one and the same practice, and in concrete behavior they are inseparably connected with each other. The technology of a practice, the ‘how’ of it, cannot be isolated from its qualifying aspect, the ‘what’ of the particular practice. As ethically qualified care, the help offered in a medical practice acquires concrete form in technical procedures and prescriptions. “One has to keep in mind,” Franklin

A normative systems approach for managing technology

369

(1990, p. 17) correctly observes, “how much the technology of doing something defines the activity itself, and by doing so, precludes the emergence of other ways of doing ‘it’, whatever ‘it’ might be.” Countless ways of doing things in our technological society know their own specific form of technological practice, including the artifacts that may form a part to it.

3.

CONTEXT OF SOCIO-TECHNICAL SYSTEMS

The context for human action and for interaction with technological artifacts is not just given with normative practices, with human ways of doing things. In contemporary society these practices are in turn often embedded in a diversity of socio-technical systems. These systems constitute an extension of the founding side of human practices in our technological society; they form a technological substructure of these practices. The technology of practices includes, to speak with Pacey (1983, p. 6), a technical and organizational component.77 The technical component pertains to the level of the things (tools, apparatuses, machines) and of the knowledge and skills required to use them. These are the sorts of things that people usually think of in connection with technology. Organization as an institutional component is, however, certainly of no less significance and may in fact sometimes be of even greater importance than the technical dimension. A classic example in which technology and organization meet is work in the factory around the machine, typical for the rise of the Industrial Revolution. Yet similar developments have occurred, for example, in education, medical care, and the agrarian sector. By organization I generally mean here “all varieties of technical (rational-productive) social arrangements” (Winner 1977, p. 12).

4.

NORMATIVE TYPES OF TECHNOLOGY

At this juncture in my reflections I return to the point introduced above, namely, the influence of fundamental choices of direction in technology. I want to consider more closely now how choices of direction that occur within a society at the collective level become translated into technological 77

In addition to the technical and organizational components mentioned here, Pacey’s concept of “technology practice” also features a cultural component. It corresponds with what I have called “direction.” In a later publication Pacey (2001, p. 7) distinguishes “personal values and individual experience of technology” from “social and cultural meanings.”

370

Chapter 34

practices. To speak of choices of direction at the collective societal level is to speak of human acts that shape the conditions for practices. These are acts that Van Asperen (1981, p. 96) refers to tellingly as regulation, which is to say “the conscious creation of a social context” within which other people act and interaction with technology occurs.

4.1

Holistic and prescriptive technology

In her fascinating book The Real World of Technology (1990) Ursula Franklin, professor emeritus of metallurgy at the University of Toronto, makes a valuable distinction between two types of technological practices having entirely different characteristics. She speaks of holistic technology and prescriptive technology. The first sort is typifiable as ‘work-related’ technology that facilitates an existing practice. As an example, Franklin mentions the replacement of the mechanical by the electric typewriter. The work remained the same but its execution was improved considerably. Prescriptive technology, in contrast, does not focus on an established work process in order to support it, but has as its goal controlling the work process itself. It is ‘control-related’ technology. An example is the introduction of the word processor. An independent personal computer is still a ‘workrelated’ technology that enables people to make all kinds of documents more easily, to be sure. Yet the moment word processors are coupled together in a network, the technology is ‘control-related’. “Now workers can be timed, assignments can be broken up, and the interaction between the operators can be monitored” (Franklin 1990, p. 18). There is a second important difference between holistic and prescriptive technology. Both types of technology feature a division of labor, but of different sorts. In holistic technology, the division of labor corresponds with specialization in the fashioning of a particular product. This type of ‘specialization by product’ is distinctive of the classical artisanship of potters, weavers, and cooks. In such vocations, the worker can decide how to proceed from beginning to end. In this type of specialization the ‘doer [remains] in total control’ of his or her own work (Franklin 1990, p. 19). Prescriptive technology is marked by precisely the opposite, namely by ‘specialization by process’ and a shift to ‘external control and internal compliance’ (Franklin 1990, p. 23). The process of fashioning, the making or doing of something is now split into steps, and as a rule a different person is responsible for each step. In this type of division of labor, which we have been familiar with since the Industrial Revolution, direction of the entire

A normative systems approach for managing technology

371

work process is in the hands of an external party.78 This external control of the work process may be accounted for, at least where modern technology is concerned, as a consequence of the overestimation of the scientific method that underlies it. The distinguishing feature of this method, namely, is control at a temporal and physical distance (Van Riessen 1949, p. 698).79 A final important difference between holistic technology and prescriptive technology concerns their scale. The scale of holistic technology is limited by conditions of natural growth. Just as for an organism, so too for holistic technology, “…growth itself cannot be commandeered; it can only be nurtured and encouraged by providing a suitable environment” (Franklin 1990, p. 27). While holistic technology is marked by a ‘growth model’, prescriptive technology is based on a ‘production model’. In this model, the emphasis is on predictability and the idea of the complete controllability of technology. Society today according to Franklin has been overrun by prescriptive technology. A torrent of technological innovations and artifacts has been directed at us, accompanied by the constant promise that these all make life easier. Reality, however, is often different. Franklin gives as an example the introduction of the sewing machine in 1851. This was praised as an apparatus that would free women from boring, time-consuming work in their homes. Yet the promise of liberation quickly turned into the exploitation of women as workers outside the home. “With the help of the new machines, sewing came to be done in a factory setting… The sewing machines at home were used less… and garments came to be readily available on the mass market… Indeed, women sew less, cook less, and have to work hard outside the home to be able to buy clothing and food” (Franklin 1990, p. 101). This example of the sewing machine illustrates that contextual factors have an important, and possibly even a decisive influence on the use of a given technology. “What turns the promised liberation into enslavement,” says Franklin, “are not the products of technology per se ʊ the car, the computer, or the sewing machine ʊ but the structures and infrastructures that are put in place to facilitate the use of these products and to develop dependency on them” (Franklin 1990, p. 102). 78

79

According to Franklin (1990, p. 20) prescriptive technology is in fact much older than the modern scientific technology that appeared with the Industrial Revolution in England. The late Roman world already knew forms of prescriptive technology for making earthenware. Even a thousand years earlier than that, prescriptive technology was applied on a large scale in ancient China, during the Shang dynasty (about 1200 B.C.) for casting bronze. Haaksma (1999) provides a valuable introduction to Van Riessen’s work as a philosopher of technology.

372

4.2

Chapter 34

Disclosive technology80

In Franklin’s studies two normative types of technological practices, replete with their underlying conflicting choices of direction, are drawn up opposite each other: holistic versus prescriptive technology. The first gives priority to the subject's freedom of action, the second desires comprehensive control of the technological system. Franklin seems to seek the solution in a compromise or balance between the two positions.81 The tension between the freedom of the individual user and the quest for behavioral control at the macro level is not abolished, however, by such an in-between position. Measures at the collective systems level quickly encounter resistance because they set boundaries for the freedom of the individual. By the same token, viewed from a macro perspective, the allowance of room for the actions of individuals involves a risk factor for control. As long as our thinking remains caught up in the contradiction in question, we will be hindered in our search for new creative solutions to the diverse systems problems of the technological society. Our frustrated efforts to solve the problem of mobility may suffice to illustrate this. The enormous growth in mobility increasingly confronts society with irresolvable conflicts between the interests of citizens in individual mobility and societal interests in accessibility and ecological sustainability. It has therefore been correctly observed that “…it will be important to lift the discussion of mobility out of the tension between individualism and collectivism” (Van der Stoep and Kee, 1997). Let us, accordingly, look more closely, finally, at the field of forces under discussion here to see if one cannot break through the polarity. It is striking and eminently worth noting, then, that both poles of the choice of practical direction are subjectively determined. Left unanswered is the question concerning what end is served by safeguarding the freedom of individuals or what the destination or goal or purpose of the claimed freedom may be. Freedom to what? Something of the same sort can be said of the quest for control. To what end is this control directed? What is the ‘cause’ or issue that we desire to control? In other words, what way of doing 80

81

‘Disclosive technology’ as a normative concept for technological practices is related to ‘disclosive systems thinking’, a normative approach to systems for intervening in these practices. Elsewhere I have expanded more broadly upon ‘disclosive systems thinking’ as a normative methodology of practices for ‘managing’ a technological society (see Strijbos 2000 and Strijbos 2003). Franklin seeks the solution for the problems of our technological world in regaining terrain for more holistic practices and launches the idea of ‘redemptive technology’. The Canadian philosopher of technology Stahl (1999) has worked this idea out further.

A normative systems approach for managing technology

373

things is at stake? And what is the normative structure of this practice? I believe this — the normative structure of a practice — is the crucial point. The choice of direction in a way of doing things must be oriented to the intrinsic meaning or destination of a particular practice. Technology must serve and create room for practical possibilities, which is to say for the disclosure of a given practice consistent with its normative structure.

REFERENCES Franklin, U. (1990). The Real World of Technology. Montreal, Toronto, New York and London: CBC Enterprises. Haaksma, H.W.H., et al. (1999). Van Riessen, filosoof van de techniek. Leende: Damon. Jochemsen, H., Glas G., and Hoogland J. (1997). Verantwoord medisch handelen. Amsterdam: Buijten & Schipperheijn. Polder, J.J., Hoogland, J., Jochemsen, H. and Strijbos, S. (1997). Profession, Practice and Profits: Competition in the Core of Health Care System. Systems Research and Behavioral Science 14 (6): 409-423. Pacey, A. (2000). ninth printing. The Culture of Technology. Cambridge, MA: MIT Press. Pacey, A. (2001). Meaning in Technology. Cambridge, MA: MIT Press. Stahl, W.A. (1999). God and the Chip: Religion and the Culture of Technology. Waterloo, Ontario, Canada: Wilfrid Laurier University Press. Strijbos, S. (1998). The Problem of Development and The Decontextualization of Technology: A World-System Approach. World Futures 52: 333-346. Strijbos, S. (1999). Kiezen en keuzen: Ethiek in de tandheelkundige praktijk. Houten and Diegem: Bohn, Stafleu, Van Loghum. Strijbos, S. (2000). Systems Methodologies for Managing our Technological Society: Towards a ‘Disclosive Systems Thinking’. Journal of Applied Systems Studies 1 (2): 159-181. Strijbos, S. (2003). Systems Thinking and the Disclosure of a Technological Society: Some Philosophical Reflections. Systems Research and Behavioral Science 20 (2): 119-131. Van Asperen, G.M. (1981). Normatieve theorie en kollektieve beslissingen. Algemeen Nederlands Tijdschrift voor Wijsbegeerte 73: 94-106. Van Riessen, H. (1949). Filosofie en techniek. Kampen: J. H. Kok. Van der Stoep, J. and Kee B. (1997). Hypermobility as a challenge for systems thinking and government policy. Systems Research and Behavioral Science 14 (6): 399-408. Winner, L. (1987). sixth printing. Autonomous Technology: Technics-out-of-Control as a Theme in Political Thought. Cambridge, MA and London, UK: MIT Press. Wolters, A.M. (1986). Creation Regained: A Transforming View of the World. Leicester, UK: IVP Press.

Chapter 35 SHAPING TECHNOLOGY-BEHAVIOR INTERACTIONS: Lessons for Policy Making

Wim Hafkamp

1.

INTRODUCTION

The objective of the final part of this volume was to derive approaches to and strategies for policymaking from the insights into the dynamics of interaction between technology and behavior. While doing so, the question of legitimization of behavior-steering technologies remained an important one. The contributions to this part were written from a ‘macro’ perspective. Brey and Strijbos focus on issues of legitimization, in particular on the moral justification of behavior-steering technology, and the role of normative structure in technology development, i.e. in the social environment. Laws and Elzen explore technology and behavior in the dynamics of ‘local’ policy processes. Popkema and Van Schagen discuss technology and behavior interactions from a cognitive, learning perspective in management systems and road design respectively. Finally, Van Vliet discusses the evolution of the role of end users, consumers in processes of technological change. The focus of this chapter is to use concepts, phenomena, applied and empirical work of the contributions by these authors in deriving ‘learning points’ for policy, and improving policy designs. Therefore, this section will begin with a brief recapitulation of the technology-behavior dilemma in environmental policy.

375 P.-P. Verbeek and A. Slob (eds.), User Behavior and Technology Development: Shaping Sustainable Relations Between Consumers and Technologies, 375-383. © 2006 Springer.

376

2.

Chapter 35

THE TECHNOLOGY-BEHAVIOR DILEMMA IN ENVIRONMENTAL POLICY MAKING

As national environmental policies have reached their limits in solving the issues of resource depletion, environmental pollution and loss of biodiversity, the question arises as to how to move beyond this critical point. We know that it is necessary to overcome the observed deficits of environmental policy in terms of enforcement, implementation, a range of instruments and strictness of measures. This is the main reason for studying the complex linkage of technology and behavior. The question becomes more pressing because national governments are losing ground to supranational entities for addressing global issues, not just environmental issues but social and economic ones as well (international trade, migration, regional conflict, security, etc.). Recent national environmental policy documents, such as the National Environmental Policy Plans (NEPP 3 and 4) in The Netherlands, suggest that a ‘quantum leap’ in policy-making is required. This was also recognized previously, in 1997, when the Dutch government published a policy document on Environment and Economy (Nota Milieu en Economie, 1997). These three policy documents build on the framework introduced in the national program on Sustainable Technology Development (Van Kasteren, 2002). The central element in this framework is the joint articulation by actors in a certain domain (e.g. transport, agriculture) of a long-term vision out of which a short term agenda is derived through a back-casting process. With the start of NIDO (the ‘Dutch National Initiative for Sustainable Development’), an attempt was made to create the necessary knowledge infrastructure for this type of transformation process. The Pathways Project, initiated by Ashford of MIT (Ashford et al., 2002), studied three cases of transformation and derived a first, generic policy agenda. For NEPP4, Rotmans et al. (2000) introduced the concept of transition management. In the context of these long-term processes of change we find a second reason to study the complex linkage between technology and behavior. Technology and behavior are often seen as a dichotomous choice for policymakers. This becomes most apparent when environmental standards for specific technologies (e.g. European standards for cars) do not prove to be sufficiently effective. Policymakers then start to argue for complementary policies that address behavior (e.g. reducing vehicle kilometers, introducing speed limits). For over 30 years, analysts and environmentalists of various ideological colors have kept this dichotomy alive.

Shaping technology-behavior interactions

3.

377

THE SCORE IN POLICY SCIENCES

Whichever the issue and school of thought in policy theory, a proper starting point is always found in thorough problem analysis, including the societal context, the actor field, actor interests, their underlying values, basic beliefs and world views. Precisely those elements appear clearly in several contributions in this part of the book. Strijbos introduces the normative structure, underlying a ‘practice’, which has a technical and an organizational component, situated in a socio-technical environment. Clearly, in his view, technology and behavior cannot be separated. Even in the more conventional approaches to policy analysis, where potential policy measures are identified, assessed and prioritized, the distinction between technology and behavior is never a division. Effectiveness counts, along with efficiency, feasibility, and similar criteria. Policy instruments are seen as interventions by government to bring about change toward public goals in an otherwise static world. Technology is treated as a given, or else an instrumental variable in the policy process (e.g. technology forcing standards). This leaves little room to question the role of technological development. The policy debate, in this approach, is likely to side step what Strijbos calls ‘choice of direction’. This is the ethical discussion among actors involved on what type of technology to adopt. In his contribution Strijbos discusses a typology of holistic and prescriptive technology. While choice of direction of technological development does not fit the concepts of conventional policymaking, it does very well fit more recent ideas of interactive policy-making. Fundamental differences in value orientation between actors are acknowledged in concepts of arena models and open planning processes. In process architecture the rules are set, procedures agreed upon, and specific techniques selected, such as joint factfinding, creative competition, negotiation, and arbitration. In the policy arenas, relevant actors operate, and their dialogue on problems and solutions is value-based from the start, it extends to issues of moralization (of technology, actors’ behavior) in a way that transcends the dichotomy between technology and behavior. The actual debate, throughout the process, circles around underlying values held by actors, and questions of responsibility, liability for environmental damage or lack of safety. So, this is not just about the behavior of end users (consumer, driver, dweller),

378

Chapter 35

but also about the behavior of other actors: retailers, government agencies, manufacturers, industrial designers, engineers, scientists, environmental organizations and citizens alike. Policymaking is thus about changes in the current, established practices that are maintained by all actors, to their collective detriment. The practice (e.g. passenger transport) is at stake, more than the technology itself (e.g. the automobile). In the current understanding of policy processes, the debate on ethical aspects of BST (behavior steering technology), discussed by Brey, should be embedded in the ethical discourse that is part of the policy process. This ethical discourse, however, is often not formally addressed, with farreaching implications for the success of the process. One example is given by the negotiations between the Dutch government, industry and NGOs on littering in 2001. Under a negotiated agreement on packaging waste, industry had agreed to specifically reduce the littering caused by single-use packaging. This involved not just cans and small plastic bottles for beverages, but also wrappers of candy bars, etc. Throughout the process, industry, government and environmentalists kept to their cast iron ‘points of view’ on who is primarily responsible: industry for pushing wasteful consumer behavior, government for failing to place and empty bins in public spaces, or consumers for plain littering. As a first policy lesson, Brey discusses some of the values at stake: freedom, responsibility and democracy. He goes on to show that BST may be acceptable to a certain extent if it affects ‘negative freedom’, which means it adds to existing limitations on an individual’s behavior. He sees this as a matter of weighing pros and cons. He questions BST if it affects ‘positive freedom’, which means that it erodes human autonomy. This is another aspect of the ethics of policy proposals, which is often lacking in the current policy discourse, a second policy lesson. The issue, however, is not only whether BST is acceptable or not, but also which ‘changes in practices’ (changes in technology in the wider sense) are acceptable. It is the policy process itself (arena, process design, actors) which resolves the first of his three moral issues, namely that of technology versus basic freedom or erosion of autonomy. The second of his moral issues holds that BST might be considered a technocratic solution, not a democratic one. Brey argues that BST can be designed in a democratic way, although he does not discuss which way that is. Again, modern policy theory would hold that this is a matter of legitimacy of the policy process, and of process design. These need to warrant the democratic nature of policy processes, and thus of the ‘changes

Shaping technology-behavior interactions

379

in practices’ adopted, whether BST or not. As a third policy lesson, the legitimacy of a technology is to be found in the development process, not in the technology itself. The third of his moral issues is about the responsibility for failure of BST. If BST is developed and adopted in an autonomous process by designers, engineers, corporations, etc., then this issue may remain unresolved. However, it is the very nature of interactive policy processes that allocation of responsibility to actors in the new situation is taken care of. Apparently, the issues put forward by Brey, and Strijbos’ question on direction of technological change, are more easily addressed and resolved in interactive policy processes than in ‘closed’, processes (fourth policy lesson). This goes for the ethical aspects of technology, as well as for the ethical aspects of behavior of all actors. Strijbos discusses the distinction between holistic and prescriptive technologies (with a reference the 1990 publication by Franklin, ‘Technology in the Real World’). At least two actor groups have an interest in prescriptive technologies: industry (market dominance, dominance in supply chains, control over workers) and government (technology as an instrument towards public goals). The interests of these two actor groups do not necessarily coincide, while their interests do not coincide with those of workers, citizens and end-users. As a fifth policy lesson, policymakers should be well aware of this distinction, as it thoroughly affects the success of a policy process and the effectiveness of a policy.

4.

INTERACTIVE PROCESSES

In the previous section, we saw that concepts introduced by Strijbos and Brey lend themselves in particular for interactive policy processes, and for interactive processes in general. Elzen’s paper applies some of the notions of Strijbos (choice of direction, role of the socio-technological system) and demonstrates the dynamics of interactivity. He rejects the dichotomy of technology and behavior, along with other authors, and introduces the concept of socio-technical learning: solutions for perceived problems through activities in the socio-technological context that aim at changing practices. Using the example of local experiments with electric vehicles (EVs) he demonstrates how technology and behavior (not just of drivers) change over time, and how all actors learn in the process. From the concepts presented in part I of this book we can explain these phenomena. While designers of products and services may in fact pass on a script to the users

380

Chapter 35

(of designed artifacts, services, (trans)actions), the user will find affordances. Script and affordance are not independent. We can see an interaction between the two concepts: script provokes one category of intended use, affordance brings other, especially unintended, categories of use. Affordance affects script. The contribution by Elzen shows how the introduction of electric vehicles ‘scripts’ a certain kind of use, but also allows, or ‘affords’ users other kinds of use. These are then used by policymakers to adapt the proposition of electric vehicles to the users. This displays an explorative process in which technology and behavior co-evolve. Interactivity makes this reciprocal relationship a learning process. At the meta-level: there is a script on the process design itself. It specifies, asks for, provokes, certain behavior by actors (information sharing, value dialogue, etc.). They, in turn, discover affordances in the process (settle an old score, proposing side deals, etc.). The sixth policy lesson, though only by observation: policy processes can be regarded in the same way as technologies: they also have ‘scripts’ and provoke ‘affordances’ with their users. Different types of learning apply which resemble the typology used in safety management, as described in the contributions by Popkema and Van Schagen. They provide a seventh policy lesson. In processes of policy and technology development it is important to know how technology and the technological environment affect skill-based, rule-based and knowledgebased behavior. This is supported by Elzen’s contribution. The scale of interactive policy processes is not necessarily the national scale. Under the influence of the disappearance of the top-down, hierarchical state as the dominant actor, these processes become local more often, have a smaller scale, and allow for multiple learning experiments. A knowledge infrastructure emerges in which different types of knowledge flow (common sense, professional experience, scientific knowledge, wisdom, intuition). Diversity in experiments, whether in niches or not, serves more purposes than learning, it also reflects diversity in consumer preferences, industrial capabilities, and cultural identities. The contribution by Laws demonstrates this eighth lesson for policy. Elzen describes a process in which policymaking is no longer a single step (‘one shot’) outside intervention in technology, behavior or its context, but a process that takes place within that context. He further articulates the diversity in interests and values that characterize the actors who encounter one another and the strategies they employ to manage their differences. Interests that are not identical do not necessarily conflict. Diverse values often overlap when they are interpreted

Shaping technology-behavior interactions

381

in concrete settings. In these pragmatic alignments, actors frequently discover a common and surprisingly stable basis for action. Laws introduces the concept of transitional spaces where actors operate in public entrepreneurship networks. In his transitional space (specific instance of arena), there is not necessarily a steering entity, a government policy agency. In different stages of the process, various actors may be pushing/pulling. Between and through the discourse ʊ values, ethics; communication on interests; fact finding; options generating ʊ the network engages in R&D, design, prototyping, pilot projects, etc. A continuous process in which there is a high degree of serendipity. This may be considered a ninth lesson for policy: public entrepreneurship may prosper in the absence of a dominant government agency. Laws identifies the various roles that must be played in such a network in order for it to be productive. One of them is ‘steward of the common good’. The end user, the consumer, plays an essential role in all of this. After all, it is the consumer whose behavior we implicitly refer to when we use the term ‘technology and behavior’. More than those by Laws and Elzen, it is the contribution by Van Vliet which shows how the roles of consumers may evolve while technology on the supply side of the electricity sector changes. His examples of ‘concerned consumers’, ‘citizen consumers’, ‘captive consumers’, and ‘co-providers’ is telling. Where was the script changed, and which affordances did that lead to?

5.

IMPLICATIONS FOR POLICY

In this chapter nine policy lessons were deduced from the six contributions in this part of the book (see box). All of these lessons lead to one overriding conclusion: policy should not accept the dichotomy of technology and behavior. An exclusive focus on technology will soon run into problems of unintended consequences, some of which are likely to undermine the stated policy goals. These may range from rebound effects ʊ the script may change, but so do the affordances ʊ to ethical discourse on the ultimate and inherently clean and safe technology that leaves no autonomy to the individual. An exclusive focus on behavior runs into moralization immediately. This was the case when in Rotterdam, on the A13, a speed limit of 80 km/hr was introduced on a section of this throughway where transport emissions and noise cause considerable health effects for residents in adjacent apartment buildings. The speed limit is

382

Chapter 35

strictly enforced, and drivers are informed about the environmental background of the speed limit. In response, some have now begun to drive there at night and honk their horns, in sheer protest. Reinterpreting Brey’s point about the moral dimension of BST, one could conclude that ethical aspects should be addressed in all policy processes that affect technology and behavior. Technology is a practice where values are brought to expression. There is a normative structure that underlies the duality of technology and behavior. Thus, by implication, any policy process that aims at changing a particular practice is a process in which values are brought to expression. In practice, we see that the value debate is often referred to politics, while policy formulation is (supposed to be) a rational process between actors. It is not. Nine policy lessons 1. In policy processes that address technology and behavior issues, values such as freedom, responsibility and democracy are at stake, and should be debated. 2. Behavior-steering technologies affect human autonomy. This aspect of the ethics of policy proposals is often lacking in the current policy discourse, and should be incorporated. 3. The legitimacy of a technology can be found in the development process of the technology, not in the technology itself. 4. Policy issues about technological change are more easily addressed and resolved in interactive policy processes than in ‘closed’, proprietary processes. This goes for the ethical aspects of technology, as well as for the ethical aspects of the behavior of all actors. 5. Policymakers should be well aware of the different interests of all involved actor groups ʊ industry, workers, citizens, consumers ʊ as these affect the success of a policy process and the effectiveness of policy. 6. Policy processes can be regarded in the same way as technologies: they also have ‘scripts’ and provoke ‘affordances’ when put into practice. 7. In policies aiming at technology development, it is important to know how technology and the technological policy environment affect skillbased, rule-based and knowledge-based behavior. 8. Experiments, whether in niches or not, serve more purposes than learning; they also reflect diversity in consumer preferences, industrial capabilities, and cultural identities. 9. In the absence of a dominant government agency, public entreentrepreneurship may prosper. However, the role of ‘steward of the common good’ must be played.

Shaping technology-behavior interactions

383

The term ‘technology and behavior’ refers to the behavior of the end users. Technology is more than the equipment used by end users. Technology is embedded in a practice, the same one in which behavior is embedded. All actors are users in this practice, not just the end users. Technology itself needs a reinterpretation, as offered by many contributions in this volume. This brings to light the explicit roles of engineers, scientists, designers, manufacturers, retailers, marketers, government agencies, and politicians. They form the actor network in which interactions between humans and artifacts take shape, are reflected upon, and adapted. Policy makers are not outside this network, intervening with policy measures aimed at public goals, but operate within that network, using their ‘means’. The ‘resultante’ of the efforts by all actors determines the outcome. The question that was formulated by the editors of this volume at the start of the introduction to this part of the book was: how can we apply the insights in technology-behavior interactions to policy design and development, and what does this imply for the organization of the policymaking process? Above we derived nine policy lessons to answer this question. In summary, we can say that we have left the era of the ‘one shot’, let alone ‘from the hip’, top-down, rational/analytical policy intervention. Interactivity, participation, recursiveness, and learning are becoming key concepts. Add to that locality, diversity and serendipity, and new patterns emerge of how society addresses its problems. Policy-making becomes ‘co-production’ among all actors involved. In this process, technology and behavior coevolve, and interactions between the two are shaped.

REFERENCES Ashford, N., W.A. Hafkamp, F. Prakke and P. Vergragt (2002). Pathways to Sustainable Industrial Transformations: Co-optimising Competitiveness, Employment and Environment, research paper, MIT, Cambridge. Kasteren, J. Van (2002). ‘Duurzame Technologie, Ontwikkeling van een Duurzame Wereld’, Wetenschappelijke Bibliotheek, Amsterdam. Nota Milieu en Economie, Ministerie van VROM, Den Haag, 1997. Rotmans, J., R. Kemp, M.B.A. van Asselt, F.W. Geels, G. Verbong and K. Molendijk (2000). Transitions & Transition Management, ICIS, Maastricht.

Chapter 36 ANALYZING THE RELATIONS BETWEEN TECHNOLOGIES AND USER BEHAVIOR: Towards a Conceptual Framework

Peter-Paul Verbeek and Adriaan Slob

1.

THE NEED FOR A SOCIOTECHNICAL APPROACH

In myriad ways, the authors of this book have discussed the intricate connections between technology design and user behavior. The specific properties of technological devices appear to have close relations with the behavior of the people using them. This makes clear that the sharp distinction that is often made between the functional world of technological artifacts, on the one hand, and the social world of human beings on the other, needs to be abandoned, both in theory and in practice. Technologies always have consequences for human behavior, and for understanding human behavior it is necessary to take into account the ways in which it is influenced by technology. For this reason, policy making that aims to influence human behavior should also pay attention to the technological context of behavior, and policy making regarding technology cannot ignore the behavioral effects of technologies. The classic distinction in policy making between social approaches which pertain to human behavior and technical approaches pertaining to technology should give way to a ‘sociotechnical approach’, as it is called in the field of Science and Technology Studies. This makes the study of technology-behavior interactions an inherently interdisciplinary activity. This accounts for why disciplines from the entire 385 P.-P. Verbeek and A. Slob (eds.), User Behavior and Technology Development: Shaping Sustainable Relations Between Consumers and Technologies, 385-399. © 2006 Springer.

386

Chapter 36

scientific spectrum have contributed to this book: from psychology to policy studies, from industrial design to philosophy, from consumer research to science and technology studies, and from safety studies to marketing research. Precisely the lack of interdisciplinary cooperation in the study of technology-behavior interactions was one of the main reasons for composing this book. In this concluding chapter we will first present an interdisciplinary vocabulary to describe and explain relations between technological products and user behavior, which we draw from the analyses in this book. After this, we will formulate some overall conclusions regarding the implications for technology design and policy-making.

2.

A VOCABULARY TO DESCRIBE RELATIONS BETWEEN TECHNOLOGICAL PRODUCTS AND USER BEHAVIOR

Knowledge about the interaction between technology and behavior exists dispersed over and fragmented in many different disciplines among which there is little interaction. An important first step for the development of interdisciplinary research in this field is to identify and create ‘common ground.’ In this section we aim to contribute to this by formulating a basic conceptual framework which links insights and concepts from all approaches involved, while still aiming to do justice to the specificity of the various disciplines. The first two parts of this book made clear that technologies and users should not be studied as separate entities, but as always interrelated. Technologies inevitably influence the behavior of their users, once they are used. And conversely, user behavior ʊ like existing habits, or interpretations of the functionality of artifacts ʊ always co-determines how technologies will eventually function. The subject to study, therefore, is not ‘technology’ or ‘user behavior’ in themselves, but the interaction between both. Moreover, many of the chapters in Part 1 and 2 make clear that this interaction always takes place in an environment, which also plays an important role in the way in which the interaction gets shape. Two types of environments can be distinguished. First, there is what we would call the technological environment of human behavior. Technologies are never able to function independently from an environment of other technologies and infrastructure. No electric device can be used without electricity networks

Analyzing the relations between technologies and user behavior

387

and power plants; no car can be used without roads and fuel stations. Second, the technology-behavior interaction takes place in a social environment. Technology users are not isolated individuals. They find themselves in social networks with social codes, and in a cultural context with specific values and preferences. This social environment, too, helps to shape the relations between human beings and technologies. Each interaction between technologies and users, therefore, depends on the specific properties of the technology and the users involved, and on the social and technological environment in which the interactions take place. Within this specific constellation, performances come about, which are both human and technological in nature. Users perform specific actions on the basis of their interaction with technology (in its socio-technical environment), and technologies perform specific tasks on the basis of the ways in which they are embedded in user practices. Figure 35-1 illustrates this interaction.

performance

user behavior

technology

technological environment

social environment

Figure 35-1. Technology and user behavior

This basic scheme to analyze interactions between technology and user behavior makes it possible to discuss and relate the various concepts and approaches used in the chapters of the first two parts of the book. We will shortly do this by highlighting some concepts and approaches to analyze (a) the interaction between technologies and users; (b) the social environment of this interaction; and (c) the technological environment.

Interactions between users and technologies In Parts 1 and 2, a large number of concepts and models are presented that can be used to analyze different aspects of the interaction between users

388

Chapter 36

and technologies. We will not discuss all these concepts here again, but only highlight concepts from the various disciplines to create a common ground. As for the interaction process itself, the concept with the longest history is affordance. This concept from environmental psychology indicates that people perceive technologies in terms of their ability to ‘afford’ specific actions: a door handle is perceived as ‘turnable’; a button as ‘pushable.’ This implies that, even at the level of conceptualization, technologies and user behavior cannot be separated. Other concepts to analyze the interaction between technologies and user behavior are: - mediation (from the philosophy of technology): when being used, technologies mediate the relationship between users and their environment, and thus help to shape human behavior; - configuring (from science and technology studies): technologies-in-use configure users, by demanding specific actions from them and delegating specific responsibilities to them; - moderation (from applied psychology), indicating how technology users are able to modify the impact of technological change; - scenarios (from safety studies): specific interactions between humans and technological appliances can be analyzed in terms of scenarios; - compatibility (from household studies), indicating the quality of the interaction between a technology and the household system; - person-environment fit (from environmental psychology), indicating the congruence between environmental resources and demands, on the one hand, and user needs and abilities, on the other. The interaction between technologies and users is informed or gets shaped by the specific characteristics of the users and of the technologies involved. On the technology side of the interaction process, there are two ways in which technologies can play a role in influencing human behavior. The first is ‘informational’ in nature. Technologies can provide users with information that can be used to make decisions on what to do. This information can be ‘static’ or ‘dynamic’. Static information consists of indicators that are ‘built into’ the technologies involved, e.g. large buttons with a marked color to indicate the power switch of a device. Dynamic information is dependent on the way the technology is used. Feedback given by technologies is the most important form of this. ‘Intelligent devices’ can provide feedback about, for instance, energy consumption or the need to clean or replace a part of the device, which can stimulate users to interact with the device in a specific way. The second way in which technological devices can evoke specific kinds of behavior is ‘material’ in nature. Technologies not only influence user behavior by addressing users on a cognitive level, but also by materially

Analyzing the relations between technologies and user behavior

389

‘demanding’ specific forms of interaction. The ‘script’ concept from science and technology studies indicates this ability of technologies. Like the script of a theater play or a movie, technologies are able to ‘prescribe’ to their users how to act, like a speed bump that contains the script ‘slow down when you approach me.’ Human behavior is not mediated by providing users with information here, but by materially shaping the physical interaction with an artifact. In a very similar way to the script concept, in Part 1 the concepts of ‘affordances’ and ‘constraints’ are used: technologies make possible specific forms of behavior and impede others. Note that the concept of ‘affordance’ is used in a somewhat different sense here than above, where it had to do with the interaction between technologies and users. On the user side of the spectrum of technology-behavior interactions, there are important influencing variables as well. A first variable is obvious: when the influence of technology is informational in nature, the specific forms and dynamics of human information processing, as studied by cognitive psychology, play an important role in the interaction: perception, cognition, evaluation, memory, and learning. A second variable is not very surprising either: the attitudes, goals, intentions and habits of users that influence their behavior when interacting with a technological device. -

-

-

There are less obvious variables too, however, such as: the pre-existing mental models of users, and their schemata and behavioral scripts; technologies never influence the behavior of users uniquely, because existing patterns of doing things inevitably play a role as well; the degree of adoption of technological products, which determines whether or not they are used in the first place, and therefore whether the technology-behavior interaction will occur at all; the ‘interpretative flexibility’ of technologies, which is indicated by the concepts of multi-stability and user logic. Users will interpret technologies before using them, and this interpretation or use logic can be totally different from the intentions of the designer. Robust car doors and safety belts are intended to enhance safety, but users can interpret this safety of their car as an invitation to drive faster. Technologies can never be said to determine human behavior, therefore, but only to help shape it.

The social environment •

In order to analyze the role of the socio-technical environment in technology-behavior interactions, this book made clear that a distinction

390

•

•

Chapter 36 should be made between two kinds of behavior: behavior regarding the use of technologies and regarding their acquisition. In both kinds of user behavior, the socio-technical environment plays a role. With respect to the role of the social environment in the acquisition of technologies, diffusion theory offers many relevant insights into the dynamics of technology adoption and diffusion. The diffusion and adoption of technologies are phenomena that do not only regard the individual user, since social structures, codes and norms play an important role here, next to individual preferences, goals and desires. The social context of the interactions between technologies and user behavior can be analyzed in terms of ‘social practices’, a concept that indicates ‘ways of doing things’ that are shared in society, like patterns of eating and cooking, or leisure, like holidays. Such practices transcend the level of individual choices: they provide a context for it, and as such help to shape the kinds of interactions between technologies and users. A third aspect of the social context is formed by the codes and norms that exist at the level of social groups and of society in general. These codes and norms also influence the adoption and diffusion of technologies and the social practices, as already mentioned.

The technological environment At several points in this book, attention turned to the role of the technological environment in the interactions between technologies and user behavior. A first and obvious factor here is technological infrastructure. The nature and availability of supporting infrastructure for technologies codetermines if and how these technologies will be used. It is more likely that people will make less use of the car if it possible to reach the same destination in about the same time by public transport. Consumers will probably be more willing to separate organic waste from other waste if both are collected separately ʊ and their willingness seems to have increased since the introduction of a biodegradable paper bag that can be used in the organic litter bin, which makes it less distasteful to empty and clean it, especially in summertime. Technologies often need other technologies to support them, and these supporting technologies can play an important role in the interaction between users and the technology in question. A second factor in the technological environment is formed by the marketing activities around technologies. Just as one finds with technological infrastructure, marketing activities help to determine if and how a technology is used. But whereas infrastructure creates specific conditions for technologies to function, marketing interferes in the user side of the interaction: the attitudes, beliefs and goals of users.

Analyzing the relations between technologies and user behavior

391

Performance

Technology

User Interaction:

Material: • artifactual scripts Informational: • indicators • feedback

• • • • • • •

affordance mediation configuring scenario’s compatibility personenvironment fit

• • •

Technological Environment • • •

infrastructure marketing system of provision

information processing (perception, cognition, etc.) attitudes, goals, intentions, habits mental models, schemata, behavioral scripts multi-stability, use logic

Social Environment • • •

dynamics of diffusion and adoption social practices codes and norms

Figure 35-2. A conceptual framework for technology-behavior interactions

In order to understand the role of technologies in their interactions with users, therefore, technology needs to be analyzed in the context of its environment, of which infrastructure and marketing are two important instances. Together, technologies and their environment are an important part of the ‘system of provision’, which makes available the resources that organize social practices. Figure 35-2 depicts the aspects of the interaction between technology and user behavior that were discussed in this section. Again, not all concepts and approaches that are elaborated in the first two parts of the book can be mentioned here; these parts are much richer than what we can present in just a few pages. We present what we consider to be the most helpful concepts to create a common ground in interdisciplinary research of the interactions between technology and human behavior.

392

3.

Chapter 36

IMPLICATIONS FOR TECHNOLOGY DESIGN AND POLICY-MAKING

Besides creating ‘common ground’ for research of technology and behavior, this book also aims to investigate how insights into the relations between technology and behavior can be made fruitful for the design of new technologies and for policy making. The central thesis developed in several chapters is that technology design and policy making should direct themselves at the interactions between technology, behavior, and the social and technological environment, rather than focusing on these aspects separately.

Designing technology-behavior interactions The integrated ‘socio-technical’ approach of technology and behavior in this book implies that the designing process cannot simply be conceptualized in terms of shaping a technological product. Design should be approached with regard to the technology-behavior interactions it helps to shape. In order to widen the boundaries of the scope of the design process, this book offers many concepts to understand design in terms of shaping the interactions between the technology-in-design and its users, within their social and technological environments, as Figure 35-3 illustrates. Several theoretical frameworks can help to further elaborate this broader perspective of design. First, technology design can be understood as the design of user plans, rather than of technological artifacts only. The activities of designers should not only be understood in technological terms, but in social terms as well: designers anticipate prospective users, and do not only design a product, but also user behavior. This process of anticipation, and the ways in which anticipations guide the development and design of technologies, is a second conceptual line that is important for understanding design as inherently connected to both technological artifacts and user behavior. The theory of domestication offers a third conceptual line where the development of products is intrinsically connected with their future use context. Domestication theory draws a distinction between the configuration of users in the design stage, on the one hand (designers anticipate the influence of technologies on user behavior), and the appropriation of a technology by users in the stage of acceptance and use, on the other (the ways in which users incorporate specific technologies in their daily lives helps to determine how their behavior will be affected).

Analyzing the relations between technologies and user behavior

393

DESIGN

technology

technological environment

user behavior

social environment

Figure 35-3. Design implications

Besides these conceptual lines, several tools and methods exist that can be used to integrate anticipation of user behavior in the process of designing new technologies. A first and obvious instrument that is always at the disposal of designers is their own imagination. When designing a product, the imagination of designers allows them to anticipate future forms of use (e.g. the Eternally Yours project discussed in this volume, which employs human imagination to design products from the perspective of the practices that might come about around them). There are also more systematic ways of connecting the use phase of technological products to their design phase. A well-elaborated method in this context is the scenario method, which offers a structured way of anticipating possible future interactions between users and technologies. Two variants of the scenario method appear to be useful. A first variant focuses on use scenarios of individual users, in which all physical actions involved in the use of the technology-in-design are taken into account. A second variant, called design-orienting scenarios, takes a systemic point of view, in which function fulfillment in specific aspects of daily life has a central position, like clothing, care, nutrition, and shelter. Whereas use scenarios focus on the physical actions of individual users, design-orienting scenarios focus on the place of technologies in the daily lives of users, and on the relations between new technologies and their socio-technical environment. A second systematic method for integrating the anticipation of user behavior in the technology design is the ‘script method’, in which the ‘scripts’ of the technology-in-design are anticipated and explicitly built into the artifact. Thirdly, actually involving users during the

394

Chapter 36

design process can be a valuable instrument for making a connection with the use stage of technologies during the designing stage.

Implications for policy making Traditionally, environmental policies were aimed at developing and implementing clean technologies. In the 1980s and 1990s, attention shifted to influencing behavior. Public information campaigns were developed in order to change consumer attitudes and behavior. This two-track approach, focusing on technologies and user behavior separately, rather than on the interactions between them, appears to be too limited. No technology nor user behavior as such should be the points of application for policy making, but the interaction between them, and the specific socio-technical context in which it takes place should provide this locus. This implies that for adequate policy-making, technology should be approached in terms of its influence on user behavior, and behavior in terms of its relation to technology, taking into account that this interaction always takes place in a socio-technical context.

Points of application The points of application for policy-making, therefore, are not only the specifications of technologies and the cognitions and attitudes behind human behavior, but also comprise the ways in which technologies help to shape user behavior. Three points of application become visible in this book. Firstly, the interaction between technology and user behavior: technologies influence human actions, and human interpretations influence the embedding and functioning of technologies. Secondly, the social context in which the interaction between technology and user behavior takes place, which can be made visible (and changed) by making stakeholder analyses and interacting closely with the main stakeholders involved. A third point of application is the technological environment of the interaction between technologies and users, like the infrastructure needed for the function (and also codetermination) of the technology. Figure 35-4 illustrates these points of application for policy making, analogously to Figure 35-3, which indicated the implications of the integrated perspective of technology and user behavior for technology design.

Analyzing the relations between technologies and user behavior

395

POLICYMAKING

technology

technological environment

user behavior

social environment

Figure 35-4. Points of application for policy-making

To illustrate this figure with a non-environmental example: in order to stimulate road safety, the focus should not only be on designing more robust cars or changing the attitudes of road users. Cars could also be designed in such a way that they help to stimulate safe driving behavior, for instance by giving feedback when drivers exceed the speed limit, or even enforcing a specific maximum speed. Moreover, by approaching the interaction between road users and their cars in its social and technological context, it becomes visible that road infrastructure is another point of application for policy making. Redesigning roads in such a way that they are less inviting to exceed the speed limit at unsafe places, for instance by building speed bumps and by physically or optically narrowing the road, could increase road safety. Measures like these can be a valuable supplement to attempts to influence behavior without taking its connection to technology and its sociotechnical environment into account, like information campaigns or financial stimuli (fines).

The policy-making process The integrated perspective of technology and user behavior not only reveals new points of application for policy-making, but also has implications for the very layout of the policy-making process. Because of the complexity of the interactions between technologies, users, and their

396

Chapter 36

socio-technical context, policy-making cannot be conceived as a linear, topdown steering process, in which predefined goals are to be reached, effects are seen as predictable, which evaluation processes take place after completion of the trajectory in order to assess the adequacy of the instruments used. From a socio-technical point of view, policy-making should rather follow the dynamics of the system, which makes the development of the process whimsical rather than linear. Policy-making is more ‘goal seeking’ than ‘goal defining’ then, and should follow a step-bystep approach with short feedback loops to enable adjustment of the process and short-term goals. Because of the manifold and complex interactions between technologies, users, and the social and technological contexts in which these find themselves, the belief in the predictability of outcomes, which has a central place in the ‘traditional’ approach, should be abandoned. From a sociotechnical point of view, phenomena like rebound effects do not come as a surprise but are inherent parts of the process. Therefore, it is important to approach policy-making as a learning process, with explicit room for experimentation. Moreover, this implies that policy-making cannot take place from one central point, but should take place in close interaction with stakeholders from all domains involved: the domains of technology design, technology use, and the socio-technical context. Steering capacity is generated not by isolated policy-makers, but in the network of actors that surrounds each technological innovation, with distributed roles, responsibilities and actions to develop or implement technologies. In order to streamline such interactive learning processes, it is important to identify the specific roles of the actors in the network involved. ‘Pioneers’, ‘public venture capitalists’, ‘superintendents’, ‘mediators’, ‘stewards of the common good’, and ‘citizen-consumers’ ʊ to mention some roles that were identified in this book ʊ all have their specific places in the network around the technology-behavior interaction at which the policy-making process should direct itself. In Table 35-1, the ‘traditional’ and ‘socio-technical’ approaches to policy-making are characterized. Policy-making and knowledge production This socio-technical approach to policy-making also has implications for the knowledge production to support the policy process. First of all, not only should the policy-making process seek interaction with the stakeholders involved, but the research process should as well, by involving stakeholders in the research process. New methods for doing this, and for dealing with viewpoints and values of stakeholders in the research process, for instance by joint fact-finding, should be further developed to support this. Secondly, research can help to structure and accelerate the learning process that is

Analyzing the relations between technologies and user behavior

397

needed in the socio-technical approach, by ‘learning evaluations’ that show results during (not after) the process, by structuring reflection and development of specific process formats, and, last but not least, by experimentation in ‘real life’. The ‘living lab’, as explained in this book, is a good example of this. Thirdly, the close interaction between research and practice that will emerge from these two points, will evoke the need for new research methodologies ʊ like a further development of action research methodologies ʊ to deal with this interaction in a fruitful way. All these aspects are closely related to the characteristics of the concept of “mode 2 science” of Gibbons et al. (1994): because of the socio-technical approach which appears to be necessary for analyzing and influencing interactions between technology and behavior, trans-disciplinarity and a close connection between theory and practice are the basic conditions for this kind of research. Table 35-1.Approaches to policy-making ‘Traditional’ Approach

‘Socio-technical’ Approach

Focus:

Technology

Process:

Leading paradigm:

Linear Long evaluation cycles Defined Goal Steering

Interactions between technology and society Whimsical Short feedback loops Goal-seeking Learning

Attitude towards uncertainty:

Reduction of uncertainty, predictable outcomes

Role of actors:

Central role of government

Acknowledgement of uncertainty, acceptance of unpredictable outcomes Distributed roles, actions, and responsibilities

Legitimacy and ethical issues A fourth aspect of policy making for technology-behavior interactions concerns the ethical issues and the legitimacy of attempts to influence human behavior with the help of technology. Behavior-steering technology can be seen as a threat to human freedom, human responsibility, and even to democracy. If technological artifacts rather than human intentions determine the actions and decisions of human beings, human freedom and responsibility seem to evaporate, and not people but technologies seem to control society. Yet, many analyses in this book have made clear that the influence of technology on human behavior is inevitable, whether we like it or not. Since virtually all aspects of our daily lives have become connected with technological artifacts, technology plays a major role in the decisions we

398

Chapter 36

make and the ways in which we act. This state of affairs should not invite us to refrain from attempts to develop technologies from the perspective of their influence on human behavior, but rather to try and do this in a morally responsible way. The concept of ‘disclosive technology,’ as developed in one of the chapters of this volume, might be helpful for doing this: technologies should not be approached as forces that completely determine human behavior, but as forces that disclose specific practices. Any form of policymaking that takes the intricate connections between technology and human behavior seriously should be guided by the question regarding what practices are worth disclosing and what practices are not.

4.

CONCLUSION

We started this book from a fascination for the interaction between technology and behavior, and the conviction that a vast variety of disciplines could contribute to a deeper understanding, description and theory-making about that intersection. Thus far, our expedition through this field has resulted in a preliminary description of a multidisciplinary conceptual framework, and in first ideas about applying this framework in the context of technology design and policy making. Both lines of research deserve to be explored further. As for the theoretical ambitions of this book: we hope that the conceptual framework for analyzing technology-behavior interactions will be elaborated further, and that the framework we have proposed will provide enough ‘common ground’ for that. Using this framework for empirical research on the interactions between artifacts and user behavior, or for design and policy making, will hopefully enable refining and expanding the vocabulary developed in this book. We especially hope that new research methodologies, design methodologies, and policy-making strategies will arise that take these inextricable connections between technologies and their social context as a starting point. As for this book’s practical ambitions: we hope that experimentation with designing technology-behavior interactions and ‘socio-technical’ policy-making will bring about interesting learning processes that help to further elaborate methodologies and strategies. Learning processes need feedback from earlier experiences, so we hope for a community of ‘technology-and-behavior’ researchers to come about, with shared ideas and experiences; where theories are developed, challenged and improved; and whose experiments

Analyzing the relations between technologies and user behavior

399

are undertaken applying the theories and concepts developed in the ‘real-life settings’ of technological innovation and policy-making. These concluding remarks end our first expedition through the world of technology-behavior interactions. We have charted a preliminary map of it, and made an inventory of the instruments available to navigate through it and to build new things within it. We hope more expeditions will follow, resulting in an ever-expanding body of knowledge about both the behavior of our technologies and the technologies of our behavior.

REFERENCE Gibbons, M., C. Limoges, H. Nowotny, S. Schwartzman, P. Scott and M. Trow (1994). The New Production of Knowledge: The Dynamics of Science and Research in Contemporary Societies. Londen: Sage Publications.

LIST OF AUTHORS

Albert G. Arnold is senior researcher and consultant at the Foundation ‘Centre of Expertise consultants for ICT and public administration’. Arnold is a cognitive psychologist and finished his Phd thesis at Delft University of Technology in 1998. He is interested in the subject of ‘e-government’ with a focus on the relation between technology and the human factor. Theo A.M. Beckers recently retired as director of Telos and as professor of Leisure Studies at Tilburg University. He now holds a special chair at Tilburg University in the field of sustainable rural development. His main research interests are in the field of policy studies for sustainable development, from the local up to the global level. Among the central research themes are the development of sustainability indicators, spatial planning, leisure, and sustainable consumption. Prof. Beckers serves as member of many advisory boards for the Dutch Government, including the ‘Raad voor het Landelijk Gebied’ and 'het Brabants Landschap’. Adrienne van den Bogaard is assistant professor at the Department of Philosophy, Delft University of Technology. Her main research interests are the history and social studies of technology. She currently leads a project on the history of information technology in the Netherlands in the 20th century. She is also involved in research on the use of mathematical expertise (quantification) in the context of public policy making. Philip Brey is associate professor and vice chair of the department of philosophy, University of Twente, The Netherlands. His research is centered on the philosophy of information and communication technology (ICT), with special attention to computer ethics and to the social, cultural and epistemological roles of information and communication technology. He is a member of the executive board of the Society for Philosophy and

401

402

List of authors

Technology and co-editor, with Thomas Misa and Andrew Feenberg, of Modernity and Technology (MIT Press, 2003). Han Brezet has a PhD on the subject of energy-innovation and works as tenure professor at the department Industrial Design Engineering of the Delft University of Technology since 1992. He is responsible for the Design for Sustainability program. This is an international program about the development of sustainable products, services and systems, including the application of new producttechnologies. Since 2001, he is also part time professor at the International Institute for Industrial Environmental Economics at Lund University in Sweden. Carja A.A Butijn is researcher at Wageningen University, Consumer Technology and Product Use. Her focus is on the handling of goods and services for task fulfillment in the household in developed and developing countries. The main topics of her research are about food safety and health, and food security of households. Erica Derijcke is researcher and advisor at IVAM Research and Consultancy on Sustainability. She works on projects on Energy and Quality of Living. Her main interest is consumers and their opinions and behaviour in the areas of renewable energy, energy saving, recreation and sustainable living. Boelie Elzen is senior researcher at the Centre for Science, Technology and Society of the University of Twente, the Netherlands. His general research interest is in understanding the dynamics of socio-technical change and in using these insights to develop suggestions on how to tackle the societal problems related to these change processes. One of his areas of interest is passenger mobility on which he has worked in various EU research projects and has consulted to various institutions in the field in the Netherlands as well as abroad.

Mirjam Fransen is PhD-student at the Department of Public Health at Erasmus MC Rotterdam. Her research topic is prenatal screening and ethnic differences. Before she was junior researcher at the Department of Health Education and Promotion of Maastricht University Valerie Frissen (1960) studied Communication Studies at the University of Nijmegen, where she also obtained her PhD-degree (1992). From 1991 to 1999, she worked as associate professor at the University of Amsterdam (Amsterdam School of Communications Research). In 1999 she started

List of authors

403

working for TNO, where she currently is Head of the Department ICT & Policy at TNO ICT. She specializes in research on the social acceptance and socio-economic implications of ICT developments and has published widely on these subjects. Since 2003 she holds a chair on ICT and Social Change at Erasmus University Rotterdam, faculty of Philosophy. She is contributing editor of the academic journals New Media and Society and Sociology. Ans (J.P.) Groot-Marcus graduated in Household and Consumer Sciences and was senior researcher at Wageningen University, Consumer Technology and Product Use. Her main research interests are household tasks and behavior, energy requirements of consumption and sustainable living conditions. Wim Hafkamp is professor of environmental economy at Erasmus University in Rotterdam. He is on the Staff of the Erasmus Centre for Sustainability and Management, and active in its International Off-Campus Ph.D. Program on Cleaner Production, Cleaner Products and Sustainable Development. His research and policy work are focused on innovation processes within and between organisations, with applications in industrial production and consumption as well as mobility. Wim J.M. Heijs is an associate professor in the department of Architecture, Building and Planning at the Eindhoven University of Technology (unit of Urban Management and Design Systems). His background lies in environmental and social psychology and his main interests are in research and development of theory concerning userenvironment interaction processes and user needs analysis with an emphasis on dwellings, residential environments and special groups within those settings (e.g. elderly). Laurie Hendrickx studied experimental psychology at the University of Amsterdam. He obtained his PhD at the University of Groningen, on research about the cognitive processes underlying risk judgment and risky decision making. Currently, he is assistant professor at the Center for Energy and Environmental Studies (IVEM) of the University of Groningen. His research interests include environmental risk perception and (determinants of) environmentally relevant behaviors. Wybo Houkes is assistant professor at the Department of Philosophy and Ethics of Technology at Eindhoven University of Technology. He participated in the NWO research programme The Dual Nature of Technical Artifacts and published on artefact functions and action-theoretical recon-

404

List of authors

structions of use and design. His current interests include the nature of technological knowledge, the ontology of artefacts, and social epistemology. Jaap Jelsma is senior researcher at the Centre of Studies of Science, Technology and Society (CSWTS) at the University of Twente, Enschede, The Netherlands. From 1999-2005 he has also been affiliated as a researcher with the Energy research Center of The Netherlands (ECN) at Petten, The Netherlands. His main research interest is on the study of use aspects of technical design, and on translating the results into tools and policy for a better environment. Nicole van Kesteren is a doctoral student at the Department of Psychology, Maastricht University, The Netherlands. Her main research interests are in the area of explaining and changing pro-environmental and health-related behavior. Remke Klapwijk is researcher at the Consultative Committee of Sectors Councils for research and development, The Hague. Her main research interests are innovation, sustainability, education and future-oriented problem solving. She published various books and articles on life cycle assessment and on education. She is currently involved in a horizon scan on future threats and opportunities for the Dutch sector councils. At the time of writing, Marjolijn Knot was a researcher at the Faculty of Design, Construction and Production (Design for Sustainability group), Delft University of Technology. Her main interest was on user-aspects of sustainable product-service systems. David Laws is a research scientist and Co-Director of the Environmental Technology and Public Policy Program at Massachusetts Institute of Technology. His research focuses on governance, dispute resolution, and deliberative democracy. Harro van Lente studied physics and philosophy at the University of Twente, The Netherlands. He is assistant professor in innovation studies at the Copernicus Institute for Sustainable Development and Innovation at the University of Utrecht. He develops and gives courses on technology dynamics, national systems of innovation and innovation policy. His research currently focuses on emerging properties of nanotechnology. Marc van Lieshout is senior researcher at TNO. His main expertise is in the field of foresight studies and strategic analysis of techno-scientific developments. Before working at TNO he has worked at the University of

List of authors

405

Nijmegen and the University of Maastricht, in which position he has been engaged with research projects dealing with social learning in ICT. During his work at the Dutch Rathenau Institute he has been involved in constructive technology assessment in the field of elderly people and ICT. He has a background in physics. Helma (W.J.) Luiten is researcher/advisor Sustainable System Innovation at TNO. The user behavior and social aspects connected to these innovations are of main interest to her. Her focus lays at generating ideas for new system innovations that apply to the needs of the user. Susan Martens works as a researcher at the Environmental Policy Group of Wageningen University, the Netherlands. Her main research interests are sustainable consumption and lifestyles, NGOs, public participation in environmental management and civil society development in Eastern and Western societies. L.T. (Teddy) McCalley is a senior research psychologist in humantechnology interaction at the Technical University Eindhoven. Her main research interest is on human motivation underlying energy conservation. She is also a lecturer on motivation through technology for the Department of Industrial Design and on consumer issues in energy conservation for the Department of Technology and Sustainable Development, Technical University Eindhoven. Ree M. Meertens, Ph.D. is associate professor at the Department of Health Education and Promotion of Maastricht University, The Netherlands. Her research interests include risk perception and communication and promotion of energy efficient behaviors. She is a member of the Dutch Health Council. Petra Mettau is researcher and consultant at the foundation ‘Centre of Expertise consultants for ICT and public administration’. She studied public administration and law. Her main interest is ICT as enabling technology for customer-oriented public services. Cees J.H. Midden is professor of human-technology interaction at the department of Technology Management at Eindhoven University of Technology. His research focus is on the interaction between users and technological products and systems and implications for the development of new products and systems, for societal and market introductions and for the consumption and use of products. He published various books and articles on environmental consumer behaviour, on the perception and communi-

406

List of authors

cation of technological risks and the social diffusion of innovations. Current topics of interest are social trust and trust in systems and emotional components in judgment and attitude formation. Henk Muis is a physicist by origin. He runs P/m, a consultancy firm on environmentally responsible product development and is an active politician for the Green Left party. He is co-founder and board member of Eternally Yours, a foundation that aims to extend the cultural lifespan of products. Marcus Popkema is teaching traffic engineering and road safety at Windesheim College, Zwolle, the Netherlands. He was trained in Science & Technology Studies. One of his main interests is in improving safe behaviour in traffic, either by traffic enforcement or road engineering. Jaco Quist is an assistant professor in technology assessment and sustainable innovation at the faculty of Technology, Policy, and Management, Delft University of Technology. His research focuses on (system) innovations towards sustainability using stakeholder and user involvement, scenarios and backcasting. In the SusHouse project he did the Dutch nutrition case study and he was involved in developing and evaluating the stakeholder workshop methodology. His teaching includes technology policy, technology and society, sustainable innovation and sustainable entrepreneurship. Ingrid van Schagen is a research psychologist and works as a senior researcher at SWOV Institute for Road Safety Research in the Netherlands. She has been involved in many national and international projects on a variety of road safety topics. One of her fields of interest is the interaction between road user behavior and road design. A.J.M. (Ton) Schoot Uiterkamp has a PhD in biophysical chemistry from the University of Groningen. He was a staff member of UNESCO in Cairo, Egypt. He has held research positions at Yale and Harvard universities. He was staff member and department head at the Environmental Division of the Netherlands Organization for Applied Scientific Research (TNO). Since 1991 he has been professor of environmental sciences at the Center for Energy and Environmental Studies of the University of Groningen. Adriaan Slob is senior researcher within TNO Built Environment and Geosciences . His main research interest is on new (interactive) policy processes needed for implementation of innovations to reach sustainability. His studies concern the role of knowledge and learning in policy processes

List of authors

407

and the development of new process designs with respect to that. He is also initiator of the Pytheas network, which involves cooperation between TNO, Erasmus University Rotterdam, University of Utrecht and the Massachusetts Institute of Technology on new policy processes. Gert Spaargaren is senior researcher in the Environmental Policy Group (ENP) at the Department of Social Sciences, Wageningen University, the Netherlands and holds a chair on “Environmental Policy for sustainable lifestyles and consumption patterns”. His research interests focus on environmental sociology, consumption, citizenship and globalization. Bea (L.P.A.) Steenbekkers is senior researcher at Wageningen University, Consumer Technology and product use. Her main research interests are ergonomics and product use and the relation with product design. Another field of interest is exposure of consumers to hazardous substances due to activities in the home. Sytse Strijbos is associated with the Philosophy Faculty at the Vrije Universiteit, Amsterdam and also with the School of Philosophy at North West University, Potchefstroom (South Africa). His publications cover philosophical issues of the systems sciences, and ethics and philosophy of technology. Forthcoming is In Search of an Integrative Vision of Technology, a book published by Springer in the series “Contemporary Systems Thinking”. Paul Swuste is associate professor at the Department of Safety Science, Delft University of Technology. He is an expert on Occupational Hygiene; he is a registered occupational hygienist and board member of the International Occupational Hygiene Association. He is also engaged in research concerning exposure and control measures to occupational hazards, occupational accidents, including its legal aspects in various branches of industry. Paul M.J. Terpstra is professor of Consumer Technology and Product Use at Wageningen University. His main research interest is on technology development in relation to function fulfilment, public health and sustainable technology and behaviour. Particular specialties in this area are function appraisal and quantitative microbial risk assessment. Jan Uitzinger is manager of the units Energy and Quality of Life at IVAM (University of Amsterdam). His main research interest is to monitor and evaluate projects on renewable energy, energy saving, recreation and sustainable living.

408

List of authors

Peter-Paul Verbeek is associate professor at the Department of Philosophy, University of Twente (The Netherlands). His work focuses on human– technology relations and the roles of technology in contemporary culture. He recently published “What Things Do: Philosophical Reflections on Technology, Agency, and Design” (Penn State University Press, 2005), in which he analyzes the mediating role of technological products in the actions and experiences of human beings, applying his analysis to the domain of industrial design. Philip J. Vergragt is currently a Visiting Scholar at MIT, Center for Technology, Policy and Industrial Development, a Visiting Senior Fellow at Tellus Institute, and a Visiting Professorial Fellow at the Manchester Business School, University of Manchester, UK. He is Emeritus Professor of Technology at TU Delft, the Netherlands. His present research interests are Visioning and Back-casting, Social Learning through Bounded SocioTechnical Experiments (BSTEs), Institutional change, Transitions to Sustainable Cities, Transitions towards the Hydrogen Economy, Sustainable Transportation, and Sustainable Consumption. Dr. Vergragt holds a Ph.D. in Chemistry from Leiden University in the Netherlands (1976), taught Chemistry and Society at Groningen University, and was a Deputy Director of the Dutch Government’s Program on Sustainable Technology Development in the 90-ies. He is a founding member of the Advisory Board of the Greening of Industry Network Pieter E. Vermaas is a postdoc researcher at the Philosophy Department of Delft University of Technology. He participated in the NWO research programme The Dual Nature of Technical Artifacts and published on function theory and design methodology. His current interests concern the notion of functional decompositions of artefacts and the use of quantum mechanics in engineering. Bas van Vliet is researcher and lecturer at the Environmental Policy Group of Wageningen University. His main research interests encompass the transformation of urban infrastructures, sustainable technology development and domestic consumption. His PhD thesis “Greening the Grid” (2002) deals with the ecological modernization of water and electricity systems in the Netherlands. He is co-author of “Infrastructures of Consumption. Environmental Innovations in the Utility Industries” (Earthscan, 2005) and co-editor of “Sustainable Consumption, the Implications of Changing Infrastructures of Provision” (Edward Elgar, 2004).

List of authors

409

Trijntje Völlink is senior lecturer at the department of Psychology at the Open University, The Netherlands. Her research interests include goal setting and feedback and promotion of energy efficient behaviors. Sienke (G.W.) Wolters is Manager Customer Services & Market Analysis at Woningbouwvereniging EMM, a public housing association in Zandvoort. She is responsible for a team of 12 employees, which rents the houses, deals with social problems and solves small maintenance problems. The paper is based on research which she did as MSc-thesis Household and Consumer Sciences at Wageningen University.