Donald A. Norman
Basic Books, 2003 ISBN 0465051359 $26.00
As a person involved in the work of designcreating things for others to use and enjoyI look to the author of The Psychology of Everyday Things for a framework that will help me to predict the judgments of others about my design workor of the designs I choose for them. I approached Don Norman's new book, Emotional Design, with this need in mind, but I did not find what I sought and I doubt anyone will. However, I did find the personal reflections of a design professional attempting to enlarge his own understanding of how design works. Taken at this level, this book is valuable and stimulating.
When Norman discusses design for use he is always worth listening to, and the section of this book on personalization and customization showcases him at his best. I found his organization of thoughts, ideas, and reactions to a well-stated and thoroughly examined problem stimulating and useful. But if you are looking for a set of tools, you may want to look elsewhere.
In this book, Norman sets out to help us understand why we love (or hate) everyday things. This requires that he explain the interactions between people and artifacts that evoke emotional reactions. To make this analysis useful to designers, he separates human responses to artifacts into three levels:
- Visceral (instinctive, unthinking, primitivethe fear of falling, the love of children)
- Behavioral (the realm of usabilitythe pleasure of doing something well or the frustration of doing it poorly)
- Reflective (the enhancement of responses due to study, memories and experienceart, architecture, literature, and design)
By working through many examples, we learn how designs may evoke emotions, a useful perspective for designers. Norman shows us how artifacts evoke emotional responses by making those perceiving the artifacts aware of the object's significance. At the visceral level, the awareness of the artifact is instinctive (eg: hunger, fear of snakes) and the responses are generally, if not universally, similar because the instincts are widely shared. At the reflective level, the artifacts relate to individual experiences and memories. Since this differs widely from person to person, emotional responses also differ. The task of emotional design might begin with designers knowing what emotion they'd like their artifacts to evoke from their audience. After all, the way in which one perceives an artifact depends on the particular person, not in the design of the object.
So far, so good. What I would expect to see next is a process for choosing the visceral, behavioral, and reflective content which make up the response which a designer wishes to evoke, and step by step, how to frame a design which evokes this content. Norman offers one solid case study, the National Football League Headset, which actually identifies the responses (content) the design is to evoke, describes how the designers went about trying to do that, and measures how well the design reaches its goal. This is great stuff because we can follow and learn from the process. Unfortunately, this type of discussion only happens once in the book.
The classification of emotional response into three levels (visceral, behavioral, and reflective) is the closest that the book comes to offering a design tool. Although the follow-up chapter on "Working with Design" deals with designing for each of the three levels of emotional response separately, the discussion doesn't go into much depth. I would have hoped that the visceral and reflective levels, which are new topics to me, would get more footage than the behavioral level. After all, Norman has already devoted an entire book, which some of us know by heart, to behavioral design. Nonetheless, the behavioral level gets almost five times the coverage of visceral and three times more than reflective. What's more, the material is largely in the form of anecdotes instead of guidelines and in-depth case studies.
Of course, quantity does not necessarily equal quality. Yet the dramatic difference in quantity was a disappointment. Familiar ideas were discussed in loving and useful detail while new ideas (new to me, at least) got short shrift. This book shows us new ideas such as visceral design in the way Moses was shown the Promised Land: We get a distant view, but the story ends before we actually arrive.
Too many of this book's key points rest on examples that are supported only by "reflective" contentmemories and experiences, which Norman assumes all designers share. In addition, the strength of each argument depends on the reader's agreement with the value of these examples, another assumption made by Norman. For rules and guidelines, this is a risky approach. Arguments that succeed when experience and beliefs are shared may falter when they are not.
Norman says "attractive things work better" and offers anecdotes as proof. We are not expected to question the proposition, ask for empirical evidence, or even to search our own thoughts for unattractive things that work "better" than attractive things (I thought of Unix versus Windows). We are expected to go with the flow.
When Norman is discussing the act of chopping vegetables or brewing tea I accept everything he says, simply because I chop vegetables and brew tea. On the other hand, when he talks about servant-master issues such as robots serving him refreshments, I don't have the necessary shared experience to readily accept his arguments. I deal with refreshments differently and servants not at all. Partly because of this difference between us, his conclusions about robots don't resonate with me. He presents no reasoning, logic, or case studies to win me over. I end up rejecting the entire premise of the "robots" chapter about the servant-master issue. Worse, I start looking back at the kitchen scenes and wondering whether Norman has put something over on me. (Well, in a way, he has.)
The author establishes the importance of "positive affect," a positive emotional response to an artifact from a user, to that user's effective and creative use of an artifact. Norman even argues that positive affect is key to survival (and it's not hard to believe that survival can depend on the use of artifacts). The discussion of the design of the affective responses that machines evoke in humans is largely sidestepped (of course, they should be positive). But do the machines themselves need the same equipment (affect/emotion) to deal most effectively with humans? Norman presents and accepts this proposition uncritically.
The only concrete evidence for the proposition I found in the book was a reference in the introduction. Norman has been influenced by the results of experiments conducted by A. R. Damasio with humans who had lost their capability for affect. Damasio found that this loss cost them their ability to make decisions in complex situations. Generalizing from damaged humans to well-designed machines, unfortunately, is a big jump.
The plausible requirement that some machines be able to read, react to, and reflect human emotions in order to deal with humans on a high level (such as the motivational aspects of teaching) is a step I am willing to take. Norman's assumption that the only and easiest way to accomplish this is by equipping artifacts with human emotions and affect remains unsupported and unquestioned. If we grant that possession of affect is a survival trait in humans, must it be one for artifacts?
The literature that Norman relies on (Asimov, Vinge, Spielberg, Shelley, and others) to give weight to his ideas is fictional and speculative. It is largely concerned with designers' failure to give machines useful affect, usually with disastrous or at least tragic results. If designing affect for machines is the direction we ought to be going in, we had better get started since the road to success looks steep and bumpy. However, we first need to be convinced that it is the right track.
When Norman buttresses his ideas with the evocation of feelings and experiences I can relate to, his speculations are cogent and meaningful, and I feel like I am part of the discussion, especially since it is a discussion, not an argument based on evidence examined and compared. The examples of food preparation make a useful bridge between Norman and this reader, but I have a radically different take on Asimov, and I have never read Vinge or seen the movie "Artificial Intelligence." When Norman begins to speculate and relies on tacit agreement in place of evidence, I can't accept his conclusions.
This is no textbook, no design manual. It is highly personal, like sitting down for a friendly chat with Norman about a range of stimulating ideas. And because the book is so stimulating on many issues, I expected great things from the discussion of emotions (or affect) in machines. We are way beyond conventional usability here, when the emotions under discussion reside in the artifacts rather than their human "users." This issue has generated considerable literature but very little organized thought, just the kind of situation in which we look to experienced people like Don Norman for guidance.
This book gives emotional design a solid footing, although it leaves the structure for others to erect. Norman discusses emotion in machines without establishing any forceful conclusions. I would think machine affect too difficult an issue for Norman if he had not also written The Psychology of Everyday Things. I hope he will take another shot.
Computer Organization and Design: The Hardware/Software Interface, 3rd Edition
David Patterson and John Hennessy
Morgan Kaufmann, 2004
ISBN: 1558606041 $64.95
Guide to Applying Human Factors Methods
Pietro C. Cacciabue and Carlo Cacciabue
Springer Verlag, 2004
ISBN: 185233705 $64.95
Jeff Hawkins and Sandra Blakeslee
Times Books, 2004
ISBN: 0805074562 $25.00
Thoughtful Interaction Design: A Design Perspective on Information Technology
Jonas Löwgren and Erik Stolterman
The MIT Press, 2004
ISBN: 0262122715 $45.00
Understanding Users: A Practical Guide to User Requirements Methods, Tools, and Techniques, 1st Edition
Catherine Courage and Kathy Baxter
Morgan Kaufmann, 2004
ISBN: 1558609350 $59.95
Usability and Internationalization of Information Technology (Volume in the Human Factors/Ergonomics Series)
Nuray Aykin, Editor
Lawrence Erlbaum Assoc., 2004
ISBN: 0805844783 $89.95
Where the Action Is: The Foundations of Embodied Interaction
Bradford Books, 2004
ISBN: 0262541785 $20.00
Will Schroeder has designed stage sets and lighting, worked as an architect and mechanical engineer and taught visual design at the Harvard Graduate School of Education. He is currently a Principal Consultant at User Interface Engineering.
To submit a book review, please email Gerard Torenvliet at firstname.lastname@example.org
©2004 ACM 1072-5220/04/0900 $5.00
Permission to make digital or hard copies of all or part of this work for personal or classroom use is granted without fee provided that copies are not made or distributed for profit or commercial advantage and that copies bear this notice and the full citation on the first page. To copy otherwise, to republish, to post on servers or to redistribute to lists, requires prior specific permission and/or a fee.
The Digital Library is published by the Association for Computing Machinery. Copyright © 2004 ACM, Inc.