Dennis Wixon, August de los Reyes
In the last column, I gave an overview of the visionary company Harmonics and the inspirational game they createdGuitar Hero. This time August de los Reyes and I explore a revolutionary approach to the design of emotionally engaging products and the research tools needed to create them. Interestingly, this approach is rooted in some pioneering insights by one of the founders of psychology and yet is perfectly congruent with an up-to-date cybernetic theory of emotion. It uses one of the most productive frameworks for game design and it is quite counterintuitive, which, in our judgment, only adds to its charm.
Typically, people think of emotions as a cause for behavior: I was afraid, so I ran away. William James, one of the founders of scientific psychology, proposed an alternative theory of emotion: that our bodies act first, and then, from our actions we draw a conclusion. These "conclusions" are emotional states. To illustrate further, the classic example goes like this: We encounter a bear in the woods and, on an instinctual level, our bodies respondadrenalin levels surge, our hearts start pounding, and we start to run. Only after these bodily responses occur do we conclude that we are afraid. It's called the James-Lange theory of emotion. Even though it seems counterintuitive, consider four advantages of adopting this viewpoint.
First, it's consistent with the thinking framework employed by modern game designers. The framework breaks design into three parts: mechanics, dynamics, and aesthetics. Mechanics are the goals and rules of the game. They are what you can put in the box. Along comes the user who brings experience and knowledge to the game, and you have dynamicswhat the user does when they play, i.e., their behavior. Dynamics then give rise to aesthetics: what the user concludes about the game. The game is hard, scary, or exciting. Feelings of power, heroism, and the joy of being evil for a change are all possible aesthetics. Notice the consistency of this model with the James-Lange theory, that behavior leads to judgments or conclusions, among which are emotions.
Second, adopting this approach helps us avoid a classic philosophical error that most psychologists and many people in general often make. That error is the confusion of an explanation with a cause. We explain behavior with hypothetical constructs like anxiety, depression, and the id/ego/superego, and then we point to our made-up constructs as causes. Another way to say this is gravity does not cause anything to fall. Gravity describes a mathematical relationship between object properties (mass) and the relationships (distance) and describes their consequent action. The problem is well expressed in a cartoon by Sydney Harris (previous page), which shows a visitor to the museum of psychology peering into display cases with labels like "Id" and "Superego."
The genius of adopting this approach is that you avoid this philosophical error. At last you can think clearly about psychological explanations.
Third, in this exploration of the Jamesian bodily feedback model, we encountered a radical approach to the concept of emotion in Marvin Minsky's latest book The Emotion Machine. In it Minsky suggests that emotions are "ways of thinking" as opposed to the commonly held belief that they are phenomena that are somehow different from rational thought. The brilliance in Minsky's discourse lies in the notion that emotions, along with other so-called "suitcase words," such as "self" and "consciousness," can be broken down into distinct mental processes. While Minsky challenges what we believe about the mind, we continue to challenge what we believe about the increasingly important role of emotion in the realms of research and design.
One final advantage of this approach urges us, as researchers and designers, to think in the most concrete and specific terms. When I create something how will the user respond and what will they "feel" about that response? As researchers, we can help identify and measure these behaviors and conclusions. As designers, we can develop the affordances to these actions.
As researchers, we can help identify and measure these behaviors and conclusions. As designers, we can develop the affordances to these actions.
Many successful products appear to follow this approachOXO products are good examples. Their products are designed to be touched; the displays for the products encourage touching. The touch is designed to feel good, and the products are designed to feel good as you use them. We use the same approach in Games User Research; we measure behavior (dynamics) and emotional conclusions from that behavior (aesthetics). You probably do this too, although you may not have thought of it in that particular way. Taking this stance can make youwhether you are a designer or a researchermore effective, because at its root, you constantly ask what the user will do and what you can do to alter that behavior and thereby improve that user's experience. It's the driving force behind the RITE method, our instrumented testing, and the ongoing partnership of research with design. Try ityou can be successful too.
Life is short. Have fun.
Microsoft Game Studios
August de los Reyes
Harvard University Graduate School of Design
About The Authors
Dennis Wixon leads a team of over 20 at Microsoft Game Studios which provides consulting and research to make games fun. He is also a member of the User Experience Leadership Team, a corporate steering group. Dennis previously worked at Digital Entertainment Corporation and has been an active member of CHI. He has authored many articles on methodology and co-edited Field Methods Casebook for Software Design (John Wiley & Sons).
August de los Reyes is a student at the Harvard University Graduate School of Design, where his research centers on design and emotion. In his spare time, he is the creative director of a hardware innovation group at Microsoft. This fall, August begins a visiting fellowship at the University of Oxford.
©2007 ACM 1072-5220/07/0900 $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 © 2007 ACM, Inc.