The "new" HCI stresses the importance of emotions. However, this emphasis might be misleading or at least not as valuable as hoped. Let me explain why.
Emotions are important. They are an essential part of our lives. We cannot function without them. There is a long tradition of accentuating the difference between "cognition and emotion," "the mind and the heart," "reason and passion." In the same vein, reformers of HCI often stress that the "old" HCI is, in essence, cognitive (i.e., focused on memory, tasks etc.) and that the future lies in emotions. But contemporary psychology understands emotion and cognition as integral parts of each other. Most complex emotions require cognitive processes. Consider satisfaction: It is the consequence of comparing an event’s outcome with one’s expectations . This comparison process necessarily involves cognition, such as comparing outcome and expectation. In this sense, emotions need cognition, and research on the role of emotions in decision-making demonstrates that cognition needs emotions, too. People, who understand "emotional design," as putting passion, desire, and seduction above reason are misled. The challenge for "emotional design" is to explore the interplay of cognition and emotion, rather than dismissing cognition entirely.
Often emotional design is understood as an explicit attempt to induce emotions through a particular product, just as marketing claims that products must touch the hearts of the customer. The most fundamental product-related emotions are attraction emotions (e.g., love, hate, liking, disliking). They are momentary and largely dependent on context. Indeed Ortony et al. conclude that "liking can be a quite ephemeral experience" . Can we really design such an "ephemeral" experience? Things loved for one reason in a particular situation, can be hated for the same reason in another. As Wright and McCarthy note in this issue of interactions, it is not possible to design an experience, only to design for an experience. Designers can shape, but they cannot determine. They can create possibilities but they cannot create certainties. The same holds true for emotional experiences. Promising that a certain set of design recommendationsif put into actionwill always result in a particular set of emotions, may be promising more than can be delivered.
What to do? Designing for needs rather than emotions
Positive emotions are important and making them happen is the "new" HCI’s primary objective. But emotions happen in context. They are volatile and ephemeral, and products alone cannot guarantee an emotion. So what to do? How can we make a wide range of positive emotions (joy, satisfaction, pride) more likely? What are "designable" promoters of those emotions?
My view is that people share a set of general needs, which may serve as a starting point, an anchor for design. I typically distinguish between needs for manipulation (goal-achievement), stimulation (personal growth, an increase of knowledge and skills), identification (self-expression, interaction with relevant others) and evocation (self-maintenance, memories). The definition of needs may vary from author to author , but, the message remains: situational fulfillment of needs promotes positive emotions. The more needs are embedded into a product, the wider the possible range of resulting emotions. A usable and useful product (i.e., high on perceived fulfillment of manipulation needs) may lead to satisfaction if a valued goal is achieved in a particular situation and at least a part of the success is attributed to the product. Pride may be the emotional result of using a presentable and classy product (i.e., high on identification) in a social situation. Joy (or "flow") may be the consequence of being excited and challenged by a product (i.e., high on stimulation) in a situation that allows for such a challenge. Whether fulfillment of a need is valued or not depends on the particular usage situation. A product designed for goal-achievement (manipulation), for example, may be appealing in a situation where important behavioral goals (e.g., getting text formatting right shortly before a deadline) are to be achieved. Identification, in turn, may become a source of appeal if the product is used in social situations (e.g. producing a stylish laptop on a train). Stimulation may become a source for appeal in situations where new solutions to problems must be found. The different types of needs are stable, but their relevance and resulting emotions are fluid and fleeting. Figure 1 summarizes my view on the relationship between product, user/owner and situation.
Appeal, attraction, and emotions are inseparable from particular situations. Underlying needs are stable. A particular product can be perceived as usable, that is, good for manipulation, no matter what the situation. Nevertheless, whether usability is valued largely depends on the particularities of the situation. Potential fulfillment of needs promotes appeal and emotions. Accordingly, designers may focus on signaling fulfillment of needs rather than "designing for emotions."
To favor a more unified, integrative approach to human-computer relationships is a worthwhile endeavor. So far, this has been well received by the HCI community. But now, more theoretical, empirical, methodological and practical work is needed to transform the claim into working practice. Models like the one presented above and new approaches such as "inspiration engineering" are only just the beginning. I hope for much more.
1. Gaver, W. W. & Martin, H. (2000). Alternatives. Exploring information appliances through conceptual design proposals. In T.Turner & G. Szwillus (Eds.), Proceedings of the CHI 2000 Conference on Human Factors in Computing (pp. 209-216). New York: ACM, Addison-Wesley.
2. Hassenzahl, M. (2003). The thing and I: Understanding the relationship between user and product. In M. Blythe, C. Overbeeke, A. F. Monk, & P. C. Wright (Eds.), Funology: From usability to enjoyment (pp. 31-42). Dordrecht: Kluwer.
Darmstadt University of Technology, Germany
©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.