Donald A. Norman’s book, Emotional Design is a great read and one that everyone in design, interaction and technology should read. His chapter on “The Multiple Faces of Emotional Design” discusses the visceral, behavioral and reflective aspects of design. In very simplified terms, visceral design is the appearance or “surface” quality of the design of an object — is it attractive or appealing? In other words, is it something we have an emotional, visceral reaction to? Behavioral design is the way in which the object is used — how effectively we operate or “behave” with the object (also software, etc.). Finally, reflective design refers to how we feel about the object or design — what emotions it evokes using or having used it. In other words, how we reflect on the object.
This analysis is amazingly clear and useful in the question of design. In looking at Apple products, it’s easy to see that one of the reasons that they are so successful is that they tend to address all three qualities — they are attractive (and therefore we have an emotional reaction to them beforehand), they are typically easy to learn and use and therefore the behavioral aspect is positive, and, because of (1) their general ease of use, and (2) cultural/societal “approval” of Apple devices, we typically feel good or “reflect positively” on having used devices.
This is also a really useful tool in analyzing why some designs fail, and also in recognizing that design does not have to successfully address all three aspects. For example, a number of the teapots that Norman refers to in his book have little practical use but are extremely attractive. Their visceral and reflective design is extremely good therefore, while their behavioral design is poor. This does not mean they have failed if the designers goal was *not* to make a “useful” teapot. On the other hand, user interfaces (such as the early incarnations of Android and Windows) had relatively good *initial* visceral appeal — people wanted to play with them. However, once users began to spend time with the devices or interfaces, they quickly found them frustrating. The visceral design was there but the behavioral design was poor. Consequently, the reflective design was poor too. Few people felt particularly good about being foiled an operating system.
This does raise one question I have about Norman’s approach: Although I can see it’s possible to design something that is heavily visceral or behavioral (attractive but not easy to use, not attractive, but easy to use), I wonder if there are any devices or designs, other than his example of momentos, that are heavily reflective. That is, is it possible to design something that is neither attractive or easy to use, but which creates a good emotional reaction having used it? While this may sound glib, in fact, the only thing that immediately comes to mind is something like the UNIX operating system. I don’t think anyone would argue that UNIX is either particularly attractive, nor that it’s particularly easy to use (assuming we’re not talking about an X-Window interface or the like). However, at least personally, I found it *enormously* rewarding to have actually wrested something out of UNIX. This may be pure masochism on my part, but actually having worked in a UNIX command line, while neither viscerally pleasing nor behaviorally pleasing, is at least something I can take some pride in. However, I suspect this was not a design consideration. Heathkits were actually more rewarding than Norman suggests, in that there was a great deal of learning accomplished and the design and layout of the instractions were usually a delight. The last suggestion of reflective design, fashion, sadly, remains a total mystery to me, but then I can’t imagine wearing anythign but the most comfortable shoes social strictures would allow.