Memory Is More Important
Than Actuality
Donald A. Norman

Nielsen Norman Group and Northwestern University | [e-mail]

“Your discussion regarding … the fact that memory of an event
is more important than the experience made me remember my trip
to Thailand a few years ago...
• I traveled for three weeks and lost 10 pounds
because I didn’t like any food.
• There were insects on steroids everywhere I turned
• And the restrooms were no joy...
However, I had the time of my life and I would go back in a second.”

(email from Tammy Guy, Nov. 10, 2008. Reproduced with permission.)

[ 1] Trope, Y. and N. Liberman. “Temporal Construal.” Psychological Review 110, no. 3 (2003): 403–421.

As interaction designers, we strive to eliminate confusion, difficulty, and above all, bad experiences. But you know what? Life is filled with bad experiences. Not only do we survive them, but in our remembrance of events, we also often minimize the bad and amplify the good. Consider this email from Tammy Guy, an audience member who heard me give a talk about the triumph of memory over actuality. Her email included photographs of fried insect treats, a huge spider, and an unseemly looking squat toilet. Would she go back to Thailand? She would. If the total experience were good enough, I’ll bet many of you would, too.

What is it about our experiences that lead us to repeat them— and recommend them—despite the bad parts? It turns out there are good psychological reasons. Let me call it “the distancing

effect.” We remember events differently when we achieve distance from them, whether the distance is time or space. We anticipate and evaluate the future, remember and reflect upon the past. Both are at a distance in time from the event itself. In anticipating events, we review the past in order to make choices for the future. In remembering events, some things fade from the mind faster than others. Details fall away faster than higher-level constructs. Emotions fade faster than cognitions. In psychology these phenomena have been studied under several rubrics, including “temporal construal theory” and “rosy remembrance.” There is considerable psychological evidence to support the notion that positive and negative events fade at different rates from memory, and that affective elements fade differently than cognitive ones (or in my

terminology, reflective memories fade most slowly) [ 1].

The aforementioned email is but one example of many. The implication for design is clear. We should not be devoting all of our time to providing a perfect experience. Why not? Well, perfection is seldom possible. More important, perfection is seldom worth the effort. So what if people have some problems with an application, a website, a product, or a service? What matters is the total experience. Furthermore, the actual experience is not as important as the way in which it is remembered.

The argument starts with a simple thought experiment. Suppose in some task, using a product or getting a service from a company, you had some perfectly horrid experience along with some positive ones. Now, just suppose you had no memory of the horrid experience. Would you go back and repeat the experience? Most people would repeat something they remembered as enjoyable. Of course, the premise is suspicious: If the experience were truly that horrible, I would maintain a memory of the negative parts. Yes, but memories for bad experiences dissipate differently than those for good ones. The negative emotions associated with the bad parts fade away more quickly than the cognitive evaluation does. So although I