User Memory Design: How to Design Ages

Hi, Habr! I present to your attention the translation of the article User Memory Design: How To Design For Experiences That Last

If we understand the difference between experience and memory, we can improve our user experience design (UX) skills.

In a psychological study of 1996, an experiment was conducted - a vivid example of the fact that the human experience of knowing the world at the moment may be very different from our perception of this experience in the future. If we understand the difference between direct experience and memories and understand how they are interrelated, we will better design the UX. The graphs below have changed my thinking about the thought process.

This time I will give some recommendations for creating a UX that will be remembered for a long time . But first I will explain these two graphs.

The intensity of pain experienced by two patients during a painful medical procedure.

The graphs reflect the intensity of pain that two patients experienced during a painful medical procedure. Patients per minute evaluated the intensity of pain on a scale from 0 (“no pain”) to 10 (“highest degree of pain”).

One glance at these two graphs says that patient B spent the time of the experiment much worse: he suffered 3 times longer, and the peak of pain was the same as that of patient A. If you had to choose which patient to turn out to be, you Would choose A. Reasonable choice.

Only for some reason people rarely think rationally.

After the procedure, patients A and B, as well as 150 other participants in the experiment, were asked to estimate the total amount of pain experienced. An analysis of the statistics of the responses received showed that these estimates are in no way related to the duration of the procedure or the total amount of pain experienced (red area on the graph).

On the contrary, the estimates of the total amount of pain obtained at the end of the experiment decreased to the average value of the intensity of pain in two points: the peak of pain and the end of the procedure.

How patients remembered the pain: the average value between peak and final moment.

The main researcher, Nobel Prize winner psychologist Daniel Kahneman, noted that the study confirms the psychological heuristic law of peak-end : human memories of the experience are based on a rough averaging of the most intense moment (peak) and the last moment of activity (end). And the duration of the experience does not affect the memories of it (the concept of neglect of duration ).

Twenty years of additional research in this area have shown that the peak-termination law works in experiments related not only to pain, but also to many other states (for example, with pleasure). Of course, the experience can include both pleasant and unpleasant moments, as reflected in the graph below with positive and negative values ​​along the OY axis. This type of chart is called experience-loop ( experience profile ).

The outlines of experience reflect positive and negative experiences over time.

Streams and frames

From studies similar to those conducted by Kahneman, we realized that the memory of the experiences we received is not related to the sum of good and bad among them. Instead, memory relies on a few key points and ignores the rest. Experience is a stream. Memory - a set of frames.

Psychologists describe the experience with phrases like “a constant stream of“ self-digging ”” or “a stream of passing states that vary from moment to moment . But when we remember something, we do not “wind off the tape” and do not re-experience this stream. Our brain loses the key moments, gradually discarding the noises of instant experiences and giving weight to the peak and final moments.

Two "I"

As Kahneman believes , two “I” live in each person: the experiencing I and the remembering Self . These parts of our consciousness look at the world from different, sometimes even conflicting points of view.

On the one hand, the experiencing I ask: “How do I feel now?”, And every moment learns the world through feelings of pleasure, boredom, frustration and fear. On the other hand, the memorable I wondered: "How am I in general?". It analyzes experiences and experience after an event has occurred, not taking into account its duration and concentrating on bright moments (peak and end).

From this pair, the memory self is in charge , since it determines what we extract from the experience gained. It creates a story about this experience, which we later tell ourselves that will determine our subsequent decisions. In other words, the memorable self of the buyer decides whether he likes your product, whether he will use the service again and what he will tell other people. Experiencing I am only an outsider in these decisions.

Custom memory design

If the memory of the experience is different from the experience itself, and the memorable self makes the final decisions, should we give up the focus on the UX design? Who should we be: user experience or user memory designers?

My answer is simple: we must be both.

Let's be realistic. I work in a web, fast environment where people can close the browser tab at any time. It makes no sense to focus on completion, if people never get to it. In the design we need to focus on reducing dissatisfaction at every moment. That's why things like usability testing are important: you need to design it so that people continue to use the product and get to complete the script.

But you also need to consider how we are going to translate the experience into positive memories that will help make a choice in favor of our product again and tell friends about it. Design for memory and experience .

Do not spoil the ending

The ending is the moment. The user completes the actions in this session and leaves for a while. Think about the places where the design of the ending, and what problems may arise there with the user. If you do everything right, users will have a lot of endings, because users return. Returned - they added the last ending to the box, supplemented their interaction experience. This concept is called experience scalability .

This does not negate the fact that the beginning is also important in shaping the perception of the user design. Many different studies have shown that the first impression of the site has a powerful anchor effect for later experience. Kahneman’s study made it clear that the beginning is not the only important thing. Completions matter too. And, based on my experience, completion more often needs to be corrected during the design process.

A few examples of good completions:

  • After reading the article, the user sees several relevant stories.
  • After making a purchase, the user can verify the purchase data as a guest, without authorization.
  • Sites like GitHub and Gmail protect user data from accidental deletion: the user confirms the deletion.

Here are some examples of not very good endings:

  • At the end of the article, the reader sees several tempting links to irrelevant sponsored content.
  • When a user tries to leave a site, a window appears that is desperately trying to get hold of him by email.
  • After the user passes the training part in a new application, a short welcome letter (in text format) comes. Hello in hindsight. At least sometime.

Finally, if the user wants to unsubscribe from your advertising letters, do not be a monster and do not complicate the process of unsubscribing! There is a whole Tumblr Spot the Unsubscribe dedicated e-mail in which it is ridiculously hard to find a link to unsubscribe.

Slow down when necessary

Remembering his impressions of using something, a person focuses on key points and ignores the duration. It turns out that in some situations, when we need to control the user's experience, we have more freedom than we think: if necessary, specifically slow down the interaction in order to improve the final impression.

Difficult said, I will explain with an example.

On the web, everything happens very quickly - so much so that sometimes it is difficult to capture the user's attention. But do not rush to do everything according to the principle "Do not make me think." Instead of quickly impressing the user, add what Andrew Grimes calls the meta moments : “small moments of reflection that make us aware of what we are experiencing.” For example, Slack uses pulsating tooltips throughout training, which take extra time and attention, but help educate completely new users, and more advanced ones give energy to start using those Slack functions that they had not previously used.

Psychologist Dan Ariely, in his blog, tells the story of how he met with a mechanic and talked to him about work:

The better and faster the locksmith worked, the less tips he was given, and the more they complained. Strange, but people estimated the value of his work in inverse proportion to the amount of time spent opening the door. Arily explains that the moral here is that people evaluate the result as the amount of effort spent by the mechanic to do the work.

A vivid example of creating artificial delays to create the illusion that you are influencing something is Intuit's TurboTax. Interestingly, the elaboration of this illusion is manifested in small dialogues (similar to this one) that constantly appear in TurboTax:

TurboTax uses artificial waiting time to create the illusion that you are affecting something.

Of course, this software does not need to double check the data. But when you look at the screen, do you have a feeling that TurboTax does a lot more to take care of you? In other words, this dialogue is a dummy without meaning, but it also improves the impression of use. In this case, the increase in task execution time is a pleasant addition that will remain in memory after use. And TurboTax is not the only company using this method.

Create peak moments

Peak moments are good because they are remembered. It turned out that memorable moments have one thing in common: they are emotionally weighty. For years, studies have shown that human memory is focused on events related to emotions .

Don Norman and Aaron Walter have been writing about design elements based on emotions for several years. Emotional design advises to go beyond the benefits and convenience to create moments that will be remembered for a long time. Properly used elements of emotional design will help reinforce the impression.

One of the techniques of emotional design — my favorite — is to create amazing moments, allowing users to make their own discoveries in the user interface. When my colleagues and I created JamBells , a collective online game for mobile devices, we added a secret to the list of playable songs. Check on the smartphone: can you find it? ..

Another example: Snapchat lens, a killer feature, which is almost impossible to find on your own in the user interface. I recently spent the whole evening with four other UX designers trying to figure out how to make a swap face. Despite the lack of convenience in the usual sense of the word, we felt very cool when we finally guessed.

For a designer, the theory of peak moments seems to say: “Don't be a perfectionist in trifles, look wider.” Several years ago, Don Norman wrote about how he asked people to tell him about things that they did not like when traveling to Disney or in a new iPhone. Everyone was ready to tell a couple of scary stories about a trip or a new operating system, but when asked if they would recommend to their friends, they almost always answered firmly, yes.

At the end of the story, Norman gives advice: “Perfectionism is rarely worth the effort. So why not leave some problems unsolved ... The overall impression is important. ”

Our goal as UX designers is not only to look for and fix all problems. It is not only impossible, it is stupid. Recall the peak-end rule and the graphs we reviewed. This is an illustration of the fact that peak moments overlap negative ones.

For example, experiments on the graphs below differ by only one negative point. It’s easy to imagine how the designer focuses on correcting a flaw, rather than creating a peak. And it turns out a convenient, but bland experience, in which any emerging problem is clearly visible.

An experience without bright moments is sensitive to sudden problems.

On the contrary, the experience with a bright positive moment (the example on the graphs below) is more resistant to problems. Theoretically, one positive peak overlaps the rest of the negative experience.

Peak moments alleviate unexpected problems with product use.

The focus on creating high emotional peaks is distinguished by a convenient, but poorly memorable product, and a cool product that will leave a positive impression for a long time.

Be a story designer

Until recently, I was skeptical of stories. You can say my skepticism was similar to the statements of Stefan Sagmeister . I thought that the very idea that stories can teach us anything is, at best, a bit exaggerated.

But then I watched a video of Kurt Vonnegut. (Attached below. Look, it will take a maximum of 4 minutes.)

Do the familiar types of graphs that Vonnegut draws seem to you? This is the same as the graphics of impressions in the study of Kahneman.

Kurt Vonnegut depicts the life of Cinderella from the fairy tale of the same name on the chart.

A student at Tufts University went a little further, showing that this chart reflects almost every Disney cartoon. And Disney has been making money on this scheme for years.

Have you noticed the details of this successful type of story? There is a high peak and a high level of completion, or, as they say in show business, a climax and a happy ending.

Human memory divides the flow of impressions into separate moments. It turns out a sequence of moments that occur over time, clearly marked beginning and end, the alternation of good and bad moments throughout the plot. This is history. People write stories every day. We are constantly creating stories from moments that make sense to us.

How to use this advantage? A useful thing for applying the narrator's thinking in UX design is an experience map or path map. This map is a great way to connect qualitative and quantitative research and visualize a story that a user will learn with a product or service. Adaptive Path made a useful guide .

Experience cards can be complex and detailed, but the most important thing here is the experience schedule. Where do the peaks come from? Where do differences arise? How does the experience end?

Finally, the thoughts expressed by the narrator should not be vague and indistinct. Donna Liko notes that any sequence of steps taken by the user for some time can be described both by data (facts, statistics) and in a narrative format with elements of stories (description, presentation, culmination). Data and stories support each other and share information. Analytics tells “What”, stories - “How”, and together they inspire new design ideas and test hypotheses.


Impressions and memories of him are closely related, but they are constantly different. In each person there live two I: the experiencing I and the remembering I; memorizing teaches us, judges something and makes decisions. Memory is a set of frames, among which the most intense moments of experiences and the final moment are significant. UX design is still important, but you need to strive to think more in terms of memory design.

How to apply it: first we pay attention to the endings to make sure that we will not disappoint users at the last moment. Then slow down as much as possible, and show users that we spend time taking care of them. Then we go beyond usability and use elements of emotional design to create peaks, focusing on the picture as a whole and getting a design that is more resistant to unexpected problems. Finally, you need to be attentive to the chips of the story that the design narrates, and pay attention to significant moments (peaks and ends).

And to finish on a positive note:


about the author

Kurt arlege

Kurt : researcher, designer, theorist, writer and lecturer on UX design. He works at Viget's in Falls Church, Virginia for clients such as POLITICO, Massachusetts General Hospital and National Trust for Historic Preservation.

Kurt's other posts

The original article can be found here.
Translated by rishavant

Also popular now: