The Design of Everyday Things

by Donald A. Norman

The behavioral level is the home of learned skills, triggered by situations that match the appropriate patterns. Actions and analyses at this level are largely subconscious.

The reflective level is the home of conscious cognition. As a consequence, this is where deep understanding develops, where reasoning and conscious decision-making take place.

The highest levels of emotions come from the reflective level, for it is here that causes are assigned and where predictions of the future take place.

To the designer, reflection is perhaps the most important of the levels of processing. Reflection is conscious, and the emotions produced at this level are the most protracted: those that assign agency and cause, such as guilt and blame or praise and pride.

It is reflection that drives us to recommend a product, to recommend that others use it—or perhaps to avoid it.

One important emotional state is the one that accompanies complete immersion into an activity, a state that the social scientist Mihaly Csikszentmihalyi has labeled “

When in the flow state, people lose track of time and the outside environment. They are at one with the task they are performing. The task, moreover, is at just the proper level of difficulty: difficult enough to provide a challenge and require continued attention, but not so difficult that it invokes frustration and anxiety.

People are innately disposed to look for causes of events, to form explanations and stories. That is one reason storytelling is such a persuasive medium. Stories resonate with our experiences and provide examples of new instances. From our experiences and the stories of others we tend to form generalizations about the way people behave and things work. We attribute causes to events, and as long as these cause-and-effect pairings make sense, we accept them and use them for understanding future events. Yet these causal attributions are often erroneous. Sometimes they implicate the wrong causes, and for some things that happen, there is no single cause; rather, a complex chain of events that all contribute to the result: if any one of the events would not have occurred, the result would be different. But even when there is no single causal act, that doesn’t stop people from assigning one.

Conceptual models are a form of story, resulting from our predisposition to find explanations. These models are essential in helping us understand our experiences, predict the outcome of our actions, and handle unexpected occurrences. We base our models on whatever knowledge we have, real or imaginary, naive or sophisticated.

Conceptual models are often constructed from fragmentary evidence, with only a poor understanding of what is happening, and with a kind of naive psychology that postulates causes, mecha...

People try to find causes for events. They tend to assign a causal relation whenever two things occur in succession. If some unexpected event happens in my home just after I have taken some action, I am apt to conclude that it was caused by that action, even if there really was no relationship between the two. Similarly, if I do something expecting a result and nothing happens, I am apt to interpret this lack of informative feedback as an indication that I didn’t do the action correctly: the most likely thing to do, therefore, is to repeat the action, only with more force. Push a door and it fails to open? Push again, harder. With electronic devices, if the feedback is delayed sufficiently, people often are led to conclude that the press wasn’t recorded, so they do the same action again, sometimes repeatedly, unaware that all of their presses were recorded. This can lead to unintended results. Repeated presses might intensify the response much more than was intended. Alternatively, a second request might cancel the previous one, so that an odd number of pushes produces the desired result, whereas an even number leads to no result.

Modern systems try hard to provide feedback within 0.1 second of any operation, to reassure the user that the request was received. This is especially important if the operation will take considerable time. The presence of a filling hourglass or rotating clock hands is a reassuring sign that work is in progress. When the delay can be predicted, some systems provide time estimates as well as progress bars to indicate how far along the task has gone. More systems should adopt these sensible displays to provide timely and meaningful feedback of results.

Some studies show it is wise to underpredict—that is, to say an operation will take longer than it actually will. When the system computes the amount of time, it can compute the range of possible times. In that case it ought to display the range, or if only a single value is desirable, show the slowest, longest value. That way, the expectations are liable to be exceeded, leading to a happy result.

We need to remove the word failure from our vocabulary, replacing it instead with learning experience. To fail is to learn: we learn more from our failures than from our successes. With success, sure, we are pleased, but we often have no idea why we succeeded. With failure, it is often possible to figure out why, to ensure that it will never happen again.

“Fail often, fail fast,”

Designers need to fail, as do researchers.

failures are an essential part of exploration...

If designers and researchers do not sometimes fail, it is a sign that they are not trying hard enough—they are not thinking the great creative thoughts that will provide breakthroughs in how we do things. It is possible to avoid failure, to always ...

Designers should strive to minimize the chance of inappropriate actions in the first place by using affordances, signifiers, good mapping, and constraints to guide the actions. If a person performs an inappropriate action, the design should maximize the chance that this can be discovered and then rectified. This requires good, intelligible feedback coupled with a simple, clear conceptual model. When people understand what has happened, what state the system is in, and what the most appropriate set of actions is, they can perform their activities more effectively.

The hard and necessary part of design is to make things work well even when things do not go as planned.

1. Discoverability. It is possible to determine what actions are possible and the current state of the device. 2. Feedback. There is full and continuous information about the results of actions and the current state of the product or service. After an action has been executed, it is easy to determine the new state. 3. Conceptual model. The design projects all the information needed to create a good conceptual model of the system, leading to understanding and a feeling of control. The conceptual model enhances both discoverability and evaluation of results. 4. Affordances. The proper affordances exist to make the desired actions possible. 5. Signifiers. Effective use of signifiers ensures discoverability and that the feedback is well communicated and intelligible. 6. Mappings. The relationship between controls and their actions follows the principles of good mapping, enhanced as much as possible through spatial layout and temporal contiguity. 7. Constraints. Providing physical, logical, semantic, and cultural constraints guides actions and eases interpretation.

“Don’t criticize unless you can do better.

Thinking about the causes and possible fixes to bad design should make you better appreciate good design. So, the next time you come across a well-designed object, one that you can use smoothly and effortlessly on the first try, stop and examine it. Consider how well it masters the seven stages of action and the principles of design.

Recognize that most of our interactions with products are actually interactions with a complex system: good design requires consideration of the entire system to ensure that the requirements, intentions, and desires at each stage are faithfully understood and respected at all the other stages.

People function through their use of two kinds of knowledge: knowledge of and knowledge how. Knowledge of—what psychologists call declarative knowledge—includes the knowledge of facts and rules. “Stop at red traffic lights.” “New York City is north of Rome.” “China has twice as many people as India.” “To get the key out of the ignition of a Saab car, the gearshift must be in reverse.” Declarative knowledge is easy to write and to teach.

Knowledge how—what psychologists call procedural knowledge—is the knowledge that enables a person to be a skilled musician, to return a serve in tennis, or to move the tongue properly when saying the phrase “frightening witches.” Procedural knowledge is difficult or impossible to write down and difficult to teach. It is best taught by demonstration and best learned through practice. Even the best teachers cannot usually describe what they are doing. Procedural knowledge is largely subconscious, residing at the behavioral level of processing.

Constraints are powerful tools for the designer:

Short-term or working memory (STM) retains the most recent experiences or material that is currently being thought about.

Long-term memory (LTM) is memory for the past.

The most effective way of helping people remember is to make it unnecessary.

Conscious thinking takes time and mental resources. Well-learned skills bypass the need for conscious oversight and control: conscious control is only required for initial learning and for dealing with unexpected situations. Continual practice automates the action cycle, minimizing the amount of conscious thinking and problem-solving required to act.

Approximate answers are often good enough, even if technically wrong.

Conceptual models are powerful explanatory devices, useful in a variety of circumstances. They do not have to be accurate as long as they lead to the correct behavior in the desired situation.

The unaided mind is surprisingly limited. It is things that make us smart. Take advantage of them.

The partnership of technology and people makes us smarter, stronger, and better able to live in the modern world.