The Design of Everyday Things

by Don Norman

  This example highlights the themes of this chapter. First, how do people do things? It is easy to learn a few basic steps to perform operations with our technologies (and yes, even filing cabinets are technology). But what happens when things go wrong? How do we detect that they aren’t working, and then how do we know what to do? To help understand this, I first delve into human psychology and a simple conceptual model of how people select and then evaluate their actions. This leads the discussion to the role of understanding (via a conceptual model) and of emotions: pleasure when things work smoothly and frustration when our plans are thwarted. Finally, I conclude with a summary of how the lessons of this chapter translate into principles of design.

  How People Do Things: The Gulfs of Execution and Evaluation

  When people use something, they face two gulfs: the Gulf of Execution, where they try to figure out how it operates, and the Gulf of Evaluation, where they try to figure out what happened (Figure 2.1). The role of the designer is to help people bridge the two gulfs.

  In the case of the filing cabinet, there were visible elements that helped bridge the Gulf of Execution when everything was working perfectly. The drawer handle clearly signified that it should be pulled and the slider on the handle indicated how to release the catch that normally held the drawer in place. But when these operations failed, there then loomed a big gulf: what other operations could be done to open the drawer?

  The Gulf of Evaluation was easily bridged, at first. That is, the catch was released, the drawer handle pulled, yet nothing happened. The lack of action signified a failure to reach the goal. But when other operations were tried, such as my twisting and pulling, the filing cabinet provided no more information about whether I was getting closer to the goal.

  The Gulf of Evaluation reflects the amount of effort that the person must make to interpret the physical state of the device and to determine how well the expectations and intentions have been met. The gulf is small when the device provides information about its state in a form that is easy to get, is easy to interpret, and matches the way the person thinks about the system. What are the major design elements that help bridge the Gulf of Evaluation? Feedback and a good conceptual model.

  FIGURE 2.1.The Gulfs of Execution and Evaluation. When people encounter a device, they face two gulfs: the Gulf of Execution, where they try to figure out how to use it, and the Gulf of Evaluation, where they try to figure out what state it is in and whether their actions got them to their goal.

  The gulfs are present for many devices. Interestingly, many people do experience difficulties, but explain them away by blaming themselves. In the case of things they believe they should be capable of using—water faucets, refrigerator temperature controls, stove tops—they simply think, “I’m being stupid.” Alternatively, for complicated-looking devices—sewing machines, washing machines, digital watches, or almost any digital controls—they simply give up, deciding that they are incapable of understanding them. Both explanations are wrong. These are the things of everyday household use. None of them has a complex underlying structure. The difficulties reside in their design, not in the people attempting to use them.

  How can the designer help bridge the two gulfs? To answer that question, we need to delve more deeply into the psychology of human action. But the basic tools have already been discussed: We bridge the Gulf of Execution through the use of signifiers, constraints, mappings, and a conceptual model. We bridge the Gulf of Evaluation through the use of feedback and a conceptual model.

  The Seven Stages of Action

  There are two parts to an action: executing the action and then evaluating the results: doing and interpreting. Both execution and evaluation require understanding: how the item works and what results it produces. Both execution and evaluation can affect our emotional state.

  Suppose I am sitting in my armchair, reading a book. It is dusk, and the light is getting dimmer and dimmer. My current activity is reading, but that goal is starting to fail because of the decreasing illumination. This realization triggers a new goal: get more light. How do I do that? I have many choices. I could open the curtains, move so that I sit where there is more light, or perhaps turn on a nearby light. This is the planning stage, determining which of the many possible plans of action to follow. But even when I decide to turn on the nearby light, I still have to determine how to get it done. I could ask someone to do it for me, I could use my left hand or my right. Even after I have decided upon a plan, I still have to specify how I will do it. Finally, I must execute—do—the action. When I am doing a frequent act, one for which I am quite experienced and skilled, most of these stages are subconscious. When I am still learning how to do it, determining the plan, specifying the sequence, and interpreting the result are conscious.

  Suppose I am driving in my car and my action plan requires me to make a left turn at a street intersection. If I am a skilled driver, I don’t have to give much conscious attention to specify or perform the action sequence. I think “left” and smoothly execute the required action sequence. But if I am just learning to drive, I have to think about each separate component of the action. I must apply the brakes and check for cars behind and around me, cars and pedestrians in front of me, and whether there are traffic signs or signals that I have to obey. I must move my feet back and forth between pedals and my hands to the turn signals and back to the steering wheel (while I try to remember just how my instructor told me I should position my hands while making a turn), and my visual attention is divided among all the activity around me, sometimes looking directly, sometimes rotating my head, and sometimes using the rear- and side-view mirrors. To the skilled driver, it is all easy and straightforward. To the beginning driver, the task seems impossible.

  FIGURE 2.2.The Seven Stages of the Action Cycle. Putting all the stages together yields the three stages of execution (plan, specify, and perform), three stages of evaluation (perceive, interpret, and compare), and, of course, the goal: seven stages in all.

  The specific actions bridge the gap between what we would like to have done (our goals) and all possible physical actions to achieve those goals. After we specify what actions to make, we must actually do them—the stages of execution. There are three stages of execution that follow from the goal: plan, specify, and perform (the left side of Figure 2.2). Evaluating what happened has three stages: first, perceiving what happened in the world; second, trying to make sense of it (interpreting it); and, finally, comparing what happened with what was wanted (the right side of Figure 2.2).

  There we have it. Seven stages of action: one for goals, three for execution, and three for evaluation (Figure 2.2).

  1.Goal (form the goal)

  2.Plan (the action)

  3.Specify (an action sequence)

  4.Perform (the action sequence)

  5.Perceive (the state of the world)

  6.Interpret (the perception)

  7.Compare (the outcome with the goal)

  The seven-stage action cycle is simplified, but it provides a useful framework for understanding human action and for guiding design. It has proven to be helpful in designing interaction. Not all of the activity in the stages is conscious. Goals tend to be, but even they may be subconscious. We can do many actions, repeatedly cycling through the stages while being blissfully unaware that we are doing so. It is only when we come across something new or reach some impasse, some problem that disrupts the normal flow of activity, that conscious attention is required.

  Most behavior does not require going through all stages in sequence; however, most activities will not be satisfied by single actions. There must be numerous sequences, and the whole activity may last hours or even days. There are multiple feedback loops in which the results of one activity are used to direct further ones, in which goals lead to subgoals, and plans lead to subplans. There are activities in which goals are forgotten, discarded, or reformulated.

  Let’s go back to my act of turning on the light. This is a case of event-dri
ven behavior: the sequence starts with the world, causing evaluation of the state and the formulation of a goal. The trigger was an environmental event: the lack of light, which made reading difficult. This led to a violation of the goal of reading, so it led to a subgoal—get more light. But reading was not the high-level goal. For each goal, one has to ask, “Why is that the goal?” Why was I reading? I was trying to prepare a meal using a new recipe, so I needed to reread it before I started. Reading was thus a subgoal. But cooking was itself a subgoal. I was cooking in order to eat, which had the goal of satisfying my hunger. So the hierarchy of goals is roughly: satisfy hunger; eat; cook; read cookbook; get more light. This is called a root cause analysis: asking “Why?” until the ultimate, fundamental cause of the activity is reached.

  The action cycle can start from the top, by establishing a new goal, in which case we call it goal-driven behavior. In this situation, the cycle starts with the goal and then goes through the three stages of execution. But the action cycle can also start from the bottom, triggered by some event in the world, in which case we call it either data-driven or event-driven behavior. In this situation, the cycle starts with the environment, the world, and then goes through the three stages of evaluation.

  For many everyday tasks, goals and intentions are not well specified: they are opportunistic rather than planned. Opportunistic actions are those in which the behavior takes advantage of circumstances. Rather than engage in extensive planning and analysis, we go about the day’s activities and do things as opportunities arise. Thus, we may not have planned to try a new café or to ask a question of a friend. Rather, we go through the day’s activities, and if we find ourselves near the café or encountering the friend, then we allow the opportunity to trigger the appropriate activity. Otherwise, we might never get to that café or ask our friend the question. For crucial tasks we make special efforts to ensure that they get done. Opportunistic actions are less precise and certain than specified goals and intentions, but they result in less mental effort, less inconvenience, and perhaps more interest. Some of us adjust our lives around the expectation of opportunities. And sometimes, even for goal-driven behavior, we try to create world events that will ensure that the sequence gets completed. For example, sometimes when I must do an important task, I ask someone to set a deadline for me. I use the approach of that deadline to trigger the work. It may only be a few hours before the deadline that I actually get to work and do the job, but the important point is that it does get done. This self-triggering of external drivers is fully compatible with the seven-stage analysis.

  The seven stages provide a guideline for developing new products or services. The gulfs are obvious places to start, for either gulf, whether of execution or evaluation, is an opportunity for product enhancement. The trick is to develop observational skills to detect them. Most innovation is done as an incremental enhancement of existing products. What about radical ideas, ones that introduce new product categories to the marketplace? These come about by reconsidering the goals, and always asking what the real goal is: what is called the root cause analysis.

  Harvard Business School marketing professor Theodore Levitt once pointed out, “People don’t want to buy a quarter-inch drill. They want a quarter-inch hole!” Levitt’s example of the drill implying that the goal is really a hole is only partially correct, however. When people go to a store to buy a drill, that is not their real goal. But why would anyone want a quarter-inch hole? Clearly that is an intermediate goal. Perhaps they wanted to hang shelves on the wall. Levitt stopped too soon.

  Once you realize that they don’t really want the drill, you realize that perhaps they don’t really want the hole, either: they want to install their bookshelves. Why not develop methods that don’t require holes? Or perhaps books that don’t require bookshelves. (Yes, I know: electronic books, e-books.)

  Human Thought: Mostly Subconscious

  Why do we need to know about the human mind? Because things are designed to be used by people, and without a deep understanding of people, the designs are apt to be faulty, difficult to use, difficult to understand. That is why it is useful to consider the seven stages of action. The mind is more difficult to comprehend than actions. Most of us start by believing we already understand both human behavior and the human mind. After all, we are all human: we have all lived with ourselves all of our lives, and we like to think we understand ourselves. But the truth is, we don’t. Most of human behavior is a result of subconscious processes. We are unaware of them. As a result, many of our beliefs about how people behave—including beliefs about ourselves—are wrong. That is why we have the multiple social and behavioral sciences, with a good dash of mathematics, economics, computer science, information science, and neuroscience.

  Consider the following simple experiment. Do all three steps:

  1.Wiggle the second finger of your hand.

  2.Wiggle the third finger of the same hand.

  3.Describe what you did differently those two times.

  On the surface, the answer seems simple: I thought about moving my fingers and they moved. The difference is that I thought about a different finger each time. Yes, that’s true. But how did that thought get transmitted into action, into the commands that caused different muscles in the arm to control the tendons that wiggled the fingers? This is completely hidden from consciousness.

  The human mind is immensely complex, having evolved over a long period with many specialized structures. The study of the mind is the subject of multiple disciplines, including the behavioral and social sciences, cognitive science, neuroscience, philosophy, and the information and computer sciences. Despite many advances in our understanding, much still remains mysterious, yet to be learned. One of the mysteries concerns the nature of and distinction between those activities that are conscious and those that are not. Most of the brain’s operations are subconscious, hidden beneath our awareness. It is only the highest level, what I call reflective, that is conscious.

  Conscious attention is necessary to learn most things, but after the initial learning, continued practice and study, sometimes for thousands of hours over a period of years, produces what psychologists call “overlearning,” Once skills have been overlearned, performance appears to be effortless, done automatically, with little or no awareness. For example, answer these questions:

  What is the phone number of a friend?

  What is Beethoven’s phone number?

  What is the capital of:

  •Brazil?

  •Wales?

  •The United States?

  •Estonia?

  Think about how you answered these questions. The answers you knew come immediately to mind, but with no awareness of how that happened. You simply “know” the answer. Even the ones you got wrong came to mind without any awareness. You might have been aware of some doubt, but not of how the name entered your consciousness. As for the countries for which you didn’t know the answer, you probably knew you didn’t know those immediately, without effort. Even if you knew you knew, but couldn’t quite recall it, you didn’t know how you knew that, or what was happening as you tried to remember.

  You might have had trouble with the phone number of a friend because most of us have turned over to our technology the job of remembering phone numbers. I don’t know anybody’s phone number—I barely remember my own. When I wish to call someone, I just do a quick search in my contact list and have the telephone place the call. Or I just push the “2” button on the phone for a few seconds, which autodials my home. Or in my auto, I can simply speak: “Call home.” What’s the number? I don’t know: my technology knows. Do we count our technology as an extension of our memory systems? Of our thought processes? Of our mind?

  What about Beethoven’s phone number? If I asked my computer, it would take a long time, because it would have to search all the people I know to see whether any one of them was Beethoven. But you immediately discarded the question as nonsensical. You don’t personally know Beet
hoven. And anyway, he is dead. Besides, he died in the early 1800s and the phone wasn’t invented until the late 1800s. How do we know what we do not know so rapidly? Yet some things that we do know can take a long time to retrieve. For example, answer this:

  In the house you lived in three houses ago, as you entered the front door, was the doorknob on the left or right?

  Now you have to engage in conscious, reflective problem solving, first to retrieve just which house is being talked about, and then what the correct answer is. Most people can determine the house, but have difficulty answering the question because they can readily imagine the doorknob on both sides of the door. The way to solve this problem is to imagine doing some activity, such as walking up to the front door while carrying heavy packages with both hands: how do you open the door? Alternatively, visualize yourself inside the house, rushing to the front door to open it for a visitor. Usually one of these imagined scenarios provides the answer. But note how different the memory retrieval for this question was from the retrieval for the others. All these questions involved long-term memory, but in very different ways. The earlier questions were memory for factual information, what is called declarative memory. The last question could have been answered factually, but is usually most easily answered by recalling the activities performed to open the door. This is called procedural memory. I return to a discussion of human memory in Chapter 3.

  Walking, talking, reading. Riding a bicycle or driving a car. Singing. All of these skills take considerable time and practice to master, but once mastered, they are often done quite automatically. For experts, only especially difficult or unexpected situations require conscious attention.