The Design of Everyday Things

Chapter 3 addresses the gap between “knowledge in the head” (what people know) and “knowledge in the world” (explicit design cues), arguing it is the designer’s job to bridge this gap. Norman holds that people rarely need to understand the details of complex theories or inventions—simplified approximations generally suffice.

Precise Behavior from Imprecise Knowledge

People derive information from two sources: their minds and external cues. If designers put sufficient cues into their products (“knowledge in the world”), they can prompt users who have imprecise knowledge about the product to behave in precise ways. People do not need an in-depth understanding of complex ideas and inventions, just enough to complete tasks. Norman describes two kinds of knowledge: knowledge of (declarative knowledge) and knowledge how (procedural knowledge). The former includes knowledge of rules and facts, while the latter is difficult to write down and best taught by demonstration.

Research in the field of psychology reveals that people only retain partial information because external knowledge makes precise knowledge unnecessary (80). Signifiers and mappings are perceivable cues that function as external knowledge. Norman explains that this type of knowledge is everywhere, from letters on keyboards to labels on controls. Constraints also allow people to function with incomplete knowledge. For example, several words are synonymous with “a mythical being,” but only one rhymes with the word “post” (ghost). Constraints simplify what must be retained in memory by reducing large numbers of decisions to only a few choices, making them powerful tools for designers.

Memory is Knowledge in the Head

People must memorize large amounts of arbitrary data to function in the world. As Norman observes, technology did away with the need to memorize phone numbers, postal codes, and other data, but it created new burdens by forcing people to memorize passwords. Indeed, security poses major design problems related to technology and human behavior. Those who write down their passwords are vulnerable to data theft. Even memorizing a strong password, however, does not guarantee safety in the age of keystroke logging. The use of biometric markers has not done away with the need for passwords. The safest systems require multiple identifiers, usually something one has, such as a card or a key, and something one knows, such as a password.

The Structure of Memory

For Norman, “knowledge in the head” is synonymous with memory. Psychologists have identified two types of memory: short-term memory and long-term memory. The former, also called “working memory,” pertains to one’s most recent experiences and is easy to retrieve, but extremely limited. Long-term memory is memory of the past. Studies reveal that sleep plays a key role in creating long-term memories, though its precise role remains poorly understood (96).

The mind records memories in bits and pieces, reconstructing and interpreting them as needed. Thus, all long-term memories are subject to distortions. Organization and retrieval are key problems with long-term memory, which allows people to reconstruct events in ways they would prefer to remember, rather than how they occurred. Indeed, psychological experiments reveal how easy it is to implant false memories, while studies consistently show that eyewitness testimony is unreliable (97).

Memory differs depending on content. Arbitrary knowledge, such as the letters of the alphabet and people’s names, requires rote memorization. Rote learning takes time and effort but can be made easier if the material is given a meaningful structure or a good conceptual model. Norman argues that good design either creates meaningful structures, or makes memory unnecessary by putting the required information in the world (e.g., in the form of dropdown menus). 

Approximate Models: Memory in the Real World

Conscious thinking requires time and mental effort, but approximations and the use of simplified models can help bypass this laborious process. Norman uses formulas to explain this concept. For example, substituting the formula °C = (°F–30) / 2 for °C = (°F-32) x 5/9 to convert temperatures from Fahrenheit to Celsius exemplifies simplification. Norman argues that situations demanding precision are best performed by machines. This observation gestures to a key point he makes later in Chapter 3: People and machines have complementary skills and are mutually reinforcing.

Knowledge in the Head

Effective memory combines “knowledge in the head” and “knowledge in the world,” allowing people to function even when their knowledge is imperfect. Transforming knowledge into memory through writing or other means is key to remembering and retrieving information. For designers, this might entail allowing users to enter information into a product in real time.

Prospective memory, or memory for the future, refers to remembering to do something at a future time. Strategies for reminding include keeping the knowledge in your head, using notes or calendar reminders, or setting alarms. Reminders have two parts: the signal and the message. The former is knowing that something must be remembered, while the latter refers to the act of remembering. A note, for example, provides a record of the thing to be remembered, but it does not remind the writer to look at it. As Norman observes, ideal reminders combine the two components.

The Tradeoff Between Knowledge in the World and in the Head

Norman argues that favoring external knowledge over internal knowledge comes with a tradeoff. External knowledge functions as its own reminder, helping people recover structures they might otherwise forget. Unlike internal knowledge, “knowledge in the world” requires no learning. However, “knowledge in the world” is more difficult to use than “knowledge in the head.” Moreover, it relies heavily on the consistent presence of the knowledge. Even when physically present, external knowledge can be overlooked, as evidenced by neglected piles of paper on desks. Digital information poses similar problems, forcing people to remember not the details of the information, but that the information is stored in the first place.

Memory in Multiple Heads, Multiple Devices

The partnership between people and technology creates what Norman calls “super-powerful beings” (112). According to Norman, taking away external devices makes people less capable.

Natural Mapping

This section focuses on creating user-friendly products using natural mapping. Mapping combines internal and external knowledge. Bad mapping, such as stoves that obfuscate the relationship between the controls and the burners, guarantees errors. Natural mapping, by contrast, makes the relationship between the control and the object being controlled obvious.

Natural mapping is effective because it relies on spatial cues. The best mappings place controls directly on the object being controlled. The second-best place controls nearby, while the third best arranges controls in the same spatial configuration as the items being controlled. Gesture-controlled hand dryers and faucets exemplify natural mapping. However, these mappings do not have signifiers and thus lack discoverability. For example, waving one’s hand under a towel dispenser and not getting a towel might indicate that the gesture was imprecise or that the machine is broken or out of towels. In other words, the mapping is natural, but the lack of signifiers hinders usability.

Cultural Design: Natural Mappings Can Vary with Culture

Chapter 3 concludes with an anecdote, which serves as the point of departure for a discussion of cultural specificity. Norman describes using a remote controller to advance his illustrations during a presentation in Asia. To his surprise, the top button reversed his images, while the bottom button advanced them. He polled the audience, and others around the globe, to determine what made more sense: using the top or bottom button to advance images. The audiences were split in their responses.

Norman realized that different cultures have different expectations, depending on how they conceptualize and represent time. For example, some cultures represent timelines vertically, with the future going up and the past going down, while others represent it horizontally, with the past on the left and the future on the right. As Norman notes, these visual metaphors are arbitrary and culturally-specific. What appears “natural” depends on one’s point of view and cultural context. It is possible to break these kinds of conventions, but not without a period of confusion as people adapt to a new system.