The Design of Everyday Things

The model

Donald Norman was a cognitive psychologist at the University of California, San Diego, who had studied how people make errors. In the early 1980s he was asked to help analyse the accident at the Three Mile Island nuclear plant, and he concluded that the operators had been blamed for errors that the control room’s design had all but guaranteed. That experience led to The Psychology of Everyday Things, published in 1988 and renamed The Design of Everyday Things in later editions. A substantially revised edition appeared in 2013.

The book’s model describes what a person needs in order to use something without confusion.

Affordances are the relationships between an object and a person that determine how it could be used: a chair affords sitting, a flat plate on a door affords pushing. Norman borrowed the term from the perceptual psychologist James Gibson. In the 2013 edition he added signifiers, the signals that tell people where an affordance is and how to use it, because designers had begun using “affordance” loosely to mean any visual cue.

Mapping is the relationship between controls and their effects. Natural mappings use spatial analogy, so that the arrangement of controls mirrors the arrangement of what they control. A stove with four burners in a square and four knobs in a line forces people to remember which knob controls which burner; the same knobs laid out in a square need no labels.

Feedback is immediate and informative communication of the results of an action. Without it people repeat actions, give up or make errors.

Constraints limit the possible actions: physical constraints such as a plug that fits only one way, cultural conventions, semantic meaning and logical relationships.

Conceptual models are people’s explanations of how something works. The designer has one model, the user forms another, and the only link between them is the object itself, which Norman calls the system image. When the system image is unclear, users build wrong models and make confident mistakes.

Norman puts these together in an account of action. To use anything, people must bridge two gulfs. The gulf of execution is the gap between what they want to do and figuring out how to do it. The gulf of evaluation is the gap between what has happened and figuring out whether it is what they wanted. Good design narrows both.

Seeing with it

The model’s most famous illustrations are doors. A door that needs a sign saying “push” has failed, because its design should already say so. The everyday name “Norman door”, later coined by others for a door whose design misleads, shows how far the book’s ideas have spread.

Norman also uses the model to reframe human error. Following the psychologist James Reason, he distinguishes slips, in which a person intends the right action but does something else, from mistakes, in which the intention itself is wrong because of a faulty model. Slips are often caused by designs in which similar actions have very different results, such as adjacent switches with different functions. Mistakes come from misleading system images. In both cases the response should be to change the design, not to scold the user. Norman argues that the label “human error” usually means the investigation stopped too early.

He also distinguishes knowledge in the head from knowledge in the world. People do not need to remember how to use well-designed things, because the design carries the knowledge. When designs remove knowledge from the world, such as identical buttons or hidden menus, they force people to rely on memory, which is fragile.

Where the model stops

Norman’s model was developed largely for physical objects and early computer interfaces, and the 2013 edition extends it to digital products, but some of its central assumptions have been tested by later technology. Touchscreen and voice interfaces offer few visible signifiers, and designers have had to invent new conventions. Norman himself has written about the limits of “user-centred design”, arguing in later work that design must also attend to activities, communities and emotion, as he did in Emotional Design (2004).

The model is also mostly about usability: whether people can do what they intend. It says less about whether designs should help people do what they intend, or what happens when designers apply the same principles to steer people toward actions that serve the designer. The site’s idea page The Architecture of Choice takes up that question.

Since it was published

The book became a foundation of user experience design, the profession that grew rapidly with the web and mobile apps. Norman later worked at Apple, co-founded a design consultancy with Jakob Nielsen, and founded the Design Lab at the University of California, San Diego. The 2013 edition added chapters on design thinking, the business of design and the human-centred design process. The site’s essays on Ink That Never Dries and A Computer of One’s Own show how many of these principles shaped the software people use every day.

Questions it lets you ask

Can a new user tell, just by looking, what actions are possible and where? Do the controls map naturally to what they control? When someone acts, do they know immediately what happened? What model will a user form from what they see, and is it the right one? When people make errors with our product, what in the design invited them? And where are we forcing people to remember something that the design could show?