16 · Deciding
Mental Model
People carry a picture in their head of how something works, and use it to guess what happens next.
Evidence: Rule of thumbWhat we do with it
- 01
Ask people how they think it works before you show them the design. Their words tell you what to call things.
- 02
When a label confuses people, rename it. A tooltip explaining a wrong name rarely saves it.
- 03
Teaching people a new way of thinking about a product is possible, and it's expensive. Spend that only where the new idea is the point of the product.
Examples
Getting it wrong, getting it right.
Don't
A “Save” button in an app that already saves on every change. People click it, nothing happens, and they stop trusting that their work is saved.
Do
No Save button, and a quiet “Saved” note that appears after each change.
The evidence
What the research says.
A way of thinking about users. It isn't a prediction you can test. Don Norman's 1983 observations still describe it well: the models people build are incomplete, unstable and partly superstitious, and they don't match how the system works. The research that does exist points the same way. David Kieras and Susan Bovair found in 1984 that people taught how a device worked learned its procedures faster, remembered them better and found shortcuts more often than people taught by rote. Stephen Payne's 1991 interviews about cash machines found that people build their own explanations, from direct experience and borrowed analogies.
Sources
- Craik, K. J. W. (1943). The Nature of Explanation. Cambridge University Press.
- Norman, D. A. (1983). Some observations on mental models. In Gentner & Stevens (Eds.), Mental Models. Erlbaum.
- Kieras, D. E., & Bovair, S. (1984). The role of a mental model in learning to operate a device. Cognitive Science, 8(3).
- Payne, S. J. (1991). A descriptive study of mental models. Behaviour & Information Technology, 10(1).
Last reviewed November 7, 2026. How we keep this current