Users do not experience your interface — they experience their idea of your interface. A mental model is the internal representation a person builds of how a system works, based on prior experience, observation, and inference. Don Norman popularised this concept in "The Design of Everyday Things" (1988), distinguishing between the designer's mental model (how the system actually works), the user's mental model (how the user thinks it works), and the system image (what the interface communicates). When these three align, interaction is intuitive. When they diverge, errors are inevitable.
The Psychology
The construction of mental models is a continuous process of inference from feedback. Users develop a model of your product the first time they encounter it, update it with each subsequent interaction, and apply it to predict what will happen when they take an action. A well-constructed mental model allows users to extend their knowledge to new parts of a product they have never seen before — because the model generalises. A poorly constructed one requires users to treat each new screen as an unfamiliar situation, rebuilding their understanding from scratch. This is the difference between a user who feels confident exploring an unfamiliar feature and one who feels anxious about breaking something.
In Product Design
Mental models are formed from prior experience as much as from the current product. A user who has spent years with Gmail brings a model of email that includes stars for flagging, labels for categorisation, and a conversation thread view for related messages. A new email client that does these things differently must either align with this existing model or provide enough guidance to build a new one. The cost of building a new model is significant. Users will misuse unfamiliar systems by applying their existing model, generate errors, interpret the results through their existing model (which may interpret the error as confirmation), and develop learned helplessness about the product's reliability.
How to Apply It
The design practice of mental model alignment starts with user research. Ask users to describe how they think a process works before you design the interface for it. Their answers will reveal the model they will bring — including its gaps and misalignments with the actual system logic. Where the user model and system model diverge significantly, you face a choice: redesign the system to match the user model, or invest heavily in onboarding and documentation to update the user model. The former is almost always preferable. An interface that works the way users already expect it to work requires no explanation.
Why It Matters
When mental models are ignored, error rates are high and support ticket volume is disproportionate to the interface's apparent complexity. Users who say "I thought clicking this would do X" are reporting a mental model mismatch. These reports are the most valuable data in UX research because they pinpoint exactly where the interface's system image has failed to communicate how the system actually works. The goal is not just that users can learn to use a product — it is that using the product correctly is the natural consequence of applying the mental model the interface establishes.
