A mistake, in James Reason's error taxonomy, is an error of planning — the user's intention was wrong because they misunderstood the situation, applied the wrong rule, or lacked the knowledge to choose correctly. Unlike slips (correct plan, botched execution) and lapses (correct plan, forgotten step), mistakes involve forming a wrong plan in the first place.
Reason further divides mistakes into:
- Rule-based mistakes: the user applied a rule that usually works but doesn't fit the current situation
- Knowledge-based mistakes: the user lacked the expertise to make the correct decision at all
Examples in healthcare software:
- A clinician orders a standard dose without realising the patient has renal impairment (rule-based)
- A junior doctor misinterprets lab results and orders the wrong test (knowledge-based)
Mistakes are the hardest error type to address through interface design alone. They require:
- Clear system feedback that reveals the user's misunderstanding
- Contextual information that reminds users of relevant constraints
- Clinical decision support that catches rule-based mistakes
- Training and expertise development
- Transparent reasoning so users can check their own logic
Where slips and lapses can often be engineered out of a system, mistakes require the user to have (or gain) the right mental model. Interfaces can support accurate mental models but cannot substitute for them.
Related terms: Slip, Lapse, Human Error, Mental Model, Clinical Decision Support
Discussed in:
- Chapter 10: Design Laws from Aviation and Engineering — Error Tolerance and Fail-Safe Design
Also defined in: Textbook of Usability