Mistakes — the intention rests on a wrong mental model

What it is

A mistake is a failure where execution matches the plan exactly, but the plan itself was wrong from the start — the user's belief about how the system works, or about what should be done right now, was mistaken, so even flawless execution doesn't produce the intended result. This is the mirror image of a slip: a slip is "the right idea, badly carried out"; a mistake is "the wrong idea, carried out perfectly."

Why it happens

A mistake can come from a wholesale wrong mental model (the user's belief about how the system works is simply incorrect — e.g., assuming a draft autosaves when it doesn't), or from applying an otherwise-correct rule to the wrong situation (the rule is valid somewhere, just not here, and the user failed to tell which rule this situation called for). Either way, the failure sits in the plan-formation stage; execution faithfully carries out that flawed plan, so the action itself shows nothing wrong — only checking against the system's actual behavior, or the rule's real scope, exposes where the plan went astray.

Studying it

Identifying a mistake requires reconstructing the user's mental model at the time, not just the action sequence — a common approach is to ask, right after the failure, "what did you expect to happen when you did that?" and compare the answer against what the system actually does; the gap between model and reality is the source of the mistake. The question has to target the expectation held at the time, not "what's the correct way to do it," or the user will answer with knowledge learned afterward and mask the mistaken belief they actually held.

Where it stops holding

If the user's mental model was actually correct and the failure was in recognizing the current state (say, not noticing a mode switch had happened), that isn't a mistake — the model itself was fine, the failure was in perceiving the situation. Mistake as a category only covers cases where the content of the model is wrong; a correct model applied in the wrong place belongs to a different error category.

Applying it

Once a failure is pinned down as a mistake rather than a slip, the fix has to correct the belief, not tighten the operation — making a button harder to press does nothing to a user's wrong expectation about what the system will do; next time they'll bring the same flawed plan and just find another way to enact it. The effective intervention surfaces what the system will actually do before the decision point (a preview, an explanation of the effect before confirmation), breaking the wrong model ahead of execution rather than flagging it after the fact. Verification: watch the same users after tightening the operation itself — if they keep trying workarounds to reach the same (mistakenly conceived) goal, the model hasn't been corrected and the problem remains.

Related

• Same group: A10.01.1 Slip — correct intention, deviated execution · A10.01.3 Violation — deviating on purpose despite knowing the rule · A10.01.4 Slips are fixed by interface constraints, mistakes by model correction — the remedies are not interchangeable · A10.01.5 Classification must precede any redesign · A10.01.6 Blaming "user carelessness" halts diagnosis · A10.01.7 The line between violation and mistake is whether the deviation was intentional
• Nearby: A10.02 Mode errors · A7.06 Identifying an incorrect mental model · A7.14 Diagnosing and correcting wrong mental models
• Search terms: mistake · rule-based error · mental model mismatch