Every mental model you hold was built from a sample — and samples have edges you stopped noticing the moment the model started working. Charles Sanders Peirce called this 'the fixation of belief': the moment a method of inquiry succeeds, it tends to calcify into a habit that actively resists the conditions that would falsify it. The model doesn't fail loudly. It just quietly determines which inputs count as evidence and which get filed as noise. For a product thinker, this means the mental model governing your most confident domain — how users decide, how teams align, what 'good metrics' look like — is probably built from the period when you were still genuinely uncertain. The uncertainty made you look hard. The success made you look less. Peirce's fix wasn't perpetual doubt; it was deliberately re-exposing your settled beliefs to the class of cases they were never tested against. Today, pick the mental model you'd least want a new colleague to challenge — and go find one case where it gave the wrong answer.

Think of the mental model you rely on most in a crunch — the one you reach for automatically. Can you name the last situation where it produced a prediction that didn't land? If you can't, is that because the model is flawless, or because you stopped tracking its misses?

Drawing from American Pragmatism — Charles Sanders Peirce — 'The Fixation of Belief' (1877) and 'How to Make Our Ideas Clear' (1878)