Identify which features are actually diagnostic

Separate features that genuinely differentiate categories from ones that just complete the picture.

Why it works

When a description includes many features consistent with a prototype, people treat the coherence of the picture as evidence of membership — even when most of the features carry no discriminating information. A feature is diagnostic only if it is more common in the target category than in alternatives. Distinguishing decorative features (they fit the story but don’t discriminate) from diagnostic ones activates more careful, Bayesian reasoning.

How to do it

1. List the features of the case you’re evaluating.
2. For each feature, ask: "Is this more common in the category I’m considering than in others?"
3. Mark features as "decorative" if they fit the prototype but don’t discriminate.
4. Base your probability judgment only on the diagnostic subset.

Evidence

Research on the dilution effect (Nisbett et al., 1981) shows that adding irrelevant (non-diagnostic) information actually weakens judgment accuracy — people average it in instead of ignoring it. This directly supports separating diagnostic from decorative features. (observational)

The dilution effect itself is context-sensitive; the practical discipline of identifying diagnostic features is a corrective heuristic rather than a guaranteed debiasing procedure.

Sources

• Nisbett, Zukier & Lemley (1981), The dilution effect: Nondiagnostic information weakens the implications of diagnostic information, Cognitive Psychology

Common mistake

Treating a long consistent description as strong evidence — the number of consistent features doesn’t compound into higher probability; only discriminating features do.