How does "Identify which features are actually diagnostic" work?

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 does "Recognize that random sequences don’t "owe" balance" work?

The gambler’s fallacy is a direct product of representativeness: a long run of heads doesn’t look like a "random" sequence, so the mind expects a correction. But the coin has no memory. The representativeness of the whole sequence is irrelevant to the probability of the next flip. The fallacy appears everywhere from roulette tables to hiring decisions (avoiding a third female hire because "we’ve already taken two"), and the mechanism is always the same: the expected sequence looks more representative, so it feels more probable.

How does "Treat small samples with explicit skepticism" work?

The representativeness heuristic is insensitive to sample size: people judge the probability of a sequence of results with 4 observations the same way they judge one with 40. But the smaller the sample, the higher the variance — extreme or patterned results are far more likely by chance. Recognizing that small samples are unreliable corrects the tendency to read meaning into early results and make premature generalizations.

How does "Audit person-judgments for prototype substitution" work?

The representativeness heuristic underpins stereotyping: a person is assigned the traits of the group they resemble, regardless of individual evidence. This is cognitively efficient — group base rates are often real — but it misfires when group variance is high, when the individual is atypical, or when the "group" is a cultural stereotype rather than an empirical category. Deliberately auditing for prototype substitution ("Am I seeing this person or their template?") activates individuating-information processing.

How does "Expect extreme results to regress toward average" work?

Regression to the mean is a mathematical inevitability whenever performance involves both skill and chance. The representativeness heuristic causes people to ignore it: they explain extreme outcomes as if they fully reflect the underlying reality ("they’ve turned a corner" / "this person is hopeless"), rather than partly reflecting chance fluctuation. This produces the famous intervention illusion: harsh feedback after a bad performance "works" and praise after a good one "hurts" — simply because extreme outcomes regress, regardless of the intervention.

How does "Name the prototype you’re comparing against before deciding" work?

Much representativeness reasoning is invisible: the mind silently matches a candidate, product, or plan to a stored prototype and outputs an evaluation without surfacing the template used. By naming the prototype explicitly ("I’m comparing this business idea to the prototypical tech startup"), you make the comparison visible for scrutiny — you can then ask whether the prototype is appropriate, whether the match is real, and what the base rate of success is for that category.

Kahneman & Tversky

What is the representativeness heuristic and when does it lead judgment astray?

When similarity hijacks probability — and how to correct the substitution

Short answer

The representativeness heuristic is the mental shortcut of judging probability by how closely something resembles a prototype or stereotype. It is fast and often useful, but it reliably misfires when it overrides base rates, produces the conjunction fallacy, or treats random-looking sequences as unlikely.

Kahneman and Tversky identified the representativeness heuristic as one of the most consequential shortcuts in human judgment: we assess probability by asking "how much does this look like a member of that category?" rather than "how often does this actually occur?" The heuristic is not always wrong — resemblance often correlates with membership — but it fails systematically when base rates are ignored, when sample sizes are treated as irrelevant, or when a coherent narrative obscures the underlying frequencies. The practices below address the specific failure modes, each with the cognitive mechanism responsible.

The practices (7)

Why it works

Representativeness substitutes "how typical is this?" for "how probable is this?" — an easier question that feels like the right one. The substitution goes unnoticed because the resemblance judgment is computed automatically and feels like evidence. Anchoring on a base rate first forces the harder question to the front, making it much more likely that probability judgment is correctly calibrated rather than distorted by surface similarity.

How to do it

1Name the category you’re about to assign something to ("this person is probably a doctor").
2Before using resemblance as evidence, ask: "What fraction of people in the context I encountered this person are doctors?"
3Write down that base rate as your starting probability.
4Update for specific observed features only if those features are genuinely more common in the category than outside it.

Evidence

Observational

Kahneman and Tversky’s original 1973 research on the engineer/lawyer problem showed that subjects ignored explicit base rate information (70/30 split) when they had a representative description. Many follow-up studies replicate base-rate neglect, though instruction on base rates can partially correct it.

Honest caveat: Base-rate neglect is stronger when the description is vivid and individuating; it reduces (but doesn’t vanish) when no description is given or when subjects are prompted to think like statisticians.

— Kahneman & Tversky (1973), On the psychology of prediction, Psychological Review

Common mistake: Treating resemblance as only one factor while still weighting it heavily — if you thought of the base rate at all but still landed on the representative answer, the heuristic won.

Go deeper on this practice →

Practice this with IX Coach

Reading about a practice changes nothing on its own. IX Coach turns these into a guided, adaptive routine — discerning where you are in real time and walking the practice with you, session after session.

Start with IX Coach →