Definition
Base rate neglect is the systematic tendency to underweight (or entirely ignore) statistical base rates — the prior probability of an event based on general population data — in favor of specific, vivid, representative descriptions of the individual case.
It is one of the most consequential cognitive biases because it leads to systematically miscalibrated probability judgments, even in high-stakes domains like medicine, law, and investment.
The Mechanism
When evaluating the probability of something, we typically have access to two types of information:
- Base rate (or prior probability): the statistical frequency of this kind of event in the relevant reference class
- Case-specific information: details about this particular instance
Rational reasoning (and bayes-theorem explicitly) requires integrating both. In practice, wysiati causes the vivid, specific description to crowd out the statistical background entirely.
Classic Examples
The Librarian/Farmer Problem (Kahneman)
“Steve is very shy and withdrawn. He has a need for order and structure and a passion for detail.”
Is Steve more likely a librarian or a farmer?
Most people answer: librarian. The base rate fact — that there are roughly 20 farmers for every librarian in the US — is completely swamped by the representativeness of the description. Correct Bayesian reasoning would make farmer much more likely unless the description is extraordinarily discriminating.
The Conjunction Fallacy
“John is athletic and a vocal LGBTQ rights advocate. Is John (a) a basketball player, or (b) a basketball player who is gay?”
Most people choose (b). But P(A and B) ≤ P(A) always, without exception. This is a direct logical error caused by the representativeness heuristic overriding probabilistic reasoning.
The Medical Test Problem (Cameron)
From source—notes-on-probability: A diagnostic test is 95% accurate. The disease has a base rate of 0.1% in the population. You test positive. What’s the probability you actually have the disease?
Most people intuit ~95%. The correct answer via bayes-theorem: ~1.94%.
The low base rate (0.1%) means false positives vastly outnumber true positives, even with a highly accurate test. This example is the clearest mathematical demonstration of why base rates cannot be ignored.
Why It Happens
The representativeness heuristic is System 1’s shortcut: “How much does this resemble the category?” This is fast and often useful, but it systematically overrides base rates.
This is a specific instance of wysiati: the specific description is “what you see,” and the statistical background is invisible.
High-Stakes Consequences
- Medicine: Doctors order tests without adjusting for base rates, generating false positives that lead to unnecessary treatment and patient anxiety
- Criminal justice: Eyewitness confidence is taken as strong evidence without accounting for base rates of error in eyewitness identification
- Investment: “This company has great management and a compelling story” overrides the base rate that most startups fail
- Hiring: Compelling interview performance overrides base rates of interview-performance-to-job-performance correlation (which is weak)
The Mathematical Fix: Bayes’ Theorem
bayes-theorem is the formal solution to base rate neglect. It forces explicit integration of base rates (priors) with case-specific evidence (likelihoods):
P(hypothesis | evidence) = P(evidence | hypothesis) Ă— P(hypothesis) / P(evidence)
The prior P(hypothesis) is the base rate. Bayesian reasoning makes it impossible to ignore it — it’s a required input.
Connections
bayes-theorem
Bayes is the mathematical antidote: base rates are the prior probability P(H), which must be explicitly multiplied with the likelihood ratio. The medical test example quantifies exactly how large the error is when the prior is ignored.
wysiati
WYSIATI is the psychological mechanism: the vivid, specific description is “all there is,” crowding out the invisible statistical background.
probability-theory
Base rates are expressions of frequentist probability — the reference class matters enormously. The “inside view” (this specific case) must be balanced with the “outside view” (what happens to cases like this in general).
mental-models
charlie-munger lists “Fermat/Pascal probability math” as one of his most important mental models, and base rate reasoning is central. His “outside view” discipline is a direct antidote to base rate neglect.
anchoring-bias
Both anchoring and base rate neglect involve over-weighting salient available information at the expense of more relevant statistical information.