Base-Rate Neglect

Base-rate neglect is a cognitive bias in which people ignore prior probabilities — the "base rates" — when making judgments under uncertainty, relying instead on individuating information. The classic demonstration is the Linda problem: subjects are given base-rate information about a population and then ignore it when evaluating a specific case. The bias is typically attributed to the representativeness heuristic — the tendency to judge probability by similarity to a prototype rather than by statistical logic. But from the perspective of ecological rationality, base-rate neglect may be a rational response to environments where individuating information is genuinely more diagnostic than population statistics. The question is not whether minds ignore base rates but whether the environment rewards or punishes doing so.

The numerical cognition literature suggests that base-rate neglect is amplified when base rates are presented in formats that do not match natural frequency representations. When statistics are presented as relative frequencies ("1 in 20") rather than as conditional probabilities ("5%"), base-rate neglect is reduced. This implies that the bias is not purely cognitive but is partially a framing effect — a consequence of how information is presented rather than of how minds process it.