Error Management Theory

principle

Evolution can favor systematic false alarms when overlooking one real threat is costlier than repeatedly fleeing harmless ones. Some biases persist because they minimize the more consequential error, not because they maximize accuracy.

One ancestor fled 999 times from dangers that were not there. The skeptic stayed calm—and on the thousandth occasion was eaten.

E1

Survival rewards the cheaper mistake

Error Management Theory begins with two possible failures: reacting to a threat that is absent, or ignoring one that is present. When those errors carry radically different costs, selection does not optimize for an even-handed judgment; it favors a decision threshold tilted toward the less disastrous mistake. In the ancestor story, unnecessary flight wastes effort, while one missed predator ends the lineage. Repeated filtering through Natural Selection can therefore preserve a mind that over-detects danger. The resulting bias looks irrational if you count incorrect judgments, but less so if you weight each judgment by its possible cost.

E1

Where it shows up

The rustle that becomes a predator

The cautious ancestor treats ambiguous signals as threats and survives despite being wrong almost every time. The example reveals the model's central asymmetry: many cheap false positives can be preferable to one fatal false negative.

E1

The asymmetry has to be real

A bias is not automatically adaptive merely because it can be described as caution. The model applies when the two errors genuinely have unequal consequences; if false alarms become costly or missed threats become recoverable, the survival-weighted threshold may produce needless overreaction instead.

E1

Price both ways of being wrong

Before correcting an apparently biased decision, write down the false-positive and false-negative outcomes separately and estimate their costs. Set your threshold according to the more damaging miss rather than demanding equal certainty for both errors.

E1

Episodes that teach this