Glossary

Take the Best Heuristic

Published 1 min read

What is the take the best heuristic?

The take the best heuristic compares two alternatives by checking cues in order of predictive validity. It stops at the first cue that distinguishes them and uses that cue to make the inference.

How it works

Gigerenzer and Goldstein's take-the-best model skips a cue when it does not distinguish the alternatives. The search can therefore inspect several cues, even though one determines the answer. Lower-ranked cues do not outweigh the deciding cue.

Example

Gigerenzer and Goldstein’s German-city analysis included cues such as national-capital status, having been an exposition site, and having a major-league soccer team, in that order of predictive validity. Imagine comparing two recognized cities, A and B, using those cues:

  1. National capital? Neither city is the capital, so this cue does not distinguish them.
  2. Exposition site? City A has been one and City B has not. This cue distinguishes them, so infer that A has the larger population.
  3. Stop. Do not check the soccer-team cue or combine lower-ranked cues to overturn the answer.

The city labels and cue values here are hypothetical; the cue types come from the original study. “Best” means the highest-ranked available cue that distinguishes this pair, not a favorite attribute or a guarantee of the correct answer.

Why it matters

A decision rule can be selective without being arbitrary. Its accuracy depends on the task, cues, and cue order. Good performance in a simulation does not establish that everyone uses the rule or that it suits every choice.

Sources and further reading