Cue validity

Cue validity is the probability that a given cue — a feature, a sign, a piece of information — correctly predicts a criterion when it discriminates between alternatives. In the context of the take-the-best heuristic, cue validity is the ranking principle: cues are searched in order of their validity, and the search stops at the first cue that discriminates. A cue with high validity is one that, when it points to one alternative, is usually right. A cue with low validity is one that is unreliable even when it discriminates.

Cue validity is not a property of the cue alone. It is a property of the relationship between the cue and the criterion in a specific environment. A cue that is highly valid for predicting city population may be useless for predicting stock price. The ecological rationality of a heuristic depends on the match between the cue validities and the heuristic's search strategy. The ecological rationality program argues that the rationality of a decision strategy cannot be evaluated in the abstract; it must be evaluated relative to the cue structure of the environment in which it operates.

The concept of cue validity is central to the debate between the heuristics and biases program and the fast-and-frugal program. The heuristics-and-biases program treats cues as inputs to a general-purpose reasoning system; the fast-and-frugal program treats cues as tools matched to specific environments. The difference is not merely theoretical. It determines whether the appropriate response to a bias is to train the decision maker or to redesign the environment. If the problem is cue validity mismatch, the solution is not better reasoning but better cue structures — clearer signals, more salient valid cues, less noise from invalid ones. The cognitive engineering of decision environments begins with the analysis of cue validity, not with the diagnosis of cognitive failure.