Adapting Cutoffs to the Choice Environment: The Effects of Attribute Correlation and Reliability

Abstract

Consumers frequently simplify complex choices by setting attribute cutoffs, which are minimum acceptable levels that an alternative must possess to be considered further. We explore the extent to which consumers adapt these cutoffs to the choice environment. We show that, as the reliability of information about the attribute increases, consumers make more severe cutoffs (i.e., fewer attribute levels are acceptable). Further, positive correlations between attributes elicit more severe cutoffs than negative correlations do, and consumers' expectations about the choice outcome partially mediate this relationship. The format of correlational information also affects adaptation: consumers adapt their cutoffs when they are given direct information about the correlation, but not when they are allowed to infer the correlation from a set of alternatives. Overall, consumers appear to adapt to information about reliability and correlations, but they have difficulty assessing correlation from the choice environment.