When Temporal Contiguity Matters

Abstract

Whereas several recent studies have examined the role of various personal and situational variables as moderators of the predictive validity of implicit measures, the role played by methodological factors has attracted relatively less attention. A two-session study (N = 104) investigated the influence of temporal contiguity between measurement and criterion on the validities of two implicit measures, the SC-IAT and the IAT, and an explicit measure, for predicting the results of a rapid picture-choice task. Because temporal contiguity between the independent and dependent variables can be framed in terms of accessibility, it was expected to moderate the predictive validities of the implicit but not of the explicit measures. The hypothesis was confirmed: The contiguity moderated the predictive validities of both implicit measures for the rapid picture-choice task. This moderation effect did not occur for explicit measures. Results also demonstrated that the SC-IAT did not show robust predictive validity whereas the IAT had an incremental validity for self-reported behaviors. Results are discussed in terms of the role played by construct accessibility in the predictive validity of implicit measures.