ROC curves show that the revelation effect is not a single phenomenon

The revelation effect describes the increased tendency to call itemsold when a recognition judgment is preceded by an incidental task. In theory, the effect could come about either from a more liberal response bias or from a change in underlying memory sensitivity. Using analyses of receiver-operating characteristic curves, we show that the revelation effect occurs for each of these reasons, but under different empirical conditions. A shift in response bias fully accounts for the revelation effect when revealed items are unrelated to the subsequent recognition probes. However, a change in memory sensitivity contributes to the effect when revealed items are identical to the recognition probes. Thus, the revelation effect encompasses at least two distinct phenomena.