Interpretive habit is strengthened by cognitive bias modification.

We investigated the nature of the memory mechanisms underlying cognitive bias modification by applying Jacoby's (1991) process-dissociation procedure to responses during the transfer task. In the two training conditions (negative and benign), students imagined themselves in 100 ambiguous scenarios, most with potentially negative resolutions; the ambiguity was resolved in a consistently negative or benign direction by completing the fragment of a final word. Control participants completed non-ambiguous, non-emotional scenarios. Next all participants responded on a final training block, where half of the scenarios were completed negatively and half benignly. Transfer was assessed by examining choices in the completion of test scenarios when participants were instructed to respond in the same way as they had to a final-block training scenario that was situationally similar. Benign training facilitated correct responding to benign analogues and impaired correct responding to negative analogues. Performance in the negative-training and control conditions was similar. Process-dissociation procedures revealed that this newly established habit and not controlled recollection provided the basis for transfer.