Abstract
To date most studies within the misinformation paradigm have used the visual presentation of a to-be-remembered event that is later tested verbally or visually. However, the well-established encoding specificity hypothesis predicts that congruence between encoding and test phases should lead to fewer memory errors. In Study 1, we examined the susceptibility to misinformation after encoding original information in 1 of 4 different formats: as a film, slides, and as a written or auditory narrative. All participants were tested verbally, and those who encoded original information pictorially (as a video or slides) were more likely to incorrectly accept verbally suggested information. This might be a consequence of encoding–retrieval format match. In Study 2, using either verbal or pictorial modality during encoding, postevent information, and test (fully crossed design), we partially supported the encoding–retrieval format match hypothesis; however, auditory presentation of original or postevent information modified the effect, showing that a memory trace created after auditory description was the strongest.