fMRI evidence supporting the role of memory conflict in the déjà vu experience.

Attempts to generate déjà vu experimentally have largely focused on engineering partial familiarity for stimuli, relying on an ensuing, but unprompted evaluation of conflict to generate the experience. Without verification that experimentally-generated familiarity is accompanied by the awareness of stimulus novelty, these experimental procedures potentially provide an incomplete déjà vu analogue. We used a modified version of the Deese-Roediger-McDermott (DRM) false memory procedure to generate both familiarity and novelty within a déjà vu analogue - we coupled experimentally-generated familiarity with cues indicating that the familiarity was erroneous, using this additional source of mnemonic information to generate cognitive conflict in our participants. We collected fMRI and behavioural data from 21 participants, 16 of whom reported déjà vu. Using univariate contrasts we identified brain regions associated with mnemonic conflict, including the anterior cingulate cortex, medial prefrontal cortex and parietal cortex. This is the first experiment to image an analogue of the déjà vu experience in healthy volunteers. The increased likelihood of déjà vu reports to DRM critical lures correctly identified as "new", and the activation of neural substrates supporting the experience of cognitive conflict during déjà vu, suggest that the resolution of memory conflict may play an integral role in déjà vu.