Dissociated process for implicit and explicit recognition of novel events

We studied recognition memory process for contextually novel events by measuring event-related brain potential (ERP). Brain electrical activity was recorded while 13 healthy subjects watched frequent triangles, infrequent slanted triangles, and infrequent novel pictures. The subjects were instructed to press a button to the occurrence of infrequent slanted triangles while ignoring other stimuli. Following the oddball detection task, the subjects performed a recognition test for the novel pictures that presented (old) or not (new) during the previous session. ERPs for the recognition test were averaged separately according to the correct and false recognition of novel pictures. We examined the differences between the ERPs for correct recognition of new pictures and those for false recognition of old pictures. Moreover, the differences between the ERPs for correct recognition of old pictures and the ERPs for false recognition of old pictures were examined. The correct recognition rate of pictures was 64%. There were significant differences in the ERPs to correctly recognized new pictures and falsely recognized old pictures in the latency range of 300-600 ms after stimulus onset. The difference wave was distributed in the bilateral inferior frontal lobes. The ERP differences between correctly and falsely recognized old pictures were maximal in the bilateral inferior temporal lobes. These results suggest that the inferior frontal lobe may contribute to the unconscious or implicit recognition process of the contextually novel events.