THE SIMILARITY OF BRAIN ACTIVITY ASSOCIATED WITH TRUE AND FALSE RECOGNITION MEMORY DEPENDS ON TEST FORMAT

Event-related potentials (ERPs) were compared for correct recognitions of previously presented words and false recognitions of associatively related, nonpresented words (lures). When the test items were present ed blocked by test type (old new, lure), waveforms for old and lure it ems were different, especially at frontal and left parietal electrode sites, consistent with preview position emission tomography (PET) data (Schacter, Reiman, et al., 1996). When the test format randomly inter mixed the types of items, waveforms for old and hire items were more s imilar. We suggest that test format affects the type of processing sub jects engage in, consistent with expectations from the source-monitori ng framework (Johnson, Hashtroudi, & Lindsay, 1993). These results als o indicate that brain activity as assessed by neuroimaging designs req uiring blocked presentation of trials (e.g., PET) do not necessarily r eflect the brain activity that occurs in cognitive-behavioral paradigm s, in which types of test trials are typically intermixed.