Validation of an Event-related Potential memory assessment procedure: Effects of incidental and intentional learning

The main aim of this study was to investigate whether an ERP-based memory assessment procedure (van Hooff et al., 1996) could be used to detect memories for items that (1) did not receive a behavioural recognition response and (2) were not intentionally learned. Participants were first required to study a short list of "neutral," semantically unrelated words that were subsequently tested for recognition. Next, participants studied a second, different word list, which again was tested for recognition (learned targets) in a following test. In this second test, the distractors consisted of the words that were studied prior to - and had to be recognized in - the first test (learned nontargets), words that were used as distractors in the first test (repeated nontargets), and words that had not been learned or presented before (new words). Targets and both types of nontargets were presented less frequently than the new words, so essentially this test had an oddball character. Results showed that learned targets as well as learned nontargets elicited a P300 component that was largely absent for the new words and for the repeated nontargets. These results seem to imply that successful use of the ERP-based memory assessment procedure did not rely on behavioural indications of recognition, but did depend on the level of contemplation during initial study/presentation.