Article

In a pair of recent studies, frontally distributed event-related potential (ERP)
indices of two distinct post-retrieval processes were identified. It has been
proposed that one of these processes operates over any kinds of task relevant
information in service of task demands, while the other operates selectively over
recovered contextual (episodic) information. The experiment described here was
designed to test this account, by requiring retrieval of different kinds of
contextual information to that required in previous relevant studies.
Participants heard words spoken in either a male or female voice at study and
ERPs were acquired at test where all words were presented visually. Half of the
test words had been spoken at study. Participants first made an old/new judgment,
distinguishing via key press between studied and unstudied words. For words
judged 'old', participants indicated the voice in which the word had been spoken
at study, and their confidence (high/low) in the voice judgment. There was
evidence for only one of the two frontal old/new effects that had been identified
in the previous studies. One possibility is that the ERP effect in previous
studies that was tied specifically to recollection reflects processes operating
over only some kinds of contextual information. An alternative is that the index
reflects processes that are engaged primarily when there are few contextual
features that distinguish between studied stimuli.
CI - Copyright (c) 2011 Elsevier Ltd. All rights reserved.

Langue : ANGLAIS