| |
Abstract:
Event-related potentials were recorded while 20 participants
completed a modified version of the Deese procedure (Deese, 1959).
In the Deese paradigm, participants are presented with lists of
semantically related words. In the test phase, participants are
asked to make recognition judgements for words initially presented
(old), never presented (new) and never presented but highly
semantically related (lures). There were two major modifications to
the Deese procedure. First, as in Johnson, Nolde, Mather, Kounios,
Schacter and Curran (1997), we divided participants into blocked
and random testing groups. Second, during the initial presentation,
for each list of semantically related words, half were read by a
male voice and half by a female voice. In the subsequent test
phase, when a word was judged "old", participants were asked if the
word was presented in a male or female voice. This source question
was not a part of the Johnson et al. (1997) study, yet differences
between blocked and random conditions emerged. This difference was
attributed to processing of source information in the blocked when
compared to random condition. The focus of the present study was to
examine if differences in the blocked and random condition would be
diminished if participants were required to make source judgements
in both conditions. Behavioral and electrophysiological results are
discussed in terms of a source monitoring framework.
|
