| |
Abstract:
Event-related potential (ERP) studies of recognition memory
demonstrate enhanced positivity for correctly classified old items
compared to new items, maximal at left parietal sites between 400
and 900 msec post-stimulus. The left parietal effect is regarded as
a correlate of recollection, for example it is larger in amplitude
for items that attract accurate source judgements than for items
that do not. It is currently unclear, however, if recollection of
source is sufficient to elicit the left parietal effect. Two
experiments manipulated strategies employed during a recognition
memory test. Participants in both experiments completed a 'deep'
non-target encoding task, followed by a second target encoding task
which was 'deep' in experiment 1 and 'shallow' in experiment 2. A
recognition memory test required one response for target items, and
another for non-target or new items. Memory for targets was high in
experiment 1 and low in experiment 2. The rejection rate for
non-targets was high in both experiments, yet a left parietal
effect was only observed for non-targets in experiment 2. We argue
that the left parietal effect was observed for non-targets when
participants attended to this information in the absence of
reliable memory for targets. These findings suggest that source
information must be attended to and employed in a goal-directed
manner in order for the left parietal effect to be observed.
|
