Abstracts 2005 - The Psychonomic Society

Friday Noon Posters 2049–2055
studied and 15 were not. Each test item appeared for 30 or 50 msec,
in between 100-msec pre- and postmasks. Skin conductance responses
(SCRs) were measured at test stimulus onset. Following stimulus offset,
participants were asked to rate their confidence that the quickly
flashed item was studied, using a scale of 0 (definitely not studied) to
10 (definitely studied). Then they were asked to identify the word.
Among items that went unidentified, both the SCR latencies and the
recognition ratings discriminated between studied and nonstudied
items.
(2049)
Source Memory in the Absence of Successful Cued Recall.
GABRIEL I. COOK, Claremont McKenna College, RICHARD L.
MARSH, University of Georgia, & JASON L. HICKS, Louisiana
State University—The authors used a paired associate learning paradigm
in which cue–target word pairs were studied and target recall
was requested in the presence of the cue. When target recall failed,
participants were asked to make a predictive source judgment of whether
a male or a female had spoken the item. In general, predictive accuracy
was at or very close to chance. By contrast, if cue–target pairs
were studied multiple times or participants knew in advance of learning
that a predictive judgment would be required, predictive source
accuracy was well above chance. These data suggest that context information
may not play a very large role in metacognitive judgments,
such as feeling-of-knowing ratings or putting one into a tip-of-thetongue
state, without strong and specific encoding procedures.
(2050)
The Effect of Wishful Thinking on Source-Monitoring Ability in
Older and Younger Adults. RUTHANNA GORDON, Illinois Institute
of Technology, & MARA MATHER, University of California,
Santa Cruz—When making source-monitoring decisions, people are
often biased to misattribute desirable claims to reliable sources and
undesirable claims to unreliable sources (Gordon, Franklin, & Beck,
2005). This “wishful thinking effect” is one way that emotional investment
leads to biased reasoning. Emotional goals may play a
greater role in memory attribution processes as we age (e.g., Mather
& Johnson, 2000). We compared younger and older adult sourcemonitoring
abilities when emotional preferences were present. Participants
read desirable and undesirable predictions made by reliable
and unreliable predictors. Later, they were asked to say, for each prediction,
which source had originally produced it or whether it was
new. Both populations showed a wishful thinking effect; however, the
effect size was larger for older than for younger adults. Increases in
this type of error with age have implications for both information
evaluation processes and subsequent decision making.
(2051)
Metamemory Monitoring and Parkinson’s Disease. CELINE
SOUCHAY, University of Leeds, MICHEL ISINGRINI, University of
Tours, & ROGER GIL, University of Poitiers (sponsored by Martin A.
Conway)—Metamemory monitoring (i.e., subjects’ ability to predict
their memory performance) in Parkinson’s disease was assessed with
both before-study prediction (global prediction), reflecting subjects’
beliefs concerning their memory, and after-study prediction (feeling
of knowing), reflecting subjects’ ability to monitor their memory performance.
Subjects were given cued recall and recognition tests of 20
cue–target words. The global prediction consists of asking participants,
prior to the study phase, to predict the number of words they
could recall. The feeling-of-knowing prediction consists of asking
participants, during the recall phase, to predict their capacity to recognize
on a subsequent recognition test the nonrecalled words. Prediction
accuracy was measured by comparing the prediction with the
memory performance. The results showed that for Parkinson patients,
feeling-of-knowing predictions are inaccurate at predicting future performance,
whereas global predictions are accurate at predicting future
performance. These results suggest that Parkinson patients are particularly
impaired in monitoring their memory performance.
77
• ATTENTION •
(2052)
The Role of Search Mode in Dimension Weighting on Visual Feature
Search. TAKATSUNE KUMADA, National Institute of Advanced
Industrial Science and Technology—In a visual feature search
task, it is known that reaction times to a singleton target were shorter
when the defining features of targets were known in advance than
when they were unknown (Müller, Heller, & Ziegler, 1985). However,
the prior-knowledge effect of targets was eliminated when a target discrimination
task was performed, instead of a simple detection task
(Kumada, 2001). In this study, I examined whether the priorknowledge
effect in a target discrimination task was affected by a participant’s
search mode (Bacon & Egeth, 1994). When participants performed
a target discrimination task with a singleton detection mode,
no prior-knowledge effect was found, consistent with a previous
study. However, when the task was performed with a feature search
mode, the prior-knowledge effect was found. This suggests that the dimension
weighting of a target-defining feature was modulated by the
participant’s mode of visual search. I discuss implications of these effects
for the mechanism of top-down control of visual feature search.
(2053)
Early Selection in Visual Search: Evidence From Event-Related
Potentials. CLAYTON HICKEY & JOHN J. MCDONALD, Simon
Fraser University, ROBERT W. TAIT, University of Northern British
Columbia, & VINCENT DI LOLLO, Simon Fraser University—Electrophysiological
research has shown that attentional selection in visual
search can occur within 200–300 msec of display onset. Earlier eventrelated
potential effects have been observed but have been attributed
to methodological factors, such as imbalances in sensory activity arising
from the use of asymmetrical search displays. Here, we investigate
whether selection in visual search can occur at latencies shorter
than 200 msec when sensory imbalances are minimized by using stimuli
that are isoluminant with the background. When an isoluminant
stimulus served as the target, an enhanced P1 component was found,
showing that selection began within the first 100 msec. When the
same stimulus served as a distractor, no such P1 effect was found, indicating
that the P1 difference was unrelated to display imbalances.
These results allow for the possibility that early, feedforward signaling
in the visual cortex can be influenced by attention.
(2054)
Methods for Solving the Serial/Parallel Issue in Visual Attention.
THOMAS L. THORNTON & DAVID L. GILDEN, University of Texas,
Austin (sponsored by David L. Gilden)—In this work, we show how
the serial/parallel issue in visual attention may be decided, using an
improved variant of search based on multiple targets (van der Heijden,
La Heij, & Boer, 1983). Data from an ensemble of 29 search tasks are
analyzed within a novel computational framework where sequential
sampling models with parallel and serial architectures compete to explain
observed patterns of search speed and error. The method and
analysis are extremely powerful and provide (1) a continuous ordering
of experiments in terms of attentional limitation and (2) a principled
categorization of tasks into serial and parallel classes on the basis
of likelihood. Ensemble analyses reveal that the majority of search
tasks are best explained using a parallel, limited-capacity architecture.
These tasks are contrasted with an isolated subset of searches based
on rotation direction and spatial configuration, which are better explained
as serial processes.
(2055)
New Objects Can Capture Attention Without a Unique Transient.
CHRISTOPHER C. DAVOLI & RICHARD A. ABRAMS, Washington
University—Recent findings suggest that new objects may capture
attention only when they are accompanied by a unique transient. To
test this, we inserted a pattern mask between a placeholder display and
a search display that contained a new object, so that the new object