Abstracts 2005 - The Psychonomic Society

Posters 2028–2034 Friday Noon
(2028)
The Importance of Preexisting Long-Term Memory Representations
in Directed Forgetting. KIRSTEN BURGHARDT & FABRICE
B. PARMENTIER, University of Plymouth—In a directed forgetting task
using verbal stimuli, Oberauer (2001) showed that participants are
slower to reject to-be-forgotten items than new probes. This “intrusion
cost” was attributed to a slowly decaying residual activation of longterm
memory traces corresponding to the to-be-forgotten stimuli. The
present study tested this assumption by using spatial stimuli for which
participants had no existing memory representations and the same
spatial stimuli when participants had developed transient representations
of these items. The results showed that the intrusion cost was
mediated by the existence of stable memory representations for the
stimuli: Participants unfamiliar with the stimuli showed no distinction
between to-be-forgotten and new probes, whereas participants trained
to learn the set of locations showed the intrusion cost. This finding is
discussed with regard to current models of memory.
• FEEDBACK AND MEMORY •
(2029)
The Testing Effect: Enhanced Retention or Attenuated Forgetting?
SHANA K. CARPENTER, HAROLD PASHLER, & JOHN T. WIXTED,
University of California, San Diego—The present study investigated
the effect of an intervening memory test versus a restudy opportunity
on the rate of forgetting of obscure facts. After encoding 60 facts, participants
completed a cued-recall test with feedback over half of the
facts and received an additional study presentation over the other half.
A final cued-recall test over all of the facts was given after 5 min,
1 day, 2 days, 7 days, 14 days, or 42 days. At each of the six intervals,
the proportion of items retained from test versus restudy conditions
was calculated, and power functions were fit to the data to describe
the rates of forgetting following test versus restudy trials. Results help
to answer the question of whether the typical advantage in retention
for tested over restudied items is due to enhanced retention or attenuated
forgetting for tested items relative to restudied items.
(2030)
Successfully Applying the Testing Effect in a College Classroom.
KAREN S. CLIFTON, DAVID L. TRUMPOWER, & STEVEN P.
MEWALDT, Marshall University—Repeated testing has been found
to improve recall of a list of words and of simulated classroom material
in laboratory settings. The present study evaluated the effect of
prior testing on subsequent performance on unit exams in eight introductory
psychology classes across 12 weeks of class. Approximately
halfway through a 4-week unit and near the end of a unit, students
took an online quiz consisting of fill in the blank type multiple
choice questions that were easy or relatively difficult; in a control condition,
subjects read statements with the blanks filled in, or they received
no practice. Both easy and hard quiz items significantly improved
performance on the unit exams, when compared with the
control items.
(2031)
Can Controlled Processes Improve Performance in the Misinformation
Paradigm? JOHN B. BULEVICH, Washington University (sponsored
by Mitchell Sommers)—Misleading postevent information
often produces erroneous recollection of memory for an original event
(Loftus, Miller, & Burns, 1978). Furthermore, susceptibility to misinformation
is modulated by age of the participant, with older adults
demonstrating poorer memory for the original event after misinformation
than younger adults (Mitchell, Johnson, & Mather, 2002). The
present study examined whether subjective experience and cognitive
control could reduce age differences in misinformation susceptibility.
In three experiments, participants viewed an original event and received
misinformation. Utilizing a multiple test procedure (Koriat &
Goldsmith, 1996), participants received a forced report test with con-
74
fidence ratings followed by a free report test where they were allowed
to pass. Retrieval demands and performance incentives were manipulated
across experiments. The results demonstrate that both older and
younger adults can improve their performance when allowed to pass
and that older adult’s deficits are largely driven by miscalibration of
their subjective experience.
(2032)
Suspicious Minds Reduce the Postidentification Feedback Effect.
JEFFREY S. NEUSCHATZ, University of Alabama, Huntsville,
MICHAEL P. TOGLIA, SUNY, Cortland, CHARLES A. GOODSELL,
DEAH S. LAWSON, & JESSICA K. SWANNER, University of Alabama,
Huntsville, & JOSEPH S. NEUSCHATZ, Roger Williams University—We
examined techniques designed to diminish the postidentification
feedback effect. Participants viewed a video event and then
were asked to identify a suspect from a target-absent photo lineup.
Following their identification, some participants were told that their
selection was correct, whereas others were not given feedback concerning
the accuracy of their identification. Some participants who received
confirming feedback were also given reasons to entertain suspicion
regarding the motives of the lineup administrator, either
immediately (Experiment 1) or after a 1-week retention interval (Experiment
2). Suspicious subjects did not demonstrate confidence inflation
effects typically associated with confirming postidentification
feedback. In Experiment 3, the confidence prophylactic effect was
tested both immediately and after a 1-week delay. The impact of the
confidence prophylactic procedure varied with retention interval.
That is, it eliminated the effects of postidentification feedback immediately,
but not after the retention interval. Both theoretical and
practical implications of the results are discussed.
(2033)
Retaining Versus Replacing Incorrect Choices in Multichoice Items:
Implications for Learning. DOMINIC A. SIMON, New Mexico State
University—A feature of multiple-choice test items is that the incorrect
responses may become associated with the question and later may
be wrongly thought to be the correct answer, particularly when endorsed
for items for which the correct answer was not known a priori.
This possibility was explored in an experiment in which participants
(n = 77) saw 60 multiple-choice general information items. The questions
were presented in random order over four rounds. If correctly answered
twice, an item was dropped from the pool. Items that were apparently
not known a priori were treated in two ways across subsequent
rounds: The original alternatives were retained, or the wrongly endorsed
choice was replaced. Contrary to prediction, replacement/
retention of incorrect alternatives did not affect the probability of correct
responding on a test of all the original items following a filled
delay ( p = .093). Number of presentations was related to test performance
( p < .001).
(2034)
Feedback Benefits Correct Responses Made With Low Confidence.
ANDREW C. BUTLER & HENRY L. ROEDIGER III, Washington
University (sponsored by Jason M. Watson)—The level of confidence
in a response has a substantial effect on the perception and use of feedback.
For example, high-confidence incorrect responses lead to longer
feedback study times and tend to be “hyper-corrected” on subsequent
tests. Although most contemporary investigations of verbal learning
focus on the role of feedback as a mechanism for error correction, the
present research sought to demonstrate the importance of feedback on
the retention of correct responses, especially those made with low confidence.
Participants took an initial four-alternative multiple-choice
test on general knowledge facts, indicating their level of confidence
after each response. Feedback was given for half of the responses.
Learning was assessed on a delayed (5 min) cued recall test. The results
indicate that feedback provides substantial benefit to the retention
of correct responses, especially those made with low confidence.