Abstracts 2005 - The Psychonomic Society

Posters 5016–5022 Saturday Evening
fect of long-term learning (over 1 year). The results showed that the
effects of even slight word learning were accumulated unconsciously.
Furthermore, individual differences were clearly described by the
progress of achievement for each student.
(5016)
Examining the Applicability of a New Method for Measuring the
Long-Lasting Effects of Learning. TETSUYA YOSHIDA, Tokoha
Gakuen University, TAKAFUMI TERASAWA & ATSUSHI KATSUBE,
Okayama University, & NOBUO OHTA, Tokyo University of Social
Welfare (sponsored by Nobuo Ohta)—This study shows that a new experimental
design, which we call the micro-step method, is widely applicable
for measuring the effects of learning with various study materials.
This experimental design was developed to measure minutely
the effects of relatively long-lasting repeated learning, for weeks or
months, by controlling various timings in the presentation of study
materials. Yoshida et al. (2003) examined the long-lasting effects of
learning second-language (English) word acquisition by Japanese senior
high school students for 6 months, using the micro-step method.
This study examines the same effects of learning for the acquisition
of scientific terms (terms in psychological domain) by Japanese undergraduates.
The expected result is that the main effects of learning
will certainly be accomplished and that individual differences will be
large.
(5017)
Training Effects on Implicit Stereotype Bias in the Decision to Shoot.
KIMBERLY M. FENN & BERND WITTENBRINK, University of
Chicago (sponsored by Howard C. Nusbaum)—Stereotypes affect the
way we act toward others, and in the extreme, holding negative stereotypes
can lead to violent outcomes. In a video game setting, participants
were quicker to press one button to “shoot” an African-American target
and to press a different button to “not shoot” a Caucasian target
(Correll et al., 2002). However, real guns have a single trigger, and
choice decisions are different from go/no-go decisions. Does the
racial shooter bias occur with a more realistic response? With a go/nogo
task, participants are both quicker and more accurate in shooting
African-American targets than Caucasian targets. Does the probabilistic
association of race and gun change this shooter bias? Stereotypic
or counterstereotypic training resulted in a significant change in
the size and direction of the shooter bias. Counterstereotypic target
distributions reduce shooter bias; stereotypic training increases bias.
Training can modify implicit stereotype effects and may ameliorate
shooter bias.
(5018)
Causal Learning: Cue Competition With Two- Versus Three-Cue
Compounds. LEYRE CASTRO & EDWARD A. WASSERMAN, University
of Iowa—Pairings of Cue A with an outcome, either before or
after pairings of Compound Cue AB with the same outcome, lead to
weakened responding to Cue B in testing. Our experiments consider
the case in which the compound comprises three cues (i.e., ABC) instead
of two, a situation that poses interesting questions: Will judgments
to both Elemental Cues B and C decrease when Cue A is paired
with the outcome? Will judgments to Compound Cue BC also decrease?
What about Compound Cues AB and AC? In general, participants’
judgments reflected the value of the elemental cues. Moreover,
low judgments to Elemental Cues B and C were observed when the
elemental phase preceded the compound phase, but not when the order
was reversed. These results raise difficulties for both associative and
statistical accounts of causal learning.
(5019)
Word Associations are Formed Incidentally During Sentential Semantic
Integration: A Behavioral and Electrophysiological Study.
ANAT PRIOR-UNGER, Carnegie Mellon University, & SHLOMO
BENTIN, Hebrew University of Jerusalem—We explored the manner
in which sentential context facilitates the incidental formation of word
126
associations. In Experiment 1, unrelated word pairs were embedded
in either coherent or semantically anomalous sentences, including either
local or global anomalies. At study, participants performed a sentence
categorization task. The strength of the incidental associations
formed between sentence constituents was probed by assessing their
associative priming effect in a subsequent unexpected explicit recognition
test for single words. Significant priming was found for noun
pairs studied in coherent sentences, but not for those appearing in
anomalous sentences, regardless of anomaly type. Experiment 2 used
the N400 component to further investigate the difference between the
types of semantic anomaly. Global anomalies were found to disrupt
online word integration more than did local anomalies, but both types
of anomalies were equally detrimental to whole-sentence integration.
Thus, sentence constituents may be incidentally associated during
sentence processing, particularly as a result of whole-sentence integration
and mental model consolidation.
(5020)
Memory-Contingent Patterns of Brain Activity During Integrative
and Isolated Encoding of Word Pairs. BRIAN G. HOWLAND & IRA
S. FISCHLER, University of Florida (sponsored by Ira S. Fischler)—
College students were shown object–location noun pairs (e.g., elephant–
library; apple–beach) and were asked either to image each pair in a
single common event or scene or to image each named object separately.
Words in a pair were shown individually for 3 sec each, while
EEG was recorded. Associative memory (recognition of intact vs. rematched
pairs), but not item memory, was better following integrative
encoding. Slow-wave potentials in the ERPs differentiated the two
tasks. Memory-contingent differences in the ERPs were found for
both associative and item memory, but the latency and topographic
pattern of these memory-contingent ERPs differed across the specific
encoding task and type of memory test. Results are discussed in terms
of the nature and functional anatomy of Baddeley’s “episodic buffer”
and the distinction between item and relational processing.
(5021)
The Associative Memory Basis of Cognitive Skill Learning: Adult
Age Differences. WILLIAM J. HOYER, JOHN CERELLA, & SERGE
V. ONYPER, Syracuse University—It has been established that “item
learning” underlies cognitive skill development and that there are substantial
age deficits in the rate of item learning. In a between-groups
design, we compared learning rates for the same set of items in skill
training (ST) and paired-associate training (PAT). We found main effects
due to condition (PAT items were acquired earlier) and to age
(older adults’ learning was delayed), but no condition � age interaction.
Thus, the age deficit in ST can be accounted for by the age
deficit in associative memory; no further explanation is needed. We
also analyzed fast and slow retrieves in ST and PAT and found that the
frequency of fast retrieves did not differ. The overall advantage of PAT
was due to the use of slow retrieves, which were absent in ST. We conjecture
that the skill algorithm displaces the recollective process in ST.
(5022)
Item Learning in Skill Training: Strategic or Automatic? BRANDY
BESSETTE-SYMONS, WILLIAM J. HOYER, & JOHN CERELLA,
Syracuse University (sponsored by William J. Hoyer)—Repeated items
are learned individually in the course of cognitive skill training, allowing
the skill algorithm to be bypassed. There is a question whether
this learning occurs automatically or requires a deliberate strategy. We
tested one implication of the latter: Repeats from a mixed list of repeated
and novel items should be learned less rapidly than repeats
from a pure list of all repeated items, because the strategy fit is only
partial. In three between-groups experiments, we found no condition
difference in learning rate for mixed and pure lists equated on interitem
lags. This was true for both short and long lists and for both younger
and older adults. We concluded that item learning proceeds autonomously,
driven by item repetition independently of list composition
or task-level context.