Abstracts 2005 - The Psychonomic Society

Thursday Evening Posters 1036–1042
converged on either the dominant or the subordinate meaning of a
nonpresented critical theme word. Only younger adults and high
frontal functioning older adults benefited from the contextual information
provided by the dominant meaning conditions and showed reductions
in false memory susceptibility. However, when gist-based
processing of list items was made less accessible, as in the subordinate
meaning condition, all groups showed similar reductions in false
memory susceptibility. These results suggest that low frontal functioning
older adults encode item information but overrely on accessible
gist information.
(1036)
Hybrid False Recollection: Evidence From ERP. STEVEN HAMIL-
TON, University of Sussex, ROBYN E. HOLLIDAY, University of Kent,
& FAY JOHNSON & BRENDAN S. WEEKES, University of Sussex
(sponsored by Brendan S. Weekes)—Two experiments examined whether
semantic and phonological false memory effects are additive by using
hybrid lists of written words presented using the DRM paradigm. Hybrid
lists were constructed so that each critical lure was related in both
meaning and sound to studied words. In Experiment 1, 36 adults were
tested, and results showed a reliable false memory effect with the semantic,
phonological, and hybrid lists. However, there was no evidence
of an additive effect across lists. In Experiment 2, EEG recordings
were taken while participants performed the same tasks. The
results showed differential ERP patterns during recognition performance
across the hybrid and nonhybrid items. Of most interest was a
pattern of bilateral activation during false recollection for hybrid
items that was not observed during semantic and phonological false
recollection. We interpret the data in terms of models of recollection
and argue that hybrid lists generate a stronger sense of gist-based recollection
than do nonhybrid lists.
(1037)
“Hybrid” False Memories Following Primed Naming: Expectancy
at Encoding or Bias at Retrieval? PATRICK A. O’CONNOR, CHI-
SHING TSE, & JAMES H. NEELY, SUNY, Albany (sponsored by James
H. Neely)—False memories of nail in recognition are higher when the
semantically related prime HAMMER and the form-related target mail
have been incidentally studied together versus separately in a prior
naming task. This contiguity-based increase in semantic-form “hybrid”
false memories (HFMs) occurs at a 750-msec prime–target SOA
with a high, but not with a low, study phase relatedness proportion
(RP; Humphreys, Burt, & Lawrence, 2001). We find this contiguitybased
increase in HFMs with a high RP at both a 760-msec SOA (6%
increase) and a 175-msec SOA (10% increase), but not with a low RP
and a 175-msec SOA (1% decrease). Because the 175-msec SOA
should minimize expectancy at encoding, the large 10% contiguitybased
increase in HFMs obtained in the high-RP/175-msec SOA condition
favors a bias-at-retrieval account over an expectancy-at-encoding
account of why contiguity inflates semantic-form HFMs only when
the study phase RP is high.
(1038)
The Development of False Memories in Bilingual Children and
Adults. NADINE GAGNON, Lakehead University, MARK L. HOWE,
Lancaster University, & LISA PAQUETTE, Lakehead University
(sponsored by Mark L. Howe)—The effects of bilingual (English–
French) study and test on rates of children’s and adults’ true and false
memories were examined. Children aged 6 through 12 and universityaged
adults participated in a standard DRM false memory task using
free recall. The results showed that (1) both true and false memories
increased with age, (2) true recall was higher in unilingual than in
bilingual conditions for all ages, and (3) there were fewer false memories
in bilingual than in unilingual conditions for the youngest children,
no differences for the 7- and 12-year-olds, and among adults,
more false memories in bilingual than in unilingual conditions. These
findings are discussed in the context of language organization models
and bilingual memory development.
57
(1039)
The False Recall Effect Following Multitrial Serial Learning.
JOHN P. TAYLOR, KARA DESOUZA, SARA HABER, & WILLIAM
P. WALLACE, University of Nevada, Reno—Lists of associates to key
words produce large intrusion rates of the key words during later free
recall. Participants with a well-practiced retrieval plan for a learned
list should be able to exclude intrusions from recall. In the present experiment,
nine lists of the 12 most frequent associates to key words
(e.g., sour, candy, sugar, etc., for the key word sweet) were taken from
Roediger and McDermott (1995). Three lists were presented for one,
three, and eight trials each. Twenty-four participants read each word
aloud, and 24 attempted to anticipate the next word in the series. Lists
were cued for final recall by presenting the first word on each list. On
final recall, the proportion of key-word intrusions following the eight
trials of serial anticipation was .21. For all other conditions, values
ranged from .44 to .49. Imposing serial order on recall over eight trials
considerably reduced the intrusion rates commonly found with associate
lists.
(1040)
Congruency Effect of Presentation Modality on Haptic, Visual,
and Auditory False Memory. TOMOHIRO NABETA, Hiroshima
University, KEIJI OKADA, Aichi University, & JUN KAWAHARA,
Hiroshima University—False recognition of an item that is not presented
(the lure) can be produced when participants study and are then
tested on their recognition of a list of items related to the lure, which
is not included in the list. False recognition has been shown to be reduced
when the study and test modality are congruent (e.g., both visual)
rather than being different (e.g., visual study and auditory test).
The present study examined whether such a congruency effect occurs
for visual, haptic and auditory modalities. After studying items presented
haptically, visually or auditorily, participants took a recognition
test in one of the three modalities. The congruency between the
study and test modalities reduced false recognition when studied haptically
or visually. These results suggest that distinctive cues encoded
through the haptic and visual modality help the participants to recognize
the studied words more accurately, resulted in less false recognition.
(1041)
Memory Illusions and Repeated Testing in Children: The “Reversed”
DRM Effect. ROBYN E. HOLLIDAY, University of Kent, Canterbury,
TIMOTHY N. ODEGARD, University of Texas, Arlington, & CHARLES
J. BRAINERD & VALERIE F. REYNA, Cornell University—The
Deese/Roediger–McDermott (DRM) paradigm was used to investigate
the effects of an encoding manipulation (read versus generate)
and a retrieval manipulation (test repetition) on 10- to 12-year-olds’
true and false recognition. Children studied two blocks of four DRM
lists, one “read” (i.e., whole words) and the other “self-generated”
(word fragment completion). Each block was followed by either one
or three recognition tests. False recognition of critical distractors was
lower in the generation condition than in the read condition. These
false-memory responses were doubly dissociated by encoding condition
and test repetition: Repetition increased false alarms in the read
condition but decreased them in the generate condition. Such findings
are consistent with fuzzy-trace theory’s opponent processes account
of false memory.
(1042)
Dissociations in Children’s True and False Recall. MARK L. HOWE,
Lancaster University—Across three experiments, the role of associative
versus categorical relations in children’s veridical and false memories
was examined. Children studied DRM lists or categorized word
or picture lists, with or without category labels as primes. For veridical
recall: (1) There were developmental increases in recall regardless
of list type, presentation modality, or priming; (2) regardless of age,
categorized lists were better recalled than DRM lists, regardless of
modality or priming. For false recall: (1) There were developmental
increases in recall, except for pictures; (2) no differences were pres-