Abstracts 2005 - The Psychonomic Society

Posters 4057–4063 Saturday Noon
(4057)
Feature Binding Within Visuospatial Short-Term Memory: Asymmetry
Between Shape and Location Encoding. JANE ELSLEY &
FABRICE B. PARMENTIER, University of Plymouth (sponsored by Fabrice
B. Parmentier)—Binding processes play an important role in memory.
Using a probe recognition task (Prabhakaran et al., 2000), we investigated
visual-to-spatial binding by presenting three irregular shapes
simultaneously in irregularly distributed locations, followed by a single
probe. Participants had to judge the probe in terms of both the visual
and the spatial features. In Experiment 1, robust binding effects
were observed. Two further experiments investigated whether binding
processes operate automatically. Results indicated that when participants
focused only on shapes (Experiment 2A), binding occurred automatically.
However, when participants focused on locations (Experiment 2B),
no binding effect was observed. Therefore, shape processing was not
independent of the processing of location, whereas location could be
attended to in isolation. This asymmetry in association is discussed
with regard to recent perceptual and attentional accounts of binding.
(4058)
Training on Relations Eliminates Interference of Elements on Relational
Processing. AARON S. YARLAS, Grand Valley State University,
& VLADIMIR M. SLOUTSKY, Ohio State University—Previous studies
(Yarlas & Sloutsky, 2004) incorporating a Garner interference task
with undergraduates showed clear evidence for asymmetry of interference
when elements and relations were processed: Status of elements
(i.e., objects in an arrangement) could not be ignored during relational
processing, whereas status of relations (i.e., patterns of
arrangements among objects) had no impact on the processing of elements.
The present study, in which the same interference task was
preceded by either training focusing on relations (e.g., labeling relations,
sorting by relations) or control training, yielded a much different
outcome, such that after relational training (although not after control
training), the interference of elements with relations disappeared,
and there was, in fact, an emergence of interference of relations with
elements. These findings provide strong evidence that distinctions between
element and relational processing are not fixed a priori but,
rather, are flexible and are likely a product of learning and/or directed
attention.
(4059)
Architecture and Retrieval Dynamics of Working Memory: Exploring
the Role of Pointer Predictability in Redirecting Attention.
CHANDRAMALLIKA BASAK & PAUL VERHAEGHEN, Syracuse
University—Working memory consists of a focus of attention and an
active outer store (Cowan, 1995). We conducted a series of four experiments
to explore the differential capacity and retrieval dynamics
of these two subsystems of working memory and whether a passive
memory load affects the retrieval speed or activation of these subsystems.
We found that when the retrieval sequence is predictable, the
focus switch cost was additive over increasing set size, but when it is
unpredictable, a controlled search slope of about 230 msec/set-size
emerges. Moreover, with extended practice, the focus can be expanded
from one to four in a predictable-sequence task, but not in an
unpredictable-sequence task. We propose that a pointer is needed to
redirect attention to the correct cell in the memory array; resource demands
on a task may depend upon whether the pointer moves predictably
or not.
(4060)
The Role of Attention During Retrieval in Working Memory Span:
A Dual-Task Study. M. KARL HEALEY, University of Toronto, &
AKIRA MIYAKE, University of Colorado, Boulder—We tested the
hypothesis that operation span (OS) performance involves attentiondemanding
retrieval of target words from long-term memory (LTM),
rather than active maintenance of all target words. Participants completed
the standard OS task and a modified, word-span-like version
114
in which all equations preceded all target words. Recall took place
under either full attention or divided attention conditions (easy vs. difficult).
Secondary-task performance was worse for the standard OS
task than for the modified version, with the difficult secondary task
showing the larger decrement. Recall also suffered under divided attention,
with the recall decrement greater for the more difficult secondary
task. Moreover, the time taken to start recalling the first word
was considerably longer for the standard version than for the modified
version. These results are consistent with the proposal that successful
OS performance involves, in part, the attention-demanding use
of retrieval cues to recover target items from LTM.
(4061)
Comparison of Standardized Measures of Working Memory Across
the Life Span. SANDRA HALE, LISA J. EMERY, JOEL MYERSON,
DESIREE A. WHITE, & DUNEESHA RANATUNGA, Washington
University—Analyses of children’s normative working memory data
from the WISC III revealed parallel developmental changes in forward
and backward digit span, extending our previous analyses of the
WAIS III. Notably, the difference between forward and backward span
remains approximately constant across the life span. Comparison of
performance on the Digit Span and Letter–Number Sequencing subtests
revealed a different pattern. Performance on the two subtests develops
in tandem from the age of 8 years through young adulthood;
scores diverge from middle-age on, however, with performance on
digit span tasks showing a much slower decline than that on
letter–number sequencing tasks. Taken together, the present findings
suggest that at least two mechanisms are involved in age-related differences
in working memory. Our findings pose problems, however,
for the view that one of these mechanisms, distinct from short-term
memory and associated with the frontal lobes, develops more slowly
in childhood and declines more rapidly in later adulthood.
(4062)
Do Phonological Processes Contribute to Aging Effects on Verbal
STM? DANIEL KLEINMAN & DEBORAH M. BURKE, Pomona College
(sponsored by Deborah M. Burke)—Age deficits in phonological
activation processes have been implicated in word production failures
in old age. We tested this phonological deficit hypothesis by examining
the effect of phonological neighborhood density of words on
short-term memory (STM) in young and older adults. Words with
dense neighborhoods share phonemes with many other words; this
benefits STM because interactive excitatory processes among phonologically
similar words maintain activation of words in STM (e.g.,
Roodenrys, Hulme, Lethbridge, Hinton, & Nimmo, 2002). Young and
older adults recalled word lists varying in word frequency of occurrence
and neighborhood density, in a memory span procedure. Frequency
effects were the same across age, but high neighborhood density
improved STM more for young than for older adults. Implications
for the processes underlying frequency, neighborhood, and aging effects
in STM are discussed.
• ATTENTION •
(4063)
Attentional Modulation of the Amygdala Varies With Personality.
STEVEN B. MOST & MARVIN M. CHUN, Yale University, &
MATTHEW R. JOHNSON & KENT A. KIEHL, Institute of Living,
Hartford Hospital—Prominent theories suggest that the amygdala responds
to emotional stimuli independently of attention. In contrast, we
found evidence for attentional modulation of the amygdala, but this
varied as a function of personality. We employed an attentional task
where participants always had to ignore emotional and neutral distractors
while searching for a target. The ease of ignoring distractors
was manipulated through the degree of specificity with which targets
were described to participants; more specific information about targets
generally leads to better ignoring of distractors (Most et al., in