Abstracts 2005 - The Psychonomic Society

Saturday Evening Posters 5107–5113
mance, and greater task difficulty increased the cost, but these patterns
did not vary between the groups.
(5107)
Alcohol-Induced Impairment of Behavioral Control: The Motor
Program and Impulsivity. CECILE A. MARCZINSKI & BEN D.
ABROMS, University of Kentucky, MARK G. VAN SELST, San Jose
State University, & MARK T. FILLMORE, University of Kentucky—
Model-based assessments of behavioral control have been used to
study the ability to execute and inhibit behavioral responses. Response
inhibition appears to be more vulnerable to the impairing effects of
alcohol than does response execution. Current information-processing
models have not accounted for this observation. The present study examined
whether the particular vulnerability of response inhibition to
the effects of alcohol occurs at the level of the action (motor program).
Participants performed a cued go/no-go task that required quick responses
to go targets and suppression of responses to no-go targets.
Response requirements were manipulated by varying the nature of the
action required (pressing or releasing a key). Dose-dependent increases
in commission errors were observed only with response engagement
(keypress) and not with response disengagement (releasing a key).
Therefore, not all aspects of motor processing requiring inhibition are
equally impaired by alcohol, and the duration of the motor program
appears to be fundamental to observing alcohol-induced impulsivity.
(5108)
Incidental Learning and Subliminal Motor Priming. FRIEDERIKE
SCHLAGHECKEN, ELISABETH BLAGROVE, & ELIZABETH A.
MAYLOR, University of Warwick—Subliminal priming effects in the
masked-prime paradigm can be obtained only when primes are part of
the task set (e.g., arrow primes elicit priming effects only when arrows
also appear as targets). We investigated whether the relevant feature of
the task set needs to be explicitly instructed (e.g., “left-hand response
to a left-pointing arrow”), or whether participants would extract it automatically
in an incidental learning paradigm. Primes and targets were
symmetrical arrows (�� and ��), with target color, not shape, the
response-relevant feature. Whereas shape and color covaried for targets
(e.g., �� always blue, �� always green), primes were always black.
Over time, a negative compatibility effect (NCE; response benefits for
prime–target pairs with different shapes) developed, indicating that
primes began to affect the motor system. When target shape and color
varied independently (control condition), no NCE occurred, confirming
that the NCE reflects motor processes, not perceptual interactions.
(5109)
Contribution of Ipsilateral Motor Areas to Response Initiation: Test
of the Hemispheric Coactivation Hypothesis. JEFF O. MILLER, University
of Otago—Miller (2004) proposed a “hemispheric coactivation”
hypothesis to explain the surprising finding that redundancy gain is
larger when redundant visual stimuli are presented to different hemispheres
of individuals without a functioning corpus callosum than
when such stimuli are presented to normals. According to this hypothesis
(and contrary to the standard assumption of contralateral
hand control), unimanual responses are actually bilaterally controlled,
with ipsilateral, as well as contralateral, motor areas making a substantial
contribution to response initiation. An EEG experiment was
conducted to test for the hypothesized contribution of ipsilateral
motor areas to response initiation. The results suggest that ipsilateral
motor areas do contribute to response initiation in a unimanual simple
RT task, but not in a between-hands choice RT task.
• PSYCHOLINGUISTICS •
(5110)
Is There a Verb/Noun Access Dissociation by Modality in Normals?
CHRIS WESTBURY, University of Alberta, & SIGNY SHELDON,
University of Toronto—Studies of aphasics suggest that word class (e.g.,
noun and verb) processing can be dissociated by sensory modality via
139
brain injury. Does modality also interact with word class in normal
language processing? We hypothesized that that the aphasic production
deficits might reflect an underlying dissociation in representation,
which could be studied in normal subjects by using a comprehension
task that facilitates experimental control. We use a syntactic
priming task, in which nouns and verbs words were preceded by either
the word to or the. Subjects were asked to make lexical decisions. Syntactic
priming effects were seen in both the visual and the auditory
modalities, but they were much larger in the auditory than in the visual
modality. RTs were much longer to primed than to unprimed words in
the visual modality only. The results constitute evidence of modalityspecific
grammatical class access differences in normal subjects.
(5111)
Coercion Without Type Shifting: The Role of the Subject in Enriched
Interpretations. STEVEN FRISSON, BRIAN D. MCELREE, &
PREETI THYPARAMPIL, New York University (sponsored by Brian D.
McElree)—Interpreting expressions like “began the book,” in which
an event-selecting verb is paired with a complement of a different semantic
type (an entity), requires more processing than do expressions
with no mismatch in semantic type (e.g., McElree et al., 2000). Recent
data suggest that it is the time required to generate additional semantic
content (e.g., “began TO READ the book”) that is costly. Two
eyetracking experiments further investigated this hypothesis with sentences
like “{The honest witness}/{The court reporter} completed the
testimony about . . .” Neither of these sentences requires type shifting,
since the verb is event selecting and the complement expresses an
event. In (1), the agent can plausibly control the actual event. In (2),
the agent controls another unspecified event associated with the complement
(e.g., recording the testimony). We observed increased reading
times for (2), as compared with (1), and to different controls, indicating
coercion effects for sentences that do not require type shifting.
(5112)
Evaluation of Unsupervised Semantic Mapping of Natural Language
With Leximancer. ANDREW E. SMITH & MICHAEL S.
HUMPHREYS, University of Queensland (sponsored by Andrew J.
Heathcote)—The Leximancer system is a new method for unsupervised
transformation of natural language into semantic patterns in an
unsupervised manner. The technique employs two stages of lexical cooccurrence
information extraction that are performed in sequence,
using a different algorithm for each stage. These can be characterized
as semantic extraction followed by relational extraction. In each stage,
the data consists of actual episodic co-occurrence records. The algorithms
used are statistical but employ nonlinear dynamics and machine
learning. This paper seeks to validate Leximancer output, using
a set of evaluation criteria taken from content analysis that are appropriate
for knowledge discovery tasks. There are several reasons why
one would want an automated system for content analysis of text.
Human decision makers are potentially subject to influences that they
are unable to report (Nisbett & Wilson, 1977). Furthermore, mitigation
of subjectivity in human analysis currently requires extensive investment
in the analysis process.
(5113)
The Effect of Manipulation Features on Lexical-Semantic Processing:
An Eyetracking Study on Normal and Brain-Damaged Populations.
JONG-YOON MYUNG, SHEILA E. BLUMSTEIN, & JULIE C.
SEDIVY, Brown University (sponsored by Steven Sloman)—An eyetracking
method was used in both normal and brain-damaged populations
to investigate whether manipulation features are a part of the
lexical-semantic representation of objects and whether they are related
to physical manipulations of objects. Participants viewed a visual display
on a computer screen and were asked to touch the corresponding
object on the display in response to an auditory input. Normals fixated
on an object picture (e.g., “typewriter”) that was manipulationrelated
to a target word (e.g., “piano”) significantly more often than
an unrelated object picture (e.g., “bucket”), as well as a visual control