S1 (FriAM 1-65) - The Psychonomic Society

Papers 15–21 Friday Morning
age. Measures that assessed global gist-based processing produced
negative correlations (higher risk perceptions associated with less risk
taking), and increased with age. Endorsement of simple, categorical
principles correlated most strongly with risk avoidance. The results
support a dual-processes interpretation in which relations depend on
developmental shifts in verbatim versus gist processing.
9:20–9:35 (15)
Decision Processes in Eyewitness Identification. STEVEN E. CLARK
& RYAN RUSH, University of California, Riverside—Eyewitness
identification accuracy depends not only on the accuracy of the witness’s
memory, but also on the kinds of decision processes that witnesses
utilize. We varied the composition of target-absent lineups to
evaluate Wells’s (1984) distinction between absolute versus relative
judgments. Specifically, our experiments were designed to make contrasting
predictions for a best above criterion (BAC) model versus a
difference model. A BAC model makes an identification if the bestmatching
lineup member is above a set criterion, whereas the difference
model makes an identification if the best-matching lineup member
is sufficiently better than the next-best matching lineup member,
irrespective of how good the matches are on an absolute scale. Our results
favored the difference model over the BAC model.
9:40–9:55 (16)
A Cognitive Theory of How People Learn to Select Strategies. JÖRG
RIESKAMP, Max Planck Institute for Human Development—It has
been widely assumed that people possess a strategy repertoire for inferences.
The strategy selection learning theory specifies how people
select strategies from this repertoire (Rieskamp, 2006, JEP:LMC, 32;
Rieskamp & Otto, 2006, JEP:G, 135). The theory holds that individuals
select strategies proportional to their subjective expectations of how
well the strategies solve particular problems and the expectations are
updated by reinforcement learning. The theory is compared to an adaptive
network model that assumes information integration according to
a connectionist network, whose weights are modified by error correction
learning. These theories were tested against each other in an experimental
study with a dynamic environment in which the performance
of inference strategies changed. In this situation a quick adaptation to
the new situation was not observed; rather, individuals stuck to the strategy
they had successfully applied previously, as was most strongly predicted
by the strategy selection learning theory.
Visual Perception
Beacon B, Friday Morning, 8:00–10:00
Chaired by Stephen R. Mitroff, Duke University
8:00–8:15 (17)
Staying in Bounds: Contextual Constraints on Object File Coherence.
STEPHEN R. MITROFF, JASON T. ARITA, & MATHIAS
S. FLECK, Duke University—Coherent visual perception necessitates
the ability to track objects as the same entities over time and motion.
Calculations of such object persistence are constrained by specific
rules and here we explore these rules through object files: visual representations
that track entities over time and motion as the same persisting
objects and store information about the objects. We present
three new findings. First, objects files are constrained by “boundedness”;
persisting entities should maintain a single closed contour. Second,
object files are constrained by “containment”; all the properties
of a persisting object should reside within the object itself. Third, object
files are sensitive to the context in which an object appears; the
same physical entity that can instantiate object-file formation in one
context cannot in another. This contextual influence demonstrates for
the first time that object files are sensitive to more than just the physical
properties contained within any given display.
8:20–8:35 (18)
The Role of Surface Features in Establishing Object Correspon-
3
dence Across Motion and Occlusion. ANDREW HOLLINGWORTH,
University of Iowa, & STEVEN L. FRANCONERI, Northwestern University—Visual
perception is frequently disrupted by saccades, blinks,
and occlusion, generating a problem of object correspondence: How
does the visual system establish the mapping between objects visible
before and after the disruption? According to the Object File Theory
of Kahneman, Treisman, and Gibbs (1992), objects are addressed by
spatial position, and perceptual correspondence is established solely
by spatiotemporal continuity. Here, we demonstrate that objects also
can be addressed by their surface features, such as color. In a modified
version of the Kahneman et al. paradigm, participants viewed two
colored disks that were filled briefly by novel shapes before moving
behind an occluder, which was removed to reveal the two disks containing
two shapes. In a shape change-detection test requiring access
to remembered shape information, performance was superior when
shape–color consistency was maintained before and after occlusion.
These data indicate that spatiotemporal information is not necessarily
privileged in object correspondence operations.
8:40–8:55 (19)
Capacity Limits of Encoding and Retrieval During Change Detection.
MELISSA R. BECK, Louisiana State University—It has been
proposed that about 4 objects can be successfully tracked over time in
a visual change detection task. However, this capacity limit may be
caused by limits on the ability to encode information and on the ability
to retrieve information. The capacity limits of encoding and retrieval
were examined by varying the number of objects that were relevant
to the change detection task and by providing a retrieval cue in
the postchange scene on half of the trials (an arrow pointing at the potential
postchange object). When a retrieval cue was provided, only
limits on encoding affected change detection performance. When a retrieval
cue was not provided, limits on both encoding and retrieval affected
performance. Performance was high and equivalent on the
postcue and the no postcue trials when 4 items needed to be encoded
for the change detection task, indicating that 4 items can be encoded
and retrieved efficiently. However, accuracy was 9% higher in the
postcue condition when 7 items needed to be encoded for the change
detection task and 14% higher when 10 items needed to be encoded.
Using Pashler’s (1988) formula for calculating capacity limits, it was
found that averaged across set sizes 7 and 10, the capacity for encoding
was 4.6 objects (postcue condition), but only 2 of these objects
could be successfully retrieved (no postcue condition). Therefore,
when the number of objects that need to be encoded outstripped the
limits of encoding, retrieval limits further impaired performance.
9:00–9:15 (20)
Testing the Neural Synchrony Hypothesis for Visual Grouping and
Selective Attention. MARIUS USHER & SAMUEL CHEADLE,
Birkbeck College, University of London, & FRANK BAUER, LMU
Munich—According to the neural synchrony hypothesis Gestaltgrouping
and selective attention are mediated by synchronous
gamma-band neural activity. To examine this theory we carried out a
number of psychophysical experiments testing temporal binding (the
ability to detect targets which are distinguished from background only
by temporal information), temporal judgments of flickering stimuli
and the effect of gamma-band flicker on spatial attention within a Posner
type paradigm. We find that (1) although for unstructured target
elements the temporal-binding resolution is slower than 60 msec,
when the target elements are organized according to smooth contours,
the resolution is increased to 10–20 msec, (2) the ability to judge the
phase of flicker of elements is affected by the spatial organization of
intervening elements, (3) a subliminal 50-Hz flicker triggers spatial
attention in a Posner type paradigm. The results are discussed in relation
to the neural synchrony hypothesis.
9:20–9:35 (21)
Figure–Ground Segregation Can Occur Under Inattention. RUTH
KIMCHI, University of Haifa, & MARY A. PETERSON, University