Abstracts 2005 - The Psychonomic Society

Saturday Morning Papers 155–161
Visual Cognition
Civic Ballroom, Saturday Morning, 8:00–10:20
Chaired by James R. Brockmole, Michigan State University
8:00–8:15 (155)
Contextual Cuing in Real-World Scenes. JAMES R. BROCKMOLE
& JOHN M. HENDERSON, University of Edinburgh—We investigated
whether contextual cuing in real-world scenes is driven by memory
for local objects and features or whether a scene’s gist guides attention
to targets. During learning, twice the number of repetitions
were required to observe maximal learning benefits for inverted
scenes (making them harder to interpret), as compared with upright
scenes, indicating a role for semantic memory in cuing. Following
learning, scenes were mirror reversed, spatially translating features
and targets while preserving gist. Observers first moved their eyes toward
the target’s previous position in the display. This localization
error caused an increase in search time, since additional fixations
were required to locate the target. The disruption was not absolute;
when initial search failed, the eyes quickly moved toward the target’s
new position. This suggests that the scene’s gist initially guides attention
to the target and that localized feature information is used if
the gist association fails.
8:20–8:35 (156)
Varieties of Emergent Features in Visual Perceptual Organization.
JAMES R. POMERANTZ & MARY C. PORTILLO, Rice University—
At the heart of Gestalt effects in perception are nonlinearities, where
the whole is perceived differently from the sum of its parts or, put differently,
where elements are perceived and discriminated differently
in different contexts. We study such Gestalt organization by searching
for configural superiority effects (CSEs), which are instances
where stimuli are discriminated from each other more quickly and accurately
in the presence of context elements even when those elements
by themselves provide no information relevant to the discrimination.
In most instances, adding such context hinders performance: Discriminating
A from B is easier than discriminating AC from BC
(where C is the context). We demonstrate a dozen or so cases where
context actually helps (sometimes greatly), and we try to extract general
principles to explain CSEs in terms of emergent features arising
when elements group.
8:40–8:55 (157)
Fixation Position Changes During Fixations in Reading. AL-
BRECHT W. INHOFF, & ULRICH WEGER, SUNY, Binghamton,
SETH GREENBERG, Union College, & RALPH R. RADACH, Florida
State University—Readers move the eyes to apply high acuity vision
to different text segments and they keep the eyes relatively stationary
(fixated) after that so that linguistic detail can be obtained. The study
examined whether small position changes that occur during fixations
are a consequence of oculomotor reflexes and/or ongoing processing
demands. Two tasks were used: silent reading and reading plus detection
of a target letter. The left eye was more stable than the right eye
in both tasks. A strong directional bias was present in the reading task,
with larger and more frequent position changes toward the right than
toward the left. This bias was eliminated when letter detection was
added to the task. An ongoing “pull-mechanism” appears to shift the
eyes toward to-be-identified words during fixations in reading. However,
when readers are forced to consider linguistic detail such as a target
letter, the pull-mechanism is apparently suppressed.
9:00–9:15 (158)
Forgetting Pictures: Visual Versus Conceptual Information. MARY
C. POTTER & LAURA F. FOX, Massachusetts Institute of Technology—
Rapidly presented pictures are quickly, but not immediately, forgotten
(Potter, Staub, & O’Connor, 2004). We assessed memory for concep-
25
tual gist and memory for visual and spatial information in experiments
using title or picture recognition tests. In addition, subjects made a
forced choice between two versions of a picture that differed in visual/
spatial features, such as color, left–right orientation, or presence/absence
of details. Although gist memory declined over testing, visual/spatial
memory (poorer to begin with) did not, suggesting that if gist is remembered,
so is any visual/spatial information encoded with the gist.
Hence, the conceptual gist of a picture is extracted early and determines
what is remembered. We also compare gist and visual/spatial
information in longer term memory for pictures.
9:20–9:35 (159)
Tunnel Vision During Visual Scanning: Do Smaller Views Yield
Smaller Representations? HELENE INTRAUB & KAREN K.
DANIELS, University of Delaware—Regions of real 3-D scenes delimited
by “windows” are remembered as having included more of the
world than was actually shown (boundary extension [BE]; Intraub,
2004). Does the scope of each exploratory fixation constrain BE? In
Experiment 1, viewers (N = 60) studied scene regions with multiple
objects for 30 sec each. Vision was binocular (normal) or monocular
through tunnel vision goggles with large (3 cm) or small (0.6 cm)
apertures. Binocular vision encompassed the entire region and surrounding
space; tunnel viewing required effortful head movements to
examine the regions. At test, viewers reconstructed the windows to
recreate the delimited areas. In all conditions, viewers remembered
seeing about one third more area; dramatic differences in visual scope
did not influence spatial extrapolation. Experiment 2 (N = 40) replicated
the results and demonstrated that viewers were experiencing true
BE—not simply creating “conventionally sized” regions. Layout extrapolation
was constrained by the targeted view—not the scope of the
input.
9:40–9:55 (160)
The Role of Visual Short-Term Memory in Gaze Control. ANDREW
HOLLINGWORTH, ASHLEIGH M. RICHARD, & STEVEN J. LUCK,
University of Iowa—It is well established that object representations
are maintained in visual short-term memory (VSTM) across saccades,
but the functional role of VSTM in gaze control is not well understood.
Saccades are often inaccurate, and when the eyes miss a saccade
target, multiple objects will lie near fixation, especially in cluttered,
real-world scenes. We tested the hypothesis that VSTM stores
target information across the saccade to identify the target among
other local objects, supporting an efficient corrective saccade to the
target. A new paradigm was developed to simulate saccade error. During
a saccade, a stimulus array was shifted so that the eyes landed between
the saccade target object and a distractor. Accurate gaze correction
to the target required transsaccadic memory for the target’s
visual form. VSTM-based gaze correction in this paradigm was accurate,
fast, and automatic, demonstrating that VSTM plays an important
role in the direction of gaze to goal-relevant objects.
10:00–10:15 (161)
Fixational Eye Movements at the Periphery. JUDITH AVRAHAMI
& OREN FLEKSER, Hebrew University of Jerusalem—Although the
fact that the eye is moving constantly has been known for a long time,
the role of fixational eye movements (FEMs) is still in dispute. Whatever
their role, it is structurally clear that, since the eye is a ball, the
size of these movements diminishes for locations closer to the poles.
Here, we propose a new perspective on the role of FEM, from which
we derive a prediction for a three-way interaction of a stimulus’ orientation,
location, and spatial frequency. Measuring time to disappearance
for gratings located in the periphery we find that, as predicted,
gratings located to the left and right of fixation fade faster
when horizontal than when vertical in low spatial frequencies and
faster when vertical than when horizontal in high spatial frequencies.
The opposite is true for gratings located above and below fixation.