Shifting Visual Attention Between Objects and Locations

1
Shifting Visual Attention Between Objects and
Locations

• --Evidence From Normal and Parietal Lesion
  Subjects

Robert Egly, Jon Driver, and Robert D. Rafal
2
Introduce

• The dispute between space-based and object-based
  models of visual attention
• Space-based
• object-based

3
Previous Evidence for Space-Based Attention 1

• The cuing paradigm
• Cue allows orienting mechanisms to allocate
  resources to the cued locus.
• Visual attention can operate on a purely spatial
  representation of the visual field.
• Accommodate spatial cuing effects within
  object-based frameworks.
• Tipper, Driver, and Weaver (1991)Object-centred
  inhibition of return

4
Previous Evidence for Space-Based Attention 2

• The spatial nature of attention deficits after
  brain damage appears broadly consistent with
  space-based models.
• Unilateral neglect
• Neglect can result from damage to a number of
  distinct components in a network controlling.
  (Posner et al.)
• Disengage ?unilateral parietal lobe
• Move ? midbrain retinotectal pathway
• Engage ?pulvinar nucleus of the thalamus

5
Previous Evidence for Object-Based Attention 1

• Duncan (1984)?
• Watt (1988)
• Duncans results might therefore imply
  spatial-frequency restrictions on attention
  rather than object-based restrictions.
• Baylis and Driver (1993)
• Kramer and Jacobson (1991)

6
Previous Evidence for Object-Based Attention 2

• Neuropsychological evidence
• Farah, Wallace, Brunn, and Madigan (1989)
• Patients with bilateral lesions of the parietal
  lobes (simultaneous agnosia)

LNOA
GHDF
7
Summary

• The experiment was designed to measure both
  space-based and object-based components of covert
  visual orienting within a single task.
• The purpose of Experiment 1 was to examine how
  cuing one part of an object affects the
  processing of other parts of that object and
  equidistant parts from another object.
• The purpose of Experiment 2 was to examine
  neuropsychological deficits in both the space-
  and object-based components of visual attention.

8
Experiment 1 1

• The subjects task was to detect the emergence of
  the target
• Design
• Target-present trials (640)
• 75 valid cue(480), 25 invalid cue(160)
• Catch trials (128)
• Result
• Initial analysis
• Three-way within-subjects ANOVA
• cuing (valid or invalid), target field
  (left or right), orientation of rectangles
  (vertical or horizontal)
• Cuing effect analysis ?
• Three-way within-subjects ANOVA
• rectangle (cued or uncued), target field
  (left or right), direction of the required shift
  in attention (vertical or horizontal)

9
Experiment 1 2

• Discussion
• Purely spatial component ?attention shift to a
  new locus
• Object-based component ?additional cost of
  between-objects shifts
• The between-objects cost cannot be attributed to
  any difficult in moving attention across
  contours.

10
Experiment 2 1

• Procedure and design were identical to those of
  Experiment 1
• Patients with unilateral damage to parietal
  cortex
• Having a pathological spatial disengage operation
• Detection should be abnormally slow for
  contralesional targets after an invalid cue.
• Evidence for spatial component ( a disengage role
  for the parietal lobe)
• Invalidly cued target in contralesional versus
  ipsilesional
• Evidence for non-spatial component
• Additional cost for between versus within object
  shift

11
Experiment 2 2

• Subject ?
• Result
• Initial analysis
• four-way mixed ANOVA
• cuing (valid or invalid), target field
  (ipsilesional or contralesional), orientation of
  rectangles (vertical or horizontal)
• side of lesion (left or right) ?
• Cuing effect analysis
• Three-way within-subjects ANOVA
• rectangle (cued or uncued), target field
  (ipsilesional or contralesional), direction of
  the required shift in attention (vertical or
  horizontal)

interaction
interaction
12
Experiment 2 3

• A signifisant three-way interaction among side of
  lesion, target field, and cued versus uncued
  rectangle.
• Right-lesion
• There were significant effects of both field and
  rectangle but no interaction.
• Left-lesion
• Interaction between target field and cued versus
  uncued rectangle.

13
General Discussion

• Right-parietal damage appears to impair only the
  space-based component, whereas left-parietal
  damage appears to impair both the space-based and
  object-based components.
• The left hemisphere is relatively specialized for
  shifting attention between objects compared with
  the right hemisphere.

14
Inhibition of return
Attention might apply to the cued box rather than
to the cued location per se, because the delayed
inhibitory effects of peripheral cues could
follow the cued box when it moved to a new
location.
15
Duncan (1984) Baylis and Driver (1993)
16
Stimuli and procedure
Wait for response
1000ms
100ms
200ms
17
Result of experiment 1
The additional cost must reflect a time cost for
shifting attention between objects ?object-based
component
18
Cilnical Date for patients
19
(No Transcript)
20
(No Transcript)
21
(No Transcript)
22
The difference between the contralesional and
ipsilesional object effects
Object effect (uncued rectangle invalid-cue
RT)-(cued rectangle invalid-cue RT)