CONSTRAINTS ON IMAGERY II

1
CONSTRAINTS ON IMAGERY II

• David Pearson
• Room T10, William Guild Building
• d.g.pearson_at_abdn.ac.uk

2

• Although concept of mental resources is widely
  accepted within cognitive psychology, there is
  considerable debate as to what these resources
  may consist of.

3
(No Transcript)
4
(No Transcript)
5

• If specialised cognitive resources exist, we
  would expect selective interference effects to
  occur between two cognitive tasks which try and
  use the same resource at the same time.

6
Working Memory

• A theoretical model which can usefully be applied
  to the study of resource limitations is that of
  working memory (Baddeley Hitch, 1974 Baddeley,
  1986).

7
Working Memory (Baddeley Hitch, 1974)
8

• Multi-component model of short-term memory
  comprises three processing components
• Central Executive a general-purpose
  attention-based system which is responsible for
  the co-ordination and control of all conscious
  cognitive activity.
• In addition to this amodal system are two
  modality specific slave systems which are
  responsible for short-term storage within working
  memory.

9

• Phonological Loop responsible for the
  short-term storage and maintenance of verbal
  material i.e., words, text, sounds etc.
• Visuo-spatial Sketchpad responsible for the
  storage and maintenance of visual and spatial
  material i.e., pictures, colour shades, spatial
  locations etc.
• Both slave systems are functionally independent
  of each other, although they may interact
  together during any given cognitive task.
• All three components are subject to capacity
  limitations on the amount of information that
  they can process or store.

10
(No Transcript)
11
Working Memory Mental Imagery

• In the working memory model the generation and
  maintenance of visual images is the
  responsibility of the specialised visuo-spatial
  component.
• As there are limited resources available to the
  visuo-spatial component, this will constrain the
  complexity of the image transformations that can
  be carried out using imagery alone.
• A capacity limitations hypothesis can be used to
  account for constraining effects in mental
  rotation (Rock, 1973) and mental synthesis
  (Roskos-Ewoldsen, 1993).

12

• Roskos-Ewoldsen examined the recognition of
  emergent shapes in constructed patterns under
  imagery or perceptual conditions.

13
(No Transcript)
14
(No Transcript)
15
PRESENTATION OF PARTS
16
PRESENTATION OF PARTS
17
PRESENTATION OF PARTS
18
PRESENTATION OF PARTS
19
TEST PROBE OLD
20
TEST PROBES NON-COMPONENT
21
TEST PROBES EMERGENT
22
Results of imagery-perception comparison
23

• Roskos-Ewoldsen also examined the effect of
  goodness and badness of patterns and parts on
  the recognition of emergent shapes
• This dimension referred to the degree of
  perceptual organisation of the parts and patterns
  as rated by independent judges.

24
(No Transcript)
25
(No Transcript)
26
Results for good and bad parts
27

• Roskos-Ewoldsen argues these findings result from
  capacity limitations of the imagery system.
• Good parts, with clear perceptual organisation,
  require less mental resources to combine together
  than bad parts with poor perceptual
  organisation.
• ......goodness of parts may reduce the
  processing load required to imaginally construct
  the parts and resulting pattern, leaving more
  processing capacity, within a limited-capacity
  system, to discover an emergent part.
  Roskos-Ewoldsen (1993)

28
Results for good and bad patterns
29

• This does not seem to apply for good patterns,
  however.
• In this case a strong perceptual organisation
  may make it harder to detect emergent parts
  compared to a more poorly organised structure.
• This can be related to the frames of reference in
  which good and bad patterns are mentally
  understood.

30
(No Transcript)
31
(No Transcript)
32
PERCEPTUAL REFERENCE FRAMES

• In addition to resource limitations, discoveries
  from imagery may also be constrained by the
  nature of images in comparison to percepts.
• While pictures may be ambiguous, percepts always
  represent a specific configuration.
• Perceptual reversals of ambiguous figures involve
  alternating between different percepts of a
  single depiction.

33
(No Transcript)
34

• Percepts result from the continuing
  interpretation of in-coming sensory stimuli. A
  percept therefore goes beyond the information
  given (Bruner, 1957)
• Peterson, Kihlstrom, Rose, and Glisky (1992)
  argue that percepts exist only within the context
  of a perceptual reference frame.
• This specifies the information needed to
  interpret the percept i.e., its orientation, its
  figure/ground organisation, its configuration in
  depth etc.
• The reference frame directly shapes our conscious
  experience of the stimuli i.e., Goldmeier,
  1937/1972 Rock, 1983.

35
(No Transcript)
36
REFERENCE FRAMES IN IMAGERY

• The theory of perceptual reference frames can
  offer an account of constraints in imagery
  discovery because images, unlike pictures, cannot
  be inherently ambiguous (Reisberg, 1996).
• Images are always created and interpreted within
  a specific frame of reference.
• This can severely limit the discovery of any
  potential emergent patterns or properties if they
  are incompatible with the reference frame in
  which the image is understood.

37
(No Transcript)
38
(No Transcript)
39
(No Transcript)
40
(No Transcript)
41
THE ROLE OF EXTERNAL SUPPORT

• Stimulus support can take many forms, ranging
  from simple sketching to the use of sophisticated
  computer-generated virtual environments.
• Stimulus support allows an individual to
  externally represent some or all aspects of their
  mental image.
• Support reduces mental load on the cognitive
  systems involved and also provides external
  support to permit reference frame reversals
  i.e., Chambers Reisberg, 1985 Pearson et al.,
  1996).

42

• Although stimulus support is beneficial for image
  restructuring tasks, it appears less beneficial
  for image combination i.e., mental synthesis.
• Anderson and Helstrup (1993) report that allowing
  participants to draw during the creative
  synthesis task resulted in fewer creative
  productions compared to participants just using
  imagery on its own.
• Verstijnen et al. (1998) also report equal or
  even deteriorated performance of the creative
  invention task using drawing support compared to
  imagery alone.

43
Why should drawing support fail to enhance
imagery?

• Image reference frames may incorporate verbal or
  semantic information associated with the image
  which is not available using a sketch or drawing.
• Usefulness of drawing support depends on the
  expertise of the participant involved (e.g.,
  Verstijnen et al., 1998).

44
STATIC VS. DYNAMIC EXTERNAL REPRESENTATION

• Image combination is a dynamic process, but
  drawing or sketching can only provide static
  external support.
• Using sketching during a creative task results in
  an interplay between internal and external
  representations.
• Potential for computer-generated virtual
  environments to enhance creative discovery by
  allowing us to externalise object transformations
  that would be impossible to perform with real
  objects.

45
Molecular Docking in VR systems
46
Course Summary for Lectures 7-9

• There is considerable anecdotal evidence that
  mental imagery can play a fundamental role during
  human creative thinking.
• Neomentalist research offers evidence which
  suggests a structural equivalence between images
  and percepts.
• Empirical studies have provided evidence that
  participants can discover emergent patterns from
  mentally-combined images to a level at least
  equivalent to that of perception.

47
Course Summary for Lectures 10-12

• Finke suggests that the generation and
  interpretation of mental images can directly lead
  to the discovery of highly creative and original
  emergent properties.
• While the combination of images is relatively
  easy, the restructuring of an existing image is
  much more difficult without some form of external
  support.
• Two possible factors which can constrain imagery
  are the capacity limitations of the cognitive
  systems involved, and the reference frames in
  which the images are interpreted.