PowerPoint-Pr

1
?
Cognitive Psychology Winter 2004 -Discussion
Section-
2
Perception Imagery

3
Cognitive functions

• Perception

Emotion Motivation Action

• Attention

• Memory

• Imagery

• Decision-making

• Reasoning, problem-solving

• Language

4
Perception

Perception, Imagery, Shepard Metzler and
Review. There is no way that I can go in any
detail in Perception.
So lets just review the most important principle.
5
Perception

What we perceive is NOT a exact copy of the
external world, it is a selective RECONSTRUCTION.
It is a MENTAL REPRESENTATION. (of aspects of the
external world)
6
Implications
4 fundamental non-correspondences

1) Things that we perceive might not be actually
present in the environment.
2) Things that are actually present in the
environment might not be perceived (see
Attention examples of last time).
3) Changes in the environment might not result in
changes in perception.
4) Changes in perception might not result from
changes in the environment.
7
Perception

4) Shepards tables
1) Kanisza triangle
8
Perception

4) Motion-induced blindness
http//www.keck.ucsf.edu/yoram/mib.html
9
Perception

3) Color-constancy
10
Perception

Fundamental Reason 1 Duham-Quine Paradox
11
Perception
Reason Limited Sampling. Inference with limited
information. ? Aliasing. Needs assumptions to
disambiguate.

0 10 0 7 6 1 9 0

0 10 0 7 6 1 9 0
12
Perception
1 Physical Stimulus ? 2 Filter1 (Static) ? 3
Filter2 (Variable) ? 4 Filter3 (Variable) ? 5
Mental representation ? 6 Filter4 ? 7 Mental
representation ? 8 Filter5 ? 9 Action.

1. e.g. Light
2. e.g. Photoreceptors
3. e.g. Eye movements
4. ATTENTION!
5. Commonly refered to as PERCEPTION!
6. ATTENTION!
7. MEMORY!
8. Action selection
9. e.g. Catching a ball

13
Perception

On the other hand, this allows us to study the
assumptions, the PROCESSING by looking at
illusions and how people perceive them.
14
Perception

A huge brain machinery is devoted to process
visual information alone. 30 of the brain. At
least. Together with other perceptual
information Easily 50. To allow us to act
efficiently on impoverished input.
15
Imagery
Some demos...
16
The Problems

• Too subjective, idiosyncratic

• Too introspective

• Too qualitative

17
High stakes...
Stephen Kosslyn, John Lindsley Professor of
Psychology at Harvard University
18
Mental scanning
Kosslyn, Ball Reiser (JExP, 1978)
19
The imagery debate(s)
Do mental representations of images retain the
depictive properties of the image itself?
20
The propositional perspective
Pylyshyn
!?
21
The arguments
Mental images represented by relations between
symbols (language-like)
Mental images represented in analogous
form (vision-like)

• Distortions by labels, heuristics
• Demand characteristics
• Computer metaphor

• Moot point 1 Anderson (PsycRev,1978) RT not
  diagnostic due to Representation/Process
  tradeoffs, emulating both.

• Moot point 2 Representation Process. Need to
  look at neurophysiology for the CODE.

22
Resolution 1 - Paradigm
Visual mental imagery activates the same areas
as visual perception (Kosslyn, et al.,
JCogNeuro, 1993)
23
Resolution 1 Paradigm
Visual mental imagery activates the same areas
as visual perception (Kosslyn et al., JCogNeuro,
1993)
24
Resolution 2
Visual mental imagery activates many visual
areas, INCLUDING V1 (Kosslyn et al., Nature
1995 Kosslyn et al., Science 1999)
25
Current state

• Perception Classically bottom up
• Retina ? LGN (Thalamus) ? V1 ? V2 ? V4 ? IT ? ? ?
  ? ? ? Frontal cortex

• Imagery Top down!

• Basically the same systems
• Frontal cortex ? ? ? ? ? ? IT ? V4 ? V2 ? V1.

26
Shepard
The grand old man of psychology
27
Mental rotation

• Shepard Metzler (Science, 1971)

28
Mental rotation

• Issue?

The nature of mental representations? Can they be
manipulated like one would expect from the laws
of physics? Does the internal world follow the
principles of the external world (it doesnt have
to!)
29
Exam review

• Purpose?

To scare you into studying. You cant wing it.
30
Exam review

• Ground rules

No cheating. You WILL regret it!
31
Exam review

• Format

• 6 essay questions. One from each field (Intro,
  Reason, Attention, Perception1, Perception2,
  Imagery). Do 5 of them.
• 10 Multiple Choice Questions.
• 1 Bonus Question.

32

• Attention

Explains many things and nothing. A theoretical
fudge factor. Equals trophic factors.
There are many models, all of them fail to
capture essentials of the phenomenon. Prominent
Treismans Attenuation theory, Broadbents early
selection theory.
Most useful to think of as a variable filter
that generates mental representations for action
selection.
There are processes that are automatic, others
need attentional resources. They reflect a
tradeoff.
Attention involved in many phenomena Visual
Search, Popout, Change blindness, etc.
33

• Perception

Perception is the formation of a mental
representation of the environment.
This representation is NOT isomorphic, but
subject to many correspondence errors.
To overcome the inherent ambiguity in sensory
data, the brain makes assumptions about the
world. (Learnt in phylogeny, ontogeny). Like
Gestalt rules.
These assumptions can be uncovered by research
with visual illusions.
34

• Imagery

Activation of a mental representation from
memory. In ABSENCE of a stimulus.
Paradigmatic case of TOP-DOWN processing.
Debate between Pylyshyn and Kosslyn whether
mental representation in imagery analog or
propositional. (Since 30 years!)
Most brain areas active in perception are also
active in imagery.
35

• Reasoning

There are two ways of reasoning, deductive
(Rule ? Particular instance) and inductive
reasoning (Data ? Rule). They both have
advantages and disadvantages.
People are subject to many biases in reasoning,
most particularly the confirmation bias, which
usually affects them in hypothesis testing.
(Elicited by Wason selection task). Also common
Baserate neglect.
People should try to falsify, not to verify
their hypotheses.