cogsci psy-3

8/14/2019 CogSci Psy-3

1/32

26 September, 2000 HKU 1

Categorization

Assigning things (percepts, concepts,

objects, etc.) to distinct groups in a

principled (rule-based) manner.


2/32


Categorization and

Perception


3/32

26 September, 2000 HKU 3How do we know what to look for?

The world is so full of information...


4/32


The world is very complex

1 We cannot process all information available to us

we must selectively attendto what is important

to us.

How do we know what to attend to?

2 We cannot keep track of every individual item in

the world we must group similar things

together. What rules are used to group objects, and how do these

rules operate?


5/32


6/32


Environment

The Horizon ratio.

Most people see the

buildings as the same

size, and the tower as

taller.

The ratio above to

below the horizon

always gives good

information about

height (except in

illusions).


7/3226 September, 2000 HKU 7

Our Biology

Neural detectors for perceptual

properties

Vision: Horizontal lines, vertical lines,dots, directional motion, retinal location

(Hubel & Wiesel, 1959, 1962); also color,

brightness, simple shapes & solids, etc. Audition: loudness, pitch, frequency

sweeps.

Etc.



Experience

Experience with making categories

causes changes in perception.

Acquired equivalence within categories.

Acquired distinctiveness between them.

Eleanor Gibson (1969)

Robert Goldstone (1998)



Categorical Perception

Perceiving a continuous range

of stimuli as members of

discrete categories.

(Harnad, 1987)



Some Physical Continua

Color

Pitch

Loudness

Brightness

Angle

Weight

Etc.



Some physical phenomena

are perceived continuously

0

20

40

60

80

100

0 1 2 3 4 5 6 7 8 9 10

Magnitude of Stimulus(e.g. Loudness)


12/32


Some are not

010

20304050

607080

90100

0 1 2 3 4 5 6 7 8 9 10

token


13/32


For Example

A continuum ranging from

/da/ to /ga/

(after Delattre, Liberman, & Cooper, 1955)

Good /ga/Good /da/

1 2 3 4 5 6 7 8


14/32


Identification:

Discontinuity at Boundary

010

20304050

607080

90100

1 2 3 4 5 6 7 8

token


15/32


Pairwise Discrimination:

Same or Different?


16/32


Pairwise Discrimination

(same/different)

0

1020

30

40

50

6070

80

90

100

1_2 2_3 3_4 4_5 5_6 6_7 7_8

Pair of stimuli


17/32


What Happened?

1 2 3 4 5 6 7 8

1 2 3 4 5 6 7 8

Physical World

Perceptual Representation


18/32



Identification

determinesDiscrimination


19/32


Traditional view of CP

Discrimination is only possible

(above chance) across a category

boundary.

Within a category, all tokens are

perceived as identical.


20/32


Another Example

Level Tone Continuum

Tone 1Tone 6 Tone 3

1 2 3 4 5 6 7 8 9 10


21/32


Identification of tones in

context(after Francis, Ciocca, & Ng, in prep)Identification

0.00

0.10

0.20

0.30

0.40

0.50

0.60

0.70

0.80

0.90

1.00

1 2 3 4 5 6 7 8 9 10

Stimulus

low

mid

high


22/32


Pairwise Discrimination

(same/different)Level tone discrimination

0.00

0.10

0.20

0.30

0.40

0.50

0.60

0.70

0.80

0.90

1.00

1_2 2_3 3_4 4_5 5_6 6_7 7_8 8_9 9_10

Stimulus pair


23/32


Cantonese Tones

Show distinct categories in

identification.

Do not show any evidence ofcategory effect on discrimination.

Conclusion (for now): Perception of

tone categories does not dependonly on changes in perceptual

abilities.


24/32



The use of categories in perception

is a cognitive process that involves

interaction between perceptualinformation and higher-level

knowledge of objects in the world.


25/32


What rules are used to

group objects?

Definitions (Feature Lists)

Family Resemblance

Similarity to Prototypes

Exemplar models


26/32



27/32


28/32


Definitions and Features

Dogs are animals that have four legs,

have fur, bark, wag their tails

Animal Four legs Fur Barks Wags tail

Banyan Tree X X X X X

Goldish X X X X

Persian Cat X X

German Shepherd

Dachshund

Mexican Hairless X

Barking Deer ?


29/32


Family Resemblance

Ludwig Wittgenstein (1953) proposed

that games could not be defined or

categorized by features. Rather, any game shares some

resemblance to some (but not all)

other games.


30/32


Similarity to Prototypes

Rosch (1978). Prototype is a central,

average, representation (real or

constructed) of a category. Tokens sufficiently similar to the

prototype are considered members

of that category. Memory for specific exemplars.


31/32


Exemplar Models

To remember a category, just

remember all the members of the

category. Head-filling-up problem.

Evidence for abstractions.


32/32

26 S t b 2000 HKU 32

Bibliography

Francis, A. L., Ciocca, V. & Ng, B. K. C. (in prep). On the noncategorical perception of

Cantonese tones.

Gibson, E. J. (1969). Principles of Perceptual Learning and Development. New York,

Appleton-Century-Crofts.

Goldstone, R. (1998). Perceptual learning. Annual Review of Psychology, 49, 585-

612. Harnad, S. (1987). Psychophysical and cognitive aspects of categorical perception: A

critical overview. In Harnad, S. (Ed.) Categorical Perception: The Groundwork of

Cognition. Cambridge, MA, The MIT Press.

Hubel, D. H. & Wiesel, T. N. (1959). Receptive fields of single neurones in the cats

striate cortex. Journal of Physiology, 148, 574-591.

Hubel, D. H. & Wiesel, T. N. (1962). Receptive fields, binocular ineraction, and

functional architecture in the cats visual cortex. Journal of Physiology, 160, 106-154.

Rosch, E. (1978). Principles of categorization. In Rosch, E. & Lloyd, B. (Eds.)

Cognition and Categorization. Hillsdale, NJ, Lawrence Erlbaum Associates.

Wittgenstein, L. (1953). Philosophical Investigations. New York, Macmillan.

cogsci psy-3

Documents