The World Is Very Complex

Categorization

Assigning things (percepts, concepts, objects, etc.) to distinct groups in a principled (rule-based) manner.

The world is very complex

1We cannot process all information available to us – we must selectively attend to what is important to us.

–How do we know what to attend to?

2We 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?

What determines what we attend to?

•The environment

–Some features of the environment give us necessary information.: The Horizon Ratio

•Our biology

–Some aspects of our biology determine what we are capable of attending to: Neural feature detectors (horizontal lines, left-right motion, frequency sweeps, etc.)

•Experience

–It turns out that the problem of having to group things together helps us learn what things in the world to attend to: Acquired Eqivalence, Acquired Similarity. Gibson (1969); Goldstone, (1998)

Categorical Perception:

Perceiving a continuous range of stimuli as members of discrete categories. (Harnad, 1987)

Some physical phenomena are perceived continuously, while some are not. For example: da/ to /ga/ ContinuumIdentification: Discontinuity at Boundary

Pairwise Discrimination: Peak at Category Boundary

Categorical Perception (Traditional)

Identification determines Discrimination

•Discrimination is only possible (above chance) across a category boundary.

•Within a category, all tokens are perceived as identical.

Another Example: Cantonese Tones

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

•Show distinct categories in identification.

•Do not show any evidence of category effect on discrimination.

•Conclusion (for now): Perception of tone categories does not depend only on changes in perceptual abilities.

Categorical Perception

•Using categories in perception is a cognitive process that involves interaction between perceptual information and higher-level knowledge of objects in the world.

What rules are used to group objects?

•Definitions (Feature Lists)

•Family Resemblance

•Similarity to Prototypes

•Exemplar models

Definitions and Features

•Dogs are animals that have four legs, have fur, bark, wag their tails…

Animal / Four legs / Fur / Bark / Wag tail
Banyan Tree / X / X / X / X / X
Goldish / √ / X / X / X / X
Persian Cat / √ / √ / √ / X / X
German Shepherd / √ / √ / √ / √ / √
Dachshund / √ / √ / √ / √ / √
Mexican hairless dog / √ / √ / X / √ / √
Barking deer / √ / √ / √ / √ / ?

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.

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.

•One problem is that people seem to retain memories for specific exemplars.

Exemplar Models

•To remember a category, just remember all the members of the category.

•Head-filling-up problem.

•Evidence for abstractions.

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 cat’s striate cortex. Journal of Physiology, 148, 574-591.

Hubel, D. H. & Wiesel, T. N. (1962). Receptive fields, binocular ineraction, and functional architecture in the cat’s 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.