Knowledge representation

Question 1

Collins & Quillian model

Answer 1

Hierarchical network model with nodes the have links to categories and their properties.

The directer the connection the faster the RT.
(RT depends on node distance)

Question 2

Critics for C&Q model

Answer 2

The frequency of association turned out to be more important than the distance in the hierarchy

The model doesn’t explain typicality effects. That which is seen as typical in your life/culture. the RT differs between people depending on the frequency of ass.

Theory also can’t explain how NO answers are created.

Question 3

Spreading activation model (Collins and Loftus)

Answer 3

Not so much hierarchy, but there’s lines between associated words. Everything on an equal field. Still can’t explain differences in RT very well

Question 4

Semantic priming

Answer 4

Prime makes info more available when presented w the target. 2 types;

1. Associative priming
   Don’t actually semantically connect, do prime based on ass. Eg cats and dogs is prime for weather
2. Semantic priming
   - Dog primes labrador, breed of dog

In a lexical decision task, the RT is shorter when primed with either types of prime.
- When learned to expect a word, the RT is quicker when its congruent w expectation. If unexpected; not

THUS the cause is the expectation created by the prime

Question 5

Categorization level

Answer 5

Superordinate;
- Category eg fruit
Basic; Central tendency
- most common eg apple
Subordinate (graded membership)
- type eg granny smith

Question 6

Mini quiz w categories

Answer 6

Studies that aim to find how humans form categories
Example w artificial stimuli
- four random dot images that belong to the same category
- recognition task follows, asked if its same category

Patterns are derived from a prototype pattern and dots around it are randomly shifted a little.
- The pattern is categorized as part of the others, whereas highly distorted are not

Question 7

Learning regularities (mini quiz conclusions)

Answer 7

• Good classification is reached after many examples (40+)
• Quite difficult learning due to randomness of dots
• Prototype is typically misrecognized as seen even if it was not

Suggests that subjet induce a prototype (= prototype theory) BUT anterograde amnesia can do this too