Knowledge

Question 1

How is knowledge organized?
Category
Exemplar
Concept

Answer 1

Category: Group of objects that belong together
Exemplar: Item in the category
Concept: Mental representation of objects, ideas, and events

Question 2

Classical view of categorization (definitional approach)
Defining features

Answer 2

Defining features: Feature that us necessary for category membership
- Any other attributes not required for category membership are sufficient

Question 3

Problems with classical view of categorization (definitional approach)

Answer 3

1) Not all categories have a list of defining features
2) Typicality effects: Some items are more typical examples of a category than others

Question 4

Prototype theory
Central tendency
Family resemblance

Answer 4

Categories are fuzzy/have graded structure
Exemplars have characteristic features: Feature that members may possess but isn’t required for category membership
Determine category membership by comparing to prototype (avg of all members) stored in mem

Central tendency: Where exemplars w/ most characteristic features are found

Family resemblance: All category members can belong to same category without being typical members

Question 5

3 levels of categories (prototype theory)

Answer 5

Superordinate: Broad category
(Mammal, plant)

Basic level: Moderately specific; informative and distinctive
(Dog, tree)

Subordinate: Specific instances of basic level category
(Poodle, maple)

Question 6

Exemplar theory
Problems with the theory

Answer 6

We store actual examples of items we encountered in the past
- Explains context effects bcuz it depends on personal exp

Problems:
1) Ppl can give typicality ratings to clearly defined categories
2) People can make categories depending on the features they want to compare

Question 7

Knowledge-based / Explanation-based theories of categorization

Answer 7

Psychological essentialism: Categories have underlying true nature that can’t be stated explicitly (bio things can’t be changed to be put into a diff bio category)
- Accounts for why we judge some features as more important than others for membership

Question 8

Semantic Network Models

Answer 8

Concerned w/ how diff items are related to each other

Nodes contain info and connect to each other by directional pathways
- Nodes activated w/ spreading activation

Question 9

Hierarchical model (Collins and Quillian) - Property inheritance

Semantic relatedness model (Collins and Loftus)

Answer 9

Property inheritance: Moving down hierarchy, concepts inherit properties from concepts higher up in hierarchy
- Helps conserve cog resources

Nodes are organized based on strength of relationship
- Stronger association or Typical exemplars = Shorter pathway

Question 10

Artificial neural network (ANNs)

Answer 10

Computing system modelled after neurons
- Composed of input, output and hidden layers
- Connections are weighted
- Each unit can be excitatory, inhibitory, or inactive

ANNs don’t store knowledge in nodes
- Stores in distributed of weights as pattern of activation
- New info changed weights via back propagation