Where is the hierarchical network of concepts stored?
In the semantic memory.
1 of 25
What are the three theories of concept storage?
Feature lists in memory, as prototypes in memory, as labelled memory traces of instances
2 of 25
What do concepts enable us to do?
Generalise from past experience of instances to predict properties of new instances.
3 of 25
What must categories be in order to use them for prediction?
Natural kinds, meaning there is some basis for generalisation.
4 of 25
Why do conceptual hierarchies provide economy of representation?
Properties common to all members can be stored just once, and subordinate categories "inherit" properties of superordinate.
5 of 25
What is the economy principle?
RTs for verifying prepositions are less when the link between preposition and article is closer. Retrieval time is determined by number of links between a category and the property.
6 of 25
What else can verification time be influence by (3)?
When concepts are acquired in life, how familiar a concept is, typicality.
7 of 25
At what level to people tend to name at?
Basic level.
8 of 25
What are the other two levels people learn later in life?
Abstract and specific.
9 of 25
How do we classify things?
By reference to a mental definition listing features of a category's members.
10 of 25
What are necessary/ essential features?
Features of a category that are more than merely frequently associated with a member.
11 of 25
What are sufficient features?
Multiple features of a category that equally are sufficient for placing a member in that category.
12 of 25
What are some priori arguments against the classical view (3)?
For many concepts its hard to come up with common-sense defining factors, some categories seem to have no common core of defining features (e.g. game).
13 of 25
Why might all members of a category not be equal (3)?
Fuzziness of category boundaries, typicality effects, similarity ratings.
14 of 25
What are the effects of typicality?
When asked to generate members of a category, people are more likely to name typical members first. Children learn typical members earlier.
15 of 25
How to RTs differ between more and less typical instances?
More typical instances produce faster RTs.
16 of 25
What is modern feature-list theory?
Categorisation by probabilistic or weighted combination of characteristic features.
17 of 25
How is category membership determined in feature-list theory?
By which category's features produce the highest weighted sum of matching features.
18 of 25
What are the implications of feature-list theory (4)?
Typical members have more high-weight features, category boundaries are fuzzy, context can influence weight of features, similar members have more features in common.
19 of 25
What is prototype theory?
The idea that concepts are represented in the mind by an average or ideal member of the category, known as the "prototype."
20 of 25
How can similarity be visualised?
As distance in multidimensional space.
21 of 25
What does prototype theory accommodate (2)?
Fuzziness of boundaries, typicality.
22 of 25
When might prototype theory not be applicable?
When features are discrete, as opposed to continuous.
23 of 25
What theory combines prototype and feature-list theory and why might it be better?
Hybrid theory, because it includes discrete and continuous properties.
24 of 25
What are instance/ exemplar theories?
Involve no abstract representation, but suggests that we retain in memory record of many labelled instances. We assign a stimulus based n how similar it is to previous instances.
25 of 25
Other cards in this set
Card 2
Front
What are the three theories of concept storage?
Back
Feature lists in memory, as prototypes in memory, as labelled memory traces of instances
Card 3
Front
What do concepts enable us to do?
Back
Card 4
Front
What must categories be in order to use them for prediction?
Back
Card 5
Front
Why do conceptual hierarchies provide economy of representation?
Comments
No comments have yet been made