Use LEFT and RIGHT arrow keys to navigate between flashcards;
Use UP and DOWN arrow keys to flip the card;
H to show hint;
A reads text to speech;
24 Cards in this Set
- Front
- Back
|
What is the classic view of concept formation |
1. Concepts are mentally represented as definitions; an object has necessary and jointly sufficient for membership of that category 2. No in-between cases; either in or out 3. every member is equally representative member of the category |
|
|
What are the problems with the classic view of concept formation |
Empirical findings have found that; 1. Concepts are mentally represented as distributions of overlapping features/properties 2. Category boundaries are not clear-cut - they have 'fuzzy' edges 3. Some category members are more typical or better representations than others |
|
|
What is the difference between concepts and catagories |
Concepts are mental/conceptual representations of categories Categories are classes of objects in the real world |
|
|
What is hierarchical category structure? |
A system of grouping things according to levels and orders (hierarchy). Common: super-ordinate, basic, sub-ordinate |
|
|
Give an example of a super-ordinate, basic and sub-ordinate category |
Mammal Dog Labrador |
|
|
Which level do categories carry the most information? |
Basic; objects are able to be differentiated from each other most often at this level |
|
|
What is the prototype approach? |
Our conceptual knowledge is stored at an abstract level, an abstract representation of the central tendency of a particular item (sum of all examples of that category) |
|
|
What is the exemplar approach |
We store every example of a given category we encounter and our conceptual representation consists of all the individual members of a category |
|
|
What is the difference between prototype and exemplar views and which models do each encompass? |
Prototype is low memory cost, less detail. Exemplar is high memory cost, more detail. They are 2 ends of a continuum |
|
|
What type of model is the family resemblance model? |
Exemplar- it predicts typicality as a function of featural overlap between category members, typicality is measured by mean distance between category member and each other category member |
|
|
What type of model is the polymorphous concept model? |
Prototype - it predicts typicality as a function of featural overlap between category members and feature list representing the abstracted category, typicality is measured between each category member and the central tendency |
|
|
What happens in category learning experiments |
uses easy to manipulate stimuli like colour but bears little resemblance to the real world Learning - participants are shown stimuli from 2 categories, asked to choose cat 'a' or 'b', given feedback and learn Transfer - participants are shown mix of old & new stimuli and asked to categorise them |
|
|
What do category learning experiments tell us about conceptual representation in the real world? |
provides insight into our ability to generalise from stored category representation to novel stimuli manipulating category structure allows us to test different theories about underlying categorisation and generalisation processed\s |
|
|
What are the 2 most common category learning computational models and which abstraction models do they map onto? |
GCM (exemplar) - prob is a weighted function of distance between target and other members of 2 categories MPM (prototype) - prob is a weighted function of distance between target and prototypes (central tendencies) or 2 categories |
|
|
What does it mean to say categories have 'graded structure'? |
Categories are not clear cut, they are 'fuzzy around the edges and some category members are more typical or 'better' category members than others and due to inter-item similarity relations of the category members |
|
|
What is the contrast category effect? |
categories located at the same level of abstraction in the hierarchy influence each others structure eg fruit and vegetables are contrast categories within the domain of foods ie similar vs distinctive |
|
|
What does it mean to say 'necessary and sufficient' features? |
Necessity - if any features are missing - it is NOT a member of the category Sufficient - if all features present then it IS a member of the category |
|
|
Which variables do and do not contribute to (support) graded structure? |
DO: Typicality, Categorisation Response time and generation frequency DO NOT: Image-ability & Familiarity |
|
|
what are the polymorphous and family resemblance model special cases of? |
the contrast model |
|
|
What are some criticisms of similarity? |
1. Not flexible enough - what about theories/rules? - surely our semantic representations are based upon far richer data 2. Too flexible - by explaining everything it explains nothing |
|
|
What is category-based induction and name 2 theories on this? |
making inferences about novel (new) stimuli:
1. the similarity between premise and conclusion category (eg robins vs ducks vs geese) 2. the typicality of the premise category (eg robins more typical therefore extension is higher - not vice versa |
|
|
What are some of the empirical based findings of category based induction? |
Premise monotonicity - when more categories are added to the premise the argument is stronger or when negative evidence is added, the probability should decrease Premise diversity - when the premises are distributed across the category we are more confident it is true of whole category |
|
|
What are some limitations of the similarity (induction) paradigm? |
*Biological similarities are overshadowed by behavioural similarities.. *Thematically related categories (tooth and toothbrush) trump similar categories * Knowledge effects (eg coin vs pizza - referential size) |
|
|
What is the difference between deduction and induction? |
Deduction = given true premises to reason other statements are true eg all men are mortal, socrates is a man - therefore socrates is mortal Induction = given the premises, the conclusion is probable (inferring) eg 7% men are colourblind, Joe is a man, therefore 7% chance he's colourblind |
