Knowledge Flashcards Preview

Cognitive Psychology > Knowledge > Flashcards

Flashcards in Knowledge Deck (22)
Loading flashcards...


A group of objects that belong to the same class of objects


What categories are useful for

Understanding the world
Understanding new cases
Tools for making inferences ("What does this look like?")
Understanding strange behaviors


Three approaches to categorization

Definitional approach
Prototype approach
Exemplar approach


Definitional approach

Each category has a definition
If an item fits the definition, it belongs in that category


Limitation of definitional approach

Some things can't be categorized
Not all things in a particular category exactly match the definition


Family resemblance

Things in the same category resemble one another, even if they don't fit the definition
Solves problem to definitional approach: allows for some variation within a category


Prototype approach

Membership in a category is determined by comparing the object to a prototype that represents the category
Prototype isn't an actual member of the category, but an "average" representation of the category


High vs. low prototypicality

High: category member closely resembles the category prototype (it is like a "typical" member of the category
Low: category member doesn't closely resemble a typical member of the category


Typicality effect

Rating on 7 point scale: how representative different examples of category are
(low number: representative- apple or robin; high number: not representative-pumpkin or bat)
We tend to have a pretty good agreement on what is prototypical
The closer the exemplar is to the prototypical, the faster we will be able to identify it



Presentation of 1 stimulus facilitates the response to another stimulus that usually follows closely in time


Results of priming experiment (prototypical approach)

People identify matches of colors that fit given prototype ("blue") more quickly than matches of colors that don't fit given prototype (yellow when priming word was "green")


Exemplar approach

Determining whether object is similar to actual examples of that object (exemplars)
Examples closer to exemplar are classified faster
Advantage over prototype approach: handles atypical cases


Hierarchical organization

Larger, more general categories are divided into smaller, more specific categories
Example: carnivores -> felines -> cat, puma, lynx


3 category levels

Superordinate -> basic -> subordinate


Basic category level

Specific examples (ex- bed, chair, table)
Below superordinate, but above subordinate


Subordinate category level

More specific than basic
Not much is gained from adding more detail (example: road bikes aren't that much different from trail bikes)
Expertise of a certain subject is at subordinate level


Superordinate category

Less specific than basic
Can't be super specific (ex- vehicle instead of truck, car, or bike)
Little kids often work at this level


Semantic networks: Collins and Quillian's model

Nodes (specific categories or concepts) are connected by links
Hierarchical model (levels are arranged so that more specific concepts are at the bottom and more general concepts are at the top)


Cognitive economy

Semantic networks
Information is stored at the highest level possible (example: put "has gills" with fish instead of with shark)


How a semantic network works

Distance between concepts predicts how long it takes to retrieve info about concepts
Example: takes longer to answer "Is a canary an animal?" than "Is a canary a bird?"


Spreading activation

Related to semantic networks: activity spreads out along any link that is connected to an activated node
Example: priming with "bird" activates robin, ostrich, and animal (faster to name the related concepts)
Things that don't match primer take longer to name (example: dog, insects, etc.)


Connectionist networks

Best exemplifies how learning occurs (trial and error)
Input units (ex-canary) are connected to representation/relation units (ex- can) which are connected to hidden units which are connected to output units (ex-grow, move, sing, fly)
Connection weights determine how signals sent from one unit either increase or decrease the activity of the next unit
If answer is wrong (ex- child incorrectly identifies canary as robin), error signals are sent back to the hidden and representation units to provide info on how connection weights should be changed so only correct units will be activated