1
Q
Category definition + example
A
- Class of thing that share a similarity
- how you organize concepts
- what is what, match incoming info
2
Q
Concept def
A
- Mental rep of an object, event, or pattern
- decreases amount of info needing to be learned
- allows us to make predictions
- content of thinking/thoughts
3
Q
Definitional approach (theory of categorization)
A
- Form concepts by finding necessary/defining features
- figure out defining features, if has said features it belongs
- if it meets definition it belongs
- cons: have rigid boundaries (all or none, similar to template theory), acts as a good start point but hard to adapt, some times defining features are had to come up with, sometimes things don’t fit definition fully but are in the concept, not all members are equally good examples, disagreements on members in category
4
Q
Probabilistic theories (2)
A
- Prototype theory
- exemplar theory
5
Q
Prototype theory/view
A
- Idealized representation
- use prototype to make inferences
- prototype= average of experiences (only thing held onto in brain)
- items can be either low-prototypical (less likely to think of) or high-prototypical (more likely to be thought of)
- some examples are ‘better’ or recalled faster
6
Q
Typicality effects
A
- Separated into two categories: typical (faster/easier recall) and atypical (slower/harder recall, further from prototype, less frequent)
- graded structure (based on feature, you have a spread)
- measure protypicality by measuring it’s distance from the prototype
7
Q
Problems with prototypes
A
- They are organized around averages
- certain members of other categories may look similar to prototypes of other categories than the prototype of their own category
- lots of overlap
8
Q
Exemplar view
A
- Store all instances/experiences of a category
- prototype is generated/abstracted as needed (not stored)
- levels to categories are formed
- has all prototype benefit plus allowance for more rewance
- holds anto features
- explains typicality effect
- takes atypical cases into account
- deals w/ variable categories
9
Q
Characteristics of categories (exemplar modal)
A
- Graded membership (typicality effect, some more typical than others)
- family resemblance ( category w/ lots of resemblance = closer to prototype and share a set of common features)
- related concepts (central tendency [prototype] and typicality effects)
10
Q
Levels of organization
A
- Superordinate (more broad/general than basic level)
- basic (level where members share most attributes of the category, most common answer)
- subordinate (more specific than basic level)
11
Q
What level of organization would an expert automatically think of for items in their expertise
A
Subordinate
12
Q
T or f: experts respond at the subordinate level at the same speed as an average person responds at the basic level
A
True
13
Q
Semantic network
A
- Info related is linked together
- everything is organized in hierarchal structures
14
Q
Semantic hierarchical theory
A
- All info is organized in a hierarchal structure
- uses nodes and links for the hierarchy
15
Q
Nodes + links
A
- Nodes: representations of concepts
- links: representations of relationship, between/connects nodes
16
Q
Spread activation
A
- Process that involves one node activation which spreads to all others, making connected nodes easier to access
17
Q
What does sentence verification depend on according to semantic network theory
A
How close nodes are
18
Q
Property inheritance
A
- Greater distance and more connections = longer it takes
19
Q
Generalizations for semantic hierarchal theory
A
- DRM paradigm (key word not present)
- lexical decision task (as you go through the list, there is a spread of activation to see if word is real or not)
20
Q
Cons of semantic hierarchal theory
A
- Can’t explain typicality effects
- cognitive economy not always true
21
Q
Connectionist model
A
- Explains how brain can process info in an algorithm
- info can be processed in parallel (not serially as previously theorized)
- Ex parallel processing model
- units correspond to how neurons function in the brain
- collection of nodes representing info (not individual nodes)
22
Q
__________ determine at each connection how strongly en candy signal will activate the next unit
A
- Weights
- ex. excitatory=positive, inhibitory=negative
23
Q
connectionist model consists of what three units
A
- input units: activated by stimulation from enviro
- hidden units: receive input from input units
- output units: receive input from hidden units
24
Q
activation rule
A
- specifies output activation based on input activation
25
t or f: hidden units have clear interpretations
false, they act as abstract entities devoid of interpretation
26
how are neural networks like the brain
- graceful degradation: can cut parts to look at brain process, breaking system does not halt performance but slows it
- generalization: varying response
- distributed: memory and identity are distributed and redundantly stored not localized and unique
27
cryonics
- precise reconstruction of brain may not be necessary to restore memory and identity
28
sensory functional hypothesis
- what matters is how info is processed
- knowledge is accessed based on what is needed more
- separate semantic stores for sensory/perceptual properties and functional info/use
- living: visual sensory features
- non-living: functional features
29
semantic category approach
- how ppl perceive info (visual or non-visual) and organize it (perceptual level)
- splits different modalities and categories
- no separation of sensory and function
30
multiple factor approach
- looks at how concepts are divided up within a category
- each category is defined by combo of large number of factors
Cog Psych
flashcards
Decks in class (10)
# Cards
