7 Flashcards

(50 cards)

1
Q

what is knowledge defined in memory terms

A

semantic memory. the organization of knowledge is the structure of semantic memory

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
2
Q

most of what we know is what type of memory

A

semantic

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
3
Q

category

A

external grouping of objects that belong together and have something in common. perceptually, biologically, or functionally similar

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
4
Q

concept

A

internal grouping that depends on how you assign and also what’s happning externally. you make assignments based on what is useful (i.e. all cars are one concept to me but foam tip and brush tip eyeliners are not. fuck foam tip fr). if you are given a category, you interalize so it technically immediately becomes a concept

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
5
Q

exemplar

A

an item in a category. exchagable with little consequences. such as porcelain to paper cup.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
6
Q

why are categories useful

A

it allows us to predict what is likely in new situations because they treat all exemplars as similar

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
7
Q

classical view of categorization

A

categories are defined by a list of necessary (items must have defining features) and sufficient (any other attributes not needed) features

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
8
Q

why is the classical view of categorization ass

A

because you can lose many characteristics and an object will still be in a group. not all categories have defining features (but then again maybe we just suck at listing the defining features). some items are more typical examles of a category than others, leading to typicality effects.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
9
Q

how are typical exemplars different from nontypical ones

A

typical exemplars: 1. are responded to faster (put into categories faster)
2. generated more frequently
3. can be used to prime atypical items (but this cannot be reversed)
4. show typicality effects that begin early in life
5. tend to be named first

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
10
Q

prototype theory, how is it different from classical categorization

A

categories are fuzzy (graded; tomatoes can be fruit or veg). have a set of characteristic (likely but not required) features rather than defining features. things are matched by matching item with prototype stored in memory

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
11
Q

central tendency

A

categories have a central tendency where exemplars with the most characteristic features are found (basically most average exemplar)

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
12
Q

family resemblance

A

something all category members have even if they are not typical members. says that for one to be part of a category, it must share at least one feature with at least one other member

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
13
Q

prototype

A

average of all category members. most typical member of a category. doesn’t have to exist. mental representation

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
14
Q

issues with prototype theory

A

typicality depends on context. for birds: robins are common in north america but sparrows are common in china. conflicts with theory that typicality depends only on number of shared features.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
15
Q

exemplar theory

A

we store specific exemplars and can create a protoype if necessary. if it is similar to other existing category members and not with other categories, then it counts.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
16
Q

similarities between exemplar and prototype theory

A

both believe we can store exemplars and can use prototype. what we do use to determine membership is different

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
17
Q

does exemplar theory require storing too much information

A

no LTM has lots of room

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
18
Q

why can exemplar theory explain context effects

A

assumes categorization based on personal experience

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
19
Q

how to make a prototype with exemplar theory

A

consider all exemplars stored in memory

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
20
Q

problems with organization theories based on similarity (prototype and exemplar)

A

similarity based theories were proposed because of typicality ratings. if this is the case, then if typicality ratings were due to experimental procedure, the reasoning behind the similarity theory falls apart. both prototype and exemplar don’t tell you what features to compare specifically. categories may be arbitrary

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
21
Q

why can exemplar theory explain typicality effects

A

because typical items are similar to many other category members, so it will be easy to retrieve from memory.

22
Q

explanation based theories for categorization

A

knowledge and beliefs determines what is important for category membership. complex implicit ideas, including an idea of the cause of category membership (psychological essentialism, basically the dogginess of something). accounts for why we judge some features as more important than others for category membership

23
Q

psychological essentialism works where and does not work where

A

it applies to natural categories, but social categorization (stereotypes) assume human essense, but these are not natural categories

24
Q

how does psychological essentialism work

A

things have some form of essense that determines membership

25
three basic levels of categories. how do they differ in specificity and broadness
basic, in the middle for both. superordinate, a broad category that is distinct but not informative. subordinate, a specific category that is informative but less distinctive
26
are the levels of categories the same for everyone
no, expertise can influence. the more expert, the more likely to report subordinate
27
where does the levels of categories come from
prototype theory
28
what level of categorization leads to typicality
basic level, because members often have many traits in common
29
semantic network models for categorization
concerned with how different items are related to each other
30
what do all network models have in common
nodes contain info and are connected by directional pathways. nodes are activated by spreading activation
31
collins and quillian's hierarchical model
one mode activated by the environment causes info spread through nodes. takes time for info to spread, further away the node, slower the spread.
32
property inheritance
you store properties once at the highest level and inherit it so you can conserve cognitive resources. everything under a node takes its properties too
33
cognitive economy
tendency to conserve cognitive effort and resources
34
issue with collins and quillian's hierarchical model
atypical objects don't have latencies of info spread through nodes. you respond faster to "a chicken is an animal" than "a chicken is a bird". failed to account for typicality effects
35
collins and loftus' semantic relatedness model
environment still activates nodes and spreads, but there is no hierarchy. only semantic relatedess. nodes are organised based on the strength of relationship, and stronger associations/typical exemplars are closer. this accounts for typicality effects. going further is still slower
36
artificial neural net
computing systems that are models after how neurons in the brain are connected. based on how brains work. basically a node is a neuron, and either excitatory or inhibitory (no info transmission). it is either 0, -1 or 1
37
ANN layers, how are they connected
nodes in input, output, and hidden layers. connected by weights
38
how do ANN weights work, how do they get tuned
the nodes have totally random connection weights at first, then yields a random output. it is corrected, back correction info flows back, randomly changing connection weights, leading to slightly different layer propagation, and now we learn
39
how many hidden layers can you have in ANN
one or more. you need at least one because you need 3 total
40
how is knowledge contained in ANN
in the distribution of weight between connected nodes. info is stored as pattern of activation, thus dictating output
41
what do circles in diagrams mean generally
unobservable behavior
42
what happens if cells are damaged in ANN
pattern of activation changes, but not all knowledge is lost because there are still many functioning cells. only specific categories of knowledge are damaged
43
embodied cognition
a theory closer to neural net than classical view that tries to capture grounding by considering information between brain, body, and environment in shaping thought
44
representational perspectives on knowledge have what in common
symbolic (something in world that you represent, but the representation doesn't have to physically look like it), amodal (don't need visualization in your brain; does not matter if its audio or vision), decontextualized (situation doesn't matter), stable (info accessed is always the same, as a consequence of decontextualization)
45
embodied cognition differs from representational perspectives on sorting knowledge how
simulations (mentally simulating stimuli as if the thing is there), modality specific (smell and see are different), situated (depends on context, apples in an orchard/bakery are different), dynamic (info accessed depends on situation)
46
barsalou's situated simulation theory
a stronger version of embodied cognition. the body and sensorimotor system have a causal role in generating intelligent behavior. cognition requires activation and simulation of sensorimotor system. distributed brain activations from sensory and motor sensors are representations of concepts (modal, depends on sense). knowledge is flexible and goal driven
47
zwann, stanfield and yaxley study on how context matters for mental representations
if an egg was in the right physical configuration as expected in the sentence presented to them (chef saw egg in fridge/pan), people were faster in saying that the egg was mentioned in the sentence
48
hauk et al study on modality specific brain activations for verbal labels
performed actions using foot, finger, tongue. then read action words kick, pick, lick (silently without motor output). active brain areas are very similar. indicates that knowledge of concepts require some stimulation. says that brain activation energy is not arbitrary, it is distributed by brain pattern
49
how is organization different from hauk et al's brain activity studies and classical categorization
hauk says it is organized based on mode (like arm/leg/tongue) instead of by concept (cat, dog)
50
how does barsalou's situated simulation theory (strong) differ from the weak version of embodied cognition
weak version says that the body influences cognition (i.e. no stimulation of the brain in motor-similar patterns required), while the strong version says that the body causes thought.