chapter 9: knowledge Flashcards
define schematic knowledge
general knowledge about something in the world that we gain through experience
name the functions of semantic knowledge
- use prior knowledge to make inferences
- helps us predict how to behave in a new situation
- helps us create an organized, connected understanding of the world
define category
set of items that are perceptually, biologically, or functionally similar.
- all items are equivalent
- individual items are called exemplars
define concept
mental representation of objects, ideas or events
- groupings of things “inside” our minds
- depends on how each person assigns meaning to the world
what is the commonsense knowledge problem
computers don’t possess the same commonsense knowledge as humans because knowledge has to be explicit with AI compared to human’s ability to infer implicit knowledge
which of the following statements is correct?
a. poodle is an exemplar of the dog category
b. poodle is an exemplar of a dog feature
c. poodle is a feature of the dog category
d. poodle is a category of a dog concept
a. poodle is an exemplar of the dog category
explain the classical view of categorization
categories are well-defined by clear features that are necessary and sufficient
use the example of the grandmother category to explain the necessary and sufficient features
must have a child who is a parent (necessary)
but no other criteria: can be senior, have gray hair, etc (sufficient)
Wittgenstein identified a problem with the classical view of categorization. what was it?
nearly impossible to identify defining category features for most categories
- look for features that exemplars have in common to determine category membership
what is a typicality rating
an experimental task in which Ps rate how good an exemplar is of a category
define typicality effects
observations that we behave differently toward typical items compared to atypical ones
describe the procedure called lexical decision task
Ps are shown a string of letters on a computer and are asked to indicate whether they spell a real word or not
define semantic priming
occurs when a person’s response time on a task is faster if it is preceded by a semantically related word compared to a word that isn’t semantically related
- phenomena of lexical decision task
give 4 examples of typicality effects
- name typical category members before atypical ones
- faster to put typical members into categories than atypical ones
- typical exemplars show larger priming effects than atypical ones
- infants learn typical category members first
what is the prototype theory of categorization
consider which features are most likely among category members
- characteristic features: likely to belong but not required
- fuzzy boundaries
- family resemblance
according to the prototype theory of categorization, what is a prototype?
average of all category members so as a consequence, most typical member of category
according to the prototype theory of categorization, how do you determine whether an item belongs to a category
compare it to the prototype to look for overlapping characteristics
- the more similar, the more typical a category member it is
name 2 problems of the prototype theory of categorization
- typicality depends on the context
- doesn’t account for atypical category members
name and describe the alternate theory of categorization to the prototype theory
exemplar theory of categorization
- compare new items with the ones in memory and look for similarities between features
how does the exemplar theory of categorization explain the problems from the prototype theory
typicality effects: typical items are similar to other category members so it’ll be easier to retrieve those members
context effects: depends on personal experience
what was the conclusion drawn from Dopkins and Gleason’s rectangle categorization experiment
Ps base their categorization on similarity to previously seen exemplars rather than similarity to prototype
what are two things both the prototype theory and the exemplar theory fail to account for
- typicality rating
- doesn’t specify how we decide which features to compare
explain the knowledge-based theory of categorization
- rely on our broad knowledge base to explain reasons for category membership
- implicit
define psychological essentialism
categories have a natural underlying true nature that cannot be stated explicitly
what is one consequence of categorization based on psychological essentialism
risk of applying “essential” qualities to social categories in the same way that we do to biological categories
people with higher essentialist beliefs are more or less likely to endorse a variety of stereotypes
more likely
According to Rosch, name the three levels of categories items can belong to and differentiate them
superordinate: distinctive, but not informative
basic: distinctive and informative
subordinate: informative, but not distinctive
explain Collins and Quillian’s hierarchical model
- knowledge is stored as concepts called “nodes”
- network of interconnected “nodes”
- hierarchically organized
- each concept has properties
what is cognitive economy and why does the hierarchy model demonstrates it
tendency to conserve cognitive resources and efforts
- bc of limited memory storage
what is property inheritance
nodes inherit properties of nodes higher in the hierarchy
explain the sentence verification task used by Collins and Quillian to test the hierarchical semantic network model
Ps presented with sentences
- asked to press one button if true, another if false
Ps responded faster to sentences that could be answered between fewest levels in hierarchy
Name one problem with the hierarchical model
fails to account for typicality effect
what is the spreading activation model
model in which concepts are organized based on their semantic similarity to each other
- no hierarchy
explain the conclusion drawn from the lexical decision task
Ps are faster to respond to the word “butter” if it was presented with the word “bread” before compared to the word “nurse”
nodes are connected via semantic relatedness
how does the spreading activation model explain the typicality effects
typical exemplars are semantically similar to each other and therefore, activation will spread quickly between them
explain the schema view of knowledge
knowledge is organized into groups of related info called “schemata”
- schema is everything we know about a particular thing
explain the repeated reproduction task
Ps shown a stimulus
- asked to reproduce after a delay over and over again
- reproductions become less similar to stimulus and more like a familiar object
- details are lost from memory but our schemata help guide memory retrieval
what is the symbol grounding problem
symbol systems need a way to connect to the real world to avoid an endless cycle of symbolic representation
how did AI address the symbol grounding problem
robotics
- receive sensory input
- manipulate objects to be in direct access with env
explain the neural network model
knowledge is contained in distribution of weights between connected nodes
- weight determines which pattern from inputs produce specific outputs
describe the phenomenon known as graceful degradation
damage to part of the network results in relatively few deficits because info is distributed across network and no single node contains info
define category-specific deficits
loss of semantic knowledge from one category but not another as a result of brain damage
what is the “black box” problem of neural network models
difficult to determine why a neural network made the response it did
- info represented in values of distributed weights, not meaningful semantic units
explain the embodied cognition theory that supports the idea of interaction between brain, body and environment
weak view: body influences cognition
- body position at encoding and retrieval
causal view: cognition is grounded in sensory experiences
- knowledge uses similar sensorimotor neurons as perception and action
what is the major difference between embodied theories and classic theories (prototype and semantic)
embodied: knowledge is driven, flexible and context-dependent
classic: knowledge is abstract and independent of context
what is semantic dementia
deficit in naming ability, comprehension in language and object use
- anomia
semantic dementia is associated with the degeneration of neurons in which part of the brain
anterior temporal lobe (ATL)
why isn’t it possible to say semantic memory resides in the ATL
- damage to ATL doesn’t always result in the same pattern of semantic knowledge loss
- low activation of ATL during tasks involving semantic memory
what is the hub-and-spoke model
hub: generalized and abstract knowledge is stored
spoke: context-dependent and modality-specific detail is stored
What is the transcranial magnetic stimulation (TMS)
- non invasive technique
- use magnetic field to stimulate cortical neurons in localized part of brain
how did the use of TMS support the hub-and-spoke model
ATL stimulated: time to name all objects increased
IPL stimulated: time to name non-living things that could be manipulated with hands increased.
- non living things that can’t be hand manipulated unchanged
which brain region was developed in experienced taxi drivers
posterior hippocampal volume ↑ as the years of experience ↑
what is highly superior autobiographical memory (HSAM)
- enhanced autobiographical memory
- no mnemonic strategies
what are some downside to a detailed memory
- ocd
- difficulty forming social networks
define generalization
process of deriving a concept from a limited number of specific cases and applying it more widely to help you label new instances
what is conceptual expansion
process of thinking outside traditional conceptual boundaries
what is the perceptual symbol theory
access diff features of a concept for a given task
- bc diff features r processed differently
describe the property verification task
- presented with 2 concept-senses match
Ps faster to respond if previous trial asked abt a feature from same percept
