Quiz 3 Flashcards
What do concepts allow us to do?
- apply general knowledge to new cases
- draw broad conclusions from experiences
concept of dog
how do you know what a dog is?
why do you have expectations when you know that a creature is a dog?
definitional- dog
an animal with four legs that barks and wags its tail
problem with the definitional approach
it is always possible to find exceptions to our defintions
- tables are flat surfaces with four legs- but what about a table with 3 legs
solution for definitional approach
- focus on family resemblance of members within a category
- no “defining” features
- characteristic features across family members
characteristic features
e.g., brothers in the Smith family USUALLY have dark hair and glasses- key word is usually
prototype theory
the category is characterized by a central member that possesses all the characteristic features (the prototype)
- the most salient features creates one prototype
- does not have to be real
typicality
category membership is judged based on typicality
graded membership
objects close to a prototype are “better” members of the category than objects farther from the prototype
sentence verification task
- used to test the prototype theory- based on how quick the response is
SVT- robins are birds
quick response as it is typical, or similar to the prototype of birds
SVT- penguins are birds
slower as it takes longer for the signal to travel because it is further from the prototype
distance and judgements correlation
judgements about items that are more distance from the prototype take more time to make
production task
-testing the prototype theory- based on category members
- fruits and birds
- participants generally name the most typical category members first- with faster response times
rating tasks
items that are closer to the prototype are rated as more typical of the category
fruit
when you think of fruit, you generally think about apples
basic level categories- bench
too general (superordinate)- furniture
in the middle- bench
too specific (subordinate)- the metal bench outside HSC
basic level categories
- represented by a single word
- default for naming objects
- easy-to-explain commonalities
- basic categories learned first
exemplar-based reasoning
in some cases, categorization relies on knowledge about specific category members (exemplars) rather than the prototype
prototypes vs. exemplars
- prototypes provide an economical summary of the category
- exemplars provide information about category variability, but less economical, and are easier to adjust the categories based on exemplars than prototypes
conceptual knowledge
- a mix of exemplar and prototype
- early learning often involves exemplars
- experience often involves averaging exemplars and prototypes
- with more experience, we can use both exemplars and prototypes to ascertain category membership and recognize objects.
typicality aides
- can aide judgment in prototype/exemplar in category membership, but are often distinct.
atypical features
- do not necessarily exclude category members
Greta Thunberg
is/was a highly atypical teenager
typical features
Presence of all typical features does not guarantee category membership
Children can agree that
a skunk cannot be turned into a raccoon
a toaster can be turned into a coffeeepot
beliefs and prior knowledge
- belief in what is typical and required for category members is the most important
- beliefs are developed through previous experiences
example of beliefs: santa
a person dressed as Santa is still a person
but some children may disagree because they have different beliefs
resemblance
similarity judgments based on resemblance can be uninformative- we can make mistakes
wall outlets example
- if you focus on the right properties, faces and wall outlets are pretty much the same thing
essential properties
-we must base similarity judgement on important, essential properties
what is important/essential
- based on your prior knowledge
typicality influences category judgments
we focus on the features that are important, using resemblance as well
typicality effects reveal the substantial role of prototypes and exemplars
helps us see the effect of having prototype, reaction time
when using a prototype or exemplar, you rely on a judgment of resemblance
fundamental to theory
that judgment of resemblance depends on other knowledge
which attributes to pay attention to? which attributes to ignore?
categorization enables us to
apply general knowledge to new cases (see a new dog but know how to approach it safely due to previous experiences)
draw broad conclusions from prior experiences (could have bad experiences with dog now fear dogs)
inferences can be guided by
typicality (more likely to infer from typical case in typical category)
theories/broader beliefs (making connections based on observation using prior knowledge)
Concepts can be characterized by
features
goal-derived categories
relational categories
event categories
natural kinds versus artifacts
living versus nonliving things
embodied cognition
proposal that our concepts include representations of perceptual properties and motor sequences
hub and spoke model
a “hub” connects and integrates more specialized information (the “spokes”) from other brain areas
damage to the hub (anterior temporal lobe)
leads to loss of general knowledge- the category itself
damage to the spoke (areas outside of atl)
leads to loss of specific knowledge- items within a category
connecting concepts
a link between concepts is knowledge
travelling the knowledge network
The knowledge we have is embedded within the links there is a push to remove redundancies.
If they all the same, common, characteristics, why create multiple nodes? You only need one.
There is different distances between different concepts, robin- is a bird, therefore can fly, concept of flying is further away from robin than it is to bird.
It should take time to activate the features that are further up. By recognizing links, we should predict the time is should take.
some results are problematic with the travelling across network theory
- sentence verifications are faster if the sentence is about a more prototypical stimulus - a robin is close to the prototype, goose is further, longer to make judgments for geese.
- the principle of “nonredundancy” does not always hold- there are situations where the common feature seems to be redundantly replicated differently- peacocks, feature of feathers tightly linked, rather than robin
propositions
the smallest unit of knowledge that can either true or false
nodes
represent concepts and link to form complex concepts
propositional networks
local representations- each node represents one concept or idea
connectionist networks
distributed representations- each idea is represented by a pattern of activation across the network
parallel distributed processing (PDP)
organization of language
sentence
phrase
word
morpheme
phoneme
morpheme
smallest unit that carries meaning
phoneme
smallest unit of sound, not carrying any meaning
categorization of speech sounds based on:
voicing (using sounds)
manner of production (how the air flow is restricted)
place of articulation (where you position your lips and tongue)
speech segmentation
the “slicing: of a continuous speech stream into appropriate segments
coarticulation
in producing speech, adjacent phonemes overlap
perception of speech
prior knowledge and expectations supplement the input
regularly used vocabulary
surprisingly limited
not primed to hear long or infrequent words
phonemic restoration effect
top-down processes change what people hear
nozzle vs novel brain activation
gaps
you replace gaps between sounds with noise, but feel that the sound did not break, so you hear it in completeness
context
aides word recognition
categorical perception
people are better at hearing differences between categories of sound (g vs. k) than within sound categories (different pronunciations of d)
combining phonemes
phonemes-morphemes-words
only some combinations of phonemes are acceptable within a language
ex. the sequence tl is not acceptable within a syllable in English- it cannot make one sound
rules for adjusting phonemes are based on combinations
ex. the s sound becomes a z in words like “bags”
for each word that a speaker knows, the speaker”
knows the words sound
usually knows the meaning
knows the rules of syntax
knows the words semantics
generativity
the capacity to create an endless series of new combinations from a small set of fundamental units
combinations to align with syntax
when new words are created, they can be combined with other phonemes/morphemes to align with the languages syntax
ex. someone who “hacks” would be called a “hacker”
syntax
rules that govern the structure of a phrase or sentence
meaning does not depend on syntax
ex. “me hungry” can still be understood
syntax does not depend on meaning
ex. “colourless green ideas sleep furiously” - Noam Chomsky- cannot be understood
syntax
phrase structure:
noun phrase-verb phrase-noun phrase
prescriptive rules
rules describing how something is “supposed to be” in the language
- preference, rules change with time, acceptance with saying new things
descriptive rules
rules describing the language as it is ordinarily used by fluent speakers and listeners
ex. phrase-structure rules
constant- do not change
parsing
the process of determining each others syntactic role in a sentence
occurs as you read/hear the words
garden-path sentences
initially suggest an interpretation that turns out to be incorrect
ex. you need to revisit your parsing
examples of garden-path sentences
the secretary applauded for his efforts was soon promoted.
the old man the ships.
The horse raced past the barn fell.
interpretations are further influences by
function words
morphemes signalling syntactic roles (ex. “-ly” signals adverbs
background knowledge
people tend to
seek the simplest interpretation
assume active-voice
extralinguistic context
the physical and social setting in which we encounter sentences
prosody
the patterns of pauses and pitch changes that characterize speech production
pragmatic rules
rules that govern how people actually use language
ex. conversational “rules” expect you to be cooperative when conversing
common ground
participants in conversations fill in gaps by drawing on their common ground
often established in early stages of conversation
biology of language
fluent language seems to depend on innate neural machinery
evidence of innate neural machinery
aphasias: nonfluent (Broca’s) vs. fluent (Wernicke’s)
most kids converse well at age 3-4
specific language impairment (interferes with the development of language skills in children do not have hearing loss, if there is no other cause)
neural machinery seems to depend on language exposure
wild/wolf children
ex. Ramu who was raised by wolves
- we cannot remake this as it is unethical
active learning
children are sensitive to patterns and regularities, deriving broad principles from what they hear
add “-ed” to make verb past tense by approx 3 years old
overregularization error
an error in which a person perceives or remembers a word/event as being closer to the “norm” than it really is
ex. “yesterday, i thinked,” or “look at those cute gooses”
linguistic relativity
the hypothesis that people who speak different languages think differently as a result
eg. a language’s colour vocabulary and categorization may affect how its speakers perceive and remember colour
spatial terminology
absolute directions (east or west) versus relative directions (left or right)
descriptions of events
active voice (sam made a mistake) versus passive voice (mistakes were made)
memory for agents in accidents
bilingualism
children raised in bilingual homes learn both languages as quickly as monolingual children learn one language
tendency to have temporarily smaller vocabularies than monolingual children at an early age
expectation that bilingual children would confuse language has not been upheld
