PSYCH 315 Exam 3

conceptual knowledge

enables us to recognize objects and events and to make inferences about their properties

concepts

categories in the mind (i.e. “bird” or “chair”)

classic view of concept representation

concepts are definitions represented as a list of defining features; object has to have defining features of category to be part of that category (membership is all or none)

issues with classic view of concept representation

categories do not have clearly defined boundaries, some categories are hard to define, category membership is seen as all-or-none when it is not

influences of typicality on cognition (typical items are/have)

thought to be category members; faster speed of categorization; learned before atypical ones; makes learning a category easier; understood more easily in language comprehension; tend to be said before atypical one in language production; influences judgements about attractiveness

sentence verification task

apple is a fruit is faster recognition because more typical than pomegranate

what determines typicality?

family resemblance, item frequency, feature frequency

family resemblance

members of a category have a family resemblance to each other

prototype theory of concept representation

potential members of the category are identified by how closely they resemble the prototype

prototype

average of category members that are commonly experienced (summary/set of weighted features (ex. prototype “bird” vs. three real birds will share similar features))

evidence that we create a prototype

posner & keele: participants saw dot patterns, then shown old examples, prototype, and new examples → correct classifications: old 87%, prototype 85%, new 67%

issues with prototype theory of concept representation

doesn’t capture boundaries of concepts, typicality is context-dependent, typicality and category membership don’t always go together (prototype says they do)

exemplar-based reasoning concept representation

individual instances (exemplars) are stored in memory, rather than a prototype or rule; to categorize a new instance, you match it to stored exemplars

category hierarchies

subordinate → basic → superordinate

basic-level categories

default, intermediate level of specificity (i.e. chair vs. desk chair), easy-to-explain commonalities

evidence that basic-level are privileged

preferred level for naming objects, recognized faster, easier to state commonalities, first learned by children

basic-level categories around the world

which level is basic is not universal due to cross-cultural and individual differences

typicality and category membership

don’t always go together → armstrong: rate each number on list for how good of an example it is for category “even numbers” → some numbers rated better than others, even though this is mathematically absurd

membership is independent of

resemblance

membership without resemblance

a flattened and striped lemon/white and yellow striped shoe can be categorized as a lemon

resemblance without membership

counterfeit money → real and fake look alike but not categorically same

knowledge of the world is used in learning and thinking about concepts

when we learn new concepts, we try to connect them to knowledge we already have

psychological essentialism

the belief that members in a category have an unseen property that causes them to be in the category and to have the properties associated with it

example of psychological essentialism

kelli: preschool kids asked to turn skunk into a raccoon → a skunk painted like a raccoon is still a skunk → kids believe identity is essence, not appearance

issues with psychological essentialism

not applied to all properties of a category (we can be a doctor without having doctor parents), not our only belief that plays a role but also knowledge in general matters, can help when dealing with most of the world but less so when applied to humans

concepts as theories for concept representation

organization of concepts is knowledge-based not similarity based, concepts are theories that describe the facts/beliefs about categories and why those members cohere

inferences based on concepts as theories

categorization is important bc it lets us make inferences (can apply knowledge to new cases, draw broad conclusions, etc.), inferences about categories are guided by typicality and our background knowledge that relates to the concept

semantic knowledge network

collins and quillian: nodes are concepts, links are relationships (i.e. animal → “can move” → bird → “can fly”)

principle of cognitive economy in knowledge networks

properties are stored at highest possible level, concepts below inherit these properties (i.e. “has wings” is stored at “bird” but not at “robin” or “ostrich” because they inherit the properties) → avoids redundant storage and males network more efficient

retrieval of data in knowledge networks

the farther you have to travel in the network, the longer it takes (going by properties (canary can sing, fly, has skin) is faster than categories (canary is canary, then bird, then animal))

issues with semantic knowledge network

typicality effect (ideally all category members should be equal, but in some cases, faster to identify robin as bird than ostrich); hierarchical structure (expects more levels to take longer, but sometimes pig=animal vs. pig=mammal takes same time); association effects (speed should depend on distance but “robin has feathers” is a stronger association than “peacock has feathers”); nonredundancy may not hold (properties should be stored once at highest level, but often stored in multiple levels)

communication system

transmission of a signal (sound, motion, etc.) that conveys information

language vs. communication system

humans are not hard-wired to only learn language of biological parents, can talk about anything (not present, lies, language, etc.), productivity

productivity

we can create an infinite number of utterances by combining a finite number of discrete linguistic units in different ways

animal vs. human communication systems

no animal communication system has all features of human ones → no animal has language

phoneme

smallest unit of sound that distinguishes words (i.e. /p/ vs. /t/ in “pin” vs. “tin”)

morpheme

smallest unit that carries meaning (i.e. “talked” is talk+ed)

differences between humans and other great apes on language

universal acquisition in children, variable acquisition in apes; differ in ease of learning; children experiment, apes copy; differences in usage

kanzi learning language

bonobo who learned lexigrams (symbols) and understood some commands, but has limited grammar and less productivity than humans

nativist view on language learning

noam chomsky: innate language learning device, language input we receive is too poor to learn language, language and cognition are independent

anti-nativist view

general-purpose learning device; we receive enough info in language input to acquire language if we are actively engaged in our environment; language and cognition are interlinked

broca’s (nonfluent/expressive aphasia) aphasia

left inferior frontal cortex, dysfluent agrammatic speech, good comprehension but trouble with grammar

wernicke (fluent/receptive) aphasia

posterior left temporal lobe, poor comprehension, fluent but often meaningless speech

williams syndrome

genetic anomaly (1 in 7,500-20,000 births), cognitive disability (IQ 50-70), hypersociability, better vocab but poorer syntactic processing compared to mental age controls

developmental language disorder/specific language impairment

diagnosed when child’s language development is deficient for no apparent reason, cause unknown but seems hereditary (7 in 100 births), typical cognitive abilities but impaired language at all levels

what is the general-purpose learning device that helps us learn language?

statistical learning

statistic learning in artificial language learning paradigm

listen to artificial language stream and identify what follows a phrase → sequences of syllables occur more often together within than across words, able to detect patterns in input

statistical learning findings

present at birth, domain-general, not unique to humans

difficulties with speech perception

coarticulation (overlapping sounds), invariance problem (same sound varies across contexts) → still understand speech easily (achieve perceptual constancy)

mental lexicon

stores sounds, spellings, meanings, and syntax of words → likely 50,000+ words stored

visual word paradigm for lexical competition

eye movements track word recognition

allopenna study

word “beaker” activates cohort (“beetle”) and rhyme (“speaker”)

word recognition in bilinguals

words from both languages can compete for recognition even when just listening to one language, but monolinguals do not have cross-language competitors

benefits of being bilingual

better cognitive control throughout lifespan (flanker test), less age-related decline in cognitive control

sapir-whorf hypothesis

language determinism vs. relativism

language determinism

language determines certain nonlinguistic cognitive processes (i.e. the way we think, remember, and perceive) (language determines thought)

linguistic relativism

to the degree that languages differ from each other, different languages will determine nonlinguistic processes in different ways (language influences thought)

vocabulary differentiation

some cultures have single words available for concepts that others may take many words to describe

issues with vocabulary differentiation

counts have been inflated, difficult to define what constitutes a word, evidence is weak that difference in words is linked to difference in thinking, and if there is a link problems in determining causality

color words study

speakers of languages with one name for two colors can still tell those apart (i.e. green/blue), but learning another language that has two names will improve their color perception

strongest version of sapir-whorf hypothesis

language determines thought → not supported bc speakers can still understand concepts they don’t have words for

weaker version of sapir-whorf hypothesis

language affects only perception

weakest version of sapir-whorf hypothesis

language only affects processing on tasks where linguistic encoding is important

color perception winawer task

russian and english asked to determine which samples match other samples; russian had 2 colors, english had 2 shades of 1 color

color perception winawer task results

russian speakers did better if samples labelled differently rather than same, english speakers showed no difference; but if russians had to rehearse numbers during task the difference went away → supports weakest version of whorf’s hypothesis

fausey and boroditsky study

english vs. spanish → english-speaking participants remember the agent of the accident (“she” broke the vase) better than spanish-speaking participants (“the vase broke itself”)