SPAU 3303 - Exam 4

Normal Language Development

phonetics

study of speech sounds

prosody

the intonation contour of speech, including pauses and changes in stress and pitch (the overall melody of language, stress patterns)

morphology

a system of rules for combining the smallest units of language into words

semantics

meaning of words and sentences

syntax

grammatical rules of putting words together; the system for combining words into sentences

grammar

-the umbrella term for syntax and morphology; structure of a language that includes syntax and morphemes
-a set of rules that specify how the units of language can be combined to produce meaningful messages

chomsky's theory of language aquisition

-countered skinner's argument with an extreme nativist/nature argument
-believed that language is biologically programmed in humans and is innate
-noted similarities between languages
-believed that behaviorist theories are wrong because language develops: rapidly, effortlessly, and without direct instruction

chomsky's universal grammar

rules underlying the structures of all languages

skinner and behaviorist approaches

-language is learned through imitation, rewards, and punishments
-try to show that correct imitation is rewarded with social interactions (first looked at animals and then thought it would be the same for children)

connectionist approaches

-a theoretical perspective that holds that thinking consists of activating connections in a network of interconnected nodes and of activation spreading in this network along paths determined by the strengths of the connections among these nodes
-a network of interconnected nodes between which activation spreads to give "cognition" or thoughts (associations and concepts)
-thought or concept \= particular pattern of activation
-learning \= strengthening connections between nodes (words or concepts)

example of connectionist model

-start to associate certain adjectives to people you previously had interactions with
-you met an American who was a jerk so then you start associating Americans to being jerks

nativist approaches

nativism is the view that knowledge is innate, as opposed to being learned from experience

nature vs. nurture debate in language development

-two contrasting views of the determinants of development
-nature view: development is biologically determined
-nurture view: development is shaped by experience and environment

statistical learning in infancy

-it suggests that language is NOT innate, and is an ability that we can learn
-counting the frequency with which one stimulus is followed by another
-it is a mechanism NOT a theory; it is what you think that they do with the statistical learning that makes it a theory
-learning of the co-occurrence probabilities of experienced stimuli
-one mechanism for learning the patterns in the input that could contribute to learning language
-e.g. babies presented with sequences of sounds appear to learn the conditional probability of one sound following another in the sequence they heard

word segmentation from statistical learning

speech is not produced with spaces between the words --\> the child must find the word boundaries in a continuous stream of sound

methods for studying language production in children

-high tech options (TED talk by Deb Roy)
-parental report (dairies)
-tests in labs (studies to collect speech samples; standardized: PPVT - Peabody Picture Vocabulary Test)
-speech samples: child language data storage system (CHILDES)

habituation/dishabituation

-when does a child get used to a stimulus that they don't react to it anymore
-a general term, in this case, based on boredom
-habituation: become acclimated to (bored of)
-dishabituation: regain interest (e.g. visual stimuli and measuring looking time)
-what this tells us: infants can categorize and differentiate (can tell when something is different)

high amplitude sucking

-study of habituation
-when an infant is interested in something, the sucking rate is faster but when an infant is not interested, the sucking rate is slower
-sucking rate increases with new sound if it is heard as a different sound
-e.g. mom's voice vs. voice of a stranger (all babies go the necessary speed to hear their own mom's voice)

head turn paradigm/procedure

-a procedure used to assess infants' abilities to discriminate stimuli
-the procedure begins by presenting an infant with sound changes that we know infants can detect
-as soon as the sound changes, an interesting and rewarding visual stimuli is presented
-infants quickly learn to anticipate the interesting visual stimulus following a sound change, and they turn their heads in anticipation
-then researchers can use this conditioned response to see when infants can discriminate a change in other stimuli

preferential looking paradigm

-a procedure that assesses language comprehension
-infants are shown side-by-side slides or videos, as the infants hears an audio presentation that matches only one of the videos
-if infants consistently look longer at the matching video, it is taken as evidence that they understand the language of the audio

longitudinal studies

following one subject for a long time (e.g. following/gathering data from the same kid for a very long time)

cross sectional studies

a study in which people of different ages are compared with one another

pidgin language

-used between speakers of different languages
-a form of communication for commerce "primitive grammar" (nouns and verbs)
-a structurally simple language that arises when people who share no common language come into contact

creole language

-grammatically complex language created by the people using a pidgin over time (unique specific language made from a mix of different cultures)
-a language that develops when children acquire a pidgin language as their native language and which is grammatically more complex than the original pidgin language

what we learn from creoles

-language does not need to be imported (humans, mainly children, can create them)
-children create the language (adults use one of the original languages)(the kids are the ones who add the structure, not the adults - the kids change it to creole language; if the adults are the only ones who use the pidgin language then that language would not change to creole)

nicaraguan sign language

-a sign language developed in the 1980s when Nicaragua opened their first school for deaf children
-the children individually already formed their own sign languages; children created their own full language (they had a full grammar and syntax) that was not the same as the one that the teachers were trying to teach them
-over 20 years, the language evolved from being a pidgin language to a creole language
-the children were building the language when they were developing and needing language
-the children were the most fluent compared to the adults and the older kids at the school
-the little kids came up with the rules of the language
-the older people could learn the words but couldn't really grasp the syntax and grammar (as you grow older, you lose the capacity to do so)

double dissociation

-damage to broca's area: syntax is damaged, semantics is fine
-damage to wernicke's area: semantics is damaged, syntax is fine
-indicates the independence of function (language isn't just one thing)
-very simple and historic view

language processes that occur in the LEFT hemisphere

-in order for the LEFT hemisphere to process language, it has to be heard from the right ear
-left hemisphere is the "logical brain"; right hemisphere is the "creative brain"

equipotentiality hypothesis

hypothesis that, at birth, both hemispheres of the brain have equal potential for acquiring language

invariance hypothesis

-the theory that holds that the left hemisphere of the brain has the adult specialization for language at birth
-the left hemisphere is lateralized (one hemisphere is dominantly able to process a certain type of information) from birth

neuroplasticity

the ability of parts of the brain to take over functions they wouldn't normally serve (kids have higher plasticity, other areas of the brain can take over)

synaptogenesis

-formation of the connections between neurons (increase in the connections between the neurons)
-starts 5th week after conception
-largely complete by around age 2
-from 3 months after birth to 2 years of age, 1.8 million new synapses per second

synaptic pruning

-development
-used synapses become stronger and faster
-synapses that are not used will die away
-from toddlerhood to teens you lose about 20 billion synapses a day (those that aren't used)
-benefit: a fast, efficient brain based on the environment
-benefit: parts of the brain can take over new abilities when you are young (neuroplasticity)
-downside: loses flexibility with age

critical period in language

-biologically determined period during which a behavior must appear
-the hypothesis that there is a biologically determined period during which language acquisition must occur, if it is to occur normally

support for critical period hypothesis

-children unexposed to language will get words but not the grammar when taught later
-children who are deaf but born to hearing parents, those who start earlier would be more proficient in sign language (younger learner \= more proficient) --\> not about the years/experience
-e.g. now 12 years old, started signing at 2 (would be more proficient) vs. now 32 years old, started signing at 12)

neurolinguistics

the relationship between the brain and language

dichotic listening task

-identifies hemisphere differences
-if you play "ba" in the left ear and "ga" in the right ear, which sound would you hear? (left hemisphere processes language) you will report hearing "ga" (from the right ear - the right ear sends information to the left hemisphere)

split brain studies

-the corpus callosum which connects the two hemisphere is split
-shows what hemisphere is processing information and how there is no language (it is "mute") in the right hemisphere
-only done for severe epilepsy patients
-video: patient looks at the dot and the word on the left side, the patient had to close his eyes to draw it but if it appears on the right side then he does know what it is and can verbally name it

ERP

-event-related potentials (ERPs)
-an imaging technique
-measures electrical activity (portions of the EEG related to cognition)
-pros: measures electrical activity; measuring how neurons are working; good temporal precision - good for measuring what is temporary in the brain and the sense of time (1 sample per milliseconds); across lifespan; cheap
-cons: bad at showing where in the brain something happens (bad at localizing activity); the output is hard to understand without training

MMN

-mismatch negativity
-shows the differentiation between two stimuli
-typical output from ERPs
-e.g. can kids/the brain of the kid differentiate between "ba" and "ga"

fMRI

-functional magnetic resonance imaging
-magnets to measure blood flow in the brain (not measuring activity)
-an imaging technique
-pros: measures blood flow; good at localizing activity
-cons: expensive; limited movement (problematic when studying children); slow (1 sample per 2 seconds, language takes 2-3 words per second)

PET

-positron emission tomography
-inhale or inject low-level radioactive substances; used like MRI to measure blood flow
-an imaging technique

NIRS

-near infrared spectroscopy
-an imaging technique
-"baby MRI"
-uses light to transmit through the skull and blood to determine blood flow (only applicable to kids and infants since adults' skulls are thicker)

joint attention

-two people attend to the same object or event
-3 criteria: 2 people and an object (can be parent and child lead; e.g. eye gaze or pointing to elicit attention)
-increases from 10-15 months; predicts language
-joint attention supports language but isn't necessary; only 25% of child/parent interactions have learning
-e.g. gaze following and communicative pointing

gaze following

-child looks where a person looks
-around 10-15 months

communicative pointing

-child points to an object and looks at a person
-10 months to 1 year
-a request or provide information (for the want to share experience and to communicate)

intention reading

-realizing others have intentions
-babies at a young age already know humans are intentional people (and those intentions have to be fulfilled)
-early evidence that infants attribute intentions to others comes from the finding that they imitate others' intended acts (e.g. string in the cup experiment)
-discerning what intentions are can take longer for kids to understand

intention reading used for language learning

-new labels go to what the adult is looking at, not what the child is looking at (around 24 months)
-social information and eye gaze for labeling

what infants prefer

-language (human voice) to other audio
-mother's voice to other women
-stories they've heard in the womb (hearing the overall prosody/rhythm of the story)
-native language (the overall prosody/rhythm of the language)
-familiarity (based fully on what they are exposed to)

infant abilities

-differentiate: all languages' phonemes \= "universal listener"
-link visual face cues to language sounds at around 5 months of age
-prefer linked audio and visual (e.g. we hate it when the audio doesn't match up with the visual)
-linking up how the human mouth has to move in order for sound to be produced
-but after 5 months, they get tricked by the McGurk effect (integrating the visual and audio information together)

phonological knowledge

knowledge of the sounds and sound patterns of a language (phonological differences among languages are not just in what sounds are used but also how the sounds are used)

phones

-all of the different sounds that can be used in a language
-200 possible, each language uses about 45
-e.g. [p], [pᑋ], and [b]

phonemes

-a phonetic feature that creates a difference in meaning (speech sounds that signal a difference in meaning in a particular language)
-distinctive feature of language
-e.g. /b/ and /p/ - "bat" and "pat"

allophones

-sounds/phones that occur in a language but do not differentiate meaning (varies across languages)
-as English speakers, we produce both phonemes and allophones all the time but we don't hear the difference easily and we have to consciously work to produce the difference

/r/ and /L/ sounds

-in English, they are distinct and differentiate meaning (phoneme); e.g. "rent" vs. "lent"
-Japanese has both but they do not differentiate meaning (they are allophones in Japanese)

/z/ and /s/ sounds

-phonemes in English
-allophones in (some dialects of) Spanish

\[pᑋ] vs. [p]

-[pᑋ] aspirated (burst of air) like in the word "pin"
-[p] non-aspirated like in the word "spin"
-allophones in English
-but phonemes in Thai: [paa] means forest, [pᑋaa] means to spit

phonological rules

-other rules of how sounds go together in a language
-voicing assimilation, plural, and past tense

voicing

-a feature of sound production in which the vocal cords vibrate as air is released in the production of a consonant
-/z/ is voiced; /s/ is voiceless

voicing assimilation

-when two consonants are together in a word, they match in terms of voicing
-e.g. "s" is pronounced /s/ after all voiceless consonants and pronounced /z/ after all voiced sounds (like "wug" -\> wug/z/)
-by the age of 4, English-speaking children display this

phonological advances in production made between 16 to 30 weeks

vocal play

vocal play

-the period between 16 weeks to 30 weeks
-increase in variety of different consonant-like and vowel-like sounds (still not distinct)
-practice, not communicative
-all babies sound the same at this point regardless of language and dear or hearing

phonological advances in production are made between 6 to 9 months

canonical babbling

canonical babbling

-aka reduplicated babbling
-happens at 6 to 9 months
-presence of true syllables
-these syllables are typically produced in reduplicated series of the same consonant and vowel combinations
-e.g. [bababababa] or [dadada]
-rare in babies who are deaf (but they would babble with their hands)
-self-stimulating (non-communicative)
-messing with sounds, but as a parent imitates them, it encourages the baby to keep babbling

phonological awareness

-conscious awareness of the phonological properties of language; such as the ability to count the number of syllables in a word and to identify rhymes
-individual's awareness of the phonetic structure of words
-knowledge of sounds and syllables and of the sound structure of words
-begins at age 2
-predict a child's success in learning to read

mental lexicon

the mental dictionary of words you know and their meanings

word

a sound sequence that symbolizes meaning and can stand alone (morpheme)

features of a word

-arbitrary symbol
-reference

arbitrary symbol

stands for something without being a part of it (e.g. cat - talking about the properties, not just one cat)

reference

-"word to world mapping"
-symbol standing for a whole concept

referential words

-referential: not bound to a specific use (using words how adults use it)
-e.g. the word "dog" for all dogs

context-bound words

-context-bound: tied to a particular context (doesn't understand the meaning)
-e.g. the word "dog" only for the dog in a specific picture book

open-class words

-aka lexical categories
-hold semantic meaning; new items can be created
-nouns, adjectives, verbs, adverbs (e.g. "texting", "hangry", "mansplaining")
-kids learn these first

closed-class words

-aka functional words
-serve grammatical function; new items cannot be added
-prepositions, articles, conjunctions (e.g. "a", "the", "over", "under")

overextension

-using a word too broadly
-e.g. dog \= dog, cow, horse, cat
-e.g. apple \= every fruit in the grocery store

misleading information in overextensions

-they may not truly reflect the child's meaning
-e.g. can't find the word "cow" fast enough so they say "dog"
-it is really hard to take production and assume their comprehension

underextension

-using a word too narrowly (only the category is off but understanding references, not situational/context bound)
-e.g. dog \= my dog Fluffy

word spurt

-around 50 words, children will increase in word learning
-early word learning of 8 to 11 words per month would change to 22 to 37 words per month (word spurt)
-some children show a big increase, but some are more linear

causes of the word spurt

-have a naming insight
-figure out the rules/constraints about how language works
-better at identifying category boundaries
-better phonological abilities

naming insight

they realize that all things have names (realize that everything has labels)

constraints of language

-they realize there are rules to how language works
-e.g. whole object bias: the label applies to the whole thing and not only just one part and anything

category boundaries

successful production of a category with new items

phonological abilities

they get better at storing and producing new words

whorfian hypothesis

-aka the linguistic relativity hypothesis
-the theory that the language you speak dictates how you view the world
-the hypothesis that language influences thought and, therefore, that differences among languages causes differences in the cognition of speakers of those languages
-current view: we do have differences in the languages we speak but it is not a huge difference

mapping problem

-how do I map a word I hear to something in the environment or to the right object?
-indeterminacy of word meaning: the problem that for any particular instance of word's use, multiple meanings are possible in the situation

fast mapping

quickly identifying something in the environment that the word goes to

whole object assumption

-the assumption that words refer to whole objects and not their component parts or characteristics
-e.g. eliminates "white", "handle", "being held in a hand" as possible meanings of "cup"

mutual exclusivity assumption

-a word learning constraint according to which children assume that objects can have only a single name (novel word refers to the novel object)
-e.g. so if a child knows the word "cup" and his or her mother says "this handle" the child won't take "handle" to be a synonym for "cup" but will look for something else to be the referent of the new term

principle of contrast

-a pragmatic principle that, by hypothesis, leads children to assume that different words have different meanings
-allows for multiple labels ("dog" vs. "chihuahua" vs. "animal" - they all have different meanings but could label the same thing)

syntactic bootstrapping

-determining meanings of words based on syntax
-the hypothesis that children find and use clues to the meaning of new words in the syntactic structure of the sentences in which new words are encountered
-e.g. the burmigs glorped in the glemish blicket → burmigs \= turtle in the picture, glorped \= swimming in the picture, glemish \= green murky in the picture, blicket \= water in the picture

comprehension of syntax

-understanding precedes production
-hard to track, hard to test

comprehension strategies

-children don't fully understand, but they can appear to understand
-3 strategies: response strategies, word order strategies, world knowledge

response strategies

-respond to a complex sentence with an action based on key words that they know, not syntax
-e.g. dad asks "do you need to change your diaper?" --\> kid runs to get a diaper --\> changed diaper --\> dad says "there you go, you have a new diaper" --\> kid runs to grab another diaper because he heard the word "diaper"

word order strategies

-treats whatever is mentioned first in the sentence as the subject and whatever is mentioned second as the object
-includes order of mention strategies: what is first in the sentence is what is going to happen first
-e.g. this strategy allows children to correctly act out "the swing bumps the kitty" and "the kitty bumps the swing"

world knowledge strategy

using common sense to determine meaning (e.g. "put you shoes on after your socks")

generativist theory of grammar

children DO have abstract rules and categories from very early on, and they use them to be highly productive (chomskian approach)

constructivist theory of grammar

-children have very limited productivity
-children do NOT initially have abstract rules or categories (more likely to think that kids are using the semantic rule)

verb island theory

(Tomasello) nouns are used in abstract ways by kids but not verbs (they have little islands and they have to learn rules for each verb; learning each verb independently)

verb island hypothesis evidence

-kids learn each verbs' arguments individually
1.) fails to extend argument structures to newly learned verbs (e.g. e.g. learn "she blicks the ball", only use "blicks" in that syntactic structure; "mom blicks the teddy" not "the ball blicks")
2.) children don't spontaneously use even well-known verbs in all forms but they do for nouns (e.g. may use: "playing" and "played" but have to learn when to use them; this inconsistency leads researchers to believe this)

development of different sentence forms

-imperative sentences (commands, e.g. "gimme juice") tend to be the most frequent forms at first
-declaratives become the most frequent sentence by 30 months
-questions are always the least frequent category but they become more frequent as children get older