Language Acquisition Flashcards
Major milestones
6 months - babbling 1 year - first word 1.5 - word spurt 2 - combine sentences and words There is lots of variability
Ways of using language
Comprehension - understanding
Production - actually speaking
Comprehension precedes production - understand more than can speak
What is there to learn? (5)
Phonology - unit of sound that distinguishes meaning (Bear vs Pear)
Grammer/synax - rules for combining words
Semantics - word meanings
Morphology - smallest units of meaning in language
Pragmatics - conventions, social rules
What are the theories of development?
Nativist argument
Behaviourist
Interactionist
Nativist argument
Believes language is too complex to be learned so quick by children Chomsky - language acquisition device, module in the brain for learning, starts out with parameters and your language tunes them for development
Behaviorist argument
Language is learned through CC and OC
Skinner: correct usage is reinforced, incorrect is not
But: challenged by: poverty of the stimulus (come up with this we’ve not heard) and negative evidence - parents don’t always say that’s wrong
Interactionist approach
Bates - language is genes and the environment. Perceptual and learning abilities molded by culture
How do we test how infants develop speech perception?
High amplitude sucking
24 hour infants, headphones and dummies hooked to a computer
DeCasper and Fifer 1980 - HAS
Infants worked to produce mothers voice - learned cause and effect between sucking and hearing voice
Infants had a preference for mothers voice - could distinguish between mum and stranger
Shows prenatal auditory experience helps shape voice preferences and parent infant interactions after birth
DeCasper and Spence 1987
Tested 55hr old infants, HAS procedure
pregnant mother read one Dr Suess story aloud twice a day during last trimester. Infants heard their own mother or another mother read the familiar story or novel story
Infants worked to produce familiar story - who read the story had no effect
Control group had no preference but preferred mothers voice
Prenatal learning occurs as no stories were read postnatally
What do infants prefer? Order
Mums voice filtered to how it sounded in the womb
Mother over father
Mum/dad over stranger
Stranger’s speaking language than another language
What is categorical perception?
The classification of continuous stimuli (colours e.g.) into distinct categories with sharp boundaries. There aren’t actual boundaries as it is continuous
The sounds that we hear are on a continuum
Categorical perception of sounds
BA and PA are the same sound except for when people start making the sound. Only learn there is a difference because it means someone to us in language. early on we are sensitive to all differences but later on realise some don’t matter to us. We perceive anything with VOT>25 as P and VOT<25 as B
Can infants discriminate speech sounds
Yes all infants can do this
Development of distinguishing speech sounds
Around 6-8 months, infants can distinguish contrasts that aren’t used in their native language
This ability declines over the first year, by 12 months, only retain contrasts used in their native language - use it or lose it
What do infants categories result from?
The distribution of phonemes in their environment
English adults can discriminate the Hindu phonemes with practise
Japanese can discriminate r/v with training
Need to be getting the input and listening
Segmenting the speech stream
In words, can be seen as one long sentence
Hard to make sentences individual words
Infants find the breaks very quickly but without context, adults find it hard
Saffron et al 1996 - speech stream
How do we know where the breaks are?
Prosody - changes in pitch
Pauses - in between words
Correlations - statistical regulations
Familiarised infants to 2 mins of speech stream with 3 repeating non sense words
The frequency of sounds that span a word boundary is lower than the frequency of sounds within a word
Eg: pretty baby pre-ty -within a word, higher frequency
Ty-ba - spans 2 words, frequency lower
Testing infants with words vs non words
Words - Padoti and golabu
Part words - Sakura and bubida
Infants preferred the part words, they noticed a difference between part words and words, even though they were both familiar
Infants can use correlations from the environment to learn language
What is infant directed speech?
highly grammatical simple structure exaggerated slower rate vary pitch and loudness
Why is learning words difficult?
The possibilities of word meanings are endless
How do they know what a word means?
Mutual exclusivity
Process of elimination
How to aid children’s word learning
Adults - scaffolding: pointing, use voice, use routines, refer to previous experience, joint attention
Children - be active learners, follow adults gaze, point, pick up objects and ask, look at objects
Baldwin - do children use joint attention?
Two novel toys to play with
1 in bucket, one in view
Experimenter waits for child to focus on visible toy, then labels: visible toy (follow condition) or bucket toy (mismatch condition)
Novel label (peri or toma) repeated 4 times
At test, presented both toys in a neutral location and asked ‘get the toma’
Follow in condition - easily learned the words
Mismatch - more errors
Child look at experimenter more in mismatch, mismatch of attention and experimenters voice - children are sensitive to joint attention
What is fast mapping?
Ability to quickly link a novel name to a novel object, typically by applying known information
Fast mapping experiment
‘Bring me the chromium tray, not the blue’ they picked the green tray-13/14
One week later asked ‘which one is chromium’- 9/13 chose green or olive green
How longer do children remember words from mutual exclusivity for?
5 minutes
Number of competitors
30 month old children show objects and asked ‘can you find the dite?’ using 2, 3 and 4 competitors
Only children with few competitor (2) retained the new words. Competitors hinder word learning via mutual exclusivity (fast mapping)
Replication with icub and same results found
Number of illustrations
3-5 year old children, 12 naming tokens Read children books with either: two pages of illustrations one page of illustrations one a3 page At test 'can you point to the tamin?'
Children with fewer illustrations learned significantly more words than multiple illustrations
3D - did well
2 illustrations - didn’t do well
Problems with shared story reading
Child doesn’t know what page the adult is on, so doesn’t know where to look etc, can’t have a joint focus of attention
Solution to shared story reading
Gesture what page you are reading
Changed the results - 2 illustrations, could learn the words
Shows children can learn words in complicate ways, but they really do need support
What is most children’s early vocab?
Nouns for object categories with are solid, shape based and names with count noun syntax (being able to count, eg 5 cookies)
What is the shape bias?
The tendency to name things on the basis of a shape
What is the shape bias a function of?
Vocabulary - before knowing 120 nouns (2 years) they don’t show a shape bias
If it is learned by learning words, we should be able to teach a shape bias by changing the words children know
Samuelson 2002
Taught children 12 real nounss once a week for 9 weeks and a 1 month follow up
Multiple instances of each category, categories usually learned much later (26 months)
Used shape training: bucket, pear, ladder
Material training: lotion, jello
Results: showed a shape bias for shape nouns but not for material bounds
1 month later - if they were taught shape nouns they developed a precoscious shape bias
Teach material nouns - no material bias
How do children learn syntax?
How words go together
Have to be able to represent something abstract about the words
What nouns, verbs, adjectives and adverbs are
What the rules are about putting them together
Young children begin showing signs of syntax with simple sentences - usually two words, noun plus noun or verb
Telegraphic speech- saying as few words as possible to get your point acrosss
Noun and a verb
Morphology
How do you change words to reflect subtle parts of meaning
Often hardest part
They go through phases off learning - past tense verbs
Past tense verbs
We have an unusual way
Most common verbs are irregular, they do not follow the ed rule for past tense
Verbs kids learn early on - come came did do get got give gave go went
They tend to show a u shaped trend in past morphology
First correctly conjugate irregular words
Then over regularise irregular words
Finally return to correct usage
WOGG test
Shows how children learn rules of language to nonsense words
Showing them images of something ‘there are 2..’ what are there?
See if they understand the tense
Using novel words makes it easier to see whether they are learning by using words they could’ve never heard before
