Language
Acquisition
Nov 24, 2008
L1 Language Acquisition
How can we research child language
acquisition?
1. Observations
2. Surveys/interviews
3. Corpora
4. Experimental
a. pre linguistic
b. post linguistic
Getting subjects. . .
Usually easy to get college aged students .
. . Harder to get infants/children
1.
2.
3.
4.
BYU daycare (or other organizations)
Wards/friends
Birth announcements
Schools
How is getting data from
children/infants different from
getting data from adults?
Can’t always tell you what they are thinking
Lots of attrition (they get bored, tired, scared)
Parental permission, IRB
You have to entertain them more than adults
You have to provide incentives more than
adults
6. They may lie (a lot)
1.
2.
3.
4.
5.
Experimental techniques to
test pre-linguistic infants
1.
2.
3.
4.
5.
Heart rate (pre-natal)
High amplitude sucking/foot kicking
Preferential looking
Conditioned head turn
2 alternative anticipatory eye movement
response
Why would you study pre
linguistic infants?!
Stages of L1 language acquisition:
1.
2.
3.
4.
5.
6.
7.
Prenatal (especially last trimester)
Crying (0-2 months)
Intonational (2-3 months)
Cooing (4-6 months)
Babbling (6-8 months)
Non-reduplicated babbling (9-12 months)
First words (12-18 months)
How well infants discriminate sounds predicts their abilities to learn words—
their ability to learn words predicts overall language and reading abilities later
on
1. Heart Rate
Typical set up:
Mother or Father asked to read story, talk
to, or other language task for 3-4 weeks
in 3rd Trimester
Mother comes to lab and is hooked up to
fetal heart rate monitor
Same and different stories are played to
baby and measure heart rate
1. Heart Rate
Basic findings . . .
Recognition of mother’s voice (Mehler et al.,
1984)
Recognition of prose passage before birth
(DeCasper et al. 1994)
Recognition of native language after birth (Mehler
et al., 1986)
Recognition of difference between male/female
voices (DeCasper et al., 2002)
1. Heart Rate
Pros?
Only way to measure prenatal language
Fairly unintrusive
Cons?
Don’t know for sure that this is best way to
measure language skills prenatally
2. High Amplitude Sucking
http://psych.rice.edu/mmtbn/language/sPerception/infantsucking_h.htm
2. High Amplitude Sucking
Basic Findings . . .
have found for
most types of
consonants
2. High Amplitude Sucking
Pros?
High level of reliability (used a lot for a long time)
Fairly unintrusive
Only way to test newborns
Cons?
High dropout rate
Need equipment
3. Preferential looking
http://psych.rice.edu/mmtbn/language/sPer
ception/infantHeadturn_h.html
3. Preferential looking
Some findings . . .
Preferences for
 Mother’s voice at 3 weeks (DeCasper & Fifer, 1980)
 Own-language prosody at 6 months
 Own language stress pattern at 9 months (Jusczyk et
al, 1993)
 Own language phonology at 9 months (Juscyzk et al.,
1993)
 Own language phonotactics at 9 months (Juscyzk et
al., 1994)
3. Preferential looking
Pros?
Can test lots of things (even up to 2-3
years)
Easy to administer
Cons?
Requires complicated equipment
4. Conditioned head turn
http://beta.vtap.com/video/Learning+From+Babies/CL0173196706_4078e8f9a_V0lLST
E4NjE0fmluOjM4
4. Conditioned head turn
Some typical findings . . .
4. Conditioned head turn
Pros?
Works well with testing sound
discrimination
Easy to administer
Cons?
Only works with testing discrimination
5. Two alternative
response
 http://psych.rice.edu/mmtbn/language/sP
erception/infantlooking_h.html
5. Two alternative
response
some findings . . .
Babies associate highly frequent words with
familiar objects by 6 months (Jusczyk, 1999)
8-10 month olds can tell the difference between
passive and active sentences and transitive
and intransitive verbs (Fisher, 2003)
Golinkoff et al. 1987
*
*
“look! cookie monster’s tickling big bird”
“look! big bird’s tickling cookie monster”
see also Naigles (1990), Fisher (2000)
4. Two alternative
response
Pros?
Works well with almost any type of
language
Cons?
Difficult to administer—works best with 1824 year olds
L1 Acquisition—12 36 months
Stages of First Language
Acquisition—12 months on
 Holophrastic stage 12-24 months




idiomorphs
mutual exclusivity and whole object bias
overgeneralizations, undergeneralizations
referential vs. expressive children
 Two-word Stage—24-30 months




Subject-verb ‘Mary go.’
Verb-modifier ‘Push truck.’
Possessor-possesed ‘Mommy sock’
Content words, no function words
 Telegraphic Stage—30-36 months
 2-5 words with little extra morphology
 Morphological overgeneralization
 Easier, more productive morphemes first
Basic methods for studying
children post linguistically
1.
2.
3.
4.
Observations
Wug tests
ERPs (Event related potentials)
Experimental trickery
1. Observations
Many of these studies done in 1950’s –
70’s
Usually linguists or psychologists would
follow own children around with tape
recorder and analyze their language
development
1. Observations
At 1 year; 1 month (babbling) (in IPA):
Production
Production
/ava/
/baewa/
/aelu/
/daevu/
/aw/
/gigi/
/n/
/paba/
At 1 year, 6 months:
Gloss
Production
baby
/bebi/
bear
/baeu/
bib
/bIb/
blue
/pu/
Gloss
go
big
Brenda
walk
Production
/go/
/gIg/
/pEnt/
/wak/
M: Did you tell Daddy what you had for tea?
N: aga (eggs) and gagadoodoodoo (cockadoodledoo = cornflakes [because of the
picture on the box] cockadoodledoo also means a weathervane on a church
spire and so he remembers his walk as well) tika (sticks) too.
M: You didn’t have cornflakes for tea! And you didn’t have sticks either!
N: ho (holes) too. (laughing—thinking this is a joke)
M: You didn’t have sticks and holes for tea!
N: doba (toast) Go wakin’, dada. (“I went walking, daddy”)
F: What did you see on your walk?
N: see ka (“I saw a car”)
F: Yes, you went for a ride in a car, didn’t you? What else did you see?
N: piti dedi mamma on gara (“a pretty flower (daisy) for momma in the garden”).
[The flower wasn’t a daisy, but he calls all flowers daisies]
(Parents get N. ready for bed)
N: Help? Need help? (meaning “I need help).
M: What is it you can’t find? Is it something under there?
N, looking under couch: ba (“a ball”)
(Later, looking at a book with his mother)
N: ‘ats dat? (“What’s that?)
M: That’s butter.
N: (repeating) taba (butter) ‘ats dat? (“What’s that?”)
M: Some ducks and a doggie.
N: Sa kuks (“some ducks”) No goggie (but there isn’t a doggie)
1. Observations
Basic findings:
Language Development: Stages and Rule
Systems
 Brown’s (1973) Stages Mean Length of
Utterance is a good index of child’s language
maturity. Stages indicate growth of language
complexity.
Stage 1 - 12 to 26 months of age = MLU 1.00 to 2.00
Stage 2 - 27 to 30 months of age = MLU 2.00 to 2.50
Stage 3 - 31 to 34 months of age = MLU 2.50 to 3.00
Stage 4 - 35 to 40 months of age = MLU 3.00 to 3.75
Stage 5 - 41 to 46 months of age = MLU 3.75 to 4.50
Irregular past-tense
Rules governing the use of irregulars follow
a developmental U-shaped curve
1.
2.
3.
4.
went
goed
wented
went
men
mans
mens
men
worst
baddest
worstest
worst
Irregular past-tense
100
Developmental U-shaped curve
90
Proportion correct
80Developmental U-shaped curve
70
60
50
40
30
1
2
3
4
Time
5
6
2. Wug Studies
http://www.maccs.mq.edu.au/~gtesan/AcquisitionMq/videos/Berkosexp.mov
2. Wug Studies
2. Wug Studies
2. Wug Studies
3. ERPs
Attach hat to babies head
Hat has electrodes that measure
Electrical movement across the
Skull
Electrical movement tells us
1. What part of the brain is activated
2. When the brain reacts to the stimulus and
how much it reacts
3. ERPs
Basic findings . . .
Stages of First Language
Acquisition
 Telegraphic Stage—2-5 years
 learning 20-30 words per day
 more complex syntax
 Fine-tuning--5-10 yrs.
 refining grammar
 building vocabulary
4. Experimental Trickery
 http://www.maccs.mq.edu.au/~gtesan/Ac
quisitionMq/videos/MedialWHquestions.
mov
Descargar

Slide 1