week 11- NATURE VS. NURTURE CHILDREN AND MULTILINGUALISM

Question 1

evidence for nature argument

Answer 1

Babble as a universal phenomenon: Babble starts out as a motoric behaviour
(Jakobson, 1968)

Early consonant production not language specific – based on sounds infants produce most easily
(Davis & MacNeilage, 199

Poverty of the stimulus: Children know more about language than what is heard in the input
(Berwick et al., 2011)

Question 2

evidence for nurture argument

Answer 2

Babble becomes a communicative phenomenon and determines later word acquisition
(McGillion et al., 2017)

Early vowel production matches the vowels produced in the ambient language
(Boysson-Bardies et al., 1989)

Children’s words, structures and phrases match what is most frequent or salient in the input
(Ambridge et al., 2015; Cameron-Faulkner et al., 2003)

Question 3

what did Chomsky say about an updated theoretical model?

Answer 3

“The question is not whether learning presupposes innate structure – of course it does, that has never been in doubt – but rather what those innate structures are in particular domains” Chomsky, 1975

Question 4

what does it mean if something is domain general?

Answer 4

Mechanisms that function with different inputs, across modalities

things that you do to create associations. eg. see a cat, hear the word, know what is means and attach meaning

Question 5

which learning mechanisms are domain general?

Answer 5

◦ Associative learning: making inferences from repeated combinations in the input (e.g. ‘cat’ + )
◦ Statistical learning: sensitivity to regularities in the input

◦ Rapid processing: sensory perception is rapid and automatic

◦ Abstraction: generalising from the input to create novel structures

These general mechanisms are thought to be innate

Question 6

what does it mean if something is domain specific?

Answer 6

Mechanisms that function independently, such that one domain does not affect another domain

Domain specificity means that adaptations evolve to solve problems in particular domains, and therefore are less well suited to solve problems in other domains.

Question 7

which learning mechanisms are domain specific?

Answer 7

◦ Physiological factors necessary for reproduction: respiration, immunity, oxygenation, etc.

◦ Sensory capacity: sight vs. smell vs. hearing vs. taste – all function differently

◦ Some developmental capacities: if something can be impaired independently it must be domain specific

◦ Some linguistic capacities: if something can be impaired independently it must be domain specific

Question 8

how can we tell the difference between what is domain general and domain specific?

Answer 8

The difference between what is domain general and what is domain specific is unclear and difficult to determine from scientific evidence.
Example: dyslexia as a hereditary processing problem

Learning in siblings (twins and adoptees): exemplifies the role of nature
Learning when key components are disrupted: exemplifies the mechanisms required for learning to occur

Question 9

what are the three types of siblings?

Answer 9

Monozygotic twins (identical): 100% gene match
Dizygotic twins (non-identical): 50% gene match
Environment is also very similar, but not identical
Adoptees: 0% gene match, but environment is similar

Question 10

what is the nature and the nurture hypothesis for siblings?

Answer 10

Nature hypothesis Twins will behave more similarly than non-twin infants in all respects

Nurture hypothesis
Twins will behave more similarly than non-twin infants in behaviours for which they share the same input, but will be no different from non-twin infants in others

Question 11

discuss a study around Specific Language Impairment in twins

Answer 11

(Newbury et al., 2005)
SLI in twins: A methodological approach
1. Select one infant with SLI who happens to be a twin
2. Calculate likelihood that the other twin also has SLI (‘concordance rate’)
3. If Monozygotic concordance rate > Dizygotic concordance rate, genes likely to play a role

Question 12

Do MZ twins have similar language achievement in relation to DZ and NT children?

Answer 12

MZ = Monozygotic 
DZ = Dizygotic 
NT = non-twin

(Tomblin & Buckwalter, 1998)
- Participants: 65 sets of MZ and DZ twins/triplets, 25 NT pairs, all < 13;0
- One member of each pair/triplet showed poor language skills
- Concordance rate for language impairment calculated for each child
- Hypotheses: MZ twins will show more similar language skills than DZ twins
DZ twins will show more similar language skills than NT children

Results
MZ x MZ = 96% concordance rate
DZ x DZ = 69% concordance rate

Question 13

How do adopted children (different genes, same learning environment) compare?

Answer 13

Thompson et al., 1985
- Participants: 182 adopted children 164 non-adopted (control) children Parents of all children included in study (adoptive and biological)

Method: Language measures at age 1;0 and 2;0 – lexical knowledge, verbal skills Parents also tested on verbal skills, memory, cognitive capacity…

Hypothesis: Correlations between adopted children and their parents would show environmental influences in language learning; correlations between biological parents and their adopted children would show genetic influences in language learning

Results

• strong correlation between learning and cognitive ability
• Results suggest a strong role for nature by 2;0, and insignificant influence of nurture

Question 14

what happens to Infants who can vocalise but cannot hear?

Answer 14

Oller & Eilers, 1988
Participants: 21 normal-hearing infants, 9 hearing-impaired infants

Method: All infants were recorded regularly to trace the onset and development of babbling

Results: All infants babbled
Hearing infants began babbling between 0;6 and 0;10
Deaf infants began babbling between 0;11 and 2;2, but this was not canonical
Only 3 deaf infants continued to babble 5-6 months after babble onset; these infants all developed some speech capacity

Question 15

what happens to Infants who can hear but cannot vocalise?

Answer 15

Jiang & Morrison, 2003
Participants: 39 children who had been tracheotomised (tube down throat) during infancy (mean age = 3;0) All aged under 5y at time of procedure
16/39 neuroatypical, 23/39 neurotypical

Method: Tested on spoken language and language comprehension in later development (age not specified) in relation to age of decannulation (tube removed)

results
Children with typical language-use had earlier and shorter tracheostomies
Mean age at decannulation of children with normal speech: 1;2
Mean age at decannulation of children with delayed speech: 1;11

These results suggest that language acquisition can take place so long as conditions are sufficient in the environment.

Question 16

what is bilingualism?

Answer 16

“Ability to speak two languages; the habitual use of two languages colloquially” (OED.com)

‘A scientific definition is hard to pin down’ Bassetti & Cook, 2011

Question 17

how many children are being raised bilingually worldwide?

Answer 17

It is thought that around 66% of children across the world are being raised bilingually (Dana Foundation, 2012)

Question 18

How does the bilingual child sort out the two languages he or she is hearing?

Answer 18

• Long-standing view: bilingual children differentiate their languages only gradually, by age two.
• Alternative view: language systems are separated from the start; exposure to each language sets the appropriate triggers.

Question 19

Can bilingual babies discriminate their two languages

Answer 19

Bosch & Sebastián-Gallés, 1997

Experiment: Can bilingual infants discriminate between languages at 0;4?

Hypothesis: bilingual babies will be able to discriminate their two languages
Spanish/Catalan bilingual infants tested on English (Unfamiliar condition) vs. Spanish/Catalan stimuli (familiar condition) AND on Spanish vs. Catalan stimuli (familiar v familiar)

Unexpected results: bilinguals were slower to respond overall. BUT they were faster to respond to non-native (English) stimuli than to Spanish/Catalan stimuli They showed no discrimination of Catalan and Spanish stimuli

Question 20

Do bilingual babies differ in word learning abilities?

Answer 20

Fennell et al., 2007

• Tested Canadian bilingual infants age 14-20m
• Compared results with monolinguals from previous experiment (Werker et al., 2002)
• Switch procedure
• infants taught ‘bih’ and ‘dih’
• will be exposed to opposite image and meaning- do they notice the mistake?

Results: Bilinguals were slower to detect differences between novel words than monolinguals
This suggests a bilingual disadvantage to word learning (or at least, fast-mapping)

Question 21

Heritage language

Answer 21

when a child is raised in a home where the non-majority language is spoken, they are a heritage speaker of that language (if they possess proficiency)

Heritage language speakers might speak the language fluently, but be unable to read or write it

They may have [had] tuition in the language

Question 22

statistical learning

Answer 22

sensitivity to regularities in our input, not specific to language, visual or acoustic patterns. pick up patterns easily.