Language

Question 1

In what way is language hierarchical, componential and compositional?

Answer 1

Hierarchical – there are multiple levels of analysis, each of which builds on the other – e.g. phonology, morphology, syntax

Componential – each level contains a finite set of units

Compositional –these units are combined productively with a finite set of rules

Question 2

2-month-olds ______, 6-month-olds _______ and 12-month-olds pronounce their _____ ______.

Answer 2

2-month-olds coo, 6-month-olds babble and 12-month-olds pronounce their first words.

Question 3

Native-language-specific babbling begins around…

Answer 3

10 months

Question 4

When infants are born they can already recognise the _______ of their own language and prefer ______ to _________.

Answer 4

When infants are born they can already recognise the prosody of their own language and prefer speech to non-speech. Measured by rate of sucking.

Question 5

Discriminating ________ phonemic categories is easy; Discriminating ________ phonemic categories is hard.

Answer 5

Discriminating across phonemic categories is easy; Discriminating within phonemic categories is hard.

Question 6

How does infants’ phonemic perception change across the first year of life?

Answer 6

In first half of first year they show categorical perception across all world’s language sounds. By 10 months, only discriminate phonemic contrasts relevant to their language.

Question 7

From 12-18 months, kids learn about one word per ____

After 18 months this rapidly accelerates to about one word per ___.

Answer 7

From 12-18 months, kids learn about one word per day.

After 18 months this rapidly accelerates to about one word per hour.

Question 8

At age 2, kids have ____-_____ words. By age 7 they have from ______-______ words.

Answer 8

At age 2, kids have 100-2,000 words. By age 7 they have from 5,000-20,000 words.

Question 9

What words are chosen by kids at 18 months?

Answer 9

2-word combinations of most important content words. e.g. Daddy work.

Question 10

What’s the difference between inflectional and derivational morphemes?

Answer 10

Inflectional morphemes change number and tense –e.g. -s, -ed. Derivational morphemes change word category – e.g. destroy -> destruction.

Question 11

What is the pattern whereby kids learn irregular words?

Answer 11

First they say correctly via imitation – e.g. teeth. Then they overgeneralise morphemic rules –e.g. tooths.

Question 12

Kids can do complex pragmatics – e.g. metaphor, irony, sarcasm –by around age __

Answer 12

Kids can do complex pragmatics – e.g. metaphor, irony, sarcasm –by around age 8.

Question 13

How does the distinction between simultaneous and componential expression tell us whether a new sign language is bona fide?

Answer 13

Language divides meaning into components to recombine them to generate more meanings. For a language to be complete it must be componential.

The first generation of the NSL sign language used simultaneous expression (e.g. ‘the ball rolled down the hill’ is a downward rolling gesture). The second generation used separate gestures for ‘down’ and ‘rolling’. This suggests a full systematic linguistic structure.

Question 14

Why must there be a universal grammar, according to Chomsky?

Answer 14

Reinforcement learning and sensorimotor learning mechanisms cannot account for the abstract syntactical rules that generate human language, e.g. recursion

Question 15

What specifically is in universal grammar?

Answer 15

Structured representations with abstract syntactic categories. e.g. sentences are composed of NP+VP, VPs are composed of Vs and NPs.

Question 16

What is a phrase?

Answer 16

A group of words that form a functional/structural unit.

Question 17

How can you tell what’s in a phrase?

Answer 17

Pronomial reference.

Question 18

Give a piece of evidence that structure dependence is an ‘innate schematism’?

Answer 18

Children never make mistakes forming questions:
‘The boy who is smoking is crazy’ ->
‘Is the boy who smoking is crazy’?
They understand that ‘the boy who is smoking’ is a NP that cannot be split up.

Question 19

What’s an early abstraction account of how kids acquire syntax?

Answer 19

There are abstract notions of agents and patients and an innate bias to link event roles to nouns. Each semantic role needs a noun. Generally, agents come before a verb and patients after. Then these cues can be used to learn novel verbs, e.g. the frog is gorping the bear.

Question 20

In Gertner et al. (2006). ‘The frog is gorping the bear’. Children look longer to the frog as causal actor (between two pictures). To learn the new verb, the kids must demonstrate…?

Answer 20

A link between abstract syntax (word order and agency) and semantics (the meaning of ‘gorping’).

Question 21

What evidence is there for a genetically driven maturational sensitive period?

Answer 21

Deaf children not exposed to sign language before puberty never learn language.

Question 22

What is the statistical learning hypothesis of language?

Answer 22

Children track the transitional probabilities between sounds, words and phrases, and how they are distributed more globally, to learn grammatical categories and phrase structure rules.

Question 23

What is a transitional probability?

Answer 23

The probability that after X, Y will occur.

Question 24

How was it demonstrated that infants can use transitional probability to infer word boundaries?

Answer 24

Kids given string of syllables, some of which co-occurred consistently. E.g. bidaku. Then tested repetitions of bidaku against repetitions of novel sequence, e.g. kupado. Infants showed novelty preference to ‘kupado’.

Question 25

How can grammatical categories be revealed via distribution in language?

Answer 25

The __ is usually a noun.

Is ____ing usually a verb. etc.

Question 26

How did Ambridge et al. disprove Crain’s claim that structural dependence was invariantly obeyed in question form? And what does it suggest?

Answer 26

Changed statement to ‘The boy who can smoke can drive’. SD error results in possible word pair. No one ever says ‘who smoking’, but people do say ‘who smoke’.

Some kids had up to 43% errors.

Universal grammar would not predict word-specific errors – their role in the phrase structure is all that matters.

Question 27

What’s a constructivist account of how children develop abstract linguistic structure? Beginning with item-based syntax…

Answer 27

Kids don’t start using words freely, but in set phrases they have heard before. They start noticing patterns of word use and generalise individual words to novel contexts. Gradually their understanding moves from the concrete to the abstract.

Progression:

I kick it, I kick ball
I kick [OBJECT], I hit [OBJ]
I [ACTION] [OBJECT]
Mummy [ACTION] [OBJ]
[SUBJ] [VERB] [OBJ]

Question 28

What is the constructivist idea of schematization?

Answer 28

Schematization:
Forming only partially productive schema specific to each word - combining with finite number of words:
e.g. Kick it + Kick ball + Kick Mummy = Kick X

Question 29

How can the idea of a verb-schema be tested?

Answer 29

Teach young kids a novel verb (e.g. the sock is tamming) in one construction and see if they can break the schema and use it in another (e.g. he is tamming the sock). If they cannot, this is evidence of a verb schema, and support for constructivism. If they CAN, this is evidence they have an abstract verb-general rule, supporting nativism.

Question 30

How does verb schematization vary across age? And what does this suggest?

Answer 30

Only 3/16 kids at 2 could transform ‘the sock is tamming’ into ‘he’s tamming the sock’. All 4-year-olds can do it.

This suggests children start with small-scale schemas, rather than general abstract syntactical rules.

Question 31

What happens when you teach kids an invented noun, e.g. toma?

Answer 31

They can slot it into a verb-specific schema.

‘I want toma’, ‘I see toma’.

Question 32

What are the two principal claims of the constructivist approach to language?

Answer 32

Claim 1: Children can learn syntactic categories through the distributional/statistical patterns of words in their input.

Claim 2: Children’s initial syntactic knowledge is item- based (based on individual words) and abstract constructions are formed slowly by generalizing across such item-based constructions.

Question 33

How did Dittmar et al. fail to replicate the findings of Gertner (gorping)? And what does this suggests?

Answer 33

Did the same test without a warm-up in which kids where shown similar pictures with known verbs (e.g. washing). Found no discrimination between images.

Suggests the kids don’t have verb-general abstract structures. If they did, they would not require this warm-up.