Lingüística Interna y Externa, Apuntes de Psicolingüística

¡Descarga Lingüística Interna y Externa y más Apuntes en PDF de Psicolingüística solo en Docsity!

L1 Acquisition II: Error Correction, Word

Learning and Grammatical Conservatism

ERROR CORRECTION

Lots of positive evidence: types of syllables, things you move to make a

question. Negative evidence is no good:

1. We are bad at telling kids what goes wrong: we agree on successful

communication without direct correction of anything the kid is saying ( Me

want more ) because correction would be so clumsy and is contrary to the

nature of the interaction. We prime communication over accuracy. But

parents do provide right models indirectly by repeating it correctly. Maybe

recasting the utterance is the way toward grammaticality.

2. We accept grammatical errors if points are true. In studies it has

been shown that parents correct meaning and pronunciation not

syntactic correctness.

3. 73% of utterances by kids are not corrected and correction does

not work either: nobody don’t like me …… nobody don’t likes me (8 times).

Children don’t see them as mistakes. For them it is a grammar internally

consistent at the time and cannot be overridden by adult corrections.

Give good positive data to compare with to refine your hypothesis.

WORD LEARNING

Arbitrariness of word signs except for onamatopeia like plop or achoo.

Sounds unpredictable with respect to info about the world. The info about

words must be a storage in our brains in the lexicon made up of lexical

entries: pron, meaning, det, pl by default rule (wugs) or irregular.

How is it learnt?: split it into its phonemes to be mapped into their

unpredictable meaning. Constraints here (look there, rabbit, run?):

1. Whole object assumption, not to one part or attribute.

2. Type assumption: a type, not a particular thing. Children offered

dax (for cow) to signal either a pig or a bottle (pig 65%), using label to

group things together.

3. Basic level assumption: new words apply to basic level categories

(not mammal, fex).

4. Mutual exclusivity assumption: objects have just one label.

Children can also use the morphosyntax already known (pl, determiner):

1&1/2 year olds knew that a zav refers to a type while Zav to a name.

GRAMMATICAL CONSERVATISM

• UG: collection of constraints on the form of possible grammars. Babies can distinguish

between different sounds but not pronounce all or utter their thoughts. As they grow older

they pick up language very fast but it takes time. Their vocal tract has to get ready for the very

precise task of performing speech. They realize that they’re not able to perform some sounds

correctly.

• They have entrenched beliefs about how language works and that’s why correction doesn’t

work for they don’t have that flexibility. For the next few years, they are in linguistic flux,

exposed to a lot of data and we notice some mistakes in a sea of correct utterances. Those few

mistakes are not only predictable but follow the same pattern: they will leave things out rather

than put things in wrong because kids are grammatically conservative. We only may notice

mistakes in experimental data when forcing kids to do something.

• They have subconscious ideas about how the data they receive are put together to form the

L1. They may have a different hypothesis and make mistakes in experimental data. But kids are

much more sophisticated when it comes to communication. Like us in L2, they only use

constructions they know belong in the TL. There must be some structural ppl underneath for

them to feel comfortable about what they’re doing.

GR. CONSERVATISM

This goes very deep: noun compounding comes for

free once they’ve learnt verb particle constructions:

children have to notice what kind of language they’re

learning. English kids start using both constructions

simultaneously (2007) but going one way: you can

have the big compound construction without the

verb particle but not vice-versa. So they can start

using compounds early, at 2 ½ years of age before

Japanese kids who also have them but not verb pcl

constructions and they nail it right from the start.