Human simulations of vocabulary learning

1
Human simulations of vocabulary learning
Gillette, Gleitman, Gleitman, Lederer

• Présentation Interface Syntaxe-Psycholinguistique
• Y-Lan BOUREAU

2
Outline

• Some background
• The problem to be solved
• Facts nouns acquisition precedes verbs
  acquisition
• Existing theory
• Gillette et al.s hypothesis
• Simulation experiment
• Learning from observation
• Learning from linguistic hints
• Discussion

3
The problem of language learning

• Children learn language from scratch
• Traditional hypothesis
• Children hear adults speak
• They spot that cat is uttered most frequently
  when there is a cat around
• They infer that cat means cat
• But babies vocabulary does not reflect input
  frequencies
• much more nouns than verbs in babies vocabulary

4
Nouns are learnt earlier

• A conceptual hypothesis
• Verbs are conceptually more difficult
• So cannot be learnt until babies display adequate
  conceptual knowledge
• Alternative hypothesis information requirements
• Verbs require some syntax to be already acquired
  (e.g. I know that Mommy is coming)

5
Pairing word to world

• Three sources of information
• Nonlinguistic evidence (e.g. Mommy says cat
  when the cat is there)
• Linguistic evidence
• Co-occurrence of semantically related words in
  sentences (e.g. food names usually appear with
  verbs like eat )
• Syntactic structures in which words occur (e.g. a
  verb with one subject and two complements is
  likely to be of the give kind)

6
Hypothesis

• Hypothesis the baby
• (1) acquires a small stock of nouns by
  word-to-world pairing
• (2) uses that stock of nouns as a scaffold for
  constructing representations of the linguistic
  input that will support a more efficient learning
  procedure
• Support correlation of changes in vocabulary
  size with appearance of multiword speech

7
A simulation experiment

• Principle
• Adult learners
• (no conceptual issues any more)
• Trying to guess most frequently used nouns or
  verbs
• Observational clues video clips
• Linguistic clues co-occurring words, syntactic
  frame

8
First experiment

• Only videoclips
• Adults trying to guess 24 nouns and 24 verbs

9
Results, Part I Nouns win

• Nouns are guessed with much better results than
  verbs

10
Imageability rules

• Provided clues are exclusively visual
• Nouns of the set (e.g. elephant, plane, bag) are
  a lot more imageable than verbs (e.g. think,
  know, wait)

11
Results, part II
Nouns
Nouns
Verbs
Verbs
12
Conclusion of experiment I

• The one relevant factor seems to be imageability
• Not that surprising from a video, you learn
  imageable things a thing that is not imageable
  would be hard to picture !!

13
Linguistic clues vs. Observational clues

• All nouns removed.
• 6 conditions
• 1 videoclips (but with a bip for the verb)
• 2 alphabetical lists of nouns
• 3 12 (videoclips alphabetical lists)
• 4 syntactic frames with all nonsense words
• 5 sentences with only the verb as nonsense
• 6 15 (videoclips sentences)

14
Linguistic clues vs. Observational clues
15
Linguistic clues vs. Observational clues

• 6 conditions
• 1 videoclips (but with a bip for the verb)
• 2 alphabetical lists of nouns
• 3 12 (videoclips alphabetical lists)
• 4 syntactic frames with all nonsense words
• 5 sentences with only the verb as nonsense
• 6 15 (videoclips sentences)

16
Results
Nouns reintroduced
No nouns provided !! No more visual information !!
Visuals reintroduced
17
Linguistic clues vs. Observational clues

• Remarkably leap between 3 and 4, whereas the
  reverse could have been expected !
• Interestingly, those verbs that were best learnt
  in the observational learnings show a decrease
  between 3 and 4

18
Discussion

• Verbs complementary distributions (12 never
  learnt with visual clues 12 best learnt with
  linguistic clues)
• This distribution corresponds to the
  imageability criterion
• Quite logically, you can learn visually only what
  is visually representable
• Verbs that use higher level linguistic
  representations have to wait until those can be
  constructed

19
Discussion general scheme

• First, imageable words are learnt on a
  word-to-world pairing basis
• Those imageable words are mostly nouns
• That would explain why nouns make up most of
  young infants vocabulary
• Second, this first set of words allows learning
  of new words on a sentence-to-world pairing
• Thus conceptual words can be learnt as well

20
Some reservations

• The argument structure is the same across
  languages (logical requirements), but
• Adults already know the words, so they could try
  to guess the verbs by exhaustive search with all
  the information given (e.g. the best
  performance is for look , and it is probably
  due to the use of look with at )

21
The End