Lecture 5 - Early Language Development Flashcards

(52 cards)

1
Q

Who investigated speech processing before birth

A

De Casper and Spence 1986

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
2
Q

Outline the method of De Casper and Spencer 1986

A

12 pregnant women read passage from the Cat in the Hat. Chosen because regular rhyme
2/3 days after birth babies tested recognition sucking response
Played either recording cat in the hat or another unfamiliar passage

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
3
Q

Outline the results of De Casper and Spencer 1986

A

Babies altered sucking pattern to hear familiar passage but not unfamiliar
Occurred irrespective whether mother or unfamiliar person reading
Recognising prosody - intonation, rhythm, stress
Babies are actively processing speech before birth

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
4
Q

Who investigated speech recognition in Utero

A

De Casper, Lecanuet and Busnel 1994

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
5
Q

Outline De Casper, Lecanuet and Busnel 1994 study on speech recognition in Utero

A

Changes in foetal heart rate.
35th week pregnancy recited one of two nursery rhymes.
Tested end 4 weeks by playing recording both rhythms using speakers close mothers abdomen.
Foetal HR decreased significantly to familiar rhyme only = due rhythm

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
6
Q

Who investigated preference for familiar language

A

Mehler, Jusczyk, Dehaene-Lambertz et al 1988

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
7
Q

Outline Mehler, Jusczyk, Dehaene-Lambertz et al 1988 study on preference for familiar language

A

4-day-old French babies presented samples French and Russian
Preferred French.
Sensitive to prosodic info - rhythm and intonation

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
8
Q

Who investigated telling languages apart

A

Christophe and Morton 1998

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
9
Q

Outline Christophe and Morton 1998 study on Telling Languages apart

A

Presented 2-month-old babies with 2 different language comparisons
English vs Japenese - different prosody
English vs Dutch
Sucking decreased = habituation. Only change sucking went from English to Japenese.
Tell difference between English and Japenese.
Prosody to distinguish

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
10
Q

Why is it useful for infants in bilingual environments to distinguish languages

A

Ability understand different rules, stress patterns for different languages, work out patterns in language

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
11
Q

What are Phonemes

A

Smallest sound unit distinguish between 1 meaning and another
e.g. b and p

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
12
Q

Why is it crucial to tell apart different phonemes

A

Understand different meanings and differentiate different words and be able to relate them to things in world

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
13
Q

Why is it crucial to perceive different variations of same phoneme

A

Allows flexibility and not be confused between irrelevant differences

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
14
Q

Outline categorical perceptions of phonemes

A

Phoneme boundaries - where physical parameter changes perception one phoneme to another
20 ms difference across phoneme boundary easy distinguish.
20 ms difference within phoneme category hard distinguish

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
15
Q

Outline High Amplitude Sucking (HAS) Paradigm

A

Test infants discrimination speech sounds.
Habituation/bored decline in sucking rate.
Introduce novel sound and notice difference sucking increase.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
16
Q

Outline Eimas et al 1971 study using High Amplitude Sucking Paradigm to look at infants categorical perception of speech

A

Phase 1: 1-4 months presented single sound /pa/. Increased sucking, then reduces to baseline
Phase 2: habituated. New similar sound played. Half heard different phoneme. Half heard variant same phoneme

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
17
Q

Outline results of Eimas et al 1971 results

A

Heard different phoneme e.g. ba increased sucking rate
Like adults perceiving different categories of sound.
Cannot distinguish within categories

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
18
Q

Is early discrimination speech sounds unique to humans?

A

NO!
Kuhl and Miller 1975 -
Macaques, Chinchillas and Quail do it

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
19
Q

Phonetic Discrimination in Native and Foreign Languages

A

Newborns potential make phonetic discrimination
Adults do not.
Often unable hear phonetic distinctions occur other languages but not their own
e.g. Japenese do not use ‘L’ or ‘R’ = difficulty name Lara

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
20
Q

Outline Conditioned Headturn Paradigm

A

Trained whenever change stimulus electric toy lit up and activated
Infants trained look toy when hear change
Target items played, observer (cannot hear) judge whether infant heard stimulus change by infants actions

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
21
Q

Outline developmental change study - Werker and Tees 1984

A

Compared babies from language communities where phonemes differed. English vs Hindi. Da vs Da

22
Q

Outline developmental change study - Werker and Tees 1984 Results

A

6-8 months tell difference
10-12 months could not tell difference
6 months universal listener
10 months stop being universal listeners

23
Q

Why does developmental loss of knowledge occur? explain findings of Werker and Tees

A

Experience dependent
Good thing - more efficient tune into language you actually acquire. Do not need to be sensitive to all contrasts = not all relevant

24
Q

Outline Kuhl et al 2006 study on effects of age and language experience: facilitation for native and decline for non-native discrimination

A

Japanese vs English. Discriminating r and l

10-12 months Japanese infants decline, American infants improve. Get better own language.

25
Outline Kuhl 2004 Neural Commitment Model
Neural tissue response becomes committed to particular contrast in own language. Any contrasts don't hear are not strengthened and we lose them
26
Outline Kuhl et al 2003 experience foreign language can reverse decline in non-native speech production
10-12 months Mandarin an English infants. 1/2 exposed English. 1/2 exposed Mandarin American perform just as well as Chinese when exposed Mandarin Experience dependent Importance live interaction.
27
Outline Perceptual Narrowing in Infancy
Developmental loss of knowledge Face processing - 3 month olds discriminate faces no matter ethnic group. 9 months only discriminate own ethnic groups. Own-race effect Adults and 9 month olds discriminate between human faces. 6 month old good discriminating between human and monkeys faces
28
Outline Speech Segmentation
How babies tell where 1 word ends and next begins
29
Outline Jusczyk and Aslin 1995 study on infants ability extract words from fluent speech
Familiarised 7.5month infants repetitions of sentences. Contain 2 target words. Tested on target and novel individual words. Shown words isolation later and asked if recognise them?
30
Outline Jusczyk and Aslin 1995 study on infants ability extract words from fluent speech Results?
Show difference in response to words that had been embedded in familiarisation sentences 7.5 months some ability detect word when they occur in fluent speech contexts
31
Outline Preferential Listening Paradigm
Infants sit on caregivers lap in test booth Each trial, one side lights flash, when infant orients to light, sounds come from that speaker. Experimenter records how long infant looks at source of sounds as measure infants preference AKA Headturn Preference Paradigm
32
How can babies tell where one word ends and next begins
Helped by infant directed speech 1. higher pitch, more exaggerated intonation contours 2. shorter utterances 3. longer pauses 4. simplified sentence structure
33
Outline Thiessen, Hill and Saffran 2005 study on the effect of infant-directed speech
Preference speaker using infant-directed speech. | 8month olds listening to words and part words
34
Outline Thiessen, Hill and Saffran 2005 study on the effect of infant-directed speech Results
Easier tell apart words from part-words when listened infant-directed speech. Showed surprise when listening part-words later on. No difference in adult directed speech - did not cement understanding Infant directed speech a facilitator
35
Outline issues of effect of infant-directed speech
Familiarity preference. Infants prefer familiarity. Result complexity prefer familiarity
36
How can babies tell where one word ends and next begins?
Implicit discovery of cues in language input e.g. use prosodic cues e.g. syllable stress attend to transitional probabilities
37
Outline Syllable Stress
Prosodic cue to word boundary | English language puts stress in first half of words. Cue for new words
38
Jusczyk, Cutler and Redanz 1993 study on Infants sensitivity to syllable stress
6 months no preference. 9 months spent longer listening to strong-weak lists. Prefer words conformed to their language. Does not confirm learning cemented
39
Outline statistical learning of transitional probabilities
Probability 1 syllable following another Certain consequences syllables occur more often than others More commonly occurring sequences likely to be words
40
Outline Saffran et al 1996 and Johnson and Jusczyk 2001 study number 1 into Transitional Probabilities
'Pretty Baby' Probability 'ty' following 'pre' high as pretty is a word. Probability 'ty' and 'ba' following unlikely as not word. Transitional probability lower.
41
Outline Saffran et al 1996 and Johnson and Jusczyk 2001 study number 2 into Transitional Probabilities
Transitional probabilities in detection of word boundaries by 8 months Invented words taking 12 syllables and combining them into 4 sequences Presented different orders = affect probability
42
Outline Saffran et al 1996 and Johnson and Jusczyk 2001 study number 2 into Transitional Probabilities RESULTS
Transitional probabilities be used to distinguish 'words' from other syllable sequences Probabilities higher between syllables and within words. Lower across word boundaries
43
Outline Saffran et al 1996 and Johnson and Jusczyk 2001 TEST STUDY into Transitional Probabilities
Infants presented 4 words. Then heard part words where syllables from 2 of words were recombined form new word Showed part-words surprising. Novelty preference. Discriminate between full and part words. Process and understand difference
44
Implications of Saffran et al 1996 and Johnson and Jusczyk 2001 study
8months able segment continuous stream speech based on stat cues alone Bias selectively attend to certain properties in acoustic signal
45
What does Saffran, Werker and Werker 2006 state
Absence of external guidelines, our perceptual systems use auditory environment for our development of communication. All occurs during 1st postnatal year
46
What is a word
Comprehension | Production .
47
Outline Comprehension
Consistent, specific response. Must be word rather nonverbal. Identify word from speech stream. Remember word to recognise. Link word with consistent event
48
Outline Production
Consistent and specific context. Identify word from speech stream. Remember word so recognise. Link word with consistent event. Repeat sound of word. Say word in appropriate context.
49
How to measure Comprehension
1. Lab - ask infants choose named object from array. Preferential looking paradigm 2. Home observations/Video recordings 3. Parental Reports - Communicative Development Inventory - CDI. What words child uses and understands
50
Outline Tincoff and Jusczyk 1999 study on early word comprehension for socially salient words
6 months. Hear recordings voice saying mummy or daddy with 2 monitors. Infants looked more video matching word heard. Intermodal matching. 6-9months also comprehend words associated with common foods and body parts.
51
Outline Fernald et al 1998 study on Time course of infant word recognition
Hear target word and eye-tracked. Look efficiently towards target. Infants rapidly look named target. Dramatic gains 2nd year. 24 months gaze shifts before offset of word
52
Outline development of comprehension
Understand 1st words 6-8months Still remember after 1 instance novel object. Through elimination understand context and select novel word. Understanding increases slowly then spurts 16months - 70-270 words Fast mapping - ability form quick and rough meanings