Topic 10: Early Speech Perception Flashcards

1
Q

What is developmental psycholinguistics?

A

The study of the underlying cognitive processes involved in word learning and language acquisition.

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

What are phonemes and prosody?

A

Phonemes – smallest unit of sound that is meaningful in native language
Prosody - speech qualities (irony, mood), rhythm and rate of syllables

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

At what age is the auditory system functional? and how is this measured

A

Around 24 weeks foetus will respond to auditory stimulation measured by increase in heartrate or shift of position.

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

What is the auditory environment like inside the womb?

A

▪ noisy environment
▪ external sounds are ‘muffled’
▪ lower frequency of speech
▪ maternal voice is loud, salient – but unintelligible & dissimilar to ‘outside’
* cannot detect phonemes but rather the prosody of language

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

What tools/ methods are used in fetal research?

A

fetal ultrasound or electrocardiogram (prone to distortion) Or, fetal magnetocardiography – relies on magnetic field and is more precise

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

Study on fetal familiarity of mother’s use of native language (Minai et al, 2017)

A

Using magnetocardiography
they measured the consistency of fetal heartrate when reading a passage in english-english or english- japanese. Results show less consistency when novel language is spoken, suggesting familiarity to native language.

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

What are neonates?

A

A newborn child (less than 4 weeks old)

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

How can we measure preferences in neonates?

A

By monitoring rates of non-nutritive sucking or by assessing them using the headturn preference procedure.

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

What auditory preferences have neonates shown to have through non nutritive sucking research?

A

Human voice > synthetic sounds
maternal voice > male voice
maternal language > non native

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

De Caspar & Spence (1986) Dr. Suess trans natal learning.

A

In later stages of pregnancy mothers read a Dr.Suess book consistenly aloud. After birth neonates showed preference for familiar Dr, Suess passage suggesting they were learning the rhythm of language in utero.

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

What is perceptual learning?

A

long-lasting changes in (auditory) perception that arise from experience or practice. e.g. discrimination between speech sounds of different languages.

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

Study on neonatal discimination between different languages.(Nazzi et al., 1998)

A

French neonates non-nutritive sucking rates were measured in response to non native langauge being spoken. Initially higher sucking rates for english, then initially higher rates for japanese but not dutch. Dutch and english have similar timing properties whereas japanese has a different rhythmic structure. So by 2 mths infants are able to discriminate between langauges.

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

How do infants segment speech and find the
fundamental units in speech streams?

A

Categorical Perception: a perceptual phenomenon whereby events (e.g., sounds, colours) that lie along a continuum are perceived as belonging to one category or another.
So infants categorise phonemes into groups that help them to understand words and sentences.

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

How do we have stable categories of speech sounds? e.g. pa-ba

A

speech sounds are on a physical continuum, most of us place the phoneme ‘boundary’ around the same place on this continuum

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

Study of infants ability to extract perceptual categories from speech sounds (Eimas et al. 1972)

A

1-4mths old infants given continuous speech stream (ba-ba-ba) that merges into another syllable (da). Non-nutritive sucking rate increased much more when new syllable introduced compared to control group. Suggests infants can discriminate across phoneme boundaries.

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

How do we know that categorisation is a general property of audition?

A

Chinchillas are very good at categorical perception of speech so it is not unique to humans. Categorisation is a general property of audition that captures the discontinuity in sounds.

17
Q

How do we know that categorisation is a general property of audition?

A

Categorisation captures the discontinuity in sounds. Chinchillas are very good at categorical perception of speech so it is not unique to humans. Therefore is a general property of audition.

18
Q

What is the Headturn Preference procedure and what can they show?

A

A method for detecting preferences (sounds, words, or sentences) in infants.
They can show discrimination of a change within a continuous sound stream e.g. /da-da-da-ba-ba-ba/.
And after training (habituation) to one sound (word or sentence) they can test dishabituation with a new sound

19
Q

Study: Infants ability to discriminate speech sounds of their native language and non-native languages (Werker & Tees, 1984)

A

Used a modified headturn procedure to measure American infants ability to percieve syllabic changes in native language and Hindi or Sallish.
Younger infants at 6-8 mths (80-90%) perceived both contrasts whereas fewer older infants at 10-12 mths (10-20%) perceived non-native contrasts.
Suggesting a big decline in distinguishing sounds from non-native languages through development.

20
Q

when do infants become more sensitive to the sounds of their native language? study (kuhl et al., 2006)

A

(kuhl et al., 2006) conducted an experiment with american and japanese infants on their ability to distinguish between r and l sounds (sounds that are ambigious in japanese). They found found that at 6-8 months infants from both countries have a similar level of ability but by 10-12 months american infants were far better at recognising the distinction and the japanese infants much worse. Suggesting USA infants are discriminating between the two sounds because they hear the words more freequently.

21
Q

Two types of statistics available from speech:

A
  1. Distributional probability. Infants can capture how often some phonemes and syllables ‘show up’ in their native language.
  2. Transitional probability of syllables that co-occur often together. Infants can capture what speech sounds/syllables are likely to be heard together in combination.
22
Q

What social factors help an infant learn langauge?

A

Child-directed speech (motherese) adults/carers use this when talking to children e.g. fewer words,
higher pitch (speech register), slower/ more pauses, more
repetition, exaggerated intonation patterns
Talking with others - helps orient the attention of infants e.g eye gaze, joint attention.

23
Q

Do infants actually need people to learn speech contrasts? study (kuhl, Tsao & Liu, 2003)

A

Attemped to teach mandarin to USA infants (10-12mths) either by live exposure to mandarin speaker or by TV exposure to mandarin speaking.
Only ‘live’ exposure to Mandarin (from a real person) enabled USA infants to discriminate the Mandarin contrasts when later tested.
Suggesting that language development is a combination of picking up the statistics of speech as well as the social environment.