Infant Speech Perception Flashcards
Brith -6 months
- 25-29 weeks: fetus show consistent response to auditory stimulation
- –full term new borns have over 2 months of auditory experience
- neonates respond differnently to native and non-native variants of vowels
- —> most info is about 1K so freks up to 2600 Hz in utero
Perception of suprasegmentals: Evidence of perception in infants–> Nursery Rhyme studies
DeCasper & Spence (1986)
- moms read nursery rhyme 3X/day 6 weeks before due date
- baby showed preference of the familiar nursery rhyme
- follow up study: 4 weeks is enough for fetus o encode some properties of the rhyme.
Kruger & Garvan (20140 at 38 weeks fetuses demonstrated memory of a nursery rhyme that was regularly presented from 28 -34 weeks gestation
Perception of suprasegmentals: Evidence of perception in infants–> piano study
Granier-Deferre et al (2011)
-fetuses exposed to piano melody at 35-37 weeks showed memory of the melody 6 weeks later
IDS vs ADS
Greater attention to IDS during the 1st 6 months of life.
-mixed evidence for attention up t age 16 mos
Perception of segmental info
What is it? Acoustic properties of speech taht differentiate phonemes
Young infants are sensitive to fine-grained changes in segmental info despite lack of experience with high freqs (due to filtering in utero)
Categorical Perception
- infants may have some sensitivity to particular acoustic-phonetic cues—> can be influenced by linguistic input
- Consonant contrast are perceived categorically, but vowels are perceived more continuously.
Effects of Lang experience on speech discrimination: Consonants
Younger adults can discriminate phonemic contrast that are difficult for adults (ex. English /d/ vs Hindi /d/)
10-12 months: only discriminate contrast that were linguistically relevant
—> consonant discrimination is affected by lang input
Effects of Lang experience on speech discrimination: Vowels
Kuhl et al. (1992) different vowel perception patterns in English [i] and Swedish [y]
- -> show impact of lang background
- —> suggested a “perceptual magnet effect”
However.
Polka and Bohn (1996, 2003 and 2011) found something diff w. english vs german infants
—> vowels on the periphery of the F1/F2 acoustic space serves as universal perceptual attractors —-> “Universalist view of infant speech discrim”
Perceptual magnet effect
distribution of vowel variants in the lang shapes infants’ perceptual systems such that they perceive variants within a vowel category to be more like the prototype of that category
Universalist View
Infants are born able to discriminate any phonemic contrast that could potentially be relevant to any language
-with experience, they loose the ability to discriminate contrasts that are not relevant for their language.
Universalist View Limitations
- doesnt conside sub-phonemic info relevant for other aspects of speech perception and language
- -> allophones (context dependent variants of a phoneme/// ie. how the production of a phoneme changes given its position in the word) ex. nite rates vs nitrates
- some contrasts are not discriminable universally during early infancy, but require language experience for discrimination
Ex. Spanish [a] vs Swedish [a] in 6-12 month olds can be done, but at 4.5-6 months, cant discriminate a distinctive spanish VOT contrast
Point: VOT boundaries shfit to whats appropriate for that lang around 8-10 months of age
Others have demonstated that some contrasts improve with language experience from infancy- adulthood.
Take away re: infants and learning lang experience
The early perceptual system is able to discriminate most of the world’s languages.
Then, through experience with language input, infants become more sensitive to sounds that are relevant for their language and less sensitive to contrasts that are not linguistically relevant
Mechanisms of Learning in Infants: Recognition Memory
Basic; to recognize something, it must be encoded in memory
Ex: preference for monther’s voice and native language
-improves significantly during 1st year of life and correlates with language and cognitive outcomes
-representaions of speech sounds become more generalizable w/ experience and development (e.x. rec words across talkers at 10.5 not 7.5 months)
Mechanisms of Learning in Infants: Associative Learning
Ex. word learning.
vocal affect and facial expressions
older: complex strings of speech associated with objects, actions, attributes, expereinces, etcs
—-> important for language acquistion
Mechanisms of Learning in Infants: Statistical
- the probablity of y given x
- sensitive to speech sounds of their lang
- saffran, aslin, and newport (1996) tested 8 month old’s ability to detect syllable sequences within a continuous stream of synthetic syllables
