Skip to main content

Speech-sound Processing in Infant Development and Evolution

Periodic Reporting for period 1 - SPIDE (Speech-sound Processing in Infant Development and Evolution)

Reporting period: 2016-05-01 to 2018-04-30

• What is the problem/issue being addressed?
Speech is composed of two categories of sound, i.e. consonants and vowels, which have different properties and serve different linguistic functions. This consonant/vowel asymmetry, which is established in adults, has led to the hypothesis of a division of labor between consonants and vowels present early in infancy would whereby facilitate lexical acquisition thanks to a specific targeting of consonants (i.e. the Consonant bias), thus to acquire language.
• Why is it important for society?
It is of major importance to science and society to understand how the ability to learn language arises in our children. In addition this research project comprises a significant impact for applied research on impaired language development (e.g. identifying the early signs of speech sound processing delays in babies) and language education (e.g. how exposure to more than one language might impact the use of consonants and vowels).
• What are the overall objectives?
First, we have explored the developmental validity of this hypothesis by studying its origins and emergence in Spanish infants during the course of the first year (infant studies 1 -3) ; second, we have investigated its evolutionary origins in a non-human animal capable of fine speech processing: Long Evans rats (Rattus Norvegicus; animal study 4).
• Firstly, our behavioral infant studies show that Spanish learning infants switch from an overall advantage for vowels in familiar words at 5 months, to a more adapted consonant bias at 12 months (Exp. 1 - 3). Secondly, our comparative work using the same stimuli in Long-Evans rats demonstrate significantly greater sensitivity to a vowel mispronunciation than to a consonant mispronunciation (Exp. 4). Thus, vowels are a better input for lexical recognition both for human infants at 5 months and in a non-human species, suggesting that the ability to target consonants preferentially in words may be a uniquely human ability. Our results contribute to the understanding of the developmental and evolutionary origins of consonant/vowel functional asymmetry.
This project contributes to the significant improvement of our knowledge of language acquisition and of the evolution of the human predisposition for language.
(1) On the origins of language in infants in particular, SPIDE has shown that the C/V asymmetry that the consonant bias for lexical processing is an essential and powerful strategy of word learning that infants are already using by the end of the first year.
(2) On the uniqueness of language in humans as opposed to other species in particular, SPIDE showed that the ability to assign a particular role to consonants in words is not present in non-human animals even when they are able to perceive and categorize consonant and vowel sounds and memorize word-forms, such as Long-Evans rats, suggesting that this ability is part of uniquely-human language ability.
Several educational and clinical impacts may be predicted. First, it may be crucially important to evaluate the C/V functional asymmetry in infants and children to understand the precursors of specific language impairments, dyslexia and other language development related deficits, and to develop methods of remediation for these populations. Second, our findings suggest that in the frequent case of bilingual exposure the best strategy use of consonants in words in one of the languages might be conflicting with the other language, which might be predictive of delays in development.
Objectives and results achieved by the SPIDE project