Both French and English 16-month-olds have a consonant bias when learning words: An eye tracking study
Thursday, June 20th, 2013 [15:40 - 16:00]
TALKS
Both French and English 16-month-olds have a consonant bias when learning words: An eye tracking study
Poltrock, S. 1, 2 , Delle Luche, C. 3 , Floccia, C. 3 & Nazzi, T. 1, 2
1 Université Paris Descartes, Sorbonne Paris Cité, Paris, France
2 CNRS, Laboratoire Psychologie de la Perception, Paris, France
3 School of Psychology, Plymouth University, United Kingdom
The proposal by Nespor et al. (2003) that consonants are more important than vowels for lexical processes has been supported by both infant and adult studies. However, in infants, differences between the two most investigated languages, namely French and English, have been repeatedly reported: While in French a consonant advantage has been demonstrated from the age of 16 months on (Havy & Nazzi, 2010), in English-speaking children this has not been observed before 30 months of age (Nazzi et al., 2009). Here we present a cross-linguistic study using the same task at the same age (16 months) in both French- and English-acquiring infants to investigate the potential role of language input for the emergence of the consonant bias. Using eyetrackers, we recorded eye movements of 24 French- and 18 English-learning infants while they were watching animated cartoons of a computer-controlled word learning task. The experiment included eight pairs of CVC pseudo-words associated with unfamiliar objects, four pairs involving a vocalic, the other four involving a consonantal contrast. In the testing phase, infants were presented simultaneously to the two objects while one was named. For French-learning infants, analyses revealed an increase of target looking from the pre- to the post-naming phase in the consonant-contrast trials (t(23)=1.69, p= .05), but not in the vowel-contrast trials (t(23)<1). Preliminary results for the English-learning infants go in the same direction: While the proportion of target looking increases significantly for consonant-contrast trials (t(17)=2.16; p= .02), no effect is found in the vowel-contrast condition (t(17)<1). These data suggest that the lexical consonant bias is a language-general linguistic constraint rather than a language-specific one. Methodological considerations will also be discussed to account for differences between studies.