The consonant bias in word learning is not determined by position within the word: Evidence from vowel-initial words

Abstract

The current study used an object manipulation task to explore whether infants rely more on consonant information than on vowel information when learning new words even when the words start with a vowel. Canadian French-learning 20-month-olds, who were taught pairs of new vowel-initial words contrasted either on their initial vowel (opsi/eupsi) or following consonant (oupsa/outsa), were found to have learned the words only in the consonant condition and performed significantly better in the consonant condition than in the vowel condition. These results extend to Canadian French-learning infants the consonant bias in word learning previously found in French-learning infants from France and, crucially, shows that vocalic information has less weight than consonant information in new word learning even when it is the initial sound of the target words, confirming the consonant bias at the lexical level postulated by Nespor et al. (2003). The current findings also suggest that French-learning infants are able to segment vowel-initial words as early as 20 months of age.

The development of early phonological networks: An analysis of individual longitudinal vocabulary growth

“…Thus, for instance, the difference between FDL and FDMS centered around the inclusion of vowel similarity scores as well as the comparison of the phonological similarity of syllables in words regardless of the position of the syllable within a word in FDMS. To the extent that these factors then do not contribute to the perceived similarity of words, this would suggest that consonants and syllable position are key determinats of lexical processing Indeed, this is in line with other studies that report a consonant bias in lexical processing, i.e., that consonants shape the acquisition of the words a child will learn and are privileged for lexical processes (Hochmann et al, 2011;Nazzi & Polka, 2018;Nazzi, 2005). Furthermore, we note that better insights into phonological similarity of words could also consider more detailed information about words.…”

Cross‐Situational Learning of Phonologically Overlapping Words Across Degrees of Ambiguity

“…Notably, the first research to address this issue suggests that position effects do not account for consonant‐vowel asymmetries. Twenty‐month‐old Canadian English–French bilinguals failed to learn VCVC words that differed in their initial vowel, which also occurred as the initial segment, but learned words that differed in their initial consonant, which occurred as the second segment (Nazzi & Polka, ). Together with previous work showing that French‐learning 20‐month‐olds show a consonant bias when learning new words—differentiating between CVCV words that differ in either consonant—but struggle to differentiate between words that differ in their vowel (Nazzi, ), this supports poorer encoding of vowel information regardless of segment position.…”

Interacting processes and developmental biases allow learners to crack the “what” code and the “who” code in spoken language

“…In consonant–vowel–consonant (CVC) words, French-learning infants give more weight to both the onset and the coda consonants compared to the vowel (Nazzi & Bertoncini, 2009). Moreover, in VC.CV words, (Canadian) French-learning infants give more weight to the coda consonant of the initial syllable than to its onset vowel, even though it has the advantage of being in word-initial position (Nazzi & Polka, 2018).…”