Consonant, vowel, and tone cues in early wordform recognition: Evidence from Cantonese-learning infants

Phonological biases in word recognition vary across languages, however, most work has examined consonant and vowel information, with little attention to how lexical tones contribute to word recognition in tone-language learners. The present study asked how consonants, vowels, and tones are weighted using a familiar-wordform recognition task in 11-month-old Cantonese-learning infants. In a baseline experiment (Experiment 1), infants listened longer to familiar wordforms than to pseudowords, confirming successful recognition of familiar wordforms. In Experiment 2, infants were tested on consonant and vowel mispronunciations of the same familiar words. Infants listened longer to vowel than to consonant mispronunciations, indicating that vowel changes were more disruptive to recognition than consonant changes, a V-bias. In Experiment 3, we compared vowel and tone mispronunciations and found no difference in infants' listening behaviour, suggesting comparable weighting of vowel and tone information at this age. These findings provide the earliest developmental evidence of phonological asymmetries in a tone language: vowels are more critical than consonants for recognizing familiar wordforms by 11 months, whereas vowels and tones appear to contribute similarly. Considered together with prior findings that older Cantonese-learning children rely less on tone contrasts during word learning, the present results suggest that the relative weighting of phonological cues reorganizes over development in tone-language learners.