Using TRACE to Model Infant Sensitivity to Vowel and Consonant Mispronunciations

Very young infants possess a capacity to discriminate contrasts that are not present in their native language. Later in development, they lose this capacity while improving the discrimination of sounds in their native language and progressively tuning their speech sensitivity to increase the phonological specificity of their lexical represenations. Recent evidence suggests a symmetry in infant sensitivity to vowel and consonant mispronunciations of familiar words from early in the second year of life. We investigate this question from a modelling perspective, using a continuous mapping model; TRACE. Our results support Mani and Plunkett’s (2007) claim that both vowels and consonants constrain lexical access to familiar words in the infant lexicon. However, TRACE predicts that infants should become increasingly sensitive to onset mispronunciations (usually consonants in English) of familiar words as vocabulary develops, whereas their sensitivity to non-onset (often vowels) mispronunciations should remain relatively stable during the second year of life. Interestingly, this effect is purely driven by the structure and size of the lexicon, as TRACE is not a developmental model.