Consonant clusters in children's spontaneous and imitative speech

Abstract

Although differences between a child's spontaneous speech and his imitative speech have been debated at least since the beginning of articulation testing, there is still no consensus on the issue. Some researchers suggest that the two modes of speech do not differ, since both are produced in terms of a child's own linguistic system; others argue that imitating a model calls on skills different from those employed in spontaneous speech. Furthermore, the effects of imitation may show themselves on the phonological or only on the acoustic‐phonetic levels. In order to obtain further insight into the issue, consonant clusters, thought to be particularly susceptible to differences between the two speech modes [J. D. Kresheck and G. Socolofsky, “Imitative and Spontaneous Articulatory Assessment of Four‐Year‐Old Children,” J. Speech Hear. Res. 15, 729–733 (1972)], in young children's spontaneous and imitative speech were compared. Ten children, five two‐year‐olds and five three‐year‐olds, were recorded individually in a play situation; any spontaneously produced words beginning with an initial consonant cluster were also recorded imitatively at the end of the session. Transcription of the spontaneously and imitatively produced words showed no differences in the proportion of correct productions of clusters for either the younger or the older children; measurement of voice onset time from wideband spectrograms also gave no indication of a systematic acoustic‐phonetic difference between the two modes of speech.