Development of phonological categories in children's perception of final voicing in English

Research into speech perception over the last fifty years has increasingly tended to support the conclusion that the adults’ perception of phonological categories is characterised by the use of multiple acoustic properties. There is typically no one acoustic property which is necessary and sufficient to signal the value of a phonological category, and there is a high degree of redundancy among these acoustic properties as cues. For example, the number of acoustic properties which have been shown to affect the perception of intervocalic voicing is at least sixteen (Lisker 1986). The use of these properties is complicated by the lack of acoustic invariance in speech: the extent of evidence for a category available from any one property in a given dialect or language varies as a function of phonological context, and differs subtly from one utterance to the next, and from speaker to speaker. Adults adjust for all this variability in using multiple acoustic properties to perceive phonological categories. What is the course of development of this ability, in childhood