Abstract
Lisker and Abramson [Word 20, 384-422 (1964)] showed that voice onset time (VOT) for word-initial stops labeled as voiced and voiceless differ across languages, thus demonstrating that VOT is learned. Over the years, several studies have examined the developmental course of acquiring VOT for several languages, but questions remain regarding the time course of this development. Earlier studies of VOT acquisition have demonstrated that English learning infants initially produce stops with short-lag VOTs only, then go through a stage of ambiguous VOTs, finally developing two mature categories of voiced and voiceless. This study analyzed mean VOT and variability in word-initial stops produced in spontaneous samples by three infants, taped at approximately 2-month intervals between the ages of 12 months and 30 months. As predicted, these infants produced primarily voiced stops (VOT<40 ms) with moderate variability until approximately 21 months. Then longer lag stops began to appear and variability increased. Outliers (both voiced stops with extreme prevoicing and very long lag voiceless stops) were found. By the last recording session, all three infants showed mean voiced and voiceless VOTs approaching adult values, but still exhibited greater variability. [Work supported by NIDCD Grant No. DC-00633.]
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have