Children learn separate aspects of speech production at different rates: Evidence from spectral moments

One theoretical perspective on speech development holds that segmental structure gradually emerges in the utterances of children as their first words, realized as roughly specified patterns of gesture, are differentiated into patterns of precisely specified and carefully coordinated gestures. The purpose of the present study was to test one aspect of this theoretical position, namely that children’s articulatory gestures are not as precisely specified as are those of adults. In addition, a related hypothesis was tested, namely that some patterns of gesture achieve adultlike precision sooner than others. To test these hypotheses, the first, third, and fourth spectral moments were derived for fricative (/s/ and /∫) and for stop‐burst (/t/ and /k/) noises in the speech samples of children (3 to 7 years of age) and of adults. First spectral moments for fricatives showed stronger consonant effects (i.e., /s/ vs /∫/) for adults’ than for children’s samples. This acoustic finding replicated a previous result [Nittrouer et al., J. Speech Hear. Res. 32, 120–132 (1989)], and provided support for the hypothesis that some articulatory gestures are not as precisely specified in children’s speech as in that of adults. First spectral moments for /t/ and /k/ revealed no age‐related differences in the magnitude of the consonant effect, providing support for the hypothesis that some gestures achieve mature status sooner than others. Although not a focus of the present study, it was also found that children’s /k/ first moments varied more than adults’ as a function of vowel environment (/i/, /ɑ/, or /u/). Thus further support was obtained for a finding described elsewhere: For at least some articulatory gestures, there is greater influence of one on another for children’s than for adults’ speech [e.g., Nittrouer et al., J. Speech Hear. Res. 32, 120–132 (1989); S. Nittrouer, J. Speech Hear. Res. 36, 959–971 (1993)].