Longitudinal data on the production of content versus function words in children's spontaneous speech

Abstract

Allen and Hawkins (1978; 1980) were among the first to note rhythmic differences in the speech of children and adults. Sirsa and Redford (2011) found that rhythmic differences between younger and older children’s speech was best accounted for by age-related differences in function word production. In other on-going work (Redford, Kallay & Dilley) we found an effect of age on the perceived prominence of function words in children’s speech, but no effect on content words. The current longitudinal study investigated the effect of word class (content versus function words) on the development of reduction in terms of syllable duration and pitch range (a correlate of accenting). Spontaneous speech was elicited for 3 years from 36 children aged 5; 2–6; 11 at time of first recording. There were effects of word class (content > function) and of time on median duration, but no interaction between these factors. The median duration decreased 13% in function words from the 1st to 3rd year; a similar decrease (15%) was found for content words. Pitch range only varied systematically with word class. Other spectral measures are being collected to further investigate the development of reduction in children's speech. [Work supported by NICHD.]