Acoustic-Lexical Characteristics of Child-Directed Speech Between 7 and 24 Months and Their Impact on Toddlers' Phonological Processing

Abstract

Speech-language input from adult caregivers is a strong predictor of children's developmental outcomes. But the properties of this child-directed speech are not static over the first months or years of a child's life. This study assesses a large cohort of children and caregivers (n = 84) at 7, 10, 18, and 24 months to document (1) how a battery of phonetic, phonological, and lexical characteristics of child-directed speech changes in the first 2 years of life and (2) how input at these different stages predicts toddlers' phonological processing and vocabulary size at 2 years. Results show that most measures of child-directed speech do change as children age, and certain characteristics, like hyperarticulation, actually peak at 24 months. For language outcomes, children's phonological processing benefited from exposure to longer (in phonemes) words, more diverse word types, and enhanced coarticulation in their input. It is proposed that longer words in the input may stimulate children's phonological working memory development, while heightened coarticulation simultaneously introduces important sublexical cues and exposes them to challenging, naturalistic speech, leading to overall stronger phonological processing outcomes.

Keywords: acoustics; child-directed speech; lexicon; nonword repetition; phonological development; phonological neighborhood density; speech clarity.

Figure 1

Word type and token count in child-directed speech between 7 and 24 months. Large, gray points indicate mean; whiskers indicate 1 SD from mean. Notches indicate median. *p < 0.05; **p < 0.01; ***p < 0.001.

Figure 2

Phono-lexical characteristics of child-directed speech between 7 and 24 months. Large, gray points indicate mean; whiskers indicate 1 SD from mean. Notches indicate median. Individual datapoints indicate median word frequency and phonological neighborhood density (left and center figures), or mean word length (right figure). *p < 0.05; **p < 0.01; ***p < 0.001.

Figure 3

Acoustic characteristics of child-directed speech between 7 and 24 months. Large, gray points indicate mean; whiskers indicate 1 SD from mean. Notches indicate median. Individual datapoints indicate each caregivers' vowel space size (left figure), or median phone duration and coarticulation (center and right figures). *p < 0.05; **p < 0.01; ***p < 0.001.

Figure 4

Lexical predictors at 18 months of nonword repetition accuracy and vocabulary size at 24 months.

Figure 5

Acoustic-lexical predictors of nonword repetition at 24 months. On the left graph, increased spectral distance indicates less coarticulation. Gray regression represents model predictions. Colored points and local regressions represent original data. Ribbons represent 95% confidence intervals. *Word types containing 4+ phonemes.

Figure 6

Acoustic predictors of nonword repetition accuracy and vocabulary size at 24 months.