Acoustic Characteristics of Maternal Speech to Young Children with Typical Development and Young Children with Autism

Background:

Child-directed (CD) speech is often used rather than adult-directed (AD) speech in caregiver-child interactions and is distinguished by its higher pitch, slower tempo and exaggerated intonation (Fernald, 1985; Kitamura et al., 2002). Typically-developing (TD) children prefer listening to CD over AD speech or non-speech analogues (Cooper & Aslin, 1990; Pegg et al., 1992; Newman & Hussain, 2006), whereas children with autism spectrum disorder (ASD), or at risk for ASD, show a preference for AD speech (Paul et al., 2007, Nadig et al., 2007), background noise (Klin, 1991), or non-speech analogues (Kuhl et al., 2005). As CD speech can facilitate language development (Thiessen et al., 2005; Tsao et al., 2004), the focus of this study is to examine whether natural speech input to children with autism involves the acoustic properties of CD speech, and if it is related to the child’s language development.

Objectives:

We compare the acoustic properties of maternal speech to children with ASD or TD, with a focus on the extent of the production of CD modification relative to their AD speech by individual mothers. Two exploratory predictions were evaluated. One, mothers of children with ASD may increase their use of CD modification, since it is likely difficult to attract their child’s attention (Nadig et al., 2007). Alternatively, mothers of children with ASD may produce less CD modification if it is driven by child responsivity (Lam & Kitamura, 2008). CD modification by individual mothers was also expected to be positively related to her child’s vocabulary growth over a 1-year period.

Methods:  

Target enrollment for this study is 25 families for both the ASD and TD groups.  The groups of children are matched on language level, with an age range of 2-6 years for ASD and 18-41 months for TD. CD samples were collected during a 10-minute storybook session where mothers read to their child. AD samples were collected via a 5-minute interview in which the experimenter asked the mother questions about the storybooks. Individual words that were produced in both CD and AD contexts were extracted and analyzed in PRAAT for mean pitch, pitch range, and duration. The MacArthur CDI (Fenson et al., 2004) were collected at the beginning of the study and 1 year later, providing a measure of vocabulary growth.

Results:  

Preliminary analyses (ASD=4; TD=15) indicate that mean pitch and pitch range are significantly higher for CD than AD speech, but only for mothers of the TD children. Ten of 15 mothers in the TD group, and 2 of 4 in the ASD group consistently increased their pitch in speech to their child, and 9 of the mothers in the TD group and 2 of 4 in the ASD group increased their pitch range.

Conclusions:  

We observed significant individual differences with respect to CD modification among both groups of mothers. Data collection is ongoing and acoustic analysis of CD speech will be conducted, as will a comparison of CD modification by each mother with her child’s vocabulary development over a 1-year period.