Poster #12 - Speech characteristics predict social competence in non-autistic, but not autistic children

• In Event: Poster Session 16

Abstract

When we communicate with those around us, changes in our speech prosody (e.g., vocal pitch) can provide valuable information to the listener about our emotional state and our intentions (Nadig & Shaw, 2012). Early descriptions of autistic individuals’ social communication style mention impairments in prosody, and atypical speech prosody can influence others’ judgements of one’s social competence, which may contribute to the social difficulties experienced by autistic individuals (Asperger & Frith, 1991). Poor prosodic control may also have downstream effects on social development, as it may interfere with the establishment and maintenance of social relationships (Hubbard et al., 2017).

The objective of the current study was to: (1) examine whether social competence in children both with and without autism may be predicted from speech prosody and (2) to explore how prosodic properties of parents’ speech may be indicative of their children’s autism characteristics and/or social competence.

Forty children with and without autism, and their parents, participated in this study. Children in both groups did not significantly differ in age (t(38)=.53, p=.60) or IQ (t(38)=1.53, p=.14). Parents completed the Multidimensional Social Competence Scale (MSCS) to measure their child’s social competence. Children and parents were asked to build a Lego object. Mean fundamental frequency (F0), pitch range, pitch variability, and the total duration of spontaneous speech were extracted from recordings of the dialogue between the parent and child as they completed this task.
A hierarchical linear regression indicated that age and sex did not predict MSCS scores for children with or without autism. When the children’s mean F0, pitch range, pitch variability, and total amount of spontaneous speech were added to the model, these speech characteristics increased the significance of the model for the children without (R2=.777, F(6,12)=6.981, p=.002), but not the children with, autism (R2=.025, p=.999). A significant proportion of the variance in the MSCS scores of the children without autism was accounted for by mean F0 (p<.05), and pitch range (p<.01).

To assess the relationship between the parents’ speech characteristics and their children’s social competence, another hierarchical regression was conducted to see whether the parent’s age and their speech characteristics could predict their child’s social competence. Parents’ speech characteristics, but not their age, accounted for significant variance in MSCS scores of the children with, but not without, autism (R2=.653, F(5,14)=5.267, p=.006). Specifically, the total amount of spontaneous speech produced by the parents accounted for significant variance in their child’s social competence (p<.05).

The results suggest a potential relationship between the development of speech characteristics and social competence in children without autism. The fact that this pattern was not found for autistic children suggests an atypical relationship between the development of their speech communication skills and social development. When examining the speech of parents, a different pattern was found. Specifically, the total amount of spontaneous speech to children with autism, but not without autism, was found to predict a child’s social competence. This second pattern may reflect greater parent scaffolding of their child’s speech communication when that child has autism.

Authors

• Alyssa Janes, University of Western Ontario
  Presenting Author

• Elise K. McClay, Simon Fraser University

• Mandeep Kaur Gurm, Simon Fraser University

• Troy Q Boucher, Simon Fraser University

• Henny Yeung

• Grace Iarocci, Simon Fraser University

• Nichole E Scheerer, Western University