19700
The Role of Gestures in Early Language Development in Children with ASD

Saturday, May 16, 2015: 10:30 AM
Grand Ballroom D (Grand America Hotel)
A. Goodwin1, S. Goldin-Meadow2, D. A. Fein3 and L. Naigles4, (1)Waisman Center, University of Wisconsin, Madison, WI, (2)University of Chicago, Chicago, IL, (3)Psychology, University of Connecticut, Storrs, CT, (4)University of Connecticut, Storrs, CT
Background:  

Gestures play an important role in language development. For example, word-gesture combinations have been shown to precede word-word combinations in both typically-developing (TD) children and children with early brain injury (Özçalışkan, Levine & Goldin-Meadow, 2013). However, gesturing is known to be impaired in children with ASD (de Marchena & Eigsti, 2010).  Nevertheless, language development is heterogeneous in children with ASD.  Some children with ASD may use gesture to facilitate later language, while others do not.

Objectives:  

This study investigates whether young children with ASD differ from language-matched TD children in gesture use, and how early gestures relate to later language development.

Methods:  

Sixteen children with ASD (M age = 32.67 months) were divided into high-verbal (HV) and low-verbal (LV) groups based on Mullen expressive-language scores. Sixteen TD children (M age = 23.52 months) were matched on language level and also divided into two groups. Children were visited every four months across a two-year period.  Mother-child interactions were video-recorded at each visit and coded for gesture at visit 1 and children’s speech (i.e., word types & MLU) at visits 1-6. Gestures were assigned codes indicating whether they: reinforce-speech, add-meaning-to-speech, disambiguate-speech, emphasize-speech, serve as a functional-act, or occur with no-speech. Counts and ratios were calculated for each type. HV- and LV-ASD group means were compared to means of the TD language-matches, and relationships between early gesture and later speech were investigated for the ASD groups.

Results:  

The HV-ASD group produced significantly fewer total gestures (M=20.75) than their language-matched TD peers (M=44.25; F(1)=7.722, p=.015). They also had lower proportions of total utterances (M=.084), words (M=.053), and communicative acts (M=.080) that were accompanied by gestures than the TD children (Ms>.097; Fs > 4.75, ps < .047).  The LV-ASD group did not produce significantly fewer gestures than their LV-TD language-matches (Ms=10.13 & 18.63, respectively; p=.065); these groups differed significantly only in no-speech gestures (Ms=5.88 & 14.88, respectively; F(1)=6.787, p=.031). Language measures at visits 2-6 were regressed on gesture measures from visit 1, after controlling for word types & MLU at visit 1. For the HV-ASD group, children who produced more add-meaning-to-speech gestures and a higher ratio of disambiguate-speech gestures produced more word types at later visits than children who produced fewer of these gestures (ΔR2=.357, p=.049 and ΔR2=.258, p=.011, respectively). Children who produced more disambiguate-speech gestures had higher MLUs at visit 4 (ΔR2=.158, p=.037). No gesture measures predicted later language in the LV-ASD group.

Conclusions:  

 LV-ASD and LV-TD children produced similarly few gestures. HV-ASD children produced significantly fewer gestures than their HV-TD language-matches, suggesting that HV children with ASD may not rely on gesture to bootstrap early language to the same extent as TD children. However, some gesturing of the HV-ASD children positively predicted their later language levels. The LV-ASD group did not show the same effects.  These findings contribute to the large body of research demonstrating relationships between language and gesture, while showing that not all children with ASD show the same pattern as found in typical development.