The Effects of Parent Sentence Diversity Input on Grammatical Development in Toddlers with ASD in Early Intervention Programs

Background: Parents’ sentence diversity (operationalized as unique combinations of subjects and verbs) serves as a lexical model, promoting grammatical development in children (Hadley & Walsh, 2014). Parent-implemented intervention increases sentence diversity in typically developing toddlers, and more diverse lexical noun phrase subjects in parents significantly predict children’s increased sentence diversity (Hadley et al., 2017). This relationship between parents’ lexical flexibility and children’s early grammatical growth is promising for children with ASD given that many early interventions are parent-mediated (Schreibman et al., 2015). However, it is unknown how parents’ language input contributes to early grammar in children with ASD over the course of early intervention.

Objectives: We 1) measured changes in the sentence diversity in toddlers with ASD over the course of treatments; 2) examined the impact of caregivers’ sentence diversity on children’s language; and 3) identified the predictors of grammatical changes in parents and children.

Methods: Participants were drawn from 72 caregiver-child dyads with ASD who participated in early intervention programs. Preliminary data included 10-minute play sessions from 34 dyads (Time 1: M=30 months, SD=2.6; Time 2: M=36 months, SD=2.4). Sentence diversity was quantified with a number of unique subject-verb combinations for parents and children from the play sessions while adapting the existing coding scheme from Hadley et al. (2017). The Brief Observation of Social Communication Change (BOSCC; Grzadzinski et al., 2016) was used to assess children’s social communication. Children were divided into two groups; those whose sentence diversity scores improved (“Improving” n=20) versus those whose were stable (“Stable” n=14) over time using the median score. Baseline cognitive and social communication skills of these two groups of children were compared using t-tests. Pearson’s r correlations between parents’ and children’s sentence diversity change scores were examined.

Results: Preliminary analyses showed that sentence diversity in children increased significantly over the six months of treatment with a medium effect size (p<0.001, Cohen’s d=.5). Parent sentence diversity improvement was marginally significant (p=0.055, d=.3). Parent and child sentence diversity scores were significantly correlated both at Time 1 (r=.5, p<.05) and Time 2 (r=.6, p<.001). Improvements in child sentence diversity from Time 1 to Time 2 were correlated with higher levels of parent sentence diversity at Time 2 (r=.4, p<.05). The Improving group had a higher baseline non-verbal IQ (Stable: M=73, SD=23.2; Improving: M=93, SD=11.5; p<0.01) and less social communication deficits (BOSCC Social Communication; Stable: M=27, SD=7.3; Improving: M=19, SD=5.5; p<0.001). Parents with higher education levels made more significant improvements in their sentence diversity (p<0.05).

Conclusions: Improvements in sentence diversity were examined in both children with ASD and their parents over the course of early interventions. Results also highlight the reciprocal interaction between increased parent sentence diversity and more advanced grammatical skills in toddlers with ASD. Higher cognitive skills and less social impairments may facilitate early grammatical growth in these children. Parental education level may also moderate changes in their lexical flexibility. The final analyses will employ generalized linear mixed models with a larger sample from an on-going study.