16360
Respiratory Sinus Arrhythmia As a Predictor of Language Outcomes in Initially Nonverbal Children with Autism
Absence of functional spoken language beyond the preschool years places children with autism spectrum disorders (ASD) at grave risk for poor social and adaptive outcomes in adulthood. Although considerable research has provided insight into predictors of later spoken language outcomes among children with ASD, minimal attention has been given to biologically-based predictors. Examining such predictors may provide insight into underlying mechanisms accounting for variable spoken language outcomes in ASD. Porges (1994, 2007) proposed that respiratory sinus arrhythmia (RSA) indexes influences of the neocortex on a complex social engagement system that includes the vagus. We have been especially interested in attention to child-directed speech (CDS) among children with ASD as a predictor of later language outcomes. Arguably, RSA indexes a key physiological substrate for sustained attention in social contexts. We previously demonstrated that RSA during exposure to CDS accounted for significant variance in expressive language outcomes of preschoolers with ASD that was unexplained by amount of looking at these stimuli (Watson et al, 2010).
Objectives:
To determine the value of RSA during CDS, combined with non-biological variables, in predicting expressive language outcomes in initially nonverbal preschoolers with ASD.
Methods:
Fifty-two initially nonverbal preschoolers with ASD were assessed 5 times in 4-month intervals over 16 months. Expressive vocabulary size was estimated using a parent report tool (i.e., McArthur-Bates Communicative Development Inventory, Words and Gestures), collected at each assessment. Mixed level modeling was used to quantify individual vocabulary growth curves. The Time 5-centered intercept of expressive vocabulary was selected as the outcome measure. Putative predictors, measured at Times 1 and/or 2, included child communicative acts, response to joint attention, consonant inventory, oral-motor skills, motor imitation, vocabulary comprehension, and parent linguistic responsivity, as well as percent of time spent looking at CDS vignettes (3 minutes) and RSA during these vignettes. Based on prior analyses (excluding CDS variables) of unique predictors of expressive vocabulary (Yoder & Watson, in preparation), the final regression model for the current analysis included four putative predictors: child communicative acts, parent linguistic responsivity, percent of time looking at CDS, and RSA during CDS.
Results:
The full model accounted for 36.6% of the variance in expressive vocabulary outcomes, F(4,46) = 6.6, p < .001. Child communicative acts, parent responsivity, and RSA during CDS each accounted for significant unique variance (ps <.05), whereas the percent of time spent looking at CDS did not (p = .42).
Conclusions:
This study expands on our previous findings to demonstrate that RSA accounts for variance in expressive language outcomes beyond that accounted for by other significant predictors. The physiological substrates underlying attention to CDS appear to influence ongoing development of expressive language. Based on Porges (1994, 2007), children with ASD who have higher RSA during CDS possibly are those experiencing stronger social affiliation. Merely “paying attention” (i.e., looking and presumably listening) to language input may be insufficient for optimal expressive language development in children with ASD; perhaps instead, the extent to which children process language input as a positive social phenomenon is key to their outcomes.