Early Predictors of Parental Linguistic Mapping in Preschoolers with Autism Spectrum Disorders

Background: Parents may interpret children’s frequent use of coordinated attention to communicative partner and referent (i.e. triadic attention) as an indicator that children will attend to, understand, and benefit from linguistic responses, which may in turn facilitate subsequent language development. Two types of triadic attention are (a) attention following and (b) intentional communication. One type of parental linguistic response occurs when parents “put into words” the presumed meaning of children’s immediately preceding communication acts: Parental linguistic mapping (PLM). Frequency of PLM predicts later language in several populations of children with disabilities, including children with ASD. Identifying the predictors of PLM is one piece of evidence needed to identify pivotal skills for language development in children with ASD. 

Objectives: Predictive relations were examined between (a) early frequency of intentional communication and attention following and (b) later frequency of PLM. 

Methods: The study used a longitudinal correlational design (3 assessment periods conducted over 12 months). Although participants could get treatment, our project did not provide any during the study. Eighty-one minimally verbal preschoolers (age M = 35.2 months, SD= 7; mental age M = 12.3 months, SD = 5) participated in the current study. All children received a clinical diagnosis of autism (95.1%) or PDD-NOS (4.9%) from a licensed psychologist who used gold standard instruments, clinical best judgment according to DSM-IV-TR criteria. The two triadic attention skills were measured at Times 1 and Time 2 (4 months from Time 1) using the Communication and Symbolic Behavior Scale, Early Social Communication Scales and responsive examiner-child language samples. Frequency of PLM was measured at Time 2 and Time 3 (8 months from Time 2) using a semi-structured 10-minute parent-child snack session. Two multiple regression analyses using frequency of PLM at Time 2 as a control variable, one of the triadic attention skills as the predictor of interest, and frequency of PLM at Time 3 as the criterion variable were conducted. 

Results: Initial frequency of PLM (partial r = .44) and early attention following (partial r = .27) predicted later frequency of PLM, R²=0.29, p<0.01. Initial frequency of PLM (partial r = .45) and early intentional communication (partial r = .25) predicted later frequency of PLM, R²=0.28, p<0.01. 

Conclusions: The results are consistent with the hypothesis that the two triadic skills may encourage parents to use PLM. Identifying the predictors of PLM may sensitize parent trainers to the reasons some parents use more PLM than others. It is also possible that teaching parents to use another type of linguistic response (e.g., talking about children’s focus of attention - follow-in utterances) might be a more adaptive parental skill during a period in which their children are not using triadic skills frequently. Indeed, past research in preschoolers with ASD found that PLM was predictive of later vocabulary only if children had a threshold level of comprehension (a probable correlate of triadic attention). Instead, follow-in utterances were predictive of language, even for the low-comprehension children (McDuffie & Yoder, 2010). More research is needed to address these possibilities.