Skip to main content

Predicting Useful Speech in Children with Autism


The Interagency Autism Coordinating Committee (IACC) "roadmap" set a long-term goal of helping 90% of children with ASD achieve "useful speech" (Health and Human Service, 2004), which has been determined to mean words that are frequent, communicative, and referential. Vocabulary comprehension has proven to be a primary predictor of useful speech (Wetherby, Watt, Morgan, & Shumway, 2007). Identifying early predictors of variation in developing useful speech and vocabulary comprehension may help address this goal, particularly when guided by a carefully considered conceptual model. In children with autism, the search for predictors of later language has been restricted almost exclusively to child factors. The predictive model for the project includes child and parent putative predictors. The theoretical model for the funded research project includes three of the most promising parental variables: (a) frequency of parental linguistic mapping of child intentional communication, (b) frequency of parental nonobligatory talk that is synchronous with the child's focus of attention, and (c) parental tendency to use responsive strategies. The child predictor constructs include the replicated child predictors of later language development in children with autism and other disabilities: (a) intentional communication, (b) attention following, and (c) object play. Also included are less investigated variables specific to the autism population: (d) child attention to child-directed speech and (e) child motor skills. The longitudinal correlational design will require multiple assessments of the variables at five 4 month intervals to address a total of nine aims. The first two aims will address the overall goal of examining the predictive validity of the child factors and the potential mediating role of parental input on (1) growth rate of child communicative word use and (2) growth rate of child comprehension. The third aim will examine whether the association between early parental behaviors and child communicative word use is mediated through child development of vocabulary comprehension. Aims four through six test longitudinal correlations and will examine the effects of child behaviors (e.g. intentional communication, attention following, and object play) on parental use of facilitative behaviors (e.g. linguistic mapping, synchronous nonobligatory talk, and responsiveness). Finally, aims seven through nine will test various alternative models against key predictions of the research model by looking at the associations between early behaviors/skills and later behaviors/skills.

This project provides a Minority Investigator Supplement to support an African-American graduate student currently completing their PhD at the University of North Carolina at Chapel Hill. The goal of this supplement is to provide additional training and research experiences in order to help the candidate achieve their goal of becoming an independent scientific investigator in the field of autism.

Journal of Consulting and Clinical Psychology 

American Journal of Speech-Language Pathology