Predictors of Growth in Communication Complexity over 16 Months for Children with ASD

Background: One of the core deficits of autism spectrum disorder (ASD) is in the area of communication. As a result children with ASD often struggle to develop early language skills, interact with others, and form relationships with both peers and adults. The importance of understanding the trajectory of communication development cannot be understated due to the role of communication in future social and academic progress.  Research also shows that play (e.g., toy play, symbolic play), joint attention (i.e., sharing attention, following and directing the attention of others), and imitation are strong correlates with later language skills (Kasari, Freeman, and Paparella, 2006; Toth, Munson, Meltzoff, & Dawson, 2006). The frequency of intentional communication that is weighted by the complexity of the form (lower weight for nonverbal and highest weight for multi-word) used to communicate has been shown to grow in a positive linear or quadratic fashion during the nonverbal to early multiword stages of communication development in typically developing children and in infant siblings of children with ASD (Yoder, Stone, Walden, Malesa, 2009). Growth of weighted intentional communication has not been model, nor have its correlates been identified, in children with ASD.

Objectives: Do measures of play, imitation, and communication positively co-vary with growth curves of weighted intentional communication (WIC) of participants 16 months after study entry, controlling for IQ and ADOS social communication algorithm?

Methods: Our participants included 86 pre-school initially nonverbal or minimally-verbal children with a diagnosis of ASD.  Their scores on the ADOS diagnostic algorithm ranged from 6 to 28 with a mean of 22.57 (SD=3.85).  Standard scores on the Mullen Scales of Early Learning ranged from 49 to 122 with a mean of 51.56 (SD = 8.49).  A measure of WIC was taken at 5 time-points, each 4 months apart.  Weighting was accomplished by assigning one point to nonverbal communication, two points to single word communication, and three points to multi-word communication.  All predictors were measured at Time 1.  Communication was measured using scales 4-7 of the Communication and Symbolic Behavior Scale (CSBS; Wetherby & Prizant, 1993).  Imitation was measured using Motor Imitation Scale (MIS; Ousley & Littleford, 1997), and number of differentiated play activities was measured from the Development Play Assessment (DPA; Lifter, 2000). Mixed level modeling was used to test whether play, imitation, and communication accounted for unique variance in growth of WIC. ASD severity and cognitive impairment levels were analyzed as control variables.

Results: Preliminary analyses indicate that play action diversity, motor imitation, and communication predicted growth of WIC over the 16-month period, even after controlling for severity of autism and IQ.

Conclusions: Play skills, motor imitation, and measures of communication are important predictors of growth in WIC for young children with ASD.  These results can be used to focus interventions on specific skill areas for these children.