Teacher Responsivity to Child Communication Acts in Autism Preschool Classrooms

Background: Free-play within preschool classrooms provides important, developmentally appropriate intervention opportunities for preschoolers with Autism. With communication deficits being a core feature of autism, teacher responsiveness to child communicative acts during play presents an important opportunity to reinforce and encourage early communication skills.

Objectives: The purpose of this study was to examine the naturally occurring relations between child and teacher communicative behaviors during play with a specific focus on exploring how those relations may change as a function of children’s overall language skills.

Methods: Twenty-nine preschool aged children with autism were videotaped during a 30-minute play period as it naturally occurred during the typical classroom routines. During the play period, children had the opportunity to engage in free-play with peers and toys with adults facilitating the play period as would be typical for each classroom. Videos were then coded using the MOOSES (Multi-Option Observation System for Experimental Studies) recording system. Child communication acts were coded as directed towards a peer, directed towards an adult, or undirected. Teacher intervention codes included commenting, expansion, mapping, repeating, or directives.  A second observer independently coded 30% of the videos for interrater reliability.

Results: Sequential dependencies were examined to explore the relation between children’s communicative acts and teacher intervention.  Yule’s Q was used as the index for this relation based on recommendations from Yoder & Fuerer (2000).  Yule’s Q is a transformed odds ratio bound by -1.0 and 1.0 that controls for the number of coded behaviors which may also be used in subsequent statistical analyses.  A score closer to 1.0 suggests a high degree of dependency (e.g., any child communicative act was responded to by the adult) and a score closer to -1.0 suggests a high degree of independence (e.g., the child exhibited communicative acts, but they are never responded to by a teacher). When examined as a group, pooled Yule’s Q scores indicated that the relation between child communicative acts and teacher intervention tended occur more independently of the other (Q = -.30).  Hypothesizing that this relation may be influenced by the communication skill level of the child, children were divided into two groups (Average and Below Average Language Skills) based on PLS-4 scores.  When compared, sequential dependencies between child communication acts and teacher intervention for the Average skills group remained similar to findings for the whole group (Yule’s Q = -.33) and the Below Average skills group experienced a near zero relation (Yule’s Q = .02). 

Conclusions: From a behavioral perspective, a higher degree of sequential dependency between child communicative acts and teacher intervention may be important to reinforcing and teaching early language and communication skills.  In this study, despite the occurrence of child communicative acts and the occurrence of teacher behaviors meant to support skill development (i.e., language mapping, expanding, commenting, etc.), there was little sequential dependency between those sets of behaviors.  Preliminary findings do suggest, however, that there may be some difference in the nature of those dependencies as a function of the child’s overall language proficiency.