Exploring Problem Behavior and Language Development in Preschoolers with Autism Spectrum Disorder

Background: Children with autism spectrum disorder (ASD) consistently demonstrate high levels of externalizing behavior problems compared to their typically developing peers, though significant heterogeneity is present (e.g., Anderson, Maye, & Lord, 2011). Behavior problems occur in many domains of functioning for children with ASD and have the potential to interrupt learning opportunities, social interactions, and other critical points of engagement.

Objectives: The present study examined the contribution of children’s problem behaviors to later expressive language functioning. These data were drawn from a larger study preschool intervention project for children with ASD, however the present focus is on individual differences.

Methods: The sample includes 137 preschool children (85% male, M = 4.2 years, SD = .62) diagnosed with ASD and enrolled in one of 78 participating classrooms across four states. Children participated in an assessment battery at the beginning of the school year targeting general developmental level (MSEL), problem behaviors (CBCL CTRF), and autism symptoms (ADOS-2). At the end of the year, problem behaviors were reassessed and expressive language was evaluated using the PLS-5. At the beginning of the year 37 children used less than 5 words during the ADOS assessment (27%), 76 regularly used single words (55.5%), and 24 were using phrase speech (17.5%).

A hierarchical regression was conducted with expressive language as the dependent variable (PLS-5 Expressive Communication Standard Score), child age and developmental level (MSEL Standard Score) entered as predictors on the first step, and problem behaviors at the beginning of the year (CBCL CTRF Total T-score) on the second step.

Results: The full model predicted 46.8% of the variance in expressive language, F(3,133) = 40.94, p < .001. Problem behaviors explained an additional 2.5% of the variance beyond the lower level predictors, β = -.162, p < .05, F change(1,133) = 6.39, p < .05. Even when considering only the children with minimal expressive language at the start of the year (completed ADOS Module 1, n = 113), the full model predicted 36.2% of the variance in expressive language scores, F(3,109) = 22.19, p < .001. Problem behaviors explained an additional 3.7% of the variance beyond the lower level predictors, β = -.195, p < .05, F change(1,109) = 6.56, p < .05. Problem behaviors at the end of the year were significantly related to problem behaviors at end of the beginning of the year, r = .61, p < .001, and to expressive language at the end of the year, r = -.32, p < .001.

Conclusions: Results show that children exhibiting more reported problem behaviors at the beginning of the year demonstrated lower expressive language skills at the end of the year. It appears that problem behaviors have a significant impact on language both concurrently and across time. Significant problem behaviors may be arising in the absence of matured expressive language, and continued communication interference may exacerbate these issues highlighting the need for early communication interventions. However, further exploration is needed around stronger predictors of language development in children with more significant language delays.