Sensory Processing Difficulties Predict Internalizing Symptoms, ADHD Symptoms, and Repetitive Behaviors in 4-10-year old Children with Williams Syndrome

• ¹ University of Louisville, Department of Psychological and Brain Sciences, United States

Purpose: Children with Williams syndrome (WS) frequently meet DSM-IV diagnostic criteria for psychiatric disorders. For example, Leyfer et al. (2006) found that 54% of 119 children aged 4 – 16 years met DSM-IV criteria for specific phobia, 12% for generalized anxiety disorder, and 65% for Attention Deficit Hyperactivity Disorder (ADHD). McNally & Mervis (2006) found that based on parent report, 95% of children with WS engage in at least one type of repetitive behavior. As there is likely significant co-morbidity among these symptoms, it is important to investigate factors which may help explain the links between them. 52% of children with WS with specific phobia have a phobia of loud noises, suggesting sensory processing as a possible predictor. The purpose of this study was to investigate the relations between sensory abnormalities in children with WS and internalizing symptoms (IS; includes anxiety and depression), ADHD symptoms (ADHD), and repetitive behaviors (RB), using path analysis.

Methods: Participants were 42 children with WS aged 4.12 – 10.95 years (mean: 8.12 years, SD=1.79). Children’s scores for IS, ADHD, RB, and SP were determined based on questionnaires completed by their parents. IS was operationalized as the Internalizing Problems T score from the Child Behavior Checklist-Parent Report (Achenbach, 2001), ADHD as the ADHD index T score from the Conner’s Parent Rating Scale-Revised (Conners, 1997). RB as the total raw score for the Repetitive Behavior Scale-Revised (Bodfish et al., 1999), and SP as the total raw score for the Short Sensory Profile (Dunn, 1999).

Results: Correlational analyses were conducted to examine the relations among SP, IS, ADHD, and RB. Significant correlations were found between all of the variables, with the exception of RB and ADHD, which approached significance. To test the hypothesis that sensory processing abnormalities predicted IS, ADHD, and RB, a path analysis was conducted. The resulting model provided a significant fit to the data (χ2 = 2.80, p > .05; CFI = 1.0; RMSEA < .05), with each individual path yielding a significant contribution (β = -.60 for IS, p < .001; β = -.45 for ADHD, p < .01; β = -.63 for RB, p < .001; see figure). Given the bivariate relations between the dependent variables, a second model was conducted to estimate the correlation among error terms for IS, ADHD symptoms, and RB. All error correlations within the model were nonsignificant, suggesting that these variables do not share any additional variance not accounted for by SP.

Discussion: The path analysis results indicate that much of the shared variance in the comorbid symptoms of IS, ADHD, and RB may be explained by sensory-related processing difficulties. Theoretical implications and applications will be discussed.