Preschool Language and Cognitive Predictors of Middle Childhood Reading Abilities in ASD and Typically Developing Children

Background: A growing body of research has demonstrated that reading abilities in individuals with autism spectrum disorder (ASD) and no intellectual disability (ID) are heterogeneous and that basic reading skills such as rate, accuracy and fluency may display greater developmental dissociation from higher order reading comprehension skills in ASD than in those with typical development (TD). Variability in reading skills has been shown to relate to concurrent oral language skills. However, little is known about the developmental trajectory of this association.

Objectives: The purpose of the current study was to examine preschool language and cognitive predictors of reading achievement more than eight years later, in middle childhood, in a sample of children with ASD compared to those with TD.

Methods: Participants were part of the Autism Phenome Project (APP) longitudinal cohort, and included children with ASD (N=32) and TD (N=43) without ID (middle childhood FIQ range 70-140), who were first assessed at ages 24-54 months (mean age=36.99 months, SD=5.73 months) and then followed-up in middle childhood at ages 115-163 months (mean age=138.26 months, SD=9.75 months). Time 1 assessments: IQ, Mullen Scales of Learning (MSEL); receptive vocabulary, Peabody Picture Vocabulary Test (PPVT); and expressive vocabulary, Expressive One Word Picture Vocabulary Test (EOPVT). Middle childhood assessments: IQ, Differential Ability Scales, Second Edition (DAS-II); reading rate, accuracy, fluency, and comprehension, Gray Oral Reading Test, Fifth Edition (GORT-5).

Results: While the ASD group had average middle childhood FIQ (M=96.06, SD=15.10), they scored significantly lower than TD, therefore ANCOVA’s controlling for FIQ were conducted on all reading variables. GORT-5 reading rate, accuracy and fluency were not significantly different between ASD and TD groups. However, ASD scored significantly lower than TD on reading comprehension (f=9.19, p=.003). Separate hierarchical linear regressions were conducted for ASD and TD groups examining T1 predictors (MSEL VIQ & NVIQ, PPVT, EOWPVT) of middle childhood GORT-5 reading rate, accuracy, fluency, and comprehension. None of the T1 variables predicted significant variance in any TD reading measure. However, in the ASD group, when controlling for other T1 variables, the PPVT was a significant predictor of reading rate (β=.882, p=.048), reading accuracy (β=1.093, p=.011), reading fluency (β=1.043, p=.018), and reading comprehension (β=1.018, p=.008). This was the only significant preschool predictor of reading achievement in the sample.

Conclusions: The current study provides evidence that, consistent with previous research, basic reading skills are a relative strength for children with ASD, but reading comprehension is a relative weakness and the patterns of association between basic and higher order skills are weaker in ASD than TD. Furthermore, while the preschool cognitive and language variables in this study did not explain significant variance in the TD reading scores, preschool receptive vocabulary was a significant predictor of all reading variables in the ASD sample. This indicates a tighter association between early receptive oral language skills and later reading (receptive written language) skills in ASD than TD samples and has implications for the role both language and reading interventions may play in the development and education of children with ASD.