Exploring the Narrative Writing Skills of Students with High-Functioning Autism Spectrum Disorders

Background: No genre is as dependent on social cognition as the narrative genre. For students with autism spectrum disorder (ASD), core weaknesses in social understanding and communication could make writing narratives one of their most challenging academics areas. Despite this, no studies to date have examined the narrative writing skills of children and adolescents with ASD, beyond a few using standardized tests that suggested global writing deficits in this population. Furthermore, previous research has confounded language ability with a diagnosis of ASD. Systematic description of the strengths and weaknesses in the narratives of students with ASD is critical to disentangling oral language impairments from autism within the writing problems of students with ASD, and this is critical in providing a foundation for developing targeted interventions.

Objectives: The aim of this study was to compare the narrative writing of students with ASD to their typically-developing (TD) peers when both groups were balanced in terms of language ability. We hypothesized that ASD would contribute to written language strengths and weaknesses beyond what would be predicted by oral language skill alone. A second aim was to explore the variables that best predicted writing quality across both groups.

Methods: Twenty-four students with ASD and 22 TD controls aged 8 to 17 years participated in this study. Participants completed standardized assessments of oral language skill, performance IQ, and ASD symptomology, as well as an untimed narrative writing task. Their narratives were coded on 20 text variables across four lower order categories: productivity, syntactic complexity, lexical diversity, and writing conventions, as well as two higher order composites: overall narrative clarity and overall narrative quality.

Results: Omnibus MANCOVA (controlling for age) revealed that the students with ASD showed greater lexical diversity in their texts compared to their TD peers (d = +0.9 SD). ANCOVA revealed that the texts of students with ASD had lower overall narrative quality than controls (d = -0.7 SD). The groups did not differ significantly on any other variables. Forward multiple regression revealed that age, language ability and ASD symptomology (but not PIQ) were significant predictors of overall narrative quality (R²=.557).

Conclusions: This was the first detailed investigation of the narrative writing skills of students with ASD and the first to ensure both groups were equivalent on language ability. Results revealed many similarities in the texts of students with ASD and controls. However, individuals with ASD tended to have difficulty with their use of narrative elements and form, i.e., problems with narrative structure and organization, character development and integrating the inner worlds of their characters with the events in the story.

We theorized that individuals with ASD may have underdeveloped narrative concepts and structures, and that they may engage in narrative thought less often. Providing repeated opportunities for youth with ASD to engage in narrative thought, such as through the creation of written narratives, could help them to better understand human acts and events. As such, developing interventions to support the narrative abilities of students with ASD may have far-reaching consequences.