Investigating Relationships Between Linguistic and Pictorial Symbolic Domains in Children with Autism Spectrum Disorder and Typical Development

Thursday, May 11, 2017: 12:00 PM-1:40 PM
Golden Gate Ballroom (Marriott Marquis Hotel)
C. Hartley1, A. Trainer2 and M. L. L. Allen3, (1)Psychology, Lancaster University, Lancaster, UNITED KINGDOM, (2)Northumbria Healthcare NHS Foundation Trust, North Tyneside, United Kingdom, (3)Lancaster University, Lancaster, UNITED KINGDOM

If children are to become effective communicators, it is vital that they learn to comprehend and produce linguistic and pictorial symbols. For typically developing (TD) children, early understanding of pictures is scaffolded by their superior and early-emerging understanding of language. However, autism spectrum disorder (ASD) is often characterised by severe and prolonged language impairments, and recent evidence shows that minimally verbal individuals also have an atypical understanding of pictures. Here, we explore the possibility that deficits in linguistic and pictorial domains are causally related in ASD.


For the first time, this study modelled concurrent predictive relationships between comprehension and production of language and pictures (plus non-verbal intelligence, chronological age, and autism severity) in linguistically-delayed children with ASD and TD controls.


Participants were 30 children with ASD (M age = 11;4) and 24 TD children (M age = 4;5). The Mullen Scales of Early Learning measured language comprehension (ASD: M = 3;7; TD: M = 4;5) and language production (ASD: M = 3;4 ; TD: M = 4;8). The Leiter-R measured non-verbal intelligence (ASD: M score = 63.5; TD: M score = 50.46) and the Childhood Autism Rating Scale measured autism severity (ASD: M score = 36.93; TD: M score = 15.42). Picture comprehension was assessed by asking children to identify the 3-D referents of line drawings. Picture production was assessed by asking children to draw unfamiliar 3-D objects and having independent raters identify their referents. Predictive relationships between domains were modelled via hierarchical regressions.


For children with ASD, pictorial understanding was predicted by language abilities (β = .41-.64, p = .03-.001), and autism severity (β = -.45, p = .02). Language comprehension and production predicted each other (β= .65-.85, p < .001), and the latter was also predicted by autism severity (β = -.22, p = .02). For TD children, picture comprehension was predicted by language abilities (β = .38-.42, p = .045-.05), which in turn were predicted by non-verbal intelligence and chronological age (β = .34-.48, p = .04-.02). Comparisons between subgroups (N = 16 per population) matched on language comprehension and production revealed no differences in pictorial understanding. However, when matched on non-verbal intelligence, children with ASD showed significantly reduced pictorial understanding relative to TD peers (t = 5.2-32.3, p = .03-<.001).


As in typical development, pictorial understanding in ASD is scaffolded by language. This between-domain relationship is similar to that observed in young TD children, and aligns with Vygotskian social-cultural theories of development rather than domain-specific theories. The observed predictive relationships indicate that referential language impairments may elicit downstream deficits in picture comprehension and production. Consequently, children with severely impaired language may not understand the representational nature of pictures. From an applied perspective, this has important implications for children’s learning and usage of picture-based communication interventions. However, improving language through targeted interventions may elicit advances in understanding across multiple symbol systems.