Social Communication and Autistic Triad Trait Patterns in Preschoolers with Severe Visual Impairment

Background:  

Severe visual impairment (VI) is associated with significant difficulties in social communication; 11-40% of children with VI are diagnosed with autism spectrum disorder (ASD). Relatively little is understood of the nature and aetiology of social communication and ASD in VI. Of debate is whether children with VI and ASD have a similar phenotype to Sighted ASD children or whether there are qualitative differences.  There have been few systematic studies of the social communicative phenotype of young children with VI due to methodological challenges and population rarity.

Objectives:  

Using a newly developed and validated Visual Impairment Social Communication schedule (VISS) (Absoud et al 2010), this study aims to investigate the social communicative phenotype of preschoolers with severe VI. Patterns within the triad traits are considered in the context of ‘At Risk’ or ‘Not At Risk’ for ASD diagnosis, vision level (profound; severe VI) and developmental quotient (DQ).

Methods:  

A consecutive clinic sample of 35 preschoolers with VI (mean age 39.4 months, range 1.5 to 5 years) attending a specialist paediatric developmental vision clinic was included. A trained Research Assistant used the observational VISS to code social interaction (SI), communication and language (C) and play and behaviour (B) during the play based assessment with the clinician.  The coded data was entered into a database and the sample was divided into ‘At Risk’ and ‘Not at Risk’ of ASD subgroups, according to whether they scored above or below the clinical threshold on total VISS score (Absoud et al). Statistical analysis was undertaken between and within the two subgroups.

Results:  

43% of the sample was found to be ‘At Risk’ and 57% ‘Not At Risk’ for ASD, with a significant difference in mean score, indicative of a bimodal distribution.  A mixed between-within subjects analysis of variance revealed that children ‘At Risk’ performed consistently worse across all triadic domains, in comparison to children ‘Not at Risk’, F (1, 33) = 107.05, p <.0005, partial eta squared = .76. A main effect for triadic area was also revealed (Wilks Lambda = .62 F (2, 32) = 9.78, p < .0005, partial eta squared = .38), with both groups showing more impairments in C relative to SI and B domains.  The triad domains were significantly positively correlated within the ‘At Risk’ group, whereas there were no significant associations in the ‘Not at Risk’ group. DQ was not associated with VISS scores in either group.

Conclusions:  

43% of the sample was found to be ‘At Risk’ for ASD, in line with previous studies.  As in classical Kanner’s autism, the scores were extremely low across all three triad domains, suggesting that VI-ASD phenotype was similar to the phenotype of Sighted-ASD preschoolers. However, both the ‘At Risk’ and ‘Not At Risk’ subgroups showed a greater impairment in social communication relative to social interaction and behaviour domains. This finding argues for possible ‘fractionation’ of the triad in the more successfully developing preschoolers and raises important theoretical questions about the vulnerability and aetiology of social communication and ASD in VI preschoolers.