Autism is a severe, neurodevelopmental disorder that often confers a profound burden on individuals with autism, their families, and society. Research aimed at uncovering the pathogenesis of this condition may lead to evidence based approaches to prevention or treatment, and is therefore of great importance. Strong evidence supports a genetic etiology in autism, and twin and family studies have also shown that genetic liability appears to be expressed among unaffected relatives of people with autism through features that are milder, but qualitatively similar, to the defining characteristics of autism. This constellation of subclinical language and personality features is commonly referred to as the ‘broad autism phenotype’ or ‘BAP’. Importantly, whereas by definition autism involves serious impairment across all three symptom domains, evidence suggests that such features may decouple and segregate independently in unaffected (with autism) relatives with the BAP. Therefore, studies of relatives of individuals with autism can help to simplify the complex autism phenotype and identify component traits which are more amenable to genetic dissection than the full clinical syndrome. In this study, we focused on defining genetically meaningful language phenotypes among individuals with autism and their relatives, that may be applied in genetic studies.
Study period 1: Using a family study design, we implemented a detailed psycholinguistic assessment battery for use in families of individuals with autism and controls. This battery of objective, experimentally derived psycholinguistic measures of language processing produced findings that throw into sharper relief current understanding of key mechanisms underlying the language impairments associated with autism and the BAP. Results also provided quantitative measures that may be used in genetic studies, and which could be targeted in clinical intervention efforts. With senior coinvestigators with expertise in genetics, we established a Biobank including these rich phenotypes and DNA samples from all families that will be used for future genetic studies, and more immediately, to follow up on promising findings sure to emerge from several large scale Genomewide studies of autism underway.
Study period 2: Differences in pragmatic (i.e., social) language use are the language deficit most robustly associated with autism and the Broad Autism Phenotype (or, BAP, a constellation of subclinical features relatives that parallel in quality the defining characteristics of autism and are believed to reflect genetic liability in unaffected relatives). Findings from the prior funding period indicate that our archival, retrospective measures of parents’ childhood language and cognitive development, together with our battery of computational linguistic and psycholinguistic measures of pragmatic ability are highly promising markers of autism endophenotypes, measurable in individuals with autism and parents (even in childhood), and therefore may help significantly refine phenotypes that can be targeted in genetic, neurobiological, and treatment studies. This renewal builds on findings from the original grant in several specific ways. First, we will capitalize on an additional, extremely valuable resource for archival language, cognitive, and social information on relatives of individuals with autism to track markers of genetic liability to autism through multiple generations. Second, we will investigate the biological basis of our candidate pragmatic language endophenotypes through analysis of auditory brain stem response related to pragmatic language. Third, we will measure a candidate protein expressed in the brain, with strong connections to the auditory brain stem, the Fragile X Mental Retardation Protein, and which preliminary data show may relate to pragmatic language features in family members of individuals with autism.