Exploring the Cognitive Underpinnings of the Autism Phenotype

Background: Autism Spectrum Disorders (ASD) are currently diagnosed behaviourally based on the presence of the triad of impairments; social impairments, communication impairments and restricted and repetitive behaviours and interests (RRBIs). Three key cognitive theories; a theory of mind deficit, executive dysfunction, and weak central coherence, have been proposed to explain the triad of impairments in ASD. However, surprisingly few studies have investigated whether different cognitive profiles are related to the triad of impairments in ASD.

Objectives: To explore the strength of the relationship between performance on theory of mind, executive function and central coherence tasks, and each impairment of the triad in ASD.

Methods: Participants were drawn from a large population-based sample of adolescent twins, which selected for 111 twin pairs with at least one twin with an ASD and 79 control twin pairs. The ASD group (83% male, average age = 13.5 years) consisted of 87 individuals with a diagnosis of autism, 23 with an ASD, and 32 with a broad spectrum diagnosis (i.e., did not quite meet clinical cut-offs) to cover the complete autism spectrum.  All twin pairs were administered an extensive cognitive battery to measure IQ, language ability, theory of mind ability (mentalising and false belief), executive functioning (planning, mental flexibility, mental initiation, and inhibitory control), and central coherence (local and global processing). The ASD group were behaviourally assessed for ASD symptomatology in the three symptom domains using parent report (Autism Diagnostic Interview-Revised, ADI-R) and direct observation (Autism Diagnostic Observation Schedule-Generic, ADOS-G). For the analyses, the effects of age and IQ were controlled for using regression analyses. Four composite measures of the cognitive tasks were then created based on a priori assumptions about the cognitive ability that each task measured; (1) theory of mind, (2) executive functioning, (3) local coherence and (4) global coherence, to reflect both aspects of central coherence. Linear regression models examined whether the composite measures individually predicted symptom severity for the three ASD symptom domains. Hierarchal regression analyses were also conducted to examine the combined contribution of the composite measures to the three ASD symptom domains. 

Results: Disparate results between the ADOS-G and ADI-R were found. Theory of mind ability alone accounted for statistically significant variation in the severity of ADOS social and communication impairments. Executive function together with local coherence, accounted for statistically significant variation in ADOS-G RRBI symptoms. In contrast, global coherence together with theory of mind ability, accounted for statistically significant variation in both ADI-R social and communication symptoms, and local coherence accounted for statistically significant variation in ADI-R RRBI symptoms.

Conclusions: The findings complement the ‘fractionable’ theory of ASD proposed by Happé and Ronald (2008) that different cognitive deficits/styles may explain different parts of the triad of impairments. However, the three cognitive theories did not fully account for all of the variance in ASD symptoms and therefore may be insufficient to fully explain ASD symptomatology. The cognitive underpinnings of the autism phenotype may be much more complex than a simple unidirectional relationship between cognition and behaviour.