Restricted and Repetitive Behaviors: Restricted By IQ?

Background:  

With the DSM-5, increased attention in autism spectrum disorder (ASD) classification has been given to Restricted and Repetitive Behaviors (RRBs), largely divided into Repetitive Sensory and Motor behaviors (RSM; e.g. rocking, flapping) and Insistence on Sameness (IS; e.g. resistance to change, compulsions).  Further delineation of RRB’s into Circumscribed Interests (CI) and Self-Injurious Behavior has also been supported.  Empirical evidence consistently reveals a strong negative correlation between RSM and Full Scale IQ (FSIQ), yet findings are mixed regarding the interaction of IS, CI, and FSIQ.  Additional research is needed to clarify this relationship and the behavioral phenotype of RRB’s in ASD.    

Objectives:  

The primary aim of the present research is to extend analysis of the relationship between IQ and RRBs in a large multi-study sample of individuals with ASD.  By elucidating covariates of RRBs, phenotypic patterns may emerge that would aid in the classification and diagnosis of ASD.      

A secondary aim of this research is to investigate associated genetic factors that correlate with high levels of RSM, IS, or CI.  These genetic factors will be explored using 518 individuals from Utah families (186 with autism and 332 unaffected relatives genotyped with the Illumina HumanExome chip, and 61 with autism and 27 unaffected relatives with whole exome sequence data from Agilent SureSelect and Illumina GAIIx).

Methods:  

Subjects for this research were ascertained from ongoing studies under the Utah Autism Genetics Project.  The inclusion criteria for the current research are a diagnosis of ASD, a standardized and norm-referenced measure of cognitive ability, and the completion of the Autism Diagnostic Interview-Revised for the RRB scale items.  490 individuals from the Utah Autism Research Database fulfilled the inclusion requirements with the following demographics (mean, standard deviation, range):  Years of Age (15.2, 11.8, 1.3-65.8); FSIQ (87.2, 26.3, 25-155); IS (5.8, 3.7, 0-15); and RSM (5.1, 3.6, 0-14).  Statistical analyses will be conducted in SAS 9.3 and include descriptive, correlational, and regression analyses.  Genetic analyses will be done using PlinkSeq, VAAST, and pVAAST.

Results:  

With the exception of Full Scale IQ and IS (r=-0.04, p=0.36), all correlations reveal significant, negative correlations between RRB’s (IS, RSM) and cognitive ability (FSIQ, VIQ, PIQ) within at least the 0.01 level.  Correlation coefficients ranged from -0.04 (FSIQ, IS) to -0.45 (FSIQ, RSM).  Individual correlation between the 11 specific ADI-R RRB items and FSIQ revealed Circumscribed Interests to have the only positive relationship (r=0.12, p=0.006) amongst all items.  Genetic analyses are pending and results will be reported upon study completion. 

Conclusions:  

This current evidence supports previous documentation of a significant, negative relationship between intellectual ability and RRB’s, and this relationship appears stronger for RSM behaviors than for IS.  Interestingly, the only RRB item showing a positive correlation with IQ was Circumscribed Interests, a finding which supports treatment of this trait as a separate factor of RRBs.  Given the current classification criteria of ASD, these findings reveal a potential quandary in the phenotypic presentation and assessment documentation of RRB’s in those with higher cognitive ability.