New Parent Report Measure of Cognitive Inflexibility and Related Symptoms in High Functioning ASD

Background: Restricted, repetitive, behaviors and interest (RRBI) symptoms are core to autism spectrum disorders (ASD), particularly higher-order RRBIs which discriminate ASD from other neurodevelopmental disorders. Furthermore, lab based and ecologically valid measures show a relationship between RRBIs and cognitive and behavioral inflexibility.  Accurate phenotyping of these symptoms on continuous scales is useful for measuring the effect of treatments and identifying gene-behavior relationships.  The Autism Diagnostic Interview (ADI) provides a gold standard diagnostic evaluation of RRBI symptoms and has been relied on heavily to characterize the RRBI phenotype, but it is intended as a diagnostic, not a quantitative measure.  The Behavior Rating Inventory of Executive Function (BRIEF) is a standardized instrument that provides quantitative measurement of cognitive and behavioral flexibility, and has consistently identified flexibility deficits in ASD, but it does not measure RRBI symptoms.  A measure tapping both these aspects of flexibility with a larger set of items could increase power and specificity when describing the inflexibility phenotype in ASD.

Objectives: To investigate the internal consistency, and discriminant and construct validity of a new 50 item parent report measure of flexibility, the Flexibility Questionnaire (FQ), in children with high functioning ASD.

Methods: Total raw FQ scores were compared in children with ASD (n= 95; 81% male; mean age =10.1 years, SD=1.8; mean full scale IQ=107.0, SD=19.8) and typically developing children (n= 29; 66% male; mean age=10.6 years, SD=1.9; mean full scale IQ=119.2, SD=11.9). Diagnosis in the ASD group was confirmed with the ADI-R and ADOS (ADI Social interaction mean=19.6, SD=5.7; ADI RRBI mean=5.7, SD=2.5; ADOS Communication and social interaction total mean=12.2, SD=5.3). Children with ASD were also assessed using the BRIEF Shift domain (mean T-score=68.0, SD=13.8), the Repetitive Behavior Scale-R (RBSR) compulsive subscale severity raw score (mean=2.8, SD=3.0), and the Social Responsiveness Scale (SRS) Mannerisms subscale (mean T-score=79.8, SD=17.1). 

Results: Chronbach’s alpha statistics indicate good internal consistency for the FQ in both the ASD (0.88) and TD (0.75) groups.  The FQ clearly distinguishes individuals with ASD from the TD group (ASD mean=63.7, SD=20.5; TD mean=17.5, SD=10.1; F=121.9, p<.0001) after controlling for full scale IQ.  The FQ shows convergent validity with the ADI RRBI domain (spearman rho=0.31, p<.005); BRIEF Shift domain (pearson r=.71, p<.0001), SRS Mannerisms Scale (0.59 p<.0001), and RBS-R compulsive subscale (.544, p<.001).  This distinguishes it from the BRIEF Shift scale which is not significantly related to the ADI RRBI domain (spearman rho=.11). 

Conclusions: Preliminary data are promising for the FQ as a continuous parent report measure of cognition, behaviors and symptoms that are related to inflexibility in high functioning ASD.  It has the potential to provide a fine grained and integrated measure of inflexibility in autism.  Future investigations are required to investigate discriminant validity of the FQ with other populations of developmentally disabled children that exhibit RRBIs, assess the relationship of FQ ratings to age and cognitive ability, and investigate the factor structure of the measure.