IQ As a Moderator of the Presence, Severity, and Fluctuation Rates of 3 Individual Rrbs: Bouncing, Lining up Objects, and Aversion to Loud Noises

Background:  The presence of restrictive and repetitive behaviors (RRBs) is required for a diagnosis of Autism Spectrum Disorders (ASD) and can have a devastating impact on children and their families (Richler, Heurta, Bishop, & Lord, 2010). Previous researchers observed that IQ moderated the presence of groups of RRBs (insistence on sameness and repetitive sensory motor, Bishop et al, 2006 Bishop et. al 2013). There continues, however, to be a lack of research assessing the characteristics of individual RRBs to help clarify ASD and other syndrome profiles.

Objectives:  The current study examined the moderating relationship between NVIQ and VIQ and the presence, severity, and fluctuation of three prevalent, individual RRBs.

Methods:  A sample of 458 children were used in analysis. All participants had a research confirmed ASD using ADOS and ADI; Comparison Severity Score (CSS) was used from the ADOS. IQ was assessed using standardized cognitive measures. The prevalence of individual RRBs was analyzed in a previous study that used the Behavior and Sensory Interest Questionnaire (BSIQ, Hanson et. al. 2015). This analysis used three very common RRBs: Bouncing, Lining up Objects and Aversion to Loud Noises (all found in > 50% of the population). We fit logistic regression models to determine the moderating relationship of IQ on presenting RRBs using two interaction variables: a) CSS and VIQ; b) CSS and NVIQ. RRB presence (0-1), severity (0-3) and fluctuation (0-1) were included in separate models as dependent variables. ANOVAs were run to assess the differences between groups. All statistics were run using SPSS.

Results:  None of our regression models reached significance. NVIQ, VIQ, and CSS were not associated with a significant increase or decrease in the presence, severity, or fluctuation rates of Bouncing, Lining up Objects, or Aversion to Loud Sounds. While our ANOVAs and T-tests were also not significant, they revealed trends between NVIQ and VIQ and our behaviors. Mean NVIQ decreased approximately ten points between individuals who did not endorse Bouncing and those who did; there was no trend between VIQ and Bouncing. In contrast, mean VIQ decreased 5 points between individuals who did not endorse Lining up objects and those who did; there was no trend between NVIQ and Lining up objects. Finally, mean NVIQ decreased 4 points and mean VIQ decreased 7 points between individuals who did not endorse Aversion to loud noises and those who did.

Conclusions:  Our results indicate that continuing to analyze RRBs on an individual basis reveals important information about distinct RRB characteristics that studying them on a grouped basis alone does not, especially in terms of the moderating effect of IQ. While none of our regression models or ANOVAs reached significance, the models assessing the fluctuation rates were the closest. This suggests to us that NVIQ and VIQ may be contributing factors of behavior change over time. Research examining the moderating role of IQ on all 74 RRBs of the BSIQ is in progress.