25298
Examining Cognitive Inflexibility and Anxiety in Relation to Restricted, Repetitive Behaviours in Autism Spectrum Disorders in a Large Multi-Site Study
Restricted and repetitive behaviours (RRBs) are core features of autism spectrum disorder (ASD). Previous studies have suggested distinct subtypes of RRBs, such as ‘lower-order’ (e.g. motor stereotypies) and ‘higher-order’ RRBs (e.g. restricted interests and insistence on sameness; Turner, 1999; Rutter et al., 2003). Despite various descriptions of RRBs, their origins are less well understood. This may be because: (1) different facets of RRBs have different underpinnings; or (2) the same facet has different underpinnings in different individuals. As an example: for some, cognitive inflexibility may underlie insistence on sameness (D’Cruz et al., 2013), whereas for others, anxiety in uncertain situations may be a factor (Lidstone et al., 2014).
Objectives:
(1) To examine the profile and severity of RRBs in children, adolescents and adults who participated in the EU-AIMS Longitudinal European Autism Project (LEAP); and (2) to investigate cognitive inflexibility and anxiety as possible underpinnings of RRBs.
Methods:
Participants were 431 individuals with ASD aged 6-30 years and 298 age-matched controls. RRBs were measured using the Repetitive Behaviour Scale-Revised (RBS-R). The RBS-R restricted interests and rituals/sameness subscales formed the ‘higher-order’ RRBs composite and the stereotyped and self-injurious behaviour subscales formed the ‘lower-order’ composite. IQ was measured using the Wechsler Scales and adaptive behaviour using the Vineland Scales. Cognitive inflexibility was assessed as the number of perseverative errors made on a computer-based probabilistic reversal learning (PRL) task (den Ouden et al., 2013). We classified perseverative errors made by ASD individuals using a normative modelling approach, which assessed each ASD participant’s performance as the deviation from the age-estimated control group mean. Anxiety was measured using the Beck’s Anxiety Inventory.
Results:
As expected, ASD participants scored significantly higher than controls on the RBS-R (p<.001, d=2.06). Within the ASD group: (1) participants had significantly greater ‘higher-order’ RRBs than ‘lower-order’ RRBs (p<.001, d=0.86; Figure 1); (2) RBS-R total scores significantly decreased with age (r=-.25, p<.001); and (3) RBS-R total scores were significantly negatively related to adaptive behaviour, independent of IQ (r=-.31, p<.001). On the PRL task, the ASD group made significantly more perseverative errors than the control group (p<.001, d=0.35). However, when stratifying ASD individuals (Figure 2), 62% performed within +/-1SD of the control group mean, and 91% within +/-2SDs. Only 9% performed below 2SDs. No significant relationship was found between perseverative errors and RRBs in the ASD group, nor did the 9% with extremely high perseveration differ in their RRBs profiles. Across age groups, RRBs – and in children notably rituals/sameness behaviours – were significantly related to levels of anxiety in ASD individuals without intellectual disability (all r’s>.41,p’s<.001).
Conclusions:
We found no evidence for a role of perseveration in RRBs in ASD. Despite significant case-control differences, only a small ‘subgroup’ of ASD participants demonstrated clinically meaningful (<2SDs) impairments in cognitive inflexibility; and they were unrelated to RRBs. Instead our findings provide proof of concept that anxiety may be a main driver of RRBs, and particularly insistence on sameness. If correct, this suggests an alternative treatment approach in ASD – targeting anxiety to reduce RBBs.