Types of Restricted Interests and Preoccupations: Are There Differences across Sex?

Background:

It is unclear if females are less likely to be diagnosed with autism spectrum disorder (ASD) because of a protective, female effect or the result of a unique female autism phenotype that current diagnostic tools and criteria do not detect (Van Wijngaarden-Cremers et al., 2014). With regard to restricted, repetitive behaviors (RRBs), parents and clinicians anecdotally report differences between males and females (Jamison et al., 2017). However, research findings are inconsistent (e.g., Van Wijngaarden-Cremers). Research suggests that there are differences in total number of RRBs, but this depends on the particular measure used and differences may only be apparent at the symptom level (e.g., Frazier et al., 2014). Given the inconsistent research results, researchers have proposed that difficulties in detection of RRBs is a result of females having more socially appropriate and difficult to detect RRBs (Halladay et al., 2015).

Objectives:

The aim of the current study was to investigate whether differences in types of restricted, circumscribed interests or preoccupations are associated with sex (i.e., male vs female), particularly restricted, circumscribed interests and preoccupations (CIs).

Methods:

Multivariate analyses were completed for the entire Simons Simple Collection (SSC; N=2648) to replicate previous findings (see Frazier et al., 2014). Qualitative data were collected from a local sub-sample of caregivers at a North American, SSC site (N=186). Qualitative phenomenological investigations are suitable when interested in a discovery-oriented approach of how individuals construct meaning of an experience (Creswell et al., 2007). In this particular case, qualitative methodology is appropriate to gain an in-depth understanding of how caregivers perceive and describe CIs. Responses describing restricted interests and unusual preoccupations on the Autism Diagnostic Interview – Revised (ADI-R) (i.e., questions 67 and 68) were transcribed and categorized using qualitative methodology. Categories were sorted by sex (females=30) and proportions of individuals with CIs in each category were calculated. Additional analyses will be completed to test for significant differences in obtained proportions and to investigate how differences in CI across gender might be associated with other phenotypic measures.

Results:

Multivariate analysis from the entire SSC was significant F(10,2648), p < .00, partial ɳ² = .40) with significant differences evident at the RBS-R Restricted Interest subscale level (p <.00) and ADI-R Encompassing preoccupation/circumscribed interest subscale level (p < .00). Males had more symptoms (i.e., higher scores) in both instances. Analysis of qualitative data found similar proportions of males and females with intense interests in television/movies; animals; and, food. Males were proportionally more likely to have CIs in categories such as factual information (13% vs 0%, respectively). Females had more CIs in several categories including arts and crafts (16.7% vs 5.77%),

Conclusions:

Preliminary analysis indicates that there are differences in the types of CIs described by parents that appear to be associated with sex. As suggested in previous research, it is plausible that females demonstrate socially acceptable restricted interests. Further, parent completed measures may not be capture socially acceptable circumscribed interests. Alternatively, there may be real differences between males and females in the number of restricted interests.