Preserved Play in Females at High Risk for ASD Across the Range of Symptom Severity

Background: Previous research indicates that the ratio of autism spectrum disorder (ASD) in males to females is 4.3:1 across the full range of intellectual functioning and is even more skewed, 5.5:1, in individuals without comorbid intellectual disability (Fombonne, 2005). Females with similar levels of autism symptomatology as their male counterparts are less likely to receive an ASD diagnosis (Dworzynski et al., 2012). Explanations of this gender disparity include hypotheses about the roles of genetic and biological factors, but their empirical support is inconsistent (Kreiser & White, 2014). Recent work has discussed a potential bias in our conceptualization of ASD as consistent with the symptom profile common in males, while females with ASD may have a different constellation of symptoms. Therefore, further investigation of the bias of the “male autism phenotype” is warranted (Robinson et al, 2013). Symbolic play (i.e. use of imagination to pretend that an object is something else) has previously been implicated in ASD (Lam & Yeung, 2012). However, gender differences in symbolic play in infants and toddlers at high and low risk for ASD have not been sufficiently explored.

Objectives: To examine symbolic play skills in male and female infants and toddlers between 12 and 24 months who are at high risk (HR) and low risk (LR) for developing ASD.

Methods: All children (n= 207, 62.3% males) were administered the Autism Diagnostic Observation Schedule (ADOS-T) by psychologists at 12, 18, and 24 months. ADOS item C2 (Imagination/Creativity) indexed symbolic play; scores range from zero to three, zero indicating well-developed imaginative play skills and three indicating none. HR children were divided into two groups based on their 24-month ADOS calibrated severity score (CSS): HR-Elevated (HR-E; CSS 4-10; n=51) and HR-Non-elevated (HR-NE; CSS 1-3; n=87). LR children all had non-elevated CSS (LR-NE; n=69).

Results: A linear mixed model for item C2 indicated a main effect of ASD severity group (F(2, 565) = 3.457, p = .032), a main effect of age (F(2, 565) = 90.061, p < .001), and a main effect of gender (F(1, 565) = 21.709, p < .001). Pairwise planned contrasts revealed that the LR-NE group had significantly better play skills than the HR-E group (p = .009); symbolic play skills improved over time (12-18mo, 18-24mo, 12-24mo) (ps < .01), and females displayed more advanced symbolic play than males (p< .001).

Conclusions: This study is unique in longitudinally investigating play in infants as young as 12 months and found that females across all ages with both high and low ASD symptom severity showed significantly stronger symbolic play skills than males. Stronger play abilities in females, regardless of ASD symptomatology, may be a factor in the clinical presentation of females with increased autism symptoms. In other words, play skills appear to be relatively preserved in females with elevated ASD symptoms. Future studies should examine the possibility that intact play skills are part of a female autism phenotype.