Evaluating Gender Differences in Perceptual Profiles of Individuals with Autism Spectrum Disorder

Background:    Concurrent demonstrations of superior performance on visuo-cognitive, non-social tasks (i.e., Block Design Task (BDT); Shah & Frith, 1993), and decreased performance on some social tasks (i.e., face identification and discrimination; Dawson et al., 2005) are considered as defining perceptual characteristics of Autism Spectrum Disorders (ASD). These dichotomous performances can, at least in part, be explained by a local processing bias in ASD. However, since ASD’s sex ratio is as high as 16 males to 1 female for high-functioning individuals (Rivet & Matson, 2011), a strong argument can therefore be made that the central tenets of perceptual theories based on the aforementioned findings in ASD are based largely on male abilities.

Objectives:    Assess whether ASD-specific perceptual profiles based on non-social and social task performance are equivalent in both males and females with ASD.

Methods:    Participants. Ten females and 10 males with ASD were recruited and matched on age and Full-scale Wechsler IQ. All participants completed the same two tasks. Task 1. A non-social, computerized reversed BDT: a target block design (matrix size of 4, 9, or 16) was presented centrally on a touch-sensitive screen, with four possible probes (one matching the target) presented simultaneously around it. The perceptual coherence of the block designs was also manipulated by varying the number of ‘adjacencies’ of opposite-coloured edges; low-coherence (LC) designs necessitating increased local analysis to be resolved (block-by-block-processing) relative to high-coherence (HC) designs. Task 2.A social, face identity discrimination task (FID): a target face image was presented centrally, with four possible probes (one matching the target) presented simultaneously around it in same view (biasing local analysis) or viewchange (biasing global analysis) orientations. For both tasks, participants were asked to choose the probe matching the target as quickly and accurately as possible. Reaction time (RT) was measured as dependent variable for both tasks. A control, motor task was included to ensure that participants had similar motor reactivity. 

Results:    No gender differences were found on control motor task across conditions. Task 1: A significant main effect of Condition was found (p < .001), with mean RTs significantly shorter for LC trials. No significant main effect of Gender (p = .689), nor a Gender x Condition interaction was found (p = .397). Task 2: A significant main effect of Condition was found (p < .001), with RTs for same view trials significantly shorter than for viewchange trials. No significant Gender x Condition (p = .454) nor main effect of Gender was found (p =.520.)

Conclusions:    Although non-significant, females with ASD tended to respond faster than males on the more global, view-change condition of the FID task. Although previous studies have suggested gender differences for traditional BDT performance (Koyama et al., 2009) for individuals with ASD, and correlations between autistic traits and impairment in face identity in males (but not in females) (Rhodes et al., 2013), the results of this study, at this point, suggest that perceptual profiles based on social and non-social task performance are equivalent in both males and females with ASD.