Social Attention in ASD Females: The Same, Yet Different

Background: Differences in attention to social stimuli are a hallmark feature of ASD and have been extensively studied using eye-tracking. Chevallier et al. (2015) found that some stimuli are better at capturing social attentional differences in children with ASD than others, with dynamic videos generating the largest group differences.

Due to a sex imbalance in ASD diagnoses, very few studies have used eye-tracking to examine sex differences in social attention. Emerging literature suggests that girls with ASD socialize differently than boys (Dean et al., 2017) and experience heightened social motivation (Sedgewick et al., 2016), mirroring some of the differences found in typically developing girls and boys. Therefore, a dynamic eye-tracking paradigm may reveal differences in social attention and motivation in girls with ASD.

Objectives: The purpose of this study was to examine whether behavioral and clinical differences in ASD males and females were mirrored in their distribution of attention to social, dynamic stimuli – particularly faces – using a validated eye-tracking paradigm.

Methods: 65 children aged 6 to 10 years were included in the study [37 ASD (16 female); 28 typically developing (TDC; 14 female). Children completed a Visual Exploration eye-tracking paradigm that included 22 silent video clips of 11 sibling pairs of school-aged children playing together (Joint condition) or in parallel (Parallel condition; Chevallier et al., 2015). Our primary dependent variable of interest was the proportion of gaze duration to faces relative to gaze duration toward the full screen, overall and by condition. We also compared the proportion of gaze duration to hands playing with objects, background objects, and the “space between” people and objects, overall and by condition.

Results: Diagnosis (ASD vs. TD) and sex (Male vs. Female) predicted a number of indices of social attention. Children with ASD spent proportionally less time attending to faces overall and across conditions (Joint and Parallel; all ps <0.02, ƞ²_p .08 to .17). However, sex also differentiated fixation to faces overall and in the joint play condition (ps <.05, ƞ²_p .06 & .08) with females allocating proportionally more attention to faces than males. Within the ASD group, ASD females spent proportionally more time attending to faces (t= -2.24, p =.03), particularly in the parallel play condition (t=-2.51, p =.01). Diagnostic groups looked equally at background objects and hands with toys, but the ASD group gazed longer at the “space between” AOIs than the TD group (ps<.01, ƞ²_p .07 to .15; no effect sex or interaction).

Conclusions: Using a validated, ecologically valid, eye-tracking paradigm we found both similarities and differences between ASD males and females. Our data reveal that while ASD females demonstrate enhanced social attention compared to boys with ASD, their overall pattern of attention, particularly to areas of the scene not portraying social or non-social information (“space between”), was comparable to their male counterparts. This disjointed pattern of attention aligns with clinical and behavioral reports of ASD females, with reported heightened social motivation and desire for friendships, yet similar rates of actual friendships and social success to ASD males.