Sex differences in social attention in autism spectrum disorder.

The final analysed group was 77 children, so several effects were small and some were only trends. Two boys with autism did not complete the task, and only children who looked at the screen for at least 20% of the time were analysed. The typically developing group was younger than the autism group; age was controlled, but the authors said results may have differed if age was more tightly matched. The task used static faces and generic interest images, not dynamic social scenes or each child's own interests. The interest images were based on earlier reports from mostly male samples, so they may not have been salient for girls. Diagnosis came from local clinical records that varied in timing and tools, and the researchers used a parent questionnaire to distinguish groups, which they said may lose validity in older children. The study cannot tell whether girls' face looking was true social interest or learned camouflage.

The supplied metadata lists NIH extramural and non-U.S. government research support. No author conflict declaration is provided in the text.

Do not conclude that face-attention goals should be dropped for girls with autism. This was an eye-tracking task, not a treatment study, and it cannot tell whether the girls' face looking was true interest or learned camouflage.