Association between Subclinical Autistic Traits and Cognitive Functions in General Children: Focusing on Sex Differences

Background: Although sex difference in autism spectrum disorders and children with sub-clinical autistic traits has attracted a variety of research; few studies have addressed sex difference in the associations between AT and cognitive functions.

Objectives: This study aimed to: 1) explore the associations of autistic trait with intelligence profiles and executive functions among general children; and 2) examine whether the associations differed by sex.

Methods: Quantitative autistic traits of 429 general children aged 6-9 years from mainland China were measured using Social Responsiveness Scale. Children were divided into low- and high-autistic-trait (HAT and LAT) groups according to the Chinese cut-off score for ASDs screening. Their intelligence profiles and executive functions were assessed with Wechsler Intelligence Scale for Children-Fourth Edition (WISC-Ⅳ) and Computerized Version of the Wisconsin Card Sorting Test (WCST) respectively. Demographic information was reported by parents. A series of hierarchical linear regression models for each sex separately were performed to examine the associations between autistic trait and cognitive profiles.

Results: Girls were related to a decreased odds ratio of being in the HAT group (OR=0.54, 95% CI: 0.33-0.88). After controlling for demographic covariates, we observed no significant associations between autistic trait and executive functions; children with HAT had significant lower scores than those with LAT in full IQ and three indices of WISC-Ⅳ (β ranged -3.74 to -6.02, all P<0.05), except for Processing Speed Index. These associations persisted in subgroups stratified by sex, except for Working Memory index. We found no sex-differences in the interrelations of autistic trait and relevant intelligence profiles (all P>0.05). However, the associations of HAT with certain intelligence profiles were much more stronger in girls than those found in boys, including full scale IQ (girls: β=-8.61, SE=2.54, P<0.001; boys: β=-4.43, SE=1.75, P=0.012), Verbal Comprehension Index(girls: β=-8.76, SE=3.06, P=0.005; boys: β=-4.81, SE=2.14, P=0.025) and Perceptual Reasoning Index (girls: β=-9.25, SE=2.49, P<0.001; boys: β=-3.44, SE=1.63, P=0.036).

Conclusions: In general population, autistic trait was negatively associated with certain intelligence profiles, both in boys and girls. Additionally, these associations were much stronger in girls than those in boys. These findings further our understanding of sex difference among individuals along the autistic trait continuum and underline that examining sex difference in the social-cognitive domain should not neglect eliminating the confound effects of sex difference in intelligence profiles.