Preservation of Higher-Order Cognition in Autistic Females: Evidence from Two Deeply-Phenotyped Cohorts in Hong Kong and the United States
Listed in
This article is not in any list yet, why not save it to one of your lists.Abstract
Background Sex differences in the clinical presentation of autism spectrum disorder (ASD) are well documented. Although prior studies have linked neurocognitive differences to these phenotypic variations, it remains unclear whether such differences are specific to ASD or attributable to co-occurring psychopathology and intrinsic sex differences in the general population. Methods This study evaluated multi-domain cognition with the Penn Computerized Neurocognitive Battery and clinical assessments in 1,101 Chinese children in Hong Kong (701 with ASD, 400 without ASD; aged 6–12 years), with concurrent analysis in an independent US cohort of 1,090 participants (218 with ASD, 872 without ASD; aged 8–21 years with diverse ethnicity). Group differences were analysed using MANCOVA followed by univariate ANCOVA, with rigorous adjustment for demographics and co-occurring psychopathology. Results Across both cohorts, a consistent Group × Sex interaction was observed specifically in the cognitive domain Complex Reasoning and Executive Function (CREF), with boys with ASD showing the greatest impairment, while girls with ASD performed comparably to non-ASD girls. Exploratory analyses in the Hong Kong cohort, stratifying ASD into ASD-only and ASD+ADHD subgroups, revealed a stepwise significant decline in CREF among boys (non-ASD > ASD-only > ASD+ADHD). In girls, CREF was largely preserved in the ASD-only group relative to non-ASD controls, while the ASD+ADHD subgroup showed lower performance that did not reach statistical significance after correction for multiple comparisons. Conclusions These findings indicate that relatively preserved higher-order cognition in autistic females may help explain subtler clinical presentations observed in some females with ASD. Given evidence linking higher-order cognition to compensatory processes, our results have implications for improving clinical detection of autistic females, addressing their specific needs, and understanding biological mechanisms underlying sex differences in ASD.