Cognitive functioning and Anxiety/Depression: Accounting for Genetic Confounding in the ALSPAC Cohort.

Aims: Depression and anxiety are leading contributors to the global burden of disease, yet the mechanisms underlying these conditions, including the role of cognitive processes, remain uncertain. Cognitive deficits in domains such as emotion recognition, working memory, and response inhibition are commonly observed in individuals with depression and anxiety, but it is unclear whether these reflect causal risk factors, consequences of symptoms, or shared aetiological influences. We investigated whether associations between cognitive function and mental health outcomes persist after accounting for genetic confounding using data from the Avon Longitudinal Study of Parents and Children (ALSPAC). Methods: Participants completed computer-based assessments of emotion recognition, working memory, and response inhibition at age 24. Depression and anxiety diagnoses were derived at age 24 using the Clinical Interview Schedule - Revised. Logistic regression models examined associations between cognitive function and each mental health outcome, followed by models adjusting for sociodemographic confounders, cognitive confounders, and polygenic scores (PGS) for depression and anxiety. Genetic sensitivity analyses (Gsens) were conducted to estimate the extent to which associations were attributable to shared genetic liability. Results: Higher working memory performance was consistently associated with lower odds of anxiety, and this association was robust to adjustment for observed confounders and PGS, with Gsens analyses indicating no evidence of genetic confounding. Evidence for an association between working memory and depression was weak and disappeared after accounting for genetic confounding. Emotion recognition showed heterogeneous patterns: better sadness recognition was associated with higher risk of depression, while weaker happiness recognition showed preliminary associations with depression but these were fully explained by shared genetic influences. For anxiety, genetic sensitivity analyses suggested negative (masking) genetic confounding for some emotion recognition domains, such that associations became stronger after accounting for genetic liability. Slower response inhibition showed weak associations with depression, which Gsens analyses suggested were entirely attributable to genetic confounding. Conclusions: Overall, these findings indicate that the relationship between working memory and anxiety is unlikely to be explained by shared genetic liability, supporting its plausibility as an intervention target. In contrast, several associations involving emotion recognition and response inhibition with depression/ anxiety appear to reflect shared genetic origins.