Examining the association between cognitive ability and emotional problems across childhood using a genetically informative design: could there be a causal relationship?

Abstract

BACKGROUND Emotional problems co-occur with difficulties in verbal and nonverbal cognitive ability, yet the pathways underlying their association remain poorly understood: It is unclear whether effects may be causal, and to what extent they may run from cognition to emotion, or vice versa. METHODS Our preregistered analyses included 5,124 twin pairs from the Twins Early Development Study (TEDS). At ages 7, 9 and 12, emotional problems were assessed through the strengths and difficulties questionnaire, and cognition was assessed using task-based measures. Cross-lagged models examined the influence of cognition and subdomains of verbal and nonverbal abilities on emotional problems and vice versa, across development. Genetic cross-lagged models examined the effect of cognition on emotional problems and vice versa, after controlling for shared genetic and environmental influence. RESULTS Cross-lagged paths in both directions were observed between cognitive ability and emotional problems (from -0.11 to -0.05). Cross-lagged associations that persisted after accounting for common genetic and environmental influences were between nonverbal ability and emotional problems. Higher emotional problems at age 7 predicted lower nonverbal ability at age 9, with 22% of the phenotypic association remaining. This, in turn, predicted greater emotional problems at age 12, with 13% of the association remaining. CONCLUSIONS Genetic and environmental factors accounted for a large proportion of the cross-lagged associations. Emotional problems in early childhood could result in a cascade effect, leading to lower nonverbal cognition in middle childhood, which increases the risk of emotional problems in late childhood. These findings highlight the importance of age- and domain-specific interventions.