Emerging Executive Functions: an investigation of latent structure in toddlerhood and prediction from prenatal stress exposure, sex and early maternal caregiving behaviours.

The study of Executive function (EF) in early child development is not fully established, with much still to be confirmed regarding its latent structure and early precursors. Proposed mechanisms underlying reported inverse associations between prenatal stress (PNS) and general cognitive development suggest that brain regions associated with early EF may be specifically vulnerable, yet this is a largely unstudied area. Postnatal caregiving may modify PNS effects on EF providing a possible early intervention target to improve developmental outcomes. This thesis comprises three linked studies. Study 1 examined latent EF structure in toddlers. Study 2 investigated associations between different forms of PNS in each trimester of pregnancy and toddler EF, and tested whether they were sex-dependent. Study 3 tested the moderating influence of early maternal caregiving on associations observed in study 2. Methods: This research was embedded within the prospective longitudinal Wirral Child Health and Development Study. Study 1: 254 toddlers (mean age 31.4m) completed a battery of EF tasks. Study 2: Mothers reported anxiety symptoms (STAI) at 20 weeks’ and 32 weeks’ gestation, and number of stressful life events (SLEs) via an investigator-led interview for each pregnancy trimester and at multiple postnatal timepoints. Toddler latent EF abilities derived in study 1 were the cognitive outcome. Study 3: Mothers reported frequency of stroking their baby at 4 and 9-12 weeks old, yielding a tactile stimulation index of early caregiving. At 6 months, maternal sensitivity was observed and rated during free play. Results: Study 1: CFA applied to EF data yielded 2 distinguishable yet moderately correlated (r = .43) factors:- working memory (WM) and inhibitory control (IC). Study 2: Multiple linear regression models revealed that 32-week STAI scores in interaction with sex predicted toddler WM (β =0 .55; p < .05), accounting for 2% of the variance in scores after accounting for pre and postnatal confounders; and 1st trimester SLEs in interaction with sex predicted IC (β = 0.45; p = .05), accounting for 2% in outcome. Both interactions arose from a similar pattern of opposite associations between PNS and EF in males and females. Study 3: PNS effects were not moderated by maternal tactile stimulation. The prediction of toddler IC by the 1st trimester SLE by sex interaction was moderated by maternal sensitivity. This 3-way interaction accounted for 3% of variance after accounting for confounders. Conclusions: EF exhibits an integrative latent structure in toddlerhood; PNS was associated with poorer EF in males and relatively enhanced EF in females; exposure to higher maternal sensitivity in the first months of life eliminated the observed sex-dependent PNS effects on toddler IC but not WM. Findings are discussed in the context of Developmental Origins of Health and Disease model and evolutionary perspectives on sex-dependent development