| dc.description.abstract | Background: The three core executive functions – inhibition, working memory, and cognitive flexibility – are employed in higher-level cognitive processes such as problem-solving. Executive functions undergo substantial development during adolescence and decrements play a key role in many psychiatric and behavioral disorders. There is some evidence that prenatal exposure to individual chemicals may adversely impact executive functions among children, but few studies have explored the association of co-exposure to multiple chemicals with executive function in adolescence, a time when decrements may become discernable. In addition, few studies have investigated the association of prenatal chemical exposures with higher-level executive functions or potential mediators of this association. Determining the neuropsychological mechanisms that are most sensitive to chemical-related impacts and characterizing the relationships between core and higher-level executive functions are both essential to understanding the relation of chemical exposures with executive function. Therefore, the purpose of this dissertation was to identify modifiable environmental risk factors that may impact executive functions and to determine how the relationships between the core and higher-level executive functions may play a role in these relationships. We did this by first investigating the association of prenatal exposure to chemical mixtures composed of prevalent and neurotoxic organochlorines and metals with the core executive functions. We then investigated the association of prenatal exposure to manganese (Mn) with problem-solving among adolescents and assessed the extent to which the core executive functions mediate this association.
Methods: Among a diverse group of 373 adolescents living near the New Bedford Harbor Superfund site in Massachusetts, we investigated the association of dichlorodiphenyldichloroethylene (DDE), hexachlorobenzene (HCB), polychlorinated biphenyls (PCBs), lead (Pb), and Mn with inhibition, working memory, and cognitive flexibility measured with subtests of the Delis-Kaplan Executive Function System (D-KEFS) and Wide Range Assessment of Memory and Learning, 2nd Edition (WRAML2). We did this by using Bayesian kernel machine regression (BKMR) as an exploratory tool to inform a traditional multivariable regression approach. We also assessed effect modification by sex and social disadvantage as well as the association of a secondary chemical mixture that included methylmercury (MeHg) and arsenic (As) in addition to the aforementioned chemicals with executive function. As we observed Mn to be consistently adversely associated with the core executive functions, we focused on Mn in our subsequent analyses. We used multivariable linear regression to assess the relation of prenatal Mn exposure with problem-solving skills followed by causal mediation methods to investigate whether the core executive functions may be potential mediators of this association.
Results: In BKMR models, we observed adverse joint associations of the chemical mixture with primarily verbal, but not non-verbal, measures of inhibition, working memory, and cognitive flexibility. In co-exposure and covariate-adjusted regression models, a twofold increase in cord blood Mn was associated with worse performance on a verbal inhibition task and multiple working memory and cognitive flexibility tasks. There were no other consistently adverse associations between biomarkers of prenatal exposure to the other targeted chemicals and executive functions. There was also little evidence of effect modification by sex but some evidence of effect modification by social disadvantage in associations of organochlorine chemicals with working memory and cognitive flexibility. Cord blood Mn was also associated with lower problem-solving scores. In a subsequent mediation analysis, the combination of inhibition, working memory, and cognitive flexibility mediated 44% of the total effect of prenatal Mn on problem-solving. When each mediator was analyzed individually, working memory mediated a larger proportion of the effect than inhibition or cognitive flexibility.
Conclusions: This dissertation provided evidence of an adverse joint association of a chemical mixture with core executive functions, particularly when assessed with verbal tasks, and implicated Mn as more adverse than other chemicals in the mixture. It was also among the first to find evidence of an adverse association of prenatal exposure to Mn with a higher-level executive function and to identify certain neuropsychological mechanisms as mediators on the pathway of this association. | |
