Gene-environment correlations and interactions in cognitive development

Cognitive development is influenced by both genetic and environmental factors, which may have additive effects, but which may also interact in non-additive ways, and which furthermore are sometimes partly confounded. Yet, most studies of environmental effects on cognitive development have not taken into account the potential correlations and interactions with genetic factors, potentially yielding inaccurate estimates of environmental effects.

This situation has changed with genome-wide association studies (GWAS) and the now widespread availability of polygenic scores (PGS) for various phenotypes in a number of developmental cohorts. The present project aims to leverage these new sources of information to revisit some classic questions and investigate new questions on the etiology cognitive development and neurodevelopmental disorders.

As an example, an environmental factor that has a well-established effect on reading ability and on the risk of dyslexia is the so-called Home literacy environment. This composite measure typically aggregates information such as the number of books owned in the household, as well as parents’ literacy practices (in particular, reading books to their children). Yet, parents who own many books and who spontaneously read stories to their children are not the same kind of parents as those who don’t: they typically are literate, highly-educated people who may have good genetic predispositions for literacy and who may have transmitted these predispositions to their children. In other words, the environmental effect is potentially confounded by a genetic effect. The project will therefore aim to use a polygenic score for reading ability/dyslexia constructed based on previous GWAS as an additional regressor in statistical models in order to disentangle genetic from environmental effects, and also to test the interaction between the PGS and the home literacy environment.

Analyses will be based on cohorts that include genetic data and the relevant environmental and phenotypic data, such as ALSPAC, TEDs, or ABCD. These data will be acquired before the start of the internship.

This project requires very strong statistical skills and a good mastery of the R language.