Attention deficit Hyperactivity disorder (ADHD) is a neuro-developmental disorder whose origins remain not well understood. Popular theories of ADHD assume that ADHD symptoms are caused by the dysfunction of specific neuro-cognitive mechanisms like the ability to represent and learn from positive and negative feedback (also called prediction errors) or the ability to hold information for further manipulation in the working memory. These neuro-cognitive functions depend on the availability of neurotransmitters (signalling molecules), in particular dopamine, in the brain. Yet, even though there is ample evidence for genetic influence on neurotransmitter availability, and twin studies indicate the heritability of ADHD, the specific genetic contributions to ADHD remain poorly understood.
This project shall use a “computational psychiatry” approach to improve the understanding the specific genetic contributions to ADHD. Computational psychiatry uses mathematical models of cognition and learning that are consistent with behavioural and neurobiological data to identify which neuro-cognitive sub-processes cause impaired learning or cognition. For example, impaired reward-based learning could be explained by the inability to learn from positive feedback or feedback, or to the inability to make choices consistent with learned reward-values. Low performance in cognitive tests could be caused by a low ability (resulting e.g. from noisy working memory representations), or by a preference for fast as opposed to accurate responding.
Because the mathematical models employed in this project have been shown to be good models of the cognitive and neuro-biological processes underlying learning and decision making, they can build a bridge between ADHD symptoms and genes by showing how neuro-cognitive deficits underlying ADHD are in turn associated with genes. We expect stronger associations between neuro-cognitive processes and genes than between ADHD-Symptoms and genes, because the former are likely more directly influence by genes.
Hence, our goals are to describe ADHD phenotypes in terms of neuro-cognitive sub processes obtained through mathematical modelling and to identify genes associated with these phenotypes.
For this aim we will model data from neuropsychological and experimental tests collected in the ALSPAC Teen focus data collections to first identify in which neuro-cognitive processes ADHD patients have the most important deficits. In a second step, we will investigate genetic associations for the neuro-cognitive processes with the largest deficits.

In sum, this project will use mathematical modelling of task performance to identify neuro-cognitive processes that mediate the influence of genes on ADHD symptoms. Modelling neuro-cognitive functioning will improve the understanding of ADHD and contribute to the definition of phenotypes that facilitate the detection of genetic contributions to ADHD.