Funded Grants

Embracing complexity in neurodevelopment

Between 14-30% of children and adolescents experience barriers to learning, like struggling to maintain attention. Many of these young people receive a diagnosis, such as autism, ADHD and dyslexia. These diagnostic categories have provided the lens through which neurodevelopmental differences are viewed. Our research, theories and interventions are all organised around the assumption that the diagnostic framework reflects underlying reality – that each diagnostic category is associated with a specific pattern of learning difficulties, caused by a distinctive neurocognitive foundation. As a result, researchers have spent decades searching for causal, ‘core’ deficits for each diagnostic group. The hope is that unlocking an underlying mechanism will lead to downstream benefits in learning and life chances. This quest has been largely unsuccessful. Purportedly ‘core’ deficits are remarkably inconsistent, interventions targeting them do not yield cascading benefits, and supposedly ‘unique’ challenges are frequently found across multiple diagnostic groups, and in children with no diagnosis.

The fundamental problem is the assumption of diagnostic ground truth is not valid. In fact, any system that assigns young people to discrete categories fails to capture the complexity of development. Moreover, this assumption has hampered our efforts to understand underlying brain and cognitive mechanisms, because it has shifted our attention towards explaining diagnosis and away from understanding the common barriers that young people face in everyday life.

The current proposal marks a radical departure from the assumption of diagnostic ground truth. We will use a JSMF Opportunity Award as a springboard to a new theoretical framework, that embraces complexity in neurodevelopment. By applying innovative data-driven and complex modelling techniques we will make new discoveries about the nature of, and barriers to, learning. Two linked projects will build a hierarchical framework in which surface level characteristics (e.g. language difficulties) can have diverse origins at multiple levels (genetic, neurological, cognitive). The first project uses existing large-scale datasets, in tandem with recent advances in data modelling, to map characteristics across these levels. The second project picks up on these mappings and builds a new theoretical model using generative network modelling. To date, most theories of neurodevelopment are descriptive accounts of groups of individuals, weaving together average characteristics with a narrative theory of their shared origins. In contrast, the second project will bring about a step-change in theoretical approach by building generative models of neurodevelopment. In simulating the growth of a connectome, this approach can capture complexity, model developmental processes, and link multiple datatypes (e.g. incorporating genetics).

Our long-term goal is not only to make discoveries to benefit the lives of young people experiencing barriers to learning, but to bring about a step-change in approach within the wider field. The work afforded by a JSMF Opportunity Award will act as a beacon for research in neurodevelopment. In addition to the programme of science, we have planned community-facing outputs, an international conference, a series of methods tutorials, and we propose to co-edit a special issue in major outlet in the field.