Towards informative translation from basic neuroscience to mental health

Reinforcement learning as a bridge between rodent and human flexible behaviour.

Multiple forms of early life stress (ELS), such as social isolation, food scarcity, and violence, significantly increase lifetime risk of developing a mental illness. Rodent models - where ELS can be introduced and controlled experimentally - show that these experiences change specific brain circuits in an extensive and long-lasting manner. In parallel, human research reveals that ELS profoundly impairs flexible, goaldirected behaviour in adulthood, such that individuals struggle to appropriately adapt their behaviours to changing environmental demands. Together, these findings suggest that alterations to neural circuits associated with flexible behaviour following ELS may be a key biomarker for the risk of mental illness.

However, testing this hypothesis is currently hindered by a key issue. Rodent models, while excellent for controlled investigation of risk factors and their influence on neural circuitry, often use simple behaviours with poor validity to measure the consequences of such alterations, leading to highly variable results with limited translatability. 

By combining rodent models, appropriate behavioural tasks and newly-developed computational models, this project will describe animal behaviour using the same framework as used in humans. In this project, the team will combine these two advances to produce a pilot study that asks three important questions:

• How does rodent behaviour in the ID/EDS task compare to human performance?
• How does ELS influence task performance?
• Are the computations underlying task performance altered by ELS?

This project will generate pilot data using an approach that links rodent systems neuroscience to behavioural tests performed by practicing clinicians in a meaningful way using computational psychiatry.

By developing much needed computational models that provide mechanistic insights into how the brain computes behavioural flexibility and how early life stress disrupts these computations, accelerating the development of interventions that could rescue behavioural deficits and ultimately improving mental health outcomes for individuals that have experienced early life stress.