We are delighted to share that Dr Samuel Liebana, Research Fellow at the Gatsby Computational Neuroscience Unit working with Professor Andrew Saxe, has received a Wellcome Early-Career Award. This prestigious award recognises Dr Liebana’s innovative research on how brains – both natural and artificial – learn. The award will support his research at the Gatsby Unit and the Sainsbury Wellcome Centre, where the experimental side of the work will be carried out in collaboration with Professor Jeffrey Erlich.

“As an emerging scientist, the Wellcome Early Career Award comes as an empowering gift that opens the doors to independent research. I am honoured to be an awardee this year and deeply thank all of the students, postdocs and professors across Gatsby/SWC that have helped me achieve this,” commented Dr Liebana. 

How the brain learns remains much of a mystery. Sometimes we master a new skill quickly, while other things take a long time to learn. Individuals vary in how they learn, differing in the trajectories they take to expertise, their relative speed, and the solution strategies they develop. Can we understand the source of this variability, predict where learners will struggle, and design curricula that speed up learning? 

Dr Liebana's research sits at the intersection of these questions, combining experimental and theoretical approaches. During his PhD, he investigated the role of dopamine in long-term learning, showing how it acts as a teaching signal that shapes diverse yet systematic learning trajectories. He also showed how long-term learning mechanisms can be understood theoretically, using artificial neural networks. 

Dr Samuel Liebana

“Remarkably, the advent of deep learning has revealed that artificial neural networks (ANNs) can flexibly learn complex tasks and share many similarities with the brain. As a growing community of theorists uncover the principles governing learning in ANNs, this presents a unique opportunity to investigate whether similar principles underlie learning in the brain,” explains Dr Liebana.

The Early-Career Award will allow Dr Liebana to test emerging theories of deep learning experimentally, to see if they hold true in the brain. To do this, he has developed behavioural tasks for rodents that probe key learning phenomena: long-term learning through stages, the transfer or forgetting of previous knowledge, and the learning of non-linear computations. This battery of tasks will test whether ANNs can help explain both the behavioural trajectories and neural mechanisms underlying learning over long timescales and multiple tasks.

The Wellcome Early-Career Award, starting in January 2027, will provide Dr Liebana with funding to support his salary and research expenses for five years. 

“This work has the potential to transform our understanding of learning in the brain by constructing a formal framework to describe learning across neural circuits. Especially in our modern world of screens and machines, much of what we do is learned over long periods of time, so understanding this process could help us understand many aspects of behaviour”, commented Dr Andrew Saxe, Group Leader at SWC and the Gatsby Unit.

To find out more about Dr Liebana and his research, visit his profile.

About Wellcome Early-Career Awards

Wellcome Early-Career Awards provide funding for early-career researchers from any discipline who are ready to develop their research identity. Through innovative projects, they will deliver shifts in understanding related to human life, health and wellbeing. By the end of the award, they will be ready to lead their own independent research programme. For further information, please visit the Wellcome website: Wellcome Early-Career Awards.

About Dr Liebana

Samuel is a Research Fellow in Theoretical Neuroscience working with Andrew Saxe. His work lies at the intersection of theoretical and experimental neuroscience, aiming to uncover the computational and biological principles of individual/collective learning in the brain.

He completed his PhD in Neuroscience at DPAG, University of Oxford, supervised by Armin Lak, Andrew Saxe, and Rafal Bogacz. During his PhD, he studied the computational role of dopamine in long-term learning.