Interview with Dr. Sami Barrit—Winner of the Brain Sciences Best Paper Award

We are honored to announce that Dr. Sami Barrit has been selected as the winner of the Brain Sciences Best Paper Award 2025.

The following is an interview with Dr. Sami Barrit:

1. Can you please provide us with a short self-introduction?
I'm a young neurosurgeon born and raised in Brussels, specializing in stereotactic and functional neurosurgery with a particular focus on epilepsy surgery. I trained across several countries—France, Japan, and now the US—and have recently joined the Computational Precision Health program at UC Berkeley and UCSF. My research interest lies at the intersection of human and artificial intelligence, from clinical decision support to brain–computer interfaces.

2. Congratulations on your published paper and the award it has received! Could you briefly introduce your research focus and the key findings of the award-winning paper? What inspired this work?
In this study, we showed that a specialized, well-constrained LLM leveraging a curated neurology corpus outperformed practicing neurologists in a blinded evaluation of complex diagnostic cases with no hallucinations detected (that was still a major thing, back in the not-so-long-ago days). The project was motivated by a simple question: Can we build AI that is not just impressive in general, but genuinely trustworthy in the kind of difficult reasoning neurologists deal with every day?

3. What were the biggest challenges you faced during this research, and how did you overcome them?
Devising a paradigm rigorous and fair enough to be meaningful was the first challenge — but honestly, the hardest part was, as very busy residents, coordinating very busy clinicians across institutions to participate and be evaluated under proper blinded conditions. That only worked thanks to an efficient and trusting collaboration with my co-author and friend (and smartest neurologist I know), Nathan Torcida, MD, and a shared commitment to getting it right rather than getting it done fast.

4. What are the potential applications or long-term impacts of your award-winning research? How might it benefit industry or society?
Beyond the now widely debated benefits and risks of AI-driven clinical support, what I find most humbling in these findings is what they suggest about differential diagnosis itself—long considered the pinnacle of medical intellect—and how AI challenges us to rethink it. Not as a threat, but as an invitation to develop new practices of co-intelligence that, well-mastered, I believe represent a genuine opportunity for both better care and broader access to it.

5. Based on your experience, what advice would you give to early career researchers working in neurosciences?
Trust your gut and know that effort pays. Find your natural interest, your method, and—most importantly—your people, because the process matters as much as the outcome. Focus on your passion and mission, and don't let the noise distract you from it.

6. Why did you choose to publish your work in Brain Sciences? What factors influenced your decision?
Open access, a broad audience across clinical neuroscience, and an editorial process that was both efficient and respectful of authors' time, which, I'll be honest, is becoming rarer than it should be.

7. As a Best Paper Award winner, what suggestions do you have for Brain Sciences to further enhance its service or impact?
Produce richer multimedia content alongside publications—podcasts, interviews, short videos featuring guest editors or authors of landmark papers. The science is there; making it more alive and accessible to a wider audience would meaningfully extend its reach.