Soyolsaikhan Odkhuu, Ph.D. — Research Assistant Professor, Department of Psychiatry, Jeonbuk National University Medical School, Republic of Korea
I did not begin in medicine. I trained as a physicist. As an undergraduate in Mongolia, I studied not only physics but also electronics, building and analysing analogue circuits and learning to solve a problem by taking a system apart to see how its pieces fit together. Later, during a master's degree, I worked on the computational side of physics. What eventually drew me toward psychosis research was, in a way, a physicist's question. The brain is the most intricate network we know of, and I became fascinated by what happens to that network when a person's sense of reality begins to change. The field, I found, was full of questions that had barely been explored. I did not want to chase techniques for their own sake; I wanted to use them to understand the brain and mental health. Moving from physics into psychiatry let me bring the mathematics of networks to one of the hardest problems in medicine.
And it is a hard problem, partly because schizophrenia is not a single thing. Two people with the same diagnosis can follow very different paths — different symptoms, different responses to treatment, different chances of recovery. To me, one of the most important questions in our field is whether we can find biological signs that capture those differences: markers that tell us not only that someone has psychosis, but what kind, how far it has progressed, and what is likely to come next. Today we still rely mostly on interviews and observed symptoms. We do not yet have that biological map.
My research tries to build part of it by studying the brain as a network. Picture a transport map: brain regions are the stations, the connections between them are the routes, and a handful of major hubs hold the whole system together. Using scans taken while a person is simply resting, I measure how resilient that network is — and to do this, my colleagues and I "stress-test" it on a computer, removing the busiest hubs one at a time and watching how quickly the rest of the map falls apart. A resilient network reroutes gracefully; a fragile one comes apart early.
Across several studies, a consistent picture has emerged. In people experiencing their first episode of a schizophrenia-spectrum disorder, these core networks were measurably less resilient than in healthy individuals — and the people whose networks held up best also tended to have sharper thinking and milder symptoms. In another study, I looked at people who had recovered well enough to stop their medication; those whose networks were less resilient were more likely to relapse afterward, while those with sturdier networks tended to stay well. And when I compared schizophrenia with a milder, harder-to-classify form of psychosis, that second group sat in between — their networks partly preserved, as if reflecting a genuinely different biology rather than simply a milder version of the same illness.
Why does this matter outside the laboratory? Because a measure of network resilience could eventually help with real decisions. It might help a clinician weigh who can safely come off medication and who needs closer follow-up. It might flag, early on, who is most vulnerable, so that support arrives sooner rather than later. Imagine a young person who develops psychosis twenty years from now: the hope behind this work is that, instead of a long stretch of trial and error, their care could be guided by a read-out of how their brain is actually coping — personalized from the very start.
This past year I had the privilege of presenting some of this work at the SIRS Annual Congress in Florence as an Early Career Awardee. While I presented, many researchers took an interest and asked me question after question — a fantastic and motivating experience for a young researcher, and one that sent me home with new ideas. What I value most about SIRS is that it brings together people from every corner of the world who are circling the same difficult questions, and it makes it possible for someone early in their career to be part of that conversation.
I am now leading a project to turn these resilience measures into a practical toolbox — one that can sort psychosis into clearer subtypes and stages of illness. If it succeeds, the goal is that when a patient first comes to the clinic, we could identify which subtype they have and how far the illness has progressed and begin to anticipate what lies ahead. My hope is that the brain's quiet ability to stay standing under strain — its resilience — turns out to be something we can measure, protect, and one day help to strengthen.
