Transcranial magnetic stimulation (TMS) is an established treatment for depression. But an important scientific question remains: what exactly happens in the brain when a TMS pulse is delivered?
The familiar explanation is relatively simple. A magnetic pulse stimulates a specific area of the cortex, such as the dorsolateral prefrontal cortex (DLPFC), with repeated stimulation ultimately producing a therapeutic effect. Yet the underlying biology is considerably more complex.
In this Neurocare Academy conversation, Dr. Trevor Brown speaks with Dr. Corey J. Keller, Associate Professor of Psychiatry at Stanford University, about what researchers are learning as they look beyond the stimulation site and into the wider brain networks affected by TMS.
Their discussion explores some of the most fundamental questions in neuromodulation: how stimulation travels through the brain, how repeated TMS may produce neuroplasticity, whether measurable biomarkers could help guide treatment, and how today's largely standardized protocols might eventually become more personalized.
When clinicians describe TMS to patients, treatment is often explained in terms of stimulating a particular cortical region. For depression, that commonly means the left DLPFC.
Dr. Keller's research is helping us look beyond that initial stimulation site.
Through collaborations involving patients undergoing intracranial monitoring for epilepsy, his team has had a rare opportunity to combine TMS with recordings from individual neurons deep within the human brain.
This work has now been published in PNAS in the study Transcranial magnetic stimulation to the dorsolateral prefrontal cortex modulates single-neuron activity in humans.
The researchers recorded activity from approximately 200 individual neurons in four patients, across brain regions including the thalamus and limbic structures. Following a single TMS pulse applied to the DLPFC, they observed changes in activity in around half of the neurons recorded.
The timing of these responses offers a particularly interesting view of how stimulation may propagate through the brain. Neurons within the thalamus responded rapidly following stimulation, while later responses were observed in regions involved in mood and emotional regulation.
Rather than viewing TMS solely as stimulation of an isolated cortical target, these findings provide direct evidence that a pulse delivered at the brain's surface can influence neuronal activity within deeper, interconnected brain structures.
The results also need to be understood in context. The study involved a small number of participants and examined responses to single TMS pulses. A clinical course of TMS involves repeated stimulation delivered across multiple treatment sessions.
The findings therefore provide important new insight into the immediate neural response to TMS, rather than an explanation of the complete mechanism through which a therapeutic course produces clinical improvement.
Understanding what happens immediately after one TMS pulse is only part of the puzzle.
Clinical treatment involves hundreds of pulses within a session and repeated treatment sessions over time. The next question is how those individual effects accumulate and lead to longer-lasting changes in brain function.
This is where neuroplasticity becomes particularly important.
Dr. Keller's team is investigating how stimulation patterns such as intermittent theta burst stimulation affect neural activity, how those effects evolve across treatment, and ultimately how they relate to changes in clinical symptoms.
The goal is to connect different levels of the TMS response: from the immediate activity of individual neurons, through changes within larger networks, to the longer-term effects that matter to patients.
Understanding these mechanisms also raises an important question for the future of TMS: could clinicians one day measure an individual's response to stimulation and use that information to guide treatment?
A major focus of Dr. Keller's research is combining TMS with electroencephalography (EEG).
Intracranial electrodes provide researchers with an exceptional view of activity deep inside the brain, but they clearly cannot form part of routine psychiatric treatment. Scalp EEG, by contrast, offers a non-invasive way of measuring brain activity.
The challenge is determining whether signals recorded from the scalp can reliably tell us something meaningful about the brain's response to stimulation.
Dr. Keller's team is investigating measures of prefrontal excitability and whether EEG-based biomarkers could eventually provide an objective measure of the neural response to TMS.
If reliable biomarkers can be established and validated, they could give researchers new tools to compare stimulation parameters and better understand why individual brains respond differently.
Most patients receiving TMS today are treated using established protocols selected before treatment begins. But the brain itself may change throughout the course of therapy.
This raises an intriguing possibility: could treatment eventually adapt alongside those changes?
Dr. Keller describes a future in which a personalized protocol could potentially be paired with measurements of the brain's response. Researchers could then investigate whether the stimulation parameters remain optimal as treatment progresses.
A reliable marker of neuroplasticity could help answer questions such as whether a particular stimulation pattern is producing the intended neural response, whether that response changes throughout treatment and how stimulation might be adapted accordingly.
This remains a research direction rather than established clinical practice. But it illustrates why understanding the mechanisms behind TMS matters. A clearer picture of how the brain responds could eventually help make treatment more measurable and individualized.
Where and how stimulation is delivered may not be the only variables that matter. Researchers are also asking whether the state of the brain at the moment of stimulation could influence its response.
Dr. Keller's laboratory is exploring this question using several experimental approaches.
One line of research has examined stimulation synchronized with the timing of musical beats. In studies involving the motor system, stimulation delivered at particular points relative to the beat produced stronger measures of excitability, while theta burst stimulation synchronized in this way produced a stronger and longer-lasting plasticity effect.
The team is also exploring cognitive tasks that activate the DLPFC, as well as stimulation during different stages of sleep.
These findings open interesting research questions about whether engaging particular neural circuits at the time of stimulation could influence the resulting response.
For now, however, this is an emerging area of research. There is not yet sufficient clinical evidence to recommend changing what patients do during TMS sessions based on these findings.
That distinction matters. Understanding new possibilities in neuroscience is different from translating them into clinical recommendations.
Despite the many questions still being explored, TMS has already become an important therapeutic option in mental healthcare. The next challenge is understanding whether treatment can become even more precise.
Many current protocols remain relatively standardized, while patients differ in their symptoms, neurobiology and response to stimulation.
Research into TMS mechanisms and biomarkers could eventually add new dimensions to treatment personalization. Future approaches might take into account not only where stimulation is delivered, but how an individual's brain responds and whether that response changes throughout treatment.
Dr. Keller describes the longer-term ambition as connecting several pieces of the puzzle: understanding what happens deep within the brain, determining how repeated stimulation changes neural activity, and identifying how those changes appear in measurements that can be made non-invasively.
The newly published single-neuron findings add an important piece to that picture by giving researchers a direct view of how cortical TMS can influence neurons within deeper human brain circuits.
Perhaps one of the most compelling ideas from the conversation is how differently clinicians may be able to describe TMS in the future.
Rather than thinking only in terms of stimulating a single brain region, researchers are building a more detailed understanding of the networks involved, the timing of their responses and the changes that occur with repeated stimulation.
There is still considerable work ahead. But understanding the mechanisms of TMS is more than an academic exercise. It could provide the foundation for more measurable, adaptive and personalized approaches to neuromodulation.
And with researchers now able to observe the response of individual neurons to TMS directly in the human brain, another piece of that complex picture is beginning to come into focus.
What does TMS really do to the brain, and how could a better understanding of its mechanisms shape treatment in the future?
Watch the full Neurocare Academy conversation with Dr. Corey J. Keller and Dr. Trevor Brown to explore the latest research on TMS mechanisms, neuroplasticity, EEG biomarkers, brain state and the ongoing search for more personalized neuromodulation.
Subscribe to Neurocare Academy for more conversations with researchers and clinicians exploring the science and practice of neuromodulation and mental healthcare.