Transcranial magnetic stimulation (TMS) uses magnetic pulses to generate tiny electrical currents inside the brain, activating neurons and triggering a cascade of changes in brain chemistry, connectivity, and even structure. Unlike medications that circulate through the entire body, TMS targets specific brain regions directly. It’s most commonly used for treatment-resistant depression, and its effects range from the immediate firing of nerve cells to long-lasting rewiring of neural circuits over weeks of treatment.
How Magnetic Pulses Become Brain Activity
A TMS device works through electromagnetic induction. A coil placed against the scalp passes a brief, powerful current that generates a rapidly changing magnetic field. That field passes painlessly through the skull and induces a small electric field in the brain tissue underneath. If that electric field is strong enough, it pushes a neuron’s voltage past its firing threshold and triggers an action potential, the same electrical impulse neurons use to communicate naturally.
The induced electric field is strongest in the superficial layers of the cortex and the white matter just beneath it. TMS activates the axons of both excitatory and inhibitory neurons, meaning it doesn’t simply “turn on” a brain region. Instead, it shifts the balance of activity in the targeted area. The exact effect depends on the orientation of the coil, the anatomy of the brain folds underneath, and the electrical properties of each individual neuron at the moment of stimulation.
Rewiring Synapses Over Time
A single TMS pulse causes neurons to fire in the moment, but the therapeutic power of the treatment comes from repetitive stimulation delivered over many sessions. Repeated TMS (rTMS) changes synaptic plasticity, the brain’s ability to strengthen or weaken connections between neurons. This happens through two well-established processes: long-term potentiation (LTP), which strengthens a synapse, and long-term depression (LTD), which weakens it.
The frequency of the pulses determines which process dominates. High-frequency stimulation (5 Hz and above) tends to produce LTP-like effects, making targeted circuits more responsive. Low-frequency stimulation (around 1 Hz applied for more than 10 minutes) produces LTD-like effects, quieting overactive circuits. The shift hinges on calcium levels inside the receiving neuron: stronger calcium influx tips the balance toward strengthening, while a more modest rise favors weakening. This frequency-dependent control gives clinicians a way to dial brain activity up or down in specific regions.
Animal studies confirm that these plasticity changes are real, not just theoretical. They involve altered expression of genes and proteins tied to a key receptor involved in learning and memory. In other words, rTMS doesn’t just temporarily nudge brain activity. It changes the molecular machinery that governs how neurons talk to each other.
Shifts in Brain Chemistry
TMS also changes neurotransmitter levels in ways that overlap with what antidepressant medications do, though through a different route. Stimulating the left dorsolateral prefrontal cortex (DLPFC), the standard target for depression treatment, increases dopamine release in the striatum, a deep brain structure involved in motivation and reward. This has been confirmed in both healthy volunteers and people with depression using brain imaging with radiotracers.
Serotonin responds as well. High-frequency TMS over the left DLPFC modulates serotonin release across several limbic regions involved in mood regulation. Animal studies show increased serotonin in the hippocampus after prefrontal stimulation, suggesting some mechanistic overlap with the class of antidepressants that block serotonin reuptake. That said, serotonin modulation from TMS has not yet been conclusively demonstrated in people with depression specifically.
TMS also affects the brain’s two main signaling molecules that govern excitation and inhibition. In people with treatment-resistant depression, 10 Hz stimulation over the left DLPFC has been shown to increase levels of the brain’s primary inhibitory chemical in the medial prefrontal cortex. In healthy subjects, the same protocol increases the primary excitatory chemical in that region. One study found that changes in the ratio of these two chemicals after a full course of TMS correlated with how much patients’ symptoms improved, and those changes persisted at a six-month follow-up.
Reaching Deep Brain Structures Through Networks
One of the most important things TMS does isn’t at the site where the coil sits. It’s what happens downstream. The left DLPFC has strong connections to deeper structures that regulate emotion, particularly the subgenual anterior cingulate cortex (sgACC), a region consistently found to be overactive in depression. TMS can’t reach the sgACC directly because it sits too deep, but stimulating the DLPFC on the surface modulates it through their shared circuitry.
Brain imaging studies show that the strength of the natural connection between the DLPFC stimulation site and the sgACC predicts how well a patient will respond to treatment. After a course of TMS, connectivity between the DLPFC and the anterior cingulate increases, at least temporarily. One study found this boost was detectable 15 minutes after stimulation but had faded by 30 minutes, suggesting that sustained courses of treatment are needed to make these connectivity changes stick.
Resetting Communication Between Brain Networks
Depression involves disrupted communication between large-scale brain networks, and TMS appears to partially restore normal patterns. Two networks are especially relevant: the central executive network (CEN), anchored in the DLPFC, which handles goal-directed thinking, and the default mode network (DMN), which is active during rumination and self-referential thought. In healthy brains, these two networks have an inverse relationship. When one is active, the other quiets down. In depression, that seesaw breaks, and the DMN tends to stay active even when it shouldn’t.
Research shows that TMS induces anticorrelations between the DLPFC and key DMN regions, including the ventromedial prefrontal cortex and posterior cingulate cortex. In depressed patients, these anticorrelations were absent before treatment and emerged after a course of TMS. In practical terms, this means TMS helps restore the brain’s ability to shift out of ruminative, inward-focused states and engage with the external world. The treatment also changes connectivity within both networks, suggesting a broad reorganization rather than a single targeted fix.
Physical Changes in Brain Structure
Perhaps the most surprising effect of TMS is that it can change the physical volume of brain tissue. A study of high-frequency left prefrontal rTMS found a 3.4% volume increase in the left hippocampus, the brain’s memory center, on the same side as stimulation. The hippocampus is one of the brain structures most consistently found to be shrunken in depression, so this finding is particularly relevant.
The hippocampus sits far from the stimulation site on the scalp. The volume change is thought to occur through a nerve fiber bundle that connects the prefrontal cortex to the hippocampus, representing a remote neuroplastic effect. Whether this reflects the birth of new neurons, increased branching of existing ones, or other cellular changes is still being worked out. But the structural change itself adds weight to the idea that TMS does something lasting to the brain, not just a temporary shift in electrical activity.
How Treatment Sessions Work
Standard high-frequency rTMS for depression delivers pulses at 10 Hz or higher to the left DLPFC, with sessions lasting about 37.5 minutes and treatment courses running five days a week for four to six weeks. Each session delivers between 1,600 and 3,000 pulses. A newer protocol called intermittent theta-burst stimulation (iTBS) compresses 600 pulses into just over three minutes by using rapid triplet bursts at 50 Hz. The FDA approved iTBS for treatment-resistant depression in 2018, and head-to-head trials show it performs comparably to standard rTMS in a fraction of the time.
Response rates to TMS for major depression range between 50% and 55%, meaning roughly half of patients experience meaningful symptom improvement. Full remission rates fall between 30% and 35%. These numbers are notable because TMS is typically offered to patients who have already failed to improve on one or more antidepressant medications.
Safety and Seizure Risk
The most serious potential side effect of TMS is a seizure, but the risk is very low. A survey covering more than 586,000 treatment sessions across 25,500 patients recorded 18 seizures total, an overall rate of about 0.31 per 10,000 sessions, or roughly 0.71 per 1,000 patients. The risk varies by device type. Standard figure-8 coils had a combined seizure rate of 0.14 per 1,000 patients, while the deeper-penetrating H-coil had a higher rate of 5.56 per 1,000 patients. Even the higher figure represents a risk well under 1%. The most common side effects are scalp discomfort at the coil site and mild headaches, both of which typically fade after the first few sessions.