A comprehensive review of how modern brain stimulation reshapes neural circuits—from synapses to brain networks.
By Gavin Hart
25-minute read
Introduction
Transcranial magnetic stimulation, or TMS, can influence brain activity without surgery by delivering magnetic pulses through a coil placed against the scalp. It is already used to treat several psychiatric conditions, particularly depression, and researchers are increasingly exploring its potential in neurological rehabilitation. But one question has the potential to shape the future of the field: how can a magnetic pulse lasting only an instant produce effects that continue long after the stimulation ends?
Neuroplasticity is widely believed to provide much of the biological foundation for these lasting effects. Although the exact mechanisms remain only partially understood, decades of research have begun to reveal how precisely patterned magnetic stimulation can alter cortical excitability, strengthen or weaken neural connections, and reshape communication across larger brain networks. This article examines what is currently known about TMS and neuroplasticity, what newer methods such as quadripulse stimulation and quadripulse theta-burst stimulation have revealed about the importance of pulse timing, and how these discoveries may lead to more precise and personalized treatments.
Section 1: Neuroplasticity
The human brain is not a fixed organ. Throughout life, it changes in response to experience, learning, environmental demands, and injury. This ability to adapt is known as neuroplasticity. At its most basic level, neuroplasticity involves changes in how strongly neurons communicate with one another. When these changes occur across many neurons, they can reshape the activity of entire brain circuits.
Two important forms of neuroplasticity are long-term potentiation, or LTP, and long-term depression, or LTD. LTP is a lasting increase in communication between neurons. When connected neurons are repeatedly activated together, their connection may become more responsive. LTD produces the opposite effect by reducing the strength of communication across a synapse. Early experiments showed that repeated patterns of stimulation could produce long lasting increases in synaptic strength, helping establish the biological basis for LTP (Bliss and Lømo, 1973).
LTD does not necessarily destroy a synapse. Instead, it weakens the connection, allowing the brain to reduce activity that is no longer useful or that interferes with more important signals. LTP and LTD therefore work together. LTP can reinforce useful patterns of communication, while LTD can weaken competing or unnecessary ones. These processes support learning, memory, and the continuing refinement of neural circuits (Suppa et al., 2016).
Although LTP and LTD are often described at the level of an individual synapse, neuroplasticity also occurs across larger networks. The cerebral cortex consists of interconnected regions that continually exchange information. As the strength and timing of these connections change, so can movement, sensory processing, cognition, and emotional regulation.
This network perspective is particularly important when considering psychiatric and neurological disorders. Depression, for example, is not simply caused by one brain region being too active or not active enough. Research suggests that it involves disrupted communication among several regions involved in mood, attention, memory, and emotional regulation. Fox et al. found that the effectiveness of different TMS targets for depression was related to their functional connection with the subgenual cingulate, a deeper region involved in mood regulation (Fox et al., 2012).
Neuroplasticity can be beneficial. It allows practice to strengthen the pathways needed for a new skill, and it can help surviving brain regions reorganize following an injury such as a stroke or traumatic brain injury. However, plasticity is not always helpful. The brain can also strengthen patterns that contribute to chronic pain, compulsive behavior, or persistent negative thinking. Conditions such as major depressive disorder, obsessive compulsive disorder, and chronic pain may therefore involve maladaptive plasticity, in which dysfunctional patterns become repeatedly reinforced.
Because neuroplasticity can strengthen both healthy and unhealthy patterns, it has become an important treatment target. An effective therapy may need to weaken maladaptive activity while strengthening healthier communication within the brain. TMS is believed to influence this balance.
Section 2: How TMS Changes the Brain
A TMS pulse begins as a rapidly changing magnetic field produced by a coil placed against the scalp. The magnetic field passes through the scalp and skull and creates a small electric current in the outer layers of the brain through a process called electromagnetic induction. This electric current can activate neural structures, particularly axons, which carry signals between neurons. The response depends on the position and angle of the coil, the direction of the induced current, the shape of the pulse, and the anatomy of the targeted area (Ilmoniemi and Kicić, 2010).
TMS does not activate every neuron beneath the coil in the same way. Instead, it influences a mixture of excitatory and inhibitory circuits. When stimulation is applied over the motor cortex, it can produce signals called indirect waves, or I waves, within local cortical circuits. These signals occur at precise intervals, which helped inspire stimulation methods such as quadripulse stimulation. Research on I waves has shown that the timing between pulses and the direction of the induced current can affect which circuits are recruited and how strongly they respond (Sakai et al., 1997; Hamada et al., 2007).
A single pulse creates only a brief electrical response. Repeated pulses, however, may engage the chemical systems that allow synapses to become stronger or weaker. One important chemical is glutamate, the brain’s main excitatory neurotransmitter. Glutamate acts on several receptors, including AMPA and NMDA receptors, located on the receiving neuron.
AMPA receptors help activate the receiving neuron. If the neuron becomes sufficiently active, NMDA receptors allow calcium to enter the cell. Calcium then acts as an internal signal. Under some conditions, it leads to the addition or strengthening of AMPA receptors, making the synapse more responsive and producing an LTP-like effect. Under other conditions, AMPA receptors are removed, weakening the connection and producing an LTD like effect (Suppa et al., 2016).
Figure 1. Simplified model of long-term potentiation and long-term depression. Calcium entering through NMDA receptors can strengthen a synapse by increasing AMPA receptors or weaken it by reducing them. This is a leading model of TMS induced plasticity, although these receptor changes cannot be observed directly during routine treatment. Adapted from Molnar and Gair, Figure 16.18.
The NMDA receptor study by Huang et al. supports this model. Participants received either memantine, a drug that blocks NMDA receptor activity, or a placebo before theta burst stimulation. Memantine largely prevented the lasting changes in cortical excitability normally produced by the stimulation, suggesting that NMDA receptors help maintain TMS induced plasticity (Huang et al., 2007).
Glutamate is only part of the process. Brain activity is controlled through a balance between excitation and inhibition. While glutamate generally makes neurons more likely to fire, GABA generally makes firing less likely. The interaction between these systems may help determine whether a stimulation pattern increases or decreases cortical excitability and may also help explain why individuals respond differently to the same protocol.
The effects of TMS are also not limited to the area directly beneath the coil. The stimulated region communicates with other areas through existing neural pathways, allowing local changes to influence larger networks. In depression, for example, TMS is commonly applied to the dorsolateral prefrontal cortex, which is connected to deeper regions involved in emotional regulation. The relationship between effective treatment targets and the subgenual cingulate supports the idea that TMS acts through connected networks rather than through one isolated area (Fox et al., 2012).
TMS is therefore better understood as a signal introduced into an active communication network than as a switch that turns one region on or off. Repeated stimulation may alter synaptic responsiveness, the balance between excitation and inhibition, and communication between distant regions. Researchers need reliable measurements to determine whether those changes have occurred.
Section 3: Measuring Neuroplasticity
Because individual synapses cannot be observed directly during routine TMS, researchers compare measurements taken before and after stimulation. Each method captures a different part of the response, so no single test can provide a complete picture.
One of the most common measurements is the motor-evoked potential, or MEP. When TMS is applied to the part of the motor cortex that controls the hand, it can produce a small muscle response that is recorded using electrodes placed on the skin. Researchers deliver the same test pulse before and after a stimulation protocol and compare the size of the responses. A larger MEP generally suggests that the motor pathway has become more excitable, while a smaller MEP suggests reduced excitability (Hamada et al., 2007).
MEPs provide a simple way to track changes over time, but they measure only the motor system. The final response is influenced by the motor cortex, spinal cord, peripheral nerves, and muscles. An MEP can therefore show that a pathway has changed, but it cannot identify the exact synapses responsible or reveal what is happening in brain regions involved in mood, memory, or attention.
Researchers can examine cortical activity more directly by combining TMS with electroencephalography, or EEG. EEG uses electrodes placed across the scalp to record electrical activity produced by groups of neurons. When a TMS pulse is delivered during EEG recording, it produces responses known as TMS evoked potentials, or TEPs. These responses allow researchers to examine activity near the stimulation site and observe how it spreads to connected regions (Ilmoniemi and Kicić, 2010).
Figure 2. Motor-evoked potentials and TMS evoked potentials provide two different ways of measuring responses to brain stimulation. MEPs record a response from a muscle, while TEPs use EEG to record electrical activity from the brain. Adapted from Hommelsen (2021), Figure 1, Panel A.
TMS EEG also presents challenges. The magnetic pulse can interfere with the recording equipment, while the click of the coil and the sensation on the scalp can create auditory and sensory responses that must be separated from the brain signal (Ilmoniemi and Kicić, 2010). Resting EEG and functional MRI provide additional information by tracking ongoing brain rhythms and communication between larger networks.
No single measurement directly proves that neuroplasticity has occurred. The strongest evidence comes when several methods show related changes. These measurements have also revealed one of the most important limitations of TMS: the same protocol does not produce the same response in every person.
Section 4: Why New Stimulation Protocols Were Developed
Conventional explanations often state that high frequency rTMS increases cortical excitability while low frequency rTMS decreases it. These patterns may appear when researchers average the responses of an entire group, but individual results can differ substantially. One participant may show the expected response, another may show little change, and another may respond in the opposite direction (Bergmann et al., 2019; Müller Dahlhaus et al., 2017).
Responses can also vary when the same individual receives the same protocol on different days. Differences among participants are known as between person variability, while changes within the same participant are called within person variability. Brain anatomy, baseline activity, sleep, medication use, attention, recent physical activity, and previous stimulation may all affect the outcome. Technical factors such as coil position, current direction, pulse shape, intensity, and total pulse number can introduce additional variation.
This variability creates both a scientific and a clinical problem. A study may report a clear average increase in MEP size even though many individual participants did not show that response. A change in an MEP also does not automatically mean that symptoms or behavior will improve. MEPs measure the motor system, while conditions such as depression, pain, or cognitive impairment involve broader networks.
Researchers developed newer protocols in an effort to produce stronger, faster, and more predictable effects. Theta burst stimulation shortened treatment time by delivering pulses in brief repeating bursts, but later studies showed that theta burst responses can also vary considerably between individuals (Huang et al., 2005; Suppa et al., 2016).
Attention therefore shifted from overall frequency alone to the precise timing of individual pulses. Researchers began asking whether controlling the spacing, direction, and shape of pulses could produce more dependable plasticity. Quadripulse stimulation and quadripulse theta burst stimulation became important tools for investigating this question.
Section 5: Quadripulse Stimulation
Early TMS research often described higher-frequency repetitive TMS as increasing cortical excitability and stimulation near 1 Hz as decreasing it. Researchers refer to these changes as LTP-like and LTD-like because they resemble long-term potentiation and long-term depression in their direction and duration. However, these are general tendencies rather than fixed rules. The outcome can vary according to stimulation intensity, pulse number, pulse pattern, target, medication use, and the person’s brain state (Bergmann et al., 2019; Müller-Dahlhaus et al., 2017).
This uncertainty led researchers to investigate whether controlling the timing of individual pulses more precisely could produce stronger and more dependable effects. Quadripulse stimulation, or QPS, was developed to test whether four carefully timed pulses could create more durable changes than simpler pulse patterns. Hamada et al. found that four pulses separated by 1.5 milliseconds produced a larger and longer-lasting increase in motor cortical excitability than a comparable paired-pulse protocol (Hamada et al., 2007).
A standard QPS session delivers a burst of four monophasic pulses every five seconds for approximately 30 minutes. This produces 360 bursts and 1,440 pulses in total. Researchers measure MEPs before and after the session to determine whether the motor pathway has become more or less responsive.
Figure 3. Basic structure of a QPS session. Each burst contains four pulses and is followed by a five-second interval. The bursts continue for approximately 30 minutes, while MEPs are measured before and after stimulation. Adapted from Ugawa (2015).
One of the strongest pieces of evidence connecting patterned TMS with neuroplasticity came from the duration of the response. Hamada et al. found that MEP amplitudes increased to more than twice their baseline level following QPS and remained significantly elevated for approximately 75 minutes (Hamada et al., 2007). By 90 minutes, the response had returned close to baseline. As shown in Figure 4, stimulation changed the responsiveness of the motor pathway well beyond the period in which the pulses were being delivered.
Figure 4. Changes in normalized motor-evoked potential amplitude following quadripulse stimulation. The dotted line represents the baseline response before stimulation. MEP amplitudes remained significantly elevated for approximately 75 minutes and returned close to baseline by 90 minutes. Asterisks indicate statistically significant differences from the prestimulation measurement. From Hamada et al. (2007), Figure 5.
The prolonged response is important because it suggests that QPS does more than briefly activate neurons. It produces a lasting change in how readily the pathway responds to later stimulation. Researchers call this an LTP-like effect rather than claiming that QPS reproduces natural LTP exactly, since an MEP cannot reveal every molecular change occurring at individual synapses.
QPS also demonstrated that very small changes in pulse timing could reverse the direction of the response. Short intervals of 1.5, 5, or 10 milliseconds generally increased MEP amplitudes, producing an LTP-like effect. Longer intervals of 30, 50, or 100 milliseconds generally reduced MEP amplitudes, producing an LTD-like effect (Hamada et al., 2008). QPS with a five-millisecond interval, known as QPS5, produced the strongest facilitation. QPS with a 50-millisecond interval, known as QPS50, produced the strongest suppression (Hamada et al., 2008).
Figure 5. Changes in motor pathway excitability following QPS5 and QPS50. QPS5 increased MEP size, while QPS50 reduced it. A value above the original baseline represents increased excitability, while a value below baseline represents reduced excitability. Adapted from Ugawa (2015).
These findings showed that the total number of pulses does not fully determine the biological effect. A difference of only a few milliseconds can influence whether a pathway becomes more or less responsive. Later research also found that the interval between bursts, the duration of the session, and whether the pulses were monophasic or biphasic affected the outcome. In one study of 35 healthy participants, 80 percent showed the expected increase in excitability following QPS5 under standard conditions (Nakamura et al., 2016).
QPS therefore provided both evidence for lasting TMS-induced plasticity and a more controlled way to study it. It established an important principle: timing is not merely a technical setting. It is part of the stimulation dose. This insight led researchers to combine the four-pulse structure of QPS with the repeating rhythm used in theta-burst stimulation.
Section 6: QuadriPulse Theta Burst Stimulation
While QPS demonstrated the importance of timing within a four-pulse burst, theta-burst stimulation, or TBS, showed that brief repeating patterns could also produce lasting changes in cortical excitability. (Huang et al.) found that a relatively short period of TBS could influence motor cortical activity after the stimulation had ended (Huang et al., 2005).
Pharmacological evidence strengthened the connection between TBS and neuroplasticity. In a later study, participants received either memantine, a drug that blocks NMDA receptor activity, or a placebo before TBS. Memantine largely prevented the lasting changes in cortical excitability that the stimulation would normally produce (Huang et al., 2007). Because NMDA receptors contribute to many forms of learning-related synaptic plasticity, this result suggests that patterned TMS engages at least some of the same biological systems involved in strengthening and weakening neural connections.
Quadripulse theta-burst stimulation, or qTBS, was developed by combining the four-pulse structure of QPS with the repeating 5 Hz rhythm used in TBS. The goal was to determine whether researchers could gain greater control over neuroplasticity by managing both the timing within each burst and the rhythm between bursts.
Jung et al. delivered 360 bursts containing four pulses each, resulting in 1,440 total pulses. The bursts were repeated every 200 milliseconds. The researchers compared pulses separated by 1.5 milliseconds, equal to approximately 666 Hz, with pulses separated by 5 milliseconds, equal to 200 Hz. They also reversed the direction of the induced current from posterior to anterior, or PA, to anterior to posterior, or AP (Jung et al., 2016).
Figure 6. Structure of quadripulse theta-burst stimulation. Each burst contains four closely spaced pulses, and the bursts are repeated every 200 milliseconds. The lower portion shows how the direction of the induced current can be reversed. From Jung et al. (2016), Figure 2.
The 1.5-millisecond interval was selected because it approximately matches the timing of the brain’s naturally occurring I-waves. At this interval, changing the current direction reversed the physiological result. AP stimulation increased MEP amplitudes, while PA stimulation decreased them (Jung et al., 2016). Nearly identical pulse sequences therefore produced opposite effects depending on the direction in which the current entered the cortical circuit.
The findings were different when pulses were separated by 5 milliseconds. At this interval, both AP and PA stimulation increased MEP amplitudes for at least 60 minutes (Jung et al., 2016). Pulse timing therefore appeared to influence how strongly the response depended on current direction.
Figure 7. Changes in MEP amplitudes following qTBS. In the upper panels, pulses separated by 1.5 milliseconds produced opposite effects depending on current direction. In the lower panels, pulses separated by 5 milliseconds increased excitability in both directions. From Jung et al. (2016), Figure 3.
Jung et al. also found no significant change in resting motor threshold following qTBS (Jung et al., 2016). This suggests that the protocol altered the responsiveness of particular cortical circuits rather than simply making the entire motor cortex easier to activate.
Together, the QPS and qTBS studies demonstrate why TMS cannot be understood as a simple switch that turns brain activity on or off. The outcome depends on the timing between pulses, direction of the induced current, pulse waveform, and circuits recruited by the coil. The lasting changes resemble LTP and LTD, while the NMDA receptor findings provide biological support for a connection to neuroplasticity. Researchers nevertheless describe the results as LTP-like and LTD-like because the complete molecular process cannot be directly observed during human stimulation.
qTBS remains an experimental protocol. The Jung study involved a small group of healthy volunteers and measured motor cortical physiology rather than improvements in clinical symptoms. Its value lies primarily in what it revealed about neuroplasticity: changes of only a few milliseconds can determine which circuits respond and whether excitability rises or falls.
Section 7: Clinical Applications Beyond Depression
The ability of TMS to influence brain networks has encouraged researchers to study it in conditions involving movement, language, rehabilitation, and pain. In most neurological applications, TMS is being investigated as an addition to medication, physical therapy, or other established treatment rather than as a replacement.
Stroke rehabilitation is one of the most actively studied applications. A stroke can disrupt communication between the damaged hemisphere and the opposite side of the brain. One approach applies higher frequency stimulation to the motor cortex of the damaged hemisphere in an effort to increase its activity. Another applies lower frequency stimulation to the unaffected hemisphere in an effort to reduce activity that may interfere with recovery. Researchers now recognize that this balance differs according to the location and severity of the stroke, so the same strategy may not be appropriate for every patient (Li et al., 2024).
When TMS is paired with physical or occupational therapy, stimulation may temporarily make the brain more receptive to practicing a movement. Rehabilitation then provides the activity that the brain is being asked to strengthen. A recent meta analysis found that rTMS can support motor recovery after stroke, with encouraging results when treatment begins within six months and in patients with more severe initial impairment (Zhang et al., 2025).
TMS has also been studied for language loss following stroke. An analysis of 17 studies involving 682 patients found small but significant improvements in poststroke aphasia, although the results depended partly on the stimulation protocol used (Wang et al., 2024). Research has also explored TMS for swallowing problems following stroke, but the evidence remains promising rather than conclusive because many studies have been small and have used different methods (Georgiou et al., 2024).
Parkinson’s disease is another major area of research. The condition affects networks connecting the cortex, basal ganglia, and cerebellum, producing symptoms such as slowed movement, balance problems, and freezing of gait. A 2024 analysis found improvements in walking speed and several measures of mobility following TMS, although the studies used different targets and stimulation methods (Liu et al., 2024).
Because Parkinson’s disease affects a connected movement network, researchers are exploring targets beyond the primary motor cortex. These include frontal, parietal, and cerebellar regions involved in movement planning, coordination, and gait (Panda et al., 2024).
Figure 8. A network view of movement control in Parkinson’s disease. Movement depends on communication among cortical regions, the basal ganglia, and the cerebellum. The figure also illustrates several possible TMS targets. Adapted from Panda et al. (2024), Figure 1.
Researchers are also investigating TMS for chronic neuropathic pain and other neurological symptoms. Evidence suggests that stimulation of the motor cortex can reduce some forms of peripheral and central neuropathic pain, but the quality and durability of the results differ among conditions (da Cunha et al., 2024).
These clinical applications test whether the plasticity observed in motor cortex experiments can be guided toward meaningful recovery in damaged or disrupted networks. The results are encouraging, but they also show that a physiological effect in a laboratory does not automatically become a dependable clinical treatment.
Section 8: What We Still Do Not Know
One of the largest unanswered questions is why people respond so differently to the same stimulation. A protocol that increases excitability in one person may produce little change or the opposite effect in another. Responses may also vary when the same person receives the same protocol on different days. Brain anatomy, medication use, sleep, attention, recent activity, coil placement, and the brain’s activity at the moment of each pulse may all influence the result.
The biological mechanisms also remain incomplete. MEPs, EEG, pharmacological studies, and brain imaging suggest that TMS engages systems involved in synaptic plasticity and changes communication across larger networks. However, researchers cannot routinely observe receptor movement, calcium signaling, or synaptic strengthening directly during human treatment. A measurable change in motor cortical excitability also does not necessarily explain an improvement in speech, movement, pain, or mood.
Translating research into clinical practice presents additional challenges. Studies often differ in coil type, target location, intensity, session number, patient characteristics, and the method used to measure improvement. A protocol may appear effective when responses are averaged across a group even though only some participants experience a meaningful benefit. Follow up periods are also often short, leaving uncertainty about how long improvements last and whether maintenance treatment is needed.
One possible solution is to personalize stimulation according to each patient’s anatomy, network connectivity, and current brain activity. Conventional TMS generally delivers pulses according to a predetermined schedule. Brain state-dependent TMS instead uses EEG to monitor ongoing rhythms and delivers stimulation when the targeted network appears most receptive.
In a 2024 proof of concept trial involving 30 people with chronic stroke, EEG triggered stimulation was feasible and produced improvements comparable with an established stroke protocol, although it was not shown to be superior (Mahmoud et al., 2024).
Future systems may go beyond triggering a pulse at a selected moment. Real time TMS EEG could eventually measure the brain’s response and adjust the timing or intensity of later pulses, creating a closed loop between measurement and stimulation. These systems remain experimental, and researchers must still solve technical problems involving stimulation artifacts, signal interpretation, safety, and speed (Wischnewski et al., 2024).
The future of TMS may therefore depend less on identifying one protocol that works for everyone and more on finding the correct target, timing, dose, and accompanying therapy for each individual. Achieving that goal will require larger controlled trials, standardized measurements, longer follow up, validated biomarkers, and clearer evidence that physiological changes lead to improvements patients can experience in daily life.
Section 9: Conclusion
Research on TMS has shown that the brain’s response to stimulation depends on far more than whether pulses are delivered quickly or slowly. The timing between pulses, direction of the induced current, position of the coil, and activity of the brain at the moment of stimulation can all shape the outcome. Evidence from MEPs, EEG, brain imaging, and pharmacological studies suggests that TMS can produce lasting changes in cortical excitability and network communication. These changes share important features with LTP and LTD, supporting the view that neuroplasticity is a major part of how TMS works.
Newer protocols such as QPS and qTBS have deepened this understanding. QPS showed that changing the interval between pulses can reverse the direction of the response. qTBS demonstrated that pulse timing and current direction can interact to recruit different cortical circuits. Together, these studies suggest that the detailed organization of stimulation matters as much as the total number of pulses.
At the same time, these findings have revealed the limits of current knowledge. A strong physiological response in the motor cortex does not always lead to clinical improvement, and the same protocol can produce different effects in different people. Researchers still do not fully understand which biological changes are most important, how long they last, or how they should be measured across different disorders.
The future of TMS is therefore likely to involve greater precision rather than simply more stimulation. Brain imaging, EEG, and other biomarkers may help identify the most appropriate target, timing, and dose for each person. Closed loop systems may eventually adjust stimulation according to the brain’s immediate state and response.
TMS has already demonstrated that a magnetic pulse can influence the brain beyond the brief moment when it is delivered. The next challenge is learning how to guide that plasticity more reliably. As researchers better understand the interaction among pulse timing, brain state, and network structure, TMS may become a more personalized tool for strengthening healthy pathways, weakening maladaptive patterns, and supporting recovery across a wider range of psychiatric and neurological conditions.
Works Cited
Bergmann, Til Ole. “Brain State-Dependent Brain Stimulation.” Frontiers in Psychology, vol. 9, 2018, article 2108. https://doi.org/10.3389/fpsyg.2018.02108.
Bliss, T. V. P., and T. Lømo. “Long-Lasting Potentiation of Synaptic Transmission in the Dentate Area of the Anaesthetized Rabbit Following Stimulation of the Perforant Path.” The Journal of Physiology, vol. 232, no. 2, 1973, pp. 331–356. https://doi.org/10.1113/jphysiol.1973.sp010273.
da Cunha, Pedro Henrique Martins, et al. “Neuromodulation for Neuropathic Pain.” International Review of Neurobiology, vol. 179, 2024, pp. 471–502. https://doi.org/10.1016/bs.irn.2024.10.013.
Fox, Michael D., et al. “Efficacy of Transcranial Magnetic Stimulation Targets for Depression Is Related to Intrinsic Functional Connectivity with the Subgenual Cingulate.” Biological Psychiatry, vol. 72, no. 7, 2012, pp. 595–603. https://doi.org/10.1016/j.biopsych.2012.04.028.
Georgiou, Anastasios M., et al. “The Effectiveness of Transcranial Magnetic Stimulation for Dysphagia in Stroke Patients: An Umbrella Review of Systematic Reviews and Meta-Analyses.” Frontiers in Human Neuroscience, vol. 18, 2024, article 1355407. https://doi.org/10.3389/fnhum.2024.1355407.
Hamada, Masashi, et al. “Quadro-Pulse Stimulation Is More Effective than Paired-Pulse Stimulation for Plasticity Induction of the Human Motor Cortex.” Clinical Neurophysiology, vol. 118, no. 12, 2007, pp. 2672–2682. https://doi.org/10.1016/j.clinph.2007.09.062.
Hamada, Masashi, et al. “Bidirectional Long-Term Motor Cortical Plasticity and Metaplasticity Induced by Quadripulse Transcranial Magnetic Stimulation.” The Journal of Physiology, vol. 586, no. 16, 2008, pp. 3927–3947. https://doi.org/10.1113/jphysiol.2008.152793.
Hommelsen, Maximilian. Inducing and Detecting Neuroplasticity: Insights from TMS-EEG and RS-EEG. 2021. University of Cologne, doctoral dissertation. Kölner UniversitätsPublikationsServer, https://kups.ub.uni-koeln.de/54163/. Accessed 12 July 2026.
Huang, Ying-Zu, et al. “Theta Burst Stimulation of the Human Motor Cortex.” Neuron, vol. 45, no. 2, 2005, pp. 201–206. https://doi.org/10.1016/j.neuron.2004.12.033.
Huang, Ying-Zu, et al. “The After-Effect of Human Theta Burst Stimulation Is NMDA Receptor Dependent.” Clinical Neurophysiology, vol. 118, no. 5, 2007, pp. 1028–1032. https://doi.org/10.1016/j.clinph.2007.01.021.
Huang, Ying-Zu, et al. “Plasticity Induced by Non-Invasive Transcranial Brain Stimulation: A Position Paper.” Clinical Neurophysiology, vol. 128, no. 11, 2017, pp. 2318–2329. https://doi.org/10.1016/j.clinph.2017.09.007.
Ilmoniemi, Risto J., and Kaisu Kicić. “Methodology for Combined TMS and EEG.” Brain Topography, vol. 22, no. 4, 2010, pp. 233–248. https://doi.org/10.1007/s10548-009-0123-4.
Jung, Na Young, et al. “Quadri-Pulse Theta Burst Stimulation Using Ultra-High Frequency Bursts: A New Protocol to Induce Changes in Cortico-Spinal Excitability in Human Motor Cortex.” PLOS ONE, vol. 11, no. 12, 2016, article e0168410. https://doi.org/10.1371/journal.pone.0168410.
Li, Ran, et al. “The Stratified Effects of Repetitive Transcranial Magnetic Stimulation in Upper Limb Motor Impairment Recovery after Stroke: A Meta-Analysis.” Frontiers in Neurology, vol. 15, 2024, article 1369836. https://doi.org/10.3389/fneur.2024.1369836.
Liu, Zicai, et al. “The Effects of Transcranial Magnetic Stimulation for Freezing of Gait in Parkinson’s Disease: A Systematic Review and Meta-Analysis of Randomized Controlled Trials.” Frontiers in Aging Neuroscience, vol. 16, 2024, article 1304852. https://doi.org/10.3389/fnagi.2024.1304852.
López-Alonso, Virginia, et al. “Inter-Individual Variability in Response to Non-Invasive Brain Stimulation Paradigms.” Brain Stimulation, vol. 7, no. 3, 2014, pp. 372–380. https://doi.org/10.1016/j.brs.2014.02.004.
Mahmoud, Wala, et al. “Brain State-Dependent Repetitive Transcranial Magnetic Stimulation for Motor Stroke Rehabilitation: A Proof of Concept Randomized Controlled Trial.” Frontiers in Neurology, vol. 15, 2024, article 1427198. https://doi.org/10.3389/fneur.2024.1427198.
Molnar, Charles, and Jane Gair. “How Neurons Communicate.” Concepts of Biology: 1st Canadian Edition, BCcampus, 2015, ch. 16.2, fig. 16.18. OpenTextBC, https://opentextbc.ca/biology/chapter/16-2-how-neurons-communicate/. Accessed 12 July 2026.
Nakamura, Koichiro, et al. “Variability in Response to Quadripulse Stimulation of the Motor Cortex.” Brain Stimulation, vol. 9, no. 6, 2016, pp. 859–866. https://doi.org/10.1016/j.brs.2016.01.008.
Panda, Rupsha, et al. “Improving Efficacy of Repetitive Transcranial Magnetic Stimulation for Treatment of Parkinson Disease Gait Disorders.” Frontiers in Human Neuroscience, vol. 18, 2024, article 1445595. https://doi.org/10.3389/fnhum.2024.1445595.
Sakai, Kazuhisa, et al. “Preferential Activation of Different I Waves by Transcranial Magnetic Stimulation with a Figure-of-Eight-Shaped Coil.” Experimental Brain Research, vol. 113, no. 1, 1997, pp. 24–32. https://doi.org/10.1007/BF02454139.
Suppa, Antonio, et al. “Ten Years of Theta Burst Stimulation in Humans: Established Knowledge, Unknowns and Prospects.” Brain Stimulation, vol. 9, no. 3, 2016, pp. 323–335. https://doi.org/10.1016/j.brs.2016.01.006.
Ugawa, Yoshikazu. “Synaptic and Axonal Plasticity Induction in the Human Cerebral Cortex.” Innovative Medicine: Basic Research and Development, edited by Kazuwa Nakao, Nagahiro Minato, and Shinji Uemoto, Springer, 2015, ch. 24. https://doi.org/10.1007/978-4-431-55651-0_24.
Wang, Cuicui, et al. “The Therapeutic Effect of Transcranial Magnetic Stimulation on Post-Stroke Aphasia and the Optimal Treatment Parameters: A Meta-Analysis.” Archives of Physical Medicine and Rehabilitation, vol. 105, no. 7, 2024, pp. 1388–1398. https://doi.org/10.1016/j.apmr.2023.11.006.
Wischnewski, Miles, et al. “Real-Time TMS-EEG for Brain State-Controlled Research and Precision Treatment: A Narrative Review and Guide.” Journal of Neural Engineering, vol. 21, no. 6, 2024, article 061001. https://doi.org/10.1088/1741-2552/ad8a8e.
Zhang, Jia Jia Y., et al. “Repetitive Transcranial Magnetic Stimulation for Motor Recovery after Stroke: A Systematic Review and Meta-Analysis of Randomized Controlled Trials with Low Risk of Bias.” Neuromodulation: Technology at the Neural Interface, vol. 28, no. 1, 2025, pp. 16–42. https://doi.org/10.1016/j.neurom.2024.07.010.
