Not two sides of the same coin: Divergent effects of motor state on corticospinal and cortical responses to TMS of motor cortex

25

Transcranial magnetic stimulation (TMS) of the motor cortex can elicit motor evoked potentials 26 (MEPs) in target muscles, reflecting corticospinal excitability. MEP amplitudes increase with TMS intensity and 27 can be facilitated by tonic muscle pre-activation. Since conventional transcranial evoked potentials (TEPs) also 28 grow with increasing TMS intensity, cortical and corticospinal responses are often considered two facets of the 29 same process. If this were true, changes in physiological motor state should modulate TEPs and MEPs in a 30 similar manner. 31

To compare the state-dependency of cortical and corticospinal responses to single -pulse TMS, we 32 simultaneously recorded TEPs and MEPs in 16 healthy young adults during relaxation and isometric 33 contraction of the right first dorsal interosseous (FDI) muscle . For each condition, 100 biphasic TMS pulses 34 were delivered to the left primary motor hand area at five different intensities centered around the resting motor 35 threshold. 36

TEP and MEP amplitudes increased with stimulation intensity. As predicted, tonic muscle contraction 37 consistently facilitated MEP. On the contrary, muscle contraction attenuated two key peaks of the TEP (N15 38 and N100). The state-dependent effects of corticospinal and cortical responses were not correlated. 39

Both TEPs and MEPs are reliably modulated by motor state, yet they differ in direction and their 40 magnitudes do not scale with each other. These findings challenge the assumption that cortical and 41 corticospinal responses are two aspects of the same process. MEP facilitation during contraction likely reflects 42 increased spinal excitability, whereas TEP attenuation may reflect reduced responsiveness of cortico-cortical 43 or cortico-subcortical networks. 44 45 46 47 48 49 .CC-BY-NC-ND 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted September 12, 2025. ; https://doi.org/10.1101/2025.09.07.674734doi: bioRxiv preprint 3

50 Transcranial magnetic stimulation (TMS) uses electromagnetic induction to non-invasively excite neural tissue 51 in the human cortex. The direct neural excitation generated by a single TMS pulse in the cortical target site 52 propagates to connected cortical, subcortical and spinal structures and back causing both local, remote, and 53 reverberant activity [1]. TMS applied over the primary motor hand region (M1 -HAND) at sufficient intensities 54 creates a volley of action potentials that travel down the corticospinal tract and activate spinal motoneurons to 55 produce a motor evoked potential (MEP) in the targeted hand muscle measured via electromyography (EMG). 56 The possibility to combine TMS with electroencephalography (EEG) [2,3] enables a direct characterization of 57 cortical responses to stimulation [3]. The TMS -evoked potential (TEP) consists of a series of positive and 58 negative peaks or deflections in the averaged EEG time-series [4–6] that result from the activation of multiple 59 neural sources with distinct spatiotemporal profiles. 60 While MEPs and TEPs are both measuring physiological responses to TMS , they may represent different 61 neurophysiological phenomena. The MEP is a peripheral motor readout reflecting the excitability and signal 62 propagation along the entire corticospinal pathway. This includes the targeted cortical circuitry in the precentral 63 cortex, the corticospinal projections, the spinal motoneuron pool and possible interneuron relays and the 64 peripheral motor axon [1,7]. Hence, alterations in the state of the corticospinal pathway can alter MEP 65 amplitudes without this necessarily implying an excitability change within the initially stimulated cortical volume. 66 On the other hand, the mechanisms generating the characteristic TEP peaks – N15, P30, N45, P60 and N100– 67 remain unknown. Given their latency, occurring more than 10ms after the TMS pulse, these TEP peaks likely 68 do not reflect the initial activation of principal cells or inhibitory interneurons by the TMS -induced electric field 69 in the motor cortex that underlies MEP generation. Instead, they may reflect reverberations of excitability 70 changes within activated cortico -subcortical circuits. This implies that MEPs and TEPs may not respond in a 71 similar manner to “extrinsic” TMS variables like stimulus intensity, or “intrinsic” brain variables such as changes 72 in brain states. 73 A characteristic feature of MEPs is that their amplitudes increase with increasing TMS intensities and with 74 voluntary pre-activation of the target muscle [8,9]. TEP peak amplitudes have also been shown to increase 75 with stimulus intensity [6,10,11]. The comparable dose-response relationship is compatible with the idea that 76 corticospinal and cortical responses are just different aspects of the same underlying phenomenon. If this was 77 the case, TEPs and MEPs should also share a similar sensitivity to a change in physiological state. A prominent 78 state-dependent feature is a consistent and prominent facilitation of MEP amplitudes during voluntary pre-79 activation of the target muscle relative to the MEP evoked during muscle relaxation and can be largely 80 attributed to spinal mechanisms [12]. 81 This prominent example of state dependency motivated this investigation into whether voluntary motor activity 82 differentially modulates cortical and corticospinal responses to TMS in healthy volunteers. We simultaneously 83 recorded MEPs and TEPs across a range of stimulus intensities during two physiological states, namely while 84 participants were at rest (i.e., relaxed target muscle) and performed a tonic contraction (i .e., steady muscle 85 engagement). The simple state manipulation revealed distinct modulations of specific cortical and corticospinal 86 .CC-BY-NC-ND 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted September 12, 2025. ; https://doi.org/10.1101/2025.09.07.674734doi: bioRxiv preprint 4 responses, indicating that TEPs and MEPs reflect separate, not equivalent, aspects of TMS -induced 87 neurophysiological activity. 88

89 Participants 90 Twenty-eight healthy volunteers (10 males; age range: 23 -32 years; mean age = 26.3 ± 1.9 years) with no 91 history of neurological or psychiatric disorders participated in the experiment after giving informed consent, 92 receiving written information, and being screened for magnetic resonance imaging (MRI) and TMS 93 contraindications. All experimental procedures were carried out in accordance with the Helsinki declaration 94 and were approved by the regional ethical committee (Capital Region of Denmark; Protocol number H -95 15008824). 96 97 Transcranial magnetic stimulation (TMS) 98 Single-pulse TMS was delivered every 2 seconds (jitter ± 0.2 seconds) to the hand representation of the left 99 primary motor region (M1 -HAND) using biphasic pulse waveforms (MagVenture X100 with MagOption, 100 MagVenture A/S, Farum, Denmark) and a 70 -mm figure-of-eight coil (MC-B70 coil, MagVenture A/S, Farum, 101 Denmark). Neuronavigation (TMS Navigator, Localite GmhB, Bonn, Germany) was used to monitor the coil 102 position throughout the experiment using individual T1w structural images (3T MRI, Siemens PRISMA , 103 Erlangen, Germany ). The precentral stimulation site was determined using an individualized mapping 104 procedure (for similar approach, see [13]). First, the precentral motor “hand knob” was anatomically localized 105 [14]. Then, a mini-mapping procedure around this spot was performed to localize the MEP hotspot, i.e., the 106 spot with most consistent and largest motor evoked potentials (MEPs) in the first dorsal interosseus (FDI) 107 muscle. Next, the resting motor threshold (RMT) was estimated defined as the intensity eliciting MEPs >50 µV 108 in 5/10 stimuli [15]. After having identified the individual MEP hotspot and the RMT, EEG data were inspected 109 using the rt-TEP tool [16] to evaluate whether scalp muscle artefacts were triggered. Scalp muscle artefacts 110 following TMS of M1 -HAND are characterized as biphasic responses with topographic extremes in EEG 111 electrodes close to the temporal muscle [13,17]. If scalp muscle responses were observed, the coil was moved 112 or tilted away from the activated temporal muscle. This successfully eliminated scalp muscle artefacts in 16 of 113 the 28 participants. In the remaining 12 participants,minor coil adjustments failed to avoid muscle artifacts, and 114 the experiment was discontinued. 115 Subsequently, TMS was delivered at 5 intensities and during 2 motor states. Per intensity and motor state, 116 100 stimulations were delivered, resulting in 1,000 pulses per participant. Intensities were defined in relation 117 to the RMT by changing the stimulator output in steps of 4% maximum stimulator output (%MSO): -8% MSO; 118 -4% MSO; RMT, +4% MSO; +8% MSO. TMS was delivered while participants were at rest or while they 119 performed a sustained isometric voluntary contraction of the index finger of the right hand at 10% of maximum 120 voluntary contraction (MVC). Feedback was provided in the form of a rectified and smooth EMG trace from the 121 FDI (Figure 1). 122 .CC-BY-NC-ND 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted September 12, 2025. ; https://doi.org/10.1101/2025.09.07.674734doi: bioRxiv preprint 5 123 Figure 1 | Experimental setup and measurements. Single biphasic TMS pulses were delivered at the 124 optimized hot spot of the M1-HAND in the absence of TMS induced cranial muscle artifacts. TMS was given 125 at five intensities while participants were at rest or performed a voluntary isometric contraction of the index 126 finger. The cortical (TEP) and corticospinal responses (MEP) to TMS were recorded. Abbreviations: FDI = 127 First dorsal interosseous; RMT = Resting motor threshold; MVC = Maximal voluntary contraction. RMT = 128 Resting motor threshold. 129 130 During recording blocks, participants were asked to keep their eyes open, avoid blinking, and fixate their gaze 131 at a fixation spot on the screen approximately 1 m in front of them. Participants were equipped with modified 132 earplugs playing a masking sound consisting of pink noise with added TMS clicks. The masking sound was 133 generated using TAAC software [18] and the sound pressure was adjusted so that participants reported not 134 being able to hear the sound of the discharging TMS coil or that it was greatly reduced while the coil was 135 floating over the head of the participants. Additionally, a thin layer of foam (approx. 1-2 mm) was used between 136 the coil and the EEG cap to reduce bone conduction of sound and vibration from the discharging coil [19]. After 137 each recording block, participants were asked to rate their perceived TMS sensations on a numerical rating 138 scale ranging from 0 -10. Specifically, participants rated the audibility of the TMS “click” sound; the sense of 139 focality or “spread” of the TMS pulse on the head; the vibrations caused by the discharge of the coil; and the 140 pressure of the coil on the head. 141 142 Electroencephalographic (EEG) and electromyographic (EMG) recordings 143 EEG activity was recorded at a sampling rate of 5 kHz from 61 passive Ag/AgCl C-slit electrodes placed in an 144 equidistant EEG cap (M10 cap layout, BrainCap TMS, Brain Products GmbH, Germany) using a TMS -145 .CC-BY-NC-ND 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted September 12, 2025. ; https://doi.org/10.1101/2025.09.07.674734doi: bioRxiv preprint 6 compatible EEG amplifier (BrainAmp DC, Brain Products GmbH, Germany) and BrainVision Recorder 146 software. The reference and ground electrodes were placed on the right and left side of the participants’ 147 forehead, respectively. Electrodes were prepared using electroconductive and abrasive gel to lower the 148 electrode impedance (<5 kΩ). Electrode impedance levels were checked regularly throughout the experiment. 149 EMG was acquired from the right first dorsal interosseus (FDI) using a belly -tendon montage with the ground 150 electrode positioned on the right processus styloideus ulnae. EMG signals were amplified (x500), bandpass 151 filtered (10-2000 Hz), and sampled at 2000 Hz (Digitimer D360, Digitimer Ltd., Hertfordshire, UK) using Signal 152 software (v4.17, Cambridge Electronic Design, Cambridge, UK). 153 154 EEG preprocessing 155 EEG preprocessing was carried out in EEGLAB (v. 2021) running in MATLAB (v. R2020a) using functions from 156 the TMS-EEG signal analyser (TESA)[20] toolbox. First, the TMS pulse artifact was removed by replacing the 157 signal from -2 to 8 ms around the stimulation with interpolated data points using cubic interpolation. 158 Subsequently, EEG data was band-pass filtered from 1-48 Hz using a 2nd order Butterworth filter and epoched 159 from -500 to 500 ms around stimulation. Following this, data were visually inspected and electrodes displaying 160 excessive noise were removed (median of 2 channels excluded per participant). Missing channels were 161 interpolated using spherical interpolation. Trials containing eye blinks or excessive noise were removed based 162 on visual inspection of signal amplitudes and topographies (median of 11 epochs excluded per condition). 163 Baseline correction was applied from -205 to -5 ms to the stimulation. 164 165 EEG data analysis 166 Two measures were computed to quantify TMS-evoked cortical activity. Global mean field power (GMFP) [21] 167 was computed to extract a single reference -free cortical response measure. For the analysis of TEP peak 168 amplitudes, we focused on data from a single electrode close to site of stimulation characterized as one of the 169 most responsive when scrutinizing the TMS-evoked EEG topographies (Fig 2B). Peaks were estimated as the 170 largest absolute value within pre -specified time windows for the prototypical TEP peaks following stimulation 171 of M1 [4,6,22], N15 (10-20 ms), P30 (15-45 ms), N45 (30-60 ms), P60 (45-75 ms) and N100 (75-125 ms). EEG 172 spectral power in the time-period leading up to stimulation (-500 to -10 ms) was also computed using Welch’s 173

during both rest and contraction . This was done to obtain an electrophysiological validation of our 174 state-modulation, as desynchronization of alpha and beta power is a hallmark EEG feature accompanied by 175 motor activity [23], and to evaluate whether individual modulations of alpha and beta power during voluntary 176 sustained contractions were associated with individual differences in TEP and MEP amplitudes. We computed 177 the area under the power spectrum curve for the alpha-band (8-12 Hz) and beta-band (15-35 Hz) to estimate 178 modulation of band-limited power during movement execution. 179 180 Statistical analysis 181 Two separate linear mixed effect models were used to model the effect s of stimulation conditions for the 182 extracted MEP and TEP amplitudes, respectively. 183 .CC-BY-NC-ND 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted September 12, 2025. ; https://doi.org/10.1101/2025.09.07.674734doi: bioRxiv preprint 7 184 For TEPs, a mixed model was fitted to the extracted amplitudes with an interaction between three independent 185 factors: peak (5 levels: N15, P30, N45, P60, N100); intensity (5 levels: -8%, -4%, RMT, +4%, +8%) and motor 186 state (2 levels: rest vs. contraction). Individuals (ID) were added to the model as random intercepts: 187 188 TEP amplitude ~ Peak x Intensity x Motor State + (1|ID) (1) 189 190 Where x represents an interaction and (1|ID) represents the random intercepts for each participant. 191 A similar model was fitted for the MEP amplitudes, but without the interaction with Peak. Furthermore, MEPs 192 were log-transformed to adhere to assumptions of normality of residuals. 193 194 log(MEP amplitude) ~ Intensity x Motor State + (1|ID) (2) 195 196 Mixed effect models were fitted in R using the lme4-package [24]. F-statistics and p -values for main effects 197 and interactions were estimated using the lmerTest-package [25] and post-hoc comparisons were performed 198 using the emmeans-package [26]. In addition, Pearson correlation analyses were performed to investigate 199 potential associations between state-dependent modulations of TEPs, MEPs, and alpha/beta power. P-values 200 were adjusted for multiple comparisons using the Bonferroni correction unless otherwise stated. For all 201 analyses, alpha was set to 0.05. 202 203

204 205 Impact of stimulation intensity of MEP and TEP peak amplitudes 206 TMS over M1-HAND evoked MEPs at threshold and suprathreshold intensities at rest, while TMS also evoked 207 MEPs at subthreshold intensities during voluntary contraction. MEP amplitudes increased with stimulation 208 intensity during both rest and contraction (Figure 2A). This was reflected by a significant main effect of 209 stimulation intensity in the fitted mixed effect model (F = 105.1; P < 0.001). GMFPs revealed peaks at various 210 time points during both rest and contraction (Figure 2B). The topographical maxima of activation were located 211 close to the site of stimulation in the left sensorimotor cortex. When comparing the TEP waveforms extracted 212 from electrodes close to this area, a general increase in peak amplitudes were observed with stimulation 213 (Figure 2C). This was confirmed statistically by comparing the extracted peak amplitudes revealing an 214 interaction between stimulation intensity and peak (F = 31.4; P < 0.001) (Figure 2D). Although all peak 215 amplitudes increased with intensity, the interaction indicated that intensity had a differential effect across 216 different TEP peaks (Figure 2D). 217 218 219 .CC-BY-NC-ND 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted September 12, 2025. ; https://doi.org/10.1101/2025.09.07.674734doi: bioRxiv preprint 8 220 221 Figure 2 | Motor evoked potentials (MEPs) and transcranial evoked potentials (TEPs) increase with stimulation 222 intensity. (A) Effects of stimulation intensity on MEP amplitudes. (B) Average global mean field power (GMFP) and EEG 223 topographies during rest and contraction. (C) Averaged local TEP waveforms across stimulation intensities during rest and 224 contraction. (D) Averaged amplitudes for prototypical TEP peaks across stimulation intensities during rest and contraction. 225 .CC-BY-NC-ND 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted September 12, 2025. ; https://doi.org/10.1101/2025.09.07.674734doi: bioRxiv preprint 9 Impact of voluntary motor activity on MEP and TEP peak amplitudes 226 As expected, MEP peak-to-peak amplitudes were consistently greater during sustained voluntary contraction 227 compared to rest (Figure 3A). The fitted mixed effect model revealed a significant main effect of state for MEP 228 amplitudes (F = 401.8; P < 0.001). In contrast, GMFP of the TMS evoked EEG response was not enhanced 229 by isometric tonic contraction of the target muscle. Two components of the TEP were actually smaller during 230 contraction than at rest (Figure 2B), namely the negative peaks around 15 ms (N15) and 100 ms (N100) after 231 the pulse. The linear mixed model showed a significant interaction between the motor states and the extracted 232 TEP peak amplitudes (F = 4.04; P = 0.001). Post-hoc comparisons confirmed that the state -dependent 233 differences in TEP amplitudes were driven by an attenuation (i.e., less negativity) of the N15 peak ( βrest vs. 234 contraction= -2.40 ± 0.61 µV; t = 4.62; P < 0.001) and N100 peak ( βrest vs. contraction= -1.58 ± 0.61 µV; t = 3.28; P = 235 0.01). No significant differences were observed for other peaks of interest (p-values range: 0.54 -0.89). The 236 state-dependent difference in N15 and N100 amplitudes did not depend on stimulation intensity as 237 demonstrated from the lack of a three -way interaction between intensity, state, and peak (F = 0.67; P = 0.83) 238 (Figure 3A). Together, these results show that MEP amplitudes increased during voluntary contraction, 239 whereas amplitudes of the N15 and N100 TEP peaks decreased during voluntary contraction compared to 240 muscle relaxation. 241 242 To investigate potential associations between individual state -dependent changes in TEP peaks and MEP 243 amplitudes across stimulation intensities a Pearson correlation matrix was computed. Significant correlations 244 were observed between the individual differences of the state difference in the N15 and N100 amplitude ( r = 245 0.56; P = 0.02), the N45 and P60 amplitude (r = 0.74; P = 0.002), and the N45 and MEP amplitude (r = -0.54, 246 P = 0.03). These correlations are illustrated in Figure 3B. 247 248 .CC-BY-NC-ND 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted September 12, 2025. ; https://doi.org/10.1101/2025.09.07.674734doi: bioRxiv preprint 10 249 Figure 3 | Effects of motor state on motor evoked potentials (MEPs) and transcranial evoked potentials (TEPs). (A) 250 Amplitudes of MEPs (left) and prototypical local TEP peaks (right) during rest (red) and contraction (blue) averaged across 251 stimulation intensities. MEP amplitudes were consistently greater during contraction compared to rest, whereas amplitudes 252 of the N15 and N100 peaks were consistently smaller during contraction compared to rest. (B) Correlations between 253 individual differences in TEP peak amplitudes between rest and contraction. Significant correlations were observed 254 between the differences in the N15 and N100 amplitudes, between the N45 and P60 amplitudes and between the N45 and 255 MEP amplitudes. P-values are corrected for the false-discovery rate. 256 257 Pre-stimulus EEG power in the alpha (8-13 Hz) and beta (15-35 Hz) frequency bands were also modulated by 258 motor state (Figure 4A). A significant desynchronization of the alpha (t(31) = 4.47; P<0.001) and beta bands 259 (t(31) =2.86; P=0.007) occurred during voluntary contraction compared to rest. This desynchronization was 260 most apparent in the electrodes covering the bilateral sensorimotor cortices (Figure 4A). This indicates that 261 our behavioral manipulation effectively altered sensorimotor brain activity. No significant correlations were 262 .CC-BY-NC-ND 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted September 12, 2025. ; https://doi.org/10.1101/2025.09.07.674734doi: bioRxiv preprint 11 observed between pre-stimulus power changes in the alpha or beta EEG bands, extracted from the electrode 263 closest to the stimulated target region, and changes in TEP peak amplitudes for the N15 and N100 peak across 264 stimulation intensities (Figure 4B). Correlation coefficients and p-values ranged from 0.027-0.35 and 0.18-0.92 265 (uncorrected), respectively. 266 267 268 269 Figure 4 | Effects of motor state on pre-stimulus alpha and beta power and associations to TEPs and MEPs. (A) 270 Power spectrum estimated from electrode overlying the stimulated sensorimotor cortex during the pre -stimulus period. 271 Differences in alpha (middle) and beta (right) power during rest vs. contraction and corresponding topo graphical plots 272 showing differences in band-limited power across states across the scalp. (B) Individual differences in alpha and beta 273 power were not associated with individual state-differences in N15, N100 or MEP amplitudes. 274 275 .CC-BY-NC-ND 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted September 12, 2025. ; https://doi.org/10.1101/2025.09.07.674734doi: bioRxiv preprint 12

276 This study directly compared the effects of tonic muscle contraction on two distinct responses to single -pulse 277 TMS of the M1-HAND region. Across multiple stimulation intensities, we simultaneously recorded corticospinal 278 motor evoked potentials (MEPs), evoked by transsynaptic activation of the corticomotor pathway, and cortical 279 transcranial evoked potentials (TEPs), reflecting the brain's direct response to TMS. Both responses scaled 280 with stimulation intensity but were differentially modulated by voluntary muscle activation. MEP amplitudes 281 were consistently enhanced during tonic contraction, whereas the early (N15) and late (N100) peaks of the 282 TEP were reliably attenuated. These findings demonstrate that TEPs and MEPs capture distinct facets of state-283 dependent TMS-induced activation of the motor system. 284 Effect of isometric tonic contraction on the transcranially evoked response 285 Alterations in the brain's functional state can markedly influence its responsiveness to TMS. A prominent 286 example is the transition from wakefulness to slow-wave sleep, causing pronounced changes in the 287 spatiotemporal dynamics of the TEP [27–29]. In contrast, the manipulation of motor state investigated in the 288 present study elicited only modest, yet highly consistent, changes in the TEP magnitudes. Specifically, the 289 transitioning from rest to voluntary isometric contraction reduced the magnitude of the evoked responses but 290 not its spatiotemporal profile. Notably, TMS administered during the active motor state reliably attenuated the 291 N15 and N100 TEP peak amplitudes across a broad range of stimulation intensities. This reduction was 292 consistently observed across participants, underscoring the robustness of the effect. 293 The precise neural generators of the N15 and N100 TEP peaks and the mechanisms rendering these peaks 294 sensitive to changes in motor state remain to be delineated [5]. Based on their latency, it can be excluded that 295 the N15 and N100 peaks reflect the initial direct inductive activation of local cortical neurons responsible for 296 generating the MEPs [30]. Instead, recent evidence suggests that immediate TEPs (i TEP), which emerge 297 within 2-8 ms following the TMS pulse [13], may serve as a more direct indicator of immediate TMS-induced 298 cortical activation [13]. In the present study, we were unable to assess iTEPs due to the presence of 299 stimulation-related artifacts within the first 5 -6 ms after the TMS pulse , due to the sampling rate of 5 kHz. 300 Emerging evidence suggests that both early TEP peaks (e.g., N15) and late TEP peaks (e.g., N100) may 301 instead reflect reverberant activity within cortico-thalamic-cortical circuits triggered by the TMS pulse [31–33]. 302 Studies in mice and humans comparing electrical and magnetic stimulation during rest and movement have 303 shown a consistent suppression of both early and late response amplitudes during movement [31,32]. The 304 authors also performed a detailed mapping of the network responsible for mediating these effects. In mice, 305 stimulation evoked early spiking in cortico -thalamic projection neurons, followed by sensorimotor thalamic 306 activity, with this relationship reversing during later time windows [32]. Furthermore, optogenetic inhibition of 307 sensorimotor thalamic neurons altered the cortical response profile [31]. Together, these findings support the 308 notion that thalamic-cortical feedback is involved in shaping the TEP in a state -dependent fashion. While the 309 precise timing and definition of the peaks in those studies differ from the ones reported in the present work, 310 the cross-species characterization of movement -modulated circuits provides valuable insights into state-311 dependent modulation. We propose that the attenuation of N15 and N100 amplitudes by voluntary contraction 312 .CC-BY-NC-ND 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted September 12, 2025. ; https://doi.org/10.1101/2025.09.07.674734doi: bioRxiv preprint 13 may reflect the dynamic engagement of sensorimotor cortico-thalamic-cortical circuits. These TEP peaks may 313 serve as sensitive markers of motor state transitions and potentially offer novel insights into motor dysfunction 314 in neurological disorders. Further investigation is warranted to clarify their underlying mechanisms and clinical 315 relevance. 316 State-dependent effects of voluntary muscle activation on MEPs and TEPs 317 While MEP amplitudes were facilitated by voluntary contraction [34], the N15 and N100 peak amplitudes of 318 the TEP were suppressed during tonic isometric contractions compared to rest. Preparation and execution of 319 voluntary movement is characterized by a desynchronization (i.e., decrease of power) of cortical network 320 activity at alpha (8-13 Hz) and beta (15-35 Hz) frequencies [23,35], and an increase in the excitability of spinal 321 motoneurons. Cortical desynchronization likely reflects a transition from a synchronized oscillating state 322 towards a state of more asynchronous neural processing that is needed to generate and maintain a tonic level 323 of motor output. The greater MEPs following TMS of M1 during low intensity sustained voluntary contractions 324 can be attributed to increased excitability of alpha motor neurons in the spinal cord [36]. Although generally 325 representing a more active cortical state than rest, the evoked N15 and N100 TEP peaks were reduced during 326 periods of isometric voluntary contraction relative to rest. This may be related to intracortical inhibition, primarily 327 mediated by GABAergic interneurons, dampening excitatory post -synaptic activity. A potential contribution of 328 cortical inhibitory circuits to the N15 and N100 peaks is supported by both pharmacological and paired pulse 329 TMS studies [37–39]. Of note, a relative attenuation of the N100 peak has been found during both movement 330 preparation [40–42] and execution [43,44]. The present results confirm and extend these N100 findings by 331 showing that tonic isometric contraction also reduces the amplitudes of the early N15 peak. Together, it can 332 be concluded that both early and late cortical responses to TMS are sensitive to the level of motor activity and 333 attenuated by tonic voluntary motor activity. Since contraction -related attenuation of the N15 and N100 334 amplitude correlated in our participants, the mechanisms mediating state-dependent modulation of these two 335 peaks may be at least partly overlapping. Alternatively, the attenuation could be generated by distinct 336 mechanisms that are influenced to a comparable extent by tonic contraction. 337 It is worth noting that the interval between the N15 and N100 peaks roughly aligns with the alpha frequency 338 band (8-13 Hz). In our study , voluntary tonic contraction of an intrinsic hand muscle led to a reduction in 339 sensorimotor EEG power in both, the alpha and beta bands. However, individual decreases in alpha and beta 340 power ipsilateral to stimulation did not correlate with the attenuation of N15 and N100 TEP components, nor 341 with the individual facilitation of MEP amplitudes during tonic contraction. This lack of correlation suggests that 342 individual changes in task -related regional cortical alpha and beta power do not account for the observed 343 modulation of TEPs or MEPs. This finding is particularly relevant in light of recent work using combined EEG-344 TMS approaches to examine how oscillatory brain activity shapes TEP and MEP responses at rest. While 345 several EEG-informed TMS-MEP studies have shown that MEP amplitudes are influenced by the power and 346 phase of the pericentral alpha rhythm [45–48], similar effects have not been consistently demonstrated for 347 TEPs [49,50]. 348 .CC-BY-NC-ND 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted September 12, 2025. ; https://doi.org/10.1101/2025.09.07.674734doi: bioRxiv preprint 14 Dose dependency of MEP and TEP responses 349 MEPs were only elicited at threshold and suprathreshold intensities during rest, while MEPs were observed at 350 all stimulation intensities during voluntary contraction, including the two intensities that were below the resting 351 motor threshold. The latter suggest that the entire intensity range used in this study was above the threshold 352 for evoking activity in pyramidal tract neurons. Accordingly, TMS over M1-HAND elicited EEG responses close 353 to site of stimulation at all intensity levels, even in the subthreshold resting state conditions, evoking a well -354 described sequence of TEP peaks with latencies at 15, 30, 45, 60 and 100 ms [4,22,51]. While some peaks 355 were clearly discernible, others were more difficult to disentangle such as the N45 and P60, as they were 356 superimposed on a positive wave [52]. This effect seemed more pronounced following suprathreshold 357 stimulation compared to subthreshold stimulation and likely also contribute to explaining the correlation of the 358 state-dependent modulation between the two. 359 Increasing the intensity of TMS enhanced the response magnitude for both MEPs [34] and TEPs [10]. Stimulus 360 intensities scaled positively with MEP and TEP amplitudes at rest and during a sustained voluntary contraction 361 due to a more efficient excitation of cortical neurons. The excitation of a larger population of pyramidal cells at 362 higher stimulus intensities should translate into larger amplitudes of both MEPs and TEPs. For TEPs, other 363 non-transcranial stimulation effects become more pronounced at higher stimulation intensities. For example, 364 the TMS-evoked muscle twitches in the contralateral hand or forearm muscles increase with stimulus intensity, 365 causing stronger re-afferent somatosensory feedback that may increase middle-to-late parts of the TEPs (from 366 approx. 40ms and onwards) [53,54]. Stronger stimulation intensities also produce greater mechanical 367 vibrations in the coil casing which produce larger “click” sounds [55] that are more easily perceived by 368 participants, as also reported in Table 1. Multimodal sensory inputs likely become more relevant at higher TMS 369 intensities and contribute to the stimulus -response relationship of TEP components that occur at latencies 370 above 40 ms through peripherally-induced cortical co-activation if not masked sufficiently [11,56–59]. 371 Table 1 | Sensory perception of TMS pulses. 372 Loudness of pulse “click”# Scalp “spread” of pulse Coil pressure Coil vibration Rest Contraction Rest Contraction Rest Contraction Rest Contraction -8% MSO 1.46 ± 1.76 1.19 ± 1.72 5.87 ± 2.72 6.20 ± 2.65 1.87 ± 1.76 1.50 ± 1.26 0.13 ± 0.35 0.13 ± 0.34 -4% MSO 1.31 ± 2.09 1.50 ± 2.03 5.94 ± 3.02 6.31 ± 2.21 1.56 ± 1.36 1.69 ± 1.74 0.13 ± 0.34 0.31 ± 0.60 RMT 1.63 ± 1.93 1.69 ± 1.78 6.06 ± 2.67 6.31 ± 2.57 1.75 ± 1.53 1.63 ± 1.31 0.13 ± 0.34 0.25 ± 0.58 +4% MSO 1.86 ± 1.63 2.07 ± 1.65 6.06 ± 2.46 6.56 ± 2.25 1.56 ± 1.21 1.94 ± 1.39 0.31 ± 0.79 0.50 ± 1.21 +8% MSO 2.20 ± 2.57 2.80 ± 2.37 6.02 ± 2.48 6.40 ± 2.59 1.93 ± 1.79 1.87 ± 1.92 0.13 ± 0.35 0.33 ± 0.82 Sensations reported by the participants using a numerical rating scale (0 -10). For loudness, 0 corresponded 373 to not hearing the click sound, while 10 represented a very loud sensation. For focality, 0 represented a very 374 focal sensation (i.e., “sharp”) and 10 very diffuse (i.e., “blunt”). For coil pressure, 0 represented no pressure at 375 all and 10 very severe pressure. For coil vibration, 0 represented no vibration at all and 10 very intense 376 vibration. Data reported as means and standard deviations. No significant differences between motor states 377 were observed between subjective perceptions of auditory or somatosensory stimuli. # represents the 378 significant effect of stimulation intensity across motor states for perceived loudness of the TMS coil “click”. 379 380 .CC-BY-NC-ND 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted September 12, 2025. ; https://doi.org/10.1101/2025.09.07.674734doi: bioRxiv preprint 15 Methodological considerations 381 Recording TEPs after stimulation of M1 -HAND can be methodological challenging as the coil is placed on 382 lateral aspects of the head in proximity to scalp muscles (e.g,, the temporal muscle ) and the nerves that 383 innervate them [1]. This increases the risk of coactivating scalp muscles resulting in large amplitude compound 384 muscle action potentials that can obscure the EEG signal in the tens of milliseconds following stimulation 385 [5,17]. Higher stimulation intensities are more likely to cause scalp muscle activation which further complicates 386 mapping input-output relationships. By individualizing coil position through minor tilts and coil movements away 387 from the temporal muscle [17] while concurrently visualizing the corresponding EEG data [16], we managed 388 to acquire scalp muscle artifact free data in 16 out of screened 28 participants at a range of different intensities. 389 In these individuals, we obtained short latency TEP responses (earliest peak at ~15 ms) with amplitudes that 390 resemble those seen after individualized and optimized targeting of peri -sagittal cortical areas (containing no 391 muscles) [60]. 392 We argue that our TEP responses were not confounded by scalp muscle activations. First, the amplitudes and 393 topographical distribution do not resemble those of prototypical TMS -evoked scalp muscle artifacts [5,17,61]. 394 Second, the fact that the N15 response was consistently modulated by the behavioral state of participants 395 indicates that it is unlikely to represent scalp muscle artifacts which should not be altered by the central motor 396 state manipulation. 397 In the present study, we compared TEPs during rest and sustained motor behavior. It is well established that 398 sensory perception and attention are affected during motor activities [62–65]. For example, auditory evoked 399 cortical potentials are smaller during periods of movement compared to rest [64,65]. In the present study, we 400 used active noise masking to minimize the influence of auditory evoked potentials on the TEP [18,19]. 401 Furthermore, no differences were observed in the subjective perception of any of the assessed sensory stimuli 402 caused by TMS, including coil clicks. However, as we were not successful in completely masking the TMS -403 evoked click sound in all participants at all intensities, we cannot entirely exclude that the modulations of TEP 404 peaks could reflect gating of sensory inputs caused by TMS (peripheral evoked potentials) or shifts in attention 405 rather than a modulation of the transcranial constituent of the TEP per se. Peripherally evoked potentials 406 primarily influence later TEP peaks [11,56–59]. Therefore, contributions of sensory gating of these potentials 407 may be more relevant for the modulation of later (N100) rather than earlier (N15) TEP peaks. 408

409 MEPs and TEPs are commonly used readouts of target engagement following stimulation of M1 and reflect 410 the combined effect of the stimulation dose and the ongoing brain state . Here we show that the cortical 411 responses (TEPs) and the cortico -motor pathway responses (MEPs) diverge when the physiological state 412 switches from relaxation to tonic activation. The marked differences in state dependency show that TEPs and 413 MEPs are distinct and do not offer two perspectives on the same physiological process . The state sensitivity 414 of the classical TEP peaks (N15 and N100) may be a useful cortical probe of cerebral sensorimotor network 415 dynamics in healthy individuals and patients. 416 .CC-BY-NC-ND 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted September 12, 2025. ; https://doi.org/10.1101/2025.09.07.674734doi: bioRxiv preprint 16

417 This study is funded by the Innovation Fund Denmark (IFD) for the Grand Solution project “PRECISION-BCT” 418 (Grant number: 9068-00025B) and the project “ADAptive and Precise Targeting of cortex-basal ganglia circuits 419 in Parkinson´s Disease - ADAPT-PD” from The Lundbeck Foundation (collaborative project grant, grant n r. 420 R336-2020-1035). Mikkel M. Beck is funded by a grant from the Capital Region of Denmark (Region 421 Hovedstaden) and a post doc grant from The Lundbeck Foundation (grant nr. R449-2023-1487). Sybren Van 422 Hoornweder is funded by two grants from the Research Foundation Flanders (FWO) (Grant nr. G1129923N 423 and nr. V426023N). Leo Tomasevic was partially funded by the dtec.bw – Digitalization and Technology 424 Research Center of the Bundeswehr [MEXT project]. The dtec.bw is funded by the European Union – 425 NextGenerationEU. 426 427 Disclosures: 428 M.M Beck; None. L. Christiansen; None. M. Heyl; None. A. Mastropasqua; None. S. Van Hoornweder; 429 None. A. Thielscher; None. L. Tomasevic; None. H.R. Siebner; Has received honoraria as speaker from 430 Sanofi Genzyme, Denmark, Lundbeck AS, Denmark, and Novartis, Denmark, as consultant from Sanofi 431 Genzyme, Denmark, Lophora, Denmark, and Lundbeck AS, Denmark, and as editor -in-chief (Neuroimage 432 Clinical) and senior editor (NeuroImage) from Elsevier Publishers, Amsterdam, The Netherlands. He has 433 received royalties as book editor from Springer Publishers, Stuttgart, Germany and from Gyldendal Publishers, 434 Copenhagen, Denmark. 435 436 437 .CC-BY-NC-ND 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted September 12, 2025. ; https://doi.org/10.1101/2025.09.07.674734doi: bioRxiv preprint 17

438 [1] Siebner HR, Funke K, Aberra AS, Antal A, Chen R, Classen J, et al. Transcranial magnetic stimulation 439 of the brain: What is stimulated? – a consensus and critical position paper. Clin Neurophysiol 2022. 440 https://doi.org/10.1016/J.CLINPH.2022.04.022. 441 [2] Cracco RQ, Amassian VE, Maccabee PJ, Cracco JB. Comparison of human transcallosal responses 442 evoked by magnetic coil and electrical stimulation. Electroencephalogr Clin Neurophysiol Evoked 443 Potentials 1989;74:417–24. https://doi.org/10.1016/0168-5597(89)90030-0. 444 [3] Ilmoniemi RJ, Virtanen J, Ruohonen J, Karhu J, Aronen HJ, Näätänen R, et al. Neuronal responses to 445 magnetic stimulation reveal cortical reactivity and connectivity. Neuroreport 1997;8:3537 –40. 446 https://doi.org/10.1097/00001756-199711100-00024. 447 [4] Farzan F, Bortoletto M. Identification and verification of a “true” TMS evoked potential in TMS -EEG. J 448 Neurosci Methods 2022;378:109651. https://doi.org/10.1016/J.JNEUMETH.2022.109651. 449 [5] Hernandez-Pavon JC, Veniero D, Bergmann TO, Belardinelli P, Bortoletto M, Casarotto S, et al. TMS 450 combined with EEG: Recommendations and open issues for data collection and analysis. Brain Stimul 451 2023;16:567–93. https://doi.org/10.1016/j.brs.2023.02.009. 452 [6] Beck MM, Heyl M, Mejer L, Vinding MC, Christiansen L, Tomasevic L, et al. Methodological choices 453 matter: A systematic comparison of TMS -EEG studies targeting the primary motor cortex. Hum Brain 454 Mapp 2024;30:2024.03.08.583873. https://doi.org/10.1002/HBM.70048. 455 [7] Bestmann S, Krakauer JW. The uses and interpretations of the motor -evoked potential for 456 understanding behaviour. Exp Brain Res 2015;233:679–89. https://doi.org/10.1007/s00221-014-4183-457 7. 458 [8] Siebner HR, Funke K, Aberra AS, Antal A, Bestmann S, Chen R, et al. Transcranial magnetic 459 stimulation of the brain: What is stimulated? – A consensus and critical position paper. Clin 460 Neurophysiol 2022;140:59–97. https://doi.org/10.1016/J.CLINPH.2022.04.022. 461 [9] Groppa S, Oliviero A, Eisen A, Quartarone A, Cohen LG, Mall V, et al. A practical guide to diagnostic 462 transcranial magnetic stimulation: Report of an IFCN committee. Clin Neurophysiol 2012;123:858 –82. 463 https://doi.org/10.1016/j.clinph.2012.01.010. 464 [10] Komssi S, Kähkönen S, Ilmoniemi RJ. The Effect of Stimulus Intensity on Brain Responses Evoked by 465 Transcranial Magnetic Stimulation. Hum Brain Mapp 2004;21:154 –64. 466 https://doi.org/10.1002/hbm.10159. 467 [11] Biabani M, Fornito A, Mutanen TP, Morrow J, Rogasch NC. Characterizing and minimizing the 468 .CC-BY-NC-ND 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted September 12, 2025. ; https://doi.org/10.1101/2025.09.07.674734doi: bioRxiv preprint 18 contribution of sensory inputs to TMS -evoked potentials. Brain Stimul 2019;12:1537 –52. 469 https://doi.org/10.1016/j.brs.2019.07.009. 470 [12] Di Lazzaro V, Restuccia D, Oliviero A, Profice P, Ferrara L, Insola A, et al. Effects of voluntary 471 contraction on descending volleys evoked by transcranial stimulation in conscious humans. J Physiol 472 1998;508:625–33. https://doi.org/10.1111/J.1469-7793.1998.625BQ.X. 473 [13] Beck MM, Christiansen L, Madsen MAJ, Jadidi AF, Vinding MC, Thielscher A, et al. Transcranial 474 magnetic stimulation of primary motor cortex elicits an immediate transcranial evoked potential. Brain 475 Stimul 2024;17:802–12. https://doi.org/10.1016/j.brs.2024.06.008. 476 [14] Yousry TA, Schmid UD, Alkadhi H, Schmidt D, Peraud A, Buettner A, et al. Localization of the motor 477 hand area to a knob on the precentral gyrus. A new landmark. Brain 1997;120:141 –57. 478 https://doi.org/10.1093/brain/120.1.141. 479 [15] Rossini PM, Burke D, Chen R, Cohen LG, Daskalakis Z, Di Iorio R, et al. Non -invasive electrical and 480 magnetic stimulation of the brain, spinal cord, roots and peripheral nerves: Basic principles and 481 procedures for routine clinical and research application: An updated report from an I.F.C.N. Committee. 482 Clin Neurophysiol 2015;126:1071–107. https://doi.org/10.1016/j.clinph.2015.02.001. 483 [16] Casarotto S, Fecchio M, Rosanova M, Varone G, D’Ambrosio S, Sarasso S, et al. The rt-TEP tool: real-484 time visualization of TMS -Evoked Potentials to maximize cortical activation and minimize artifacts. J 485 Neurosci Methods 2022;370:109486. https://doi.org/10.1016/J.JNEUMETH.2022.109486. 486 [17] Mutanen T, Mäki H, Ilmoniemi RJ. The effect of stimulus parameters on {TMS}-{EEG} muscle artifacts. 487 Brain Stimul 2013;6:371–6. https://doi.org/10.1016/j.brs.2012.07.005. 488 [18] Russo S, Sarasso S, Puglisi GE, Dal Pal ù D, Pigorini A, Casarotto S, et al. TAAC - TMS Adaptable 489 Auditory Control: A universal tool to mask TMS clicks. J Neurosci Methods 2022;370:109491. 490 https://doi.org/10.1016/J.JNEUMETH.2022.109491. 491 [19] ter Braack EM, de Vos CC, van Putten MJAM. Masking the Auditory Evoked Potential in TMS -EEG: A 492 Comparison of Various Methods. BRAIN Topogr 2015;28:520 –8. https://doi.org/10.1007/s10548-013-493 0312-z. 494 [20] Rogasch NC, Sullivan C, Thomson RH, Rose NS, Bailey NW, Fitzgerald PB, et al. Analysing concurrent 495 transcranial magnetic stimulation and electroencephalographic data: A review and introduction to the 496 open-source TESA software. Neuroimage 2017;147:934 –51. 497 https://doi.org/10.1016/j.neuroimage.2016.10.031. 498 [21] Lehmann D, Skrandies W. Reference -free identification of components of checkerboard -evoked 499 multichannel potential fields. Electroencephalogr Clin Neurophysiol 1980;48:609 –21. 500 .CC-BY-NC-ND 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted September 12, 2025. ; https://doi.org/10.1101/2025.09.07.674734doi: bioRxiv preprint 19 https://doi.org/10.1016/0013-4694(80)90419-8. 501 [22] Ilmoniemi RJ, Ki čić D. Methodology for combined TMS and EEG. Brain Topogr 2010;22:233 –48. 502 https://doi.org/10.1007/s10548-009-0123-4. 503 [23] Pfurtscheller G, Lopes Da Silva FH. Event -related EEG/MEG synchronization and desynchronization: 504 Basic principles. Clin Neurophysiol 1999;110:1842–57. https://doi.org/10.1016/S1388-2457(99)00141-505 8. 506 [24] Bates D, M ächler M, Bolker BM, Walker SC. Fitting linear mixed -effects models using lme4. 507 ArXiv:14065823v1[StatCO]23 2014:1–51. https://doi.org/10.1177/009286150103500418. 508 [25] Kuznetsova A, Brockhoff PB, Christensen RHB. lmerTest Package: Tests in Linear Mixed Effects 509 Models . J Stat Softw 2017;82:1–26. https://doi.org/10.18637/jss.v082.i13. 510 [26] Lenth R. Emmeans: estimated marginal means. Aka Least -Squares Means 2019:https://cran.r -511 project.org/package=emmeans. https://doi.org/10.32614/CRAN.PACKAGE.EMMEANS. 512 [27] Massimini M, Ferrarelli F, Huber R, Esser SK, Singh H, Tononi G. Breakdown of cortical effective 513 connectivity during sleep. Science (80- ) 2005;309:2228–32. https://doi.org/10.1126/science.1117256. 514 [28] Bergmann TO, M ölle M, Schmidt MA, Lindner C, Marshall L, Born J, et al. EEG -guided transcranial 515 magnetic stimulation reveals rapid shifts in motor cortical excitability during the human sleep slow 516 oscillation. J Neurosci 2012;32:243–53. https://doi.org/10.1523/JNEUROSCI.4792-11.2012. 517 [29] Massimini M, Ferrarelli F, Esser SK, Riedner BA, Huber R, Murphy M, et al. Triggering sleep slow 518 waves by transcranial magnetic stimulation. Proc Natl Acad Sci U S A 2007;104:8496 –501. 519 https://doi.org/10.1073/PNAS.0702495104/SUPPL_FILE/02495FIG7.PDF. 520 [30] Mueller JK, Grigsby EM, Prevosto V, Petraglia FW, Rao H, Deng Z De, et al. Simultaneous transcranial 521 magnetic stimulation and single -neuron recording in alert non -human primates. Nat Neurosci 522 2014;17:1130–6. https://doi.org/10.1038/nn.3751. 523 [31] Russo S, Claar LD, Furregoni G, Marks LC, Krishnan G, Zauli FM, et al. Thalamic feedback shapes 524 brain responses evoked by cortical stimulation in mice and humans. Nat Commun 2025;16:1 –19. 525 https://doi.org/10.1038/s41467-025-58717-2. 526 [32] Claar LD, Rembado I, Kuyat JR, Russo S, Marks LC, Olsen SR, et al. Cortico -thalamo-cortical 527 interactions modulate electrically evoked EEG responses in mice. Elife 2023;12. 528 https://doi.org/10.7554/eLife.84630. 529 [33] Casarotto S, Turco F, Comanducci A, Perretti A, Marotta G, Pezzoli G, et al. Excitability of the 530 .CC-BY-NC-ND 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted September 12, 2025. ; https://doi.org/10.1101/2025.09.07.674734doi: bioRxiv preprint 20 supplementary motor area in Parkinson ’s disease depends on subcortical damage. Brain Stimul 531 2019;12:152–60. https://doi.org/10.1016/J.BRS.2018.10.011. 532 [34] Devanne H, Lavoie A, Capaday C, Experiments A. Input -output properties and gain changes in the 533 human corticospinal pathway 1997:329–38. 534 [35] Conway BA, Halliday DM, Farmer SF, Shahani U, Maas P, Weir AI, et al. Synchronization between 535 motor cortex and spinal motoneuronal pool during the performance of a maintained motor task in man. 536 J Physiol 1995;489:917–24. https://doi.org/10.1113/jphysiol.1995.sp021104. 537 [36] Di Lazzaro V, Oliviero A, Profice P, Insola A, Mazzone P, Tonali P, et al. Effects of voluntary contraction 538 on descending volleys evoked by transcranial electrical stimulation over the motor cortex hand area in 539 conscious humans. Exp Brain Res 1999;124:525–8. https://doi.org/10.1007/s002210050649. 540 [37] Premoli I, Castellanos N, Rivolta D, Belardinelli P, Bajo R, Zipser C, et al. TMS -EEG Signatures of 541 GABAergic Neurotransmission in the Human Cortex. J Neurosci 2014;34:5603 –12. 542 https://doi.org/10.1523/JNEUROSCI.5089-13.2014. 543 [38] Rogasch NC, Daskalakis ZJ, Fitzgerald PB. Mechanisms underlying long -interval cortical inhibition in 544 the human motor cortex: A TMS -EEG study. J Neurophysiol 2013;109:89 –98. 545 https://doi.org/10.1152/JN.00762.2012/ASSET/IMAGES/LARGE/Z9K0021317430004.JPEG. 546 [39] Šulcová D, Salatino A, Ivanoiu A, Mouraux A. Evaluation of GABAAR-mediated inhibition in the human 547 brain using TMS -evoked potentials. BioRxiv 2022:2022.10.31.514225. 548 https://doi.org/10.1101/2022.10.31.514225. 549 [40] Bender S, Basseler K, Sebastian I, Resch F, Kammer T, Oelkers -Ax R, et al. Transcranial magnetic 550 stimulation evokes giant inhibitory potentials in children. Ann Neurol 2005;58:58 –67. 551 https://doi.org/10.1002/ana.20521. 552 [41] Bruckmann S, Hauk D, Roessner V, Resch F, Freitag CM, Kammer T, et al. Cortical inhibition in 553 attention deficit hyperactivity disorder: new insights from the electroencephalographic response to 554 transcranial magnetic stimulation. Brain 2012;135:2215–30. https://doi.org/10.1093/BRAIN/AWS071. 555 [42] Yamanaka K, Kadota H, Nozaki D. Long -latency TMS-evoked potentials during motor execution and 556 inhibition. Front Hum Neurosci 2013;7:751. https://doi.org/10.3389/fnhum.2013.00751. 557 [43] Kičić D, Lioumis P, Ilmoniemi RJ, Nikulin V V. Bilateral changes in excitability of sensorimotor cortices 558 during unilateral movement: Combined electroencephalographic and transcranial magnetic stimulation 559 study. Neuroscience 2008;152:1119–29. https://doi.org/10.1016/j.neuroscience.2008.01.043. 560 [44] Nikulin V V, Kicić D, Kähkönen S, Ilmoniemi RJ. Modulation of electroencephalographic responses to 561 .CC-BY-NC-ND 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted September 12, 2025. ; https://doi.org/10.1101/2025.09.07.674734doi: bioRxiv preprint 21 transcranial magnetic stimulation: evidence for changes in cortical excitability related to movement. Eur 562 J Neurosci 2003;18:1206–12. https://doi.org/10.1046/j.1460-9568.2003.02858.x. 563 [45] Karabanov AN, Madsen KH, Krohne LG, Siebner HR. Does pericentral mu-rhythm “power” corticomotor 564 excitability? – A matter of EEG perspective. Brain Stimul 2021;14:713 –22. 565 https://doi.org/10.1016/j.brs.2021.03.017. 566 [46] Schaworonkow N, Triesch J, Ziemann U, Zrenner C. EEG-triggered TMS reveals stronger brain state-567 dependent modulation of motor evoked potentials at weaker stimulation intensities. Brain Stimul 568 2019;12:110–8. https://doi.org/10.1016/j.brs.2018.09.009. 569 [47] Zrenner C, Kozák G, Schaworonkow N, Metsomaa J, Baur D, Vetter D, et al. Corticospinal excitability 570 is highest at the early rising phase of sensorimotor µ-rhythm. Neuroimage 2023;266. 571 https://doi.org/10.1016/j.neuroimage.2022.119805. 572 [48] Bergmann TO, Lieb A, Zrenner C, Ziemann U. Pulsed facilitation of corticospinal excitability by the 573 sensorimotor μ-alpha rhythm. J Neurosci 2019;39:10034 –43. 574 https://doi.org/10.1523/JNEUROSCI.1730-19.2019. 575 [49] Perera ND, Alekseichuk I, Shirinpour S, Wischnewski M, Linn G, Masiello K, et al. DISSOCIATION OF 576 CENTRALLY AND PERIPHERALLY INDUCED TRANSCRANIAL MAGNETIC STIMULATION 577 EFFECTS IN NONHUMAN PRIMATES. J Neurosci 2023;43:8649 –62. 578 https://doi.org/10.1523/JNEUROSCI.1016-23.2023. 579 [50] Desideri D, Zrenner C, Gordon PC, Ziemann U, Belardinelli P. Nil effects of μ-rhythm phase-dependent 580 burst-rTMS on cortical excitability in humans: A resting -state EEG and TMS -EEG study. PLoS One 581 2018;13. https://doi.org/10.1371/journal.pone.0208747. 582 [51] Beck MM, Heyl M, Mejer L, Vinding MC, Christiansen L, Tomasevic L, et al. Methodological choices 583 matter: A systematic comparison of TMS -EEG studies targeting the primary motor cortex. BioRxiv 584 2024;30:2024.03.08.583873. https://doi.org/10.1101/2024.03.08.583873. 585 [52] Perera ND, Wischnewski M, Alekseichuk I, Shirinpour S, Opitz A. State dependent motor cortex 586 stimulation reveals distinct mechanisms for corticospinal excitability and cortical responses. BioRxiv 587 2024:2024.07.15.603619. https://doi.org/10.1101/2024.07.15.603619. 588 [53] Petrichella S, Johnson N, He B. The influence of corticospinal activity on TMS -evoked activity and 589 connectivity in healthy subjects: A TMS -EEG study. PLoS One 2017;12:e0174879. 590 https://doi.org/10.1371/journal.pone.0174879. 591 [54] Fecchio M, Pigorini A, Comanducci A, Sarasso S, Casarotto S, Premoli I, et al. The spectral features 592 of EEG responses to transcranial magnetic stimulation of the primary motor cortex depend on the 593 .CC-BY-NC-ND 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted September 12, 2025. ; https://doi.org/10.1101/2025.09.07.674734doi: bioRxiv preprint 22 amplitude of the motor evoked potentials. PLoS One 2017;12:e0184910. 594 https://doi.org/10.1371/journal.pone.0184910. 595 [55] Koponen LM, Goetz SM, Tucci DL, Peterchev A V. Sound comparison of seven TMS coils at matched 596 stimulation strength. Brain Stimul 2020;13:873–80. https://doi.org/10.1016/j.brs.2020.03.004. 597 [56] Conde V, Tomasevic L, Akopian I, Stanek K, Saturnino GB, Thielscher A, et al. The non -transcranial 598 TMS-evoked potential is an inherent source of ambiguity in TMS -EEG studies. Neuroimage 599 2019;185:300–12. https://doi.org/10.1016/J.NEUROIMAGE.2018.10.052. 600 [57] Rocchi L, Di Santo A, Brown K, Ib áñez J, Casula E, Rawji V, et al. Disentangling EEG responses to 601 TMS due to cortical and peripheral activations. Brain Stimul 2021;14:4 –18. 602 https://doi.org/10.1016/j.brs.2020.10.011. 603 [58] Nikouline V, Ruohonen J, Ilmoniemi RJ. The role of the coil click in TMS assessed with simultaneous 604 EEG. Clin Neurophysiol 1999;110:1325–8. https://doi.org/10.1016/S1388-2457(99)00070-X. 605 [59] Biabani M, Fornito A, Goldsworthy M, Thompson S, Lynton Graetz ;, Semmler JG, et al. Characterising 606 the contribution of auditory and somatosensory inputs to TMS -evoked potentials following stimulation 607 of prefrontal, premotor and parietal cortex. BioRxiv Neurosci 2024. 608 https://doi.org/10.1101/2023.11.09.566371. 609 [60] Rosanova M, Casali A, Bellina V, Resta F, Mariotti M, Massimini M. Natural frequencies of human 610 corticothalamic circuits. J Neurosci 2009;29:7679 –85. https://doi.org/10.1523/JNEUROSCI.0445 -611 09.2009. 612 [61] Rogasch NC, Thomson RH, Daskalakis ZJ, Fitzgerald PB. Short -latency artifacts associated with 613 concurrent TMS-EEG. Brain Stimul 2013;6:868–76. https://doi.org/10.1016/j.brs.2013.04.004. 614 [62] Cohen LG, Starr A. Localization, timing and specificity of gating of somatosensory evoked potentials 615 during active movement in man. Brain 1987;110:451–67. https://doi.org/10.1093/brain/110.2.451. 616 [63] Starr A, Cohen LG. “Gating” of somatosensory evoked potentials begins before the onset of voluntary 617 movement in man. Brain Res 1985;348:183–6. https://doi.org/10.1016/0006-8993(85)90377-4. 618 [64] Tapia MC, Cohen LG, Starr A. Attenuation of auditory-evoked potentials during voluntary movement in 619 man. Int J Audiol 1987;26:369–73. https://doi.org/10.3109/00206098709081565. 620 [65] Hazemann P, Audin G, Lille F. Effect of voluntary self -paced movements upon auditory and 621 somatosensory evoked potentials in man. Electroencephalogr Clin Neurophysiol 1975;39:247 –54. 622 https://doi.org/10.1016/0013-4694(75)90146-7. 623 .CC-BY-NC-ND 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted September 12, 2025. ; https://doi.org/10.1101/2025.09.07.674734doi: bioRxiv preprint 23 624 .CC-BY-NC-ND 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted September 12, 2025. ; https://doi.org/10.1101/2025.09.07.674734doi: bioRxiv preprint