Pinging the Brain with Transcranial Magnetic Stimulation Reveals Cortical Reactivity in Time and Space

Single-pulse transcranial magnetic stimulation (TMS) elicits an evoked electroencephalography (EEG) potential (TMS-evoked potential, TEP), which is interpreted as direct evidence of cortical reactivity to TMS. Thus, combining TMS with EEG may enable the mechanistic investigation of how TMS treatment paradigms engage network targets in the brain. However, there remains a central controversy about whether the TEP is a genuine marker of cortical reactivity to TMS or the TEP is contaminated by responses to peripheral somatosensory and auditory inputs. Resolving this controversy is of great significance for the field and will validate TMS as a tool to probe networks of interest in cognitive and clinical neuroscience. Here, we delineated the TEP’s cortical origins by localizing successive TEP components in time and space and modulating them subsequently with transcranial direct current stimulation (tDCS). We collected both motor evoked potentials (MEPs) and TEPs elicited by suprathreshold single-pulse TMS to the left primary motor cortex (M1). We found that the earliest TEP component (P25) was localized on the TMS target location (left M1) and the following TEP components (N45 and P60) largely were localized on the primary somatosensory cortex, which may reflect afferent input by hand-muscle twitches. The later TEP components (N100, P180, and N280) largely were localized to the auditory cortex. To casually test that these components reflect cortical and corticospinal excitability, we applied tDCS to the left M1. As hypothesized, we found that tDCS modulated cortical and corticospinal excitability selectively by modulating the pre-stimulus mu-rhythm oscillatory power. Together, our findings provide causal evidence that the early TEP components reflect cortical reactivity to TMS.

marker of cortical reactivity to TMS or the TEP is contaminated by responses to peripheral 23 somatosensory and auditory inputs. Resolving this controversy is of great significance for 24 the field and will validate TMS as a tool to probe networks of interest in cognitive and clinical 25 neuroscience. Here, we delineated the TEP's cortical origins by localizing successive TEP 26 components in time and space and modulating them subsequently with transcranial direct 27 current stimulation (tDCS). We collected both motor evoked potentials (MEPs) and TEPs 28 elicited by suprathreshold single-pulse TMS to the left primary motor cortex (M1). We found 29 that the earliest TEP component (P25) was localized on the TMS target location (left M1) 30 and the following TEP components (N45 and P60) largely were localized on the primary 31 somatosensory cortex, which may reflect afferent input by hand-muscle twitches. The later 32 TEP components (N100, P180, and N280) largely were localized to the auditory cortex. To 33 casually test that these components reflect cortical and corticospinal excitability, we applied 34 tDCS to the left M1. As hypothesized, we found that tDCS modulated cortical and we administered single-pulse TMS to the left M1 while recording TEPs and MEPs in three 82 sessions. The tDCS condition (anodal, cathodal, and sham tDCS) was randomized for each 83 session in a double-blind, cross-over study design. For tDCS, we used the conventional two-84 electrode montage (Fig. 1c, left, referred to as M1-SO, one for the M1, and another for the 85 supraorbital cortex). We applied 100 single pulses of TMS before and after tDCS during 86 each session. At the end of each session, we collected EEG electrode locations using a 87 stereo-camera tracking digitizer to improve accuracy of source localization. baseline period (-100 to 0ms). We refer to these peaks by their canonical names: P25, N45, 98 P60, N100, P180, and N280. 99 To investigate these TEP components' spatial distribution on the scalp, we computed 100 topographical distributions at each TEP time point (Fig. 1b). We found that the left 101 sensorimotor area was activated predominantly up to 60ms (P25, N45, and P60). After P60, 102 the centroid of activation drifted towards the midline until it centered entirely at 280ms (N100, 103 P180, and N280). 104 As the sensor-space representation captures the summed cortical activity on the scalp, we 105 next localized the TEPs on the cortex (source space). We first localized the TEPs to 106 individual cortex models (15000 voxels) for each participant and then projected the localized 107 TEPs to a template cortex model (15000 voxels, FsAverage) for group analysis and 5 108 computed the grand-averaged TEP (5255 epochs). For each component depicted in sensor 109 space (Fig. 2b), we projected the grand-averaged TEP onto the template cortex model (Fig.   110 2c). We found that P25, the earliest TEP component, was localized to the hand area of the 111 left M1 (TMS target, Fig. 1b). N45 showed activation that spread between the M1 and the 112 primary somatosensory cortex (Fig. 2c, second column). Next, P60 was localized to primary 113 somatosensory cortex (Fig. 2c, third column). In contrast, the N100 and P180 peaks largely 114 were localized to the auditory cortex and reflected the N100-P180 auditory complex [6,7]. 115 The final TEP component (N280) also was localized to the auditory cortex, but exhibited 116 additional activation in the frontal cortex. These findings demonstrate that single-pulse TMS 117 on the hand area of the M1 elicits multiple TEP components and that the earliest (P25) 118 reflects genuine cortical reactivity to TMS. We hypothesized from this finding that the N45 119 and P60 reflect the afferent signal from the corticospinal tract attributable to hand-muscle 120 twitches. In contrast, the later TEP components (N100, P180, and N280) may reflect 121 auditory processing of the coil's clicking sound. each EEG channel. We found positive correlation clusters for the P25 (9 EEG channels) and 132 P60 (6 EEG channels), and a negative cluster for the N45 (5 EEG hannels) in the left 133 sensorimotor area (Fig. 3a, top row, r-value topographical maps). The black dots in the 134 topographical maps indicate significant EEG channels (p<0.05). In contrast, we found no 135 significant cluster for the N100, P180, and N280 (Fig. 3a, bottom row, p>0.05). 136 To understand the relation between cortical reactivity and the corticospinal response better, 137 we selected the significant EEG channels for each TEP component and averaged them to 138 obtain scatter plots with MEP amplitude (Fig. 3b, n=54 for each TEP component). As 139 expected, we found significant positive correlations for the P25 (r=0.52, p<0.001) and P60 6 140 (r=0.51, p<0.001), and a significant negative correlation (r=-0.58, p<0.001) for the N45. Note 141 that right green y-axis corresponds to the N45 amplitude (negative amplitude) 142 Next, we investigated how the localized TEP components in source space were correlated 143 with MEPs. First, we defined a region of interest (ROI, Fig. 3c) for the TEP components 144 (P25, N45, and P60) based on the source-localized TEPs (Fig. 2c). Using these ROIs, we 145 performed correlation analyses between the localized TEP components and MEPs (Fig. 3d). 146 We found significant positive correlations for the P25  sensorimotor area (averaged 7 EEG channels described previously) for the entire epoch 170 and calculated the ratio (post/pre) for each tDCS condition. We found that the period of the 171 TEP from 25 to 60ms differed significantly for "condition" (Fig. 4b, shaded period; linear 172 mixed-effect model, F 2,28 =129, p<0.0001), but not for "session" (F 2,28 =1.12, p=0.34), or their 7 173 interaction (F 4,28 =1.31, p=0.29). In contrast, we found no significant difference for the other 174 TEP components across tDCS conditions (100 to 280ms, p>0.05). 175 To investigate the modulated TEPs' spatial representation for each tDCS condition, we next 176 computed topographical distributions for the P25, N45, and P60. We found that the left 177 sensorimotor area for the P25, N45, and P60 differed significantly in the anodal tDCS  Cathodal tDCS attenuated the magnitude of TEP components that contained M1 activation.

185
In the sham tDCS condition, we found no significant EEG channels for the P25, N45, or P60   and third columns). After cathodal tDCS, the N45 was modulated significantly in the primary 207 somatosensory cortex (Fig. 4e, second row, second column), but the P60 did not differ 208 significantly (Fig. 4e, second row, third column). We found no such significant differences in 209 the sham tDCS condition (Fig. 4e, third row). Similarly, we found no statistical difference for 210 the N100, P180, and N280 in all tDCS conditions ( Supplementary Fig. 2b). These findings 211 indicate that tDCS modulates localized cortical reactivity by single-pulse TMS in the early 212 TEP components. 213 We then performed correlation analyses to investigate how the modulation of localized TEPs    The remaining TEP components (N100, P180, and N280) were localized primarily to the 271 auditory cortex. Importantly, tDCS modulated the first two TEP components (P25 and N45) 272 selectively depending upon polarity in our double-blind, placebo-controlled study. In addition, 273 we found evidence that cortical reactivity played a causal role in predicting corticospinal 274 excitability. Thus, our findings demonstrate that the early TEP reflects genuine cortical 275 reactivity and later TEP components are associated with somatosensory and auditory 276 processing in the brain.

277
A recent study that investigated neural effects at the single-cell level has shown that 278 suprathreshold single-pulse TMS elicits a stereotyped burst of action potentials within the 279 first 30ms (10-30ms) after TMS onset in the macaque parietal cortex [ previous findings. We also observed N45 and P60 components that were localized primarily 287 in the primary somatosensory cortex and reflected afferent input by hand-muscle twitches 288 produced by suprathreshold TMS. We demonstrated further that these somatosensory-11 289 evoked potentials were correlated with MEP amplitude (Fig. 3b) and comparable to the 290 conventional somatosensory evoked potentials with respect to response latency [16]. For the 291 later TEP components, although we applied auditory masking using white noise that 292 removed the auditory perception of TMS pulses, we obtained the typical N100-P180 auditory 293 complex[6] by single-pulse TMS (Fig. 2a), which was localized in the auditory cortex ( Fig.   294 2c). This phenomenon may derive from inevitable bone-and air-conducted sound from the 295 TMS coil [7]. The amplitude of these potentials (>5uV) was comparable with the N100 296 amplitude in our study. Thus, we conclude overall that each TEP component single-pulse 297 TMS elicits has a distinct network representation in the brain and the P25 represents 298 genuine cortical reactivity from TMS to the M1.

299
Since the first attempt to modulate motor cortex excitability by weak direct current on the 300 scalp[17], it has been shown consistently that tDCS modulates motor cortex excitability 301 depending upon polarity [8,9,[18][19][20][21][22][23]. We hypothesized that if a TEP elicited by single-pulse 302 TMS on the M1 is genuine motor-related cortical reactivity, then tDCS to the M1 should 303 modulate it. We found that tDCS successfully modulated the P25 in the stimulated cortical 304 area in a polarity-dependent manner (Fig. 4e). tDCS also modulated the N45 in the same 305 manner, but only anodal tDCS modulated the P60. Consistent with the findings for the P60, 306 the relation between changes in MEP and P60 amplitude was not significant in both the 307 sensor (r=-0.14, p=0.57) and source (r=0.41, p=0.09) space. We assume that this 308 unexpected finding might be caused by the reduction of post-stimulus mu-rhythm (around 309 200 to 300ms after onset) by cathodal tDCS (Fig. 5b, second row, time-frequency t-value 310 map). We hypothesized that tDCS could modulate only the pre-stimulus mu-rhythm, but 311 cathodal tDCS actually reduced the post-stimulus mu-rhythm, which was not found in the  In our study, we used the M1-SO montage with two smaller electrodes (5x5cm, 25cm 2 ) to 321 increase efficacy via a greater current intensity [20]. We performed electric field modeling 12 322 with structural MR images and confirmed that the induced electric field is comparable to that 323 in previous tDCS studies (Fig. 1c). As an exploratory analysis, we investigated how the      379 We performed a crossover, double-blind, sham-controlled study with three tDCS 380 conditions (anodal, cathodal, and sham tDCS) at the University of North Carolina at 381 Chapel Hill, which the Biomedical Institutional Review Board at the university approved.

382
The study protocol was registered before participants were recruited (ClinicalTrials.gov, 383 NCT03481309). We recruited 19 healthy, right-handed, male participants free of any 384 neurological disorders. All participants provided written informed consent before 385 participation. After telephone screening to assess their eligibility for the study, structural

399
Based on the structural MR images, we performed brain segmentation and determined an  The EEG and MEP recording procedures were performed both before and after tDCS.

426
Transcranial direct current stimulation (tDCS) 427 We applied two carbon-silicone electrodes (5x5cm) to the scalp with Ten20 conductive     All data, as well as analysis codes that were used to perform analyses, can be made 533 available from the corresponding author upon reasonable request.