Repetitive transcranial magnetic stimulation (rTMS) protocols, such as continuous theta-burst stimulation (cTBS), represent a non-invasive way to reduce cortical excitability in human volunteers (Huang et al., 2005) and to explore neuroplasticity in several patient populations (Edwards et al., 2006; Oberman et al., 2010; Orth et al., 2010; Suppa et al., 2011; Koch et al., 2012; Munneke et al., 2013; Opie et al., 2013; Chuang et al., 2014; Mori et al., 2014; Suppa et al., 2014a; Suppa et al., 2014b). However, a number of major hurdles prevent its widespread application in basic research and in clinical care. For example, despite more than 10 years of studies in human volunteers, it is still unclear how neural activity changes in the cortex after cTBS (Pitcher et al., 2021). It is generally assumed that the reduction in neuronal excitability may resemble the changes observed in long-term depression (LTD; Fitzgerald et al., 2006; Thut and Pascual-Leone, 2010; Vlachos et al., 2012), but no study has ever tested in vivo whether this is true. The reduction in the amplitude of the motor evoked potential (MEP) after cTBS also grows over time (Huang et al., 2005), which does not resemble the immediate reduction in neuronal excitability after conditioning observed in LTD. Moreover and more importantly, the physiological effects of cTBS are notoriously variable, with inhibitory effects in some subjects and facilitatory effects in other subjects (Hamada et al., 2013). Several factors could contribute to this variability (reviewed in Suppa et al., 2016). Genetic variation, differences in the intracortical network activated by the TMS pulses (measured in the so-called late I-waves), previous levels of activity, circadian effects, and differences in the brain state may all contribute to the inter- and intra-individual variability reported in cTBS studies.

We sought answers to the aforementioned questions by recording single-neuron activity before and after cTBS in parietal cortex of awake behaving rhesus monkeys. We probed the neuronal excitability using single-pulse TMS (sTMS), and measured how this excitability changed up to 2 hr after cTBS. To the extent possible, we standardized the factors potentially contributing to the known variability associated with cTBS: the positioning of the coil on the skull, the level of ongoing motor activity, the stimulation hour, and the brain state were equalized as much as possible across sessions and animals. We observed highly reproducible changes in neuronal excitability, in which individual neurons would progress through phases of hypo- and hyperexcitability, in some cases followed by recovery. At the population level, the reduction in neuronal excitability grew over time, reaching its maximum 30–40 min after cTBS, consistent with studies in humans. Thus, in a standardized experimental setting, cTBS induced reproducible and long-lasting changes in neuronal excitability.

We recorded single-unit activity in parietal area PFG of two monkeys during sTMS in 90 experimental sessions. Across these sessions, a total of 86 neurons (51 neurons in monkey Y; 35 in monkey A) showed a significant increase in their firing rate in response to sTMS, and were further recorded before and after cTBS during passive fixation.

Effect of cTBS on individual neurons

Before applying cTBS, we tested the excitability of each neuron using sTMS administered at light onset above the object (Figure 1).

Figure 1

Download asset Open asset

Figure 2 illustrates, with two example neurons, the typical results obtained in this study. The first example neuron (Figure 2A) generated a brief burst of action potentials almost immediately after sTMS (top row) but did not respond to light onset in the absence of sTMS. After 20 s of cTBS, however, the excitability of this neuron was markedly reduced (second row, sTMS-evoked response pre- compared to post-cTBS: z=3.88, p=1.06 e−04, r=0.60; Wilcoxon test). Indeed, in the first 10 min post- CTBS, the activity in high-stimulation trials did not differ anymore from the activity in no-stimulation trials (z=1.81, p=0.07, r=0.17). Over the subsequent time intervals, the excitability of this neuron gradually recovered (no-stimulation compared to stimulation trials; 20 min post-cTBS: z=4.99, p=6.01e−07, r=0.46; 30 min post-cTBS: z=6.58, p=4.57e−11, r=0.60; 40 min post-cTBS: z=5.27, p=1.35e−07, r=0.48; 50 min post-cTBS: z=4.74, p=2.17e−06, r=0.43), but only after 60 min did the excitability of the neuron return to the pre-cTBS level (sTMS-evoked response pre- compared to 60 min post-cTBS: z=1.63, p=0.103, r=0.26). Thus, cTBS caused a marked and immediate reduction in excitability in this parietal example neuron, which recovered over the course of 1 hr.

Figure 2

Download asset Open asset

The second example neuron also responded to sTMS before we applied cTBS (Figure 2B, top panel, notice also the weak response to light onset in the no-stimulation trials), but was not affected by cTBS in the first 10 min after cTBS (Figure 2B, second panel, sTMS-evoked response pre- compared to post-cTBS: z=–0.61, p=0.548, r=–0.10). Only in the 20 and 30 min intervals after cTBS, this neuron’s excitability dropped considerably (no significant difference between stimulation and no-stimulation trials at 20 min post-cTBS: z=–0.84, p=0.399, r=–0.07, and at 30 min post-cTBS: z=–0.20, p=0.839, r=–0.02), although its visual response (to light onset: baseline vs. post-light onset activity in no-stimulation trials; z=–4.72, p=2.32e−06, r=–0.32) remained. Surprisingly, in the 40 and 50 min intervals, the excitability of this neuron temporarily increased significantly (sTMS-evoked response pre-compared to post-cTBS at 40 min post-cTBS: z=–2.24, p=0.025, r=–0.355 and at 50 min post-cTBS: z=–2.68, p=0.007, r=–0.42). Only in the 60 min interval did the neuron return to its pre-cTBS excitability level (z=–2.61, p=0.08, r=–0.41, pre-compared to 60 min post-cTBS in the first 40 ms after TMS onset). A third example neuron (Figure 3A) showed yet another pattern of excitability changes after cTBS. The strong burst of activity evoked by sTMS was immediately reduced after cTBS (Figure 3A, compare top panel with second panel: z=3.64, p=2.741e–04, r=0.68). However, this reduction in excitability grew markedly over time and reached its peak only after 50 min. At the end of our standard 60 min recording epoch, the neuron did not respond anymore to sTMS, without any sign of recovery, although we could still sporadically detect spikes (Figure 3B). In addition to the growing suppression of the sTMS response in the first hour after cTBS, we also observed a significant reduction in the baseline activity in this neuron (by 91%, pre-compared to 40 min post-cTBS: z=3.64, p=1.251e–05, r=0.65). Because the activity of the neuron was extremely reduced, we verified that the neuron was not lost during the recording session by comparing the spike waveforms recorded before (Figure 3B, upper panel) and 60 min after cTBS (Figure 3B, lower panel). The firing rate of this neuron in the 60 min post-cTBS interval was reduced to barely 4 spikes/s, but the spike waveform was virtually identical compared to the pre-cTBS epoch, which confirms that we still recorded from the same neuron. Notice also a very weak response to sTMS at 60 min post-cTBS. Overall, cTBS induced a compelling reduction in the excitability of parietal neurons, with highly variable onset time and recovery.

Figure 3

Download asset Open asset

Population analysis

Because the results were highly comparable in the two animals, we pooled all neurons recorded, but we also provide statistics for each animal separately (see Figure 4—figure supplement 1). In the first 10 min post-cTBS, almost half of the PFG neurons (43%; 39% in monkey Y and 49% in monkey A) showed a significant (two-sided Wilcoxon rank sum test p<0.05) change in their sTMS response (either hypo- or hyperexcitability). However, the proportion of neurons in which cTBS affected the sTMS response gradually increased over time, such that in the 60 min post-cTBS interval, virtually all neurons (85/86) were significantly affected by cTBS (Table 1; with Bonferroni correction, the proportions of affected neurons grew from 24% at 10 min post-cTBS to 62% at 60 min post-cTBS). Similarly, cTBS induced an immediate and significant change in the baseline activity (i.e., the spike rate before sTMS) in about one third of the neurons (34%, two-sided Wilcoxon rank sum test p<0.05), which reached its maximum 1 hr post-cTBS (93% of the neurons, Table 1). Overall, 21% of the neurons showed an effect both in the sTMS response and the baseline activity in the first 10 min post-cTBS, compared to 83% of the neurons showing a combined effect 1 hr post-cTBS.

Table 1

	10 min post-cTBS	20 min post-cTBS	40 min post-cTBS	60 min post-cTBS
sTMS effect	43% (37/86) *24% (21/86) Y: 20/51; A: 17/35 *Y: 9/51; A: 12/35	53% (46/86) *33% (28/86) Y: 23/51; A: 23/35 *Y: 14/51; A: 14/35	86% (74/86) *57% (49/86) Y: 43/51; A: 31/35 *Y: 26/51; A: 23/35	99% (85/86) *62% (53/86) Y: 50/51; A: 35/35 *Y: 29/51; A: 24/35
Baseline effect	34% (29/86) *22% (19/86) Y: 17/51; A: 12/35 *Y: 12/51; A: 7/35	50% (43/86) *33% (28/86) Y: 24/51; A: 19/35 *Y: 18/51; A: 10/35	77% (66/86) *44% (38/86) Y: 40/51; A: 26/35 *Y: 24/51; A: 14/35	93% (80/86) *50% (43/86) Y: 48/51; A: 32/35 *Y: 30/51; A: 13/35
Hyperexcitability in sTMS response	10% (9/86) *3% (3/86) Y: 6/51; A: 3/35 *Y: 2/51; A: 1/35	10% (9/86) *7% (6/86) Y: 6/51; A: 3/35 *Y: 4/51; A: 2/35	19% (16/86) *15% (13/86) Y: 10/51; A: 6/35 *Y: 8/51; A: 5/35	28% (24/86) *23% (20/86) Y: 15/51; A: 9/35 *Y: 12/51; A: 8/35
Hyperexcitability in baseline activity	*10% (9/86) 6% (5/86) Y: 5/51; A: 4/35 *Y: 3/51; A: 2/35	*12% (10/86) 10% (9/86) Y: 5/51; A: 5/35 *Y: 5/51; A: 4/35	*15% (13/86) 13% (11/86) Y: 8/51; A: 5/35 *Y: 7/51; A: 4/35	*24% (21/86) 21% (18/86) Y: 13/51; A: 8/35 *Y: 11/51; A: 7/35

1. Results on the Wilcoxon test without Bonferroni correction calculated for 10’ intervals.

2. *

   Results on the Wilcoxon test with Bonferroni correction calculated for 10’ intervals.

On average, cTBS significantly reduced the neuronal excitability assessed with sTMS, an effect that grew over the standard 1 hr post-cTBS period (Figure 4 and Figure 4—figure supplement 1). In the first 10 min interval post-cTBS, the average sTMS response was 24% lower than pre-cTBS (z=2.11, p=0.03, r=0.16), and this reduction peaked and plateaued 30–50 min post-cTBS (32% reduction). Similarly, the reduction in the average baseline activity induced by cTBS emerged already in the first 20 min post-cTBS (by 14%; z=1.42, p=0.04, r=0.15), became more apparent 40 min post-cTBS (28% reduction, z=3.23, p=0.001, r=0.25), and recovered partially at 60 min post-cTBS (20% reduction, two-sided Wilcoxon rank sum tests comparing the pre-cTBS cell response with that measured at 40 min post-cTBS; for the two animals at 40 min: z=2.66, p=0.01, r=0.26 for monkey Y and z=1.86, p=0.06, r=0.22 for monkey A).

Figure 4 with 1 supplement see all

Download asset Open asset

The cTBS effect was highly similar in both animals (maximum 32% reduction in sTMS response in monkey Y, and 29% reduction in sTMS response in monkey A), but the timing of the effect differed slightly, since the maximum reduction appeared at 30 min post-cTBS in monkey Y and at 40 min post-cTBS in monkey A (Figure 4—figure supplement 1).

To investigate whether the cTBS effect on the neuronal sTMS response and the cTBS effect on the baseline activity were based on similar neuronal mechanisms, we plotted the changes in the sTMS response and in the baseline activity against each other and calculated the correlation between the two effects in every epoch post-cTBS (Figure 5; see Table 2). In each individual animal (data not shown) and across all neurons combined, we measured significant and high correlation coefficients ranging from 0.54 (at 60 min post-cTBS) to 0.80 (at 30 min post-cTBS), suggesting that cTBS induced a general change in neuronal excitability, which was manifested in both sTMS response changes and alterations in spontaneous activity. This correlation increased significantly over time (20, 30, 40, and 50 min interval compared to the first 10 min post-cTBS), indicating that the cTBS effect became more apparent in both the baseline activity and the sTMS evoked responses. At 60 min post-cTBS, the correlation declined significantly compared to the preceding intervals, most likely because neurons started to recover from the reduction in excitability induced by cTBS.

Figure 5

Download asset Open asset

Table 2

95% CI	+10′	+20′	+30′	+40′	+50′	+60′
Lower bound	0.51	0.69	0.70	0.69	0.67	0.36
r	0.66	0.79	0.80	0.79	0.77	0.54
Upper bound	0.76	0.86	0.86	0.86	0.85	0.68

Figure 5 also illustrates that, although cTBS in general reduced neuronal excitability (53% of the neurons showed hypoexcitation in all time epochs post-cTBS), transient phases of hyperexcitability were not uncommon in our neuronal population. Overall, almost half of the neurons (47%) were hyperexcitable in at least one interval post-cTBS (Table 1). The large majority of these neurons (91%) were initially less excitable followed by an epoch of hyperexcitability, as the example neuron in Figure 2B. However, in a small number of neurons, hyperexcitability appeared immediately after cTBS, either as the only effect throughout the entire recording session (6%) or followed by hypoexcitability after a variable time interval (10–40 min, 3%). Significant increases in baseline activity were also not uncommon, since 44% of the neurons showed hyperexcitability in at least one epoch post-cTBS (Table 1).

cTBS may not only influence the average neuronal response to sTMS, but also the spike timing, that is, the variability of the interspike intervals. We could not detect any significant effect of cTBS on the variance to mean ratio (the Fano factor) of the neurons, nor on the distribution of the interspike intervals (ISIs, data not shown). However, the power spectrum of the spike trains changed significantly after cTBS, both in the stimulation and in no-stimulation trials (Figure 6). In every epoch post-cTBS, the power in the low frequencies (below 5 Hz) was significantly reduced compared to pre-cTBS. Thus, cTBS also induced changes in the low-frequency oscillatory activity of parietal neurons.

Figure 6

Download asset Open asset

Effect of cTBS on task-related neural activity

Since a motor task may cause small movements of the electrode and since we wanted to give priority to the stability of the neural recordings, we chose to use a passive fixation task for the recordings, in which the monkeys were required to simply fixate an object illuminated in front of them to obtain a fluid reward. Despite the absence of a grasping movement, a subset of the neurons we recorded in PFG showed significant task-related activity (i.e., object responses) after light onset (N=18). The presence of object responses allowed us to test the effect of cTBS on neuronal excitability without applying sTMS. To capture all task-related responses, we tested the effect of cTBS in two intervals, an early (0–80 ms after light onset) and a late interval (100–500 ms after light onset) and analyzed the response difference in the 40 ms around the maximal response. The later task-related activity (100–500 ms after light onset) differed significantly compared to the pre-cTBS epoch in every epoch from 30 to 60 min post-cTBS, whereas the early interval after light onset did not show a significant difference in any epoch post-cTBS (Wilcoxon signed rank test, Figure 7A). When probed with sTMS, this subpopulation of task-related neurons behaved similar to the rest of the population, since the neuronal excitability was significantly reduced at 30 min post-cTBS (Figure 7B). Overall, cTBS also induced a reduction in task-related activity in parietal neurons, comparable to the effects observed with sTMS.

Figure 7

Download asset Open asset

Effect of cTBS on neuronal excitability beyond 1 hr

Our standard recording time was 1 hr after cTBS (plus approximately 20 min of recordings pre-cTBS). However, in a subpopulation of neurons (N=34; Figure 8A), we could test the effect of cTBS on neuronal excitability for up to 90 min post-cTBS. Because neurons can be lost in the course of such a long time interval due to small brain movements, we only included units which the signal-to-noise ratio was at least 5:1 for the entire duration of the recording session, and we compared the spike waveforms pre- and post-cTBS to verify that the neuron was still present (see example spike waveforms in Figure 8B). Even 90 min after cTBS, most neurons (18/34, 53%) showed significantly reduced sTMS responses compared to pre-cTBS, and the average normalized sTMS-evoked response was significantly reduced at this late time period post-cTBS (z=1.99, p=0.02, r=0.24). Moreover, we also detected a significant reduction in neuronal activity after the sTMS-evoked burst (in the interval 200–500 ms after sTMS; z=−2.58, p=0.01, r=–0.32). A smaller subset of neurons (N=15) was monitored for up to 2 hr post-cTBS, but again no recovery was detectable in this subpopulation (data not shown). Thus, the cTBS-induced reduction in neuronal excitability is very prolonged and may not recover for several hours post-cTBS.

Figure 8

Download asset Open asset

We observed a pronounced and long-lasting reduction in neuronal excitability after cTBS in macaque parietal neurons. This reduced excitability of individual neurons grew over time, consistent with previous studies in humans, but some neurons exhibited periods of hyperexcitability during the recovery phase. These results provide the first experimental evidence on the neural effects of cTBS on single neurons in awake behaving monkeys. Importantly, while we did not control for remote or network cTBS effects as per scope of this study, our findings that cTBS did not systematically lead to inhibition of neuronal firing rate, that hypo- and hyperexcitability periods occurred at different time points in different neurons and that neuronal activity recovered to baseline, pre-cTBS levels, on average an hour post-cTBS, allow us to argue against nonspecific effect of the cTBS intervention. Moreover, in our previous work investigating the behavioral effects of cTBS on monkeys (Merken et al., 2021), we showed the differential effect of effective versus sham cTBS on motor performance, providing indirect evidence about the specificity of the underlying neural cTBS effects found in this study.

Our basic findings, both at the single-cell level and at the behavioral level (see Merken et al., 2021), are remarkably consistent with the known effects of cTBS over primary motor cortex in human volunteers (Huang et al., 2005). The average amplitude of the MEP—a measure of neuronal excitability in the primary motor cortex—is unaltered in the first minutes after cTBS, then gradually declines and recovers 25 (after 20 s cTBS) or 60 min (after 40 s cTBS) later. In parallel, the reaction time significantly increases 10 min after cTBS over primary motor cortex (Huang et al., 2005). The time course of the neural effects we measured after 20 s of cTBS in monkeys seems to be more similar to the 40 s cTBS protocol in humans, possibly due to the thinner skull of monkeys. Nevertheless, it should also be noted that we applied cTBS over parietal cortex, whereas most cTBS studies in humans have targeted primary motor cortex (but see Davare et al., 2010). Overall, the consistency of our findings with the human literature strongly suggests that single-cell recordings after cTBS in awake macaque monkeys represent a valid approach to understanding the neural effects of cTBS.

The crucial advantage of our approach is that we can measure the wide range of changes in excitability in individual neurons, while the MEP amplitude represents an excitability measure of a large population of neurons. In addition, we could demonstrate that the inhibitory influence of cTBS progressively recruits more neurons over time, even neurons that were initially unaffected. Finally, we showed that cTBS induces a general reduction in neuronal excitability, which appears in changes in baseline firing rate, sTMS-evoked responses, low-frequency oscillatory activity, and task-related activity. It is noteworthy that the neuronal object responses we measured here in PFG were relatively slower and longer-lasting, compared to neighboring area AIP (e.g., Pani et al., 2014).

Extracellular recordings do not easily allow determining the cell type of the unit that is being recorded based on the spike waveform (see, e.g., Woloszyn and Sheinberg, 2012 compared to Vigneswaran et al., 2011), nor the cortical layer in which these units were recorded. In general, we recorded from neurons located immediately under the TMS coil and generated large spike waveforms, that we could monitor for more than 1 hr and up to 2 hr post-cTBS. However, the large range of effects we observed in our population of neurons may suggest that cTBS exerts specific effects on different cell types, on neurons in different layers, and in a different orientation and/or location with respect to the TMS coil. In nonhuman primates, future studies could address these questions with advanced techniques such as calcium imaging (Ikezoe et al., 2013; Tang et al., 2020).

Previous studies have suggested that cTBS may induce LTD-like effects on cortical synapses (Huang et al., 2011). LTD, a widespread phenomenon driving synaptic plasticity both in subcortical structures and in the cortex, is typically induced by low-frequency stimulation (LFS; at 1 Hz), and its underlying molecular mechanisms may be very diverse depending on brain area and developmental stage (Massey and Bashir, 2007). In visual cortex, 15 min of 1 Hz stimulation induces LTD of synaptic responses (Kirkwood and Bear, 1994). However, the LTD effect studied in cortical slices appears immediately after the end of LFS and remains constant for up to half an hour. In contrast, the reduction in neuronal excitability we observed after cTBS grew gradually and reached its maximum 30–50 min after cTBS, similar to the reduction in the amplitude of the MEP observed after cTBS in humans (Huang et al., 2005). Almost half of the neurons in our sample were not affected at all in the first epoch post-cTBS, but became less excitable in the hour following cTBS. Therefore, the time course of the cTBS effect we measured does not seem to be compatible with a pure LTD effect. Instead, the gradual reduction in neuronal excitability seems more consistent with an increase in the local concentration of the inhibitory neurotransmitter GABA, which may slowly spread in the cortical area under the TMS coil. It is noteworthy that magnetic resonance spectroscopy has demonstrated an increase in GABA in human motor cortex after cTBS (Stagg et al., 2009). In the nonhuman primate model, future studies will be able to investigate in more detail which molecular mechanisms are responsible for the effect of cTBS.

Unexpectedly, some neurons also showed a period of hyperexcitability after an initial phase of reduced excitability caused by cTBS. We interpret these findings as evidence that nearby inhibitory interneurons showed a different time course of recovery after cTBS, such that at some point post-cTBS the neuronal excitability had recovered while the normal inhibitory inputs to the neuron were still less active. In a previous study (Romero et al., 2019), we also observed that TMS affects both large pyramidal neurons and small (inhibitory) interneurons. The temporary hyperexcitability in our data may also be related to the rare occurrence of seizures after TMS (Lerner et al., 2019).

Because we concentrated on measuring the changes in neuronal excitability after cTBS, we did not map the area of cortex under the coil that was affected by cTBS. sTMS affects a surprisingly small volume of cortex, which we estimated to be not larger than 2 by 2 by 2 mm (Romero et al., 2019). If our hypothesis of the spreading of GABA is correct, we expect that a slightly larger volume of cortex will be affected by cTBS. Furthermore, our previous observation that sTMS affects a very small volume of cortex explains why a control site was not necessary in the current study: we assessed neuronal excitability by means of sTMS, and therefore moving the TMS coil to a different location would have made this assessment impossible. In theory, the possibility exists that the reduced neuronal excitability we measured under the TMS coil was an indirect effect caused by inactivation of an input area of PFG (e.g., neighboring area AIP or 7a). Even the inclusion of a remote control site would not entirely rule out this possibility because this control site would most likely not be connected to PFG and therefore would not cause any effect in PFG. We believe this theoretical possibility is unlikely because the induced electric field was maximal immediately under the coil where we recorded neuronal activity, and therefore this mechanism would imply the recruitment of another cortical area with a lower electric field. Nevertheless, cTBS and other reversible inactivation methods certainly evoke effects on connected remote areas (Davare et al., 2010; Bestmann et al., 2015), sometimes even far away from the inactivation site (see, fe.g., Van Dromme et al., 2016 for effects in inferior temporal cortex after reversible inactivation of AIP).

It is important to note that we can rule out nonspecific factors influencing the neurons under the TMS coil. For example, cutaneous stimulation could not drive our effects since the TMS coil was positioned on the implant of the animal which was composed of dental acrylic. Moreover, we could measure the reduction in neuronal excitability 30–60 min after the administration of cTBS in task-related neurons in the absence of sTMS. It is highly unlikely that these late effects of cTBS would have resulted from cutaneous stimulation or other nonspecific factors. Note also that we previously observed highly grasp-specific effects of cTBS in line with the role of PFG in processing object properties for grasping (Merken et al., 2021).

Our results were robust (around 30% reduction in response) and highly reproducible in two animals, which is the standard in monkey electrophysiology experiments. Moreover, Merken et al., 2021 also reported highly similar behavioral results in two different animals. However, future studies will have to determine to what extent the effects of cTBS are variable in a larger number of monkeys. Although the neural effects of cTBS were highly similar in the two animals we used, the effects of cTBS in human volunteers are notoriously variable across subjects (Hamada et al., 2013; Hordacre et al., 2017; Jannati et al., 2017). The reproducibility of our results was most likely related to the very controlled conditions in which we applied cTBS. In our setup, the TMS coil was rigidly anchored to the head implant of the animal, so that we kept both the position and the orientation of the coil similar across sessions. Also, another possibility is that monkeys become highly overtrained in the grasping task, which may partially explain the similar behavioral effects of cTBS we reported in Merken et al., 2021. It is therefore plausible to assume that the larger variability inherent to human behavior is one reason underlying the variability of cTBS effects in humans, since stimulation is applied over a brain area in subjects at different levels of learning stages and behavioral performance, ultimately impacting on the susceptibility of that brain area to cTBS and increasing the inter-individual variability of the technique.

The critical advantage of the macaque monkey model is that we can measure single-cell activity and behavioral performance before and after cTBS. In a subset of neurons that we could record for more than 1 hr, we did not observe any recovery of the excitability even after 2 hr. Consistent with this observation, the behavioral effect of cTBS also lasted at least 2 hr (Merken et al., 2021). Therefore, our cTBS paradigm in monkeys may have caused longer-lasting effects than in human volunteers, possibly due to the thinner skull of the monkey, which could induce a stronger effect in parietal cortex. We did not observe any effect on the neuronal excitability on consecutive days, since we could always record single-unit activity over several weeks of recordings.

Investigating the neural effects of non-invasive brain stimulation techniques requires adequate animal models, so that behavioral measurements and detailed recordings of individual neurons can be combined with neuromodulation. In future studies, other stimulation protocols such as intermittent theta-burst stimulation and other neuromodulation techniques such as transcranial alternating current stimulation can be investigated and optimized in animal models using a similar approach.

All data generated or analysed during this study are publicly available on Dryad https://doi.org/10.5061/dryad.ns1rn8pvr.

1. 1. Pita MC
   2. Merken L
   3. Janssen P
   4. Davare M

   (2022) Dryad Digital Repository

   Neural effects of continuous theta-burst stimulation in macaque parietal neurons.

   https://doi.org/10.5061/dryad.ns1rn8pvr

1. 1. Bestmann S
   2. de Berker AO
   3. Bonaiuto J

   (2015) Understanding the behavioural consequences of noninvasive brain stimulation

   Trends in Cognitive Sciences 19:13–20.

   https://doi.org/10.1016/j.tics.2014.10.003

   • PubMed
   • Google Scholar
2. 1. Caprara I
   2. Janssen P

   (2021) The causal role of three frontal cortical areas in grasping

   Cerebral Cortex 31:4274–4288.

   https://doi.org/10.1093/cercor/bhab085

   • PubMed
   • Google Scholar
3. 1. Chuang WL
   2. Huang YZ
   3. Lu CS
   4. Chen RS

   (2014) Reduced cortical plasticity and gabaergic modulation in essential tremor

   Movement Disorders 29:501–507.

   https://doi.org/10.1002/mds.25809

   • PubMed
   • Google Scholar
4. 1. Davare M
   2. Rothwell JC
   3. Lemon RN

   (2010) Causal connectivity between the human anterior intraparietal area and premotor cortex during grasp

   Current Biology 20:176–181.

   https://doi.org/10.1016/j.cub.2009.11.063

   • PubMed
   • Google Scholar
5. 1. Edwards MJ
   2. Huang YZ
   3. Mir P
   4. Rothwell JC
   5. Bhatia KP

   (2006) Abnormalities in motor cortical plasticity differentiate manifesting and nonmanifesting DYT1 carriers

   Movement Disorders 21:2181–2186.

   https://doi.org/10.1002/mds.21160

   • PubMed
   • Google Scholar
6. 1. Fitzgerald PB
   2. Fountain S
   3. Daskalakis ZJ

   (2006) A comprehensive review of the effects of rtms on motor cortical excitability and inhibition

   Clinical Neurophysiology 117:2584–2596.

   https://doi.org/10.1016/j.clinph.2006.06.712

   • PubMed
   • Google Scholar
7. 1. Goldsworthy MR
   2. Pitcher JB
   3. Ridding MC

   (2012) The application of spaced theta burst protocols induces long-lasting neuroplastic changes in the human motor cortex

   The European Journal of Neuroscience 35:125–134.

   https://doi.org/10.1111/j.1460-9568.2011.07924.x

   • PubMed
   • Google Scholar
8. 1. Hamada M
   2. Murase N
   3. Hasan A
   4. Balaratnam M
   5. Rothwell JC

   (2013) The role of interneuron networks in driving human motor cortical plasticity

   Cerebral Cortex 23:1593–1605.

   https://doi.org/10.1093/cercor/bhs147

   • PubMed
   • Google Scholar
9. 1. Hordacre B
   2. Goldsworthy MR
   3. Vallence AM
   4. Darvishi S
   5. Moezzi B
   6. Hamada M
   7. Rothwell JC
   8. Ridding MC

   (2017) Variability in neural excitability and plasticity induction in the human cortex: a brain stimulation study

   Brain Stimulation 10:588–595.

   https://doi.org/10.1016/j.brs.2016.12.001

   • PubMed
   • Google Scholar
10. 1. Huang YZ
    2. Edwards MJ
    3. Rounis E
    4. Bhatia KP
    5. Rothwell JC

    (2005) Theta burst stimulation of the human motor cortex

    Neuron 45:201–206.

    https://doi.org/10.1016/j.neuron.2004.12.033

    • PubMed
    • Google Scholar
11. 1. Huang YZ
    2. Rothwell JC
    3. Chen RS
    4. Lu CS
    5. Chuang WL

    (2011) The theoretical model of theta burst form of repetitive transcranial magnetic stimulation

    Clinical Neurophysiology 122:1011–1018.

    https://doi.org/10.1016/j.clinph.2010.08.016

    • PubMed
    • Google Scholar
12. 1. Ikezoe K
    2. Mori Y
    3. Kitamura K
    4. Tamura H
    5. Fujita I

    (2013) Relationship between the local structure of orientation map and the strength of orientation tuning of neurons in monkey V1: a 2-photon calcium imaging study

    The Journal of Neuroscience 33:16818–16827.

    https://doi.org/10.1523/JNEUROSCI.2209-13.2013

    • PubMed
    • Google Scholar
13. 1. Jannati A
    2. Block G
    3. Oberman LM
    4. Rotenberg A
    5. Pascual-Leone A

    (2017) Interindividual variability in response to continuous theta-burst stimulation in healthy adults

    Clinical Neurophysiology 128:2268–2278.

    https://doi.org/10.1016/j.clinph.2017.08.023

    • PubMed
    • Google Scholar
14. 1. Kirkwood A
    2. Bear MF

    (1994) Homosynaptic long-term depression in the visual cortex

    The Journal of Neuroscience 14:3404–3412.

    https://doi.org/10.1523/JNEUROSCI.14-05-03404.1994

    • PubMed
    • Google Scholar
15. 1. Koch G
    2. Di Lorenzo F
    3. Bonnì S
    4. Ponzo V
    5. Caltagirone C
    6. Martorana A

    (2012) Impaired ltp- but not ltd-like cortical plasticity in alzheimer’s disease patients

    Journal of Alzheimer’s Disease 31:593–599.

    https://doi.org/10.3233/JAD-2012-120532

    • PubMed
    • Google Scholar
16. 1. Lerner AJ
    2. Wassermann EM
    3. Tamir DI

    (2019) Seizures from transcranial magnetic stimulation 2012-2016: results of a survey of active laboratories and clinics

    Clinical Neurophysiology 130:1409–1416.

    https://doi.org/10.1016/j.clinph.2019.03.016

    • PubMed
    • Google Scholar
17. 1. Massey PV
    2. Bashir ZI

    (2007) Long-term depression: multiple forms and implications for brain function

    Trends in Neurosciences 30:176–184.

    https://doi.org/10.1016/j.tins.2007.02.005

    • PubMed
    • Google Scholar
18. 1. Merken L
    2. Davare M
    3. Janssen P
    4. Romero MC

    (2021) Behavioral effects of continuous theta-burst stimulation in macaque parietal cortex

    Scientific Reports 11:4511.

    https://doi.org/10.1038/s41598-021-83904-8

    • PubMed
    • Google Scholar
19. 1. Mori F
    2. Nisticò R
    3. Mandolesi G
    4. Piccinin S
    5. Mango D
    6. Kusayanagi H
    7. Berretta N
    8. Bergami A
    9. Gentile A
    10. Musella A
    11. Nicoletti CG
    12. Nicoletti F
    13. Buttari F
    14. Mercuri NB
    15. Martino G
    16. Furlan R
    17. Centonze D

    (2014) Interleukin-1β promotes long-term potentiation in patients with multiple sclerosis

    Neuromolecular Medicine 16:38–51.

    https://doi.org/10.1007/s12017-013-8249-7

    • PubMed
    • Google Scholar
20. 1. Mueller JK
    2. Grigsby EM
    3. Prevosto V
    4. Petraglia FW
    5. Rao H
    6. Deng ZD
    7. Peterchev AV
    8. Sommer MA
    9. Egner T
    10. Platt ML
    11. Grill WM

    (2014) Simultaneous transcranial magnetic stimulation and single-neuron recording in alert non-human primates

    Nature Neuroscience 17:1130–1136.

    https://doi.org/10.1038/nn.3751

    • PubMed
    • Google Scholar
21. 1. Munneke MAM
    2. Rongen JJ
    3. Overeem S
    4. Schelhaas HJ
    5. Zwarts MJ
    6. Stegeman DF

    (2013) Cumulative effect of 5 daily sessions of θ burst stimulation on corticospinal excitability in amyotrophic lateral sclerosis

    Muscle & Nerve 48:733–738.

    https://doi.org/10.1002/mus.23818

    • PubMed
    • Google Scholar
22. 1. Oberman L
    2. Ifert-Miller F
    3. Najib U
    4. Bashir S
    5. Woollacott I
    6. Gonzalez-Heydrich J
    7. Picker J
    8. Rotenberg A
    9. Pascual-Leone A

    (2010) Transcranial magnetic stimulation provides means to assess cortical plasticity and excitability in humans with fragile x syndrome and autism spectrum disorder

    Frontiers in Synaptic Neuroscience 2:26.

    https://doi.org/10.3389/fnsyn.2010.00026

    • PubMed
    • Google Scholar
23. 1. Opie GM
    2. Catcheside PG
    3. Usmani ZA
    4. Ridding MC
    5. Semmler JG

    (2013) Motor cortex plasticity induced by theta burst stimulation is impaired in patients with obstructive sleep apnoea

    The European Journal of Neuroscience 37:1844–1852.

    https://doi.org/10.1111/ejn.12203

    • PubMed
    • Google Scholar
24. 1. Orth M
    2. Schippling S
    3. Schneider SA
    4. Bhatia KP
    5. Talelli P
    6. Tabrizi SJ
    7. Rothwell JC

    (2010) Abnormal motor cortex plasticity in premanifest and very early manifest huntington disease

    Journal of Neurology, Neurosurgery, and Psychiatry 81:267–270.

    https://doi.org/10.1136/jnnp.2009.171926

    • PubMed
    • Google Scholar
25. 1. Pani P
    2. Theys T
    3. Romero MC
    4. Janssen P

    (2014) Grasping execution and grasping observation activity of single neurons in the macaque anterior intraparietal area

    Journal of Cognitive Neuroscience 26:2342–2355.

    https://doi.org/10.1162/jocn_a_00647

    • PubMed
    • Google Scholar
26. 1. Pitcher D
    2. Parkin B
    3. Walsh V

    (2021) Transcranial magnetic stimulation and the understanding of behavior

    Annual Review of Psychology 72:97–121.

    https://doi.org/10.1146/annurev-psych-081120-013144

    • PubMed
    • Google Scholar
27. 1. Romero MC
    2. Davare M
    3. Armendariz M
    4. Janssen P

    (2019) Neural effects of transcranial magnetic stimulation at the single-cell level

    Nature Communications 10:2642.

    https://doi.org/10.1038/s41467-019-10638-7

    • PubMed
    • Google Scholar
28. 1. Stagg CJ
    2. Best JG
    3. Stephenson MC
    4. O’Shea J
    5. Wylezinska M
    6. Kincses ZT
    7. Morris PG
    8. Matthews PM
    9. Johansen-Berg H

    (2009) Polarity-sensitive modulation of cortical neurotransmitters by transcranial stimulation

    The Journal of Neuroscience 29:5202–5206.

    https://doi.org/10.1523/JNEUROSCI.4432-08.2009

    • PubMed
    • Google Scholar
29. 1. Suppa A
    2. Belvisi D
    3. Bologna M
    4. Marsili L
    5. Berardelli I
    6. Moretti G
    7. Pasquini M
    8. Fabbrini G
    9. Berardelli A

    (2011) Abnormal cortical and brain stem plasticity in gilles de la tourette syndrome

    Movement Disorders 26:1703–1710.

    https://doi.org/10.1002/mds.23706

    • PubMed
    • Google Scholar
30. 1. Suppa A
    2. Marsili L
    3. Di Stasio F
    4. Berardelli I
    5. Roselli V
    6. Pasquini M
    7. Cardona F
    8. Berardelli A

    (2014a) Cortical and brainstem plasticity in tourette syndrome and obsessive-compulsive disorder

    Movement Disorders 29:1523–1531.

    https://doi.org/10.1002/mds.25960

    • PubMed
    • Google Scholar
31. 1. Suppa A
    2. Marsili L
    3. Di Stasio F
    4. Latorre A
    5. Parvez AK
    6. Colosimo C
    7. Berardelli A

    (2014b) Primary motor cortex long-term plasticity in multiple system atrophy

    Movement Disorders 29:97–104.

    https://doi.org/10.1002/mds.25668

    • PubMed
    • Google Scholar
32. 1. Suppa A
    2. Huang YZ
    3. Funke K
    4. Ridding MC
    5. Cheeran B
    6. Di Lazzaro V
    7. Ziemann U
    8. Rothwell JC

    (2016) Ten years of theta burst stimulation in humans: established knowledge, unknowns and prospects

    Brain Stimulation 9:323–335.

    https://doi.org/10.1016/j.brs.2016.01.006

    • PubMed
    • Google Scholar
33. 1. Tang R
    2. Song Q
    3. Li Y
    4. Zhang R
    5. Cai X
    6. Lu HD

    (2020) Curvature-processing domains in primate v4

    eLife 9:e57502.

    https://doi.org/10.7554/eLife.57502

    • PubMed
    • Google Scholar
34. 1. Thut G
    2. Pascual-Leone A

    (2010) A review of combined TMS-EEG studies to characterize lasting effects of repetitive TMS and assess their usefulness in cognitive and clinical neuroscience

    Brain Topography 22:219–232.

    https://doi.org/10.1007/s10548-009-0115-4

    • PubMed
    • Google Scholar
35. 1. Van Dromme IC
    2. Premereur E
    3. Verhoef BE
    4. Vanduffel W
    5. Janssen P

    (2016) Posterior parietal cortex drives inferotemporal activations during three-dimensional object vision

    PLOS Biology 14:e1002445.

    https://doi.org/10.1371/journal.pbio.1002445

    • PubMed
    • Google Scholar
36. 1. Vigneswaran G
    2. Kraskov A
    3. Lemon RN

    (2011) Large identified pyramidal cells in macaque motor and premotor cortex exhibit “thin spikes”: implications for cell type classification

    The Journal of Neuroscience 31:14235–14242.

    https://doi.org/10.1523/JNEUROSCI.3142-11.2011

    • PubMed
    • Google Scholar
37. 1. Vlachos A
    2. Müller-Dahlhaus F
    3. Rosskopp J
    4. Lenz M
    5. Ziemann U
    6. Deller T

    (2012) Repetitive magnetic stimulation induces functional and structural plasticity of excitatory postsynapses in mouse organotypic hippocampal slice cultures

    The Journal of Neuroscience 32:17514–17523.

    https://doi.org/10.1523/JNEUROSCI.0409-12.2012

    • PubMed
    • Google Scholar
38. 1. Woloszyn L
    2. Sheinberg DL

    (2012) Effects of long-term visual experience on responses of distinct classes of single units in inferior temporal cortex

    Neuron 74:193–205.

    https://doi.org/10.1016/j.neuron.2012.01.032

    • PubMed
    • Google Scholar

Author details

1. Maria C Romero

   Laboratorium voor Neuro- en Psychofysiologie, The Leuven Brain Institute, Leuven, Belgium

   Contribution

   Conceptualization, Data curation, Formal analysis, Investigation, Visualization, Methodology, Writing – original draft

   Competing interests

   No competing interests declared

   "This ORCID iD identifies the author of this article:" 0000-0001-7758-2211

2. Lara Merken

   Laboratorium voor Neuro- en Psychofysiologie, The Leuven Brain Institute, Leuven, Belgium

   Contribution

   Data curation, Formal analysis, Investigation, Methodology

   Competing interests

   No competing interests declared

3. Peter Janssen

   Laboratorium voor Neuro- en Psychofysiologie, The Leuven Brain Institute, Leuven, Belgium

   Contribution

   Conceptualization, Resources, Supervision, Funding acquisition, Writing – original draft, Project administration, Writing – review and editing

   Contributed equally with

   Marco Davare

   For correspondence

   peter.janssen@kuleuven.be

   Competing interests

   No competing interests declared

   "This ORCID iD identifies the author of this article:" 0000-0002-8463-5577

4. Marco Davare

   Faculty of Life Sciences and Medicine, King's College London, London, United Kingdom

   Contribution

   Conceptualization, Resources, Writing – review and editing, Funding acquisition, Methodology

   Contributed equally with

   Peter Janssen

   For correspondence

   marco.davare@kcl.ac.uk

   Competing interests

   No competing interests declared

   "This ORCID iD identifies the author of this article:" 0000-0002-4670-0251

• 1,191

  views

• 177

  downloads

• 16

  citations

Views, downloads and citations are aggregated across all versions of this paper published by eLife.