Contents lists available at SciVerse ScienceDirect

Brain Stimulation

journal homepage: www.brainst imjrnl .com

Brain Stimulation xxx (2013) 1e7

Original Research

Continuous Theta Burst Stimulation of the Supplementary Motor Area: EffectUpon Perception and Somatosensory and Motor Evoked Potentials

Wynn Legon, Jennifer K. Dionne, W. Richard Staines*

Department of Kinesiology, University of Waterloo, 200 University Ave. West, Waterloo, Ontario N2L 3G1, Canada

a r t i c l e i n f o

Article history:Received 14 November 2012Received in revised form22 April 2013Accepted 23 April 2013Available online xxx

Keywords:Theta burst stimulationSomatosensory evoked potentialsN30Tactile perceptionSupplementary motor areaMotor evoked potentials

This work was supported by funding to WRS froEngineering Research Council of Canada (NSERC), tProgram, the Canada Foundation for Innovation and thand JKD were supported by graduate scholarship fund* Corresponding author. Tel.: þ1 519 888 4567x377

E-mail address: rstaines@uwaterloo.ca (W.R. Staine

1935-861X/$ e see front matter � 2013 Elsevier Inc. Ahttp://dx.doi.org/10.1016/j.brs.2013.04.007

a b s t r a c t

Background: The supplementary motor area (SMA) has been implicated in many aspects of movementpreparation and execution. In addition to motor roles, the SMA is responsive to somesthetic stimulithough it is unclear exactly what role the SMA plays in a somatosensory network.Objective/Hypothesis: It is the purpose of this study to assess how continuous theta burst stimulation(cTBS) of the SMA affects both somatosensory (SEPs) and motor evoked potentials (MEPs) and if cTBSleads to alterations in tactile perception thresholds of the index fingertip.Methods: In experiment 1, cTBS was delivered over scalp sites FCZ (SMA stimulation) (n ¼ 10) and CZ(control stimulation) (n ¼ 10) in separate groups for 40 s (600 pulses) at 90% of participants’ restingmotor threshold. For both groups, median nerve SEPs were elicited from the right wrist at rest viaelectrical stimulation (0.5 ms pulse) before and at 10 min intervals post-cTBS out to 30 min (t ¼ pre, 10,20, and 30 min). Subjects’ perceptual thresholds were assessed at similar time intervals as the SEP datausing a biothesiometer (120 Hz vibration). In experiment 2 (n ¼ 10) the effect of cTBS to SMA upon singleand paired-pulse MEP amplitudes from the right first dorsal interosseous (FDI) was assessed.Results: cTBS to scalp site FCZ (SMA stimulation) reduced the frontal N30 SEP and increased tactileperceptual thresholds 30 min post-stimulation. However, parietal SEPs and MEP amplitudes from bothsingle and paired-pulse stimulation were unaffected at all time points post-stimulation. cTBS to stimu-lation site CZ (control) did not result in any physiological or behavioral changes.Conclusion(s): These data demonstrate cTBS to the SMA reduces the amplitude of the N30 coincidentwith an increase in vibration sensation threshold but does not affect primary somatosensory or motorcortex excitability. The SMA may play a significant role in a somatosensory tactile attention network.

� 2013 Elsevier Inc. All rights reserved.

Introduction

The supplementary motor area (SMA) is classically associatedwith various aspects of movement preparation and execution [1]. Inaddition to these motor roles the SMA has been demonstrated to beresponsive to somesthetic stimuli in primates [2,3] and humans[4,5]. The SMA receives peripheral afferent input via the thalamus[6] and from post-rolandic parietal areas [7,8] supporting a role forthe SMA in a somatosensory network. In addition, SMA respon-siveness to peripheral afferent input is inferred by the N30 frontalsomatosensory evoked potential elicited by electrical median nerve

m the Natural Sciences andhe Canada Research Chairse Ontario Research Fund. WLs from NSERC.56; fax: þ1 519 885 0470.s).

ll rights reserved.

stimulation [9]. It is unclear however, exactly what role the SMAplays in a somatosensory network though the SMA has beenidentified as part of a tactile attention network [5] and hypothe-sized to link somatic sensation to action [10]. Further, lesion ofthe SMA has been associated with impairment in a temporaldiscrimination task [11]. Transcranial magnetic stimulation (TMS)of the SMA has been shown to disrupt various aspects of motorperformance [12e14] and to affect motor cortex excitability [15,16]but no literature to date has explored how transient inhibition ofthe SMA using TMS affects somatosensory evoked potentialsand tactile perception. In contrast, there are numerous papersexploring howmagnetic stimulation of the primary somatosensory[17e25], parietal [26,27] and motor cortex [17,20,28] affects tactileperception.

TMS protocols similar to those that affect behavioral perfor-mance delivered to both M1 and S1 also modulate somatosensoryevoked potentials [25,29e31]. Continuous theta burst stimulation(cTBS) [32] delivered over both primary motor and somatosensory

Figure 1. Experimental setup. Pictorial representation (not to scale) of coil place-ment(s) and time-line for each experiment. Experiment 1 examined the effect ofcontinuous theta burst stimulation (cTBS) to scalp sites FCZ (3 cm anterior to CZ) andCZ upon somatosensory evoked potentials (SEPs). Experiment 2 examined the effect ofcTBS to FCZ upon motor evoked potentials (MEPs) derived from the left primary motorcortex (M1). The bottom box outlines the time of respective recordings. (Pre) prior tocTBS and at times 10, 20, and 30 min post-cTBS.

W. Legon et al. / Brain Stimulation xxx (2013) 1e72

cortex has been shown to also affect median nerve somatosensoryevoked potentials but with differing effects. cTBS over M1 resultedin an increase of N30 amplitude and parietal SEP componentswhereas cTBS of S1 suppressed the same components [33]. Unfor-tunately, none of the above studies investigated both tactileperceptual thresholds and SEP effects and specifically effects uponthe frontal N30.

The frontal N30 SEP is a large negative potential recordedmaximally over frontal central scalp electrode sites and as such hasbeen suggested to be generated by the underlying cortex, mostnotably the SMA [9], though an exact generator has not yet beenestablished. The N30 is likely generated independently of parietalpotentials as it is spared with an S1 lesion [34] and specificallyattenuated as a result of meningioma of the falx cerebri com-pressing the SMA [35]. It is hypothesized that N30 amplitude islargely the result of proprioceptive afference [36,37] and previousresearch suggests a link between N30 amplitude and sensorimotorintegration independent of parietal potentials as it is particularlyaffected in Parkinson’s disease [38], is facilitated during movementof the contralateral limb [39,40] but attenuated by motor imageryor ideation tasks [41].

The purpose of this study was twofold: 1) to investigate howtransient inhibition of the SMA using cTBS affects the frontal N30somatosensory evoked potential and further, to determine if thisresults in any perceptual consequences, and 2) to test if cTBS of theSMA affects motor cortex excitability as a possible means ofsomatosensory effects. It was hypothesized that inhibitory stimu-lation of the SMA would attenuate the frontal N30 and that suchattenuation of somatosensory input to the SMA would be accom-panied by alterations in tactile perceptual measures. Furthermore,it was expected that cTBS of the SMA would attenuate the ampli-tude of MEPs.

Methods

Participants

A total of 27 subjects participated in the two experiments. Twogroups of subjects participated in experiment 1 on separate days.Each group consisted of 10 participants (4 female, age 24 � 3.6 yrs),(3 female, age 23 � 3.3 yrs). Three participants were included inboth groups. Ten subjects participated in experiment 2 (3 female;25.6 � 4.6 yrs) performed on separate days from experiment 1 totest the effects of cTBS to SMA on motor cortical excitability. Allsubjects were self-report right hand dominant and providedwritten informed consent to participate. None reported any historyof neurological or musculoskeletal impairments or any contra-indicators for TMS. All were paid a nominal fee for their participa-tion. The University of Waterloo Office of Research Ethics approvedall experimental procedures.

Experiment 1 e SEP and perceptual detection

Behavioral taskParticipants were seated in a desk chair with elbow and forearm

of both arms resting on a platform upon a tabletop. The platformallowed for the hand to rest over the far edge in slight flexion of thewrist and allowed for a comfortable resting of the index finger upona vibrating post-placed at the end of the raised platform. Individualsomatosensory thresholds were determined on the right indexfinger using the method of limits with a Vibratron II biothesiometer(Physitemp Instruments, Clifton, NJ, USA) (120 Hz vibration) for alltime points of testing (pre-cTBS, 10, 20, and 30 min). To ensure thatsubjects were not using a time estimation strategy to reportperception, the timing of displacement increase was random and

non-increase catch trials were interspersed amongst true increases.All subjects performed the tactile judgment task with their eyesclosed. Participants were familiarized with the vibration sensationbefore testing and instructed to relax and not move their finger orarm and to be sure they felt the vibration before reporting it. Toestablish a pre-testing baseline, participants repeated the thresholdtesting until three consecutive trials were within one vibration unitof each other. Vibration units (X) are related to the true amplitude(A) of post-excursion in microns by the following formula:A ¼ (0.5) X2. Participants were not informed of these values at anypoint of the testing and were naïve to the purposes of the study. Foreach time point of testing post-cTBS, 3 repeats of the tactile judg-ment were performed resulting in one average value. Vibrationthreshold testing was performed pre-cTBS (10 min prior) and attime points 10, 20, and 30 min post-cTBS (see Fig. 1).

Stimulation and recordingSEPs were derived from the electrical stimulation of the median

nerve of the dominant wrist. Square wave pulses of 0.5 ms duration(GRASS S88 stimulator with SIU5 stimulus isolation unit; WestWarwick, Rhode Island, USA) were delivered through a bar elec-trode, with the anode distal, fixed over the median nerve. Stimu-lation occurred at a constant rate of 1 Hz and at an intensitysufficient to produce a small but noticeable thumb twitch. Surfaceelectromyography (EMG) was recorded from the thenar muscula-ture to record the M-wave, an EMG wave resulting from the directstimulation of the motoneuronal axons serving the thenar muscu-lature to ensure consistency of stimulation. EMG recordings wereamplified (2000�), band-pass filtered (20e200 Hz), digitized andstored for later analysis. Electroencephalographic (EEG) data wererecorded from two AgeAgCl cup electrodes fixed to the scalp andreferenced to the linked mastoids. One electrode was placed 3 cmanterior to site CZ and the other over a spot corresponding toelectrode site CP3 (contralateral to MN stimulation) in accordancewith the international 10e20 system for electrode placement. Datawere amplified (40,000�), filtered (2e200 Hz) and digitized at1000 Hz (NeuroScan 4.3; Compumedics; Charlotte, NC, USA).SEPs were extracted by averaging epochs time-locked to mediannerve stimulation (�50 to 300 ms). All traces were visuallyinspected for artifact (blinks, eye movements or contraction ofscalp musculature) and any contaminated epochs were eliminatedbefore averaging. All traces were the result of 200 randomlychosen stimulations. Electrodes were removed during the cTBS

Figure 2. Group average N30 SEP amplitudes. N30 amplitude group data for experi-ment 1 (n ¼ 10) measured from electrode site FCZ. Black bars represent data for SMAstimulation; White bars for control stimulation. Ordinate is amplitude in microvolts(mV). Abscissa is time of recording before (pre-) and post-cTBS stimulation in minutes(min). Bars represent �SEM. Note: amplitudes are absolute values for display purposes.* denotes significance P < 0.05.

W. Legon et al. / Brain Stimulation xxx (2013) 1e7 3

stimulation. The position was marked on the scalp using a felt-tipped marker and electrodes were replaced and impedancesmeasured to achieve pre-cTBS values.

Transcranial magnetic stimulationFor group 1, continuous theta burst stimulation for 40 s (600

pulses) of 3 stimuli at 50 Hz repeated at 5 Hz [32] was applied usinga MagPro stimulator (Medtronic, Minneapolis, MN, USA) and ‘figureof eight’ coil (MCF-B65) at 90% resting motor threshold (RMT) toa scalp site 3 cm anterior to site CZ in accordance with the inter-national 10e20 system for electrode placement (site for SMAstimulation). This site was based upon previous neuroimaging[42,43] and TMS studies targeting the SMA [16,44]. For group 2, theintersection of the coil was placed over site CZ (site for controlstimulation). These coil placements will heretofore be referred to asSMA and CONTROL stimulation sites respectively. For both experi-ments, the handle of the coil was directed posteriorly along themid-line. Prior to theta burst stimulation, the RMT for the right firstdorsal interosseous (FDI) was determined for each participant withthe coil handle pointing backward and laterally at 45� frommidlineover the left motor cortex. RMT was determined as the loweststimulus intensity at which 5 of 10 consecutive stimuli eliciteda reliable MEP of at least 50 mV.

Data analysisSomatosensory evoked potentials. Latencies and amplitudes of thefrontal and parietal SEPs were measured from the individualparticipant averages for each time point of interest from electrodeFCZ and CP3. Latencies were measured from onset of stimulation tothe peak of each SEP component of interest (frontal: P18, N30, andN60; parietal: N20, P27, and P50). A clearly defined peak wasnecessary for inclusion. Amplitudes were measured as raw ampli-tude relative to a pre-stimulus baseline (50 ms). A mixed two-wayanalysis of variance (ANOVA) was performed for each SEP compo-nent of interest with between subject factor STIMULATION SITE(SMA, CONTROL) and within subject factor TIME (pre, 10, 20, and30 min) post-cTBS. The specific hypothesis for an effect of SMAstimulation upon N30 amplitude was tested with a one-wayANOVA for each stimulation site with TIME as the factor and post-hoc Tukey’s tests to compare differences between the individualtime points if there was a main effect of TIME.

A similar mixed two-way ANOVA was performed to testperceptual thresholds. Significant effects were explored with post-hoc one-way ANOVAs and t-tests. Further, to assess the linearrelationship between N30 amplitude and threshold levels, a Pear-son’s correlation was performed on data normalized to pre-cTBSvalues for both SMA and CONTROL collapsed across time. In allcases significance was taken as P < 0.05.

Experiment 2 e motor evoked potentials (MEPs)

Participants were seated in a desk chair with forearm and handresting palm down upon a modified armrest. The participants’ headwas supported by a chin rest and a forehead crutch to prevent headmovement during MEP testing. During testing participants wereinstructed to fixate on a point in front of them while maintaininga relaxed posture.

Stimulation and recordingPrior to cTBS, the RMT for the right FDI was determined as for

Experiment 1. Average single-pulse MEPs were the result of 10single-pulse stimulations delivered approximately 2 s apart at120% RMT and collected at time points 10 min prior to cTBSand 10 min post-cTBS. Short-interval intracortical inhibition (SICI)was assessed using the inhibitory paired-pulse paradigm with

a conditioning stimulus of 80% RMT delivered 3 ms prior to a teststimulus of 120% RMT. Average paired-pulse MEP amplitudes werethe result of 10 individual trials. The cTBS protocol was identical tothat described for Experiment 1 above (see Fig. 1).

Data analysisTo compare the effect of cTBS to SMA upon M1 excitability

individual paired t-tests were performed for both single and paired-pulse protocols pre- and post-stimulation.

Results

Experiment 1

Latencies and M-waveAll participants displayed clear frontal and parietal potentials.

There were no differences in the latencies of any of the potentials ofinterest between groups or across time. Average latencies are pre-sented collapsed across group and time. Frontal P18 (18 � 1.2 ms),N30 (33 � 4.7 ms), N60 (65 � 8.1 ms); Parietal N20 (19.7 � 1.6 ms),P27 (25.6 � 2.7 ms), P50 (47 � 7.1 ms). There were no significantdifferences in M-wave amplitudes across conditions.

Somatosensory evoked potentialsFrontal potentials (N30 and N60). For the N30 component, themixed two-way ANOVA revealed a significant main effect of TIME[F(3,54) ¼ 3.12, P ¼ 0.034] and no interaction of STIMULATIONSITE � TIME [F(3,54) ¼ 1.40, P ¼ 0.255] and no between subjectseffect of STIMULATION SITE [F(1,18) ¼ 0.84, P ¼ 0.375]. The specifichypothesis of an effect of TBS stimulation of SMA upon N30amplitude was tested with separate one-way repeated measuresANOVAs for each stimulation site. The one-way repeated measuresANOVA for STIMULATION SITE: CONTROL revealed nomain effect ofTIME [F(3,36) ¼ 1.22, P ¼ 0.301] whereas the one-way repeatedmeasures ANOVA for STIMULATION SITE: SMA revealed a maineffect of TIME [F(3,36) ¼ 3.41, P¼ 0.033] (Figs. 2 and 3, and Table 1).Post-hoc Tukey’s tests revealed that N30 amplitude was signifi-cantly attenuated at 30 min post-cTBS compared to pre-cTBS (P <

0.05), but not at 10 or 20 min post-cTBS (see Figs. 1 and 2A, andTable 1). There were no significant main effects or interactions forthe P18 or N60. STIMULATION � TIME interaction values reportedbelow: P18: [F(3,54) ¼ 0.74, P ¼ 0.545], N60: [F(3,54) ¼ 0.90,P ¼ 0.456] (Figs. 2 and 3, and Table 1).

Figure 3. Group average frontal and parietal somatosensory evoked potentials. Group(n ¼ 10) SEP traces recorded from electrode positions FCZ for frontal potentials (toptwo traces) and CP3 for parietal potentials (bottom two traces). SMA denotes cTBSdelivered to scalp site FCZ; control denotes cTBS delivered to scalp site CZ. Potentials ofinterest are marked. Vertical line represents onset of stimulation. Positive amplitude isup. Recordings presented before (pre-) and at time points (10, 20, and 30) in minutesfollowing cTBS. * denotes significant difference P < 0.05.

Table 1Frontal and parietal SEP amplitudes. Mean amplitudes are shown in bold with SEbelow.

SEP SMA Control

Pre 10 20 30 Pre 10 20 30

FrontalN30 L3.81 L3.64 L3.17 L2.63 L4.07 L4.00 L3.88 L4.03

0.48 0.46 0.63 0.67 0.50 0.67 0.71 0.68N60 L2.94 L2.79 L2.62 L2.25 L3.50 L2.74 L2.96 L3.10

0.55 0.63 0.80 0.71 0.76 0.83 0.55 0.75ParietalN20 L1.71 L1.94 L1.49 L1.62 L2.13 L1.96 L1.73 L1.82

0.21 0.29 0.24 0.24 0.28 0.27 0.21 0.25P27 3.20 2.85 2.58 2.66 3.37 3.33 3.06 3.31

0.80 0.66 0.71 0.57 0.92 1.02 0.92 1.06P50 4.96 4.84 4.69 4.51 3.30 4.19 2.93 3.30

0.61 0.83 0.76 0.64 0.40 0.65 0.33 0.69

Figure 4. Group average vibration thresholds. Group (n ¼ 10) data of behavioralperception thresholds in vibration units. Black bars represent data from cTBS to scalpsite FCZ (SMA); White bars represent results from cTBS to scalp site CZ (control).Abscissa represents time points of testing in minutes (min) before cTBS (pre-) andpost-cTBS (10, 20, and 30). Bars are �SEM. * denotes P < 0.05.

W. Legon et al. / Brain Stimulation xxx (2013) 1e74

Parietal potentials (N20, P27, and P50). There were no significantmain effects or interactions for any of the parietal potentials thoughthere was a trend for the P50 for the main effect of STIMULATIONSITE: [F(1,18) ¼ 3.02, P ¼ 0.099] where the P50 tended to be largerfor the TBS condition. STIMULATION � TIME interaction values re-ported below: N20 [F(3,54) ¼ 0.57, P ¼ 0.647]; P27 [F(3,54) ¼ 0.217,P ¼ 0.892]; P50 [F(3,54) ¼ 0.886, P ¼ 0.455] (Fig. 3 and Table 1).

Behavioral detection thresholdThe mixed two-way ANOVA revealed a strong trend for an

interaction of STIMULATION SITE � TIME [F (3,54) ¼ 2.74,P ¼ 0.052]. The interaction was explored with separate one-way

repeated measures ANOVA for each stimulation site. The one-wayrepeated measures ANOVA for CONTROL revealed no main effect ofTIME [F(3,36) ¼ 0.17, P ¼ 0.926] whereas the one-way repeatedmeasures ANOVA for SMA revealed a significant main effect ofTIME [F(3,36) ¼ 8.24, P < 0.001]. Post-hoc two-tailed paired t-testsrevealed nodifference between timepoints pre and 10 [t(9)¼�0.43,P ¼ 0.684] or 20 [t(9) ¼ �1.79, P ¼ 0.108] min post-cTBS, buta significant difference between pre- and 30 min post-cTBS[t(9) ¼ �4.09, P ¼ 0.003] (Fig. 4).

There was no statistically significant linear relationship betweenN30 amplitude and sensory threshold [r(58) ¼ 0.0821, P ¼ 0.533](see Fig. 5).

Experiment 2 e MEPs

Two-tailed paired t-tests for both single-pulse and paired-pulseprotocols revealed no effect of cTBS to SMA: Single pulse[t(9)¼ 0.357, P¼ 0.733; paired pulse t(9)¼ 0.827, P¼ 0.431] (Fig. 6).

Discussion

It was the main purpose of this study to test the effect ofcontinuous theta burst stimulation to the SMA upon frontal N30amplitude and tactile perceptual thresholds. The results demon-strated that cTBS of the SMA attenuated N30 SEP amplitudes 30minafter stimulation, which coincided well in time with an increase in

Figure 5. Correlation of N30 amplitude and sensory threshold. group (n ¼ 20) datafrom SMA and control cTBS stimulation across time points 10, 20, and 30 min. Ordinaterepresents N30 amplitude normalized to pre-cTBS values. Abscissa represents vibra-tory threshold normalized to pre-cTBS values.

W. Legon et al. / Brain Stimulation xxx (2013) 1e7 5

tactile perceptual thresholds of the index finger. Continuous TBS toSMA had neither effect upon the amplitude of any of the measuredparietal potentials (N20, P27, and P50) nor effect upon single- orpaired-pulse MEP amplitudes. The results support the hypothesisthat inhibition of the SMA attenuates the N30 coincident with anincrease in tactile perceptual threshold.

Only two studies to date have assessed the effect of TMS uponN30 amplitude. Urushihara et al. [45] delivered very low frequency(0.2 Hz) monophasic stimulation to the primary motor cortex (M1),pre-motor cortex (PMd) and SMA and reported an increase in N30amplitude for PMd stimulation only. Interestingly, biphasic 1 Hzstimulation to the same PMd location had no effect. Ishikawa et al.[33] delivered cTBS to left M1 and a point 2 cm posterior to this andreported the frontal P22/N30 to increase for M1 stimulation but todecrease for S1 stimulation as recorded from an electrode site 5 cmanterior to M1. The results of these two studies are difficult torectify with the results from the present study. The increase in N30amplitude in Urushihara et al. [45] is likely due to differences in thestimulation parameters used. Both the frequency as well as thepulse phase have been demonstrated to elicit different patterns ofdescending waves as measured from cervical epidural electrodessuggesting preferential activation of different sets of excitatorycortico-cortical fibers (for review see Ref. [46]). The absence ofeffect of the 1 Hz biphasic stimulation upon N30 amplitude may bedue to either stimulus location (PMd vs. SMA) or perhaps the timingof recording post-stimulation as effects upon N30 amplitude in thisstudy were not statistically significant until 30 min post-cTBS. The

Figure 6. Group average motor evoked potentials. Group (n ¼ 10) motor evokedpotential amplitudes recorded from right FDI for experiment 2. SP ¼ single pulse;PP ¼ paired pulse; SP TBS ¼ single-pulse after TBS; PP TBS ¼ paired-pulse after TBS.Values are presented normalized to SP (1.00). Bars are �SEM.

results of Ishikawa et al. [33] do not suggest a specific relationshipbetween SEP and MEP effects. Continuous TBS to both M1 and S1both had effects upon N30 amplitude, albeit opposite, but only M1stimulation had effect uponMEP amplitude. The relationship, if any,between cTBS effects and N30 amplitude so far suggests thata decrease in N30 amplitude from both S1 stimulation [33] and SMAstimulation from this study does not result in a concomitant or anyeffect upon cortico-motor excitability as assessed by MEP ampli-tude. cTBS to both S1 and M1 in Ishikawa et al. [33] had effect uponmultiple SEP components but only cTBS to SMA as evidenced fromthis study resulted in a specific effect upon the N30, leaving bothMEP and parietal potentials unaffected.

Behavioral effects

SEP amplitude is generally acknowledged to represent thesummation of synchronous excitatory post-synaptic potentials ofa large neuronal population though recent research has attributedN30 generation to beta/gamma phase locking [47]. Regardless ofthe specific generation, the link between SEP amplitude andbehavior is not well understood and no fixed relationship has beenestablished. Despite this, conscious attention to a supraliminaltactile stimulus has been demonstrated to increase the firing rate ofS1 neurons [48] and to facilitate early parietal potentials generatedin S1 [49e51]. There is no evidence however, that attentionmodulates the N30 SEP [52]. This is somewhat surprising as the N30does display similar effects to parietal potentials under varioussensory-motor paradigms, attention affects cortico-motor excit-ability [53] and the SMA has been previously identified as a locus ina tactile attention network [5].

One potential hypothesis for a dissociation of N30 effects fromearly parietal potentials with tactile attention may be due to thedifference in response properties of cells in S1 and SMA. Single cellstudies in non-human primates [10] as well as BOLD response inhumans [54,55] have shown that S1 responds to passive tactilestimulation but SMA only responds to tactile stimulation thatrequires or is paired to a motor response. Indeed, Romo et al. [10]have suggested that the SMA’s role may be to link perception toaction and it may be involved in the decision-making process inresponding to a stimulus and not perception per se. Under thishypothesis, a reduction in the N30 SEP in this study does notnecessarily suggest that its amplitude is directly linked to consciousperception but rather it may simply be a neurophysiological indi-cator of an inhibited SMA that may serve to alter or disrupt thedecision-making process for action. This line of reasoning isconsistent with research suggesting that the SMA is part of a fronto-parietal network activated during goal-directed task performance[56] including vibrotactile attention [57]. Lesion evidence furthersupports this. Lesion of either the thalamocortical projections to S1or S1 itself results in primary sensory deficits and impaired tactiledetection whereas lesion of the SMA leaves primary sensorydetection intact but disrupts tactile discrimination [11]. The preciserole of the SMA in a somatosensory tactile attention network isunclear but may prove substantial as detection of cutaneous stimuliis raised before movement [17,58] (at timings similar to thoseshown for SMA activation) and a decrease in left SMA activation hasbeen associated with an increase in sensory threshold [28].

Parietal potentials

It was hypothesized that transient SMA inhibition may affect S1processing due to cortico-thalamic connectivity between these twoareas [7,8] and the finding that TMS of a remote site has beendemonstrated to have effects on anatomically connected brainregions [59]. There was no effect upon any of the measured parietal

W. Legon et al. / Brain Stimulation xxx (2013) 1e76

potentials. This is somewhat surprising as cTBS of S1 has previouslybeen demonstrated to decrease N30 amplitude [33]. A reciprocaleffect upon S1 potentials may not occur because S1 sends directcortico-cortical efferents to SMA but there is no evidence for directcortico-cortical connection from SMA to S1; connection is via thethalamus. Be this as it may, the results suggest that inhibition of S1was not a direct cause for the decrease in tactile perception. This isimportant because lesion of S1 results in a loss of sensory percep-tion and ablation of early parietal SEPs [11,60] and furthermorebecause inhibition of S1 via TMS affects tactile perception[17,19e25] and SEP amplitudes [21,29e31]. An effect upon N30amplitude but not early parietal potentials suggests a distinct rolefor the SMA in conscious tactile perception that can functionindependently of the role of S1.

MEPs

The SMA is densely connected with M1 [7,8] and it is possiblethat the effect on both N30 amplitude and detection threshold isa result of indirect inhibition of primary motor cortex. It is well-established that M1 asserts considerable control over peripheralafference [61], that motor execution may increase cutaneousthresholds [58,62,63] and that TMS of M1 directly affects detectionthreshold [20,28] and SEP amplitudes [30,31,64]. Furthermore,other TMS protocols delivered to SMA [15,16] as well as cTBS to PMd[65e67] have been reported to affect M1 corticospinal excitability.The results of experiment 2 however, demonstrate that cTBS at 90%RMT to the SMA did not affect single- or paired-pulse (short-interval intra cortical inhibition) MEP amplitudes which may be theresult of cTBS location. cTBS delivered to M1 results in an effectupon both single-pulse MEP amplitudes and paired protocols [68]but if delivered to PMd, affects single-pulse MEPs only; suggest-ing that stimulation of remote sites affects a different population ofneurons within M1 as compared to M1 stimulation. In the presentstudy MEPs were assessed at 10 min post-cTBS based on theprevious finding that cTBS to PMd modulated MEP amplitudeswithin 5 min of cTBS stimulation. Thus it is unlikely, but notimpossible, that MEPs may have changed over an extended timerelative to cTBS. It is possible that successive single pulses may alterthe excitability of intracortical interneurons that could impact MEPamplitude. It is important to note that the MEP assessment proce-dures were consistent pre- and post-cTBS. This is the first study toassess cTBS of SMA upon M1 excitability and thus has no precedentfor comparison, however other protocols assessing SMA-M1 effectshave reported both inhibitory [15] and facilitatory [16] interactions.

Timing

Finally, the timing of N30 effects is at odds with other researchinvestigating the effect of cTBS on SEPs. The original paperdescribing the 40 s cTBS protocol by Huang et al. [32] reporteda maximal reduction in MEP amplitude as early as 10 min post-stimulation and Ishikawa et al. [33] studied the effect of 40 s cTBSto S1 on SEPs and reported statistically significant effects as early as3 min post-stimulation though maximal effects were observedbetween 10 and 20 min post-stimulation. The reason for the delayin effect found herewith SMA stimulation compared to studies withS1 stimulation is not clear butmay be related to the specific locationof the target areas relative to the stimulus coils. Regardless, thedelay in effect found in this study is in linewith previous behavioraleffects as a result of SMA stimulation. Terao et al. [69] showed that1 Hz rTMS at 105% RMT to SMA resulted in a delay in reaction time30 min after stimulation. This delay in the modulation of the N30amplitude changes following cTBS was a major contributor to the

lack of a significant interaction between Stimulation Site and Timepost-cTBS.

Conclusions

Continuous theta burst stimulation to a scalp site overlying SMAattenuates the amplitude of the N30 concomitant with an increasein tactile perceptual threshold. These effects are independent ofeffect upon parietal SEPs andMEPs suggesting a specific effect uponthe SMA only and not upon primary somatosensory or motorcortex. As such, the SMA may play a more substantial role ina cutaneous somatosensory network then previously appreciated.

Acknowledgments

The authors would like to thank Malte Steiner and Natalie DiezD’Aux for help with data collection.

References

[1] Tanji J. The supplementary motor area in the cerebral cortex. Neurosci Res1994;19(3):251e68.

[2] Romo R, Ruiz S, Crespo P, Zainos A, Merchant H. Representation of tactilesignals in primate supplementary motor area. J Neurophysiol 1993;70(6):2690e4.

[3] Matsuzaka Y, Aizawa H, Tanji J. A motor area rostral to the supplementarymotor area (presupplementary motor area) in the monkey: neuronal activityduring a learned motor task. J Neurophysiol 1992;68(3):653e62.

[4] Staines WR, Black SE, Graham SJ, McIlroy WE. Somatosensory gating andrecovery from stroke involving the thalamus. Stroke 2002;33(11):2642e51.

[5] Burton H, Sinclair RJ. Attending to and remembering tactile stimuli: a reviewof brain imaging data and single-neuron responses. J Clin Neurophysiol2000;17(6):575e91.

[6] Wiesendanger M, Hummelsheim H, Bianchetti M. Sensory input to the motorfields of the agranular frontal cortex: a comparison of the precentral,supplementary motor and premotor cortex. Behav Brain Res 1985;18(2):89e94.

[7] Jurgens U. The efferent and afferent connections of the supplementary motorarea. Brain Res 1984;300(1):63e81.

[8] Luppino G, Matelli M, Camarda R, Rizzolatti G. Corticocortical connections ofarea F3 (SMA-proper) and area F6 (pre-SMA) in the macaque monkey. J CompNeurol 1993;338(1):114e40.

[9] Desmedt JE, Bourguet M. Color imaging of parietal and frontal somatosensorypotential fields evoked by stimulation of median or posterior tibial nerve inman. Electroencephalogr Clin Neurophysiol 1985;62(1):1e17.

[10] Romo R, Salinas E. Touch and go: decision-making mechanisms in somato-sensation. Annu Rev Neurosci 2001;24:107e37.

[11] Lacruz F, Artieda J, Pastor MA, Obeso JA. The anatomical basis of somaesthetictemporal discrimination in humans. J Neurol Neurosurg Psychiatry1991;54(12):1077e81.

[12] Gerloff C, Corwell B, Chen R, Hallett M, Cohen LG. Stimulation over the humansupplementary motor area interferes with the organization of future elementsin complex motor sequences. Brain 1997;120(Pt 9):1587e602.

[13] Serrien DJ, Strens LH, Oliviero A, Brown P. Repetitive transcranial magneticstimulation of the supplementary motor area (SMA) degrades bimanualmovement control in humans. Neurosci Lett 2002;328(2):89e92.

[14] Steyvers M, Etoh S, Sauner D, Levin O, Siebner HR, Swinnen SP, et al. High-frequency transcranial magnetic stimulation of the supplementary motor areareduces bimanual coupling during anti-phase but not in-phase movements.Exp Brain Res 2003;151(3):309e17.

[15] Civardi C, Cantello R, Asselman P, Rothwell JC. Transcranial magnetic stimu-lation can be used to test connections to primary motor areas from frontal andmedial cortex in humans. Neuroimage 2001;14(6):1444e53.

[16] Matsunaga K, Maruyama A, Fujiwara T, Nakanishi R, Tsuji S, Rothwell JC.Increased corticospinal excitability after 5 Hz rTMS over the human supple-mentary motor area. J Physiol 2005;562(Pt 1):295e306.

[17] Cohen LG, Bandinelli S, Sato S, Kufta C, Hallett M. Attenuation in detection ofsomatosensory stimuli by transcranial magnetic stimulation. Electro-encephalogr Clin Neurophysiol 1991;81(5):366e76.

[18] Andre-Obadia N, Garcia-Larrea L, Garassus P, Mauguiere F. Timing and char-acteristics of perceptual attenuation by transcranial stimulation: a study usingmagnetic cortical stimulation and somatosensory-evoked potentials. Psycho-physiology 1999;36(4):476e83.

[19] Knecht S, Ellger T, Breitenstein C, Bernd Ringelstein E, Henningsen H.Changing cortical excitability with low-frequency transcranial magneticstimulation can induce sustained disruption of tactile perception. BiolPsychiatry 2003;53(2):175e9.

W. Legon et al. / Brain Stimulation xxx (2013) 1e7 7

[20] McKay DR, Ridding MC, Miles TS. Magnetic stimulation of motor andsomatosensory cortices suppresses perception of ulnar nerve stimuli. Int JPsychophysiol 2003;48(1):25e33.

[21] Satow T, Mima T, Yamamoto J, Oga T, Begum T, Aso T, et al. Short-lastingimpairment of tactile perception by 0.9Hz-rTMS of the sensorimotor cortex.Neurology 2003;60(6):1045e7.

[22] Hannula H, Ylioja S, Pertovaara A, Korvenoja A, Ruohonen J, Ilmoniemi RJ, et al.Somatotopic blocking of sensation with navigated transcranial magneticstimulation of the primary somatosensory cortex. Hum Brain Mapp2005;26(2):100e9.

[23] Tegenthoff M, Ragert P, Pleger B, Schwenkreis P, Forster AF, Nicolas V, et al.Improvement of tactile discrimination performance and enlargement ofcortical somatosensory maps after 5 Hz rTMS. PLoS Biol 2005;3(11):e362.

[24] Morley JW, Vickery RM, Stuart M, Turman AB. Suppression of vibrotactilediscrimination by transcranial magnetic stimulation of primary somatosen-sory cortex. Eur J Neurosci 2007;26(4):1007e10.

[25] Meehan SK, Legon W, Staines WR. Paired-pulse transcranial magnetic stim-ulation of primary somatosensory cortex differentially modulates perceptionand sensorimotor transformations. Neuroscience 2008;157(2):424e31.

[26] Nager W, Wolters C, Munte TF, Johannes S. Transcranial magnetic stimulationto the parietal lobes reduces detection of contralateral somatosensory stimuli.Acta Neurol Scand 2004;109(2):146e50.

[27] Oliveri M, Caltagirone C, Filippi MM, Traversa R, Cicinelli P, Pasqualetti P, et al.Paired transcranialmagnetic stimulation protocols reveal a patternof inhibitionand facilitation in the human parietal cortex. J Physiol 2000;529(Pt 2):461e8.

[28] Yoo WK, You SH, Ko MH, Tae Kim S, Park CH, Park JW, et al. High frequencyrTMS modulation of the sensorimotor networks: behavioral changes and fMRIcorrelates. Neuroimage 2008;39(4):1886e95.

[29] Kujirai T, Sato M, Rothwell JC, Cohen LG. The effect of transcranial magneticstimulation on median nerve somatosensory evoked potentials. Electro-encephalogr Clin Neurophysiol 1993;89(4):227e34.

[30] Seyal M, Browne JK, Masuoka LK, Gabor AJ. Enhancement of the amplitude ofsomatosensory evoked potentials following magnetic pulse stimulation of thehuman brain. Electroencephalogr Clin Neurophysiol 1993;88(1):20e7.

[31] Enomoto H, Ugawa Y, Hanajima R, Yuasa K, Mochizuki H, Terao Y, et al.Decreased sensory cortical excitability after 1 Hz rTMS over the ipsilateralprimary motor cortex. Clin Neurophysiol 2001;112(11):2154e8.

[32] Huang YZ, Edwards MJ, Rounis E, Bhatia KP, Rothwell JC. Theta burst stimu-lation of the human motor cortex. Neuron 2005;45(2):201e6.

[33] Ishikawa S, Matsunaga K, Nakanishi R, Kawahira K, Murayama N, Tsuji S, et al.Effect of theta burst stimulation over the human sensorimotor cortex onmotorand somatosensory evoked potentials. Clin Neurophysiol 2007;118(5):1033e43.

[34] Slimp JC, Tamas LB, Stolov WC, Wyler AR. Somatosensory evoked potentialsafter removal of somatosensory cortex in man. Electroencephalogr Clin Neu-rophysiol 1986;65(2):111e7.

[35] Rossini PM, Babiloni F, Bernardi G, Cecchi L, Johnson PB, Malentacca A, et al.Abnormalities of short-latency somatosensory evoked potentials in parkin-sonian patients. Electroencephalogr Clin Neurophysiol 1989;74(4):277e89.

[36] Pierantozzi M, Sabato AF, Leonardis F, Marciani MG, Cicardi C, Giacomini P,et al. Curariform peripheral block of muscular tone selectively increases pre-central N30 somatosensory evoked potentials component. A pharmacologicalstudy carried out on healthy subjects and parkinsonian syndromes. Exp BrainRes 2000;133(3):368e76.

[37] Restuccia D, Valeriani M, Barba C, Le Pera D, Tonali P, Mauguiere F. Differentcontribution of joint and cutaneous inputs to early scalp somatosensoryevoked potentials. Muscle Nerve 1999;22(7):910e9.

[38] Rossini PM, Bassetti MA, Pasqualetti P. Median nerve somatosensory evokedpotentials. apomorphine-induced transient potentiation of frontal compo-nents in Parkinson’s disease and in parkinsonism. Electroencephalogr ClinNeurophysiol 1995;96(3):236e47.

[39] Legon W, Meehan SK, Staines WR. The relationship between frontalsomatosensory-evoked potentials and motor planning. Neuroreport 2008;19(1):87e91.

[40] Legon W, Dionne JK, Meehan SK, Staines WR. Non-dominant hand movementfacilitates the frontal N30 somatosensory evoked potential. BMC Neurosci2010;11:112.

[41] Rossini PM, Caramia D, Bassetti MA, Pasqualetti P, Tecchio F, Bernardi G.Somatosensory evoked potentials during the ideation and execution of indi-vidual finger movements. Muscle Nerve 1996;19(2):191e202.

[42] Hikosaka O, Sakai K, Miyauchi S, Takino R, Sasaki Y, Putz B. Activation ofhuman presupplementary motor area in learning of sequential procedures:a functional MRI study. J Neurophysiol 1996;76(1):617e21.

[43] Lee KM, Chang KH, Roh JK. Subregions within the supplementary motor areaactivated at different stages of movement preparation and execution. Neu-roimage 1999;9(1):117e23.

[44] Hamada M, Ugawa Y, Tsuji S, Effectiveness of rTMS on Parkinson’s DiseaseStudy Group, Japan. High-frequency rTMS over the supplementary motor areafor treatment of Parkinson’s disease. Mov Disord 2008;23(11):1524e31.

[45] Urushihara R, Murase N, Rothwell JC, Harada M, Hosono Y, Asanuma K, et al.Effect of repetitive transcranial magnetic stimulation applied over the pre-motor cortex on somatosensory-evoked potentials and regional cerebralblood flow. Neuroimage 2006;31(2):699e709.

[46] Di Lazzaro V, Ziemann U, Lemon RN. State of the art: physiology of trans-cranial motor cortex stimulation. Brain Stimul 2008;1(4):345e62.

[47] Cheron G, Cebolla AM, De Saedeleer C, Bengoetxea A, Leurs F, Leroy A, et al.Pure phase-locking of beta/gamma oscillation contributes to the N30frontal component of somatosensory evoked potentials. BMC Neurosci2007;8:75.

[48] Romo R, Hernandez A, Zainos A, Brody C, Salinas E. Exploring the corticalevidence of a sensory-discrimination process. Philos Trans R Soc Lond B BiolSci 2002;357(1424):1039e51.

[49] Legon W, Staines WR. Predictability of the target stimulus for sensory-guidedmovement modulates early somatosensory cortical potentials. Clin Neuro-physiol 2006;117(6):1345e53.

[50] Meehan SK, Staines WR. Task-relevance and temporal synchrony betweentactile and visual stimuli modulates cortical activity and motor performanceduring sensory-guided movement. Hum Brain Mapp 2009;30(2):484e96.

[51] Taylor-Clarke M, Kennett S, Haggard P. Vision modulates somatosensorycortical processing. Curr Biol 2002;12(3):233e6.

[52] Kida T, Wasaka T, Nakata H, Akatsuka K, Kakigi R. Centrifugal regulation ofa task-relevant somatosensory signal triggering voluntary movement withouta preceding warning signal. Exp Brain Res 2006;173(4):733e41.

[53] Conte A, Gilio F, Iezzi E, Frasca V, Inghilleri M, Berardelli A. Attention influ-ences the excitability of cortical motor areas in healthy humans. Exp Brain Res2007;182(1):109e17.

[54] Johansen-Berg H, Christensen V, Woolrich M, Matthews PM. Attention totouch modulates activity in both primary and secondary somatosensory areas.Neuroreport 2000;11(6):1237e41.

[55] Staines WR, Graham SJ, Black SE, McIlroy WE. Task-relevant modulation ofcontralateral and ipsilateral primary somatosensory cortex and the role ofa prefrontal-cortical sensory gating system. Neuroimage 2002;15(1):190e9.

[56] Fox MD, Snyder AZ, Vincent JL, Corbetta M, Van Essen DC, Raichle ME. Thehuman brain is intrinsically organized into dynamic, anticorrelated functionalnetworks. Proc Natl Acad Sci U S A 2005;102(27):9673e8.

[57] Burton H, Sinclair RJ, McLaren DG. Cortical network for vibrotactile attention:a fMRI study. Hum Brain Mapp 2008;29(2):207e21.

[58] Williams SR, Shenasa J, Chapman CE. Time course and magnitude ofmovement-related gating of tactile detection in humans. I. Importance ofstimulus location. J Neurophysiol 1998;79(2):947e63.

[59] Bestmann S, Ruff CC, Blankenburg F, Weiskopf N, Driver J, Rothwell JC.Mapping causal interregional influences with concurrent TMS-fMRI. Exp BrainRes 2008;191(4):383e402.

[60] Allison T, McCarthy G, Wood CC, Jones SJ. Potentials evoked in human andmonkey cerebral cortex by stimulation of the median nerve. A review of scalpand intracranial recordings. Brain 1991;114(Pt 6):2465e503.

[61] Canedo A. Primary motor cortex influences on the descending and ascendingsystems. Prog Neurobiol 1997;51(3):287e335.

[62] Chapman CE, Bushnell MC, Miron D, Duncan GH, Lund JP. Sensory perceptionduring movement in man. Exp Brain Res 1987;68(3):516e24.

[63] Milne RJ, Aniss AM, Kay NE, Gandevia SC. Reduction in perceived intensity ofcutaneous stimuli during movement: a quantitative study. Exp Brain Res1988;70(3):569e76.

[64] Mochizuki H, Terao Y, Okabe S, Furubayashi T, Arai N, Iwata NK, et al. Effects ofmotor cortical stimulation on the excitability of contralateral motor andsensory cortices. Exp Brain Res 2004;158(4):519e26.

[65] Huang YZ, Rothwell JC, Lu CS, Wang J, Weng YH, Lai SC, et al. The effect ofcontinuous theta burst stimulation over premotor cortex on circuits inprimary motor cortex and spinal cord. Clin Neurophysiol 2009;120(4):796e801.

[66] Stefan K, Gentner R, Zeller D, Dang S, Classen J. Theta-burst stimulation:remote physiological and local behavioral after-effects. Neuroimage2008;40(1):265e74.

[67] Stinear CM, Barber PA, Coxon JP, Verryt TS, Acharya PP, ByblowWD. Repetitivestimulation of premotor cortex affects primary motor cortex excitability andmovement preparation. Brain Stimul 2009;2(3):152e62.

[68] Suppa A, Ortu E, Zafar N, Deriu F, Paulus W, Berardelli A, et al. Theta burststimulation induces after-effects on contralateral primary motor cortexexcitability in humans. J Physiol 2008;586(Pt 18):4489e500.

[69] Terao Y, Furubayashi T, Okabe S, Mochizuki H, Arai N, Kobayashi S, et al.Modifying the cortical processing for motor preparation by repetitive trans-cranial magnetic stimulation. J Cogn Neurosci 2007;19(9):1556e73.