Behavioural Brain Research 287 (2015) 27–33
Contents lists available at ScienceDirect
Behavioural Brain Research journal homepage: www.elsevier.com/locate/bbr
Research report
Cumulative effects of anodal and priming cathodal tDCS on pegboard test performance and motor cortical excitability Monica Christova a,b,∗ , Dietmar Rafolt c , Eugen Gallasch a a b c
Department of Physiology, Medical University of Graz, Austria Department of Physiotherapy, University of Applied Sciences FH-Joanneum, Austria Center for Medical Physics and Biomedical Engineering, Medical University Vienna, Austria
h i g h l i g h t s • • • •
atDCS during pegboard training improved motor performance. Preceding ctDCS induced an outlasting increase in cortical excitability. Preceding ctDCS improved off-line learning. Homeostatic plasticity is a possible factor for the cortical excitability effects.
a r t i c l e
i n f o
Article history: Received 5 January 2015 Received in revised form 10 March 2015 Accepted 14 March 2015 Available online 21 March 2015 Keywords: Skill learning Motor cortex Neuroplasticity Procedural memory tDCS TMS
a b s t r a c t Transcranial direct current stimulation (tDCS) protocols applied over the primary motor cortex are associated with changes in motor performance. This transcranial magnetic stimulation (TMS) study examines whether cathodal tDCS prior to motor training, combined with anodal tDCS during motor training improves motor performance and off-line learning. Three study groups (n = 36) were trained on the grooved pegboard test (GPT) in a randomized, between-subjects design: SHAM—sham stimulation prior and during training, STIM1—sham stimulation prior and atDCS during training, STIM2—ctDCS stimulation prior and atDCS during training. Motor performance was assessed by GPT completion time and retested 14 days later to determine off-line learning. Cortical excitability was assessed via TMS at baseline (T0), prior training (T1), after training (T2), and 60 min after training (T3). Motor evoked potentials (MEP) were recorded from m. abductor pollicis brevis of the active left hand. GPT completion time was reduced for both stimulated groups compared to SHAM. For STIM2 this reduction in time was significantly higher than for STIM1 and further off-line learning occurred after STIM2. After ctDCS at T1, MEP amplitude and intracortical facilitation was decreased and intracortical inhibition was increased. After atDCS at T2, an opposite effect was observed for STIM1 and STIM2. For STIM2 these neuromodulatory effects were retained until T3. It is concluded that application of atDCS during the training improves pegboard performance and that additional priming with ctDCS has a positive effect on off-line learning. These cumulative behavioral gains were indicated by the preceding neuromodulatory changes. © 2015 Elsevier B.V. All rights reserved.
1. Introduction From our experience we know that practicing a new motor task improves motor performance and that an acquired skill remains relative stable without additional practice. However the performance gains achieved within a certain training period are limited.
∗ Corresponding author at: Medical University of Graz, Department of Physiology, Harrachgasse 21/5, 8010 Graz, Austria. Tel.: +43 316 380 76 27; fax: +43 316 380 9630. E-mail address: [email protected] (M. Christova). http://dx.doi.org/10.1016/j.bbr.2015.03.028 0166-4328/© 2015 Elsevier B.V. All rights reserved.
This obviously depends on straits to reorganize cortical networks [1], but also on the utilization of cognitive resources at the early stage of motor learning [2]. Recently noninvasive brain stimulation has been probed successfully in order to enhance motor performance at various hand motor learning tasks. Often transcranial direct current stimulation (tDCS) was used in such studies [3], a technique that allows polarization of membrane potentials [4,5] and modulation of cortical excitability [6]. Basically tDCS induces a current flow through the skull, and it was suggested that cathodal polarization slightly decreases firing rate in the underlying brain tissue while anodal polarization slightly increases the firing rate [4]. After application of tDCS over the
28
M. Christova et al. / Behavioural Brain Research 287 (2015) 27–33
primary motor cortex such neuromodulatory effects can be observed via transcranial magnetic stimulation (TMS): after cathodal tDCS (ctDCS) cortical excitability decreases and after anodal tDCS (atDCS) excitability it increases [7,8]. For these effects transient changes in synaptic efficacy are held responsible. For example it was demonstrated with magnetic resonance spectroscopy that atDCS (1 mA, 10 min) locally reduces GABA, while ctDCS reduces glutamergic activity [9]. If tDCS is paired with motor learning it is crucial to consider timing and polarity of stimulation. In several studies atDCS was tested during practice in order to facilitate motor performance [10–12]. It is argued that increase of firing rates in task specific networks imposes additional strengthening of specific synaptic connections [13]. In some studies further priming tDCS (tDCS in advance of practice) was tested. Stagg et al. [14] found slower learning of a finger sequence task for both atDCS and ctDCS, while Antal et al. [15] found an improvement in early learning of a visuomotor coordination task for both polarities. Despite these inconsistencies in learning paradigms, lowering of neuronal activity with priming ctDCS seems to be advantageous as it reduces the threshold for subsequent protocols to increase cortical excitability [16,17]. Also the site of stimulation has to be considered. At tasks with high cognitive demands the anode often is placed over the premotor cortex [18,19], while at tasks depending more on sensory input placement over the primary motor cortex showed appropriate [10,12,20]. In most of the studies the impact of tDCS has been exclusively explored at the behavioral level. However to elucidate the underlying adaptive processes, it is advantageous to incorporate noninvasive neuronal measures. Therefore in the current study we measured motor evoked potential (MEP) amplitudes in order to evaluate changes of cortical excitability. As training paradigm the grooved pegboard test (GPT) was chosen. It was shown that the GPT itself evokes rather minor excitability changes [21,22] and therefore the tDCS-induced changes should dominate the picture. The study protocol was designed in order to examine whether ctDCS prior to motor training, combined with atDCS during motor training, cumulatively enhances pegboard test performance and off-line learning. We further assume that the atDCS and priming ctDCS driven changes in cortical excitability are related to the expected behavioral gains. To verify these assumptions three matched study groups were tested in a randomized, single-blind, between-subject design: one group receiving sham stimulation prior and during GPT training, a second group receiving sham stimulation prior and atDCS during the training and a third group receiving ctDCS prior and atDCS during the training. For all three groups single and paired pulse TMS assessments were performed prior to priming stimulation (sham or ctDCS), after priming stimulation, immediately after practice, and 60 min after practice. To assess off-line learning GPT performance was retested two weeks later.
2. Material and methods 2.1. Participants and study design Thirty-six healthy volunteers took part in the study approved by the Ethics Committee at the Medical University of Graz. They were screened for possible neurological disorders and contraindications to TMS and signed a written informed consent. All participants were right-handed according to the laterality quotient (LQ) from the Edinburgh Handedness Inventory [23]. According to self report none of the participants actively played an instrument or engaged in any other activity that extensively involved the left nondominant hand. Following acquisition of a novel task the same subject could not be tested again, thus the participants were randomly assigned to three study groups, (n = 12) each receiving
different tDCS protocol. The first group (8 ♀ and 4 ♂, mean age 24.92 ± 5.04 years) underwent sham tDCS (SHAM). The second group (8 ♀ and 4 ♂, mean age 27.67 ± 9.98 years) underwent sham stimulation preceding the GPT and atDCS during the GPT performance (STIM 1). The third group (6 ♀ and 6 ♂, mean age 26.00 ± 8.91 years) underwent ctDCS preceding GPT and atDCS during GPT performance (STIM2). 2.2. Grooved Pegboard test and skill training The use of GPT (Model 32025, Lafayette Instrument, USA) is well documented for the ability to generate performance curves and is also used to assess motor function in patients with motor deficits. In this study, the GPT was performed in 4 blocks (4 trials in each block) with interblock rest intervals of 2 min to avoid muscle and central fatigue. The GPT was retested two weeks after skill training in a single block (4 trials). Subjects received exact instructions and a demonstration of the test, without being provided with practice trials. All participants completed the test at equal ambient conditions at the same time of day (between 9 and 12 a.m.). The observer instructed the subjects to complete the task as fast as possible and recorded the time for each trial. 2.3. Transcranial direct current stimulation (tDCS) A bipolar electrode montage was utilized to deliver tDCS. The active electrode was placed to stimulate the right primary motor cortex M1, contralaterally to the performing left hand. The electrode was centered on C4 of the international 10–20 electroencephalogram system as it was shown in neuroimaging studies that C3/C4 corresponds to the left and right M1 [24]. The correspondence between C4 and right M1 was additionally confirmed using TMS individually for each subject and adjusted when necessary to the APB representation spot. The reference electrode was placed over the contralateral left supraorbital area. tDCS was delivered by the MAGSTIM ELDITH DC-stimulator and a pair of non-metallic, conductive rubber electrodes with water-soaked synthetic sponges (5 cm × 7 cm/35 cm2 ). Stimulations were delivered at an intensity of 1 mA (current density 0.029 mA/cm2 ). As it was shown that a stimulation period between 9 to 20 min is appropriate to induce stable after effects in the motor cortex [25], priming ctDCS was applied for 15 min. The period of atDCS during GPT practice was approximately 20 min in order to cover the overall training period. For sham tDCS the electrodes were placed in the same manner, however the stimulation was turned off after 15 s [16]. This procedure allows to blind subjects for the stimulation condition [10]. 2.4. TMS assessments For the TMS assessment, two Magstim 200 magnetic stimulators connected via a Bistim module (The Magstim Company, Whitland, Dyfed, UK) were employed. Magnetic pulses were delivered through a figure eight-shaped coil (outer loop diameter of 9 cm). The coil was positioned on the scalp over the right motor cortex at the optimal site for stimulating the contralateral left abductor pollicis brevis (APB). The intersection of the coil was placed tangentially to the scalp, with the handle pointing backward and laterally at a 45◦ angle away from the midline. The resting motor threshold (rMT) was expressed as a percentage of the maximum output of the stimulator. Using suprathreshold intensities, the coil was moved over the scalp in small steps to locate the site with the largest MEP. This position was marked on an EEG cap and the coil was fixed at that position. The resting motor threshold (rMT), MEP, short-latency intracortical inhibition (SICI) and intracortical facilitation (ICF) were examined. The resting MT was defined as the lowest stimulus
M. Christova et al. / Behavioural Brain Research 287 (2015) 27–33
29
Fig. 1. Experimental design. Grooved pegboard test (GPT) is trained in a session of 4 blocks × 4 trials and retested after 2 weeks in a session of 1 block × 4 trials. Three study groups: sham stimulation before/sham stimulation during GPT (SHAM), sham stimulation before/atDCS during GPT (STIM1) and ctDCS before/atDCS during GPT (STIM2). TMS assessments were taken at four time points: T0—at baseline, T1—after priming tDCS, T2—after tDCS/GPT and T3—60 min after tDCS/GPT.
intensity at which five out of ten consecutive TMS applications led to a MEP of at least 0.05 mV on the relaxed APB. MEPs were obtained with three stimulus intensities: 110%, 130% and 150% of rMT in a random order. Eight single pulses were delivered at each stimulus intensity. SICI and ICF were assessed using the pairedpulse paradigm with interstimulus intervals (ISI) of 3 ms and 13 ms. Eight conditioned pulses were obtained for each ISI. The conditioning and test stimuli were 80% and 120% of the corresponding rMT at the actual assessment. Four TMS assessments were performed: baseline (T0), after 15 priming tDCS (T1), after the GPT/tDCS (T2) and 60 min after the GPT/tDCS (T3) (Fig. 1). 2.5. Electromyographic (EMG) recordings MEPs were recorded from APB muscle of the left hand using 9 mm diameter Ag–AgCl surface disk electrodes in a tendonbelly manner. EMG signals were amplified, band pass filtered (8–2000 Hz), digitized (sampling rate 10 kHz) and recorded on a disk (DasyLab 8.0 software package). The EMG signal was displayed continuously during TMS assessment to ensure the absence of any voluntary motor activity. 2.6. Data analysis and statistics 2.6.1. Motor performance, on- and off-line learning Motor performance was quantified as the time to complete the GPT for each trial. It was then displayed in the form of inverted learning curves. For statistical analysis, the mean time per block was calculated. An ANOVA with a within-subject factor block (block1, block2, block3, block4, retest) and a between-subject factor group (SHAM, STIM1, STIM2) was performed. Where the ANOVAs showed significance, further one-factorial ANOVAs with Bonferoni adjustment for multiple comparisons were performed. On-line learning was quantified by the normalized time values for each block. Accordingly four values were determined by the ratio mean time block n/mean time block 1 (n = 1, 2, 3, 4). For offline learning one value was determined by the ratio mean time retest/mean time block1. An ANOVA with within-subject factors block (block1, block2, block3, block4, retest) and a between-subject factor group (SHAM, STIM1, STIM2) was carried out on these values. Where the ANOVA showed significance one-factorial ANOVAs with Bonferoni adjustment for multiple comparisons were performed. In the following text and in Fig. 3, on- and off-line learning is referred to as “performance improvement”. 2.6.2. Motor cortical excitability rMT, expressed as the percentage of maximum stimulator output (% mean ± S.E.M.), was assessed using a two-factorial repeated measures ANOVA (analysis of variance) with a within-subject factor time (4 levels: T0, T1, T2 and T3) and a between-subject factor group (3 levels: SHAM, STIM1, STIM2). The MEP amplitudes were presented in mV (mean ± S.E.M.). For each stimulation intensity (110%, 130% and 150%), the MEP amplitudes were analyzed separately using two-factorial ANOVA with
Fig. 2. Performance curves. Performance curves presenting the GPT completion time, (mean ± S.E.M) for each group: SHAM, STIM1 (sham before/atDCS during GPT) and STIM2 (ctDCS before/atDCS during GPT) in the four training blocks (4 trials per block) and in the retest (4 trials).
within-subject factors: time (4 levels: T0, T1, T2 and T3) and a between-subject factor group (3 levels: SHAM, STIM1, STIM2). In case of significant interactions, follow-up ANOVAs with Bonferoni adjustment for multiple comparisons with between-subject factor group for each time level and within-subject factor time for each group were conducted. For the conditioned MEP responses (SICI and ICF), the amplitudes were calculated as a percentage of the single pulse MEP for each subject individually (mean ± S.E.M.). For each ISI (3 and 13 ms), a repeated measures ANOVA was used to compare the conditioned MEP amplitudes with within-subject factors time (4 levels: T0, T1, T2 and T3) and a between-subject factor group (3 levels: SHAM, STIM1, STIM2). In case of significant main or interaction effects, follow-up ANOVAs with Bonferoni adjustment for multiple comparisons with between-subject factor group for each time level and within-subject factor time for each group were conducted. In all statistical tests a significance level of 0.05 was used. 3. Results 3.1. Motor performance and performance improvement The performance curves data (Fig. 2) show that overall the time to complete the GPT was longer for the sham group (69.87 ± 2.81 s) than for STIM1 (62.63 ± 2.51 s) and STIM2 (59.83 ± 2.18 s). An ANOVA revealed a significant main effect of block (F(4, 144) = 141.34, p < 0.001, 2 = 0.797), but insignificant interaction effect of block × group (F(8, 144) = 1.32, p > 0.05, 2 = 0.068). In block 1 no significant difference between the groups was found, however in the following training blocks between-groups differences were observed: block 2 (p = 0.04), block 3 (p = 0.01), block4 (p = 0.02) and retest (p < 0.001). In block 2 the GPT completion time was shorter for STIM 2 compared to SHAM p = 0.05). In blocks 2, 3 and 4 no difference was found between STIM1 and STIM2. In blocks 3, 4 and retest both stimulated groups showed shorter GPT completion time compared to SHAM (ps < 0.03). Further at retest STIM2 showed significantly shorter performance time in comparison to STIM1 (p = 0.04). To follow the performance improvement over the four blocks (on-line learning) and in the retest (off-line learning) data are displayed as normalized time values for each group separately, see Fig. 3. An ANOVA revealed a significant main effect of block
30
M. Christova et al. / Behavioural Brain Research 287 (2015) 27–33
Fig. 3. Performance improvement. On-line (blocks 1–4) and off-line improvement (retest after 2 weeks) for the three groups: SHAM, STIM1 (sham before/atDCS during GPT) and STIM2 (ctDCS before/atDCS during GPT). Data are presented as the percentage of the time in block 1 for each group (mean ± SD). The significant within group differences between the baseline training block (Block1) and the following training blocks (Block 2, 3, 4, retest) are presented * p < 0.05.
(F(4, 144) = 147.61, p < 0.001, 2 = 0.804) and a tendency for interaction effect of block × group (F(8, 144) = 1.73, p = 0.09, 2 = 0.088). Across all groups significant performance improvement occurred (21% for SHAM, 22% for STIM1, 20% for STIM2); however, no difference between the groups was found over the four training blocks. In comparison to the last training block 4, at the retest the normalized time values were significantly decreased only for STIM2 (−6%, p = 0.01) and slightly increased for STIM1 (+2%) and SHAM (+3%).
the MEP amplitude increase reached significance only at intensity 110% (p = 0.02). At time level T3 (60 min after GPT/tDCS) significant between-groups difference was observed at all stimulation intensities (ps < 0.05), where STIM2 showed increases MEP
3.2. MEP results The mean rMT values measured at baseline (T0) were 39.27 ± 1.1% for SHAM, 41.93 ± 2.2% for STIM1 and 39.00 ± 1.4% for STIM2. No significant differences were observed. MEP amplitudes at stimulation intensity 110%, 130% and 150%, obtained at time levels T0, T1, T2 and T3 for each group are presented in Fig. 4. For all stimulation intensities a two-factorial ANOVA revealed a significant main effect of time (Fs(3, 102) > 6.17, ps < 0.01, s2 < 0.290) and an interaction effect of time × group (Fs(6, 102) > 3.46, ps < 0.004, s2 < 0.310). To reveal the within-groups cortical excitability effects across the four TMS assessments, one-factorial ANOVAs with factor time were conducted for each group and for each TMS intensity. The SHAM group didn’t show any MEP amplitude changes over the time. The STIM1 group showed significant cortical excitability changes at stimulation intensities 130% and 150% (ps < 0.001) where a longlasting MEP increase was found at time levels T2 and T3. The STIM2 group showed significant MEP amplitude changes at all stimulation intensities (P ps < 0.03), where at time level T1 (after priming tDCS) the MEP amplitudes decreased, at time level T2 (after GPT/atDCS) they increased and at time level T3 (60 min after the end of GPT/tDCS) they reached their maximum values. Further to reveal the between-groups excitability effects, onefactorial ANOVAs with factor group were conducted for each time level and for each TMS intensity. At time levels T0 (baseline) and T1 (after priming tDCS) no significant differences between the four groups were found. At time level T2 (after GPT/tDCS) differences between groups were found at intensities 110% (p = 0.06) and at 130% (p = 0.04). The MEP amplitudes at these intensities were increased for STIM2 compared to SHAM. For STIM1
Fig. 4. MEP amplitudes. MEP amplitudes at stimulation intensities 110%, 130% and 150% of the rMT for the three groups: SHAM, STIM1 (sham before/atDCS during GPT) and STIM2 (ctDCS before/atDCS during GPT), at baseline (T0), after preconditioning tDCS (T1), immediately after (T2) and 60 min after (T3) the end of GPT/tDCS. MEP amplitudes (mV) are plotted for each stimulation intensity as the mean (S.E.M). The significant differences between T0 and the following TMS assessments are presented * p < 0.05.
M. Christova et al. / Behavioural Brain Research 287 (2015) 27–33
31
Fig. 5. Short-latency intracortical inhibition (SICI). Normalized paired-pulse responses (ISI 3 ms) for the three groups: SHAM, STIM1 (sham before/atDCS during GPT) and STIM2 (ctDCS before/atDCS during GPT), at baseline (T0), after preconditioning tDCS (T1), immediately after (T2) and 60 min (T3) after the end of GPT/tDCS. Values are normalized for each subject to their corresponding values in single pulse stimulation and plotted as the mean (S.E.M.). The significant differences between T0 and the following TMS assessment are presented * p < 0.05.
amplitudes in comparison to SHAM. No differences were observed for STIM1. The results from the paired-pulse stimulation are shown in Figs. 5 and 6. Separate ANOVAs were conducted for ISIs (3, 13 ms) to reveal the effect of time and group on SICI and ICF, respectively. For SICI, a significant main effect of time (F(3, 102) = 8.53, p < 0.001, 2 = 0.201) and interaction effect of time × group (F(6, 102) = 6.18, p < 0.001, 2 = 0.267) were found. For ICF, also significant main effects of time (F(3, 102) = 6.08, p < 0.001, 2 = 0.152) and interaction effect of time × group (F(6, 102) = 2.26, p = 0.04, 2 = 0.117) were observed. To reveal the within-group intracortical excitability effects across the four TMS assessments, follow-up one-factorial ANOVAs with factor time were conducted for each group. To reveal the within-groups intracortical excitability effects across the four TMS assessments, one-factorial ANOVAs with factor time were conducted for each group. For SHAM no significant
differences were found. For STIM1 group significant effect of time was found for SICI (p < 0.001) and a tendency for ICF (p = 0.05). At time level T2 a reduced effect of SICI (expressed with increased value) +20% and increased ICF +30% were found. These effects were transient and at time level T3 the SICI and ICF values reached back their baseline levels. For STIM2 group, significant effect of time was observed, both for SICI and ICF (ps < 0.001). At time level T1 (after priming tDCS) an increased effect of SICI (expressed with decreased value) −40% (p < 0.001) and a tendency for decreased effect of ICF −23% (p = 0.05) were found. Further, at time levels T2 (after GPT/tDCS) SICI effect was reduced (p = 0.03) and ICF effect was increased (p = 0.02). At time level T3 (60 min after GPT/tDCS) these effects were still present, however they reached significant level only for ICF. To reveal the between-groups differences in SICI and ICF onefactorial ANOVAs with factor group were conducted for each
Fig. 6. Intracortical facilitation (ICF). Normalized paired-pulse responses (ISI 13 ms) for the three groups: SHAM, STIM1 (sham before/atDCS during GPT) and STIM2 (ctDCS before/atDCS during GPT), at baseline (T0), after preconditioning tDCS (T1), immediately after (T2) and 60 min (T3) after the end of GPT/tDCS. Values are normalized for each subject to their corresponding values in single pulse stimulation and plotted as the mean (S.E.M.). The significant differences between T0 and the following TMS assessment are presented * p < 0.05.
32
M. Christova et al. / Behavioural Brain Research 287 (2015) 27–33
time level. At time level T0 (baseline), no significant differences between the four groups were found, either for SICI or for ICF. At time level T1 (after priming tDCS) significant effect of group was found, (ps < 0.001), whereby for STIM2 group SICI was increased (ps < 0.001) and ICF was decreased (ps < 0.03) compared to SHAM and to STIM1. For STIM1 ICF was increased compared to SHAM (p = 0.04). At time level T2 (after GPT/tDCS) significant effect of group was observed for SICI which was decreased for STIM1 group in comparison to SHAM (p = 0.02). At time level T3 no significant differences between the groups were found. 4. Discussion The purpose of this study was to examine the enhancing effects of atDCS and priming ctDCS on pegboard performance and offline learning in relation to associated changes in motor cortical excitability. As expected, pegboard training itself induced no shifts in motor cortical excitability. On the other hand atDCS increased excitability after the training and priming ctDCS reduced excitability ahead of the training. Thus for the forthcoming discussion solely these externally induced excitability shifts have to be taken into account. 4.1. Effects of atDCS during the GPT training The behavioral results show that atDCS delivered concurrently with GPT training resulted in shorter GPT performance times in relation to SHAM, both during the second half of the training session, and 14 days later at retest see, Fig. 2. Comparable gains in motor performance for example have been described for serial reaction time tasks [10,14,26] and for a visuomotor task [27]. Further comparable gains at retest have been described for an isometric pinch force task [28,29] and for a visuomotor transformation task [30]. In contrast to these gains in motor performance, performance improvement (on- and off-line learning) remained unchanged, see Fig. 3. Comparable brain stimulation effects on on-line learning values were not described, but in studies where pegboard performance was tested with the bare and the gloved hand, these values also remained unchanged [22,31]. With regard to the current results it appears that on-line learning is less prone to interventions. However the number of participants in the current study (n = 12) is rather low which is a limitation to resolve such questions. The facilitatory after effects on motor cortical excitability, as observed here, are comparable to TMS studies where atDCS was tested at rest without practicing a motor task [6,32,33]. Thus the results of our study show that atDCS is able to induce similar after effects in the background of learning related motor activity. Comparable effects, increased MEP amplitude for 60 min and decreased effect of SICI for 30 min were also described after atDCS over the nondominant M1 with the Perdue Pegboard test [34]. Further a reduction in intracortical inhibitory networks, related to significant reduction of GABA concentration within M1, was documented with magnetic resonance spectroscopy after a motor adaptation task paired with atDCS [35]. Additionally it has to be mentioned here, that the tDCS-induced after effects depend also on the type of the training task [36]. 4.2. Effects of atDCS and priming ctDCS The behavioral results show a reduction of the GPT performance time also in the first half of the training session and a further reduction of performance time at the retest. Thus a cumulative effect is observed as priming ctDCS additionally reduces performance times in relation to SHAM. A similar effect was found for off-line learning. While STIM1 was not able to influence retention of motor memory, there was a positive effect on memory in the
group who received STIM2. A limitation of this study is that priming ctDCS followed by sham stimulation was not tested by a fourth group. It is however rather unlikely that priming ctDCS alone would have produced these effects, see previous results [14,15], and discussion on homeostatic plasticity below. Off-line learning further indicates that consolidation has occurred after the training. Likewise off-line learning was reported by Butts et al. [37] who primed cortical excitability with two potentiating protocols (iTBS followed by bihemispheric tDCS) and found performance improvement on the Jebsen–Taylor hand function test with a delay of 1–7 days after training. The inhibitory effect on cortical excitability ahead of the training again is comparable to TMS studies where ctDCS was tested at rest [6]. However the most striking findings in comparison to STIM1 are the prolonged shifts of SICI and ICF, indicating increased intracortical excitability at least until 60 min after the training. As pegboard training itself showed no effect on excitability there is an evidence that priming ctDCS followed by atDCS generates this delayed neuroplastic effect. Such delayed effects are in coincidence with the concepts of homeostatic plasticity [17]. According to these concepts lowering of neuronal activity with preceding ctDCS reduces the threshold for subsequent brain stimulation protocols in order to increase the corticospinal excitability. Such delayed effects have been tested with brain stimulation protocols a rest. For example increase of MEP amplitude was found after high frequency rTMS preconditioned by ctDCS [7] where the after effects were examined within 30 min. Further the differential impact of preconditioning tDCS was shown by Siebner et al. [16], where 1 Hz rTMS increased cortical excitability when primed by ctDCS but decreased it when primed by atDCS. On the other hand the concept that the history of the network activity influences subsequent brain stimulation protocols is not compulsory. For example in a study were paired associative stimulation (PAS) was used to mimic learning learning-like synaptic plasticity [38], homeostatic modulation of motor cortical plasticity occurred when tDCS was applied simultaneously with PAS, but not when applied before PAS. Recently a few studies implemented the homeostatic plasticity rules in order to boost motor learning demonstrating both homeostatic and nonhomeostatic modulation of learning. Kuo et al. [39] after combining atDCS with NMDA agonist, found reduced performance in a sequential learning task, however after combining ctDCS with NMDA agonist no change was found. Further the time delay between priming stimulation and start of training was shown to be of importance [40]. A PAS protocol for excitability depression (PASLTD ) and potentiation (PASLTP ) facilitated subsequent thumb flexion movements when performed immediately before the task execution. But when performed 90 min before the task, PASLTD facilitated and PASLTP diminished the motor performance. In our study ctDCS preceded the GPT by 10 min, which seemed to be sufficient to enhance pegboard performance and to improve retention of motor memory. Currently the mechanisms of such delayed effects are not completely understood, obviously because the effects on learning depend on a variety of parameters such as the stimulation modality, timing and the type of the training task. For the effect on motor memory LTP-like plasticity and synaptic strengthening are of importance. Due to the prolonged increase of intracortical excitability after the training LTP-like processes will become boosted within M1. LTP involves a cascade of processes over time, which can increase the efficacy of existing synapses and promote the formation of new synapses [41]. Accordingly, LTP-like processes are thought to continue after training, thus stabilizing the memory traces of the newly learned task. Supporting evidence comes from a recent study [42] where it was shown that experimental reversal of LTP-like plasticity immediately after motor training was able to disrupt skill retention. In addition to LTP, GABAergic mechanisms play a role for the adjustment of motor
M. Christova et al. / Behavioural Brain Research 287 (2015) 27–33
cortical representations [43]. For example a pharmacological study has shown that enhancing of GABAa receptor function with administration of lorazepam results in impaired motor learning and formation of new memories [44]. Currently it is unclear why pegboard training itself left no trace in the measured MEP amplitudes. Perhaps some training-induced neuroplasticity occurs beyond the motor cortex, and atDCS acts indirectly via functional connections within a wider network? But it is also conceivable that neuroplasticity arises at circuits in the motor cortex which do not contribute to the transsynaptically evoked indirect waves elicited via magnetic stimulation with a figure of eight-shaped coil in posterior-anterior direction [45]. Further studies including brain stimulation protocols to elicit direct waves could be helpful to clarify this enigma. 5. Conclusions In conclusion this study shows that atDCS in combination with priming ctDCS decreases the time to complete a repetitive motor task which could have an impact on neurorehabilitative science. Acknowledgments This work was supported by the Research Fund of Government of Styria, Austria. References [1] Luft AR, Buitrago MM. Stages of motor skill learning. Mol Neurobiol 2005;32:205–16. [2] Seidler RD, Bo J, Anguera JA. Neurocognitive contributions to motor skill learning: the role of working memory. J Mot Behav 2012;4:445–53. [3] Reis J, Fritsch B. Modulation of motor performance and motor learning by transcranial direct current stimulation. Curr Opin Neurol 2011;24:590–6. [4] Bindman LJ, Lippold OC, Redfearn JW. The action of brief polarizing currents on the cerebral cortex of the rat (1) during current flow and (2) in the production of long-lasting after-effects. J Physiol 1964;172:369–82. [5] Priori A, Berardelli A, Rona S, Accornero N, Manfredi M. Polarization of the human motor cortex through the scalp. NeuroReport 1998;9:2257–60. [6] Nitsche MA, Paulus W. Excitability changes induced in the human motor cortex by weak transcranial direct current stimulation. J Physiol 2000;527:633–9. [7] Lang N, Siebner HR, Ernst D, Nitsche MA, Paulus W, Lemon RN, et al. Preconditioning with transcranial direct current stimulation sensitizes the motor cortex to rapid-rate transcranial magnetic stimulation and controls the direction of after-effects. Biol Psychiatry 2004;56:634–9. [8] Di Lazzaro V, Ranieri F, Profice P, Pilato F, Mazzone P, Capone F, et al. Transcranial direct current stimulation effects on the excitability of corticospinal axons of the human cerebral cortex. Brain Stimul 2013;6:641–3. [9] Stagg CJ, Bachtiar V, Johansen-Berg H. The role of GABA in human motor learning. Curr Biol 2011;21:480–4. [10] Nitsche MA, Schauenburg A, Lang N, Liebetanz D, Exner C, Paulus W, et al. Facilitation of implicit motor learning by weak transcranial direct current stimulation of the primary motor cortex in the human. J Cogn Neurosci 2003;15:619–26. [11] Galea JM, Celnik P. Brain polarization enhances the formation and retention of motor memories. J Neurophysiol 2009;102:294–301. [12] Prichard G, Weiller C, Fritsch B, Reis J. Effects of different electrical brain stimulation protocols on subcomponents of motor skill learning. Brain Stimul 2014;7:532–40. [13] Fritsch B, Reis J, Martinowich K, Schambra HM, Ji Y, Cohen LG, et al. Direct current stimulation promotes BDNF-dependent synaptic plasticity: potential implications for motor learning. Neuron 2010;66:198–204. [14] Stagg CJ, Jayaram G, Pastor D, Kincses ZT, Matthews PM, Johansen-Berg H. Polarity and timing-dependent effects of transcranial direct current stimulation in explicit motor learning. Neuropsychologia 2011;49:800–4. [15] Antal A, Begemeier S, Nitsche MA, Paulus W. Prior state of cortical activity influences subsequent practicing of a visuomotor coordination task. Neuropsychologia 2008;46:3157–61. [16] Siebner HR, Lang N, Rizzo V, Nitsche MA, Paulus W, Lemon RN, et al. Preconditioning of low-frequency repetitive transcranial magnetic stimulation with transcranial direct current stimulation: evidence for homeostatic plasticity in the human motor cortex. J Neurosci 2004;24:3379–85. [17] Ziemann U, Siebner HR. Modifying motor learning through gating and homeostatic metaplasticity. Brain Stimul 2008;1:60–6.
33
[18] Nitsche MA, Jakoubkova M, Thirugnanasambandam N, Schmalfuss L, Hullemann S, Sonka K, et al. Contribution of the premotor cortex to consolidation of motor sequence learning in humans during sleep. J Neurophysiol 2010;104:2603–14. [19] Pavlova E, Kuo MF, Nitsche MA, Borg J. Transcranial direct current stimulation of the premotor cortex: effects on hand dexterity. Brain Res 2014;1576:52–62. [20] Antal A, Polania R, Schmidt-Samoa C, Dechent P, Paulus W. Transcranial direct current stimulation over the primary motor cortex during fMRI. NeuroImage 2011;55:590–6. [21] McDonnell MN, Ridding MC. Transient motor evoked potential suppression following a complex sensorimotor task. Clin Neurophysiol 2006;117:1266–72. [22] Christova M, Rafolt D, Golaszewski S, Nardone R, Gallasch E. Electrical stimulation during skill training with a therapeutic glove enhances the induction of cortical plasticity and has a positive effect on motor memory. Behav Brain Res 2015, http://dx.doi.org/10.1016/j.bbr.2015.03.028. [23] Oldfield RC. The assessment and analysis of handedness: the Edinburgh inventory. Neuropsychologia 1971;9:97–113. [24] Okamoto M, Dan H, Sakamoto K, Takeo K, Shimizu K, Kohno S, et al. Three-dimensional probabilistic anatomical cranio-cerebral correlation via the international 10–20 system oriented for transcranial functional brain mapping. NeuroImage 2004;21:99–111. [25] Nitsche MA, Cohen LG, Wassermann EM, Priori A, Lang N, Antal A, et al. Transcranial direct current stimulation: state of the art 2008. Brain Stimul 2008;1:206–23. [26] Kantak SS, Mummidisetty CK, Stinear JW. Primary motor and premotor cortex in implicit sequence learning—evidence for competition between implicit and explicit human motor memory systems. Eur J Neurosci 2012;36:2710–5. [27] Antal A, Nitsche MA, Kincses TZ, Kruse W, Hoffmann KP, Paulus W. Facilitation of visuo-motor learning by transcranial direct current stimulation of the motor and extrastriate visual areas in humans. Eur J Neurosci 2004;19:2888–92. [28] Reis J, Schambra HM, Cohen LG, Buch ER, Fritsch B, Zarahn E, et al. Noninvasive cortical stimulation enhances motor skill acquisition over multiple days through an effect on consolidation. Proc Natl Acad Sci U S A 2009;106:1590–5. [29] Saucedo Marquez CM, Zhang X, Swinnen SP, Meesen R, Wenderoth N. Taskspecific effect of transcranial direct current stimulation on motor learning. Front Hum Neurosci 2013;7:333. [30] Galea JM, Vazquez A, Pasricha N, de Xivry JJ, Celnik P. Dissociating the roles of the cerebellum and motor cortex during adaptive learning: the motor cortex retains what the cerebellum learns. Cereb Cortex 2011;21:1761–70. [31] Phillips AM, Birch NC, Ribbans WJ. Protective gloves for use in high-risk patients: how much do they affect the dexterity of the surgeon. Ann R Coll Surg Engl 1997;79:124–7. [32] Nitsche MA, Seeber A, Frommann K, Klein CC, Rochford C, Nitsche MS, et al. Modulating parameters of excitability during and after transcranial direct current stimulation of the human motor cortex. J Physiol 2005;568:291–303. [33] Fujiyama H, Hyde J, Hinder MR, Kim SJ, McCormack GH, Vickers JC, Summers JJ. Delayed plastic responses to anodal tDCS in older adults. Front Aging Neurosci 2014;6:115. [34] Kidgell DJ, Goodwill AM, Frazer AK, Daly RM. Induction of cortical plasticity and improved motor performance following unilateral and bilateral transcranial direct current stimulation of the primary motor cortex. BMC Neurosci 2013;14:64. [35] Kim S, Stephenson MC, Morris PG, Jackson SR. tDCS-induced alterations in GABA concentration within primary motor cortex predict motor learning and motor memory: a 7 T magnetic resonance spectroscopy study. NeuroImage 2014;99:237–43. [36] Antal A, Terney D, Poreisz C, Paulus W. Towards unravelling task-related modulations of neuroplastic changes induced in the human motor cortex. Eur J Neurosci 2007;26:2687–91. [37] Butts RJ, Kolar MB, Newman-Norlund RD. Enhanced motor skill acquisition in the non-dominant upper extremity using intermittent theta burst stimulation and transcranial direct current stimulation. Front Hum Neurosci 2014;8:451. [38] Nitsche MA, Roth A, Kuo MF, Fischer AK, Liebetanz D, Lang N, Tergau F, Paulus W. Timing-dependent modulation of associative plasticity by general network excitability in the human motor cortex. J Neurosci 2007;27:3807–12. [39] Kuo MF, Unger M, Liebetanz D, Lang N, Tergau F, Paulus W, et al. Limited impact of homeostatic plasticity on motor learning in humans. Neuropsychologia 2008;46:2122–8. [40] Jung P, Ziemann U. Homeostatic and nonhomeostatic modulation of learning in human motor cortex. J Neurosci 2009;29:5597–604. [41] Nägerl UV, Köstinger G, Anderson JC, Martin KA, Bonhoeffer T. Protracted synaptogenesis after activity-dependent spinogenesis in hippocampal neurons. J Neurosci 2007;27:8149–56. [42] Cantarero G, Lloyd A, Celnik P. Reversal of long-term potentiation-like plasticity processes after motor learning disrupts skill retention. J Neurosci 2013;33:12862–9. [43] Jacobs KM, Donoghue JP. Reshaping the cortical motor map by unmasking latent intracortical connections. Science 1991;251:944–7. [44] Donchin O, Sawaki L, Madupu G, Cohen LG, Shadmehr R. Mechanisms influencing acquisition and recall of motor memories. J Neurophysiol 2002;88:2114–23. [45] Di Lazzaro V, Rothwell JC. Corticospinal activity evoked and modulated by non-invasive stimulation of the intact human motor cortex. J Physiol 2014;592:4115–28.
Contents lists available at ScienceDirect
Behavioural Brain Research journal homepage: www.elsevier.com/locate/bbr
Research report
Cumulative effects of anodal and priming cathodal tDCS on pegboard test performance and motor cortical excitability Monica Christova a,b,∗ , Dietmar Rafolt c , Eugen Gallasch a a b c
Department of Physiology, Medical University of Graz, Austria Department of Physiotherapy, University of Applied Sciences FH-Joanneum, Austria Center for Medical Physics and Biomedical Engineering, Medical University Vienna, Austria
h i g h l i g h t s • • • •
atDCS during pegboard training improved motor performance. Preceding ctDCS induced an outlasting increase in cortical excitability. Preceding ctDCS improved off-line learning. Homeostatic plasticity is a possible factor for the cortical excitability effects.
a r t i c l e
i n f o
Article history: Received 5 January 2015 Received in revised form 10 March 2015 Accepted 14 March 2015 Available online 21 March 2015 Keywords: Skill learning Motor cortex Neuroplasticity Procedural memory tDCS TMS
a b s t r a c t Transcranial direct current stimulation (tDCS) protocols applied over the primary motor cortex are associated with changes in motor performance. This transcranial magnetic stimulation (TMS) study examines whether cathodal tDCS prior to motor training, combined with anodal tDCS during motor training improves motor performance and off-line learning. Three study groups (n = 36) were trained on the grooved pegboard test (GPT) in a randomized, between-subjects design: SHAM—sham stimulation prior and during training, STIM1—sham stimulation prior and atDCS during training, STIM2—ctDCS stimulation prior and atDCS during training. Motor performance was assessed by GPT completion time and retested 14 days later to determine off-line learning. Cortical excitability was assessed via TMS at baseline (T0), prior training (T1), after training (T2), and 60 min after training (T3). Motor evoked potentials (MEP) were recorded from m. abductor pollicis brevis of the active left hand. GPT completion time was reduced for both stimulated groups compared to SHAM. For STIM2 this reduction in time was significantly higher than for STIM1 and further off-line learning occurred after STIM2. After ctDCS at T1, MEP amplitude and intracortical facilitation was decreased and intracortical inhibition was increased. After atDCS at T2, an opposite effect was observed for STIM1 and STIM2. For STIM2 these neuromodulatory effects were retained until T3. It is concluded that application of atDCS during the training improves pegboard performance and that additional priming with ctDCS has a positive effect on off-line learning. These cumulative behavioral gains were indicated by the preceding neuromodulatory changes. © 2015 Elsevier B.V. All rights reserved.
1. Introduction From our experience we know that practicing a new motor task improves motor performance and that an acquired skill remains relative stable without additional practice. However the performance gains achieved within a certain training period are limited.
∗ Corresponding author at: Medical University of Graz, Department of Physiology, Harrachgasse 21/5, 8010 Graz, Austria. Tel.: +43 316 380 76 27; fax: +43 316 380 9630. E-mail address: [email protected] (M. Christova). http://dx.doi.org/10.1016/j.bbr.2015.03.028 0166-4328/© 2015 Elsevier B.V. All rights reserved.
This obviously depends on straits to reorganize cortical networks [1], but also on the utilization of cognitive resources at the early stage of motor learning [2]. Recently noninvasive brain stimulation has been probed successfully in order to enhance motor performance at various hand motor learning tasks. Often transcranial direct current stimulation (tDCS) was used in such studies [3], a technique that allows polarization of membrane potentials [4,5] and modulation of cortical excitability [6]. Basically tDCS induces a current flow through the skull, and it was suggested that cathodal polarization slightly decreases firing rate in the underlying brain tissue while anodal polarization slightly increases the firing rate [4]. After application of tDCS over the
28
M. Christova et al. / Behavioural Brain Research 287 (2015) 27–33
primary motor cortex such neuromodulatory effects can be observed via transcranial magnetic stimulation (TMS): after cathodal tDCS (ctDCS) cortical excitability decreases and after anodal tDCS (atDCS) excitability it increases [7,8]. For these effects transient changes in synaptic efficacy are held responsible. For example it was demonstrated with magnetic resonance spectroscopy that atDCS (1 mA, 10 min) locally reduces GABA, while ctDCS reduces glutamergic activity [9]. If tDCS is paired with motor learning it is crucial to consider timing and polarity of stimulation. In several studies atDCS was tested during practice in order to facilitate motor performance [10–12]. It is argued that increase of firing rates in task specific networks imposes additional strengthening of specific synaptic connections [13]. In some studies further priming tDCS (tDCS in advance of practice) was tested. Stagg et al. [14] found slower learning of a finger sequence task for both atDCS and ctDCS, while Antal et al. [15] found an improvement in early learning of a visuomotor coordination task for both polarities. Despite these inconsistencies in learning paradigms, lowering of neuronal activity with priming ctDCS seems to be advantageous as it reduces the threshold for subsequent protocols to increase cortical excitability [16,17]. Also the site of stimulation has to be considered. At tasks with high cognitive demands the anode often is placed over the premotor cortex [18,19], while at tasks depending more on sensory input placement over the primary motor cortex showed appropriate [10,12,20]. In most of the studies the impact of tDCS has been exclusively explored at the behavioral level. However to elucidate the underlying adaptive processes, it is advantageous to incorporate noninvasive neuronal measures. Therefore in the current study we measured motor evoked potential (MEP) amplitudes in order to evaluate changes of cortical excitability. As training paradigm the grooved pegboard test (GPT) was chosen. It was shown that the GPT itself evokes rather minor excitability changes [21,22] and therefore the tDCS-induced changes should dominate the picture. The study protocol was designed in order to examine whether ctDCS prior to motor training, combined with atDCS during motor training, cumulatively enhances pegboard test performance and off-line learning. We further assume that the atDCS and priming ctDCS driven changes in cortical excitability are related to the expected behavioral gains. To verify these assumptions three matched study groups were tested in a randomized, single-blind, between-subject design: one group receiving sham stimulation prior and during GPT training, a second group receiving sham stimulation prior and atDCS during the training and a third group receiving ctDCS prior and atDCS during the training. For all three groups single and paired pulse TMS assessments were performed prior to priming stimulation (sham or ctDCS), after priming stimulation, immediately after practice, and 60 min after practice. To assess off-line learning GPT performance was retested two weeks later.
2. Material and methods 2.1. Participants and study design Thirty-six healthy volunteers took part in the study approved by the Ethics Committee at the Medical University of Graz. They were screened for possible neurological disorders and contraindications to TMS and signed a written informed consent. All participants were right-handed according to the laterality quotient (LQ) from the Edinburgh Handedness Inventory [23]. According to self report none of the participants actively played an instrument or engaged in any other activity that extensively involved the left nondominant hand. Following acquisition of a novel task the same subject could not be tested again, thus the participants were randomly assigned to three study groups, (n = 12) each receiving
different tDCS protocol. The first group (8 ♀ and 4 ♂, mean age 24.92 ± 5.04 years) underwent sham tDCS (SHAM). The second group (8 ♀ and 4 ♂, mean age 27.67 ± 9.98 years) underwent sham stimulation preceding the GPT and atDCS during the GPT performance (STIM 1). The third group (6 ♀ and 6 ♂, mean age 26.00 ± 8.91 years) underwent ctDCS preceding GPT and atDCS during GPT performance (STIM2). 2.2. Grooved Pegboard test and skill training The use of GPT (Model 32025, Lafayette Instrument, USA) is well documented for the ability to generate performance curves and is also used to assess motor function in patients with motor deficits. In this study, the GPT was performed in 4 blocks (4 trials in each block) with interblock rest intervals of 2 min to avoid muscle and central fatigue. The GPT was retested two weeks after skill training in a single block (4 trials). Subjects received exact instructions and a demonstration of the test, without being provided with practice trials. All participants completed the test at equal ambient conditions at the same time of day (between 9 and 12 a.m.). The observer instructed the subjects to complete the task as fast as possible and recorded the time for each trial. 2.3. Transcranial direct current stimulation (tDCS) A bipolar electrode montage was utilized to deliver tDCS. The active electrode was placed to stimulate the right primary motor cortex M1, contralaterally to the performing left hand. The electrode was centered on C4 of the international 10–20 electroencephalogram system as it was shown in neuroimaging studies that C3/C4 corresponds to the left and right M1 [24]. The correspondence between C4 and right M1 was additionally confirmed using TMS individually for each subject and adjusted when necessary to the APB representation spot. The reference electrode was placed over the contralateral left supraorbital area. tDCS was delivered by the MAGSTIM ELDITH DC-stimulator and a pair of non-metallic, conductive rubber electrodes with water-soaked synthetic sponges (5 cm × 7 cm/35 cm2 ). Stimulations were delivered at an intensity of 1 mA (current density 0.029 mA/cm2 ). As it was shown that a stimulation period between 9 to 20 min is appropriate to induce stable after effects in the motor cortex [25], priming ctDCS was applied for 15 min. The period of atDCS during GPT practice was approximately 20 min in order to cover the overall training period. For sham tDCS the electrodes were placed in the same manner, however the stimulation was turned off after 15 s [16]. This procedure allows to blind subjects for the stimulation condition [10]. 2.4. TMS assessments For the TMS assessment, two Magstim 200 magnetic stimulators connected via a Bistim module (The Magstim Company, Whitland, Dyfed, UK) were employed. Magnetic pulses were delivered through a figure eight-shaped coil (outer loop diameter of 9 cm). The coil was positioned on the scalp over the right motor cortex at the optimal site for stimulating the contralateral left abductor pollicis brevis (APB). The intersection of the coil was placed tangentially to the scalp, with the handle pointing backward and laterally at a 45◦ angle away from the midline. The resting motor threshold (rMT) was expressed as a percentage of the maximum output of the stimulator. Using suprathreshold intensities, the coil was moved over the scalp in small steps to locate the site with the largest MEP. This position was marked on an EEG cap and the coil was fixed at that position. The resting motor threshold (rMT), MEP, short-latency intracortical inhibition (SICI) and intracortical facilitation (ICF) were examined. The resting MT was defined as the lowest stimulus
M. Christova et al. / Behavioural Brain Research 287 (2015) 27–33
29
Fig. 1. Experimental design. Grooved pegboard test (GPT) is trained in a session of 4 blocks × 4 trials and retested after 2 weeks in a session of 1 block × 4 trials. Three study groups: sham stimulation before/sham stimulation during GPT (SHAM), sham stimulation before/atDCS during GPT (STIM1) and ctDCS before/atDCS during GPT (STIM2). TMS assessments were taken at four time points: T0—at baseline, T1—after priming tDCS, T2—after tDCS/GPT and T3—60 min after tDCS/GPT.
intensity at which five out of ten consecutive TMS applications led to a MEP of at least 0.05 mV on the relaxed APB. MEPs were obtained with three stimulus intensities: 110%, 130% and 150% of rMT in a random order. Eight single pulses were delivered at each stimulus intensity. SICI and ICF were assessed using the pairedpulse paradigm with interstimulus intervals (ISI) of 3 ms and 13 ms. Eight conditioned pulses were obtained for each ISI. The conditioning and test stimuli were 80% and 120% of the corresponding rMT at the actual assessment. Four TMS assessments were performed: baseline (T0), after 15 priming tDCS (T1), after the GPT/tDCS (T2) and 60 min after the GPT/tDCS (T3) (Fig. 1). 2.5. Electromyographic (EMG) recordings MEPs were recorded from APB muscle of the left hand using 9 mm diameter Ag–AgCl surface disk electrodes in a tendonbelly manner. EMG signals were amplified, band pass filtered (8–2000 Hz), digitized (sampling rate 10 kHz) and recorded on a disk (DasyLab 8.0 software package). The EMG signal was displayed continuously during TMS assessment to ensure the absence of any voluntary motor activity. 2.6. Data analysis and statistics 2.6.1. Motor performance, on- and off-line learning Motor performance was quantified as the time to complete the GPT for each trial. It was then displayed in the form of inverted learning curves. For statistical analysis, the mean time per block was calculated. An ANOVA with a within-subject factor block (block1, block2, block3, block4, retest) and a between-subject factor group (SHAM, STIM1, STIM2) was performed. Where the ANOVAs showed significance, further one-factorial ANOVAs with Bonferoni adjustment for multiple comparisons were performed. On-line learning was quantified by the normalized time values for each block. Accordingly four values were determined by the ratio mean time block n/mean time block 1 (n = 1, 2, 3, 4). For offline learning one value was determined by the ratio mean time retest/mean time block1. An ANOVA with within-subject factors block (block1, block2, block3, block4, retest) and a between-subject factor group (SHAM, STIM1, STIM2) was carried out on these values. Where the ANOVA showed significance one-factorial ANOVAs with Bonferoni adjustment for multiple comparisons were performed. In the following text and in Fig. 3, on- and off-line learning is referred to as “performance improvement”. 2.6.2. Motor cortical excitability rMT, expressed as the percentage of maximum stimulator output (% mean ± S.E.M.), was assessed using a two-factorial repeated measures ANOVA (analysis of variance) with a within-subject factor time (4 levels: T0, T1, T2 and T3) and a between-subject factor group (3 levels: SHAM, STIM1, STIM2). The MEP amplitudes were presented in mV (mean ± S.E.M.). For each stimulation intensity (110%, 130% and 150%), the MEP amplitudes were analyzed separately using two-factorial ANOVA with
Fig. 2. Performance curves. Performance curves presenting the GPT completion time, (mean ± S.E.M) for each group: SHAM, STIM1 (sham before/atDCS during GPT) and STIM2 (ctDCS before/atDCS during GPT) in the four training blocks (4 trials per block) and in the retest (4 trials).
within-subject factors: time (4 levels: T0, T1, T2 and T3) and a between-subject factor group (3 levels: SHAM, STIM1, STIM2). In case of significant interactions, follow-up ANOVAs with Bonferoni adjustment for multiple comparisons with between-subject factor group for each time level and within-subject factor time for each group were conducted. For the conditioned MEP responses (SICI and ICF), the amplitudes were calculated as a percentage of the single pulse MEP for each subject individually (mean ± S.E.M.). For each ISI (3 and 13 ms), a repeated measures ANOVA was used to compare the conditioned MEP amplitudes with within-subject factors time (4 levels: T0, T1, T2 and T3) and a between-subject factor group (3 levels: SHAM, STIM1, STIM2). In case of significant main or interaction effects, follow-up ANOVAs with Bonferoni adjustment for multiple comparisons with between-subject factor group for each time level and within-subject factor time for each group were conducted. In all statistical tests a significance level of 0.05 was used. 3. Results 3.1. Motor performance and performance improvement The performance curves data (Fig. 2) show that overall the time to complete the GPT was longer for the sham group (69.87 ± 2.81 s) than for STIM1 (62.63 ± 2.51 s) and STIM2 (59.83 ± 2.18 s). An ANOVA revealed a significant main effect of block (F(4, 144) = 141.34, p < 0.001, 2 = 0.797), but insignificant interaction effect of block × group (F(8, 144) = 1.32, p > 0.05, 2 = 0.068). In block 1 no significant difference between the groups was found, however in the following training blocks between-groups differences were observed: block 2 (p = 0.04), block 3 (p = 0.01), block4 (p = 0.02) and retest (p < 0.001). In block 2 the GPT completion time was shorter for STIM 2 compared to SHAM p = 0.05). In blocks 2, 3 and 4 no difference was found between STIM1 and STIM2. In blocks 3, 4 and retest both stimulated groups showed shorter GPT completion time compared to SHAM (ps < 0.03). Further at retest STIM2 showed significantly shorter performance time in comparison to STIM1 (p = 0.04). To follow the performance improvement over the four blocks (on-line learning) and in the retest (off-line learning) data are displayed as normalized time values for each group separately, see Fig. 3. An ANOVA revealed a significant main effect of block
30
M. Christova et al. / Behavioural Brain Research 287 (2015) 27–33
Fig. 3. Performance improvement. On-line (blocks 1–4) and off-line improvement (retest after 2 weeks) for the three groups: SHAM, STIM1 (sham before/atDCS during GPT) and STIM2 (ctDCS before/atDCS during GPT). Data are presented as the percentage of the time in block 1 for each group (mean ± SD). The significant within group differences between the baseline training block (Block1) and the following training blocks (Block 2, 3, 4, retest) are presented * p < 0.05.
(F(4, 144) = 147.61, p < 0.001, 2 = 0.804) and a tendency for interaction effect of block × group (F(8, 144) = 1.73, p = 0.09, 2 = 0.088). Across all groups significant performance improvement occurred (21% for SHAM, 22% for STIM1, 20% for STIM2); however, no difference between the groups was found over the four training blocks. In comparison to the last training block 4, at the retest the normalized time values were significantly decreased only for STIM2 (−6%, p = 0.01) and slightly increased for STIM1 (+2%) and SHAM (+3%).
the MEP amplitude increase reached significance only at intensity 110% (p = 0.02). At time level T3 (60 min after GPT/tDCS) significant between-groups difference was observed at all stimulation intensities (ps < 0.05), where STIM2 showed increases MEP
3.2. MEP results The mean rMT values measured at baseline (T0) were 39.27 ± 1.1% for SHAM, 41.93 ± 2.2% for STIM1 and 39.00 ± 1.4% for STIM2. No significant differences were observed. MEP amplitudes at stimulation intensity 110%, 130% and 150%, obtained at time levels T0, T1, T2 and T3 for each group are presented in Fig. 4. For all stimulation intensities a two-factorial ANOVA revealed a significant main effect of time (Fs(3, 102) > 6.17, ps < 0.01, s2 < 0.290) and an interaction effect of time × group (Fs(6, 102) > 3.46, ps < 0.004, s2 < 0.310). To reveal the within-groups cortical excitability effects across the four TMS assessments, one-factorial ANOVAs with factor time were conducted for each group and for each TMS intensity. The SHAM group didn’t show any MEP amplitude changes over the time. The STIM1 group showed significant cortical excitability changes at stimulation intensities 130% and 150% (ps < 0.001) where a longlasting MEP increase was found at time levels T2 and T3. The STIM2 group showed significant MEP amplitude changes at all stimulation intensities (P ps < 0.03), where at time level T1 (after priming tDCS) the MEP amplitudes decreased, at time level T2 (after GPT/atDCS) they increased and at time level T3 (60 min after the end of GPT/tDCS) they reached their maximum values. Further to reveal the between-groups excitability effects, onefactorial ANOVAs with factor group were conducted for each time level and for each TMS intensity. At time levels T0 (baseline) and T1 (after priming tDCS) no significant differences between the four groups were found. At time level T2 (after GPT/tDCS) differences between groups were found at intensities 110% (p = 0.06) and at 130% (p = 0.04). The MEP amplitudes at these intensities were increased for STIM2 compared to SHAM. For STIM1
Fig. 4. MEP amplitudes. MEP amplitudes at stimulation intensities 110%, 130% and 150% of the rMT for the three groups: SHAM, STIM1 (sham before/atDCS during GPT) and STIM2 (ctDCS before/atDCS during GPT), at baseline (T0), after preconditioning tDCS (T1), immediately after (T2) and 60 min after (T3) the end of GPT/tDCS. MEP amplitudes (mV) are plotted for each stimulation intensity as the mean (S.E.M). The significant differences between T0 and the following TMS assessments are presented * p < 0.05.
M. Christova et al. / Behavioural Brain Research 287 (2015) 27–33
31
Fig. 5. Short-latency intracortical inhibition (SICI). Normalized paired-pulse responses (ISI 3 ms) for the three groups: SHAM, STIM1 (sham before/atDCS during GPT) and STIM2 (ctDCS before/atDCS during GPT), at baseline (T0), after preconditioning tDCS (T1), immediately after (T2) and 60 min (T3) after the end of GPT/tDCS. Values are normalized for each subject to their corresponding values in single pulse stimulation and plotted as the mean (S.E.M.). The significant differences between T0 and the following TMS assessment are presented * p < 0.05.
amplitudes in comparison to SHAM. No differences were observed for STIM1. The results from the paired-pulse stimulation are shown in Figs. 5 and 6. Separate ANOVAs were conducted for ISIs (3, 13 ms) to reveal the effect of time and group on SICI and ICF, respectively. For SICI, a significant main effect of time (F(3, 102) = 8.53, p < 0.001, 2 = 0.201) and interaction effect of time × group (F(6, 102) = 6.18, p < 0.001, 2 = 0.267) were found. For ICF, also significant main effects of time (F(3, 102) = 6.08, p < 0.001, 2 = 0.152) and interaction effect of time × group (F(6, 102) = 2.26, p = 0.04, 2 = 0.117) were observed. To reveal the within-group intracortical excitability effects across the four TMS assessments, follow-up one-factorial ANOVAs with factor time were conducted for each group. To reveal the within-groups intracortical excitability effects across the four TMS assessments, one-factorial ANOVAs with factor time were conducted for each group. For SHAM no significant
differences were found. For STIM1 group significant effect of time was found for SICI (p < 0.001) and a tendency for ICF (p = 0.05). At time level T2 a reduced effect of SICI (expressed with increased value) +20% and increased ICF +30% were found. These effects were transient and at time level T3 the SICI and ICF values reached back their baseline levels. For STIM2 group, significant effect of time was observed, both for SICI and ICF (ps < 0.001). At time level T1 (after priming tDCS) an increased effect of SICI (expressed with decreased value) −40% (p < 0.001) and a tendency for decreased effect of ICF −23% (p = 0.05) were found. Further, at time levels T2 (after GPT/tDCS) SICI effect was reduced (p = 0.03) and ICF effect was increased (p = 0.02). At time level T3 (60 min after GPT/tDCS) these effects were still present, however they reached significant level only for ICF. To reveal the between-groups differences in SICI and ICF onefactorial ANOVAs with factor group were conducted for each
Fig. 6. Intracortical facilitation (ICF). Normalized paired-pulse responses (ISI 13 ms) for the three groups: SHAM, STIM1 (sham before/atDCS during GPT) and STIM2 (ctDCS before/atDCS during GPT), at baseline (T0), after preconditioning tDCS (T1), immediately after (T2) and 60 min (T3) after the end of GPT/tDCS. Values are normalized for each subject to their corresponding values in single pulse stimulation and plotted as the mean (S.E.M.). The significant differences between T0 and the following TMS assessment are presented * p < 0.05.
32
M. Christova et al. / Behavioural Brain Research 287 (2015) 27–33
time level. At time level T0 (baseline), no significant differences between the four groups were found, either for SICI or for ICF. At time level T1 (after priming tDCS) significant effect of group was found, (ps < 0.001), whereby for STIM2 group SICI was increased (ps < 0.001) and ICF was decreased (ps < 0.03) compared to SHAM and to STIM1. For STIM1 ICF was increased compared to SHAM (p = 0.04). At time level T2 (after GPT/tDCS) significant effect of group was observed for SICI which was decreased for STIM1 group in comparison to SHAM (p = 0.02). At time level T3 no significant differences between the groups were found. 4. Discussion The purpose of this study was to examine the enhancing effects of atDCS and priming ctDCS on pegboard performance and offline learning in relation to associated changes in motor cortical excitability. As expected, pegboard training itself induced no shifts in motor cortical excitability. On the other hand atDCS increased excitability after the training and priming ctDCS reduced excitability ahead of the training. Thus for the forthcoming discussion solely these externally induced excitability shifts have to be taken into account. 4.1. Effects of atDCS during the GPT training The behavioral results show that atDCS delivered concurrently with GPT training resulted in shorter GPT performance times in relation to SHAM, both during the second half of the training session, and 14 days later at retest see, Fig. 2. Comparable gains in motor performance for example have been described for serial reaction time tasks [10,14,26] and for a visuomotor task [27]. Further comparable gains at retest have been described for an isometric pinch force task [28,29] and for a visuomotor transformation task [30]. In contrast to these gains in motor performance, performance improvement (on- and off-line learning) remained unchanged, see Fig. 3. Comparable brain stimulation effects on on-line learning values were not described, but in studies where pegboard performance was tested with the bare and the gloved hand, these values also remained unchanged [22,31]. With regard to the current results it appears that on-line learning is less prone to interventions. However the number of participants in the current study (n = 12) is rather low which is a limitation to resolve such questions. The facilitatory after effects on motor cortical excitability, as observed here, are comparable to TMS studies where atDCS was tested at rest without practicing a motor task [6,32,33]. Thus the results of our study show that atDCS is able to induce similar after effects in the background of learning related motor activity. Comparable effects, increased MEP amplitude for 60 min and decreased effect of SICI for 30 min were also described after atDCS over the nondominant M1 with the Perdue Pegboard test [34]. Further a reduction in intracortical inhibitory networks, related to significant reduction of GABA concentration within M1, was documented with magnetic resonance spectroscopy after a motor adaptation task paired with atDCS [35]. Additionally it has to be mentioned here, that the tDCS-induced after effects depend also on the type of the training task [36]. 4.2. Effects of atDCS and priming ctDCS The behavioral results show a reduction of the GPT performance time also in the first half of the training session and a further reduction of performance time at the retest. Thus a cumulative effect is observed as priming ctDCS additionally reduces performance times in relation to SHAM. A similar effect was found for off-line learning. While STIM1 was not able to influence retention of motor memory, there was a positive effect on memory in the
group who received STIM2. A limitation of this study is that priming ctDCS followed by sham stimulation was not tested by a fourth group. It is however rather unlikely that priming ctDCS alone would have produced these effects, see previous results [14,15], and discussion on homeostatic plasticity below. Off-line learning further indicates that consolidation has occurred after the training. Likewise off-line learning was reported by Butts et al. [37] who primed cortical excitability with two potentiating protocols (iTBS followed by bihemispheric tDCS) and found performance improvement on the Jebsen–Taylor hand function test with a delay of 1–7 days after training. The inhibitory effect on cortical excitability ahead of the training again is comparable to TMS studies where ctDCS was tested at rest [6]. However the most striking findings in comparison to STIM1 are the prolonged shifts of SICI and ICF, indicating increased intracortical excitability at least until 60 min after the training. As pegboard training itself showed no effect on excitability there is an evidence that priming ctDCS followed by atDCS generates this delayed neuroplastic effect. Such delayed effects are in coincidence with the concepts of homeostatic plasticity [17]. According to these concepts lowering of neuronal activity with preceding ctDCS reduces the threshold for subsequent brain stimulation protocols in order to increase the corticospinal excitability. Such delayed effects have been tested with brain stimulation protocols a rest. For example increase of MEP amplitude was found after high frequency rTMS preconditioned by ctDCS [7] where the after effects were examined within 30 min. Further the differential impact of preconditioning tDCS was shown by Siebner et al. [16], where 1 Hz rTMS increased cortical excitability when primed by ctDCS but decreased it when primed by atDCS. On the other hand the concept that the history of the network activity influences subsequent brain stimulation protocols is not compulsory. For example in a study were paired associative stimulation (PAS) was used to mimic learning learning-like synaptic plasticity [38], homeostatic modulation of motor cortical plasticity occurred when tDCS was applied simultaneously with PAS, but not when applied before PAS. Recently a few studies implemented the homeostatic plasticity rules in order to boost motor learning demonstrating both homeostatic and nonhomeostatic modulation of learning. Kuo et al. [39] after combining atDCS with NMDA agonist, found reduced performance in a sequential learning task, however after combining ctDCS with NMDA agonist no change was found. Further the time delay between priming stimulation and start of training was shown to be of importance [40]. A PAS protocol for excitability depression (PASLTD ) and potentiation (PASLTP ) facilitated subsequent thumb flexion movements when performed immediately before the task execution. But when performed 90 min before the task, PASLTD facilitated and PASLTP diminished the motor performance. In our study ctDCS preceded the GPT by 10 min, which seemed to be sufficient to enhance pegboard performance and to improve retention of motor memory. Currently the mechanisms of such delayed effects are not completely understood, obviously because the effects on learning depend on a variety of parameters such as the stimulation modality, timing and the type of the training task. For the effect on motor memory LTP-like plasticity and synaptic strengthening are of importance. Due to the prolonged increase of intracortical excitability after the training LTP-like processes will become boosted within M1. LTP involves a cascade of processes over time, which can increase the efficacy of existing synapses and promote the formation of new synapses [41]. Accordingly, LTP-like processes are thought to continue after training, thus stabilizing the memory traces of the newly learned task. Supporting evidence comes from a recent study [42] where it was shown that experimental reversal of LTP-like plasticity immediately after motor training was able to disrupt skill retention. In addition to LTP, GABAergic mechanisms play a role for the adjustment of motor
M. Christova et al. / Behavioural Brain Research 287 (2015) 27–33
cortical representations [43]. For example a pharmacological study has shown that enhancing of GABAa receptor function with administration of lorazepam results in impaired motor learning and formation of new memories [44]. Currently it is unclear why pegboard training itself left no trace in the measured MEP amplitudes. Perhaps some training-induced neuroplasticity occurs beyond the motor cortex, and atDCS acts indirectly via functional connections within a wider network? But it is also conceivable that neuroplasticity arises at circuits in the motor cortex which do not contribute to the transsynaptically evoked indirect waves elicited via magnetic stimulation with a figure of eight-shaped coil in posterior-anterior direction [45]. Further studies including brain stimulation protocols to elicit direct waves could be helpful to clarify this enigma. 5. Conclusions In conclusion this study shows that atDCS in combination with priming ctDCS decreases the time to complete a repetitive motor task which could have an impact on neurorehabilitative science. Acknowledgments This work was supported by the Research Fund of Government of Styria, Austria. References [1] Luft AR, Buitrago MM. Stages of motor skill learning. Mol Neurobiol 2005;32:205–16. [2] Seidler RD, Bo J, Anguera JA. Neurocognitive contributions to motor skill learning: the role of working memory. J Mot Behav 2012;4:445–53. [3] Reis J, Fritsch B. Modulation of motor performance and motor learning by transcranial direct current stimulation. Curr Opin Neurol 2011;24:590–6. [4] Bindman LJ, Lippold OC, Redfearn JW. The action of brief polarizing currents on the cerebral cortex of the rat (1) during current flow and (2) in the production of long-lasting after-effects. J Physiol 1964;172:369–82. [5] Priori A, Berardelli A, Rona S, Accornero N, Manfredi M. Polarization of the human motor cortex through the scalp. NeuroReport 1998;9:2257–60. [6] Nitsche MA, Paulus W. Excitability changes induced in the human motor cortex by weak transcranial direct current stimulation. J Physiol 2000;527:633–9. [7] Lang N, Siebner HR, Ernst D, Nitsche MA, Paulus W, Lemon RN, et al. Preconditioning with transcranial direct current stimulation sensitizes the motor cortex to rapid-rate transcranial magnetic stimulation and controls the direction of after-effects. Biol Psychiatry 2004;56:634–9. [8] Di Lazzaro V, Ranieri F, Profice P, Pilato F, Mazzone P, Capone F, et al. Transcranial direct current stimulation effects on the excitability of corticospinal axons of the human cerebral cortex. Brain Stimul 2013;6:641–3. [9] Stagg CJ, Bachtiar V, Johansen-Berg H. The role of GABA in human motor learning. Curr Biol 2011;21:480–4. [10] Nitsche MA, Schauenburg A, Lang N, Liebetanz D, Exner C, Paulus W, et al. Facilitation of implicit motor learning by weak transcranial direct current stimulation of the primary motor cortex in the human. J Cogn Neurosci 2003;15:619–26. [11] Galea JM, Celnik P. Brain polarization enhances the formation and retention of motor memories. J Neurophysiol 2009;102:294–301. [12] Prichard G, Weiller C, Fritsch B, Reis J. Effects of different electrical brain stimulation protocols on subcomponents of motor skill learning. Brain Stimul 2014;7:532–40. [13] Fritsch B, Reis J, Martinowich K, Schambra HM, Ji Y, Cohen LG, et al. Direct current stimulation promotes BDNF-dependent synaptic plasticity: potential implications for motor learning. Neuron 2010;66:198–204. [14] Stagg CJ, Jayaram G, Pastor D, Kincses ZT, Matthews PM, Johansen-Berg H. Polarity and timing-dependent effects of transcranial direct current stimulation in explicit motor learning. Neuropsychologia 2011;49:800–4. [15] Antal A, Begemeier S, Nitsche MA, Paulus W. Prior state of cortical activity influences subsequent practicing of a visuomotor coordination task. Neuropsychologia 2008;46:3157–61. [16] Siebner HR, Lang N, Rizzo V, Nitsche MA, Paulus W, Lemon RN, et al. Preconditioning of low-frequency repetitive transcranial magnetic stimulation with transcranial direct current stimulation: evidence for homeostatic plasticity in the human motor cortex. J Neurosci 2004;24:3379–85. [17] Ziemann U, Siebner HR. Modifying motor learning through gating and homeostatic metaplasticity. Brain Stimul 2008;1:60–6.
33
[18] Nitsche MA, Jakoubkova M, Thirugnanasambandam N, Schmalfuss L, Hullemann S, Sonka K, et al. Contribution of the premotor cortex to consolidation of motor sequence learning in humans during sleep. J Neurophysiol 2010;104:2603–14. [19] Pavlova E, Kuo MF, Nitsche MA, Borg J. Transcranial direct current stimulation of the premotor cortex: effects on hand dexterity. Brain Res 2014;1576:52–62. [20] Antal A, Polania R, Schmidt-Samoa C, Dechent P, Paulus W. Transcranial direct current stimulation over the primary motor cortex during fMRI. NeuroImage 2011;55:590–6. [21] McDonnell MN, Ridding MC. Transient motor evoked potential suppression following a complex sensorimotor task. Clin Neurophysiol 2006;117:1266–72. [22] Christova M, Rafolt D, Golaszewski S, Nardone R, Gallasch E. Electrical stimulation during skill training with a therapeutic glove enhances the induction of cortical plasticity and has a positive effect on motor memory. Behav Brain Res 2015, http://dx.doi.org/10.1016/j.bbr.2015.03.028. [23] Oldfield RC. The assessment and analysis of handedness: the Edinburgh inventory. Neuropsychologia 1971;9:97–113. [24] Okamoto M, Dan H, Sakamoto K, Takeo K, Shimizu K, Kohno S, et al. Three-dimensional probabilistic anatomical cranio-cerebral correlation via the international 10–20 system oriented for transcranial functional brain mapping. NeuroImage 2004;21:99–111. [25] Nitsche MA, Cohen LG, Wassermann EM, Priori A, Lang N, Antal A, et al. Transcranial direct current stimulation: state of the art 2008. Brain Stimul 2008;1:206–23. [26] Kantak SS, Mummidisetty CK, Stinear JW. Primary motor and premotor cortex in implicit sequence learning—evidence for competition between implicit and explicit human motor memory systems. Eur J Neurosci 2012;36:2710–5. [27] Antal A, Nitsche MA, Kincses TZ, Kruse W, Hoffmann KP, Paulus W. Facilitation of visuo-motor learning by transcranial direct current stimulation of the motor and extrastriate visual areas in humans. Eur J Neurosci 2004;19:2888–92. [28] Reis J, Schambra HM, Cohen LG, Buch ER, Fritsch B, Zarahn E, et al. Noninvasive cortical stimulation enhances motor skill acquisition over multiple days through an effect on consolidation. Proc Natl Acad Sci U S A 2009;106:1590–5. [29] Saucedo Marquez CM, Zhang X, Swinnen SP, Meesen R, Wenderoth N. Taskspecific effect of transcranial direct current stimulation on motor learning. Front Hum Neurosci 2013;7:333. [30] Galea JM, Vazquez A, Pasricha N, de Xivry JJ, Celnik P. Dissociating the roles of the cerebellum and motor cortex during adaptive learning: the motor cortex retains what the cerebellum learns. Cereb Cortex 2011;21:1761–70. [31] Phillips AM, Birch NC, Ribbans WJ. Protective gloves for use in high-risk patients: how much do they affect the dexterity of the surgeon. Ann R Coll Surg Engl 1997;79:124–7. [32] Nitsche MA, Seeber A, Frommann K, Klein CC, Rochford C, Nitsche MS, et al. Modulating parameters of excitability during and after transcranial direct current stimulation of the human motor cortex. J Physiol 2005;568:291–303. [33] Fujiyama H, Hyde J, Hinder MR, Kim SJ, McCormack GH, Vickers JC, Summers JJ. Delayed plastic responses to anodal tDCS in older adults. Front Aging Neurosci 2014;6:115. [34] Kidgell DJ, Goodwill AM, Frazer AK, Daly RM. Induction of cortical plasticity and improved motor performance following unilateral and bilateral transcranial direct current stimulation of the primary motor cortex. BMC Neurosci 2013;14:64. [35] Kim S, Stephenson MC, Morris PG, Jackson SR. tDCS-induced alterations in GABA concentration within primary motor cortex predict motor learning and motor memory: a 7 T magnetic resonance spectroscopy study. NeuroImage 2014;99:237–43. [36] Antal A, Terney D, Poreisz C, Paulus W. Towards unravelling task-related modulations of neuroplastic changes induced in the human motor cortex. Eur J Neurosci 2007;26:2687–91. [37] Butts RJ, Kolar MB, Newman-Norlund RD. Enhanced motor skill acquisition in the non-dominant upper extremity using intermittent theta burst stimulation and transcranial direct current stimulation. Front Hum Neurosci 2014;8:451. [38] Nitsche MA, Roth A, Kuo MF, Fischer AK, Liebetanz D, Lang N, Tergau F, Paulus W. Timing-dependent modulation of associative plasticity by general network excitability in the human motor cortex. J Neurosci 2007;27:3807–12. [39] Kuo MF, Unger M, Liebetanz D, Lang N, Tergau F, Paulus W, et al. Limited impact of homeostatic plasticity on motor learning in humans. Neuropsychologia 2008;46:2122–8. [40] Jung P, Ziemann U. Homeostatic and nonhomeostatic modulation of learning in human motor cortex. J Neurosci 2009;29:5597–604. [41] Nägerl UV, Köstinger G, Anderson JC, Martin KA, Bonhoeffer T. Protracted synaptogenesis after activity-dependent spinogenesis in hippocampal neurons. J Neurosci 2007;27:8149–56. [42] Cantarero G, Lloyd A, Celnik P. Reversal of long-term potentiation-like plasticity processes after motor learning disrupts skill retention. J Neurosci 2013;33:12862–9. [43] Jacobs KM, Donoghue JP. Reshaping the cortical motor map by unmasking latent intracortical connections. Science 1991;251:944–7. [44] Donchin O, Sawaki L, Madupu G, Cohen LG, Shadmehr R. Mechanisms influencing acquisition and recall of motor memories. J Neurophysiol 2002;88:2114–23. [45] Di Lazzaro V, Rothwell JC. Corticospinal activity evoked and modulated by non-invasive stimulation of the intact human motor cortex. J Physiol 2014;592:4115–28.
