- Email: [email protected]
Dynamical changes in corticospinal excitability during imagery of unimanual and bimanual wrist movements in humans: a transcranial magnetic stimulation study
Neuroscience Letters 359 (2004) 185–189 www.elsevier.com/locate/neulet
Dynamical changes in corticospinal excitability during imagery of unimanual and bimanual wrist movements in humans: a transcranial magnetic stimulation study O. Levin*, M. Steyvers, N. Wenderoth, Y. Li, S.P. Swinnen Motor Control Laboratory, Department of Kinesiology, Group Biomedical Sciences, Katholieke Universiteit Leuven, Tervuurse Vest 101, 3001 Heverlee, Belgium Received 7 November 2003; received in revised form 29 January 2004; accepted 29 January 2004
Abstract This study explored the dynamical changes in corticospinal excitability during the imagination of cyclical unimanual and bimanual wrist flexion –extension movements. Transcranial magnetic stimulation was applied over the left motor cortex to evoke motor evoked potentials in the right wrist flexor and extensor muscles. Findings provided evidence for increased reciprocal excitability changes during imagery of symmetrical in-phase movements as compared to asymmetrical (anti-phase) or unimanual movements. This suggests that in-phase movements may reinforce whereas anti-phase movements may reduce the temporal representation of the task in the corticospinal motor networks of the brain. q 2004 Elsevier Ireland Ltd. All rights reserved. Keywords: Bimanual coordination; Motor imagery; Transcranial magnetic stimulation; Temporal representation; Corticospinal excitability
Functional imaging studies have revealed that motor imagery activates large parts of the central nervous system (CNS) motor networks such as the cerebellum, premotor cortex (PMC), supplementary motor area (SMA) and the primary motor cortex (M1) [2,3,13 –15]. While SMA and PMC are equally activated during both actual and imagined movements, the levels of activation in M1 during imagery have been found lower than those observed during actual execution of the same movements [13,15]. Complementary studies using transcranial magnetic stimulation (TMS) have demonstrated an increase in M1 excitability during imagery of a wide range of motor tasks [6,7,9,10,12, 18]. Additionally, motor evoked potential (MEP) facilitation in the targeted muscles has been shown to result mainly from increased cortical excitability and not from spinal motorneural excitability [9,10,12]. For unimanual movements, Hashimoto and Rothwell [9] demonstrated that imagination of cyclical flexion–extension movements produces temporal changes in the size of MEPs similar to those expected during actual performance of the task. * Corresponding author. Tel.: þ 32-16-329065; fax: þ 32-16-329197. E-mail address: [email protected] (O. Levin).
Unimanual and bimanual tasks employ largely overlapping neural resources [17]. With respect to bimanual coordination, brain activation levels appear to increase with task complexity, e.g. coordination of limb movements in the anti-phase mode is associated with higher activation levels in the SMA, PMC and cingulate motor area as compared to the in-phase mode [8,17, 20]. The engagement of SMA in imagining and performing coordinated movements suggests that it is likely involved in planning and preparation of action sequences, whether imagined or real [14]. Here, we address the question whether imagining cyclical bimanual vs. unimanual movements has differential effects on the temporal patterns of corticospinal excitability as measured by TMS. More specifically, we assume that the imagination of symmetric (in-phase) movements might result in stronger synchronization of the activity of bilateral motorneuron pools that are connected via the corpus callosum. On the contrary, synchronization between the same neural assemblies may be weakened during the imagination of asymmetric (anti-phase) movements, presumably via the recruitment of inhibitory mechanisms [16]. Nine right-handed volunteers (two males and seven females, age range 23–32 years) without a history of
0304-3940/03/$ - see front matter q 2004 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.neulet.2004.01.070
186
O. Levin et al. / Neuroscience Letters 359 (2004) 185–189
neurological or psychiatric diseases participated in the study. All gave their informed consent and the procedure was approved by the local ethical committee. Subjects were seated in a comfortable reclining chair with both the right and left forearms and hands positioned horizontally and stabilized in a semi-prone position. Surface electromyographic (EMG) data were recorded from the right flexor carpi radialis (FCR) and extensor carpi radialis (ECR) muscles by means of disposable disc surface electrodes. The EMG electrodes were placed 2 cm apart, over the middle portion of the muscle belly, and aligned with the longitudinal axis of the muscle. The preamplified signals (gain 80 dB) from each channel were bandpass-filtered (15 Hz to 10 kHz), sampled at 5 kHz (CED Power 1401, Cambridge Electronic Design, Cambridge, UK) and analyzed off-line. Single-pulse transcranial magnetic stimuli were delivered by means of a Dantec MagLite r-25 stimulator (Medtronic, Skovlunde, Denmark) with a figure-of-eight coil (MC-B70 magnetic coil transducer, outer radius diameter: 50 mm). The coil was positioned over the subjects’ left hemisphere, tangentially to the scalp at the optimal position for eliciting MEPs in the right FCR. An ECR response was also evoked in this position. The intersection of the two windings pointed backward and about 458 laterally away from the midsagittal line, optimally orientated to activate the primary motor cortex transsynaptically. The stimulation intensity was set at 120% of the FCR rest motor threshold (rMT). The rMT was defined as the stimulator intensity that is required to elicit an MEP of at least 50 mV in the relaxed FCR in at least five of ten consecutive stimuli. Subjects were instructed to imagine cyclical wrist movements at 1 Hz, in which flexion and extension phases were cued by a dual-tone auditory signal. Subjects were instructed to imagine cyclical flexion and extension movements with: (1) right or (2) left wrist only; (3) simultaneous flexion and extension of both wrists (in-phase); (4) flexion of one wrist with extension of the other (anti-phase); and (5) no movement but listening to the auditory pacing used in tasks 1 –4 (rest). For the unimanual (right/left) and the inphase tasks, subjects were required to imagine flexion upon the onset of a low tone and extension upon the onset of a high tone. For the bimanual anti-phase task, subjects were instructed to imagine flexion of their dominant (right) wrist and extension of their non-dominant (left) wrist upon the onset of the low tone and reverse the course of sequence upon the onset of the high tone. Trials lasted 50 s and were performed with eyes closed. Prior to the TMS sessions, subjects practiced all movement conditions. Initially, subjects performed real movements until they were able to strictly follow the pacing of the auditory cue for the flexion – extension cycles. Then, subjects received the auditory cue and were instructed to produce the required movement for 40 s and then imagine the movement for 10 s after which the required movement pattern was tested. During the successive practice trials the proportion of the trial during which real movements were
performed was gradually reduced. Following the practice session, subjects performed each trained motor imagery task six times with TMS. During practice and TMS sessions, the order of the motor imagery and rest conditions was counterbalanced among subjects. In the TMS session, eight stimuli were randomly delivered at eight different phases of the pacing cycle at the following times: 0 ms (08); 125 ms (458); 250 ms (908); 375 ms (1358); 500 ms (1808); 625 ms (2258); 750 ms (2708); 875 ms (3158). These were synchronized with the onset of the auditory cue for right wrist flexion. The phases were visited in a randomized order. Stimuli were delivered between 10 and 45 s after the onset of data recording and the minimum interval between two consecutive stimuli was 5 s. Throughout the TMS session, the coil was placed at the optimal site for eliciting a response in the right FCR. A specific requirement was that both FCR and ECR remained electrically silent throughout the TMS session. Data from a representative subject are illustrated in Fig. 1. The peak-to-peak amplitude of the TMS evoked MEP was measured for each single response and was expressed as a percentage of the maximum MEP recorded at each of the eight time intervals at the eight different phases of the pacing cycle. Data were then calculated for each of the imagery conditions and averaged across trials. A one-way analysis of variance (ANOVA) was used to test whether MEP amplitude changed as a function of phase. The a-level was set at 0.05. Mean group data of the average sizes of MEPs of the FCR and ECR across the eight tested phase conditions of the four imagery tasks and of the control task (REST) are presented in Fig. 2. For the FCR, the sizes of MEPs at the 458 or 908 phase of the movement cycle were smaller than those at the 2708 and 3158 during imagery of the unilateral right and bimanual in-phase conditions, but significance was not reached: UNIR, F7;56 ¼ 1:65; BMIN, F7;56 ¼ 1:75, both P . 0:05. This tendency was not observed during imagery of the UNIL, BMAN and REST (F7;56 # 1:28). For the ECR the average sizes of MEPs differed significantly among the eight cycle phases during imagery of the unilateral right and the two bimanual tasks: UNIR, F7;56 ¼ 2:21; BMIN, F7;56 ¼ 3:43; BMAN, F7;56 ¼ 2:57, all P , 0:05, but not during imagery of the unilateral left (UNIL) conditions or the control task (REST) (F7;56 , 1). During the UNIR conditions, there was a tendency for the sizes of mean MEPs from the ECR muscle at 908 and 1358 to be larger than those at 2708 and 3158. Post-hoc tests (Tukey, HSD) indicated a marginally significant difference (P ¼ 0:062). On the other hand, mean MEPs from the ECR muscle were significantly higher for the 908 than the 2708 (P , 0:05) and 3158 (P , 0:005) phases of the pacing cycle during motor imagery of the BMIN condition. During the BMAN conditions, mean MEPs from the ECR muscle were significantly higher for the 1358 than for the 3158 phase of the pacing cycle (P , 0:05). The histograms in Fig. 3 depict the distribution of peak
O. Levin et al. / Neuroscience Letters 359 (2004) 185–189
187
generate any phasic modulations in the corticospinal representation of the FCR and ECR. The present findings confirm previous observations that imagination of cyclical movement produces temporal changes in the size of MEPs that largely correspond to the expected reciprocal activity of the wrist muscles during the execution of rhythmical flexion and extension movements [9]. The present observations show that this reciprocal patterning becomes more pronounced when subjects imagine the in-phase movements compared to when they imagine the anti-phase movements. Given that cortical excitability during motor imagery is affected at the level where outputs of specific muscles are selected [9,18], the present results indicate a differential regulation of bimanualrelated activity in the CNS. This could be attributed to increased tuning of bilateral motor networks during mirror symmetry movements, requiring simultaneous activation of
Fig. 1. Results from a single subject (average of six trials). The electromyographic responses were recorded from the right flexor (FCR) and extensor (ECR) carpi radialis muscles during the four imagery tasks and the control task (REST). Reciprocal changes in the amplitude of the response in the FCR and ECR over time are indicated during the unilateralright and the two bimanual conditions.
MEPs among the eight points of the pacing cycle for the four imagery conditions and the control task (REST). Modulations of MEP amplitudes among the eight phase conditions that paralleled the onsets of flexion or extension were observed during motor imagery of the UNIR and BMIN tasks. For the FCR, histograms showed a peak at 3158 (i.e. prior to the onset of the auditory cue for wrist flexion). For the ECR, histograms showed a peak at 908 (i.e. prior to the onset of the auditory cue for wrist extension). As compared to the UNIR condition, imagery of the in-phase task (BMIN) amplified the phase-related reciprocal relationship between FCR and ECR MEPs. In contrast, the timing characteristics during motor imagery of the BMAN and UNIL tasks were less distinct and did not show clear reciprocity between the muscles. There were no indications for cyclical facilitatory effects on the MEPs of both the FCR and ECR during the listening task (REST). Note that the latter condition is not expected to
Fig. 2. Mean and standard deviation (n ¼ 9) of MEP data of the right flexor (FCR) and extensor (ECR) carpi radialis muscles during imagery of unilateral (right and left) and bimanual (in-phase and anti-phase) tasks and of the control task (REST). The amplitude of MEPs is expressed as a percentage of the maximum MEP recorded over the eight cycle phases.
188
O. Levin et al. / Neuroscience Letters 359 (2004) 185–189
movements involves a stronger representation of the temporal features of the task within corticospinal networks. This may be related to the increased synergy between neural structures both within and between hemispheres. In addition, the findings indirectly underscore the key role of egocentric constraints in the performance of bimanual tasks [11,19], suggesting that mirror symmetry movements would be more tightly harnessed into stable flexion and extension patterning than asymmetrical movements. In this respect, the task involving mirror symmetry movements will enhance synchronized activity in corticospinal pathways.
Acknowledgements
Fig. 3. Histogram showing the distribution of the phase offsets for which peak MEP was observed. Data were assembled across trials (n ¼ 54) with the same task conditions. The offsets are determined with respect to the onset of the first auditory cue (at 08) for flexion of the right limb. Lines were fitted to the points of the scatter plots using the MATLAB cubic spline interpolation procedure.
homologous muscle pairs, whereas suppression occurs when those muscles are activated unilaterally [4,5] or discretely [1,16]. The reduction in phasic representation of the flexion and extension cycles during motor imagery of the anti-phase task is possibly associated with interhemispheric inhibition of interconnected neural networks that normally promote synchronization among homologous effectors. As such, successful desynchronization during production of dissimilar movements with homologous effectors might depend on recruitment of inhibitory networks [16,19]. In other words, interhemispheric inhibition may be essential for performing less familiar coordination patterns. This may require the increased involvement of other brain areas. The observation that SMA, PMC and cingulate motor cortex become increasingly important as more difficult spatiotemporal relations between simultaneous limb movements are explored may help to draw the boundaries of this distributed neural network [8,17,20]. Temporal changes in the size of MEPs during imagery of bimanual movements are consistent with the observation that interhemispheric interactions occur even when a movement is not actually executed. The overall picture would then be that interhemispheric interactions affect both movement planning and movement execution. Indeed, for imagery of the in-phase task, histograms of both FCR and ECR peak MEPs showed a 908 phase lead over the auditory cue for flexion and extension (Fig. 3). This raises the possibility that the degree of interhemispheric interactions during mirror-symmetry movements coincides not only with the process of movement execution but also with movement preparation. In summary, the present findings revealed that imagery of bimanual in-phase as compared to anti-phase or unilateral
Support for the present study was provided through a grant from the Research Council of K.U. Leuven, Belgium (contract OT/03/61) and the Flanders Fund for Scientific Research (Projects G.0285.98, G.0105.00 & G.0460.04).
References [1] S. Cardoso de Oliveira, A. Gribova, O. Donchin, H. Bergman, E. Vaadia, Neural interactions between motor cortical hemispheres during bimanual and unimanual arm movements, Eur. J. Neurosci. 14 (2001) 1881–1896. [2] J. Decety, The neurophysiological basis of motor imagery, Behav. Brain Res. 77 (1996) 45–52. [3] J. Decety, D. Perani, M. Jeannerod, V. Bettinardi, B. Tadary, R. Woods, J.C. Mazziotta, F. Fazio, Mapping motor representation with PET, Nature 371 (1994) 600 –602. [4] O. Donchin, A. Gribova, O. Steinberg, H. Bergman, E. Vaadia, Primary motor cortex is involved in bimanual coordination, Nature 395 (1998) 274–278. [5] O. Donchin, A. Gribova, O. Steinberg, H. Bergman, S. Cardoso de Oliveira, E. Vaadia, Local field potentials related to bimanual movements in the primary and supplementary motor cortices, Exp. Brain Res. 140 (2001) 46–55. [6] S. Facchini, W. Muellbacher, F. Battaglia, B. Boroojerdi, M. Hallett, Focal enhancement of motor cortex excitability during motor imagery: transcranial magnetic stimulation study, Acta Neurol. Scand. 105 (2002) 146–151. [7] L. Fadiga, G. Buccino, L. Craighero, L. Fogassi, V. Gallese, G. Pavesi, Corticospinal excitability is specifically modulated by motor imagery: a magnetic stimulation study, Neuropsychologia 37 (1999) 147 –158. [8] G.W. Goerres, M. Samuel, I.H. Jenkins, D.J. Brooks, Cerebral control of unimanual and bimanual movements: an H15 2 O PET study, NeuroReport 9 (1998) 3631–3638. [9] R. Hashimoto, J.C. Rothwell, Dynamic changes in corticospinal excitability during motor imagery, Exp. Brain Res. 125 (1999) 75 –81. [10] T. Kasai, S. Kawai, M. Kawanishi, S. Yahagi, Evidence for facilitation of motor evoked potentials (MEPs) induced by motor imagery, Brain Res. 744 (1997) 147 –150. [11] J.A.S. Kelso, J.J. Jeka, Symmetry breaking dynamics of human interlimb coordination, J. Exp. Psychol. Hum. Percept. Perform. 18 (1992) 645 –668. [12] L. Kiers, B. Fernando, D. Tomkins, Facilitatory effects of thinking about movement on magnetic motor-evoked potentials, Electroenceph. clin. Neurophysiol. 105 (1997) 262 –268. [13] M. Lotze, P. Montoya, M. Erb, E. Hulsmann, H. Flor, U. Klose, N.
O. Levin et al. / Neuroscience Letters 359 (2004) 185–189 Birbaumer, W. Grodd, Activation of cortical and cerebral motor areas during executed and imagined hand movements: an fMRI study, J. Cogn. Neurosci. 11 (1999) 491– 501. [14] D.G. Nair, K.L. Purcott, A. Fuchs, F. Steinberg, J.A.S. Kelso, Cortical and cerebral activity of human brain during imagined and executed unimanual and bimanual action sequences: a functional MRI study, Cogn. Brain Res. 15 (2003) 250– 260. [15] C.A. Porro, M.P. Francescato, V. Cettolo, M.E. Diamond, P. Baraldi, C. Zuiani, M. Bazzocchi, P.E. di Prampero, Primary motor and sensory cortex activation during motor performance and motor imagery: a functional magnetic resonance imaging study, J. Neurosci. 16 (2000) 7688–7698. [16] U. Rokni, O. Steinberg, E. Vaadia, H. Sompolinsky, Cortical representation of bimanual movements, J. Neurosci. 23 (2003) 11577–11586.
189
[17] N. Sadato, Y. Yonekura, A. Waki, H. Yamada, Y. Ishii, Role of the supplementary motor area and the right premotor cortex in coordination of bimanual finger movements, J. Neurosci. 15 (1997) 9667– 9674. [18] C.M. Stinear, W.D. Byblow, Motor imagery of phasic thumb abduction temporally and spatially modulates corticospinal excitability, Clin. Neurophysiol. 114 (2003) 909 –914. [19] S.P. Swinnen, Intermanual coordination: from behavioural principles to neural-network interactions, Nat. Rev. Neurosci. 3 (2002) 348–359. [20] M. Toyokura, I. Muro, T. Komiya, M. Obara, Relation of bimanual coordination to activation in the sensorimotor cortex and supplementary motor area: analysis using functional magnetic resonance imaging, Brain Res. Bull. 48 (1999) 211–217.
Dynamical changes in corticospinal excitability during imagery of unimanual and bimanual wrist movements in humans: a transcranial magnetic stimulation study O. Levin*, M. Steyvers, N. Wenderoth, Y. Li, S.P. Swinnen Motor Control Laboratory, Department of Kinesiology, Group Biomedical Sciences, Katholieke Universiteit Leuven, Tervuurse Vest 101, 3001 Heverlee, Belgium Received 7 November 2003; received in revised form 29 January 2004; accepted 29 January 2004
Abstract This study explored the dynamical changes in corticospinal excitability during the imagination of cyclical unimanual and bimanual wrist flexion –extension movements. Transcranial magnetic stimulation was applied over the left motor cortex to evoke motor evoked potentials in the right wrist flexor and extensor muscles. Findings provided evidence for increased reciprocal excitability changes during imagery of symmetrical in-phase movements as compared to asymmetrical (anti-phase) or unimanual movements. This suggests that in-phase movements may reinforce whereas anti-phase movements may reduce the temporal representation of the task in the corticospinal motor networks of the brain. q 2004 Elsevier Ireland Ltd. All rights reserved. Keywords: Bimanual coordination; Motor imagery; Transcranial magnetic stimulation; Temporal representation; Corticospinal excitability
Functional imaging studies have revealed that motor imagery activates large parts of the central nervous system (CNS) motor networks such as the cerebellum, premotor cortex (PMC), supplementary motor area (SMA) and the primary motor cortex (M1) [2,3,13 –15]. While SMA and PMC are equally activated during both actual and imagined movements, the levels of activation in M1 during imagery have been found lower than those observed during actual execution of the same movements [13,15]. Complementary studies using transcranial magnetic stimulation (TMS) have demonstrated an increase in M1 excitability during imagery of a wide range of motor tasks [6,7,9,10,12, 18]. Additionally, motor evoked potential (MEP) facilitation in the targeted muscles has been shown to result mainly from increased cortical excitability and not from spinal motorneural excitability [9,10,12]. For unimanual movements, Hashimoto and Rothwell [9] demonstrated that imagination of cyclical flexion–extension movements produces temporal changes in the size of MEPs similar to those expected during actual performance of the task. * Corresponding author. Tel.: þ 32-16-329065; fax: þ 32-16-329197. E-mail address: [email protected] (O. Levin).
Unimanual and bimanual tasks employ largely overlapping neural resources [17]. With respect to bimanual coordination, brain activation levels appear to increase with task complexity, e.g. coordination of limb movements in the anti-phase mode is associated with higher activation levels in the SMA, PMC and cingulate motor area as compared to the in-phase mode [8,17, 20]. The engagement of SMA in imagining and performing coordinated movements suggests that it is likely involved in planning and preparation of action sequences, whether imagined or real [14]. Here, we address the question whether imagining cyclical bimanual vs. unimanual movements has differential effects on the temporal patterns of corticospinal excitability as measured by TMS. More specifically, we assume that the imagination of symmetric (in-phase) movements might result in stronger synchronization of the activity of bilateral motorneuron pools that are connected via the corpus callosum. On the contrary, synchronization between the same neural assemblies may be weakened during the imagination of asymmetric (anti-phase) movements, presumably via the recruitment of inhibitory mechanisms [16]. Nine right-handed volunteers (two males and seven females, age range 23–32 years) without a history of
0304-3940/03/$ - see front matter q 2004 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.neulet.2004.01.070
186
O. Levin et al. / Neuroscience Letters 359 (2004) 185–189
neurological or psychiatric diseases participated in the study. All gave their informed consent and the procedure was approved by the local ethical committee. Subjects were seated in a comfortable reclining chair with both the right and left forearms and hands positioned horizontally and stabilized in a semi-prone position. Surface electromyographic (EMG) data were recorded from the right flexor carpi radialis (FCR) and extensor carpi radialis (ECR) muscles by means of disposable disc surface electrodes. The EMG electrodes were placed 2 cm apart, over the middle portion of the muscle belly, and aligned with the longitudinal axis of the muscle. The preamplified signals (gain 80 dB) from each channel were bandpass-filtered (15 Hz to 10 kHz), sampled at 5 kHz (CED Power 1401, Cambridge Electronic Design, Cambridge, UK) and analyzed off-line. Single-pulse transcranial magnetic stimuli were delivered by means of a Dantec MagLite r-25 stimulator (Medtronic, Skovlunde, Denmark) with a figure-of-eight coil (MC-B70 magnetic coil transducer, outer radius diameter: 50 mm). The coil was positioned over the subjects’ left hemisphere, tangentially to the scalp at the optimal position for eliciting MEPs in the right FCR. An ECR response was also evoked in this position. The intersection of the two windings pointed backward and about 458 laterally away from the midsagittal line, optimally orientated to activate the primary motor cortex transsynaptically. The stimulation intensity was set at 120% of the FCR rest motor threshold (rMT). The rMT was defined as the stimulator intensity that is required to elicit an MEP of at least 50 mV in the relaxed FCR in at least five of ten consecutive stimuli. Subjects were instructed to imagine cyclical wrist movements at 1 Hz, in which flexion and extension phases were cued by a dual-tone auditory signal. Subjects were instructed to imagine cyclical flexion and extension movements with: (1) right or (2) left wrist only; (3) simultaneous flexion and extension of both wrists (in-phase); (4) flexion of one wrist with extension of the other (anti-phase); and (5) no movement but listening to the auditory pacing used in tasks 1 –4 (rest). For the unimanual (right/left) and the inphase tasks, subjects were required to imagine flexion upon the onset of a low tone and extension upon the onset of a high tone. For the bimanual anti-phase task, subjects were instructed to imagine flexion of their dominant (right) wrist and extension of their non-dominant (left) wrist upon the onset of the low tone and reverse the course of sequence upon the onset of the high tone. Trials lasted 50 s and were performed with eyes closed. Prior to the TMS sessions, subjects practiced all movement conditions. Initially, subjects performed real movements until they were able to strictly follow the pacing of the auditory cue for the flexion – extension cycles. Then, subjects received the auditory cue and were instructed to produce the required movement for 40 s and then imagine the movement for 10 s after which the required movement pattern was tested. During the successive practice trials the proportion of the trial during which real movements were
performed was gradually reduced. Following the practice session, subjects performed each trained motor imagery task six times with TMS. During practice and TMS sessions, the order of the motor imagery and rest conditions was counterbalanced among subjects. In the TMS session, eight stimuli were randomly delivered at eight different phases of the pacing cycle at the following times: 0 ms (08); 125 ms (458); 250 ms (908); 375 ms (1358); 500 ms (1808); 625 ms (2258); 750 ms (2708); 875 ms (3158). These were synchronized with the onset of the auditory cue for right wrist flexion. The phases were visited in a randomized order. Stimuli were delivered between 10 and 45 s after the onset of data recording and the minimum interval between two consecutive stimuli was 5 s. Throughout the TMS session, the coil was placed at the optimal site for eliciting a response in the right FCR. A specific requirement was that both FCR and ECR remained electrically silent throughout the TMS session. Data from a representative subject are illustrated in Fig. 1. The peak-to-peak amplitude of the TMS evoked MEP was measured for each single response and was expressed as a percentage of the maximum MEP recorded at each of the eight time intervals at the eight different phases of the pacing cycle. Data were then calculated for each of the imagery conditions and averaged across trials. A one-way analysis of variance (ANOVA) was used to test whether MEP amplitude changed as a function of phase. The a-level was set at 0.05. Mean group data of the average sizes of MEPs of the FCR and ECR across the eight tested phase conditions of the four imagery tasks and of the control task (REST) are presented in Fig. 2. For the FCR, the sizes of MEPs at the 458 or 908 phase of the movement cycle were smaller than those at the 2708 and 3158 during imagery of the unilateral right and bimanual in-phase conditions, but significance was not reached: UNIR, F7;56 ¼ 1:65; BMIN, F7;56 ¼ 1:75, both P . 0:05. This tendency was not observed during imagery of the UNIL, BMAN and REST (F7;56 # 1:28). For the ECR the average sizes of MEPs differed significantly among the eight cycle phases during imagery of the unilateral right and the two bimanual tasks: UNIR, F7;56 ¼ 2:21; BMIN, F7;56 ¼ 3:43; BMAN, F7;56 ¼ 2:57, all P , 0:05, but not during imagery of the unilateral left (UNIL) conditions or the control task (REST) (F7;56 , 1). During the UNIR conditions, there was a tendency for the sizes of mean MEPs from the ECR muscle at 908 and 1358 to be larger than those at 2708 and 3158. Post-hoc tests (Tukey, HSD) indicated a marginally significant difference (P ¼ 0:062). On the other hand, mean MEPs from the ECR muscle were significantly higher for the 908 than the 2708 (P , 0:05) and 3158 (P , 0:005) phases of the pacing cycle during motor imagery of the BMIN condition. During the BMAN conditions, mean MEPs from the ECR muscle were significantly higher for the 1358 than for the 3158 phase of the pacing cycle (P , 0:05). The histograms in Fig. 3 depict the distribution of peak
O. Levin et al. / Neuroscience Letters 359 (2004) 185–189
187
generate any phasic modulations in the corticospinal representation of the FCR and ECR. The present findings confirm previous observations that imagination of cyclical movement produces temporal changes in the size of MEPs that largely correspond to the expected reciprocal activity of the wrist muscles during the execution of rhythmical flexion and extension movements [9]. The present observations show that this reciprocal patterning becomes more pronounced when subjects imagine the in-phase movements compared to when they imagine the anti-phase movements. Given that cortical excitability during motor imagery is affected at the level where outputs of specific muscles are selected [9,18], the present results indicate a differential regulation of bimanualrelated activity in the CNS. This could be attributed to increased tuning of bilateral motor networks during mirror symmetry movements, requiring simultaneous activation of
Fig. 1. Results from a single subject (average of six trials). The electromyographic responses were recorded from the right flexor (FCR) and extensor (ECR) carpi radialis muscles during the four imagery tasks and the control task (REST). Reciprocal changes in the amplitude of the response in the FCR and ECR over time are indicated during the unilateralright and the two bimanual conditions.
MEPs among the eight points of the pacing cycle for the four imagery conditions and the control task (REST). Modulations of MEP amplitudes among the eight phase conditions that paralleled the onsets of flexion or extension were observed during motor imagery of the UNIR and BMIN tasks. For the FCR, histograms showed a peak at 3158 (i.e. prior to the onset of the auditory cue for wrist flexion). For the ECR, histograms showed a peak at 908 (i.e. prior to the onset of the auditory cue for wrist extension). As compared to the UNIR condition, imagery of the in-phase task (BMIN) amplified the phase-related reciprocal relationship between FCR and ECR MEPs. In contrast, the timing characteristics during motor imagery of the BMAN and UNIL tasks were less distinct and did not show clear reciprocity between the muscles. There were no indications for cyclical facilitatory effects on the MEPs of both the FCR and ECR during the listening task (REST). Note that the latter condition is not expected to
Fig. 2. Mean and standard deviation (n ¼ 9) of MEP data of the right flexor (FCR) and extensor (ECR) carpi radialis muscles during imagery of unilateral (right and left) and bimanual (in-phase and anti-phase) tasks and of the control task (REST). The amplitude of MEPs is expressed as a percentage of the maximum MEP recorded over the eight cycle phases.
188
O. Levin et al. / Neuroscience Letters 359 (2004) 185–189
movements involves a stronger representation of the temporal features of the task within corticospinal networks. This may be related to the increased synergy between neural structures both within and between hemispheres. In addition, the findings indirectly underscore the key role of egocentric constraints in the performance of bimanual tasks [11,19], suggesting that mirror symmetry movements would be more tightly harnessed into stable flexion and extension patterning than asymmetrical movements. In this respect, the task involving mirror symmetry movements will enhance synchronized activity in corticospinal pathways.
Acknowledgements
Fig. 3. Histogram showing the distribution of the phase offsets for which peak MEP was observed. Data were assembled across trials (n ¼ 54) with the same task conditions. The offsets are determined with respect to the onset of the first auditory cue (at 08) for flexion of the right limb. Lines were fitted to the points of the scatter plots using the MATLAB cubic spline interpolation procedure.
homologous muscle pairs, whereas suppression occurs when those muscles are activated unilaterally [4,5] or discretely [1,16]. The reduction in phasic representation of the flexion and extension cycles during motor imagery of the anti-phase task is possibly associated with interhemispheric inhibition of interconnected neural networks that normally promote synchronization among homologous effectors. As such, successful desynchronization during production of dissimilar movements with homologous effectors might depend on recruitment of inhibitory networks [16,19]. In other words, interhemispheric inhibition may be essential for performing less familiar coordination patterns. This may require the increased involvement of other brain areas. The observation that SMA, PMC and cingulate motor cortex become increasingly important as more difficult spatiotemporal relations between simultaneous limb movements are explored may help to draw the boundaries of this distributed neural network [8,17,20]. Temporal changes in the size of MEPs during imagery of bimanual movements are consistent with the observation that interhemispheric interactions occur even when a movement is not actually executed. The overall picture would then be that interhemispheric interactions affect both movement planning and movement execution. Indeed, for imagery of the in-phase task, histograms of both FCR and ECR peak MEPs showed a 908 phase lead over the auditory cue for flexion and extension (Fig. 3). This raises the possibility that the degree of interhemispheric interactions during mirror-symmetry movements coincides not only with the process of movement execution but also with movement preparation. In summary, the present findings revealed that imagery of bimanual in-phase as compared to anti-phase or unilateral
Support for the present study was provided through a grant from the Research Council of K.U. Leuven, Belgium (contract OT/03/61) and the Flanders Fund for Scientific Research (Projects G.0285.98, G.0105.00 & G.0460.04).
References [1] S. Cardoso de Oliveira, A. Gribova, O. Donchin, H. Bergman, E. Vaadia, Neural interactions between motor cortical hemispheres during bimanual and unimanual arm movements, Eur. J. Neurosci. 14 (2001) 1881–1896. [2] J. Decety, The neurophysiological basis of motor imagery, Behav. Brain Res. 77 (1996) 45–52. [3] J. Decety, D. Perani, M. Jeannerod, V. Bettinardi, B. Tadary, R. Woods, J.C. Mazziotta, F. Fazio, Mapping motor representation with PET, Nature 371 (1994) 600 –602. [4] O. Donchin, A. Gribova, O. Steinberg, H. Bergman, E. Vaadia, Primary motor cortex is involved in bimanual coordination, Nature 395 (1998) 274–278. [5] O. Donchin, A. Gribova, O. Steinberg, H. Bergman, S. Cardoso de Oliveira, E. Vaadia, Local field potentials related to bimanual movements in the primary and supplementary motor cortices, Exp. Brain Res. 140 (2001) 46–55. [6] S. Facchini, W. Muellbacher, F. Battaglia, B. Boroojerdi, M. Hallett, Focal enhancement of motor cortex excitability during motor imagery: transcranial magnetic stimulation study, Acta Neurol. Scand. 105 (2002) 146–151. [7] L. Fadiga, G. Buccino, L. Craighero, L. Fogassi, V. Gallese, G. Pavesi, Corticospinal excitability is specifically modulated by motor imagery: a magnetic stimulation study, Neuropsychologia 37 (1999) 147 –158. [8] G.W. Goerres, M. Samuel, I.H. Jenkins, D.J. Brooks, Cerebral control of unimanual and bimanual movements: an H15 2 O PET study, NeuroReport 9 (1998) 3631–3638. [9] R. Hashimoto, J.C. Rothwell, Dynamic changes in corticospinal excitability during motor imagery, Exp. Brain Res. 125 (1999) 75 –81. [10] T. Kasai, S. Kawai, M. Kawanishi, S. Yahagi, Evidence for facilitation of motor evoked potentials (MEPs) induced by motor imagery, Brain Res. 744 (1997) 147 –150. [11] J.A.S. Kelso, J.J. Jeka, Symmetry breaking dynamics of human interlimb coordination, J. Exp. Psychol. Hum. Percept. Perform. 18 (1992) 645 –668. [12] L. Kiers, B. Fernando, D. Tomkins, Facilitatory effects of thinking about movement on magnetic motor-evoked potentials, Electroenceph. clin. Neurophysiol. 105 (1997) 262 –268. [13] M. Lotze, P. Montoya, M. Erb, E. Hulsmann, H. Flor, U. Klose, N.
O. Levin et al. / Neuroscience Letters 359 (2004) 185–189 Birbaumer, W. Grodd, Activation of cortical and cerebral motor areas during executed and imagined hand movements: an fMRI study, J. Cogn. Neurosci. 11 (1999) 491– 501. [14] D.G. Nair, K.L. Purcott, A. Fuchs, F. Steinberg, J.A.S. Kelso, Cortical and cerebral activity of human brain during imagined and executed unimanual and bimanual action sequences: a functional MRI study, Cogn. Brain Res. 15 (2003) 250– 260. [15] C.A. Porro, M.P. Francescato, V. Cettolo, M.E. Diamond, P. Baraldi, C. Zuiani, M. Bazzocchi, P.E. di Prampero, Primary motor and sensory cortex activation during motor performance and motor imagery: a functional magnetic resonance imaging study, J. Neurosci. 16 (2000) 7688–7698. [16] U. Rokni, O. Steinberg, E. Vaadia, H. Sompolinsky, Cortical representation of bimanual movements, J. Neurosci. 23 (2003) 11577–11586.
189
[17] N. Sadato, Y. Yonekura, A. Waki, H. Yamada, Y. Ishii, Role of the supplementary motor area and the right premotor cortex in coordination of bimanual finger movements, J. Neurosci. 15 (1997) 9667– 9674. [18] C.M. Stinear, W.D. Byblow, Motor imagery of phasic thumb abduction temporally and spatially modulates corticospinal excitability, Clin. Neurophysiol. 114 (2003) 909 –914. [19] S.P. Swinnen, Intermanual coordination: from behavioural principles to neural-network interactions, Nat. Rev. Neurosci. 3 (2002) 348–359. [20] M. Toyokura, I. Muro, T. Komiya, M. Obara, Relation of bimanual coordination to activation in the sensorimotor cortex and supplementary motor area: analysis using functional magnetic resonance imaging, Brain Res. Bull. 48 (1999) 211–217.