UBC Faculty Research and Publications

Excitatory repetitive transcranial magnetic stimulation to left dorsal premotor cortex enhances motor… Boyd, Lara A; Linsdell, Meghan A Jul 7, 2009

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata


52383-12868_2009_Article_1076.pdf [ 1010.09kB ]
JSON: 52383-1.0074676.json
JSON-LD: 52383-1.0074676-ld.json
RDF/XML (Pretty): 52383-1.0074676-rdf.xml
RDF/JSON: 52383-1.0074676-rdf.json
Turtle: 52383-1.0074676-turtle.txt
N-Triples: 52383-1.0074676-rdf-ntriples.txt
Original Record: 52383-1.0074676-source.json
Full Text

Full Text

ralssBioMed CentBMC NeuroscienceOpen AcceResearch articleExcitatory repetitive transcranial magnetic stimulation to left dorsal premotor cortex enhances motor consolidation of new skillsLara A Boyd*1,2,3 and Meghan A Linsdell3Address: 1Department of Physical Therapy, University of British Columbia, Vancouver, Canada, 2Brain Research Centre, University of British Columbia, Vancouver, Canada and 3Graduate Program in Rehabilitation Sciences, University of British Columbia, Vancouver, CanadaEmail: Lara A Boyd* - lara.boyd@ubc.ca; Meghan A Linsdell - malinsde@interchange.ubc.ca* Corresponding author    AbstractBackground: Following practice of skilled movements, changes continue to take place in the brainthat both strengthen and modify memory for motor learning. These changes represent motormemory consolidation a process whereby new memories are transformed from a fragile to a morepermanent, robust and stable state. In the present study, the neural correlates of motor memoryconsolidation were probed using repetitive transcranial magnetic stimulation (rTMS) to the dorsalpremotor cortex (PMd). Participants engaged in four days of continuous tracking practice thatimmediately followed either excitatory 5 HZ, inhibitory 1 HZ or control, sham rTMS. A delayedretention test assessed motor learning of repeated and random sequences of continuousmovement; no rTMS was applied at retention.Results: We discovered that 5 HZ excitatory rTMS to PMd stimulated motor memoryconsolidation as evidenced by off-line learning, whereas only memory stabilization was notedfollowing 1 Hz inhibitory or sham stimulation.Conclusion: Our data support the hypothesis that PMd is important for continuous motorlearning, specifically via off-line consolidation of learned motor behaviors.BackgroundIt is clear that skilled practice is essential for the acquisi-tion of learned motor behaviors [1-3] and that the braincontinues to process information from practice sessionswell beyond the timeframe of motor performance [4-7].In fact, many changes take place after practice that bothstrengthen and modify the motor skill being learned.These changes represent motor memory consolidation [5-7] a process whereby new, fragile memories are trans-formed into more permanent, robust and stable state.behavior that occur in between practice sessions, and 2)memory stabilization which reduces the fragility or sus-ceptibility to interference by other motor actions whilebehavioral improvements are maintained [6-9]. Whilethese two elements of motor consolidation are not com-pletely independent of one another, the degree to whichthey interact and/or rely of unique neural structuresremains unclear.Functional brain imaging has been used to consider howthe neural structures associated with movement change asPublished: 7 July 2009BMC Neuroscience 2009, 10:72 doi:10.1186/1471-2202-10-72Received: 25 February 2009Accepted: 7 July 2009This article is available from: http://www.biomedcentral.com/1471-2202/10/72© 2009 Boyd and Linsdell; licensee BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.Page 1 of 9(page number not for citation purposes)Consolidation of motor skill memories has been pur-ported to take two forms: 1) off-line improvements inmotor learning occurs [5,10,11]. Following practice,while motor ability remains unchanged, positron emis-BMC Neuroscience 2009, 10:72 http://www.biomedcentral.com/1471-2202/10/72sion tomography shows that the brain recruits newregions to perform the task. Early in motor skill acquisi-tion prefrontal brain regions are active. Later, there is ashift in activation to premotor, posterior parietal and cer-ebellar structures [5]. Evolution of the network activatedin association with motor learning is widely believed tosupport motor consolidation or the increase in stability ofthe new skill [5-7,11].Intracortical recordings in animals and human neuroim-aging studies indicate that the premotor cortex (PMC)plays an important role in the selection of movements[12,13]. The PMC can be functionally segregated accord-ing to the type of movement being selected [14-18]. Theventral premotor (PMv) cortex is involved in graspingmovements that are externally triggered by the environ-ment [12], while the dorsal premotor (PMd) cortexappears to be particularly important in the selectionmovements that are learned [12,19].At present it is unclear whether PMd is important forselecting learned movements [20] or for learning newmovements [21,22]. Some animal work suggests that PMdis involved in motor learning. The PMd cortex may be crit-ical for holding sensory information in working memoryand then converting it to a motor program [22]. Single cellrecordings from PMd during motor task practice demon-strate the emergence of new motor programs that arebased on the sensory information acquired through prac-tice [22]. Other research in humans, suggests a dominanceof left PMd for the selection of learned movements[12,19]. Functional magnetic resonance imaging (MRI)studies show that only right PMd is active during move-ments of the left hand; however, left PMd is activated dur-ing movements of both the right and left hands [19]. Thedisparate nature of PMd activity during movements ofeach hand has also been confirmed by transcranial mag-netic stimulation (TMS) studies demonstrating that dis-ruption of left PMd alters movements in both hands [21].Though past work has purported to investigate the role ofPMd in skill acquisition it has often been limited by thefailure to consider motor learning at a separate delayedretention test [7,23] but rather has largely consideredbehavioral changes across a single day or session. Withoutdata from separate sessions it is impossible to evaluate theimpact of any intervention of the long-term, permanentability to perform new motor skills [24]. Because we wereinterested in the possibility that PMd might play a role inconsolidation of new motor learning we designed thepresent experiment to contain practice sessions and adelayed retention test, which were all conducted on sepa-rate days.pairing the delivery of an epoch of excitatory (5 Hz) repet-itive TMS (rTMS) with movement task practice. To verifythe effects of excitatory rTMS, we also trained a group ofindividuals who only received inhibitory (1 Hz) rTMS andanother cohort who received sham stimulation. Becauseof the known role of PMd in motor learning and its pur-ported role in the stabilization of newly acquired skills,we hypothesized that excitatory rTMS to PMd would facil-itate motor skill consolidation.MethodsParticipantsThirty-two healthy, neurologically intact individuals aged20 to 38 (14 men and 18 women) enrolled in the experi-ment (Table 1). All participants gave written informedconsent and the protocol was approved by the Universityresearch ethics board. Two participants were unable tocomplete the testing as a result of discomfort during initialmotor thresholding with TMS. All participants reported tobe right hand dominant; all received left sided rTMS. Par-ticipants were not enrolled if 1) they exhibited any frankor clinically evident signs of neurological impairment ordisease [25], or 2) they had any color blindness thatwould impair response ability. Participants were recruitedfrom the University, the local community and the labdatabase.Behavioural taskParticipants were seated in front of a computer monitorand engaged in continuous tracking of a target moving ina sine-cosine waveform by manipulating a joystick usingtheir right arm [26-29]. The target appeared as a white cir-cle and participant movements were represented as a reddot (Figure 1A). Joystick position sampling and all stimuliwere presented at 40 Hz using custom software developedon the LabView platform (v. 7.1; National InstrumentsCo.).The pattern of target movement was predefined accordingto a method modified from Wulf and Schmidt.[26] Aunique 33s trial was constructed from one 3s baseline andthree 10s sine-cosine segments. One block consisted often 33 second trials. Unknown to the participants, themiddle third of each tracking pattern was repeated andidentical across practice and retention. This pattern wasconstructed using the polynomial equation as describedby Wulf and Schmidt (1997) with the following generalform:Table 1: Subject characteristics for the Excitatory, Inhibitory, and Sham groupsExcite Inhibit ShamPage 2 of 9(page number not for citation purposes)To investigate the role of PMd in motor skill learning wecoupled brain stimulation with motor-skill acquisition,Age (years ± sd) 24 ± 2 24 ± 1 27 ± 7Gender 6 M, 4 F 3 M, 7 F 5 M, 5 FBMC Neuroscience 2009, 10:72 http://www.biomedcentral.com/1471-2202/10/72The middle (repeated) segment was constructed by usingthe same coefficients for every trial (Appendix 1). The firstand third segments of the tracking pattern were generatedrandomly using coefficients ranging from 5.0 to -5.0. Adifferent random sequence was used for both the first andthird segments for every trial (Figure 1B); however, toensure uniformity across participants the same set of trialswere practiced by all of the participants so that on anygiven trial the random segments were the same for eachparticipant. In each third of the tracking pattern therewere 10 separate reversals in the direction. The trajectoriesof the target and participants' movements did not leave atrail and thus, participants could not visualize the entiretarget pattern.The same trial order was employed for every participant.Participants were not informed of the existence of therepeating sequence but instructed daily to track the targetas accurately as possible by controlling the position of thecursor with the joystick.TMSApplication of TMS was performed with a 70 mm figure-8 air-cooled coil (Magstim Super Rapid2, Magstim Com-pany, Ltd.). The magnetic stimulus had a biphasic wave-form with a pulse width of 400 us. During stimulation ofboth M1 for thresholding and PMd for repetitive stimula-45 degrees. Prior to the start of the experiment each partic-ipant underwent an anatomical MRI scan on a separateday at the UBC 3T MRI Centre (T1 images TE = 5 ms, TR =24 ms, 40° flip angle, NEX = 1, thickness = 1.2 mm, FOV= 256 mm). These images were imported into Brainsight™TMS neuronavigation software (Rogue Research Inc.) toallow for stereotaxic registration of the participant's brainwith TMS coil for online control of coil positioning.Participants were instructed to remain relaxed throughoutthe application of rTMS. Surface electromyography(EMG) from participants' right flexor digitorum musclewas monitored through the output screen attached to thetranscranial magnetic stimulator (Magstim Super Rapid2,Magstim Company, Ltd.). Determination of the locationof left primary motor cortex (M1) for resting motorthreshold was performed using Brainsight. M1 was identi-fied using the axial scans by locating the "hand knob" andhook MRI images.[30-34] Resting motor threshold (RMT)was determined for each participant, as the percentage ofmaximal stimulator output to evoke a response of ³ 50 mVin 5 of 10 trials. The location and trajectory of the coil forthis spot was marked using Brainsight™ to minimize vari-ability across subsequent trials and days (Figure 2). Next,the left dorsal pre-motor (PMd) area was marked in Brain-sight™ by moving one gyrus forward from the flexor digi-torum "hot spot" identified during determination of RMT.The location of PMd was confirmed as the posterior aspectof the middle frontal gyrus (Figure 1).[21,33-38]Several steps were taken to ensure that stimulation of PMdwithout M1 during rTMS. First we used a coil that has pre-f( ) sin( ) cos( ) sin( )cos( ) sin( )x b a x b x a xb x a xo= + + ++ + ++1 1 22 622 6Kb x6 6cos( )Illustration of the behavioral taskFigu e 1Illustration of the behavioral task. A) Continuous tracking of sequences was performed using a joystick; participants were instructed to manipulate the joystick in the vertical direction to track the target (open circle) as accurately as possible. Partici-pant movements appeared as a red, closed circle. B) Continuous tracking trials were constructed from three individual sequences that were joined seamlessly to form one trace. Unknown to participants, the first and last third of each trial (10 s each) were random. The middle third was repeated on every tracking trial. Unlike this illustration, no trace or trail from move-ment was evident during tacking, only the target and current position of the participant's cursor were visible.Page 3 of 9(page number not for citation purposes)tion the TMS coil was oriented tangentially to the scalpwith the handle pointing back and away from midline atviously been shown to have a focal enough output tostimulate PMd in isolation. Application of TMS was per-BMC Neuroscience 2009, 10:72 http://www.biomedcentral.com/1471-2202/10/72formed with a 70 mm figure-8 air-cooled coil (MagstimSuper Rapid2, Magstim Company, Ltd.). Past work hasdemonstrated that a 70 mm coil can deliver focal stimula-tion [39] with a current spread small enough, 10 × 10 × 20mm [31], to stimulate M1 without PMd and viceversa.[40] Second, we oriented our coil over the PMdusing anatomic landmarks shown via each individual's T1MRI to guide us to the posterior aspect of the middle fron-tal gyrus.[21,35] Once confirmed, the location of PMdand the direction of our stimulation were maintainedboth within and across sessions by trajectory targetingusing BrainSight.Participants were randomly assigned to one of threegroups: 5 Hz excitatory rTMS stimulation (Excite group),1 Hz inhibitory rTMS stimulation (Inhibit group), or 5 Hzsham stimulation (Sham group). Sham stimulation wasperformed using a custom sham coil that looks andsounds like an active coil but does not induce any currentin the underlying cortex (Magstim Company Ltd.). Allparticipants were naive to TMS measures and wereblinded to group assignment. rTMS was performed overthe marked spot for left PMd for 15 minutes at 120% ofRMT.[21,41] If stimulation at this level caused any visiblemotor activation intensity was decreased in 5% incre-ments of RMT until there was no longer any motorresponse. rTMS stimulation intensity over PMd wasdecreased to eliminate motor response in 16 of 20 partic-ipants who received active stimulation: 10 in the Excitegroup and 6 in the Inhibit group. Across participants rTMSBecause a 15 minute bout of rTMS has been shown toinduce approximately a 15-minute after-effect,[42] indi-viduals underwent rTMS first, then immediately practicedthe motor task on each of the four practice days. Thisstructure was identical regardless of the rTMS groupassignment.Design and proceduresThe experiment lasted for five days spread over a 2-weektimeframe. Days 1–4 were training (rTMS paired withmotor task practice). In each of these days participantsperformed 3 blocks (30 trials) of tracking. Tracking wasperformed immediately after application of rTMS. Partici-pants were given 2 weeks to complete the entire 5-dayexperiment, but no more than one day lapsed betweenday 4 of practice and the retention test for any subject.That is, retention testing always occurred within 48 hoursof the last practice session. However, it was necessary toallow days between practice sessions in order to accom-modate individual participants' schedules.Illustration of the sterotaxic system and markers that guided TMS coil placementFigu e 2Illustration of the sterotaxic system and markers that guided TMS coil placement. Brainsight™ was used to locate primary motor cortex (M1) for resting motor threshold determination and also to subsequently for coil placement over PMd for rTMS. Markers were placed on day 1 of testing to ensure accuracy and repeatability of coil placement and rTMS application across days.Table 2: Mean (standard deviation) RMT from day 1 and day 5 of the experiment by group. Day 1 Day 5 p-valueExcite 59.3 (7.1) 59.0 (8.3) .93Inhibit 58.0 (7.6) 59.7 (7.5) .62Sham 59.5 (9.1) 59.8 (9.0) .94Page 4 of 9(page number not for citation purposes)stimulation was never decreased below 100% RMT(Tables 2 &3).Paired t-tests revealed that there were no statistically significant changes in RMT from day 1 to day 5 of the experiment.BMC Neuroscience 2009, 10:72 http://www.biomedcentral.com/1471-2202/10/72To better separate performance effects from more perma-nent changes in behaviour associated with learning [23],a retention test consisting of 1 block of continuous track-ing was given on a separate 5th day. No rTMS was admin-istered at the retention test.Repeated sequence awarenessOn day 5 following the retention test block, participantswere shown 10, 30s blocks (all 3 sequences) of continu-ous target movement and asked to decide if they recog-nized any as the pattern that they had seen duringpractice. Three of the 10 were "true" middle sequence i.e.,the same as the repeated practice pattern; 7 were foils.Individuals who identified the repeated sequence at a bet-ter than chance rate, i.e., 2 of 3 repeated sequences identi-fied correctly as being recognized and 4 of 7 novel,random epochs identified correctly as never having beenseen before, were considered to have gained explicitawareness of the repeating sequence.Outcome measuresMotor performance was evaluated across practice andretention. Our analysis considered changes in root meansquared error (RMSE; Appendix 2), which reflects overalltracking error in the kinematic pattern and is the averagedifference between the target pattern and participantmovements. This score was calculated separately for ran-dom and repeating sequences on every tracking trial andaveraged by block (every 10 trials)[27,28,43]. Compari-son between RMSE from the repeated and randomsequences reflects sequence specific learning. This meas-ure was used to evaluate reductions in tracking errorsacross practice and at retention.To investigate the possibility that rTMS stimulated off-linemotor learning we calculated a change score to reflect thedifference in tracking error at the end of practice with thatat the retention test. This computation was performed forboth repeated and random sequences. We assume thatcontinued further decrease in tracking error (RMSE)between the last practice block and the retention testreflects off-line motor learning associated with consolida-tion [6,7].assessed with a Kolmogorov-Smirnov test. The data werenormally distributed. Overall, our analyses were con-ducted in three steps. First, we considered performancerelated changes across the four experimental days whenpractice was paired with rTMS. Second, we assessed motorlearning at the retention test on day 5. Third, we assessedoffline learning. We defined off-line learning related gainsaccording to Robertson (2004, 2006) as the difference inRMSE in between the last block of practice on day 4 (whenrTMS was last delivered) and the retention test on day 5when there was no rTMS. Our retention test was deliveredwithin 48 hours of the last practice day to ensure that weassessed off-line learning within the accepted timeframe.[7,44]Acquisition practice. Performance of the repeatedsequence during practice was examined using two factor(Group [Excite, Inhibit, Sham] X Block [1-12]) repeatedmeasures ANOVA. This analysis was performed separatelywith repeated sequence RMSE and random sequenceRMSE as the dependent variables.Retention. Motor learning at retention was examined viaa Group [Excite, Inhibit, Sham] by Sequence [Random,Repeated] repeated measures ANOVA with RMSE or track-ing error as the dependent measure. A Bonferroni correc-tion was used for post-hoc tests to determine the locus ofsignificant group by sequence interactions. Off-line motorlearning was assessed via a one-way ANOVA using thechange score from the last block of practice to the reten-tion test as the dependent measure. This test was per-formed separately for random and repeated sequencechange scores.ResultsOverall, our data demonstrate three main results. First,regardless of stimulation condition tracking error asreflected by RMSE decreased with practice. Second, at theretention test all groups showed motor learning of therepeated sequence; however, the largest amount of changebetween repeated and random sequence tracking errorwas shown by the Excite group. Third, consideration ofgains made in tracking accuracy between the last block ofpractice and the retention test demonstrate off-line motorlearning for the Excite group but not for the Inhibit orSham groups.Acquisition practiceAll groups improved performance on the repeatingsequence across practice as demonstrated by a main effectof Block for repeating sequence tracking error (F(11, 286)= 15.23, p = .000; Figure 3A). In addition, non-specificimprovements in tracking that reflect improved motorTable 3: Mean (standard deviation) adjusted %RMT for rTMS from days 1 to 4 of the experiment for the Excite and Inhibit groups.Day 1 Day 2 Day 3 Day 4Excite 107.0 (7.2) 106.6 (4.8) 104.3 (2.7) 104.9 (2.2)Inhibit 112.1 (8.6) 109.3 (7.8) 107.5 (8.7) 108.0 (8.4)Page 5 of 9(page number not for citation purposes)Statistical analysesPrior to running analyses of variance on our motor prac-tice and learning data the normality of distribution wascontrol during random sequence tracking also was dem-onstrated by a main effect of Block (F(11,286) = 11.31, p= .000; Figure 3B). There were no significant interactionsBMC Neuroscience 2009, 10:72 http://www.biomedcentral.com/1471-2202/10/72for either random or repeated sequence tracking data overpractice.RetentionAll groups demonstrated motor learning at retention asshown by a main effect of Sequence that illustrated a sig-nificant difference between tracking error for repeated andrandom sequences (F(1,26) = 99.28, p = .000). Moreinteresting, was a Group by Sequence interaction (F(2,26)= 4.257, p = .003). Post-hoc testing revealed that the Excitegroup made less tracking error than then Inhibit group (p= .002) or Sham group (p = .012) during repeatedsequence tracking but not during random sequence track-ing at retention (Figure 4).Off-line learningBetween group differences in consolidation that occurredoff-line were illustrated by a significant one-way ANOVAusing the repeated sequence change scores from thegroups as the dependent measure (F(91,26) = 8.32, p =.002). Consolidation of motor learning occurred off-linefor the Excite group as demonstrated by the continueddecrease in tracking error that occurred between the endof practice and the retention test. This finding is con-trasted to the Inhibit and Sham groups who both showedslightly worse tracking error at retention as compared tothe end of practice (Figure 5A). There were no betweengroup differences in change scores for random sequencetracking (Figure 5B).Explicit knowledgeNone of the experimental groups gained explicit knowl-edge of the repeating sequence as demonstrated by theability to identify the repeating sequence during recogni-tion tests on the final day at chance.DiscussionEven single sessions of motor practice can lead to long-term storage of movement representations in the brain[5]. It is now clear that after practice has ended the func-tional properties and representation of skilled movementcontinues to evolve in the brain [4-7]. These changes areevident in the gradual development of resistance to inter-ference from other behaviours as time passes after taskpractice [4,5]. In some cases motor skills are not merelystabilized but can be improved through this consolida-tion process [6,7]. Indeed, this is what we discoveredwhen we paired 5 Hz excitatory rTMS to left PMd withmotor task practice; motor skill continued to improve off-line after practice. Conversely, participants who practicedthe motor task and received either inhibitory or shamstimulation showed only memory stabilization, there wasno further between session improvement, but rather a rel-role in off-line motor skill enhancement. Importantly, weseparated the short-term effects of practice from more per-manent changes in behaviour demonstrated at retentionby performing these tests on different days [23]. Thisexperimental feature allowed us to view off-line learningin the Excite group without any interference effects frompractice [7].Critically, the off-line motor consolidation demonstratedTwelve blocks of sequence tracking were performed across four days f practice (3 blocks per day; each block consisted of 10 tria s of the 30 track)Figur 3Twelve blocks of sequence tracking were performed across four days of practice (3 blocks per day; each block consisted of 10 trials of the 30 track). Root mean square error (RMSE) for repeated and random sequence tracking was calculated. A) RMSE for repeated sequences across practice and at the retention test. B) RMSE for ran-dom sequences across practice and at the retention test. Data are mean RMSE ± standard error of the mean (SEM).Page 6 of 9(page number not for citation purposes)ative preservation in motor skill level acquired via practice[7]. Though these two forms of memory consolidation arenot mutually exclusive, our data suggest that PMd has aby the Excite group was related to sequence-specific motorlearning rather than to generalized improvements inmotor control associated with task practice. Illustration ofBMC Neuroscience 2009, 10:72 http://www.biomedcentral.com/1471-2202/10/72this point is evident in the difference in tracking erroracross groups for the repeated sequence at retention (Fig-ure 4); the three groups demonstrated equivalent per-formance on random sequences at the same time.Random sequence tracking reflects generalized motor exe-cution whereas repeated sequence performance showsmotor learning [27,28,43]. Thus, the role of PMd inmotor consolidation in the present work related toimplicit sequence-specific learning rather than an overallimprovement in the generalized ability to track continu-ous sequences.There has been debate as to whether PMd activity relatesto motor learning [22] or to the recall of already learnedmovements [5,20]. Our data support the hypothesis thatPMd activity facilitates motor learning, specifically by aid-ing memory consolidation. Two features of our data sup-port our conclusion. Critically, excitatory stimulation toPMd promoted off-line learning. We expected that if PMdplayed a role in recall of learned movements rather thanin motor learning, we would have noted memory stabili-zation rather than off-line improvements. Second, theimprovements associated with excitatory stimulation toPMd were sequence-specific and not simply related togeneralized motor control improvements.range of cortico-cortical and cortic-subcortical networks.On-line rTMS-fMRI imaging has shown that excitatorystimulation of PMd increases the BOLD signal bothlocally (in PMd, PMv, supplementary motor area, somato-sensory cortex, and cingulate motor area) and distantlyRMSE for repeated and random sequences at the retention testFigure 4RMSE for repeated and random sequences at the retention test. All groups showed sequence specific motor learning as demonstrated by significantly lower RMSE for repeated as compared to random sequences at retention. However, individuals in the Excite group showed even lower tracking error for repeated sequences that those in the Inhibit or Sham stimulation groups. Data are mean RMSE ± SEM.Change in tracking error (RMSE) between the last block of sequence p tice and the retention test by group for r peat d and random sequencesFig r 5Change in tracking error (RMSE) between the last block of sequence practice and the retention test by group for repeated and random sequences. A) Sequence specific off-line learning was evident for the Excite group as demonstrated by the continued decrease in tracking error from practice to retention (show by negative num-bers). This was not the case for the Inhibit or Sham stimula-tion groups who showed slight losses in tracking accuracy as evidenced by positive changes from practice to retention. B) No non-specific generalized motor control improvements occurred off-line. For random sequences slightly higher tracking error was shown at retention as compared to prac-tice for all three groups. Data are mean change in RMSE ± standard error of the mean (SEM).Page 7 of 9(page number not for citation purposes)The influence of PMd activity on motor learning andmemory consolidation likely operated though a networkof brain regions. PMd is ideally situated to impact a broad(in contralateral PMd, cerebellum, putamen and caudate;[45]). Further, these rTMS driven modifications in hemo-dynamics occur even in the absence of overt motorBMC Neuroscience 2009, 10:72 http://www.biomedcentral.com/1471-2202/10/72responses. This pattern of brain activation associated withrTMS to PMd reflects the known anatomical and func-tional connectivity amongst these regions [14,15,45].Though we cannot ascribe the offline learning we docu-mented to any single region within this broad network, itis evident that 5 Hz rTMS stimulated motor memory con-solidation most likely via up-regulating at least some ele-ments of both local and distantly connected brain regions.We expected that the Inhibit group might have demon-strated worse behaviour than those participants in theSham stimulation condition. However, it may be that thepositive effects of practice on accuracy of motor trackingperformance countered the impact of 1 Hz stimulation.Indeed, we and others [26-28,46] have shown that motortask practice of continuous tracking tasks may be welllearned over as few as three practice sessions. It is possiblethat in the present study the effects of 1 Hz TMS was eitherovercome by motor practice or that the network of brainregions associated with motor learning [5,21,45] was ableto compensate for less PMd function following inhibitorystimulation.Though sleep may have played a role in the consolidationwe noted across our experimental groups it cannot explainthe lower tracking error for repeated sequences shownonly by the Excite group at retention. Each of our groupsslept in between the last practice day and the retentiontest; conferring the benefit of sleep on motor skill consol-idation regardless of group assignment. Further, past workhas demonstrated that off-line improvements in implicitmotor learning in young, healthy controls are not sleepdependent [47]. Instead, sleep related improvements inmotor skill may develop equally well over the day as theydo over the night [44]. Thus, we do not believe that thesleep-induced benefits that are associated with consolida-tion can account for our findings.It is also unlikely that differences in explicit knowledgeexplain any of our group differences across practice or atretention; none of the groups gained explicit awareness ofthe repeating sequence. In addition, past work [27,48] hasnot shown a benefit of explicit knowledge for motorlearning of tracking tasks. Based on the results of ourexplicit tests we are confident that our data reflect changesassociated with the implicit motor learning system.We were surprised at the large number (n = 16) of individ-uals who required TMS intensity to be reduced owing toinadvertent motor twitching during PMd stimulation.These individuals were from the Excite and Inhibit groupsalike. One possible explanation is that the threshold forstimulating primary motor cortex (from which we derivedble that PMd has a lower threshold for stimulation thanM1; thus, stimulation of PMd may have either activatedM1 via PMd-M1 connections or recruited descendingtracts from PMd that normally would not fire at lowerintensities. Future work will have to endeavor to developmethods for thresholding stimulation intensity moreappropriately for regions outside motor cortex.ConclusionTaken together, our data support a role for PMd in motormemory consolidation through the process of off-linelearning. In addition, our findings support the conceptthat motor memory consolidation may take two distinctforms (off-line improvement and memory stabilization)and that these processes may be dissociated during learn-ing of the same task [6,7]. Though it is likely that 5 HZrTMS increased activation in a network of brain regions,we did note a strong influence on excitatory stimulationon memory consolidation suggesting a role for PMd inmotor learning. Importantly, the positive effect of 5 HZrTMS to PMd was directly related to sequence-specificmotor learning and had little effect on generalized motorcontrol during continuous tracking.Authors' contributionsLB conceived of the study, designed the experiment, andprogrammed the task and analyses. ML coordinated andcollected the TMS and behavioral data. Both LB and MLdrafted the manuscript, read and approved the final man-uscript.Appendix 1b0 = 2.0, a1 = -4.0, b1 = 3.0, a2 = -4.9, b2 = -3.6, a3 = 3.9, b3= 4.5, a4 = 0.0, b4 = 1.0, a5 = -3.8, b5 = -0.5, a6 = 1.0, and b6= 2.5Appendix 2xi = participant's position in degrees at time 1, Ti = targetposition at time 1, n = the number of samples for the par-ticipant's trajectory arrayAcknowledgementsSupport from the North Growth Foundation and Michael Smith Foundation for Health Research awards (LAB) funded this work. The authors thank Sean Meehan, PhD and Brenda Wessel, MS for their comments on this man-uscript.References1. Karni A, Meyer G, Rey-Hipolito C, Jezzard P, Adams MM, Turner R,Ungerleider LG: The acquisition of skilled motor performance:RMSE = -=åÖ {( ) / } /x T ni iin2 1 21Page 8 of 9(page number not for citation purposes)our resting motor threshold) is not the same as the thresh-old in other brain regions [45,49]. Specifically, it is possi-Fast and slow experience-driven changes in primary motorcortex.  Proceedings of the National Academy of Science (USA) 1998,95:861-868.BMC Neuroscience 2009, 10:72 http://www.biomedcentral.com/1471-2202/10/722. Boyd LA, Vidoni ED, Siengsukon CF: Multidimensional motorsequence learning is impaired in older but not younger ormiddle-aged adults.  Phys Ther 2008, 88(3):351-362.3. van Raalten TR, Ramsey NF, Duyn J, Jansma JM: Practice inducesfunction-specific changes in brain activity.  PLoS ONE 2008,3(10):e3270.4. Brashers-Krug T, Shadmehr R, Bizzi E: Consolidation in humanmotor memory.  Nature 1996, 382:252-255.5. Shadmehr R, Holcomb HH: Neural correlates of motor memoryconsolidation.  Science 1997, 277(5327):821-825.6. Robertson EM, Cohen DA: Understanding consolidationthrough the architecture of memories.  Neuroscientist 2006,12(3):261-271.7. Robertson EM, Pascual-Leone A, Miall RC: Current concepts inprocedural consolidation.  Nature reviews 2004, 5(7):576-582.8. Goedert KM, Willingham DB: Patterns of interference insequence learning and prism adaptation inconsistent withthe consolidation hypothesis.  Learn Mem 2002, 9(5):279-292.9. Korman M, Raz N, Flash T, Karni A: Multiple shifts in the repre-sentation of a motor sequence during the acquisition ofskilled performance.  Proceedings of the National Academy of Sciencesof the United States of America 2003, 100(21):12492-12497.10. Pascual-Leone A, Grafman J, Hallett M: Modulation of corticalmotor output maps during development of implicit andexplicit knowledge.  Science 1994, 263(5151):1287-1289.11. Walker MP, Stickgold R, Jolesz FA, Yoo SS: The functional anat-omy of sleep-dependent visual skill learning.  Cereb Cortex 2005,15(11):1666-1675.12. Rushworth MF, Johansen-Berg H, Gobel SM, Devlin JT: The leftparietal and premotor cortices: motor attention and selec-tion.  NeuroImage 2003, 20(Suppl 1):S89-100.13. Kalaska JF, Crammond DJ: Deciding not to GO: neuronal corre-lates of response selection in a GO/NOGO task in primatepremotor and parietal cortex.  Cereb Cortex 1995, 5(5):410-428.14. Dum RP, Strick PL, Humphrey DR, Freund HJ: Premotor areas:Nodal points for parallel efferent systems involved in thecentral control of movement.  In Motor Control: Concepts andIssues John Wiley & Sons Ltd; 1991:383-397. 15. Picard N, Strick PL: Motor areas of the medial wall: A review oftheir location and functional activation.  Cerebral Cortex 1996,6:342-353.16. Koch I, Allport A: Cue-based preparation and stimulus-basedpriming of tasks in task switching.  Memory & cognition 2006,34(2):433-444.17. Crammond DJ, Kalaska JF: Differential relation of discharge inprimary motor cortex and premotor cortex to movementsversus actively maintained postures during a reaching task.Exp Brain Res 1996, 108(1):45-61.18. Thoenissen D, Zilles K, Toni I: Differential involvement of pari-etal and precentral regions in movement preparation andmotor intention.  J Neurosci 2002, 22(20):9024-9034.19. Schluter ND, Krams M, Rushworth MF, Passingham RE: Cerebraldominance for action in the human brain: the selection ofactions.  Neuropsychologia 2001, 39(2):105-113.20. Chen YC, Thaler D, Nixon PD, Stern C, Passingham RE: The func-tions of the medial premotor cortex (SMA) II. The timingand selection of learned movements.  Experimental BrainResearch 1995, 102:461-473.21. Johansen-Berg H, Rushworth MFS, Bogdanovic MD, Kischka U,Wimalaratna S, Matthews PM: The role of ipsilateral premotorcortex in hand movement after stroke.  Proceedings of theNational Academy of Sciences of the United States of America 2002,99(22):14518-14523.22. Ohbayashi M, Ohki K, Miyashita Y: Conversion of working mem-ory to motor sequence in the monkey premotor cortex.  Sci-ence 2003, 301(5630):233-236.23. Salmoni AW, Schmidt RA, Walter CB: Knowledge of results andmotor learning: a review and critical reappraisal.  Psychol Bull1984, 95(3):355-386.24. Schmidt RA, Lee TD: Motor Control and Learning: A Behavioral EmphasisChampaign: Human Kinetics; 2005. 25. Lundy-Ekman L: Neuroscience: Fundamentals for Rehabilitation Philadel-phia: WB Saunders; 1998. 26. Wulf G, Schmidt RA: Variability of practice and implicit motor27. Boyd LA, Winstein CJ: Providing explicit information disruptsimplicit motor learning after basal ganglia stroke.  Learning &Memory 2004, 11(4):388-396.28. Boyd LA, Winstein CJ: Cerebellar stroke impairs temporal butnot spatial accuracy during implicit motor learning.  Neurore-habilitation and neural repair 2004, 18:134-143.29. Vidoni ED, Boyd LA: Motor sequence learning occurs despitedisrupted visual and proprioceptive feedback.  Behav BrainFunct 2008, 4:32.30. Yousry TA, Schmid UD, Alkadhi H, Schmidt D, Peraud A, Buettner A,Winkler P: Localization of the motor hand area to a knob onthe precentral gyrus. A new landmark.  Brain 1997, 120(Pt1):141-157.31. Wassermann EM, Wang B, Zeffiro TA, Sadato N, Pascual-Leone A,Toro C, Hallett M: Locating the motor cortex on the MRI withtranscranial magnetic stimulation and PET.  NeuroImage 1996,3(1):1-9.32. Boroojerdi B, Foltys H, Krings T, Spetzger U, Thron A, Topper R:Localization of the motor hand area using transcranial mag-netic stimulation and functional magnetic resonance imag-ing.  Clin Neurophysiol 1999, 110(4):699-704.33. Dassonville P, Lewis SM, Zhu XH, Ugurbil K, Kim SG, Ashe J: Theeffect of stimulus-response compatibility on cortical motoractivation.  NeuroImage 2001, 13(1):1-14.34. Dassonville P, Lewis SM, Zhu XH, Ugurbil K, Kim SG, Ashe J: Effectsof movement predictability on cortical motor activation.Neurosci Res 1998, 32(1):65-74.35. Fridman EA, Hanakawa T, Chung M, Hummel F, Leiguarda RC, CohenLG: Reorganization of the human ipsilesional premotor cor-tex after stroke.  Brain 2004, 127(Pt 4):747-758.36. Toni I, Passingham RE: Prefrontal-basal ganglia pathways areinvolved in the learning of arbitrary visuomotor associations:a PET study.  Exp Brain Res 1999, 127(1):19-32.37. Gerschlager W, Siebner HR, Rothwell JC: Decreased corticospi-nal excitability after subthreshold 1 Hz rTMS over lateralpremotor cortex.  Neurology 2001, 57(3):449-455.38. Munchau A, Bloem BR, Irlbacher K, Trimble MR, Rothwell JC: Func-tional connectivity of human premotor and motor cortexexplored with repetitive transcranial magnetic stimulation.J Neurosci 2002, 22(2):554-561.39. Ruohonen J, Ilmoniemi RJ: Physical principles for transcranialmagnetic stimulation.  In Handbook of Transcranial Magnetic Stimu-lation First edition. Edited by: Pascual-Leone A, Davey NJ, Rothwell J,Wassermann EM, Puri BK. New York: Arnold; 2002:24. 40. Chouinard PA, Leonard G, Paus T: Role of the primary motorand dorsal premotor cortices in the anticipation of forcesduring object lifting.  J Neurosci 2005, 25(9):2277-2284.41. Foltys H, Krings T, Meister IG, Sparing R, Boroojerdi B, Thron A,Topper R: Motor representation in patients rapidly recover-ing after stroke: a functional magnetic resonance imagingand transcranial magnetic stimulation study.  Clin Neurophysiol2003, 114(12):2404-2415.42. Chen WH, Mima T, Siebner HR, Oga T, Hara H, Satow T, Begum T,Nagamine T, Shibasaki H: Low-frequency rTMS over lateral pre-motor cortex induces lasting changes in regional activationand functional coupling of cortical motor areas.  Clin Neuro-physiol 2003, 114(9):1628-1637.43. Boyd LA, Quaney BM, Pohl PS, Winstein CJ: Learning implicitly:effects of task and severity after stroke.  Neurorehabilitation andneural repair 2007, 21(5):444-454.44. Robertson EM: Skill learning: putting procedural consolidationin context.  Curr Biol 2004, 14(24):R1061-1063.45. Bestmann S, Baudewig J, Siebner HR, Rothwell JC, Frahm J: BOLDMRI responses to repetitive TMS over human dorsal premo-tor cortex.  NeuroImage 2005, 28(1):22-29.46. Shea CH, Wulf G, Whitacre CA, Park JH: Surfing the implicitwave.  Q J Exp Psychol A 2001, 54(3):841-862.47. Robertson EM, Press DZ, Pascual-Leone A: Off-line learning andthe primary motor cortex.  J Neurosci 2005, 25(27):6372-6378.48. Green TD, Flowers JH: Implicit versus explicit learning proc-esses in a probabilistic, continuous fine-motor catching task.Journal of motor behavior 1991, 23:293-300.49. Stewart LM, Walsh V, Rothwell JC: Motor and phosphene thresh-olds: a transcranial magnetic stimulation correlation study.Page 9 of 9(page number not for citation purposes)learning.  Journal of Experimental Psychology: Learning, Memory andCognition 1997, 23(4):987-1006.Neuropsychologia 2001, 39(4):415-419.


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items