Beyond Jeannerod's motor simulation theory: An approach for improving post-traumatic motor rehabilitation

(1)

HAL Id: hal-02051088

https://hal-insep.archives-ouvertes.fr/hal-02051088

Submitted on 27 Feb 2019

HAL

is a multi-disciplinary open access archive for the deposit and dissemination of sci- entific research documents, whether they are pub- lished or not. The documents may come from teaching and research institutions in France or abroad, or from public or private research centers.

L’archive ouverte pluridisciplinaire

HAL, est

destinée au dépôt et à la diffusion de documents scientifiques de niveau recherche, publiés ou non, émanant des établissements d’enseignement et de recherche français ou étrangers, des laboratoires publics ou privés.

approach for improving post-traumatic motor rehabilitation

Claire Calmels

To cite this version:

Claire Calmels. Beyond Jeannerod’s motor simulation theory: An approach for improving post-

traumatic motor rehabilitation. Neurophysiologie Clinique/Clinical Neurophysiology, Elsevier Masson,

2019, �10.1016/j.neucli.2019.01.033�. �hal-02051088�

(2)

Please cite this article in press as: Calmels C. Beyond Jeannerod’s motor simulation theory: An approach Disponibleenlignesur

ScienceDirect

www.sciencedirect.com

PERSONAL VIEW

Beyond Jeannerod’s motor simulation theory: An approach for improving post-traumatic motor rehabilitation

Claire Calmels

^∗

LaboratorySport,ExpertiseandPerformance(EA7370),ResearchDepartment,FrenchInstituteofSport (INSEP),75012Paris,France

Received8September2018;accepted12January2019

KEYWORDS Motorsimulation;

Orthopedictrauma injuries;

Recovery;

Transient immobilization;

Transient sensorimotor deprivation

Summary Thispaperdepictsanapproachaimingtoallowindividualswithorthopedictrauma injuriestooptimizerecoveryandsafelyreturntodailylife.Thecoreofthisapproachismotor simulation,usedincomplementtoconventionalphysicalrehabilitationmethods.Thispaper providesrecentscientiﬁcinsightsonthebasisofmotorsimulation,discussesbeneﬁtsofthis approachinmotorrehabilitation,andprovidesappliedperspectives.

Introduction

This paper describes an approachaimed at allowing indi- vidualswithorthopedictraumainjuriestooptimizephysical recovery and safely return to daily life. The core of this approachismotorsimulation, usedincomplementtotra- ditional physical rehabilitation. According to Jeannerod [47—49], motor simulation is characterized by action-

∗LaboratoireSEP(EA7370),unitérecherche,institutnationaldu sport,del’expertiseetdelaperformance,11,avenueduTremblay, 75012Paris,France.

E-mailaddress:[email protected]

relatedcognitive states,such asmotor imagery or action observation, which activate cortical motor systems similar tothoseinvolved duringactual action. Motorimagery relies on representational neural networks involving top- down sensorial, perceptual and affective characteristics thatareprimarilyundertheconsciouscontroloftheimager andwhich may occur in theabsence of perceptual afference,beingfunctionallyequivalenttoactualmotoraction experience.On the other hand, action observation relies on representational neural networks involving bottom-up sensorial,perceptualandaffectivecharacteristicsthatare primarily under the subconscious control of the observer andwhich occur in thepresence of afference, alsobeing functionallyequivalenttoactual motor action experience

https://doi.org/10.1016/j.neucli.2019.01.033

(3)

Please cite this article in press as: Calmels C. Beyond Jeannerod’s motor simulation theory: An approach

ARTICLE IN PRESS

2 C.Calmels

[41,42].BasedonJeannerod’stheory[47—49],motorsimu- lationmayoccurin3forms:

• whentheactionisimaged(imagedaction);

• whentheactionisobserved(observedaction)or;

• whenitisverballydescribed,eithersilentlyoraloud(ver- balizedaction).

This paper is composed of four parts. The first part presentsJeannerod’smotorsimulationtheoryonwhichthe aforementionedapproachisbased.Thesecondpartfocuses onrecentscientificfindingsoneffectsofsensoryandmotor deprivationonmotorperformance andhuman brainorga- nization.Thethirdpartdiscussesmotorsimulationtraining tocounteractdetrimentaleffectsoftransientsensorimotor deprivation,andthelastpartsuggestsappliedperspectives.

Motorsimulation Jeannerod’stheory

Thistheoryexplainshowmotor-relatedcognitivetaskssuch as motor imagery (imaging the execution of an action without physically performing it) or action observation (observinganactionexecutedbyanotherorbyoneself)are connectedtoactiontasksthatareactuallyperformed.The basisof the motor simulationtheoryis that commoncor- ticalmotorsystemsareactivatedwhenimaging,observing orexecutingan action.Jeannerodstatesthatan intended actionthatisexecutedconsistsoftwoparts:acovertstage, whichisnotdirectlyobservableandanovertstage,which isdirectlyobservable[47—49].Thecovertstageisarepre- sentationofthefuture:thegoaloftheaction,themeansof performingtheactionanditsconsequencesontheorganism andtheexternalworld.Fortheovertstage,thisrepresents the actual action, that is, an action willingly completed.

Jeannerodalsospeciﬁedthatcovertstagescouldcomprise different kinds of action, such as voluntary action that willbesubsequentlyexecuted(i.e.,thatwillleadtoovert action), imagined action or observed action (i.e., covert orsimulatedaction) [47—49].Therefore,covert andovert actionscanbeplacedonacontinuumgoingfromintentionto motorexecution.Consequently,inordertobegenerated,an overtexecutedactionmustbeprecededbyacovertstage, whereasacovertactiondoesnotnecessarilyevolveintoan overtaction.Indeed,whenimagingorobservinganaction, motorcommandinhibitorymechanismsblockmotoroutput.

Inotherwords,‘‘covertactionsareinfactactions,except for thefact theyarenot executed’’ [48].Jeannerod also advancesthatcovertandovertactionsareboth preceded byrepresentationoftheirsensorialconsequencesandfuture states. Thus, simulating an action involves a representa- tionofits currentstate aswell asa representationof its subsequentstates.

Functionalequivalenceofimaged/observedaction andexecutedaction

Over the past twenty years, neuroimaging technological developments and widespread use of meta-analyses have

allowedsomesupportforJeannerod’smotorsimulationthe- ory, showing that both movement simulation(i.e. imaged andobservedactions),andactualexecutionofmovement, activatecommoncorticalareas[15,32,36,39,48].Earlyneu- roimaging literature reviews revealed common recruited motorareas(primarymotorandpremotorareas,prefrontal and parietal areas, basal ganglia and cerebellum). How- ever, these studies were based on methodology that did notintegrateprincipledstatisticaltesting[32,48].Adecade later, Casper et al. [15] and Hetu et al. [39] conducted meta-analyses allowing more objective assessment of the evidence, by independently investigating neural networks involved duringmotorimageryandaction observationi.e.

withoutcomparingthemtoeachotherandtomotorexecu- tion.

Very recently, in their meta-analysis, Hardwick et al.

[36]examinedneuralnetworksinvolvedinactionexecution, motor imagery and action observation. Using a conjunc- tionanalysisacrosstheovertaction(i.e.,executedaction) anditstwocovertcounterparts(i.e.,imagedandobserved actions),theypinpointedaconsistentlycommonpremotor- parietal and primary somatosensory network. They also identified specific areas for each of these threedifferent tasks.Morespecifically,anetworkincludingpremotor,sensorimotor and subcortical structures(putamen, thalamus, andcerebellum)showedactivitywhensubjectsperformed anactualaction,andontheotherhandanetworkinvolving premotor,parietalandsubcorticalstructures(putamenand cerebellum)wasactivatedwhensubjectswereinstructedto imagineanaction.When anactionwasobserved,thenet- workinvolvedpremotor,parietalandoccipitalareaswithno consistentinvolvementofsubcorticalareas.

Takentogether Hardwick’s team results are broadly in accordance with those of Grezes and Decety [32] and Jeannerod [48] in showing the involvement of premotor and inferior parietal areas when imaging, observing and executinganaction.Moreover,Hardwicketal.[36]addition- allyrevealedrecruitmentofprimarysomatosensorycortex acrossthesethreetasks.

Interestingly,ventralpremotorcortex(vPM),dorsalpre- motorcortex(dPM)andpre-SMA,allareasrecruitedinmotor simulation and execution, are recognized to be involved in action preparation[44] andin learning arbitraryvisuo- motor mapping [62,77]. The vPMcortex is thought to be involved inﬁnemotor coordination[21]whereasitscoun- terpart, the dPM cortex, is presumed to be involved in action selection and to a lesser extent in action execution [69]. The pre-SMA plays a key role in self-initiated actions[40].Furthermore,itisofinterestthatinferiorpari- etalcortexistypicallyimplicatedinprocessingmultisensory information[7],preparingmovementandgeneratingmotor intention[22,23,73],aswellasinstoringkinestheticrepre- sentationstomapthemontothepremotorandmotorareas [70].Aspreviouslythought,theprimarysomatosensorycor- texisnotlimitedtoprocessingsomatosensoryinformation such asauditory, visual, tactile, proprioceptive and noci- ceptiveinformation,butratherparticipatesinintegrating, transmittingandreceivingmultimodalinformationthrough cortico-subcorticalnetworks,thusregulatingsensationsand movementsinordertomakethemaseffectiveaspossible.

This is completed in collaborationwith the motor cortex [8].

(4)

Please cite this article in press as: Calmels C. Beyond Jeannerod’s motor simulation theory: An approach Functionalequivalenceofverbalizedactionand

executed/simulated/mimedaction

Various studies have also demonstrated neuronalnetwork equivalencebetweenverbalizedandexecutedactions.For example, Hauk et al. [37] showed cortical overlap when action verbs related toarm or leg movements were pas- sively read out, and when they were actually executed:

in both cases,premotor andprimary motorcortices were somatotopicallyactivated.Interestingly,thesecorticesare recognizedtobeinvolvedinactionprogramming[44,64,69]

andmorespeciﬁcally,primarymotorcortexappearstobe inchargeofencodinglevelsofmuscleactivationaswellas movementdirection.

Ithasalsobeenevidencedthatverbalizedactionshares functional cortical networks with both simulated action (imaged action or observed action) and mimed action.

Firstly,Aziz-Zadehetal.[2]showedthatthe(left)premo- torareawasinvolvedwhenreadingactionphrasesrelatedto armorlegmovements,aswellaswhenobservingthem.Sec- ondly,Peranetal.[68]consideredthreetasks:actionverb generation,imaged action, andmimed action. Using fMRI conjunctionanalysisacrossthesetasks,theyrevealedcom- mon neuronal networks including premotor, parietal, and occipitalcortices.

Brainplasticityandtransientsensorimotor deprivation

Developmentofneuroimagingtechniquesoverthelastfew yearshasprovidedcompellingevidenceofthebrain’sability toreorganizeitsneuronalconnectionsinformandfunction throughout life [25], in response to internal and external constraints or goals. This process involves short-term changeswhichleadtomorestable,longer-termalterations, providedthat the individualis exposed for along enough periodoftimetotheseconstraintsorgoals,whetherinter- nal or environmental [66].This reconfiguration, knownas brain plasticity and driven by sensory input [26,52] and motoroutput[16,17,38],aimsatmaximizingthefunction- ingofneuralnetworksduring(human)development,during learning,and in responsetobrain lesions[66].Generally, brain plasticity is perceived as being beneficial but may alsobe damaging. When plasticityallows an individualto improve behavioral capacity [18,24,71], it is qualified as adaptive;conversely,whenlinkedtobehavioraldeteriora- tion and leading to negative consequences, it is seen as maladaptive[27,28].

Effectsofsensoryandmotordeprivationonmotorperfor- manceandhumanbrainorganizationhavebeeninvestigated over recent decades [30]. Some research groups have focusedonimpairmentsresultingfromneurologicalinjuries (e.g.strokes,limbamputations,spinalcordinjuries),whilst othershaveinvestigatedalterationsresultingfromorthope- dictraumainjuries(e.g.bonefractures,ligamentormuscle injury,joint dislocations). In thispaper, we consideronly orthopedic trauma requiring transient immobilizations of variousdurations.

Transientsensorimotordeprivationaffectsmotor andcognitiveperformance

Toexaminetheeffectsoftransientsensorimotordeprivation onmotororcognitiveperformance,ashort-termupperlimb immobilizationparadigm in healthy participants has most oftenbeenused[3,5,58,60,74].

Afterimmobilizingsubjects’leftarmsfor12hours,Moi- selloetal.[60]showedchangesinlimbkinematicsduring reachingmovements.Changeswerenotseenafter6hours ofimmobilizationbutoccurredafter12hours,atwhichtime greaterhand-pathareasandinter-jointcoordinationtiming wereobserved.Bassolinoetal.[3],usinga10-hourimmobilization periodof theright hand, showed alterations in areaching-to-grasptask.Increaseddurationofthe reaching phaseassociated witha time reduction to attain the peakvelocitywithinthisphaseoccurred;theauthorsinter- preted this as reﬂecting incorrect prediction of sensory consequences of the reaching task. During the immobi- lizationperiod,sincenomovementcan beperformed, no visual or proprioceptive information associated with arm movementisgenerated.Becausethisinformationisessen- tialinordertoaccessandmaintainmotorrepresentations stored in brain [9,50], the reaching move can no longer beupdated.Consequently, discrepancybetweenexpected and actual sensory consequences of the motor command couldexplaintheimmobilization-inducedchangesobserved duringtheexecution ofthe move.Bosbach etal.[9]also statedthatintheabsenceofsuchperipheral information, motorrepresentationsaresupposedtofade;consequently, beingdeprivedofsuchsensationspreventsindividualsfrom beingabletosimulateactions.Tenyearslater,theseﬁnd- ings were supported by Toussaint’s research group. This groupinvestigatedeffectsof48-hourleft-handimmobiliza- tioninhealthyparticipantsonsensorimotorrepresentations [57,58,74]. They used hand and foot laterality tasks to determinethe qualityof sensorimotorrepresentations, as thesetasksrequireimplicituseofmotorimageryandthus provideinformation onthe centralprocessing ofthe sensorimotor system [65]. They found that after 24hours of left-handimmobilizationsensorimotorrepresentationofthe immobilized hand was altered, and that 48hours later, representations ofboth immobilizedand non-immobilized handswereimpaired [59,74].Moreover,they alsodemon- stratedthatupperlimbimmobilizationfor48hoursdidnot impact upon sensorimotor representation of lower limbs [58].

Bidet-Ildeietal.[5]found that24hourimmobilization oftheright hand canaffectaction verbprocessing.Using asemanticdecisiontaskin whichindividualshadtojudge whetheran actionverbinvolvedhandor footmovements, theyobserved thatcontrolparticipantswhosehandswere notrestrictedrespondedmorequicklytohandactionverbs thantofootactionverbsinpre-andpost-tests.Conversely, afterimmobilization,participantsdidnotdisplaysigniﬁcant differenceinresponsetimebetweenhandandfootaction verbprocessingeventhough,liketheircontrolcounterparts, theyhadresponded morequickly tohand action verbs in thepre-test.Thelackofdifferenceobservedinimmobilized subjectswasinterpretedby alesserimprovementinhand actionverbprocessingbetweenthepre-andpost-tests,due tosensorimotor deprivation. These ﬁndings highlight that

(5)

ARTICLE IN PRESS

4 C.Calmels

highercognitivefunctions arecloselyassociatedwithsen- sorimotorexperience;thisisfullyinline withLyons etal.

[55],whocomparedexpertplayersinice-hockeytonovices and found greater activation within left premotor cortex inexpertslisteningtosentencesrelatedtospeciﬁchockey actions,whilenosuchdifferencewasregisteredwhenlisting tosentencesbasedonactivitiesofdailyliving.

Transientsensorimotordeprivationaffectsbrain organization

Tostudytheeffectsoftransientsensorimotordeprivationon humanbrainorganization,twomaingroupsofsubjectshave beenstudied:healthyindividualsdeprivedofafferentand efferentinformationbywearingacastorasplint[1,45,76];

and individuals suffering from orthopedic trauma injuries requiringacastorsplintontheinjuredbodypart[53,54].

Imposing temporary upper limb immobilization on healthy participants, Huber et al. [45] found, after 12 consecutive hours of immobilization, changes in the contralateralsensorimotor cortexindicating motorperfor- mancedeteriorationanddepression ofsomatosensoryand motorevokedpotentialsfortheimmobilizedupperlimb.

Along similarlines, after 10hours of immobilizationof the right hand, Avanzino et al. [1] found decreased cortical excitability within left primary motor cortex (M1) aswellasdiminution ofinter-hemispheric inhibition from left to right hemisphere. Interestingly, right hand immobilization led to increased excitability of right primary motor cortex as well as greater inter-hemispheric inhibition fromright to left hemisphere, only observed among subjectswhousedtheirleftarmexcessively.Thisincreased excitabilitycouldbeduetoincreasedproprioceptiveinput generated by overuse of the lefthand. Using fMRI, after 72hoursofimmobilizationof theright hand andforearm, Weibulletal. [76]showeddecreasedactivation inipsilat- eralprimarysomatosensorycortexandprimarymotorcortex whenperformingaﬁnger-tappingtaskwiththeimmobilized hand.Conversely,performingthesametaskwiththenon- immobilizedhand wasassociated withbilateral increased activationwithinprimarysomatosensory,primarymotorand sensoryassociationcortex.Thisstudyalsorevealedlossof grip strength, dexterity and tactile discrimination of the immobilizedhandassociatedwithincreasedtactilediscrim- inationofthenon-immobilizedhand.

Inindividualssufferingfromorthopedictraumainjuries, Lisseketal.[54]showed,after asixweek immobilization periodofhandandarm,impairmentintactileperception, handuseandshrinkageofsomatosensorycorticalmaps.Per- ceptualcompensationalsooccurred,reﬂectedbyenhanced tactileacuityinthenon-affected hand.Langeretal.[53]

describedreduced corticalthicknesswithinthecontralat- eral sensorimotor cortex as well as fractional anisotropy (FA)decreaseinthecontralateralcorticospinaltractafter twoweeks’immobilization of thewhole right arm. More- over, observed increasesin cortical thicknesswithin right primarymotor cortexandFAwithinrightpremotor cortex wereinterpretedascompensatoryprocesses.

Takenaltogether,transient sensoryand motordeprivation,investigatedbymeans ofupperlimb immobilization, lead to rapid remodeling of the sensorimotor system

reﬂected by functional as well as structural changes.

Decreasedactivationinsensorimotorcortexanddeteriora- tionofmotorperformancewereobserved[1,45,53,54,76].

Compensation effects reﬂected by substantial activation increasesin adjacent or contralateralareas of theimmo- bilized upper limb have also been observed, suggesting compensationforareasthataredeprivedofinput.Thishas been explained by increased use of the non-immobilized handtodealwithdailydemands[1,53,54,76].

Doestransientsensorimotordeprivationaffect motorsimulation?

To investigate whether sensorimotor deprivation affects motorsimulationprocesses,studieslookedateitherhealthy individuals deprived of sensorimotor information [11] or individuals with orthopedic trauma injuries [12]. Asking healthy participants to wear a splint on the right hand, Burianovaetal.[11]observed,after24hoursofimmobiliza- tion,changesinbrainactivitywhenthesubjectsperformed theﬁngerconﬁgurationtask,whichrequiresimplicituseof motorimagery [10].Theyfoundaresting motorthreshold increaseinmotorcortexcontralateralbutnotipsilateralto the immobilized hand, suggesting decreased corticospinal excitability in projections tothe immobilized hand. Acti- vation decrease within primary motor cortex and BA6 contralateraltotheimmobilizedhandwerealsodetected.

Curiously, imagery performances assessed by the ﬁnger conﬁguration task did not change after immobilization, regardlessoflateralityofhandimmobilization.

InCalmelsetal. [12],brainhemodynamic activitywas recorded twice in 13 national female gymnasts suffering from a lowerextremity injury at the onset of the exper- iment. The gymnasts were scanned one month after the injury andwere showngymnastic routines theywerenor- mallyablebut temporarilyunabletoperform.Sixmonths later, after complete recovery, they were scanned again andshown the sameroutinesthat bythis timetheywere abletopracticeagain.Results showedﬁrstlythat activity withininferiorparietallobuleandMT/V5/EBA(extrastriate bodyarea),areasconstitutiveoftheactionobservationnet- work,wasindependentofthegymnasts’physicalcondition.

Second,duringtheperiodofinjury,higheractivityincere- bellumwasdetected.TheequalcontributionofMT/V5/EBA andinferiorparietallobuleduringtheobservationofmove- mentsthat the gymnasts wereable or unabletopractice suggests respectively that physical provisional incapacity doesnotinterferewithperceptualprocessingofbodyshape and motion information, and that motor expertise may preventthedeteriorationofsensorimotorrepresentations.

Higheractivationsoccurredincerebellum,whichisknown to play a key role in prediction of sensory consequences [6,78]andupdatingpredictionsabout visualconsequences ofbehavior[72].Thissuggeststhatwheninjured,predicting theoutcomeofothers’viewedactionsmaybeaffected.Asa consequence,estimatederrorbetweenpredictedoutcomes andincoming consequencesoftheviewedmovement may behigher.

Takentogether,bothtypesofstudyshowthattransient sensorimotordeprivation affectsbrainorganizationduring

(6)

Please cite this article in press as: Calmels C. Beyond Jeannerod’s motor simulation theory: An approach motorsimulation.However,additionalstudiesarerequired

tobetterelucidatethisissue.

Trainingthebraintocounteractdetrimental effectsoftransientsensorimotordeprivation

Itisthereforereasonabletoconsiderthepossibilityofpre- ventingdeteriorationofmotorrepresentationsoreventheir loss,insubjectswithtransientsensoryandmotordepriva- tion.Isitpossibletogeneratesensations,suchaskinesthetic and haptic sensations, without concomitantly performing actualmovements?Canmotorsimulationallowthis?

Individual experiential testimonies indicate that sub- jectsperformingmotorsimulation,areabletoexperience sensations.Thiscanbeexplained:ﬁrst,kinestheticsensa- tions perceived during motor imagery must be the result of expected sensory consequences of the action, as they are not generated by actual movement nor by periph- eral input [29,33,48,63]. Second, when observing others’

actions,motorandsomatosensorycorticesoftheobserver are activated as if the observer was really performing theviewedaction[15,31,51].Theobserverthus simulates motoroutputandsomatosensoryinputnecessarytoproduce the viewedaction [51]. Interestingly, primarysomatosen- sory area BA2 is not limited to encoding somatosensory information as waslong believed, but is also involved in motorinformationcoding[75].Moreover,Calvo-Merinoetal.

[13,14] showed that activation of this areawhen observ- ingclassicaldancemovementswaspositivelycorrelatedto the observer’s degree of motor expertise. Finally, Guillot etal.[35]demonstratedthat combiningmotorsimulation (andmorespeciﬁcallyimagery)withmimesincreasedvivid- nessofimagery,temporalcongruencebetweenimageryand actualexecutionandtechnicalqualityandefﬁcacyofmove- ments.Thus,couplingmimedactionsinvolvinghands,upper body or whole body withmotor simulation(i.e., imaged, observed,verbalizedactions)couldinduceperipheral sensations.Thesesensationscouldbemoreintensethanthose elicitedbymotorsimulationalonebutlessthanthosegen- eratedbyactualexecutionofmovements.

To summarize, since many common cortical areas are activated both during motor simulation and execution of movements,andsincemotorsimulationcanproduceperiph- eralsensationsneededtodrivebrainplasticityforcortical reorganization,itcouldthusbepostulatedthatpracticing motor simulationconstitutestrainingfor actualexecution ofmovements.Consequently,motorsimulationcouldbea meanstocounteract negativeeffectsof transientsensori- motordeprivation.

Fewstudieshaveinvestigatedthistopicinsubjectswith no past neurological history, who experience an immobi- lizationperiodduetoatraumaticlimbinjury[56,61]orin whomsuchimmobilizationisimposedforexperimentalrea- sons[4,19,57].First,bothMoukarzeletal.[61]andMarusic etal.[56]foundsomebeneﬁtsofusingmotorsimulationin rehabilitation ofpatients whohad respectivelyundergone total kneeand hiparthroplasty surgery. More speciﬁcally, Moukarzeletal.[61]instructedpatientstoperformvisual imageryfromaninternalandexternal perspectiveaswell askinestheticimagery.Three15minute-sessionsperweek

wereprovidedfor 4weeksandthreesetsofimagerypro- gramsweredevisedbasedontheintendedgoals:

• painmanagement;

• kneeﬂexionrangeofmotionincrease;

• quadricepsstrengthimprovement.

Moukarzel et al. [61] mainly showed decreased pain and increased quadriceps strength. Marusic et al. [56]

asked patients to observe video clips depicting locomo- toractions andconcomitantly tofeelthe sensationsasif theywereactually performingtheaction. Twenty-sixses- sions, each lasting 30minutes and spread over 2 months, wereperformed.Resultsshowedbettermotorperformances in patients following the motor simulation program com- pared to control patients who did not beneﬁt from this program.However,nodifferencewasobserved whenper- formingmotoractionsoutsidethemotorsimulationtraining program.

Second, studies of healthy participants wearing a cast or a splint to induce temporary sensorimotor deprivation showdivergentresults[4,19,57].Usinganexperimentalpre- test/post-testdesignwithcontrolandexperimentalgroups, Crews and Kamen [19] placed all subjects’ left hands in a cast during a period of seven days. The experimental groupfollowed an imagery program composed of 300 tri- als in which a motor task devisedby Payton et al. [67], requiringabductionmovementsof thelittleﬁnger, hadto be mentally performed. Subjects were asked to imagine themselvesperformingthe taskfromaninternal perspec- tiveandtofeelkinestheticsensationsassociatedwiththis task.Three 30-minute trainingsessions were spread over theimmobilization period.The control group didnot fol- lowtheimageryprogram.Afterthecasthadbeenremoved, Crews and Kamen [19] found, decrease in motor evoked potentialwithinthe hand areaof the primarymotor cor- texinbothgroupswithnodifferencebetweenthese.Asfor motorperformancesofthetaskimagedbytheexperimen- talgroup,therewerenodifferencesinaccuracyscores,but errorscoreswerehigherintheexperimentalgroup.Crews andKamenthus [19]showedthatmotorimageryfailedto counteractthedetrimentaleffectsoftransientsensorimo- tordeprivation.

Bassolino et al. [4] immobilized the right arm of all subjects for 10hours and used an experimental pre- test/post-test design with three groups (control group, observationgroup,andimagerygroup).The controlgroup watched documentaries with no human agents whereas the observation group observed videos displaying a right hand froman internal perspective, reaching andgrasping objects with different kinds of grips. The imagery group wasinstructedtoimagine,witheyesclosed,reachingand graspingobjects withtheirimmobilizedhand while trying tofeelthesensationstheyexperiencedwhenactuallyper- formingtheaction.Tasksdescribedabovewereperformed ten times at the rate of one session per hour, each ses- sionlastingaround4minutes.Bassolinoetal.[4]foundthat onlyobservationwasabletocounteractthenegativeeffects of sensorimotor deprivation, as motor cortex excitability registeredin theobservation groupwasgreater thanthat recordedincontrolandimagerygroups.

(7)

ARTICLE IN PRESS

6 C.Calmels

Meugnotetal.[57]recruitedparticipantswhowereran- domlyassignedtofourgroups:

• immobilizedgrouppracticingvisualimagery;

• immobilizedgrouppracticingkinestheticimagery;

• immobilizedgroupwithnoimagerypractice,and;

• non-immobilizedgroupwithnoimagerypractice.

Immobilizedparticipantsworeasplintontheirlefthand for 24hours. In visual imagery, subjects were instructed toimagine themselves performing hand and ﬁnger motor actionswiththeir immobilizedhand by focusingonvisual information.Inkinestheticimagery,theyweretofeelthem- selvesperforminghandandﬁngermotoractionswiththeir immobilizedhand by focusing onkinesthetic information.

Imagerysessionswereexecutedjustbeforesplintremoval during3sequencesof5minutes,witheyesclosed.Ahand laterality task, recognized to provide information on the central processing of the sensorimotor system [65], was completedbyallparticipantsjustafterthesplintremoval.

Meugnotetal.[59]foundthatonlykinestheticimagerywas abletocounteractthenegativeeffectsinducedbytransient sensorimotordeprivation,reﬂectedbyslowingofsensorimo- torprocesses.

In view of these results, the beneﬁts of using motor simulation to prevent detrimental effects of transient sensorimotor deprivation have been highlighted by some researchers, whereas others take the opposite view. To accountforthesedivergentresults,threeexplanationsare suggested.

First, divergences could be attributed to recruitment:

transientsensorimotordeprivationinducedbyimmobiliza- tionbecauseoftrauma, orthatartificially reproducedfor experimental reasons, might not produce similar effects onbrainplasticity.Second,durationofimmobilizationmay playarole,sincedifferentmechanismsoccurdependingon whether immobilization is brief (i.e., 24hours) or longer (i.e.,aweek, amonthor evenseveralmonths).Third,as shown byIsaac [46], theefficacy of motor imagery train- ingdepends onindividual’s imagery ability: subjects with poorimageryabilitycouldfinditdifficulttoapplyandfol- lowinstructionsprovidedbytheexperimenters.Inthefive reviewedstudies,3assessedtheimageryabilityofthepar- ticipants andthose withpoor imageryability were either notincludedinthestudyorweretrainedtoenhancetheir ability[4,19,61].

In conclusion, though some divergent results are observed, using motor simulation in addition to tradi- tionalphysicalrehabilitationappearsofinterest,potentially allowingindividualswithtransientsensorimotordeprivation tooptimizerecoveryandsafereturntodailylifeactivities.

However,more clinicalresearchin theﬁeldoforthopedic traumaisnecessaryinordertooptimizetheimpactofmotor simulation.

Appliedperspectives

Formotorimagery,principlesasdescribedbythePhysical, Environment,Task,Timing,Learning,Emotion,Perspective (PETTLEP)[43]orMIMS(theIntegratedMotorImageryModel appliedtoSports) [34]model couldbe helpful. Thus,the

practitioner should be awareof what an imaginedaction consists of and how the subject imagines it, in order to designatailor-madeprogramforeachsubject:

• image generation (i.e. does the individual imagine an actionwithopenorclosedeyes?);

• postureoftheimager (i.e.istheindividuallyingdown, sittingorstanding?);

• visualperspectiveandviewingangle(i.e.doestheimager useaninternalorexternal,oracombinationvisualper- spective,andwhichviewingangleisfavored?);

• imagingmodalities(i.e.whichsensesareinvolvedduring mental imaging?Visual modality? Kinesthetic? Auditory?

Olfactive?Tasting?Acombination?);

• imagingspeed (i.e. are the temporalcharacteristics of theperformedmovereﬂectedbytheimagedmove?Does theimager create themove at thesame speed asthat ofactualperformance,oristheimagedmovefaster,or slower?);

• agency(i.e.doimagerspicture themselves,or another, oranavatar?);

• including physiological and emotional elements to imagery (i.e. should practitioners feature emotional content[fear,frustration,enthusiasm]and/orphysiologi- calresponses[tremors,musculartension,fatigue,sweaty palms]?).

Likewise,inordertoincreasetheefficiencyofobserva- tion,usersmustfollowanumberofprinciplesderivedfrom sport psychology and neuroscience findings [41,42]. Users willmakedecisionsbasedontheseresultsandoneachspe- cific situation and include these elements to observation sessions.Forexample:

• observationcontent(i.e.isitbettertoobserveamodel duringthelearningphase,performingcorrectexecutions, orperformingwithmistakes?);

• visualperspectiveandviewingangle(i.e.isitmoreben- eﬁcialtoobservea model froman internal or external perspective? And, in the case of an external perspective,from which viewingangle? Should the model face theobserver,orshouldtheybewatchedfrombehind?);

• natureof instructions provided prior to an observation session (i.e. observe a movement, no instruction provided?Observe a move aiming to replicate it later on?

Observeamoveinordertoimage/mimeitlater?Combin- ingobservationofamoveandconcomitantmimes?);

• agency(i.e.istheobserverwatchinghisownperformance orthatofanother?);

• expertiseleveloftheobservedmodel(i.e.doesthemodel possessthesamelevelofmotorexpertiseastheobserver?

Istheirlevellowerorhigher?);

• observationcontext(i.e.shouldthecontextbedeﬁnedor not?Ifcontextisspeciﬁed,shoulditfeatureastake?).

Inordertomaximizetheefﬁcacyofthemovementver- balizationprocess,verbalsequences,actionverbsandkey wordsshouldbemeaningfultothesubject.Thus,thesub- jectisencouragedtogenerate,withthepractitioner’shelp, theirownverbalizations,whichmayevolvewithtime.

Lastly,regardingamimedtaskwhichgeneratesperiph- eralsensations,anumberofquestionsemerge:shouldthe

(8)

Please cite this article in press as: Calmels C. Beyond Jeannerod’s motor simulation theory: An approach subjectmime theentire movement, oronly parts? Should

mimedsequencesbeperformedusinghands,upperorlower half-body, orthewholebody?Shouldan objectrelatedto activitiesofdailylivingbeused,ornot?

Thus, apersonalized supportprogram,basedonmotor simulation, is designed jointly with the injured subject.

Thisprogramtakesthesubject’sindividualityintoaccount, notably personal history, sensitivity and personal experience,notforgettingcontext.Morespeciﬁcally,choosingto useonesimulationtechniqueoveranotherwilldependon the individual’s mental resources and their mastery, and the purposeandscope of motorsimulation. Forinstance, these techniques can be used alone or in combination, withorwithoutmimedsequences,e.g.movementimagery;

movement imagery+movement verbalization; movement observation;ormovementobservation+mimedmovement.

Finally,efﬁcacydependsheavilyonthesubject’svoluntary participation;in thecase of forcedparticipation, optimal resultsarenottobeexpected[20].Currentlyimplemented withinjuredathletesonanationalcampusinFrance,this approach with motor simulation at its core has yielded promisingresults.

Conclusion

Associating motor simulation withmore traditional physi- calrehabilitationisapromisingavenueinsportandclinical settings. For example, in sports, motor simulation could help prevent injuries or allow athletes tomaximize their movement potential either by removing erroneous move- mentpatternsthatpreventthemfromattainingexcellence, orbycreatingnewones.

Disclosureof interest

Theauthordeclaresthathehasnocompetinginterest.

References

[1]AvanzinoL,BassolinoM,PozzoT,BoveM.Use-dependenthemi- sphericbalance.JNeurosci2011;31:3423—8.

[2]Aziz-ZadehL, WilsonSM, Rizzolatti G, Iacoboni M. Congru- ent embodied representations for visually presented actions and linguistic phrases describing actions. Curr Biol2006;16:1818—23.

[3]Bassolino M, Bove M, Jacono M, Fadiga L, Pozzo T.

Functional effect of short-term immobilization: kinematic changes and recovery on reaching-to-grasp. Neuroscience 2012;215:127—34.

[4]BassolinoM,CampanellaM,BoveM,PozzoT,FadigaL.Train- ingthemotorcortexbyobservingtheactionsofothersduring immobilization.CerebCortex2014;24:3268—76.

[5]Bidet-IldeiC,MeugnotA,BeauprezSA,GimenesM,Toussaint L.Short-termupperlimbimmobilizationaffectsaction-word understanding.JExpPsycholLearn2017;43:1129—39.

[6]Blakemore SJ, Frith CD, Wolpert DM. The cerebellum is involvedinpredictingthesensoryconsequencesofaction.Neu- roreport2001;12:1879—84.

[7]BlockH,BastianA,CelnikP.Virtuallesionofangulargyrusdis- ruptstherelationshipbetweenvisuoproprioceptiveweighting andrealignment.JCognNeurosci2013;25:636—48.

[8]BorichMR,BrodieSM,GrayWA,IontaS,BoydLA.Understanding theroleoftheprimarysomatosensorycortex:opportunitiesfor rehabilitation.Neuropsychologia2015;79:246—55.

[9]Bosbach S, Cole J,Prinz W, KnoblichG. Inferring another’s expectationfromaction:theroleofperipheralsensation.Nat Neurosci2005;8:1295—7.

[10]Burianova H, Marstaller L, Sowman P, Tesan G, Rich AN, Williams M, et al. Multimodal functional imaging of motor imageryusinganovelparadigm.Neuroimage2013;71:50—8.

[11]Burianova H, Sowman PF, Marstaller L, Rich AN, Williams MA,SavageG,etal. Adaptivemotorimagery:amultimodal studyofimmobilization-inducedbrainplasticity.CerebCortex 2016;26:1072—80.

[12]Calmels C, Pichon S, Grezes J. Can we simulate an action that we temporarily cannot perform? Neurophysiol Clin 2014;44:433—45.

[13]Calvo-MerinoB,GlaserDE,GrèzesJ,PassinghamRE,Haggard P.Actionobservationandacquiredmotorskills:anfMRIstudy withexpertdancers.CerebCortex2005;15:1243—9.

[14]Calvo-MerinoB,GrèzesJ,GlaserDE,PassinghamRE,Haggard P.Seeingordoing?Inﬂuenceofvisualandmotorfamiliarityin actionobservation.CurrBiol2006;16:1905—10.

[15]CaspersS,ZillesK,LairdA,EickhoffSB.ALEmeta-analysisof actionobservationandimitationinthehumanbrain.Neuroim- age2010;50:1148—67.

[16]Chen SX, Kim AN, Peters AJ, KomiyamaT. Subtype-speciﬁc plasticityofinhibitorycircuitsinmotorcortexduringmotor learning.NatNeurosci2015;18:1109—15.

[17]Cichon J, Gan WB. Branch-speciﬁc dendritic Ca(2+) spikes causepersistentsynapticplasticity.Nature2015;520:180—5.

[18]Cohen LG, Celnik P, Pascual-Leone P, Corwell B, Falz L, DambrosiaJ,etal.Functionalrelevanceofcross-modalplas- ticityinblindhuman.Nature1997;389:180—3.

[19]CrewsRT,KamenG.Motor-evokedpotentialsfollowingimagery andlimbdisuse.IntJNeuroscience2006;116:639—51.

[20]DalyJM,BrewerBW,VanRaalteJL,PetitpasAJ,SklarJH.Emo- tionaladjustementetadherencetorehabilitationfollowinga kneesurgery.JSportRehabil1995;4:23—30.

[21]DavareM,MontagueK,OlivierE,RothwellJC,LemonRN.Ven- tral premotor to primary motor cortical interactionsduring object-drivengraspinhumans.Cortex2009;45:1050—7.

[22]Desmurget M, Sirigu A. A parietal-premotor network for movement intention and motor awareness. TrendsCogn Sci 2009;13:411—9.

[23]DesmurgetM,ReillyKT,RichardN,SzathmariA,MottoleseC, SiriguA.Movementintentionafterparietalcortexstimulation inhumans.Science2009;324:811—3.

[24]ElbertT,PantevC,WienbruchC,RockstrohB,TaubE.Increased corticalrepresentationoftheﬁngersofthelefthandinstring players.Science1995;270:305—7.

[25]Elias JW, Wagster MV. Developing context and background underlyingcognitiveintervention/trainingstudiesinolderpop- ulations.JGerontolBPsycholSciSocSci2007;62S1:5—10.

[26]EspinosaJS,StrykerMP.Developmentandplasticityofthepri- maryvisualcortex.Neuron2012;75:230—49.

[27]Flor H, Elbert T, Knecht S, Wienbruch C, Pantev C, Bir- baumerN,etal.Phantom-limbpainasaperceptualcorrelate ofcorticalreorganization following armamputation.Nature 1995;375:482—4.

[28]FlorH,NikolajsenL,JensenTS.Phantomlimbpain:acaseof maladaptiveCNSplasticity?NatRevNeurosci2006;7:873—81.

[29]FrithC,DolanRJ.Brainmechanismsassociatedwithtop-down processesinperception.PhilosTRoySocB1997;352:1221—30.

[30]FurlanL,ConfortoAB,CohenLG,SterrA.Upperimmobiliza- tion: aneural plasticitymodel withrelevance topoststroke motorrehabilitation.NeuralPlast2016;2016:8176217.

[31]GazzolaV,KeysersC.Theobservationandexecutionofactions sharemotorandsomatosensoryvoxelsinalltestedsubjects:

(9)

ARTICLE IN PRESS

8 C.Calmels

single-subjectanalysesofunsmoothedfMRIdata.CerebCortex 2009;13:1239—55.

[32]GrezesJ,DecetyJ.Functionalanatomyofexecution,mental simulation,observation and verbgenerationofaction.Hum BrainMapp2001;12:1—19.

[33]Grush R. The emulation theory of representation: motor control, imagery,and perception. Behav Brain Sci 2004;27:

377—96.

[34]GuillotA,ColletC.Constructionofthemotorimageryintegra- tivemodelinsport:areviewandtheoreticalinvestigationof motorimageryuse.IntRevSportExer2008;1:31—44.

[35]GuillotA,MoschbergerK,ColletC.Couplingmovementwith imagery as a new perspective for motor imagery practice.

BehavBrainFunct2013;9:8.

[36]Hardwick RM, Caspers S, Eickhoff SB, Swinnen SP. Neural correlates of action: comparing meta-analyses of imagery, observation,andexecution.NeurosciBiobehav Rev2018;94:

31—44.

[37]HaukO, Johnsrude I,PulvermullerF.Somatotopicrepresen- tationofactionwordsinhumanmotorandpremotorcortex.

Neuron2004;41:301—7.

[38]Hayashi-TakagiA,YagishitaS,NakamuraM,ShiraiF,WuYiI, LoshbaughAL,etal. Labellingandopticalerasureofsynap- tic memory traces in the motor cortex. Nature 2015;525:

333—8.

[39]HetuS, Gregoire M,SaimpontA, CollMP, Eugene F,Michon PE,etal.Theneuralnetworkofmotorimagery:anALEmeta- analysis.NeurosciBiobehavRev2013;37:930—49.

[40]HoffstaedterF,GrefkesC,ZillesK,EickhoffSB.The‘‘What’’

and ‘‘When’’ of self-initiated movements. Cereb Cortex 2013;23:520—30.

[41]HolmesP,CalmelsC.Aneuroscientiﬁcreviewofimageryand observationuseinsport.JMotorBehav2008;40:433—45.

[42]HolmesP,CalmelsC.Mentalpractice:neuroscientiﬁcsupport foranewapproach.In:CollinsD,AbbottA,RichardsH,editors.Performancepsychologyforphysicalchallenge.London:

Elsevier;2011.p.231—44.

[43]Holmes PS, Collins DJ. The PETTLEP approach to motor imagery:afunctional equivalencemodelforsport psycholo- gists.JApplSportPsychol2011;13:60—83.

[44]Hoshi E, Tanji J. Distinctions between dorsal and ventral premotorareas:anatomicalconnectivityandfunctionalprop- erties.CurrOpinNeurobiol2007;17:234—42.

[45]Huber R,GhilardiMF, MassiminiM,Ferrarelli F,Riedner BA, PetersonMJ,etal.Armimmobilizationcausescorticalplastic changesand locallydecreasessleepslowwaveactivity. Nat Neurosci2006;9:1169—76.

[46]IsaacAR.Mentalpractice-doesitworkintheﬁeld?SportPsy- chol1992;6:192—8.

[47]Jeannerod M. The representing brain: neural correlates of motor intention and imagery. Behav Brain Sci 1994;17:182—202.

[48]JeannerodM.Neuralsimulationofaction:aunifyingmecha- nismformotorcognition.NeuroImage2001;14:S103—9.

[49]Jeannerod M. Motor cognition: what actions tell the self.

Oxford:OxfordUniversityPress;2006.

[50]KawatoM.Internalmodelsformodelsformotorcontroland trajectoryplanning.CurrOpinNeurobiol1999;9:718—27.

[51]KeysersC,KaasJH,GazzolaV.Somatosensationinsocialper- ception.NatRevNeurosci2010;11:417—28.

[52]Ko M, AnJ, Bandukwala HS, Chavez L, Aijo T, Pastor WA, etal.ModulationofTET2expressionand5-methylcytosineoxi- dation bythe CXXCdomain proteinIDAX. Nature 2013;497:

122—6.

[53]LangerN,HanggiJ,MullerNA,SimmenHP,JanckeL.Effects of immobilization on brain plasticity. Neurology 2012;78:

182—8.

[54]LissekS, Wilimzig C,Stude P, PlegerB, Kalisch T, Maier C, et al.Immobilization impairs tactileperception and shrinks somatosensorycorticalmaps.CurrBiol2009;19:837—42.

[55]LyonsIM, Mattarella-MickeA, CieslakM,NusbaumHC, Small SL, Beilock SL. The role of personal experience in the neural processing of action-related language. Brain Lang 2010;112:214—22.

[56]MarusicU,GrosprêtreS,ParavlicA,KovacS,PisotR,Taube W.Motorimageryduringactionobservationoflocomotortasks improvesrehabilitationoutcomeinolderadultsaftertotalhip arthroplasty.NeuralPlast2018;2018:5651391.

[57]MeugnotA,AgbanglaNF,AlmecijaY,ToussaintL.Motorimagery practicemaycompensate for theslowdownofsensorimotor processesinduced byshort-termupper-limb immobilization.

PsycholRes2015;79:489—99.

[58]MeugnotA,AgbanglaNF,ToussaintL.Selectiveimpairmentof sensorimotorrepresentationsfollowingshort-termupper-limb immobilization.QJExpPsychol2016;69:1842—50.

[59]Meugnot A, Toussaint L. Functional plasticity of sensorimotor representations following short-term immobilization of the dominant versus non-dominant hands. Acta Psychol 2015;155:51—6.

[60]MoiselloC,BoveM,HuberR,AbbruzzeseG,BattagliaF,Tononi G,etal.Short-termlimbimmobilizationaffectsmotorperfor- mance.JMotorBehav2008;40:165—76.

[61]MoukarzelM,DiRienzoF,LahoudJC,HoyekF,ColletC,Guil- lotA,etal.Thetherapeuticroleofmotorimageryduringthe acutephaseaftertotalkneearthroplasty:apilotstudy.Disabil Rehabil2017;24:1—8.

[62]NachevP,KennardC,HusainM.Functionalroleofthesupple- mentaryandpre-supplementarymotorareas.NatRevNeurosci 2008;9:856—69.

[63]NaitoE, KochiyamaT, KitadaR,NakamuraS,Matsumura M, YonekuraY,et al.Internallysimulatedmovement sensations duringmotorimageryactivatecorticalmotor areasand the cerebellum.JNeurosci2002;22:3683—91.

[64]Olson CR,ColbyCL.Spatialcognition.In: SquireL, BergD, BloomFE,Lac S,Ghosh A, SpitzeNC,editors.Fundamental neuroscience.4theditionLondon:Elsevier;2013.p.969—88.

[65]Parsons LM. Temporal and kinematic properties of motor behaviouralreﬂectedinmentallysimulatedaction.JExpPsy- cholHuman1994;20:709—30.

[66]Pascual-LeoneA, AmediA,FregniF,MerabetLB.Theplastic humanbraincortex.AnnRevNeurosci2005;28:377—401.

[67]Payton OD, Su S, Meydrech EF. Abductor digiti quinti shuf- ﬂeboard: a study in motor learning. Arch Phys Med Rehab 1976;57:169—74.

[68]Peran P,Demonet JF, Cherubini A, Carbebat D,Caltagirone C,Sabatini U. Mental representationsof action:the neural correlates of the verbal and motor components. Brain Res 2010;1328:89—103.

[69]RushworthMF,NixonPD,WadeDT,RenowdenS,Passingham RE.Thelefthemisphereandtheselectionoflearnedactions.

Neuropsychologia1998;36:11—24.

[70]SiriguA,DuhamelJR,CohenL,PillonB,DuboisB,AgidY.The mentalrepresentationofhandmovementsafterparietalcor- texdamage.Science1996;273:1564—8.

[71]Sterr A, Muller MM, Elbert T, Rockstroh B, Pantev C, Taub E. Changed perceptions in Braille readers. Nature 1998;391:134—5.

[72]SynofzikM,LindnerA, TheirP.Thecerebellumupdatespre- dictionsaboutthevisualconsequencesofone’sbehavior.Curr Biol2008;18:814—8.

[73]ToniI,ThoenissenD,ZillesK.Movementpreparationandmotor intention.Neuroimage2001;14:S110—7.

[74]ToussaintL,MeugnotA.Short-limbimmobilizationaffectscog- nitivemotorprocess.JExpPsycholLearn2013;39:623—32.

(10)

Please cite this article in press as: Calmels C. Beyond Jeannerod’s motor simulation theory: An approach [75] Valchev N, Gazzola V, Avenanti A, Keysers C. Primary

somatosensory contribution to action observation brain activity-combiningfMRIandcTBS.SCAN2016;11:1205—17.

[76] WeibullA,FlondellM,RosenB,BjorkmanA.Cerebralandclin- icaleffectsofshort-termhandimmobilisation.EurJNeurosci 2011;33:699—704.

[77]Wise SP, Murray EA. Arbitrary associations between antecedentsandactions.TrendsNeurosci2000;23:271—6.

[78]WolpertDM,MiallRC,KawatoM.Internalmodelsinthecere- bellum.TrendsCognSci1998;2:338—47.