Human upright posture control models based on multisensory inputs; in fast and slow dynamics

ContentslistsavailableatScienceDirect

Neuroscience

Research

j ou rn a l h o m epa g e : w w w . e l s e v i e r . c o m / l o c a t e / n e u r e s

Review

article

Human

upright

posture

control

models

based

on

multisensory

inputs;

in

fast

and

slow

dynamics

Ryosuke

Chiba

_,

_Kaoru

_Takakusaki

_,

_Jun

_Ota

_,

_Arito

_Yozu

_,

_Nobuhiko

_Haga

a_{ResearchCenterforBrainFunctionandMedicalEngineering,AsahikawaMedicalUniversity,Japan} b_{ResearchintoArtifacts,CenterforEngineering(RACE),TheUniversityofTokyo,Japan}

c_{DepartmentofRehabilitationMedicine,GraduateSchoolofMedicine,TheUniversityofTokyo,Japan}

a

r

t

i

c

l

e

i

n

f

o

Articlehistory:

Received15September2015 Receivedinrevisedform 29November2015 Accepted1December2015 Availableonline30December2015 Keywords:

Posturecontrol Multisensoryintegration Long-termalteration Bodyrepresentationinbrain Balance

a

b

s

t

r

a

c

t

Posturecontroltomaintainanuprightstanceisoneofthemostimportantandbasicrequirementsinthe dailylifeofhumans.Thesensoryinputsinvolvedinposturecontrolincludevisualandvestibularinputs, aswellasproprioceptiveandtactilesomatosensoryinputs.Thesemultisensoryinputsareintegrated torepresentthebodystate(bodyschema);thisisthenutilizedinthebraintogeneratethemotion. Changesinthemultisensoryinputsresultinposturalalterations(fastdynamics),aswellaslong-term alterationsinmultisensoryintegrationandposturecontrolitself(slowdynamics).Inthisreview,we discussthefastandslowdynamics,withafocusonmultisensoryintegrationincludinganintroductionof ourstudytoinvestigate“internalforcecontrol”withmultisensoryintegration-evokedposturealteration. Wefoundthatthestudyoftheslowdynamicsislaggingcomparedtothatoffastdynamics,suchthat ourunderstandingoflong-termalterationsisinsufficienttorevealtheunderlyingmechanismsandto proposesuitablemodels.Additionalstudiesinvestigatingslowdynamicsarerequiredtoexpandour knowledgeofthisarea,whichwouldsupportthephysicaltrainingandrehabilitationofelderlyand impairedpersons.

©2015TheAuthors.PublishedbyElsevierIrelandLtd.ThisisanopenaccessarticleundertheCC BY-NC-NDlicense(http://creativecommons.org/licenses/by-nc-nd/4.0/).

Contents

1. Introduction...97

2. Contributionsofindividualsensoryinputs ... 97

2.1. Visualsensoryinput...97

2.2. Vestibularsensoryinput...97

2.3. Proprioceptivesomatosensoryinput...97

2.4. Tactilesomatosensoryinput...98

3. Experimentalapproachformultisensoryintegration...98

3.1. Experimentalmethodforinvestigationofmultisensoryintegration...98

3.2. Amultisensoryintegratedcontrolmodel(Chibaetal.,2013)...98

4. Modelingandsimulationofmultisensoryintegration...100

4.1. Posturecontrolmodel...100

4.2. Bodymodel...100

4.3. Multisensoryintegrationmodel...101

5. Long-termalterationinposture...101

5.1. Agingandposturedisorders...101

∗ Correspondingauthor.Tel.:+81166682888.

E-mailaddress:[email protected](R.Chiba).

http://dx.doi.org/10.1016/j.neures.2015.12.002

0168-0102/©2015TheAuthors.PublishedbyElsevierIrelandLtd.ThisisanopenaccessarticleundertheCCBY-NC-NDlicense( http://creativecommons.org/licenses/by-nc-nd/4.0/).

5.2. Learningandrehabilitation...102

6. Conclusion...102

Acknowledgment ... 102

References...102

1. Introduction

The“posture control” neededto maintainanupright stance is a very important and basic requirement in daily humanlife (Wallmann, 2009). Adequate posture control requires the sen-sory and central nervous systems (CNS), and for the human bodytofunctionappropriatelyagainstgravityand environmen-tal forces (Runge et al., 1999). Posture control requires visual andvestibularinputs,aswellasbothproprioceptiveand tactile somatosensoryinputs,tocontrolposture-regulatingmusclesinthe wholebody,especiallyinthelowerlimbsandtrunk(Horakand Macpherson,1996).Thus,theCNSneedstocontrolmultiple mus-clessimultaneouslybasedoncorrespondingmultisensoryinputs. BecauseofthecomplexityoftheCNS,themechanismbywhich thisregulationoccursis stillunknowndespiteresearchers’best efforts.

Changesin multisensory inputs elicit immediate changes in posturecorrespondingtothesensoryinformation.These immedi-atealterations,referredtoasthefastdynamicsintheCNS,occur whenthebrainpredicts sensoryinputs andcorrectsthebody’s motionbasedontheerrorbetweenthepredictedandactual sen-soryinputs.Inthisprocess,themultisensoryinputsareintegrated torepresentthebodystate(bodyschema);thisisthenutilizedby thebraintogeneratemotion.Thoughaninterestingphenomenon initsownright,howthebrainintegratesmultisensoryinputsin theCNSis stillunknown.The“weightandreweight”conceptis animportantframeworkfordescribingtheabilityofmultisensory integrationtocalculatebodystate,includingparameterssuchas thecenterofmass(CoM)andheading.Ingeneral,thesensoryinputs include signal noise, and the multisensory “integrator” decides which inputs are reliable,and to whatdegree, as the “weight-ing”process.ThentheCNSreweights(changestheweights)the inputs accordingto theinternal and external conditions ofthe bodyandaroundthebodysuchaslightlevel,bodilyacceleration, etc.Thedetailsof thisconceptwillbeintroducedinthefourth section.

Anotherinterestingaspectofposturecontrolisthat multisen-sory integrationand the posture controller undergo long-term alterationsforavarietyofreasons,includingagingandlearning. Theselong-termalterationsarereferredtoastheslowdynamicsin theCNS.

Inthisreview,wewillfocusonbothfastandslowdynamics ofmultisensoryintegrationinposturecontrol.Inthissection,we introducedthebackgroundofposturecontrolstudiesand summa-rizedthecurrentgapsinourunderstanding.Next,wewillintroduce thedetailedsensoryinputsrelatedtoposture.Inthethirdsection, wewillfollowthatupbyintroducingstudiesinvestigatingthe rela-tionshipbetweenmultisensoryinputsandposturalalterations(fast dynamics),includingourstudyinvestigating“internalforce con-trol”toevokeposturalalterationswhenthemultisensoryinputsare changed.Inthefourthsection,wewillintroducetheavailable mod-elingand simulationapproachestoexplainpostural alterations, includingtheresultsofthosestudiesintroducedinthethird sec-tion.Finally,wewillalsodescribethefindingsofstudiesexamining long-termalterationsinpostureinrelationtoaging,learning,and rehabilitation,whichisafieldofresearchthathasincreasinglybeen garneringattention.Inclosing,wewillsummarizethekeypoints ofthisreview.

2. Contributionsofindividualsensoryinputs

Asmentionedabove,visual,vestibular,proprioceptive,and tac-tilesensoryinputsaretheprimarycontributorstothemaintenance ofuprightposture.Severalstudieshaveexaminedthecontribution ofeachsensoryinputtoposturemaintenanceusingavarietyof experimentalmethodologies.Duetothelargenumberofstudies relatedtothesesensorysystems,weintroduceonlyreviewpapers withexperimentalmethodsinthissection.Foreachsensoryinput, wewilldescribethewaystheinputcanbealtered,andtheresulting changesinposturalcontrol.

2.1. Visualsensoryinput

Visualsensoryinputcanbeeasilymodifiedbyopeningand clos-ingtheeyes,whichresultsinchangesinposturalstability(Horak andMacpherson,1996).Anothermethodofmodifyingvisual sen-soryinputis byprojectingscenesontoa moving screen.Wade andJones(1997)havereviewedtheeffectsofvisualstimulation onposturecontrol.Recentstudiesofvisualalterationdiscussthe contributionofvisioninmultisensoryreweightingandthehuman posturecontrolmodel.Suchstudieswillbeintroducedinthethird section.

2.2. Vestibularsensoryinput

Vestibularinputcannotbeeasilychanged,becauseitcannotbe consciouslycontrolled.Vestibularinformationtransmittedtothe brainprovidestheheadorientationrelative togravity(vertical) and,inconjunctionwithneckproprioception,makesitpossibleto estimatebodyorientation.Theheadorientation,inturn,affectseye movements.Changesinvestibularinputscanbeelicitedby elec-tricalstimulation,includinggalvanicvestibularstimulation(GVS). Contributionofthevestibularinputtoposturalcontrolhasalso beeninvestigatedbystudyingpatientswithdisordersthatresultin lossofvestibularfunction.Vestibularmechanismsandtheir rela-tionshiptoposturecontrol,includingheadmovementandwhole bodysway,arereviewedbyGreenandAngelaki(2010).Fora gen-eraloverview of vestibularfunction, please refer tothereview paper(Forbesetal.,2015).

2.3. Proprioceptivesomatosensoryinput

Proprioception,likevestibularsensation,isdifficultto modu-latein experiments.Proprioceptiongivesus informationonthe staticanddynamiccomponentsofjointposition/orientation;for posturecontrol,informationaboutlowerlimbandankle orienta-tionisespeciallyimportant.Thissensoryinputcanbedecreased experimentallybyusingatiltingplatformtoadjusttheanklejoint anglewithareferringbodyswayofthesubjects.Thereduced pro-prioceptiveinputduetothetiltingplatformevokesaposturalsway betterthanthatrelatedtoanormal,fixedplatform. Propriocep-tioncanbealteredthroughthevibrationofmusclessuchasthe soleus.Proprioceptivealterationalonedoesnotevokebodysway, butratherrequiresalterationsofvisual,vestibular,ortactile sen-sations as well.Allumet al.(1998) have reviewed studiesthat investigated proprioceptivecontributions usingthis method.To findthecontributionoftheproprioception,neckandanklemuscles

werestimulatedbyvibrationtoanalyzetherelationshipbetween neckandankleproprioception(Kavounoudiasetal.,1999). 2.4. Tactilesomatosensoryinput

Intheuprightstance,tactilecuesareconveyedtothebrainfrom thesolesofthefeet.Theinformationisbasedonthepressureon thesole,andthecenterofpressure(CoP)isaveryimportantcuefor themaintenanceofastablestance.Sole-derivedtactileinputcan beeasilychangedinexperimentsbycoveringthefloorwithfoam orsimilarsoftmaterial.Anotherstudynotedthatalighttouch any-whereonasubject’sbodycanattenuateposturalsway,evenwhen thetouchisnotphysicalbutelectricalstimulation(e.g.,ofafinger) (Shimaetal.,2013).Theeffectsofalighttoucharevery interest-ing,butthemechanismbywhichtheyworkisstillunknown.Some datasuggestthemechanismmayrelatetotheattention,whichis nearconceptto“reweighting”,dedicatedbythebraintothe sen-soryinputsmaintainingposture(WoollacottandShumway-Cook, 2002;Soto-Faracoetal.,2004).

3. Experimentalapproachformultisensoryintegration

Inthissection,wediscussstudiesthatexperimentally inves-tigated the effects of multisensory input alterations. We also introduceourstudyasanexampleoftheexperimental investiga-tion.

3.1. Experimentalmethodforinvestigationofmultisensory integration

Visual and proprioceptive sensory inputs can be changed simultaneouslyusingtheSensoryOrganizationTest(SOT),which examineshowsubjectsutilizecombinationsofthosesensory feed-backtomaintainanuprightstance(Fig.1)(Nashneretal.,1982; Nashnerand Peters,1990).In theSOT, ascreen in front ofthe subjectandaplatformonwhichthesubjectstandscanbothtilt accordingtothebodyswayofsubjects.Inonestudyfocusingon fastdynamics,brainfunctionwasmeasuredusingfunctional near-infraredspectroscopic (fNIRS)imaging to identifythe region(s) associatedwithsensoryinputprocessingduringalterationsofthe conditionsoftheSOT(Karimetal.,2013).Thisstudyfoundbilateral activationintemporal–parietalareasinconditionsofvisualand proprioceptiveinhibition,inwhichthistestandfNIRSpotentially utilizedthemeasurementsofthebrainfunctionforquietstanding. Thistestisutilizedinmanystudiestoinvestigateaging,learning,

(a)

(b

)

(c

)

(d)

Fig.1.Sensoryorganizationtest:thesubjects’CoP,CoM,andtheother measure-mentsarecalculatedusinga(a)fixedscreeninfrontofsubjectswithafixedplatform, (b)rotatingscreenreferencedsubjects’bodysway(visualalteration),(c)rotating platformreferencedsubjects’bodysway(proprioceptionalteration),(d)rotating screenandplatformreferencedsubjects’bodysway(visualandproprioception alteration).Theposturecontrolisscoredbasedonthemeasurements.

multisensory-integratedmodeling,andposturecontrolmodeling, allofwhicharediscussedinthefollowingsection.

Underconditionsofvisualstimulationbyprojectionofa mov-ingpattern,andproprioceptiveperturbationbyvibrationtothe bilateralAchillestendons,theposturalmaintenanceofyoungadult subjects wasperturbed, while thedisturbance was rescuedby galvanicvestibularstimulation(Eikemaetal.,2014).Thisresult suggeststhatvestibularinputhaslessofaneffectthantheother sensoryinputs,since thesubjectscouldnotholdtheirpostures withoutartificialvestibularstimulation.Asimilarexperimentused GVStoaltervestibularinput,screentransitiontoaltervisualinput, andvibrationtoalterproprioception(Hwangetal.,2014).Inthis study,whenoneofthesensoryinputswasdisturbed,theweights oftheothersensoryinputs increasedaspera feedbackcontrol model.Thedataindicatedthatthesensoryinputswereclearly inde-pendent,andthattheweightofeachsense’sreliabilitywouldbe requiredforcorrectmodeling.Thecombinedeffectofthevisual andvestibularsensoryinputsonposturecontrolhasbeenreviewed elsewhere(CoelhoandBalaban,2015).

Asmentioned,alighttouchincreasedposturalstabilityofthe uprightstanceinthepresenceofvisualdisturbances(Jekaetal., 2000;Oieetal.,2001;Allisonetal.,2006).Theeffectisstillobserved inpatientswithvestibularloss(Horak,2009).Thesedatasuggested that thecontributionof each of themultisensory inputs is not integratedlinearly,becausealighttouchdidnotincrease informa-tionofpostureparameters(e.g.,CoP).Rather,thisinputpossibly inducethereweightingofinformation fromeachsensoryinput. Moreover,posturecanbemaintainedbylighttouchwhethergiven orreceived,andeveninthepresenceofvestibulardisturbances (Chibaetal.,2013).Inaddition,proprioceptivedisturbanceusing afoamsurfacewithvisualstimulationstillincreasesbodysway comparedtoafixedsurfacewithnormalvision(Bronstein,1986). However,experimentscomparingpostureswitheyesclosedona fixedsurfaceversusafoamsurfacefoundnosignificantdifferences ofposturecontrolbetweenonthefixedandfoamsurfaces(Creath etal.,2005).

Finally,onestudynotedthatvibrationofthesoleusthatelicited adisturbanceofanklejointorientationunderaneyes-closed con-dition induced a variable postural body sway that varied with thevibration frequency (Capicikovaet al., 2006).However,the multisensoryreweightingwasunclearinthisstudy,making inter-pretationdifficult.

3.2. Amultisensoryintegratedcontrolmodel(Chibaetal.,2013) Here,asoneexampleoffastdynamicsstudies,weintroduceour studyexaminingchangesinposturebyalterationsinmultisensory inputsinthissubsection.Thisworkwasdonetoshowthatmuscular tonuscontrolandtactileinputsfromnotonlythefoot,butanypart ofthebody,playimportantrolesinposturalstability.

Forthisstudy,werecruited12malesubjectsintheirtwenties withoutanyimpairments.WemeasuredtheCoPandmuscle activ-itiesasshowninFig.2.Weused8experimentalconditions,with differentcombinationsofmodulatedvisual,vestibular,andtactile stimulation.Visionwasoccluded byclosingtheeyes,vestibular sensationwasdisturbedbyacalorictestthatupsetthevestibular systemthroughintroductionofcoldwaterintotheleftear cav-ity,andtactilestimulationwasintroducedbythetouchofabody partbyanotherperson.Specifically,the8conditionswere: con-dition1,normal;condition2,visualocclusion; condition3,tactile stimulation;condition4,visualocclusionandtactilestimulation; condition5, vestibular disturbance; condition6, visual occlusion andvestibulardisturbance;condition7,vestibulardisturbanceand tactilestimulation;condition8,visualocclusion,vestibular disturb-ance,andtactilestimulation.Theparticipantswerejustrequiredto maintainanuprightstanceunderallconditions.

Table1

Muscleactivitiesineachcondition:Thecombinedchangeofvisualandvestibularinputsresultsinhighactivitiesinalmostallofthemuscles(condition6).Addinglight touchdoesnotdecreasetheseactivities,althoughtheposturerecovers(condition8).

Conditions 1 2 3 4 5 6 7 8 Soleus(left) 1.00 1.19 1.18 1.34 1.38 1.71 1.40 1.84 Soleus(right) 1.00 1.12 1.11 1.21 1.29 1.48 1.32 1.60 Tibialisanterior(l) 1.00 1.26 1.40 1.33 2.29 4.51 2.47 4.41 Tibialisanterior(r) 1.00 1.29 1.25 1.23 2.35 4.44 2.15 4.33 Quadricepsfemoris(l) 1.00 1.16 1.21 1.20 1.64 3.61 1.91 4.49 Quadricepsfemoris(r) 1.00 0.91 0.90 1.10 1.88 2.81 1.93 3.35 Hamstring(l) 1.00 1.44 1.64 1.68 1.44 1.75 1.62 1.94 Hamstring(r) 1.00 1.38 1.39 1.60 1.38 1.64 1.47 1.64 Erectorspinae(l) 1.00 1.03 1.04 1.05 1.12 1.16 1.14 1.23 Erectorspinae(r) 1.00 1.02 1.01 1.04 1.15 1.21 1.15 1.26

Fig.2. Measuredmuscles:werecordedtheelectromyographsofthetibialisanterior, soleus,quadricepsdemoris,hamstring,anderectorspinaeonbothsidesofthebody.

Outsideofthecontrolcondition,theirposturesweremarkedly

altered,asshowninFig.3.Fig.3(A)showstheposturalalteration

ineachcondition,andFig.3(B)showstheCoPineachcondition. Briefly, posturewas inclined byvestibular disturbancewithout

vision,thoughmostinterestingly,theinclinedposturewasrescued withalighttouch.However,themuscleactivitiesdidnotshowthe sameeffect.Table1showsthemuscleactivitiesineachcondition. Theactivitiesofthetibialisanteriorandquadricepsfemoriswere enhanced,asweretheothermusclesincondition6andcondition8, whencomparedwithcondition1(control).Inotherwords,to main-taininguprightposture,theactivitiesofmanymusclesarerequired, andnotonlytheanklemusclesoftheleftside(themainCoPside). Moreover,themuscleactivationsincondition8donotdiffermuch fromthoseincondition6,althoughpostureisrecoveredbyleaning totheleft(thedisturbedvestibularside).Thisindicatesthatthe CNSsimultaneouslycontrolsthestiffnessofthehumanbodyas wellastheposturewhichmeansregulatingthejointangles.It fur-therindicatesthatitisimportanttomeasurethemuscleactivities ofthewholebodytoassessposturecontrol.

Anoverviewofthefeedbackmodelthatweproposeisshownin Fig.4.Theintegratedsensoryfeedbackwasinputintotwokindsof controllers,aposture-maintainingcontrollerforjointanglesanda stiffness-adjustingcontrollerformomentofjointinertia(referred toasan“internalforcecontroller”here).Astiffnesscontrolmodel in Winter etal. (1998)lacked accuracy,becausethemodel did notincludemultisensoryinputalterations.Weconsideredthatthe existenceofthisinternalforceshouldnotbeignoredwhen design-ingthisposturecontrolmodel,andthatthestiffnessofeachjointin thebodymodelshouldbeadjustable.Itseemsthatthiscontroller functionsinamannersimilartoanintermittentcontroller(Bottaro etal.,2008;Suzukietal.,2012;Nomuraetal.,2013),tocontrolthe torqueandthephaseofeachjoint.Futurestudiesexpandingonthis modelshouldinvestigatetherelationshipbetweenstiffnesscontrol andphasecontrol.

Fig.3. Posturealteration(A)andaverageCoP(B)bythemultisensoryinputschange:thesolevisualorvestibularchangesresultinlesseralterations(A(a)andB(condition 2,3)).Thecombinedchangeofvisualandvestibularinputsresultsinmarkedalterations(A(b)andB(condition6)).However,thelighttouchattheconditionresultsin recoverytonearnormalposture(A(b)andB(condition8)).

extensor muscles + flexor muscles -+ +

K+bs

θ

u

u

PID

control

Sensory

integrator

d

d

internal

controller

-Fig.4.Feedbackmodelswithstiffnesscontrolling:sensoryinputsarefeedbackto2 controllers,aPIDcontroller,whichprovidesthetorqueateachjointfortheposture maintenance,andaninternalforcecontroller,whichprovidesthestiffnessofeach joint.

Mulsensory Inputs

Posture control Body dynamics

(musculoskeletal)

Mulsensory integraon

Fig.5.Thesimplestconceptofbasicfeedbackcontrolwithmultisensory integra-tion:theposturecontrollerinducesmuscleactivitiesandthepostureisconfigured bythesemuscleactivities.Whenthebodyconfigurationchanges,sensoryinputsthe bodyandenvironmentarealtered.Thesensoryinputsintegrateintotheinformation thatisthenusedbytheposturecontroller.

4. Modelingandsimulationofmultisensoryintegration

ThefeedbackloopofposturecontroldescribedinFig.5canbe usedtounderstandmultisensoryinputmodeling,because multi-sensoryintegrationcanbeevaluatedcomputationallyusingthe bodymodelparameters.Theseparametersincludethenumberof segmentsthatequalthenumberofjointsordegreesoffreedom, segmentmass,segmentinertia,segmentlength,andothersinthe posturecontrolmodel.Thebodymodelparametersaredescribedin studiesthatinvestigatedposturealterationsbymeasuringmotion capturedata,forceplatedata,andEMGdata.Tounderstandthe multisensoryintegrationmodel,wefirstmustintroducethestudies oftheposturecontrolmodelandthebodymodel.

4.1. Posturecontrolmodel

Modelingthecontrollerhelpsusunderstandtheinternal mech-anismunderlyingthehumanuprightposture.Ifwecanestimate theinternalstateofthecontrollerbymodeling,itwouldbeuseful forrehabilitation,becausewecouldestimatehowmuch impair-mentcouldberecovered.Thebasicmodelofposturecontrolisthe sensoryfeedbackmodelthatisutilizedinmachinecontrol(foran investigationofthistheoryinthemuscleactivitiesincats,seeHe etal.(1991)).Thismodelisabletoexplainthemotionofpatients withbalancelossinrelationtothetimedelaysinthecontrolloop (VanderKooijetal.,1999;LockhartandTing,2007;Vetteetal., 2010;Lietal.,2012).Otherstudieshaveproposedafeedforward mechanismforpredictivecontrolorstrategyselection(Barin,1989; Wolpertetal.,1995;Fitzpatricket al.,1996; Gatevetal.,1999; MorassoandSanguineti,2002).Amodelwithbothfeedbackand/or feedforwardactivitywasproposedbyKuo(1995).The continu-ousfeedbackmodelsareproposedwithdeterminedparametersto accountfortheexperimentalresultsinseveralstudies(for exam-ple,musclestiffness(Morassoetal.,1999;MorassoandSanguineti, 2002)).Thefeedbackmodelsareextendedtothemodels includ-ingestimationofsensoryinputsthatisreconstructedduringtask (Diekmannetal.,2004).InHettichetal.(2011),Mahboobin(2007), andMergneretal.(2009),uprightstandingrobotsusingvestibular

Fig.6.Bodylinkmodels:theankleisthebasicjointutilizedinalllinkmodels. Thehip,knee,waist,toe,andneckjointsareincludedforadditionalmodels.These aremodeledastheinversedpendulumthatcanbemechanicallyformulatedin mathematics.

inputwereproposed,withmultiplejointsusingafeedback mech-anismsimilartothatinhumans,whichverifiedthestabilityofthe robotstance.Thesemodelswereconstructedbasedonaone-joint, invertedpendulum,whichwillbeintroducedinthenext subsec-tion.Afeedbackmodelwithatwo-jointbodymodelwasproposed inParketal.(2004)torepresentamorecomplexposturalalteration withthehipjoint.InHettichetal.(2014),theposturealterations werealsoobtainedusingafeedbackcontrollerwithatwo-joint bodymodel.

Accordingtoafeedbackcontrolmodel,humansmaystabilize postureinordertominimizemuscleactivities,ratherthanto mini-mizebodysway(Kiemeletal.,2011).Onestudyalsoconsidered ankle states (angleand torque)as the minimizedindex in the feedbackcontrolmodel(Quetal.,2009).Posturecontrolfeedback modelswithmultisensoryinputsarediscussedinanotherreview (Mergner, 2010).Finally,intermittentfeedback controlhasalso beenproposedtoaccountforposturecontrol(Bottaroetal.,2008; Suzukietal.,2012;Nomuraetal.,2013).

Currently,thecontrolmodelofhumanpostureremainstobe established;theproposedmodelsstilldependonthetasks,which include target posture,environment, adding force, and sensory inhibition. In addition, thesealso dependonthe target indices beingoptimizedinmotion.Thetargetindicesaremeasured prop-ertiesinexperiments,forexampleCoM,CoP,EMGs,andheading, thenthestudiestrytoconstructnecessaryandsufficientmodels toexplainthemeasuredproperties.Databaseandbigdata analy-sesmaybeneededtoresolvethisissuebecauseoftheextensive couplingbetweenthesedependentvariables.

4.2. Bodymodel

Itisalsoimportanttorepresentbodysegmentsmathematically andcomputationallyforsimulations.Thebasicmodelisthesingle invertedpendulummodel(e.g.,Peterka,2002;Maureretal.,2006). Thismodelhasonesegmentwithonejointattheankle,andno jointatthekneeorhip.However,thehipandthekneejointsplayan importantroleinmaintainingpostureforthetrunk,andforrelative headmovement(Creathetal.,2005;Hsuetal.,2007).Therefore, modelsincludingtwoormorejointshavebeenproposed(Fig.6) (Alexandrovetal.,2005;Kuo,2005;Hsuetal.,2007;Kilbyetal., 2015).Duringquietstanding,theCoMcanberepresentedwithtwo joints:theankleandthehipjoint(Gageetal.,2004).Kilbyetal. (2015)calculatedtheCoMin3,4,5andhigherdegreesoffreedom (DoFs)models.NotonlytheDoFs,butalsothetendonmodeling, areimportantforconstructingarealisticbodymodeltoverifythe controller(Lorametal.,2004,2005a,b).Recently,withadvances incomputingtechnology,detailedmusculoskeletalmodelssuchas OpenSim(Delpetal.,2007)havebeenproposedandutilizedto

simulatehumanmotions.Inonestudy,3DoFmodelswith mus-cleswereconstructedwithoptimizedposturecontrolusingthe GeneticAlgorithm(PashaZanoosietal.,2015).Althoughmodels withadditionalDoFscanrepresentcomplexmotion,theyrequire complexcontrollerstoreproducemotionsthatwereobservedand measuredinexperiments.Therefore, it isimportant tofindthe simplestbodymodelsthatcanadequatelyrepresentthe empiri-callyderivedmotions.Forexample,unperturbedstandingcanbe adequatelyrepresentedwithonejoint(Pinteretal.,2008). 4.3. Multisensoryintegrationmodel

Dealingwithmultisensoryfeedbackandstateestimationare challengesencounteredduringmodelingofposturecontrol. Sev-eral studies have attempted to construct a Kalman filter as a multisensoryintegrationof theinputsof theposturecontroller inhumans(VanderKooijetal.,1999,2001;Kiemeletal.,2002). Sensoryweightingandreweightingmodelsareverycommon feed-backmodels.Themultisensoryreweightingmodelsforcondition changesinhumansareproposedinmanystudiesusingBayesian modelsorlinearsummation.Reweightingofproprioceptionand graviceptionisdiscussedintermsofcontinuousfeedbackcontrol (Peterka and Loughlin,2004;Van derKooijand Peterka,2011). Themultisensoryinputsareintegratedlinearlyasweightsofeach sensoryinputinthemodel,andthemodelcanthenrepresent alter-ationsof body sway in experiments.The model isextended to separatemodelsoftheupperandlowerbodyinGoodworthand Peterka(2012).Somestudies(Mahboobinetal.,2005;Mahboobin, 2007)alsoutilizethelinearreweighting modelwithvisualand proprioceptive inputs.In Hwanget al.(2014), thevisual, vesti-bular,andproprioceptiveinputsaredisturbedinthequietstanding posture,andthegainofeachsensoryinputiscalculated repeat-edlyasboththeweightandreweight.Estimationandthreshold areimportantcontributorstoposturecontrolinseveralcommonly occurringhumanconditionsofthesensorydisturbances,andsome researchershaveproposedareweightingmethodwiththresholds toestimatetheseconditions(Blümleet al.,2006;Maureretal., 2006).Anoptimal estimation method withminimal prediction errorhasbeenproposedbyKuo(2005).IntheBayesianmodels,the reliabilityofeachsensoryinputisobtainedbyday-to-day experi-ence,andthenusedbythecontrollertoweighteachsense(Dokka etal.,2010;VilaresandKording,2011).Thismethodcanrepresent anonlinearintegrationofthemultisensoryinputs,andisadequate forrepresentingtheeffectsofaging.Therefore,thismethodis cur-rentlyconsideredthemosteffectiveavailableformodelingslow dynamics.Acomputationalmodelofthecontrolandsensory inte-grationof posture,aswellas otherkinds ofmotions,hasbeen previouslyreviewedinFranklinandWolpert(2011).

5. Long-termalterationinposture

Theposturalalterationsmentionedaboveareconsideredfast dynamics because of theirrelated immediate sensory changes. However,alterationsarealsocausedoverthelongterm,producing slowdynamics.

5.1. Agingandposturedisorders

Agingisoneofthefactorsinvolvedinlong-termalterationsin posture.Comparedtoyoungerindividuals,thereweightingof sen-soryinputsisalteredinelderlyindividuals(Eikemaetal.,2014). Therefore,therelationshipbetweenthemultisensory reweight-ingandaginghasbeeninvestigatedrecentlyusingvarioussensory inputalterations. According toreports of experimentalposture controlstudies,thelikelihoodoffallingdownincreases exponen-tiallywithage,especiallyaftertheageof60(Horaketal.,1989).

Furthermore,anincreaseinbodyswayintheSOTwasobserved in theelderly(Speerset al.,2002), especially intheabsenceof visualandproprioceptivecues(Cohenetal.,1996).Thepatterns of joint movement in elderly individuals are similarto that in youngadults,butlargerproximaljointrotationsinelderly indi-vidualsinducelargersway(Tsaietal.,2014).Backwardtransition platformexperimentsindicatethatelderlyindividualsexperience morebodyswayandemploydifferentposturecontrolstrategies comparedtoyoungerindividuals(Kasaharaetal.,2015).Moreover, astudyusingavarietyofmethodstoinhibitmultisensoryinputs reporteddifferencesinCoPbetweenelderlyandyoung individ-ualsinseveralconditions(Maitreetal.,2013).Similartoreportsof experimentalsensoryreweighting,studiesreportingthat proprio-ceptioncanbealteredinelderlyindividualshavebeenreviewed byGobleetal.(2009)andShafferandHarrison(2007).Recently, itwasreportedthattheeffectsofvisualcueswithoscillation,as wellaswithoscillationandtransition,showonlymarginal differ-encesbetweenyoung,elderly,andfall-proneelderlyindividuals (Jekaetal.,2006).Vestibulardysfunctioninelderly,healthyelderly, and healthyyounger adultshasbeencompared usingSOT, and theresultsindicatedthatvisualandvestibularfunctionsexhibited significantlymoreage-relatedweakeningthandidproprioception (Pedalinietal.,2009).Furthermore,GVSislesseffectiveonelderly subjectscomparedwithyoungsubjects(Eikemaetal.,2014). Sev-eralstudies ofage-related alterations atthecorticaland spinal levelshavebeenconducted(seePapegaaijetal.,2014forareview). The model-based approach of posturecontrol indicatesthat withProportional-Integral-Derivative(PID)control,whichisbasic feedbackcontrolmethodinengineering,improvinghipandankle stiffness mayincreasethelateralstability ofelderlyindividuals (Nishihorietal.,2012).In additiontothedifferencesinsensory inputs, thetime delayinthefeedback loop isalsosignificantly differentbetweenayounggroupandanelderlygroup(Quetal., 2009;Davidsonetal.,2011).Themodel-basedapproachofsensory reweightingindicatesthatproprioception,ratherthanothercues, isprimarilyutilizedinelderlyindividuals(Wiesmeieretal.,2015). Reductionofboththevisualandproprioceptivecuesresultsin sig-nificantdifferencesoftheposturesbetweentheyoungandelderly (Mahboobin,2007).Theabovestudiesprovideevidencethat mul-tisensoryintegrationundergoesanage-relateddeclineresultingin alterationsinposturecontrol.Ontheotherhand,multisensory inte-grationinchildrenalsodiffersfromthatinadults(Petersonetal., 2006).Thismeansthatthealterationofmultisensoryintegrationis inducedbyaginganddevelopment.

Certaindisorders alsocausealterations inmultisensory inte-gration. For example, Yozu and Haga found that the patients ofhereditarysensoryand autonomicneuropathy(HSAN)type4 or 5 have significantly different gait parameters (Zhang et al., 2013).Patientswithalossofvestibularfunctioncanmaintainan uprightposturewhenvisualandsomatosensorycuesarealtered bylong-termadaptation(Bronstein,1986;Herdman,1997). Sim-ilarly,individualswithvestibulardysfunctiondonotdifferfrom normalsubjectsintheirresponsestovisualinhibitionorafoam platform(Cohenetal.,2014).Inaddition,individualswithlossof somatosensoryfunctiondonotexhibitchangesinposture com-paredtohealthyindividualsduringquietstanding(Horaketal., 1990). However, limitations in other sensory substitutions are observedinpatientswithlossofvestibularinput(Nashneretal., 1982;Bronstein, 1986;Maureretal.,2000,2006;Blümleetal., 2006),andtheydonotemploythehipstrategythathumanutilize movementofhipjointforposturalmaintenance.Elderlypatients withlossofvestibularinputexhibitmoresignificantdifferencesin posturecontrol(Pedalinietal.,2009).Patientswithalossof ves-tibularinputcanbedividedintotwogroups,dependingonhow pronetheyaretofallinginresponsetovisualchangesandalight touch(Horak,2009).Theauthorshypothesizedthatthisdifference

mayhavebeencausedbytrainingandlearningintheindividuals’ dailylives.TheposturalstabilityofpatientswithParkinson’s dis-easehasalsobeenanalyzed,andasignificantdifferenceinposture wasobservedinmultisensoryalteration,butnotinsingle-sensory alteration(Bertolinietal.,2015).Onexposuretovisuallycongruent stimulation,patientswithAlzheimer’sdiseaseshowedadecrease inbalance,butpatientswithParkinson’sdiseasedidnot(Chong etal.,1999).Abalancetestforhemiplegiadiagnosishasinfact beenproposed(DiFabioandBadke,1990),whichindicatesthat theSOTmaybeacandidatetestforsuchpatients.Itisvery diffi-culttoproposeacommontheorytoexplaintheobservationsinall thesepatients,becauseofinter-individualvariations.

5.2. Learningandrehabilitation

Alearningeffectinposturecontrolhasbeenpreviously demon-strated,revealingthathealthyadultsimprovetheirposturecontrol significantlyintheSOT(Wrisleyetal.,2007).Thearchers(archery players)canbeinastablepositioncomparedwithnormalsubjects becauseoftraining(Stambolievaetal.,2015).Alearningeffectwas alsoobservedinPeterkaandLoughlin(2004).

Recentstudieshavefocusedonthelearningandrehabilitation applicationsofmotioncontrol,includingposturecontrol.Arecent study(Nardoneetal.,2010)reportedrehabilitationinpatientswith disorders ofperipheralneuropathy, as wellasvestibular disor-ders,usingamovingplatform.Theimprovementofbalancecontrol providedbythisinterventionissimilartothatseenwiththe appli-cationofFrankelexercises.Forvestibularrehabilitationtherapy, pleaserefertothefollowingreview(Hanetal.,2011).Recently, robot-aidedneurorehabilitationhasbeenproposed(Krebsetal., 2000), butrecoveryofposturewiththis method(model)is still insufficient.Oneofthereasonsforthisisthattherelationshipor correspondencebetweenthemodelparametersandtheendpoint goalofrehabilitationarestillunclear.Thedifferenceinthe param-etersshouldbebasedonthemotionormeasurablevaluesfromthe rehabilitation.

6. Conclusion

In this review, we introducedstudies ofposture control for themaintenance ofanuprightstance.Multisensoryinputs play important roles in the stability of humanposture. The studies presentedinvestigatedmultisensoryinputsandobservedpostural alterations.Theyalsoconstructedmodelsthatincludenotonly pos-turecontrolmodels,butalsobodymodelsandsensoryintegration models.Thesemodelsarestilldependentonthetasksinvolved, andconditionsthatincludeimpairmentsofsubjectsand environ-mentalconditions,andmeasurementssuchasCoP,CoM,EMGs,and brainimages.Inthisfield,theresearchershavecollectedevidence toprovideinsightintohumaninternalcontrol,forwhichthey pro-posemodelstobeabletoexplaintheresultsofthehumanpostural experiments.

Fastdynamicsof posturealterationby changesin multisen-soryinputshavebeeninvestigatedandmodeledinmanystudies. However,slowdynamics,thelong-termalteration,hasnotbeen sufficientlyinvestigated.We suggestthattheresearchesshould focusonthe“bodyrepresentationinthebrain”thatisutilizedto estimatethestateofthebodyviatheintegrationofsensoryinputs. Anydifferencebetweenthis “bodyrepresentationinbrain” and therealbodywillresultininappropriatebodycontrol.This differ-encemayresultfromagingorotherimpairmentsandcanbepartly reducedbyrehabilitation,usingthelong-termalterationmodel.To achievethis,aninterdisciplinaryapproachcombiningthefieldsof neuroscience,biomechanicsandrehabilitationwillberequired.

Acknowledgment

ThisworkissupportedbyJSPSKAKENHI,Grantin-Aidfor Scien-tificResearchonInnovativeAreasUnderstandingbrainplasticity onbodyrepresentationstopromotetheiradaptivefunctions(Grant Number26120004and26120006)andGrant-in-AidforScientific Research(C)(15K06131).

WewouldliketothankEditage(www.editage.jp)forEnglish languageediting.

References

Alexandrov,A.,Frolov,A.,Horak,F.,Carlson-Kuhta,P.,Park,S.,2005.Feedback equi-libriumcontrolduringhumanstanding.Biol.Cybern.93,309–322.

Allison,L.K.,Kiemel,T.,Jeka,J.J.,2006.Multisensoryreweightingofvisionandtouch isintactinhealthyandfall-proneolderadults.Exp.BrainRes.175,342–352.

Allum,J.H.J.,Bloem,B.R.,Carpenter,M.G.,Hulliger,M.,Hadders-Algra,M.,1998.

Proprioceptivecontrolofposture:areviewofnewconcepts.GaitPosture8, 214–242.

Barin,K.,1989.Evaluationofageneralizedmodelofhumanposturaldynamicsand controlinthesagittalplane.Biol.Cybern.61,37–50.

Bertolini,G.,Wicki,A.,Baumann,C.R.,Straumann,D.,Palla,A.,2015.Impairedtilt perceptioninParkinson’sdisease:acentralvestibularintegrationfailure.PLOS ONE10,e0124253.

Blümle,A.,Maurer,C.,Schweigart,G.,Mergner,T.,2006.Acognitiveintersensory interactionmechanisminhumanposturalcontrol.Exp.BrainRes.173,357–363.

Bottaro,A.,Yasutake,Y.,Nomura,T.,Casadio,M.,Morasso,P.,2008.Boundedstability ofthequietstandingposture:anintermittentcontrolmodel.Hum.Mov.Sci.27, 473–495.

Bronstein,A.M.,1986.Suppressionofvisuallyevokedposturalresponses.Exp.Brain Res.63,655–658.

Capicikova,N.,Rocchi,L.,Hlavaˇcka,F.,Chiari,L.,Cappello,A.,2006.Humanpostural responsetolowerlegmusclevibrationofdifferentduration.Physiol.Res.55, 1–6.

Chiba,R.,Ogawa,H.,Takakusaki,K.,Asama,H.,Ota,J.,2013.Muscleactivities chang-ingmodelbydifferenceinsensoryinputsonhumanposturecontrol.Adv.Intell. Syst.Comput.194,479–491.

Chong,R.K.,Horak,F.B.,Frank,J.,Kaye,J.,1999.Sensoryorganizationforbalance: specificdeficitsinAlzheimer’sbutnotinParkinson’sdisease.J.Gerontol.A.Biol. Sci.Med.Sci.54,M122–M128.

Coelho,C.M.,Balaban,C.D.,2015.Neuroscienceandbiobehavioralreviews visuo-vestibularcontributions to anxietyand fear. Neurosci.Biobehav. Rev.48, 148–159.

Cohen,H.,Heaten,L.G.,Congdon,S.L.,1996.Changesinsensoryorganisationtest scoreswithage.AgeAgeing25,39–44.

Cohen,H.S.,Mulavara,A.P.,Peters,B.T.,Sangi-Haghpeykar,H.,Bloomberg,J.J.,2014.

Standingbalancetestsforscreeningpeoplewithvestibularimpairments. Laryn-goscope124,545–550.

Creath,R.,Kiemel,T.,Horak,F.,Peterka,R.,Jeka,J.,2005.Aunifiedviewofquietand perturbedstance:simultaneousco-existingexcitablemodes.Neurosci.Lett.377, 75–80.

Davidson,B.S.,Madigan,M.L.,Southward,S.C.,Nussbaum,M.A.,2011.Neuralcontrol ofpostureduringsmallmagnitudeperturbations:effectsofagingandlocalized musclefatigue.IEEETrans.Biomed.Eng.58,1546–1554.

Delp,S.L.,Anderson,F.C.,Arnold,A.S.,Loan,P.,Habib,A.,John,C.T.,Guendelman, E.,Thelen,D.G.,2007.OpenSim:open-sourcesoftwaretocreateandanalyze dynamicsimulationsofmovement.IEEETrans.Biomed.Eng.54,1940–1950.

DiFabio,R.P.,Badke,M.B.,1990.Relationshipofsensoryorganizationtobalance functioninpatientswithhemiplegia.Phys.Ther.70,542–548.

Diekmann,V.,Jürgens,R.,Becker,W.,2004.Maintainingspatialbodyalignmenton arotatingplatformbymeansofactivecounter-circling:roleofvestibularand podokinestheticafferents.Exp.BrainRes.158,504–518.

Dokka,K.,Kenyon,R.V.,Keshner,E.A.,Kording,K.P.,2010.Selfversusenvironment motioninposturalcontrol.PLoSComput.Biol.6.

Eikema,D.J.A.,Hatzitaki,V.,Tzovaras,D.,Papaxanthis,C.,2014.Applicationof inter-mittentgalvanicvestibularstimulationrevealsage-relatedconstraintsinthe multisensoryreweightingofposture.Neurosci.Lett.561,112–117.

Fitzpatrick,R.,Burke,D.,Gandevia,S.C.,1996.Loopgainofreflexescontrolling humanstandingmeasuredwiththeuseofposturalandvestibulardisturbances. J.Neurophysiol.76,3994–4008.

Forbes,P.A.,Siegmund,G.P.,Schouten,A.C.,Blouin,J.-S.,2015.Task,muscleand frequencydependentvestibularcontrolofposture.Front.Integr.Neurosci.8, 1–12.

Franklin,D.W.,Wolpert,D.M.,2011.Computationalmechanismsofsensorimotor control.Neuron72,425–442.

Gage,W.H.,Winter,D.A.,Frank,J.S.,Adkin,A.L.,2004.Kinematicandkineticvalidity oftheinvertedpendulummodelinquietstanding.GaitPosture19,124–132.

Gatev,P.,Thomas,S.,Kepple,T.,Hallett,M.,1999.Feedforwardanklestrategyof balanceduringquietstanceinadults.J.Physiol.514,915–928.

Goble,D.J.,Coxon,J.P.,Wenderoth,N.,VanImpe,A.,Swinnen,S.P.,2009. Proprio-ceptivesensibilityintheelderly:degeneration,functionalconsequencesand plastic-adaptiveprocesses.Neurosci.Biobehav.Rev.33,271–278.

Goodworth,A.D.,Peterka,R.J.,2012.Sensorimotorintegrationformultisegmental frontalplanebalancecontrolinhumans.J.Neurophysiol.107,12–28.

Green,A.M.,Angelaki,D.E.,2010.Internalmodelsandneuralcomputationinthe vestibularsystem.Exp.BrainRes.200,197–222.

Han,B.I.,Song,H.S.,Kim,J.S.,2011.Vestibularrehabilitationtherapy:reviewof indications,mechanisms,andkeyexercises.J.Clin.Neurol.7,184–196.

He,J.,Levine,W.S.,Loeb,G.E.,1991.Feedbackgainsforcorrectingsmall perturba-tionstostandingposture.IEEETrans.Autom.Control36,322–332.

Herdman,S.J.,1997.Advancesinthetreatmentofvestibulardisorders.Phys.Ther., 602–618.

Hettich,G.,Assländer,L.,Gollhofer,A.,Mergner,T.,2014.Humanhip-ankle coordina-tionemergingfrommultisensoryfeedbackcontrol.Hum.Mov.Sci.37,123–146.

Hettich,G.,Lippi,V.,Mergner,T.,2011.Human-LikeSensorFusionMechanismsina PosturalControlRobot.

Horak,F.B.,2009.Posturalcompensationforvestibularloss.Ann.N.Y.Acad.Sci. 1164,76–81.

Horak,F.B.,Macpherson,J.M.,1996.Posturalorientationandequilibrium.Compr. Physiol.,255–292.

Horak,F.B.,Nashner,L.M.,Diener,H.C.,1990.Posturalstrategiesassociatedwith somatosensoryandvestibularloss.Exp.BrainRes.82,167–177.

Horak,F.B.,Shupert,C.L.,Mirka,A.,1989.Componentsofposturaldyscontrolinthe elderly:areview.Neurobiol.Aging10,727–738.

Hsu,W.-L.,Scholz,J.P.,Schöner,G.,Jeka,J.J.,Kiemel,T.,2007.Controland esti-mationofpostureduringquietstancedependsonmultijointcoordination.J. Neurophysiol.97,3024–3035.

Hwang,S.,Agada,P.,Kiemel,T.,Jeka,J.J.,2014.Dynamicreweightingofthree modal-itiesforsensorfusion.PLOSONE9,1–8.

Jeka,J.,Allison,L.,Saffer,M.,Zhang,Y.,Carver,S.,Kiemel,T.,2006.Sensory reweight-ingwithtranslationalvisualstimuliinyoungandelderlyadults:theroleof state-dependentnoise.Exp.BrainRes.174,517–527.

Jeka,J.,Oie,K.S.,Kiemel,T.,2000.Multisensoryinformationforhumanpostural control:integratingtouchandvision.Exp.BrainRes.134,107–125.

Karim,H.,Fuhrman,S.I.,Sparto,P.,Furman,J.,Huppert,T.,2013.Functionalbrain imagingofmulti-sensoryvestibularprocessingduringcomputerizeddynamic posturographyusingnear-infraredspectroscopy.Neuroimage74,318–325.

Kasahara,S.,Saito,H.,Anjiki,T.,Osanai,H.,2015.Theeffectofagingonvertical posturalcontrolduringtheforwardandbackwardshiftofthecenterofpressure. GaitPosture42(4),448–454.

Kavounoudias,A.,Gilhodes,J.C.,Roll,R.,Roll,J.P.,1999.Frombalanceregulationto bodyorientation:twogoalsformuscleproprioceptiveinformationprocessing? Exp.BrainRes.124,80–88.

Kiemel,T.,Oie,K.S.,Jeka,J.J.,2002.Multisensoryfusionandthestochasticstructure ofposturalsway.Biol.Cybern.87,262–277.

Kiemel,T.,Zhang,Y.,Jeka,J.J.,2011.Identificationofneuralfeedbackforupright stanceinhumans:stabilizationratherthanswayminimization.J.Neurosci.31, 15144–15153.

Kilby,M.C.,Molenaar,P.C.M.,Newell,K.M.,2015.Modelsofposturalcontrol:shared varianceinjointandCOMmotions.PLOSONE10,e0126379.

Krebs,H.I.,Volpe,B.T.,Aisen,M.L.,Hogan,N.,2000.Increasingproductivityand qual-ityofcare:robot-aidedneuro-rehabilitation.J.Rehabil.Res.Dev.37,639–652.

Kuo,A.,1995.Anoptimalcontrolmodelforanalyzinghumanposturalbalance. Biomed.Eng.IEEETrans.,1–37.

Kuo,A.D.,2005.Anoptimalstateestimationmodelofsensoryintegrationinhuman posturalbalance.J.NeuralEng.2,S235–S249.

Li,Y.,Levine,W.S.,Loeb,G.E.,Member,S.,2012.Atwo-jointhumanposturecontrol modelwithrealisticneuraldelays.IEEETrans.NeuralSyst.Rehabil.Eng.20, 738–748.

Lockhart,D.B.,Ting,L.H.,2007.Optimalsensorimotortransformationsforbalance. Nat.Neurosci.10,1329–1336.

Loram,I.D.,Maganaris,C.N.,Lakie,M.,2004.Paradoxicalmusclemovementinhuman standing.J.Physiol.556,683–689.

Loram,I.D.,Maganaris,C.N.,Lakie,M.,2005a.Active,non-spring-likemuscle move-mentsinhumanposturalsway:howmightparadoxicalchangesinmuscle lengthbeproduced.J.Physiol.564,281–293.

Loram,I.D.,Maganaris,C.N.,Lakie,M.,2005b.Humanposturalswayresultsfrom frequent,ballisticbiasimpulsesbysoleusandgastrocnemius.J.Physiol.564, 295–311.

Mahboobin,A.,2007.Computationalandroboticmodelsofhumanposturalcontrol 3300572,126.

Mahboobin,A.,Loughlin,P.J.,Redfern,M.S.,Sparto,P.J.,2005.Sensoryre-weighting inhumanposturalcontrolduringmoving-sceneperturbations.Exp.BrainRes. 167,260–267.

Maitre,J.,Gasnier,Y.,Bru,N.,Jully,J.L.,Paillard,T.,2013.Discrepancyintheinvolution ofthedifferentneuralloopswithage.Eur.J.Appl.Physiol.113,1821–1831.

Maurer, C., Mergner,T., Bolha, B., Hlavacka, F.,2000. Vestibular, visual, and somatosensorycontributionstohumancontrolofuprightstance.Neurosci.Lett. 281,99–102.

Maurer,C.,Mergner,T.,Peterka,R.J.,2006.Multisensorycontrolofhumanupright stance.Exp.BrainRes.171,231–250.

Mergner,T.,2010.Aneurologicalviewonreactivehumanstancecontrol.Annu.Rev. Control34,177–198.

Mergner,T.,Schweigart,G.,Fennell,L.,2009.Vestibularhumanoidposturalcontrol. J.Physiol.Paris103,178–194.

Morasso,P.G.,Baratto,L.,Capra,R.,Spada,G.,1999.Internalmodelsinthecontrol ofposture.NeuralNetw.12,1173–1180.

Morasso,P.G.,Sanguineti,V.,2002.Anklemusclestiffnessalonecannotstabilize balanceduringquietstanding.J.Neurophysiol.88,2157–2162.

Nardone,A.,Godi,M.,Artuso,A.,Schieppati,M.,2010.Balancerehabilitationby movingplatformandexercisesinpatientswithneuropathyorvestibulardeficit. Arch.Phys.Med.Rehabil.91,1869–1877.

Nashner,L.M.,Black,F.O.,Wall,C.,1982.Adaptationtoalteredsupportandvisual conditions during stance: patients with vestibulardeficits. J. Neurosci.2, 536–544.

Nashner,L.M.,Peters,J.F.,1990.Dynamicposturographyinthediagnosisand man-agementofdizzinessandbalancedisorders.Neurol.Clin.8,331–349.

Nishihori,T.,Aoki,M.,Jiang,Y.,Nagasaki,S.,Furuta,Y.,Ito,Y.,2012.Effectsofaging onlateralstabilityinquietstance.AgingClin.Exp.Res.24,162–170.

Nomura,T.,Oshikawa,S.,Suzuki,Y.,Kiyono,K.,Morasso,P.,2013.Modelinghuman posturalswayusinganintermittentcontrolandhemodynamicperturbations. Math.Biosci.245,86–95.

Oie,K.S.,Kiemel,T.,Jeka,J.J.,2001.Humanmultisensoryfusionofvisionandtouch: detectingnon-linearitywithsmallchangesinthesensoryenvironment. Neu-rosci.Lett.315,113–116.

Papegaaij,S.,Taube,W.,Baudry,S.,Otten,E.,Hortobágyi,T.,2014.Agingcausesa reorganizationofcorticalandspinalcontrolofposture.Front.AgingNeurosci. 6,1–15.

Park,S.,Horak,F.B.,Kuo,A.D.,2004.Posturalfeedbackresponsesscalewith biome-chanicalconstraintsinhumanstanding.Exp.BrainRes.154,417–427.

PashaZanoosi,A.A.,Naderi,D.,Sadeghi-Mehr,M.,Feri,M.,Beheshtiha,A.S., Fallahne-jad,K.,2015.Investigationoftheeffectsofhumanbodystabilityonjointangles’ prediction.MultibodySyst.Dyn.35(2),111–129.

Pedalini,M.E.B.,Cruz,O.L.M.,Bittar,R.S.M.,Lorenzi,M.C.,Grasel,S.S.,2009.Sensory organizationtestinelderlypatientswithandwithoutvestibulardysfunction. ActaOtolaryngol.129,962–965.

Peterka,R.J.,2002.Sensorimotorintegrationinhumanposturalcontrol.J. Neuro-physiol.88,1097–1118.

Peterka,R.J.,Loughlin,P.J.,2004.Dynamicregulationofsensorimotorintegrationin humanposturalcontrol.J.Neurophysiol.91,410–423.

Peterson,M.L.,Christou,E.,Rosengren,K.S.,2006.Childrenachieveadult-like sen-soryintegrationduringstanceat12-years-old.GaitPosture23,455–463.

Pinter,I.J.,vanSwigchem,R.,vanSoest,A.J.K.,Rozendaal,L.A.,2008.Thedynamics ofposturalswaycannotbecapturedusingaone-segmentinvertedpendulum model:aPCAonsegmentrotationsduringunperturbedstance.J.Neurophysiol. 100,3197–3208.

Qu,X.,Nussbaum,M.A.,Madigan,M.L.,2009.Model-basedassessmentsoftheeffects ofageandanklefatigueonthecontrolofuprightpostureinhumans.GaitPosture 30,518–522.

Runge,C.F.,Shupert,C.L.,Horak,F.B.,Zajac,F.E.,1999.Ankleandhippostural strate-giesdefinedbyjointtorques.GaitPosture10,161–170.

Shaffer,S.W.,Harrison,A.L.,2007.Agingofthesomatosensorysystem:a transla-tionalperspective.Phys.Ther.87,193–207.

Shima,K.,Shimatani,K.,Sugie,A.,Kurita,Y.,Kohno,R.,Tsuji,T.,2013.Virtuallight touchcontact:anovelconceptformitigationofbodysway.In:20137th Inter-nationalSymposiumonMedicalInformationandCommunicationTechnology (ISMICT).IEEE,pp.108–111.

Soto-Faraco,S.,Ronald,A.,Spence,C.,2004.Tactileselectiveattentionandbody posture:assessingthemultisensorycontributionsofvisionandproprioception. Percept.Psychophys.66,1077–1094.

Speers,R.A.,Kuo,A.D.,Horak,F.B.,2002.Contributionsofalteredsensationand feed-backresponsestochangesincoordinationofposturalcontrolduetoaging.Gait Posture16,20–30.

Stambolieva,K.,Otzetov,M.,Petrova,D.,Ikonomov,R.,Gatev,P.,2015.Postural Sta-bilityduringStaticUprightStanceinArchers.In:Gatev,P.,Hatzitaki,V.(Eds.), Posture,BalanceandtheBrain,InternationalWorkshopProceedings.ProconLtd, Sofia,pp.29–35.

Suzuki,Y.,Nomura,T.,Casadio,M.,Morasso,P.,2012.Intermittentcontrolwithankle, hip,andmixedstrategiesduringquietstanding:atheoreticalproposalbasedon adoubleinvertedpendulummodel.J.Theor.Biol.310,55–79.

Tsai,Y.C.,Hsieh,L.F.,Yang,S.,2014.Age-relatedchangesinpostureresponseunder acontinuousandunexpectedperturbation.J.Biomech.47,482–490.

VanderKooij,H.,Jacobs,R.,Koopman,B.,Grootenboer,H.,1999.Amultisensory integrationmodelofhumanstancecontrol.Biol.Cybern.80,299–308.

VanderKooij,H.,Jacobs,R.,Koopman,B.,VanDerHelm,F.,2001.Anadaptive modelofsensoryintegrationinadynamicenvironmentappliedtohumanstance control.Biol.Cybern.84,103–115.

VanderKooij,H.,Peterka,R.J.,2011.Non-linearstimulus-responsebehaviorofthe humanstancecontrolsystemispredictedbyoptimizationofasystemwith sensoryandmotornoise.J.Comput.Neurosci.30,759–778.

Vette, A.H., Masani, K., Nakazawa, K., Popovic, M.R., Member, S.S.S.S., 2010.

Neural-mechanicalfeedbackcontrolschemefluctuationduringquietstance. Rehabilitation18,86–95.

Vilares,I.,Kording,K.,2011.Bayesianmodels:thestructureoftheworld,uncertainty, behavior,andthebrain.Ann.N.Y.Acad.Sci.1224,22–39.

Wade,M.G.,Jones,G.,1997.Theroleofvisionandspatialorientationinthe main-tenanceofposture.Phys.Ther.77,619–628.

Wallmann,H.W.,2009.Thebasicsofbalanceandfalls.HomeHealthCareManag. Pract.21,436–439.

Wiesmeier,I.K.,Dalin,D.,Maurer,C.,2015.Elderlyuseproprioceptionratherthan visualandvestibularcuesforposturalmotorcontrol.Front.AgingNeurosci.7, 1–14.

Winter,D.A.,Patla,A.E.,Prince,F.,Ishac,M.,Gielo-Perczak,K.,1998.Stiffnesscontrol ofbalanceinquietstanding.J.Neurophysiol.80,1211–1221.

Wolpert,D.M.,Ghahramani,Z.,Jordan,M.I.,1995.Aninternalmodelfor sensorimo-torintegration.Science269,1880–1882.

Woollacott,M.,Shumway-Cook,A.,2002.Attentionandthecontrolofpostureand gait:areviewofanemergingareaofresearch.GaitPosture16,1–14.

Wrisley,D.M.,Stephens,M.J.,Mosley,S.,Wojnowski,A.,Duffy,J.,Burkard,R.,2007.

Learningeffectsofrepetitiveadministrationsofthesensoryorganizationtestin healthyyoungadults.Arch.Phys.Med.Rehabil.88,1049–1054.

Zhang,Y.,Ogata,N.,Yozu,A.,Haga,N.,2013.Two-dimensionalvideogait anal-ysesinpatientswithcongenitalinsensitivitytopain.Dev.Neurorehabil.16, 266–270.