Mental Practice for Relearning Locomotor Skills

Mental Practice for RelearningLocomotor SkillsFrancine Malouin, Carol L. Richards

Over the past 2 decades, much work has been carried out on the use of mentalpractice through motor imagery for optimizing the retraining of motor function inpeople with physical disabilities. Although much of the clinical work with mentalpractice has focused on the retraining of upper-extremity tasks, this article reviewsthe evidence supporting the potential of motor imagery for retraining gait and tasksinvolving coordinated lower-limb and body movements. First, motor imagery andmental practice are defined, and evidence from physiological and behavioral studiesin healthy individuals supporting the capacity to imagine walking activities throughmotor imagery is examined. Then the effects of stroke, spinal cord injury, lower-limbamputation, and immobilization on motor imagery ability are discussed. Evidence ofbrain reorganization in healthy individuals following motor imagery training of danc-ing and of a foot movement sequence is reviewed, and the effects of mental practiceon gait and other tasks involving coordinated lower-limb and body movements inpeople with stroke and in people with Parkinson disease are examined. Lastly,questions pertaining to clinical assessment of motor imagery ability and trainingstrategies are discussed.

F. Malouin, PhD, is ProfessorEmeritus, Department of Rehabili-tation, Faculty of Medicine, LavalUniversity, and Center for Inter-disciplinary Research in Rehabil-itation and Social Integration(CIRRIS), Rehabilitation Institute ofQuebec, 525 Blvd Wilfrid-HamelEst, Quebec City, Quebec, CanadaG1M 2S8. Address all correspon-dence to Dr Malouin at: [email protected].

C.L. Richards, PT, PhD, DU, FCAHS,is Professor and Holder of the LavalUniversity Research Chair in Cere-bral Palsy, Department of Rehabil-itation, Faculty of Medicine, LavalUniversity, and Director, Centrefor Interdisciplinary Research inRehabilitation and Social Integra-tion (CIRRIS), Rehabilitation Insti-tute of Quebec.

[Malouin F, Richards CL. Mentalpractice for relearning locomotorskills. Phys Ther. 2010;90:xxx–xxx.]

© 2009 American Physical TherapyAssociation

Perry Gait Issue

Post a Rapid Response orfind The Bottom Line:www.ptjournal.org

February 2010 Volume 90 Number 2 Physical Therapy f 1

Over the past 2 decades, muchwork has been carried out onthe use of mental practice

through motor imagery for optimiz-ing the retraining of motor functionin people with physical disabilities.Although much of the clinical workwith mental practice has focused onthe retraining of upper-extremitytasks, in this article we will reviewthe evidence supporting the poten-tial of motor imagery for retraininggait and tasks involving coordinatedlower-limb and body movements.First, we will define motor imageryand mental practice and examineevidence from physiological and be-havioral studies in healthy individu-als supporting the capacity to imag-ine walking activities through motorimagery. Then the effects of stroke,spinal cord injury (SCI), lower-limbamputation, and immobilization onmotor imagery ability will be dis-cussed. Evidence of brain reorganiza-tion in healthy individuals followingmotor imagery training of dancingand of a foot movement sequencewill be reviewed, and then the ef-fects of mental practice on gait andother tasks involving coordinatedlower-limb and body movements inpeople with stroke and in peoplewith Parkinson disease will be exam-ined. Lastly, questions pertaining toclinical assessment of motor imageryability and training strategies will bediscussed.

Defining Mental Practiceand Motor ImageryMotor imagery is the imagining ofan action without its physical execu-tion; it is an active process duringwhich the representation of an ac-tion is internally reproduced withinworking memory without any overtoutput.1 Mental practice or motorimagery practice, on the other hand,is the repetition or rehearsing ofimagined motor acts with the inten-tion of improving their physical ex-ecution.2 Mental practice of locomo-tor skills thus requires the ability to

form internal representations of lo-comotor activities. Movement repre-sentations can be made from 2 per-spectives: (1) from the third-personperspective (or external imagery), asspectator, when imagining anotherperson walking or (2) from the first-person perspective (or internal imag-ery), from the inside as if the actor,when imagining oneself walking.2–4

Each perspective has different prop-erties. The external perspective im-plies primarily a visual representa-tion of the motor task, whereas theinternal perspective entails, in addi-tion to the visual representation, thekinesthetic sensations associated withthe simulated movements, thus bothvisual and kinesthetic cues.

What Evidence Do We HaveThat Locomotor ActivitiesCan Be Imagined ThroughMotor Imagery?Experimental studies have used dif-ferent approaches to examine themental representation of locomotoractivities in individuals without dis-abilities. Our understanding of motorimagery of walking comes from neu-rophysiological and cerebral imagingstudies examining the similarities be-tween real and simulated locomotoractivities. These studies have shownthat locomotor activities, either per-formed physically or imagined, aresubject to common laws and princi-ples. For instance, autonomic studiesthat monitored changes in heart andrespiration rates while healthy indi-viduals imagined walking on a tread-mill at different speeds showedspeed-related increases during theimagination of walking.5–7 Mentalchronometric studies comparingmovement times in people walkingor imagining walking to targetsplaced at different distances showedthe duration of walking to be simi-lar in both conditions, thus indicat-ing a temporal coupling between theduration of real and imagined walk-ing conditions.3,8–12 Moreover, when

people walk or imagine themselveswalking on narrow beams,12 throughgates11 or along paths of differentwidths,3 uphill or downhill,9 and atdifferent speeds,9 both the actualand imagined walking times increaseas a function of the difficulty of thetask. The latter findings thus indicatethat Fitts’ law,13 which states thatmore-difficult movements take moretime to produce physically than eas-ier movements, also applies to motorimagery of walking.

Further confirmation of functionalsimilarity between real walking andimagined walking comes from func-tional brain imaging studies. Directcomparison of cortical activity evokedduring actual gait and the imagina-tion of gait with near-infrared spec-troscopy14 have shown that actualand simulated walking increase brainactivity bilaterally in the medial pri-mary sensorimotor cortices and thesupplementary motor area (SMA).These findings have been confirmedwith positron emission tomography(PET)15 and functional magnetic res-onance imaging (fMRI).16,17 Furtherdemonstration that cortical brainareas are engaged during locomotoractivities comes from PET and fMRIstudies that examined brain activa-tion patterns during the imaginingof standing,15,18 initiating gait,15,19

normal walking,15–17 walking withobstacles,15 precision gait,16 walk-ing along a curved path,19 and run-ning20,21 and during the imaginingof other complex movements in-volving the whole body (eg, swim-ming, dancing; lifting a heavy box).21

Altogether the findings from thefunctional brain imaging studiesconfirm that the simulation of lo-comotor activities and complexwhole-body tasks result in the activa-tion of cortical networks similarto those found during motor imageryof simple movements, thus suggest-ing that the overlapping amongneural substrates during real and

Mental Practice for Relearning Locomotor Skills

2 f Physical Therapy Volume 90 Number 2 February 2010

imagined movements also applies tocomplex body movements.21

In summary, mentally simulated andphysically executed locomotor ac-tivities share similar autonomic re-sponses and temporal organizationand activate neural networks thatgreatly overlap. Consequently, it hasbeen suggested that the benefits ofmental practice training are linked tothe activation of cerebral networksthat are comparable to those acti-vated during physical execution.2,21

Is Motor Imagery AbilityAffected by Central andPeripheral Lesions of theNervous System?Because the ability to form internalrepresentations of motor acts is nec-essary for training with mental prac-tice, there is a need to determinewhether this ability is specifically af-fected following central nervous sys-tem (CNS) or peripheral nervous sys-tem (PNS) lesions. Because of itsconcealed nature, however, motorimagery is difficult to assess.22–24

Three main approaches have beenused to assess motor imagery abilityin clinical settings: mental rotation,mental chronometry, and question-naires. Mental rotation is used tomeasure the accuracy of motor rep-resentations. In this approach, peo-ple are asked to verbally judge thelaterality of hands or feet from pic-tures in different postural condi-tions. Behavioral and functional neu-roanatomy studies have demonstratedthat mental rotation of body parts iscarried out through a sort of innermotor simulation.25,26 Mental chro-nometry involves the comparison ofmovement times during the simula-tion and execution of a motor task invarious conditions (fast and slowpaces or over short and long timeperiods); it is used to examine tem-poral organization of simulated ac-tions.10,27–30 Lastly, the clarity anddetails of the images and the inten-

sity of the sensations (vividness) per-ceived during movement simulationare assessed with motor imageryquestionnaires.24,31

After StrokeMotor imagery ability has been stud-ied extensively in people with cere-bral lesions. The findings indicatethat the representation of movementremains possible after stroke,24–26,28

even in people with chronic or se-vere motor impairments,26 suggest-ing that the mental representation ofmovement is not dependent on mo-tor activity following CNS injury. Todate, only a few patients with focal-

ized lesions in the superior region ofthe parietal cortex29 or the frontalcortex25 have shown motor imageryimpairment. Recently, findings in pa-tients with stroke and age-matchedhealthy subjects who were assessedwith the Kinesthetic and Visual Im-agery Questionnaire (KVIQ)31 re-vealed that the level of motor imag-ery vividness following stroke wassimilar to that of healthy subjects,with good and bad imagers in bothgroups24 (Fig. 1). Likewise, using aleft- or right-hand judgment task thatimplicitly requires motor imagery(mental rotation), Johnson and col-leagues26 found that people with

Figure 1.Individual visual (A) and kinesthetic (B) scores of the Kinesthetic and Visual ImageryQuestionnaire for people with a left hemispheric lesion (LHL) (n�13), a right hemi-spheric lesion (RHL) (n�19), and age-matched healthy individuals (CTL) (n�32). Thehorizontal lines indicate the 2-sided 95% confidence interval (CI). Scores above the CIlines�very good imagery ability, scores between the CI lines�good imagery ability, andscores below the CI lines�poor imagery ability. Adapted and reprinted with permissionof the American Society of Neurorehabilitation from: Malouin F, Richards CL, Durand A,Doyon J. Clinical assessment of motor imagery after stroke. Neurorehabil Neural Repair.2008;22:330–340.



stroke and healthy subjects had arange of accuracy scores that wassimilar in both groups, with very ac-curate subjects (above 90%) and less-accurate subjects (above 65%) ineach group.26 This wide range ofscores in accuracy and vividness ofmotor imagery indicates that motorimagery ability is not an all-or-nonephenomenon; rather, there is a con-tinuum in the level of performance,and in some cases the difficulty informing internal representations ofmovement can be a premorbid traitunrelated to cerebral damage.24

Although these clinical studies fo-cused mainly on the representationof simple limb movements, findingsfrom a chronometric study in agroup of people with stroke and acontrol group who were asked tophysically execute and imagine theTimed “Up & Go” Test (TUG) indi-

cate that the temporal representa-tion of this complex locomotor taskis retained following stroke.32 Thevariability in the movement timesduring imagination and executionconditions in the group with strokewas very similar to that observed inthe control group (Fig. 2: uppergraphs), and there was no significantdifference between the mean imagi-nation times and execution times foreach of the subtasks in both groups.Moreover, the relative percentage oftime dedicated to each subtask wassimilar in both groups (Fig. 2: lowergraphs). These results imply thatboth the temporal coupling and thetemporal structure of the TUG areretained after stroke. Such findingsare notable because they indicatethat the ability to rehearse mentallycomplex motor tasks is preserved af-ter stroke.32

After Complete SpinalCord InjuryFindings from behavioral10 and fMRIstudies33–35 indicate that the repre-sentation of foot movements is re-tained in people with complete SCI.Indeed, scores of motor imageryvividness after complete SCI are sim-ilar to those of control subjects, andthe extent of brain activation duringimagery of foot movements corre-lates with the vividness of their im-agery.33–35 The persistence of motorrepresentations in the disconnectedlimbs after a complete SCI is furtherevidence that motor imagery is main-tained even when voluntary move-ments are not possible.

After Limb Amputation andLimb ImmobilizationAlthough motor imagery ability fol-lowing cerebral lesions24–26,28 andspinal lesions33–35 is retained evenwhen physical movements are im-paired or impossible, findings indi-cate that motor imagery ability is di-minished by the lack of movementafter the loss of a limb36,37 or tempo-rary disuse following limb immobili-zation.37 More specifically, althoughthe representation of movements isstill present after upper-limb ampu-tation, it has been shown to be lessaccurate during a left- or right-handjudgment task in people with anupper-limb amputation, suggestingthat the absence of a limb does notprevent motor imagery, but makes itmore difficult.36 Similarly, motor im-agery vividness after lower-limb am-putation is significantly decreasedfor foot movements of the missinglimb,37 further demonstrating that al-though it is still possible to generatea mental representation of move-ment, the vividness of these imagesis weaker after the loss of a limb.Similar changes of motor imageryvividness were found in subjectswho had an ankle immobilized in acast for 2 to 4 weeks without weightbearing.37 A significant decrease inmotor imagery vividness for move-

Figure 2.Individual imagination and execution times of the Timed “Up & Go” Test (TUG) in agroup of people with stroke (CVA: top left panel) and an age-matched group of healthyindividuals (CTL: top right panel). Relative percentage of time dedicated during theimagination and execution of each subtask (lower panels). Adapted and reprinted withpermission of Oxford University Press from: Malouin F, Richards CL, Jackson PL, DoyonJ. Motor imagery for optimizing the reacquisition of locomotor skills after cerebraldamage. In: Guillot A, Collet C, eds. The Neurophysiological Foundations of Mental andMotor Imagery, Part 3: Motor Imagery in Rehabilitation. London, United Kingdom:Oxford University Press. In press.



ments of the foot (the distal segmentthat was immobilized) also was ob-served. Such findings are remark-able, given the short period of immo-bilization, and imply that changes inmotor imagery vividness with limbdisuse can take place relativelyquickly. Most interesting is the sig-nificant positive correlation (r�.79)found between the onset of walkingwith a prosthesis and scores of mo-tor imagery vividness on the ampu-tated side, which suggests that pros-thesis use helps in maintaining themental representation of the missinglimb.37 In contrast, after the immobi-lization of one lower limb, the posi-tive correlation (r�.90) betweenmotor imagery vividness and the du-ration of immobilization found in theintact limb suggests that the aug-mented use of the intact limb, due tolonger periods of immobilization,promoted imagery vividness on thatside. Overall, these findings indicatethat after limb amputation and dis-use, the mental representation of ac-tions is retained but weaker andhighly modulated by sensorimotorinputs.37

What Evidence Do WeHave That MentalPractice Training of TasksInvolving CoordinatedLower-Limb and BodyMovements InducesBrain Reorganization?To date, 2 studies have investigatedchanges in brain activation patternsin people who used mental practiceto learn sequences of leg move-ments38 and a foot movement se-quence.39 Sacco and colleagues38

found that, in contrast to subjectswho only physically practiced se-quences of leg movements throughtango lessons (45 minutes a day for 5days), those who rehearsed the se-quences mentally (15 minutes a dayfor 5 days) in addition to physicalpractice showed an expansion of thebilateral motor areas. In addition,

there was a reduction of the visuo-spatial activation in the posteriorcortex, suggesting that focusing asubject’s attention on the foot move-ments involved in dancing decreasesthe role of visual imagery processesin favor of motor-kinesthetic pro-cesses.38 Likewise, changes in brainactivity found in the medial aspect ofthe orbitofrontal cortex (increase)and cerebellum (decrease) reportedafter intense mental practice (300repetitions a day for 5 days) of asequence of foot movements furthersupport the notion that mental prac-tice through motor imagery initiallyimproves performance by acting onmotor preparation and planning.38,40

Although foot movements seem re-mote from the more-complex limband body movements in walking, arecent transcranial magnetic stimula-tion (TMS) study41 that examinedhow corticospinal excitability wasaffected by motor imagery of footdorsiflexion and motor imagery ofgait showed a close relationship inthe control of the tibialis anteriormuscle during motor imagery of sim-ple foot dorsiflexion and gait. Thefindings of that study indicate thatcorticospinal effects of a simple mo-tor imagery task can predict cortico-spinal effects of a more-complex mo-tor imagery task involving the samemuscle.

Does Mental PracticeTraining Improve thePerformance of Gaitand Other Tasks InvolvingCoordinated Lower-Limband Body Movementsin People With Strokeand in People WithParkinson Disease?GaitMental practice with motor imageryprovides an opportunity to improvelocomotor skills through safe andself-paced locomotor training in peo-ple with severe disability that ren-

ders walking practice difficult andlimited in time, especially in the earlyphase of rehabilitation.2,42,43 Yet, thepotential use of mental practice foroptimizing the relearning of activi-ties such as walking43–45 and risingfrom a chair and sitting,46,47 as wellas sequential foot movements,48 hasbeen examined mainly in explor-atory studies and case reports withsmall sample sizes.

Dickstein and colleagues43–45 havedeveloped a motor imagery trainingprogram for gait rehabilitation post-stroke. The effects of this trainingprogram were first described in acase report of a 69-year-old man withleft hemiparesis,43 and later the fea-sibility of using this motor imagerytraining at home was examined in 4case studies.44

Dunsky and colleagues45 recently in-vestigated the effects of this home-based motor imagery program in agroup of 17 people poststroke toconfirm and extend previous find-ings. Motor imagery training in thegait rehabilitation program consistedof 15- to 20-minute sessions, 3 timesa week for 6 weeks, without anyphysical intervention. Both internaland external perspectives of motorimagery were used. The main objec-tives of training were to facilitatemovements of the affected lowerlimb and improve posture by focus-ing on specific problems (eg, fore-foot initial contact, push-off) and topromote functional walking in thepatient’s own environment. Thecomplexity of the task during motorimagery was increased progressivelyfrom familiarization with motor im-agery of walking in an isolated place,on flat terrain and without distur-bance (week 1), to more-complexsituations such as imagining walk-ing toward meaningful targets in thepatient’s home and outdoors to in-crease gait speed and symmetry(weeks 5 and 6). Spatiotemporal pa-rameters (gait speed, step length,



single-leg support) and some kine-matic variables (knee extension atinitial foot contact) served as out-come measures. Most patients in-creased their gait speed, with gainsranging from about 10% to 80%, for amean increase of 15 cm/s.45 The ef-fect size was 0.64, corresponding toa moderate treatment effect. In addi-tion to a gain in gait speed, stridelength increased by 18% and single-leg stance time increased by 13%,indicating an improvement in mobil-ity and dynamic balance.45 Most ofthe gains were retained at follow-up(3 weeks after the end of training).

Such a substantial improvementraises much interest because it sup-ports the idea that walking skills canbe enhanced by mental practice.However, because it has been shownthat mental practice of upper-limbmovements led to an increase in thephysical use of the trained extremi-ty,49,50 an increase in real walkingover the 6-week training programalso is likely. Thus, as the amount ofreal walking over the 6-week trainingperiod was not monitored in thestudy by Dunsky and colleagues,45

the extent of motor improvementattributed to mental practice must beinterpreted with caution becausethe gains, in part, could be due tomore walking. Further studies con-trolling for walking activities duringthe training program are needed tovalidate the contribution of mentalpractice to the extent of the gainsreported.

In a recent controlled study of peo-ple with Parkinson disease, evidencethat mental practice could help inreducing bradykinesia during theTUG task was provided when thegroup of patients who combinedphysical and mental practice over a12-week period showed faster per-formance than the group whotrained physically only.51 An impor-tant limitation of this study and ofmany others is the lack of informa-

tion about treatment adherence overthe 12 week-program and theamount of physical locomotor train-ing received by patients in eachgroup.

Coordinated Lower-Limb andBody MovementsBeneficial effects of one session ofmental practice in 12 people withchronic stroke were found in a studythat examined the effects of mentalpractice in combination with a smallamount of physical practice (7 se-ries, each consisting of 1 physicalrepetition and 5 mental repetitions)to improve the amount of loading onthe affected leg during rising from achair and sitting down.46,47 The load-ing on the affected leg after trainingsignificantly improved by 17.9% and16.2%, respectively, when risingfrom the chair and sitting down.Gains were still significant 24 hourslater during rising (12.8%) and sittingdown (11.2%), indicating that learn-ing had occurred. Patients with def-icits in at least 2 domains of workingmemory had a smaller improvement(27% versus 72%) and showed noretention at follow-up, suggestingthat learning effects are strongly re-lated to working memory abilities.46

In contrast, the duration of the taskdid not change with training. Thelatter findings suggest that, in theearly stage of learning a complex mo-tor task, changes in motor strategiespredominate over changes in speedof execution and that, at this stage,vertical forces (limb loading) repre-sent a more-sensitive measure of per-formance than movement time. Al-though, this study had no controlgroup and did not tease out the spe-cific effects of mental practice, thegains achieved with a relatively smallamount of physical practice (7 phys-ical repetitions and 35 mental repe-titions) had a magnitude similar tothose measured after 3 weeks of reg-ular physical training.52

One reason for retention of gainswith such little physical practice isthe combination of physical practicewith mental practice that requiresthe person to mentally and explicitlyrehearse the sequence of move-ments associated with the mobilitytask. Such rehearsal makes the per-son focus each time on the prepara-tion and planning of the proper strat-egy, thus increasing his or herawareness of the required move-ments. Such an interpretation is inline with the results of Pascual-Leoneand colleagues,40 who, using TMS,demonstrated that mental practicehas preparatory effects and increasesthe efficiency of subsequent physicaltraining. Another possible explana-tion for retention of the gains afteronly one mental training sessioncould be the consolidating effect ofsleep on motor learning, which wasreported recently following mentalpractice in individuals who werehealthy.53

Additional support for the primingor added effects of motor imagery onmotor performance comes from thefindings of a pilot study54 showingthat a small number of physical rep-etitions alone (total of 120 repeti-tions over a 4-week period) did notenhance the motor performance(limb loading of the affected leg dur-ing rising-up and sitting-down tasks).When these relatively few physicalrepetitions were combined with alarge number of mental repetitions(total of 1,100 repetitions), however,the loading on the affected leg wassignificantly increased and was re-tained 3 weeks after training.

Another example pertaining to theeffect of combining physical prac-tice and mental practice on lower-limb function is a case study thatinvestigated the effect of mentalpractice on the learning of a footmovement sequence task in a 38-year-old man with a left hemorrhagicsubcortical stroke.48 During the first



2 weeks, the patient physically prac-ticed a serial response time task withthe lower limb. The next week men-tal practice was combined withphysical practice, and then the pa-tient practiced only mentally for 2weeks at home. The patient’s aver-age response time improved sig-nificantly during the first 5 days ofphysical practice (26%), but thenfailed to show improvement. Thecombination of mental practice andphysical practice during the thirdweek yielded an additional improve-ment (10.3%), and the following 2weeks of mental practice resulted ina marginal increase in performance(2.2%). These findings indicate thatthe addition of mental practice whenthe performance has reached a pla-teau can lead to further improve-ment in the performance of a se-quential motor skill.48 Moreover, theretention of the motor skill with mo-tor imagery practice alone at homesuggests that mental practice canplay a role in the retention of newlyacquired abilities.48

In summary, although results fromthese clinical studies suggest thatmental practice can lead to im-provements in gait and other tasksinvolving coordinated lower-limband body movements after stroke,randomized clinical trials withlarger samples are needed to con-firm and generalize findings aboutthe effects reported so far in a smallnumber of subjects. Yet, despiteshortcomings of their designs,these case reports and feasibilityand exploratory studies have pro-vided useful data about thepatients’ ability to adhere to di-verse training approaches, thesensitivity of outcome measures,and the amount of training requiredto obtain significant clinicalimprovements.

What Do We KnowAbout the ClinicalAssessment of MotorImagery Ability andMotor ImageryTraining Strategies?Clinical AssessmentThe inclusion of mental practice inrehabilitation training strategies isstill relatively new. Consequently,strategies and guidelines for its use asan adjunct therapy to promote therelearning of functional activities suchas walking are under development.Because mental practice throughmotor imagery requires the repre-sentation of an action that is inter-nally reproduced within workingmemory,1 good cognitive functionand communicative skills are neces-sary. Clinical studies have used dif-ferent approaches to test cognition,but a common test is the ModifiedMini-Mental State Exam, with an in-clusion score of 24/30 or more.45 Anormal working memory in 2 do-mains (eg, visuospatial, verbal, kines-thetic) also has been suggested.46 Pa-tients with severe communicationproblems were excluded from moststudies because of their difficulty inunderstanding the verbal instruc-tions and in expressing themselvesin order to actively participate in theassessment of motor imagery and inthe learning of imagery.2,39,47

The next step is to assess motor im-agery ability because it may be defi-cient as a result of the nature of theinjury or simply as a premorbid trait.24

Because of its complex nature, how-ever, more than one assessment toolshould be used to determine whethera person is able to engage in motorimagery.22–24 Recently, the combina-tion of the Time Dependent MotorImagery (TDMI) screening test55 andthe KVIQ,24,31 2 measures that aresimple and easy to use in a clinicalsetting, has been proposed as a clin-ical assessment procedure.24 TheTDMI is a chronometric screening

test wherein the examiner recordsthe number of movements imagined(eg, stepping movement) over 3 timeperiods (15, 25, and 45 seconds); itassumes that individuals who reportan increase in the number of move-ments imagined with increasing timeare able to simulate movements andlikely engage in motor imagery24,37,55

(Fig. 3). Similar chronometric testscan be applied to locomotor tasks.For instance, whether patients arereally imagining walking can be ver-ified by asking them to imagine them-selves walking along a short versus alonger path or along a wide versus anarrow path. If the patients reallyengage in motor imagery, movementtimes while imagining walking areexpected to increase with increasingdistance3,8,12 and decrease with in-creasing pathway width.3,11 Likewise,autonomic responses (heart or respi-ratory rates) also could be monitoredduring the imagination of walking atslow and fast speeds; cardiorespira-tory responses are expected to in-crease at faster walking speeds.5–7

Lastly, when scheduling patients forassessing motor imagery, care shouldbe taken to keep track of the time ofthe day because, contrary to realwalking, the duration of the imagina-tion of walking is influenced by thetime of day.56

The KVIQ is a motor imagery ques-tionnaire developed for people withphysical disabilities that assesses thevividness of motor imagery from afirst-person perspective31 and uses a5-point scale to rate the clarity of theimage (visual subscale) and the in-tensity of the sensations (kinestheticsubscale). It consists of 20 items (10movements in each subscale) repre-senting gestures with different bodyparts (head, shoulders, trunk, upperand lower limbs), and all movementsare performed from a sitting posi-tion. Both the TDMI and the KVIQhave been standardized, and theirtest-retest reliability in people withstroke has been confirmed.31,55 Of a



sample of 37 patients with chronicstroke who underwent these evalua-tion procedures, only 2 patients failedthe chronometric screening test, andthe KVIQ revealed that 3 patientswho had passed the screening testshad difficulty in forming mental im-ages of movement.24 Thus, althoughpatients may pass the chronometrictest, they may have difficulty in gen-erating vivid internal representationsof movements. Failure to adhere tothe KVIQ also has been observed oc-casionally in people who werehealthy.24 The assessment proce-dures provide not only an idea on

the motor imagery ability of the pa-tient but also a good introduction tothe notion of motor imagery that pre-pares the patient for mental practicetraining.

Validity of MotorImagery QuestionnairesThe validity of imagery question-naires for assessing motor imageryability has been questioned,23,57

given the subjective nature of self-reported ratings. However, over thepast few years, several studies exam-ining brain activation patterns33–35 ormotor cortex excitability58-60 during

the imagination of movements haveshown strong relationships betweenimagery vividness scores and thelevel of brain activations, suggestingthat ratings from imagery question-naires provide a good indication ofthe ability to generate vivid mentalimages of movement. Recently, inan fMRI study of subjects allocatedto bad and good imager groupsbased on combined scores from amotor imagery questionnaire, mentalchronometry and electrodermal re-sponses—2 distinct functional neuro-anatomical networks, each specificto either the good or bad imagers—were described.61 The latter findingsprovide further evidence of a linkbetween behavioral outcome mea-sures and functional neuroanatomi-cal networks. Finally, in a recent studyin people with lower-limb amputa-tion,37 the agreement between KVIQscores, which are based on an ex-plicit imagery paradigm (the individ-ual is asked to imagine a movement),and scores obtained by Nico and col-leagues,36 who used an implicit im-agery paradigm (ie, mental rotation,in which the individual is asked toverbally judge the laterality of handsand feet portrayed in pictures in dif-ferent postural conditions), providesadditional support for the validity ofmotor imagery questionnaires.

Training StrategiesResults from behavioral, psycho-physical, and brain imaging studiesin athletes and healthy individualsprovide some guidelines for trainingstrategies in rehabilitation.

Are All Tasks Amenable toMental Practice?A large body of knowledge support-ing the use of mental practice to en-hance skill acquisition comes fromstudies conducted in athletes and inhealthy individuals.62–67 It is gener-ally recognized that for tasks with alarge cognitive component (eg, peg-board, card sorting), mental practiceyields stronger effects (effect size�

Figure 3.Number of stepping movements imagined over 3 time periods (15, 25, and 45 seconds)in the Time Dependent Motor Imagery screening test in a group of people with stroke(CVA) and an age-matched control group (CTL). Note the increase in the number ofmovements imagined with longer time periods. Adapted and reprinted with permissionof the American Society of Neurorehabilitation from: Malouin F, Richards CL, Durand A,Doyon J. Clinical assessment of motor imagery after stroke. Neurorehabil Neural Repair.2008;22:330–340.



1.44) compared with motor taskswith an effect size of 0.43.62 Thedifficulty, however, is to determinethe size of the cognitive componentin any motor task.63 Moreover, thecognitive dimension of a task changesas the skill level of the performerchanges; a novice may be thinkingabout how to do the skills, whereasan expert is concentrating on thestrategy and tactics related to theperformance of that skill.63 It also isbelieved that other factors such asthe level of familiarity and task com-plexity interact to determine ef-fects.64 For example, the use of mo-tor imagery to train isometricvoluntary contractions, which in-duced a substantial increase in max-imal torque with the untrained(weak) abductor muscle of the fifthdigit,65 had no beneficial effects onthe strong elbow flexor muscles.66

Such findings indicate that even atask with a low cognitive component(eg, learning to contract a musclemaximally) can benefit from mentalimagery training in the early stage ofstrength training that involves a neu-ral component (eg, learning to con-tract a muscle maximally requiresspatial recruitment of existing motorunits).67 Thus, benefits are to be ex-pected with a motor task when men-tal practice targets motor planningand preparation components9,40 (eg,sequencing of complex foot move-ments, coordination of movementsfrom different body parts). Interest-ingly, meta-analyses revealed thatpositive effects for highly cognitivetasks (eg, card sorting, peg-board)were associated with very few trialsand minutes per session, whereas formotor tasks, positive effects wereobtained with longer sessions andmore repetitions per session.62

What Imagery PerspectiveShould Be Used?One issue in mental practice is theselection of perspective: should aninternal perspective or an externalperspective be used? Because the

terminology with visual (external)and kinesthetic (internal) imageryversus the type of perspective some-times is confusing,2–4,16,23,42 in thepresent article external perspectiverefers to a perspective that involvesprimarily a visual representation(third person) of the motor task,whereas internal perspective entails,in addition to the kinesthetic sensa-tions, a visual representation (fromthe inside: first person) of the simu-lated movements, thus both visualand kinesthetic cues.

Therefore, should the patient be in-structed to imagine another personwalking or to imagine himself or her-self walking from the inside? Behav-ioral, neurophysiological, and brainimaging studies have shown that,compared with the third-person pers-pective, the first-person perspectiveshares more physiological characteris-tics with those observed during theexecution of movement, and thusmovement imagery in the first-personperspective is closer to the real execu-tion of movement.2,16,61,68–72 For thatreason, the challenge with motor im-agery training of gait is to ascertainthat the patient really is imagining thetask in the first-person perspective andfocusing his or her attention on bothvisual and kinesthetic cues to promoteactivation of neural networks associ-ated with motor imagery of gait. Con-sequently, instructions should directthe patient to focus on both visual andkinesthetic components seen and feltfrom the inside. Because the vividnessof visual imagery usually is better thanthat of kinesthetic imagery,24,37,73 con-centrating on visual cues may be easierinitially, but both should be encour-aged. For instance, patients should en-vision walking within an environment(eg, imagine a path’s width, the size orthe position of the obstacles) and thedisplacement of their limbs (eg, seethe top of the feet, the inside of theswinging arms) and re-create the sen-sations associated with the task (eg,feeling the push-off, the effort to in-

crease the step height or length).These cues should be introduced grad-ually and progress according to eachindividual’s needs and ability. In addi-tion, therapists should inquire aboutwhat patients see and feel during theimagination phase to ascertain thatthey are engaged in first-person motorimagery. Checking periodically withmental chronometry or autonomic re-sponses is another way to monitorwhether the patient really is engagedin imagining a given task. These pro-cedures should not only assist in thedevelopment of vivid images but alsohelp control the mental representa-tion of the motor tasks throughout thetraining session.63

How Should We Position thePatient During Imagery Training?Another consideration during motorimagery training is the patient’s pos-ture because internal representationof a movement implies a motor planbased on a body-centered frame ofreference, which depends on visuo-kinesthetic inputs. Results from fMRIand TMS studies have shown thatwhen the position of the imaginedhand is congruent with the actualhand position, higher levels of corti-cal facilitation and brain activationsare recorded,69,72,74 implying thatmotor imagery generates motor plansthat depend on the current configu-ration of the limbs. However, a re-cent study that examined the effectsof hand posture on mental rotationof hands and feet showed that men-tal rotation of hands but not of feetwas influenced by changes in handposture.75 Their findings suggest thatpostural information coming fromthe body may influence mental rota-tion of body parts according to spe-cific somatotopic rules.75 To date,no study has investigated whetherbody orientation (sitting or standingversus lying supine) influences corti-cal activation during the imaginingof hand or foot movements. How-ever, during imagery training in aclinical setting, there are advantages



to placing the patient in a positionsimilar to that used during physicalexecution of the task. For instance,for a task such as rising to a standingposition or reaching forward, a sit-ting position will provide the visualand kinesthetic cues that will helpdevelop a mental representation ofthe task from a first-person perspec-tive.47,48 Moreover, assuming thesame position during both condi-tions (simulation and execution) ismore practical when series of mentalrepetitions are combined with phys-ical repetitions.47,48 Lastly, to promotethe concentration and relaxationthat facilitate motor imagery vivid-ness,63 care should be taken to pro-vide seated individuals with backand head support during the simula-tion condition.

Is It Important to CombineMental Practice andPhysical Practice, and, If So,How Should It Be Done?Another question concerns opti-mal conditions for motor imagerytraining. Although mental rehearsalalone can promote brain reorganiza-tion38,39,40 and have a priming effecton subsequent physical training,40

the number of sessions and repeti-tions required to observe motor im-provement may depend on the typeand complexity of the task.39,40,76,77

Evidence of improved motor per-formance was observed after 1,500mental rehearsals of a foot move-ment sequence over a 5-day trainingperiod.39 In another study, young in-dividuals who were healthy needed120 mental repetitions over a singletraining session before a significantimprovement of a complex manipu-lative skill was observed.76 A thirdstudy involving complex finger move-ment sequences showed that sub-jects who only trained mentally for 2hours a day had reached after 5 daysthe level of performance attained af-ter 3 days of physical practice.40

Most interesting is the fact that on

the fifth day, the group who hadtrained mentally needed only 2hours of physical practice to reach alevel of performance similar to thatof the subjects who had trainedphysically 2 hours a day for 5 days.40

If findings can be extended to loco-motor tasks in people with physicaldisabilities, it could mean that pa-tients who mentally rehearse walk-ing before they actually can standand walk would show improvementfaster once they start walking thanthose who did not rehearse mentally.

Although motor improvement can beobtained with motor imagery alone,better results are obtained whencombining physical and mental re-hearsals of a task.2,42,62 We also mustkeep in mind that mental practice isan adjunct to habitual therapy andthat mental rehearsal of a task doesnot replace physical practice of thesame task.2,62,63 Therefore, the choiceof strategy for combining mental andphysical repetitions is paramount.We know that the temporal featuresof imagined walking are less variablefrom trial to trial when each mentalrehearsal is separated by a physicalrehearsal, suggesting that the affer-ent information is helpful for consis-tent reproduction of the next imag-ined movement.9 Only a few clinicalstudies46–48,54 have provided detailsabout the training procedures andcontrolled for the number of physi-cal and mental rehearsals. To date inthe clinical setting, good adherenceand learning effects46–48,54 havebeen reported with training para-digms combining physical executiontrials and mental rehearsal trials inproportions ranging from 1 physicalexecution and 5 mental rehearsals to1 physical execution and 10 mentalrehearsals for retraining rising andsitting down in people followingstroke. Because mental rehearsals atthe outset of training demand muchattention and concentration, it is sug-gested to gradually increase the num-ber of mental repetitions46,47,54 be-

tween physical repetitions.Including one physical execution be-tween bouts of mental repetitionshas been found to help in maintain-ing the kinesthetic sensations of thetask.46,47,62

Teaching Motor ImageryFindings from studies carried out inathletes62–64,78 can provide someguidelines for teaching imagery inclinical settings. Because motor im-agery is a complex, multidimen-sional process, it is important to pro-vide imagery instructions withsufficient details to ensure that theindividual is imagining the task in thedesired manner (eg, with vividnessand perspective). It must be clearwhether the entire task is to be imag-ined or just specific parts, and formore-complex tasks, the proper se-quence of movements should betaught. Thus, verbal instructions arevery important, and it is essential toestablish a dialogue between theteacher and the learner, especially inthe beginning, to make sure that theinstructions are well understood. Italso is suggested that people whohave low imagery ability or who arenot familiar with motor imagery startby imagining skill tasks that they al-ready do well.62 Best outcomes areexpected when the imagery includesa positive or successfulperformance.78

ConclusionsSeveral lines of evidence point tothe beneficial effects of mental prac-tice for retraining locomotor skills.However, further clinical studies withstrong designs and larger groups areneeded to confirm and generalizethe positive findings reported so far.Clinicians must be aware that goodand bad imagers coexist after stroke,24

and thus it is imperative to evaluatemotor imagery ability before intro-ducing mental practice. Based onrecent findings that mental represen-tation of actions is highly modulatedby imagery practice,37 patients who



initially demonstrate difficulty ingenerating mental representation ofmovements eventually may improvetheir motor imagery ability with re-peated exposures. Virtual environ-ments79 and observation of gait17 mayprovide the visual reinforcementsnecessary to promote the generationof proper mental images in poor im-agers of walking or those who can-not yet walk.

The use of mental practice as anadjunct to physical practice in neu-rorehabilitation still is relatively new,and many questions remain regard-ing the optimization of training strat-egies. Should patients be lying downand relaxing while listeningto prerecorded instruction tapes, orin contrast, should they be more ac-tively engaged in their training?Should they learn to self-monitortheir training and solve problemsalong the way46,47,54,80,81 so that mo-tor imagery eventually can be prac-ticed alone48 or at home withoutsupervision?81 Given the potentialbenefits of mental practice in reha-bilitation and increasing clinician in-terest in its use, more information onstrategies and clinical guidelines thataddress the questions raised in thisarticle should be available in the nearfuture.

Both authors provided concept/idea/projectdesign, writing, fund procurement, institu-tional liaisons, and consultation (includingreview of manuscript before submission). DrMalouin provided data analysis, projectmanagement, and clerical support. The au-thors acknowledge the assistance of DanielTardif in the preparation of the figures.

This article was received January 28, 2009,and was accepted May 28, 2009.

DOI: 10.2522/ptj.20090029

References1 Decety J, Grezes J. Neural mechanisms

subserving the perception of human ac-tions. Trends in Cog Sci. 1999:3:172–178.

2 Jackson PL, Lafleur MF, Malouin F, et al.Potential role of mental practice usingmotor imagery in neurological rehabilita-tion. Arch Phys Med Rehabil. 2001;82:1133–1141.

3 Bakker M, de Lange FP, Stevens JA. Motorimagery of gait: a quantitative approach.Exp Brain Res. 2007;179;497–504.

4 Solodkin A, Hlustik P, Chen EE. Small SI.Fine modulation in network activationduring motor execution and motor imag-ery. Cerebral Cortex. 2004;14:1246–1255.

5 Decety J, Jeannerod M, Germain M, Pas-tene J. Vegetative responses during imag-ined movement is proportional to mentaleffort. Behav Brain Res. 1991:42:1–5.

6 Wuyam B, Moosavi SH, Decety J, et al.Imagination of dynamic exercise pro-duced ventilatory responses which weremore apparent in competitive sportsmen.J Physiol. 1995:482(pt 3):713–724.

7 Fusi S, Cutuli D, Valente MR, et al. Cardio-ventilatory responses during real or imag-ined walking at low speed. Arch Ital Biol.2005;143:223–228.

8 Bakker M, Verstappen CCP, Bloem B R,Toni I. Recent advances in functional neu-roimaging of gait. J Neural Transm. 2007;114:1323–1331.

9 Courtine G, Papaxanthis C, Gentili R,Pozzo T. Gait-dependent motor memoryfacilitation in covert movement execu-tion. Cog Brain Res. 2004;22:67–75.

10 Decety J, Boisson D. Effect of brain andspinal cord injuries on motor imagery. EurArch Psychiatr Clin Sci.1990;240:39–43.

11 Decety J, Jeannerod M. Mentally simulatedmovements in virtual reality: does Fitts’law hold in motor imagery? Behav BrainRes.1996;72:127–134.

12 Decety J, Jeannerod M, Prablanc C. Thetiming of mentally represented action. Be-hav Brain Res. 1989;34:35–42.

13 Fitts PM. The information capacity of thehuman motor system in controlling theamplitude of movement. J Exp Psychol.1954;47:381–391.

14 Miyai I, Tanabe HC, Sase I, et al. Corticalmapping of gait in humans: a near-infraredspectroscopic topography study. Neuro-image. 2001;14:1186–1192.

15 Malouin F, Richards CL, Jackson PL, et al.Brain activations during motor imagery oflocomotor-related tasks: a PET study. HumBrain Mapp. 2003;19:47–62.

16 Bakker M, de Lange FP, Helmich RC, et al.Cerebral correlates of motor imagery ofnormal and precision gait. Neuroimage.2008;41:998–1010.

17 Iseki K, Hanakawa T, Shinozaki J, et al.Neural mechanisms involved in mental im-agery and observation of gait. Neuroim-age. 2008;41:1021–1031.

18 Ouchi Y, Okada H, Yoshikawa E, et al.Brain activation during maintenance ofstanding postures in humans. Brain. 1999;22(pt 2):329–338.

19 Wagner J, Stephan T, Kalla R, et al. Mindthe bend: cerebral activations associatedwith mental imagery of walking along acurved path. Exp Brain Res. 2008;191:247–255.

20 Jahn K, Deutschlander A, Stephan T, et al.Brain activation patterns during imaginedstance and locomotion in functional mag-netic resonance imaging. Neuroimage.2004;22:1722–1731.

21 Szameitat AJ, Shen S, Sterr A. Motor imag-ery of complex everyday movements: anfMRI study. Neuroimage. 2007;34:702–713.

22 Guillot A, Collet C. Contribution from neu-rophysiological and psychological meth-ods to the study of motor imagery. BrainRes Rev. 2005;15:287–297.

23 Sharma N, Pomeroy VM, Baron JC. Motorimagery: a backdoor to the motor systemafter stroke? Stroke. 2006;37:1941–1952.

24 Malouin F, Richards CL, Durand A, DoyonJ. Clinical assessment of motor imageryafter stroke. Neurorehabil Neural Repair.2008;22:330–340.

25 Johnson SH. Imagining the impossible: in-tact motor representations in hemiplegics.Neuroreport. 2000;11:729–732.

26 Johnson SH, Sprehn G, Saykin AJ. Intactmotor imagery in chronic upper limb hemi-plegics: evidence for activity-independentaction representations. J Cog Neurosc. 2002;14:841–852.

27 Malouin F, Richards CL, Desrosiers J,Doyon J. Bilateral slowing of mentally sim-ulated actions after stroke. Neuroreport.2004;7:1349–1353.

28 Sirigu A, Cohen L, Duhamel JR, et al. Con-gruent unilateral impairments for real andimagined hand movements. Neuroreport.1995,6:997–1001.

29 Sirigu A, Duhamel JR, Cohen L, et al. Themental representation of hand movementsafter parietal cortex damage. Science. 1996;273:1564–1568.

30 Stinear CM, Fleming MK, Barber PA, By-blow WD. Lateralization of motor imageryfollowing stroke. Clin Neurophysiol. 2007;118:1794–1801.

31 Malouin F, Richards CL, Jackson PL, et al.The Kinesthetic and Visual Imagery Ques-tionnaire (KVIQ) for assessing motor im-agery in persons with physical disabilities:a reliability and construct validity study.J Neurol Phys Ther. 2007;31:20–29.

32 Malouin F, Richards CL, Jackson PL,Doyon J. Motor imagery for optimizing thereacquisition of locomotor skills after ce-rebral damage. In: Guillot A, Collet C, eds.The Neurophysiological Foundations ofMental and Motor Imagery, Part 3: MotorImagery in Rehabilitation. London,United Kingdom: Oxford University Press.In press.

33 Alkadhi H, Brugger P, Boendermaker SH,et al. What disconnection tells about mo-tor imagery: evidence from paraplegic pa-tients. Cereb Cortex. 2005;15:131–140.

34 Cramer SC, Orr ELR, Cohen MJ, LacourseMG. Effects of motor imagery training afterchronic, complete spinal cord injury. ExpBrain Res. 2007;177:233–242.

35 Hotz-Boendermaker S, Funk M, SummersP, et al. Preservation of motor programs inparaplegics as demonstrated by attemptedand imagined foot movements. Neuroim-age. 2008;39:383–394.



36 Nico D, Daprati E, Rigal F, et al. Left andright hand recognition in upper limb am-putees. Brain. 2003;27:120–132.

37 Malouin F, Richards CL, Durand A, et al.Effects of practice, visual loss, limb ampu-tation and disuse on motor imagery vivid-ness. Neurorehabil Neural Repair. 2009;23:449–463.

38 Sacco K, Cauda F, Cerliani L, et al. Motorimagery of walking following training inlocomotor attention: the effect of “the tangolesson.” Neuroimage. 2006;32:1441–1449.

39 Jackson PL, Lafleur MF, Malouin F, et al.Functional cerebral reorganization follow-ing motor sequence learning through men-tal practice with motor imagery. Neuro-image. 2003;20:1171–1180.

40 Pascual-Leone A, Nguyet D, Cohen LG,et al. Modulation of muscle responsesevoked by transcranial magnetic stimu-lation during the acquisition of new finemotor skills. J Neurophysiol. 1995;74:1037–1045.

41 Bakker M, Overeem S, Snijders AH, et al.Motor imagery of foot dorsiflexion andgait: effects on corticospinal excitability.Clin Neurophysiol. 2008;119:2519–2527.

42 Dickstein R, Deutsch JE. Motor imagery inphysical therapist practice. Phys Ther.2007;87:942–953.

43 Dickstein R, Dunsky A, Marcovitz E. Motorimagery for gait rehabilitation in post-stroke hemiparesis. Phys Ther. 2004:84:1167–1177.

44 Dunsky A, Dickstein R, Ariav C, et al. Mo-tor imagery practice in gait rehabilitationof chronic post-stroke hemiparesis: fourcase studies. Int J Rehabil Res. 2006;29:351–356.

45 Dunsky A, Dickstein R, Marcovitz,E, et al.Home-based motor imagery training forgait rehabilitation of people with chronicpoststroke hemiparesis. Arch Phys MedRehabil. 2008;89:1580–1588.

46 Malouin F, Belleville S, Desrosiers J, et al.Working memory and mental practice af-ter stroke. Arch Phys Med Rehabil. 2004;85:177–183.

47 Malouin F, Richards CL, Belleville S, et al.Training mobility tasks after stroke withcombined mental and physical practice: afeasibility study. Neurorehabil Neural Re-pair. 2004;18:66–75.

48 Jackson PL, Doyon J, Richards CL, MalouinF. The efficacy of combined physical andmental practice in learning of a foot-sequence task after stroke: a case study.Neurorehabil Neural Repair. 2004:18:106–111.

49 Page SJ, Levine P, Leonard A. Mental prac-tice in chronic stroke: results of a random-ized, placebo-controlled trial. Stroke. 2007;38:1293–1297.

50 Page SJ, Levine P, Leonard AC. Effects ofmental practice on affected limb use andfunction in chronic stroke. Arch Phys MedRehabil. 2005;86:399–402.

51 Tamir R, Dickstein R, Huberman M. Inte-gration of motor imagery and physicalpractice in group treatment applied tosubjects with Parkinson’s disease. Neuro-rehabil Neural Repair. 2007;21:68–75.

52 Engardt M, Ribbe T, Olsson E. Verticalground reaction force feedback to enhancestroke patients’ symmetrical body-weightdistribution while rising/sitting down.Scand J Rehabil Med. 1993;25:41–48.

53 Debarnot U, Creveaux T, Collet C, et al.Sleep-related improvements in motorlearning following mental practice. BrainCog. 2009;69:398–405.

54 Malouin F, Richard CL, Durand A, Doyon J.Added value of mental practice combinedwith a small amount of physical practiceon the relearning of rising and sitting post-stroke: a pilot study. J Neurol Phys Ther.In press.

55 Malouin F, Richards CL, Durand A, DoyonJ. Reliability of mental chronometry forassessing motor imagery ability after stroke.Arch Phys Med Rehabil. 2008;89:311–319.

56 Gueugneau N, Mauvieux B, Papaxanthis C.Circadian modulation of mentally simu-lated motor actions: implications for thepotential use of motor imagery in rehabil-itation. Neurorehabil Neural Repair. 2009;23:237–245.

57 Lotze M, Halsband U. Motor imagery.J Physiol (Paris). 2006;99:386–395.

58 Lotze M. Scheler G, Tan H-RM, et al. Themusician’s brain: functional imaging ofamateurs and professionals during perfor-mance and imagery. Neuroimage. 2003;20:1817–1829.

59 Lotze M, Flor H, Grodd W, et al. Phantommovements and pain: a fMRI study in up-per limb amputees. Brain. 2001;124:2268–2277.

60 Fourkas AD, Bonavolonta V, Avenanti A,Aglioti SM. Kinesthetic imagery and tool-specific modulation of corticospinal rep-resentations in expert tennis players.Cereb Cortex. 2008;18:2382–2390.

61 Guillot A, Collet C, Nguyen VA, et al. Func-tional neuroanatomical networks associ-ated with expertise in motor imagery.Neuroimage. 2008;41:1471–1483.

62 Feltz DL, Landers DM. The effects of men-tal practice on motor skill learning andperformance: a meta-analysis. J Sports Psy-chol. 1982;5:25–57.

63 Hall C, Schmidt D, Durand MC, Buckolz E.Imagery and motor skill acquisition. In:Sheikh AA, Korn ER, eds. Imagery in Sportsand Physical Performance. Imagery andHuman Development Series. New York, NY:Baywood Publishing Co Inc; 1994:121–134.

64 Corbin CB. Mental practice. In: MorganWP, ed. Ergogenic Aids and MuscularPerformance. New York, NY: AcademicPress; 1972:94–118.

65 Yue G, Cole KJ. Strength increases fromthe motor program: comparison of train-ing with maximal voluntary and imaginedmuscle contractions. J Neurophysiol. 1992;67:1114–1123.

66 Herbert RD, Dean C, Gandevia SC. Effectsof real and imagined training on voluntarymuscle activation during maximal isomet-ric contractions. Acta Physiol Scand.1998;163:361–368.

67 Sale DG. Neural adaptation to resistancetraining. Med Sci Sports Exerc. 1988;20:S135–S145.

68 Fourkas AD, Avenanti A, Urgesi C, AgliotiSM. Corticospinal facilitation during firstand third person imagery. Exp Brain Res.2006;168:143–151.

69 de Lange FP, Helmich RC, Toni I. Postureinfluences motor imagery: an fMRI study.Neuroimage. 2006;33:609–617.

70 Guillot A, Collet C, Nguyen VA, et al. Brainactivity during visual versus kinestheticimagery: an fMRI study. Hum Brain Mapp.2009;30:2157–2172.

71 Stinear CM, Byblow WD, Steyvers M, et al.Kinesthetic, but not visual, motor imagerymodulates corticomotor excitability. ExpBrain Res. 2006;168:157–164.

72 Vargas CD, Olivier E, Craighero L, et al.The influence of hand posture on cortico-spinal excitability during motor imagery: atranscranial magnetic stimulation study.Cereb Cortex. 2004;14:1200–1206.

73 Hall CR, Pongrac J, Buckolz E. The mea-surement of imagery ability. Hum Mov Sci.1985;4:107–118.

74 Fourkas AD, Ionta S, Aglioti SM. Influenceof imagined posture and imagery modalityon corticospinal excitability. Behav BrainRes. 2006;168:190–196.

75 Ionta S, Fourkas AD, Fiorio M, Aglioti SM.The influence of hands posture on mentalrotation of hands and feet. Exp Brain Res.2007;183:1–7.

76 Allami N, Paulignan Y, Brovelli A. Bous-saoud D. Visuo-motor learning with com-bination of different rates of motor imag-ery and physical practice. Exp Brain Res.2008;184:105–113.

77 Louis M, Guillot A, Maton S, et al. Effect ofimagined movement speed on subsequentmotor performance. J Mot Behav. 2008;40:117–132.

78 Woolfolk RL, Parrish MW, Murphy M. Theeffects of positive and negative imagery onmotor skill performance. Cogn Ther Res.1985;9:335–341.

79 Fung J, Malouin F, McFadyen BJ, et al. Lo-comotor rehabilitation in a complex vir-tual environment. Conf Proc IEEE EngMed Biol Soc. 2004;7:4859–4861.

80 Liu KP, Chan CC, Lee TM, Hui-Chan CW.Mental imagery for promoting relearningfor people after stroke: a randomized con-trolled trial. Arch Phys Med Rehabil. 2004;85:1403–1408.

81 Braun S, Kleynen M, Schols J, et al. Us-ing mental practice in stroke rehabilita-tion: a framework. Clin Rehabil. 2008;22:579–591.



Mental Practice for Relearning Locomotor Skills

Documents