546134

research-article2014

NNRXXX10.1177/1545968314546134Neurorehabilitation and Neural RepairDeconinck et al

Clinical Research Article

Reflections on Mirror Therapy: A Systematic Review of the Effect of Mirror Visual Feedback on the Brain

Neurorehabilitation and Neural Repair 2015, Vol. 29(4) 349­–361 © The Author(s) 2014 Reprints and permissions: sagepub.com/journalsPermissions.nav DOI: 10.1177/1545968314546134 nnr.sagepub.com

Frederik J. A. Deconinck, PhD1,2, Ana R. P. Smorenburg, PhD3, Alex Benham, PhD4, Annick Ledebt, PhD5, Max G. Feltham, PhD6, and Geert J. P. Savelsbergh, PhD5

Abstract Background. Mirror visual feedback (MVF), a phenomenon where movement of one limb is perceived as movement of the other limb, has the capacity to alleviate phantom limb pain or promote motor recovery of the upper limbs after stroke. The tool has received great interest from health professionals; however, a clear understanding of the mechanisms underlying the neural recovery owing to MVF is lacking. Objective. We performed a systematic review to assess the effect of MVF on brain activation during a motor task. Methods. We searched PubMed, CINAHL, and EMBASE databases for neuroimaging studies investigating the effect of MVF on the brain. Key details for each study regarding participants, imaging methods, and results were extracted. Results. The database search yielded 347 article, of which we identified 33 suitable for inclusion. Compared with a control condition, MVF increases neural activity in areas involved with allocation of attention and cognitive control (dorsolateral prefrontal cortex, posterior cingulate cortex, S1 and S2, precuneus). Apart from activation in the superior temporal gyrus and premotor cortex, there is little evidence that MVF activates the mirror neuron system. MVF increases the excitability of the ipsilateral primary motor cortex (M1) that projects to the “untrained” hand/arm. There is also evidence for ipsilateral projections from the contralateral M1 to the untrained/affected hand as a consequence of training with MVF. Conclusion. MVF can exert a strong influence on the motor network, mainly through increased cognitive penetration in action control, though the variance in methodology and the lack of studies that shed light on the functional connectivity between areas still limit insight into the actual underlying mechanisms. Keywords systematic review, sensorimotor control, visual feedback, mirror, hemiparesis, stroke, cerebral palsy

Introduction Often a source of fascination, or perhaps frustration, optical illusions have captivated people since ancient times. For instance, curved surfaces and the absence of right angles in archaic Greek temples suggest that its architects attempted to optically correct the illusion of slanted columns or curved tympanums; however, others believe these features may serve engineering purposes or reflect aesthetic preference.1 As much as they are a source of excitement, for neuroscientists optical illusions are considered a backdoor into people’s mind and provide an excellent way to study the neural mechanisms underlying perception and action.2 Interestingly, although optical illusions are known to deceive the individual, the false reality may fool the brain, such that the outcome is beneficial. One such an illusion is the mirror illusion, which has been found to have therapeutic benefits over the past 2 decades. When a mirror is placed, along the midsagittal plane in between the 2 limbs, the

reflection of the limb in front of the mirror is superimposed on the contralateral limb. Any motion of the limb in front of the mirror induces the illusion of 2 synchronously moving limbs. After Ramachandran and his colleagues found that this illusion could alleviate phantom pain in a proportion of the patients,3 mirror visual feedback (MVF) was introduced as a neurorehabilitation tool to treat other unilateral pain 1

Ghent University, Ghent, Belgium Manchester Metropolitan University, Manchester, UK 3 Burke-Cornell Medical Research Institute, White Plains, NY, USA 4 Bradford Institute for Health Research, Bradford, UK 5 VU University, Amsterdam, Netherlands 6 University of Birmingham, Birmingham, UK 2

Corresponding Author: Frederik J. A. Deconinck, Ghent University, Faculty of Medicine and Health Sciences, Department of Movement and Sports Sciences, Watersportlaan 2, Gent 9000, Belgium. Email: [email protected]

350 disorders, such as complex regional pain syndrome (CRPS). In addition, the paradigm is now used to promote motor recovery (eg, in hemiparetic patients or after hand surgery). Despite its widespread use in neurorehabilitation and the claims that MVF therapy would lead to neuroplastic changes, there is no consensus about the underlying mechanism and speculation often lacks the neuroscientific proof. The aim of this review is therefore to bring together current knowledge on the effect of MVF on the brain as has been described in neuroimaging studies, in order to explore potential processes underlying the beneficial clinical effects of MVF. To acquaint the reader with MVF and its current applications, we will first revisit Ramachandran’s rationale for MVF, followed by a narrative review of the clinical neurorehabilitation research that followed in his footsteps. At the end of this section, we introduce 3 hypotheses that have been proposed to explain the positive effects related to MVF. Part 2 provides a systematic review and discussion of studies that examined the effect of MVF on brain activation patterns using neuroimaging or electrophysiological techniques. Finally, in Part 3 we discuss the findings of the systematic review in relation to the hypotheses introduced in Part 1 and we identify where further research is required.

Part 1: Mirror Therapy—Background, Current Applications, and Potential Mechanisms The idea of using MVF for the management of phantom limb pain was inspired by early findings on the integration of perception and action, in particular the principle of reafference.4,5 Reafference is afferent sensory information caused by a motor command (eg, signals from muscle spindles in M. biceps brachii when the arm is actively flexed), as opposed to exafferent information, which results from factors outside the individual (eg, signals from muscle spindles in M. biceps brachii when the arm is flexed passively). To distinguish between these 2 sensory stimuli, it is maintained that the generation of a motor command is accompanied by a parallel signal, termed efference copy, which contains the sensory feedback to be expected due to this command. Comparison of all afferent signals with the efference copy provides a way to separate signals that originate from bodily movements and those from outside the individual. As a consequence, motor commands that are not instantaneously followed by the expected reafferent feedback will be modified in an attempt to evoke the expected sensory afference.5 It is this conflictive state that, according to some,6,7 may evolve into a form of “learned paralysis” accompanied by a feeling of painful spasms,* as experienced by a proportion of patients who have had an arm or

Neurorehabilitation and Neural Repair 29(4) leg amputated. The goal of MVF is to restore the efference– afference loop that has been interrupted. MVF of the intact limb deceives the individual and elicits the awareness that the amputated limb is still intact, not at least due to the dominance of the visual system over other modalities.8,9 Indeed, when the illusion was tested in arm amputees with complaints of “clenching spasms” and phantom limb pain, the spasms were eliminated and the pain was relieved immediately after exposure to MVF in a proportion of the sample.3 The novelty and simplicity of the idea, in combination with the far-reaching potential of MVF, prompted clinicians and researchers to replicate the initial findings of Ramachandran and colleagues. Consistent with the earlier observations, follow-up studies have confirmed that MVF treatment has the capacity to reduce phantom limb pain intensity and duration.10,11 Moreover, the notion that many neurological disorders with unilateral pain and motor symptoms may be (partly) caused by maladaptive cortical reorganization involving a disruption of the efference–afference loop, led others to apply MVF to a wide range of conditions. Hemiparesis after stroke is perhaps the most striking example. In a proportion of the patients the paresis is thought to be a form of “learned paralysis” due to a nonpermanent blocking of corticofugal fibers by swelling after the trauma.12 A recent Cochrane Review exploring the effectiveness of MVF therapy in patients after stroke (13 randomized controlled trials, 506 patients) concluded that mirror therapy indeed might be more effective in promoting motor function than a control intervention† when used as an adjunct to conventional therapy.13 Furthermore, the metaanalysis indicated that the effects were retained, up to 6 months after the intervention, and that MVF therapy had a significantly greater effect than control interventions on activities of daily living and on pain, though the latter was found only in a subgroup with CRPS after stroke. To date, MVF is administered to treat various unilateral pain and/or motor disorders, including CRPS,14-18 hemiparesis after stroke,19,20 reduced mobility after wrist fracture,21 and spastic hemiparetic cerebral palsy (SHCP).22,‡ The findings of these studies tend to corroborate the initial work, that is, a reduction in pain and improvement in motor function. Still, it should be noted that publication bias toward a *In their recent review article, Ramachandran and Altshuler9 recognize that the origin of phantom pain is still poorly understood and may be related to other factors, for example, persistence of preamputation pain and pathological “remapping” among others (see also Ramachandran and Hirstein6). The rationale to use MVF, however, is based on the notion of a mismatch between motor output and visual and/or proprioceptive feedback. † The effect of MVF therapy was significantly larger than control interventions.

351

Deconinck et al selection of positive results may be likely and additional placebo-controlled studies are needed for all conditions or symptoms. In this respect, it is worth mentioning that Brodie et al found that the attenuating effect of MVF on pain was not stronger than a control condition in lower limb amputees.23 Despite the fact that there appear to be parallels in the pathophysiology of unilateral pain and motor disorders as described by Ramachandran and Altschuler,12 the 2 phenomena should be considered separately, hence the focus of this review will be on the effects of MVF on sensorimotor control. To fully exploit the potential of MVF, a better insight into the processes that underlie the beneficial effects on motor function is required. Not only would this knowledge advance our theoretical understanding of the brain, it may also provide guidelines as to when MVF may be useful and how it should be applied. When the individual is required to perform bilateral, symmetrical motor tasks, MVF therapy may be considered a special form of bilateral training, and hence exploit similar mechanisms (see Cauraugh and Summers for a review24). However, in search of the added value of MVF researchers have invoked 3 (not mutually exclusive) hypotheses to account for the positive effects of MVF on motor recovery. A first hypothesis relates to the mirror neuron system.17,19,20 Mirror neurons fire both when an individual observes an action and when he/she performs a similar action. The network, including the premotor cortex, supplementary motor area, inferior frontal gyrus, and inferior parietal lobule of the brain, is thought to play an important role in action recognition and motor learning or rehabilitation.25 An observation/execution matching mechanism, whereby action observation activates crucial parts of the motor system, is hypothesized to induce motor learning.25,26 It is known that action observation facilitates the corticospinal pathway and this paradigm is already used in neurorehabilitation as mental practice aimed at improving motor function.27 According to this hypothesis, a “mirror box” is a means to facilitate action observation and therefore MVF is thought to activate the mirror neuron system in a similar way to action observation (Hypothesis 1). In line with this is the notion that MVF may elicit or enhance motor imagery,28 that is, internal simulation of movement without overt action. Just like action observation motor imagery has been attributed therapeutic capacities because it activates neural circuits involved in motor control.29 ‡

See Ramachandran and Altschuler9 for a list of clinical cases where the use of MVF has been observed informally but has not been described in the literature.

Second, MVF might promote recruitment of ipsilateral motor pathways.30 These motor pathways, originating in the unaffected hemisphere and projecting ipsilaterally to the paretic body-side, have been attributed a nontrivial role in the restoration of motor function in hemiparesis.31-34 It is hypothesized that MVF might facilitate the unmasking of “dormant” ipsilateral projections, which are normally inhibited (Hypothesis 2). Finally, MVF or the associated illusion is thought to increase an individual’s (spatial) attention toward the unseen (affected) limb.35 It is known that hemiparetic patients may end up in a state of “learned nonuse,” by continuously avoiding the use of the paretic hand or by pathophysiological disruption of the efference–afference loop.36 In keeping with the rationale for using constraint-induced movement therapy,36 the increased attention toward the affected limb, mediated by the illusory image of a “healed” paretic limb, may activate motor networks (Hypothesis 3). In sum, an increasing body of evidence underpins the potential of MVF to facilitate recovery of motor function. Still, the neural mechanism of MVF, whether the behavioral effect is accompanied by neuroplastic changes, and what this reorganization would involve is unclear. The hypotheses invoked to explain MVF effects are based on known concepts in neurorehabilitation, but they remain speculative. Recent experimental neuroimaging research has begun to reveal the extent of brain activation during movement with MVF, and its modulatory effects on brain processes compared with normal visual conditions. In the following part of this article, the findings of these studies will be systematically reviewed. This will serve as a validity test of the proposed hypotheses.

Part 2: The Neural Correlates of Mirror Visual Feedback Purpose The purpose of this systematic review was to identify the areas in the brain that are differentially affected or modulated by MVF compared with a condition with normal or without visual feedback.

Literature Search A literature search using the electronic databases PubMed, CINAHL, and EMBASE (1972 to January 2014) was conducted. Search terms included “mirror therapy” or “mirror visual feedback” combined with “functional magnetic resonance imaging (fMRI),” “positron emission topography (PET),” “transcranial magnetic stimulation (TMS),” “magnetoencephalography (MEG),” “electroencephalography

352

Neurorehabilitation and Neural Repair 29(4)

Search of bibliographic databases: PubMed, CINAHL, EMBASE + Personal database

Outcome N = 347 Databases: N = 341 records Personal databases: N = 6 records 1st stage screening of titles and abstracts by 2 independent reviewers

Main reasons for exclusion: - Methodology did not involve MVF - Methodology did not involve neuroimaging

N = 41 records 2nd stage screening of full texts by 2 independent reviewers Final selection: N = 33, including 12 fMRI, 9 TMS, 5 MEG, 4 EEG, 2 PET, and 2 NIRS

Main reasons for exclusion: - Neuroimaging only for diagnostic purposes, not for measuring effect MVF (N = 2) - Effect of MVF on pain or tactile perception, not motor control (N = 4) - See Supplementary Material for details.

Figure 1.  Flow diagram of the article selection process.

(EEG),” or “near infra-red spectrometry (NIRS).” In addition, we checked our personal database and the reference list of included articles. Our search was restricted to peerreviewed full articles written in English. Inclusion criteria were the following: •• Experimental studies or clinical trials •• Normal and/or motor-impaired human participants •• Use of neuroimaging techniques (fMRI, PET, MEG, EEG, NIRS) or TMS to study the effect of MVF* on cortical activation (and related motor performance or perceptual measures) Exclusion criterion was the following: •• Studies that do not assess effect of MVF on sensorimotor control, but focus on pain and/or tactile perception. The records identified by this search were screened independently by 2 authors of this systematic review (FD and AS) in 2 stages: a first stage screening of titles and abstracts and a second stage using the full text of the remaining article. The lists of eligible articles identified by the independent reviewers were compared, and any disagreements were resolved through discussion (and referral to the text of the articles in question).

Results The electronic database search yielded 347 unique articles, of which 33 were deemed eligible for this systematic review (see Figure 1 for an overview of the selection process). Across the selected articles, the most commonly used scanning technique was fMRI (12 studies).37-47 MEG was the neuroimaging modality in 5 article,48-52 EEG in 4,53-56 PET in 2,57,58 and NIRS in 2.59,60 Nine studies investigated the effect of MVF on cortical activation with TMS47,61-68; 1 study used both TMS and fMRI.47 In Tables 1 to 4, the included articles and their methodologies are listed according to modality. The majority of the studies (n = 27) examined immediate modulatory effects when exposed to MVF, of which 22 focused on healthy individuals and 5 on stroke patients (Tables 1-3). In 16 studies, MVF was provided in a bilateral fashion, that is, not obscuring the active hand. Six studies assessed neuroplastic changes in response to a bout of practice or an intervention in healthy individuals (n = 3) or stroke patients (n = 3). Here, bilateral MVF was used in all but one study (Table 4 and Supplementary Material). The variance in methodology and motor task, which may be unilateral or contralateral, complicates the overall interpretation of the observations. A factor that *Mirror visual feedback could be induced by a real mirror or using a virtual reality environment.

353

Deconinck et al

Table 1.  Outline of the Methodology, Type of MVF, Measures to Avoid Systematic Variation in Movement Across Conditions, and Known Potential Confounds in All Experimental Studies Using fMRI Included in the Systematic Review. Study Reference

Participants

Outline of Experimental Methodology Relevant to This Review

Type of MVF

(a) Measures to Avoid Systematic Variation (b) Potential Confounds

Diers et al (2010)37

9 healthy individuals aged 51.9 ± 6.9

Repetitive clenching of the fist at dominant side with (1) normal VF, (2) MVF of active hand, or (3) during motor imagery

Bilateral

Dohle et al (2004)38

6 healthy individuals aged 29.0 ± 1.5 15 healthy individuals aged 22-56

Unilateral Unilateral

(a) Metronome paced

18 healthy individuals aged 22-48 3 healthy individuals aged 26.8 and 1 stroke patient aged 70.0

Static condition vs finger-thumb opposition with (1) virtual VF of hand; (2) virtual MVF of hand Unilateral index finger-thumb opposition of either left or right hand with (1) virtual VF, (2) virtual MVF Tapping right index finger with (1) VF active hand; (2) MVF active hand Finger tapping right hand with (1) virtual VF active hand; (2) virtual MVF active hand

(a) Metronome paced (b) No statistical comparison between conditions (a) Metronome paced

Bilateral

(a) None

Unilateral

18 stroke patients aged 54.7 ± 9.9 5 healthy individuals aged 21-57 and 2 patients with brain tumor aged 47 and 67

Unimanual (nonaffected hand) and bimanual clenching movement with (1) VF; (2) MVF Unimanual and bimanual clenching of hand(s) with (1) eyes closed; (2) MVF of (one of the) active hand(s)

Bilateral

(a) None (b) Lack of control condition (a) Metronome paced

Wang et al (2013)42

15 healthy individuals aged 22-56

Unilateral

Wang et al (2013)43

15 healthy individuals aged 22-56 and 5 stroke patients aged 50-72

Movement performance task: Unilateral index finger-thumb opposition of either left or right hand with (1) virtual VF, (2) virtual MVF. Movement observation task: Observation of similar action on video Unilateral index finger-thumb opposition of either left or right hand with (1) virtual VF, (2) virtual MVF

Fritzsch et al (2013)39 Matthys et al (2009)46 Merians et al (2009)47 Michielsen et al (2011)40 Shinoura et al (2008)41

Bilateral

Unilateral

(a) None (b) No statistical comparison between conditions (a) Number of repetitions counted post hoc

(a) Number of repetitions counted post hoc

Abbreviations: MVF, mirror visual feedback; VF, normal visual feedback. Bilateral MVF, active hand and its mirror reflection; Unilateral MVF, MVF of active hand only. MVF was generated by a mirror placed in between the arms in the sagittal plane, except when stated otherwise.

adds to this difficulty is the absence of a standard control condition. In some studies, MVF is contrasted with normal full vision of the 2 limbs, whereas in others visual feedback of the active or static limb only is used as a control. Furthermore, to isolate the effect of MVF on brain activation it is crucial to keep the movement (range, frequency, intensity) constant across MVF and control conditions, which a number of studies have failed to do or to control for (see Table 1-4 for details). Instant Neuromodulatory Effects of MVF.  The neuromodulatory effects of MVF refer to changes in activity or excitation that are evoked by MVF, that is, revealed by direct comparison of MVF with a control condition (listed in Table 5). MVF evokes a conflict between expected and actual feedback (cognitive conflict) and between visual and kinaesthetic feedback (perceptual conflict). This conflict is most obvious when performing unimanual or asymmetric bimanual tasks, although even during symmetric bimanual actions, the perfect interlimb symmetry is perceived as surreal. In this latter case, MVF is accompanied with an increase of activity within the superior parietal lobe (precuneus [bilateral] and superior posterior parietal cortex [contralateral]40), the posterior cingulate cortex,58 and ipsilateral lateral sulcus51,52 compared to a condition with full vision of the 2 hands. MVF inducing a more extreme conflict (during unimanual actions) stimulates activity within primary visual and

somatosensory areas, as well as higher order processing areas in the occipital and parietal cortex ipsilateral to the moving limb.38,39,42,43,46,60 In fact, the mirror inverts the lateralization that is normally associated with the presentation of a right or left hand in these regions. Note, however, that these observations stem from studies that contrasted virtual MVF of the active hand with virtual VF displayed in the frontal plane (on a screen or onto MRI-compatible goggles). Other studies show increased activation in the right dorsolateral prefrontal cortex (DLPFC),58 the contralateral secondary sensory cortex (SII),51 the ipsilateral superior temporal gyrus (STG),46 and the contralateral insular cortex.52,57 Finally, MVF modulates activity of the primary motor cortex, both ipsilateral and contralateral to the active hand reflected into the mirror, though the findings are somewhat inconsistent. The majority of the studies suggest an increase in ipsilateral M1 excitability or increased ipsilateral activation as indicated by a change in laterality index (vs a control condition).47,53,55,56,63,65-68 Hadoush et al claim that this effect is more prominent for unimanual MVF, that is, when the active hand (reflected into the mirror) is covered.48 Others show that the increase in excitability is observed particularly when MVF and motor imagery are combined.66,68 Investigation of the potential mediators of this effect suggests that MVF has the capacity to neutralize the (interhemispheric) inhibition from contralateral to ipsilateral hemisphere. Still, it should be noted that a number of

354

Neurorehabilitation and Neural Repair 29(4)

Table 2.  Outline of the Methodology, Type of MVF, Measures to Avoid Systematic Variation in Movement Across Conditions, and Known Potential Confounds in All Experimental Studies Using TMS Included in the Systematic Review.

Study Reference Carson and Ruddy (2012)

Participants 63

12 healthy individuals aged 21.5 ± 3.4

Fukumura et al (2007)68

6 healthy individuals aged 20-39

Funase et al (2007)64

12 healthy individuals aged 19-40

Garry et al (2005)65

8 healthy individuals aged 39.6 ± 14.5

Kang et al (2011)66

30 healthy individuals aged 28.0 ± 2.4 and 30 stroke patients aged 66.0 ± 11.0

 

Kang et al (2012)67

18 healthy individuals aged 30.9 ± 2.2 and 18 stroke patients aged 61.3 ± 11.6

 

Merians et al (2009)47

1 healthy individual

Outline of Experimental Methodology Relevant to This Review Exp 1: Flexion-extension of left wrist with (1) VF of nonmoving right hand; (2) MVF of moving left hand; (3) no VF Flexion-extension of left wrist with (1) VF left hand; (2) VF left + motor imagery right hand; (3) MVF left hand; (4) MVF left + motor imagery right hand; (5) MVF left + passive movement right hand; (6) MVF + passive movement and motor imagery right hand Flexion-extension left hand or left index finger and wrist with (1) VF left side; (2) MVF left side Finger-thumb opposition with (1) VF active hand; (2) VF inactive hand; (3) neutral VF; (4) MVF active hand Thumb abduction/adduction with (1) VF active hand + motor imagery; (2) MVF active hand + motor imagery inactive hand; (3) MVF active hand + motor imagery of asymmetric movements inactive hand Passive task with (1) motor imagery of right hand; (2) AO other hand + motor imagery own right hand; (3) MVF + motor imagery other hand Exp 1: TMS across nondominant/affected hemisphere during unilateral flexionextension of the dominant/unaffected wrist with (1) MVF, (2) virtual MVF, or during (3) relaxation Exp 2: TMS across nondominant/affected hemisphere during unilateral flexionextension of the dominant/unaffected wrist with (1) continuous virtual MVF, (2) intermittent virtual MVF Finger tapping right hand with (1) virtual VF active hand; (2) virtual MVF active hand

(a) Measures to Avoid Systematic Variation

Type of MVF

(b) Potential Confounds

Bilateral

(a) Metronome paced, EMG recording

Bilateral

(a) Metronome paced

Bilateral

(a) Metronome paced

Bilateral

(a) Metronome paced, EMG recording

Bilateral

(a) Metronome paced

 

Bilateral

(a) Metronome paced

 

Bilateral

(a) None

Abbreviations: MVF, mirror visual feedback; VF, normal visual feedback; AO, action observation; Exp, experiment. Bilateral MVF, active hand and its mirror reflection; Unilateral MVF, MVF of active hand only. MVF was generated by a mirror placed in between the arms in the sagittal plane, except when stated otherwise.

studies could not find changes in activity within the ipsilateral M1 evoked by MVF.39,60,64 Neuroplastic Effects Due to Practice or Intervention With MVF.  Six studies have examined the effect of a bout of practice or treatment with MVF on motor function while also measuring the change in brain activity pre and post training (see Table 4 for details).44,45,54,61,62,69 Consistent with earlier reports training with MVF resulted in a gain in motor function of the untrained or affected hand in all studies. This gain seems to be related to an enhanced excitatory function of the (contralateral) corticospinal pathway projecting to this hand (ie, decrease in motor threshold and intracortical inhibition within M1 ipsilateral to the trained hand).61,62

After training the activation balance when moving the affected hand has shifted toward M1 of the affected hemisphere, indicating increased activation of the affected side and/or decreased activation of the contralesional side.45,54,69 This would imply a reestablishment the hemispheric balance that was disrupted by the insult. Another study suggests, however, that the improvement of untrained hand is related to the establishment of a functional connection between this hand and the ipsilateral motor cortex (ie, M1 contralateral to the trained hand).44 Dynamic causal modeling and functional connectivity analysis further indicates that this reorganization is mediated by enhanced connectivity between the premotor cortices (both left and right) and the ipsilateral supplementary motor area.

355

Deconinck et al

Table 3.  Outline of the Methodology, Type of MVF, Measures to Avoid Systematic Variation in Movement Across Conditions, and Known Potential Confounds in All Experimental Studies Using PET, MEG, EEG, and NIRS Included in the Systematic Review.

Study Reference Dohle et al (2011)

Modality 57

PET

Fink et al (1999)58

PET

Hadoush et al (2013)48

MEG

Tominaga et al (2009)49

MEG

Tominaga et al (2011)50

MEG

Wasaka and Kakigi (2012)52

MEG

Wasaka and Kakigi (2012)51

MEG

Praamstra et al (2011)55

EEG

Touzalin-Chretien et al (2008)56

EEG

Touzalin-Chretien et al (2010)53

EEG

Imai et al (2008)59

NIRS

Mehnert et al (2013)60

NIRS

Participants

Outline of Experimental Methodology Relevant to This Review

10 healthy individuals aged 19-42 10 healthy individuals aged 20-73

Unilateral circular movement of left or right arm with (1) VF; (2) virtual MVF Luria’s bimanual circle drawing task: (1) in-phase with VF; (2) in- phase with MVF; (3) out-of-phase with VF; (4) out-of-phase with MVF 10 healthy individuals Index finger extension with MVF of active aged 22-35 hand, (1) with VF of active hand or (2) without VF of active hand 11 healthy individuals Stimulation of right median nerve at wrist aged 19-34 while holding a pencil with (1) MVF of right hand; (2) VF of right hand; (3) MVF of left hand; (4) VF of left hand 13 healthy individuals Stimulation of right or left median nerve at aged 19-34 wrist while holding a pencil with (1) MVF of right hand; (2) VF of right hand; (3) MVF of left hand; (4) VF of left hand 10 healthy individuals Stimulation of left median nerve at wrist aged 24-46 while flexing-extending left thumb repetitively or at rest with (1) MVF of stationary right hand; (2) VF of both hands 10 healthy individuals Stimulation of left median nerve at wrist aged 28-46 while flexing-extending left and right thumb repetitively symmetrically or asymmetrically with (1) MVF of right hand; (2) VF of both hands 9 healthy individuals aged Extension-flexion right or left index finger 32 ± 11 with (1) VF active finger; (2) MVF active finger 11 healthy individuals Key press with right hand [except (4)] with aged 27.7 (1) VF of right hand; (2) MVF of right hand; (3) MVF of right hand in a frontal mirror; (4) VF of left hand 8 healthy individuals aged Key press with right hand [except (2)] with 32.2 (1) VF of right hand; (2) VF of left hand; (3) MVF of right hand; (4,5) MVF of right hand with left hand in incongruent position; (6) MVF of right hand in a frontal mirror 5 healthy individuals aged Unilateral grasping movements with right or 21.1 ± 1.1 left hand with (1) VF of nonmoving hand, (2) with MVF of moving limb 22 healthy individuals Unilateral index finger-thumb opposition of aged 21-40 either left or right hand with (1) virtual VF, (2) virtual MVF

(a) Measures to Avoid Systematic Variation Type of MVF (b) Potential Confounds Unilateral

(a) Metronome paced

Bilateral

(a) None

Bilateral

(a) None

Unilateral

(a) N/A

Unilateral

(a) N/A

Bilateral

(a) None

Bilateral

(a) None

Unilateral

(a) Set number of trials, EMG recording

Unilateral

(a) Set number of trials, EMG recording

Unilateral

(a) Set number of trials, EMG recording

Bilateral

(a) Metronome paced

Unilateral

(a) Metronome paced

Abbreviations: MVF, mirror visual feedback; VF, normal visual feedback; PET, positron emission topography; MEG, magnetoencephalography; EEG, electroencephalography; NIRS, near infrared spectrometry.

Discussion This systematic review aimed to identify instant and longterm neuromodulatory effects associated with MVF. The findings of the 33 articles indicate that MVF engages a distributed network within the brain, including regions related to perceptual, motor, and higher cognitive functions, though it is important to acknowledge a number of limitations. Clearly, the variety of methodologies impedes the ability to

draw firm conclusions. In addition, a large number of studies had relatively small sample sizes (N ≤ 10 in 17 out of 33 studies) and a number of studies fail to meet all methodological requirements, for example, equal performance across experimental manipulations or adequate control intervention (see Tables 1-3 and Supplementary Material). These limitations highlight that future research should primarily focus on isolating the effect of MVF from the

356

Neurorehabilitation and Neural Repair 29(4)

Table 4.  Outline of the Methodology in All Clinical or Training Studies Included in the Systematic Review. Study Reference

Modality

Bae et al (2012)54

EEG

Bhasin et al (2012)69

fMRI

Hamzei et al (2012)44

fMRI

Laeppchen et al (2012)61

TMS

Michielsen et al (2011)45

fMRI

Nojima et al (2012)62

TMS

 

Participants

Nature of Training

Frequency and Duration

Outline of Experimental Methodology Relevant to This Review

Observation of hand movements MG: 10 stroke patients MG: bilateral movements 4 weeks, 5 sessions/ during rest, while measuring µaged 55.2 ± 8.5 with MVF week, 30 minutes/ rhythm across left (C3), right (C4) session CG: 10 stroke patients CG: unilateral and (Cz) central fissure prior to and aged 52.6 ± 11.2 movements of paretic after training side with VF of 2 limbs Repetitive clenching/extension of fist MG: 20 stroke patients Bilateral hand exercises 8 weeks, 5 sessions/ at the affected side without VF preweek, 30-60 minutes/ aged 28-62 with virtual MVF on and posttraining session laptop screen 4 days, 20 minutes/day fMRI during (1) AO of grasping MG: 13 healthy MG and CG: 5 movement during rest, (2) active individuals aged 23.9; unimanual skills (RH imitation of grasping movement only) with MVF or VF CG: 13 healthy with left or right hand, pre- and of RH only individuals aged 25.5 posttraining 4 days, 20 minutes/day Test of TMS parameters (MT, ICI, MG and& CG: 5 MG and CG: 10 ICF, IHI left hemisphere) pre- and unimanual skills (RH healthy individuals posttraining only) with MVF or VF aged 24.1 of RH only 6 weeks, 5 sessions/ fMRI during clenching movement with MG: 9 stroke patients MG and CG: bimanual week, 60 minutes/day affected hand without VF pre- and aged 51.9 ± 9.3 exercises with MVF or posttraining VF of 2 hands CG: 7 stroke patients aged 59.0 ± 10.4 Test of TMS parameters (MT, ICI, IHI) Exp 1. MG and CG: 10 MG and CG: unimanual 10 sets of 30 minutes each pre- and postpractice (RH only) ball handling healthy individuals task with MVF or VF of (young adults) 2 hands Similar to Exp 1. Practice was followed Exp 2. MG1 and MG2: MG1 = MG2: unimanual 10 sets of 30 minutes each by cTBS over M1 (MG1) or occipital (RH only) ball handling 8 healthy individuals cortex (MG2) task with MVF (young adults)

Abbreviations: EEG, electroencephalography; fMRI, functional magnetic resonance imaging; TMS, transcranial magnetic stimulation; MG, mirror group; CG, control group; MVF, mirror visual feedback; VF, normal visual feedback; Exp, experiment; AO, action observation; MT, motor threshold; ICI, intracortical inhibition; ICF, intracortical facilitation; IHI, interhemispheric inhibition; cTBS, continuous theta burst stimulation. MVF was generated by a mirror placed in between the arms in the sagittal plane, except when stated otherwise.

Table 5.  Brain Areas Differentially Activated With Mirror Visual Feedback (MVF) Compared to Control Conditions, Corresponding Brodmann Area (BA) Where Known, Side of Activation Listed According to Populationa,b. Population Healthy                     Stroke    

Area Dorsolateral prefrontal cortex Precentral gyrus (M1) Postcentral sulcus, posterior wall (S1/S2) Lateral sulcus, upper wall (S2) Superior posterior parietal cortex Precuneus (V6) Superior temporal gyrus Cuneus/Lingual gyrus (V1/V2) Superior/Middle occipital gyrus (V2, 3, 5) Fusiform gyrus (V4) Insular cortex, posterior region Precentral gyrus (M1) Precuneus (V6) Posterior cingulate cortex

BA 9/46 4 2 7 7 39 17/18 19 37 4 7 30

Side 58

Right Ipsi41,47,63,65-68 Contra52 Contra51,52 Ipsi58 Ipsi38,42,43,60 Ipsi46 Ipsi38,42 Ipsi38,42,46 Ipsi38,42 Contra57 Ipsi66,67 Contra40, Ipsi40,43 Contra40

MVF

Network

Bi Uni and Bi Uni and Bi Bi Bi Uni and Bi Bi Uni Uni and Bi Uni Uni Bilateral Uni and Bi Bi

Attention Motor Attention Attention Attention Attention MNS Attention Attention Attention Attention Motor Attention Attention

Abbreviation: MNS, mirror neuron system. a The 2 final columns indicate the type of MVF that elicits this observation and the network to which specific areas are considered to belong. b Contralateral (contra) and ipsilateral (ipsi) are defined with respect to the (moving) limb that is reflected in the mirror. MVF indicates the type of MVF (unilateral or bilateral) that elicits this observation.

357

Deconinck et al mechanisms associated with bilateral training. Despite these limitations, this first systematic review of the effect of MVF on the brain reveals useful information with respect to the underlying mechanisms of MVF for both researchers and practitioners in neurorehabilitation. Overall, the findings suggest that MVF may affect perceptuo-motor control processes via (parts of) 3 functional networks. First, the increase in activity in primary and secondary visual and somatosensory areas suggests a rise in attentional resources to resolve the perceptual incongruence.51,52 This is associated with conscious awareness of sensory feedback or control of agency, as observed in the activity within the insular cortex,70 and enhanced monitoring of the movement, as found in the involvement of the right DLPFC.58,71 Furthermore, greater activation of the posterior aspect of the parietal and cingulate cortex supports the notion of greater attentional demands. The posterior cingulate cortex, highly interconnected with various brain regions, is considered a hub for information exchange72 and a prominent role in the cognitive control of behavior is attributed to this region.73 The nearby superior posterior parietal cortex and its medial extension (precuneus) are known to be involved in visuospatial information processing and directing spatial attention, especially during bimanual coordination tasks.74,75 This probably explains why the effect of MVF on the precuneus is primarily observed in studies that use virtual MVF of the hand in the frontal plane.38,39,42,43,60 Notably, a number of studies have shown that the precuneus is particularly active during motor imagery,76 which may account for the combined effect of imagery and MVF. Second, MVF seems to cause increased immediate activation of STG46 and elevated engagement of PMC44 after training. Both areas have been associated with the mirror neuron system. STG is often linked to its neighboring superior temporal sulcus and is involved in the visual identification of biological motion.77 Combined with the PMC, it forms a network that subserves the imitation of biological motion and the acquisition of motor skills.25,78,79 The activation of PMC, especially at the side of the lesion, is in keeping with earlier research that has attributed a prominent role to this region in motor recovery after stroke.80,81 A third functional network on which MVF appears to exert a modulatory effect is the motor network. The primary motor cortex (M1) ipsilateral to the active (reflected) hand, that is, M1 projecting to the unseen hand behind the mirror, is considered the final common pathway for the beneficial effect of MVF according to various studies. A complete reversal of lateralization when moving a limb that is reflected by a mirror (ie, the ipsilateral side taking over control), as put forward by some,53,56,59 is unlikely; however, there is accumulating evidence that MVF decreases the motor threshold and enhances corticospinal output of the ipsilesional M1 in stroke patients.55,63,65-68 This is probably

mediated through a reduction in interhemispheric inhibition (from contralateral/lesional to ipsilateral/lesional)63 and/or a reduction of intracortical inhibition.61 In view of the notion that functional recovery is correlated with the extent of involvement of the ipsilesional (here: contralateral) motor network,82 this seems a promising therapeutic effect of MVF. Other findings in healthy adults, however, indicate that improved motor skill of the untrained hand is achieved by establishing a functional connection with the ipsilateral motor cortex via MVF, a mechanism that has been associated with poorer motor recovery compared with normalization of the hemispheric balance.83 This discrepancy might reflect 2 stages in the recovery process or a population-specific response and warrants further investigation. The effect of MVF on brain activation is likely dependent on the specific nature of the feedback. In this respect it is remarkable that the effect on primary and secondary visual processing areas is primarily related to unilateral MVF. Bimanual MVF, in contrast, seems to engage more frontal and parietal regions related to higher cognitive functions like attention and monitoring (see Table 5). Finally, it is noteworthy that the findings for individuals with stroke are in accordance with those for healthy people, insofar as the limited number of studies allows this comparison (see Table 5). There is evidence of increased activation of higher order areas involved with attentional processes (precuneus and posterior cingulate cortex) and the ipsilateral M1.

Part 3: Summary and Future Directions Convergent evidence suggests that MVF may be used as a tool to promote functional recovery in patients with unilateral motor impairments. A systematic review of neuroimaging research was conducted to test the validity of the 3 hypotheses proposed to explain the positive effects associated with MVF. The findings of this review, suggesting substantial overlap between MVF-related activity and regions subserving attention-related processes, confirm that MVF activates a broad network dedicated to attention and action monitoring (Hypothesis 3). This is consistent with known motor learning principles, which attribute success of motor practice to attentional focus and cognitive processing. Furthermore, the positive effect on motor function is associated with facilitation of M1 contralateral to the affected or untrained hand (here: referred to as ipsilateral to the moving hand that is mirrored). However, there is also evidence to support a mechanism that exploits ipsilateral control of the affected limb, which has been associated with suboptimal recovery after other therapeutic interventions (Hypothesis 2). Regions that have been linked with the mirror neuron system (PMC, STG) may play a mediating role in connecting perceptual and motor areas (Hypothesis 1). Still, the

358 current evidence indicates that MVF therapy is certainly not a substitute for observational therapy or motor imagery, given that MVF activates only isolated parts of the MNS. Future research using recent advances in graph theory may elucidate functional connectivity within and between the involved networks.84 To date the majority of evidence stems from studies in healthy adult individuals and the few studies that have examined a patient population only considered people who survived a stroke. It remains unclear to what extent these hypotheses may be valid for other clinical conditions for which MVF has been suggested an adjunct to conventional therapy (eg, SHCP22,85-87 and CRPS14-18). The finding that MVF may have an impact on multiple functional networks may mean it can serve as a versatile tool to promote motor recovery, of which the actual mechanism is dependent on the specific condition or damage. Largescale clinical trials that include measurement of brain function and structure are to examine the efficacy and the underlying mechanisms of MVF in different populations, and potential differences between them. Although further research is warranted to fully understand and exploit the potential of MVF in neurorehabilitation, it is indisputable that MVF can exert a strong modulatory influence on the motor system. Acknowledgments The authors wish to thank Prof Karen Caeyenberghs and two anonymous reviewers for their advice and suggestions.

Declaration of Conflicting Interests The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was partly supported by a research grant from Sparks, registered charity 1003825 (England & Wales), to Geert Savelsbergh and Frederik Deconinck (Grant 09MMU01).

Supplementary Material Supplementary material for this article is available on the Neurorehabilitation & Neural Repair Web site at http://nnr.sagepub.com/content/by/supplemental-data

References 1. Thompson P, Papadopoulou G, Vassiliou E. The origins of entasis: illusion, aesthetics or engineering? Spat Vis. 2007;20:531-543. doi:10.1163/156856807782758359. 2. Gregory RL. Perceptual illusions and brain models. Proc R Soc Lond B Biol Sci. 1968;171:279-296. 3. Ramachandran VS, Rogers-Ramachandran D, Cobb S. Touching the phantom limb. Nature. 1995;377:489-490.

Neurorehabilitation and Neural Repair 29(4) 4. Von Holst E, Mittelstaedt H. The principle of reafference: interactions between the central nervous system and the peripheral organs. Naturwissenschaften. 1950;37:464-476. 5. Von Helmholtz H. Handbuch der physiologischen Optik. Leipzig, Germany: Voss; 1867. 6. Ramachandran VS, Hirstein W. The perception of phantom limbs (The D. O. Hebb lecture). Brain. 1998;121:16031630. 7. Harris AJ. Cortical origin of pathological pain. Lancet. 1999;354:1464-1466. doi:10.1016/S0140-6736(99)05003-5. 8. Gibson JJ. Observations on active touch. Psychol Rev. 1962;69:477-491. 9. Mechsner F, Kerzel D, Knoblich G, Prinz W. Perceptual basis of bimanual coordination. Nature. 2001;414:69-73. doi:10.1038/35102060. 10. Chan BL, Witt R, Charrow AP, et al. Mirror therapy for phantom limb pain. N Engl J Med. 2007;357:2206-2207. doi:10.1056/NEJMc071927. 11. Darnall BD. Self-delivered home-based mirror therapy for lower limb phantom pain. Am J Phys Med Rehabil. 2009;88:78-81. doi:10.1097/PHM.0b013e318191105b. 12. Ramachandran VS, Altschuler EL. The use of visual feedback, in particular mirror visual feedback, in restoring brain function. Brain. 2009;132(pt 7):1693-1710. doi:10.1093/ brain/awp135. 13. Thieme H, Mehrholz J, Pohl M, Behrens J, Dohle C. Mirror therapy for improving motor function after stroke. Cochrane Database Syst Rev. 2012;(3):CD008449. 14. Selles RW, Schreuders TAR, Stam HJ. Mirror therapy in patients with causalgia (complex regional pain syndrome type II) following peripheral nerve injury: two cases. J Rehabil Med. 2008;40:312-314. doi:10.2340/16501977-0158. 15. Karmakar A, Lieberman I. Mirror box therapy for complex regional pain syndrome. Anaesthesia. 2006;61:412-413. 16. Tichelaar YIGV, Geertzen JHB, Keizer D, van Wilgen PC. Mirror box therapy added to cognitive behavioural therapy in three chronic complex regional pain syndrome type I patients: a pilot study. Int J Rehabil Res. 2007;30:181-188. doi:10.1097/MRR.0b013e32813a2e4b. 17. Rosen B, Lundborg G. Training with a mirror in rehabilitation of the hand. Scand J Plast Reconstr Surg Hand Surg. 2005;39:104-108. 18. McCabe CS, Haigh RC, Ring EFJ, Halligan PW, Wall PD, Blake DR. A controlled pilot study of the utility of mirror visual feedback in the treatment of complex regional pain syndrome (type 1). Rheumatology. 2002;42:97-101. doi:10.1093/ rheumatology/keg041. 19. Sütbeyaz S, Yavuzer G, Sezer N, Koseoglu BF. Mirror therapy enhances lower-extremity motor recovery and motor functioning after stroke: a randomized controlled trial. Arch Phys Med Rehabil. 2007;88:555-559. doi:10.1016/ j.apmr.2007.02.034. 20. Yavuzer G, Selles R, Sezer N, et al. Mirror therapy improves hand function in subacute stroke: a randomized controlled trial. Arch Phys Med Rehabil. 2008;89:393-398. doi:10.1016/ j.apmr.2007.08.162. 21. Altschuler EL, Hu J. Mirror therapy in a patient with a fractured wrist and no active wrist extension. Scand J Plast Reconstr Surg Hand Surg. 2008;42:110-111. doi:10.1080/02844310701510355.

Deconinck et al 22. Gygax Jequier M, Schneider P, Newman CJ. Mirror therapy in children with hemiplegia: a pilot study. Dev Med Child Neurol. 2011;53:473-476. doi:10.1111/j.1469-8749.2011.03924.x. 23. Brodie EE, Whyte A, Niven CA. Analgesia through the looking-glass? A randomized controlled trial investigating the effect of viewing a “virtual” limb upon phantom limb pain, sensation and movement. Eur J Pain. 2007;11:428-436. doi:10.1016/j.ejpain.2006.06.002. 24. Cauraugh JH, Summers JJ. Neural plasticity and bilateral movements: a rehabilitation approach for chronic stroke. Prog Neurobiol. 2005;75:309-320. doi:10.1016/j.pneurobio.2005.04.001. 25. Buccino G, Solodkin A, Small SL. Functions of the mirror neuron system: implications for neurorehabilitation. Cogn Behav Neurol. 2006;19:55-63. 26. Iacoboni M. Cortical mechanisms of human imitation. Science. 1999;286:2526-2528. doi:10.1126/science.286.5449.2526. 27. Pomeroy VM, Clark CA, Miller JSG, Baron JC, Markus HS, Tallis RC. The potential for utilizing the “mirror neurone system” to enhance recovery of the severely affected upper limb early after stroke: a review and hypothesis. Neurorehabil Neural Repair. 2005;19:4-13. doi:10.1177/1545968304274351. 28. Stevens JA, Stoykov MEP. Using motor imagery in the rehabilitation of hemiparesis. Arch Phys Med Rehabil. 2003;84:1090-1092. doi:10.1016/S0003-9993(03)00042-X. 29. Grèzes J, Decety J. Functional anatomy of execution, mental simulation, observation, and verb generation of actions: a meta-analysis. Hum Brain Mapp. 2001;12:1-19. 30. Ezendam D, Bongers R, Jannink M. Systematic review of the effectiveness of mirror therapy in upper extremity function. Disabil Rehabil. 2009;31:2135-2149. doi:10.1080/09638280902887768. 31. Carr LJ, Harrison LM, Stephens JA. Evidence for bilateral innervation of certain homologous motoneurone pools in man. J Physiol. 1994;475:217-227. 32. Staudt M, Grodd W, Gerloff C, Erb M, Stitz J, Krägeloh-Mann I. Two types of ipsilateral reorganization in congenital hemiparesis: a TMS and fMRI study. Brain. 2002;125(pt 10):2222-2237. 33. Schwerin S, Dewald JPA, Hatzi M, Jovanovich S, Nickeas M, MacKinnon C. Ipsilateral versus contralateral cortical motor projections to a shoulder adductor in chronic hemiparetic stroke: implications for the expression of arm synergies. Exp Brain Res. 2008;185:509-519. doi:10.1007/s00221-0071169-8. 34. Benecke R, Meyer BU, Freund HJ. Reorganisation of descending motor pathways in patients after hemispherectomy and severe hemispheric lesions demonstrated by magnetic brain stimulation. Exp Brain Res. 1991;83:419-428. 35. Dohle C, Püllen J, Nakaten A, Küst J, Rietz C, Karbe H. Mirror therapy promotes recovery from severe hemiparesis: a randomized controlled trial. Neurorehabil Neural Repair. 2009;23:209-217. doi:10.1177/1545968308324786. 36. Taub E, Crago JE, Uswatte G. Constraint-induced movement therapy: a new approach to treatment in physical rehabilitation. Rehabil Psychol. 1998;43:152-170. doi:10.1037//00905550.43.2.152. 37. Diers M, Christmann C, Koeppe C, Ruf M, Flor H. Mirrored, imagined and executed movements differentially activate sensorimotor cortex in amputees with and without phan-

359 tom limb pain. Pain. 2010;149:296-304. doi:10.1016/ j.pain.2010.02.020. 38. Dohle C, Kleiser R, Seitz RJ, Freund HJ. Body scheme gates visual processing. J Neurophysiol. 2004;91:2376-2379. doi:10.1152/jn.00929.2003. 39. Fritzsch C, Wang J, Dos Santos LF, Mauritz KH, Brunetti M, Dohle C. Different effects of the mirror illusion on motor and somatosensory processing. Restor Neurol Neurosci. 2013;32:269-280. 40. Michielsen ME, Smits M, Ribbers GM, et al. The neuronal correlates of mirror therapy: an fMRI study on mirror induced visual illusions in patients with stroke. J Neurol Neurosurg Psychiatry. 2011;82:393-398. 41. Shinoura N, Suzuki Y, Watanabe Y, et al. Mirror ther apy activates outside of cerebellum and ipsilateral M1. NeuroRehabilitation. 2008;23:245-252. 42. Wang J, Fritzsch C, Bernarding J, et al. A comparison of neural mechanisms in mirror therapy and movement observation therapy. J Rehabil Med. 2013;45:410-413. doi:10.2340/16501977-1127. 43. Wang J, Fritzsch C, Bernarding J, et al. Cerebral activation evoked by the mirror illusion of the hand in stroke patients compared to normal subjects. NeuroRehabilitation. 2013;33:593-603. 44. Hamzei F, Läppchen CH, Glauche V, Mader I, Rijntjes M, Weiller C. Functional plasticity induced by mirror training: the mirror as the element connecting both hands to one hemisphere. Neurorehabil Neural Repair. 2012;26:484-496. doi:10.1177/1545968311427917. 45. Michielsen ME, Selles RW, van der Geest JN, et al. Motor recovery and cortical reorganization after mirror therapy in chronic stroke patients: a phase II randomized controlled trial. Neurorehabil Neural Repair. 2011;25:223-233. doi:10.1177/1545968310385127. 46. Matthys K, Smits M, Van der Geest JN, et al. Mirror-induced visual illusion of hand movements: a functional magnetic resonance imaging study. Arch Phys Med Rehabil. 2009;90:675681. doi:10.1016/j.apmr.2008.09.571. 47. Merians AS, Tunik E, Fluet GG, Qiu Q, Adamovich S V. Innovative approaches to the rehabilitation of upper extremity hemiparesis using virtual environments. Eur J Phys Rehabil Med. 2009;45:123-133. 48. Hadoush H, Mano H, Sunagawa T, Nakanishi K, Ochi M. Optimization of mirror therapy to excite ipsilateral primary motor cortex. NeuroRehabilitation. 2013;32:617-624. 49. Tominaga W, Matsubayashi J, Deguchi Y, et al. A mirror reflection of a hand modulates stimulus-induced 20-Hz activity. Neuroimage. 2009;46:500-504. doi:10.1016/j.neuroimage.2009.02.021. 50. Tominaga W, Matsubayashi J, Furuya M, et al. Asymmetric activation of the primary motor cortex during observation of a mirror reflection of a hand. PLoS One. 2011;6:e28226. doi:10.1371/journal.pone.0028226. 51. Wasaka T, Kakigi R. The effect of unpredicted visual feedback on activation in the secondary somatosensory cortex during movement execution. BMC Neurosci. 2012;13:138. doi:10.1186/1471-2202-13-138. 52. Wasaka T, Kakigi R. Conflict caused by visual feedback modulates activation in somatosensory areas during movement

360 execution. Neuroimage. 2012;59:1501-1507. doi:10.1016/ j.neuroimage.2011.08.024. 53. Touzalin-Chretien P, Ehrler S, Dufour A. Dominance of vision over proprioception on motor programming: evidence from ERP. Cereb Cortex. 2010;20:2007-2016. doi:10.1093/ cercor/bhp271. 54. Bae SH, Jeong WS, Kim KY. Effects of mirror therapy on subacute stroke patients’ brain waves and upper extremity functions. J Phys Ther Sci. 2012;24:1119-1122. doi:10.1589/ jpts.24.1119. 55. Praamstra P, Torney L, Rawle CJ, Chris Miall R. Misconceptions about mirror-induced motor cortex activation. Cereb Cortex. 2011;21:1935-1940. 56. Touzalin-Chretien P, Dufour A. Motor cortex activation induced by a mirror : evidence from lateralized readiness potentials. J Neurophysiol. 2008;100:19-23. doi:10.1152/ jn.90260.2008. 57. Dohle C, Stephan KM, Valvoda JT, et al. Representation of virtual arm movements in precuneus. Exp Brain Res. 2011;208:543-555. 58. Fink GR, Marshall JC, Halligan PW, et al. The neural consequences of conflict between intention and the senses. Brain. 1999;122:497-512. 59. Imai I, Takeda K, Shiomi T, Taniguchi T, Kato H. Sensorimotor cortex activation during mirror therapy in healthy right-handed subjects: a study with near-infrared spectroscopy. J Phys Ther Sci. 2008;20:141-145. 60. Mehnert J, Brunetti M, Steinbrink J, Niedeggen M, Dohle C. Effect of a mirror-like illusion on activation in the precuneus assessed with functional near-infrared spectroscopy. J Biomed Opt. 2013;18:66001. 61. Laeppchen CH, Ringer T, Blessin J, et al. Optical illu sion alters M1 excitability after mirror therapy—a TMS study. J Neurophysiol. 2012;108:2857-2861. doi:10.1152/ jn.00321.2012. 62. Nojima I, Mima T, Koganemaru S, Thabit MN, Fukuyama H, Kawamata T. Human motor plasticity induced by mirror visual feedback. J Neurosci. 2012;32:1293-1300. doi:10.1523/JNEUROSCI.5364-11.2012. 63. Carson RG, Ruddy KL. Vision modulates corticospinal suppression in a functionally specific manner during movement of the opposite limb. J Neurosci. 2012;32:646-652. doi:10.1523/JNEUROSCI.4435-11.2012. 64. Funase K, Tabira T, Higashi T, Liang N, Kasai T. Increased corticospinal excitability during direct observation of self-movement and indirect observation with a mirror box. Neurosci Lett. 2007;419:108-112. doi:10.1016/j.neulet.2007.04.025. 65. Garry MI, Loftus A, Summers JJ. Mirror, mirror on the wall: viewing a mirror reflection of unilateral hand movements facilitates ipsilateral M1 excitability. Exp Brain Res. 2005;163:118-122. doi:10.1007/s00221-005-2226-9. 66. Kang YJ, Ku J, Kim HJ, Park HK. Facilitation of corticospinal excitability according to motor imagery and mirror therapy in healthy subjects and stroke patients. Ann Rehabil Med. 2011;35:747-758. doi:10.5535/arm.2011.35.6.747. 67. Kang YJ, Park HK, Kim HJ, et al. Upper extremity rehabilitation of stroke: facilitation of corticospinal excitability using virtual mirror paradigm. J Neuroeng Rehabil. 2012;9:71.

Neurorehabilitation and Neural Repair 29(4) 68. Fukumura K, Sugawara K, Tanabe S, Ushiba J, Tomita Y. Influence of mirror therapy on human motor cortex. Int J Neurosci. 2007;117:1039-1048. 69. Bhasin A, Padma Srivastava M, Kumaran S, Bhatia R, Mohanty S. Neural interface of mirror therapy in chronic stroke patients: a functional magnetic resonance imaging study. Neurol India. 2013;60:570-576. 70. Farrer C, Frith CD. Experiencing oneself vs another person as being the cause of an action: the neural correlates of the experience of agency. Neuroimage. 2002;15:596-603. doi:10.1006/nimg.2001.1009. 71. Mima T, Nagamine T, Nakamura K, Shibasaki H. Attention modulates both primary and second somatosensory cortical activities in humans : a magnetoencephalographic study. J Neurophysiol. 1998;80:2215-2221. 72. Hagmann P, Cammoun L, Gigandet X, et al. Mapping the structural core of human cerebral cortex. PLoS Biol. 2008;6:e159. doi:10.1371/journal.pbio.0060159. 73. Leech R, Braga R, Sharp DJ. Echoes of the brain within the posterior cingulate cortex. J Neurosci. 2012;32:215-222. doi:10.1523/JNEUROSCI.3689-11.2012. 74. Andersen RA. Encoding of intention and spatial location in the posterior parietal cortex. Cereb Cortex. 1995;5:457-469. 75. Wenderoth N, Debaere F, Sunaert S, Swinnen SP. The role of anterior cingulate cortex and precuneus in the coordination of motor behaviour. Eur J Neurosci. 2005;22:235-246. doi:10.1111/j.1460-9568.2005.04176.x. 76. Cavanna AE, Trimble MR. The precuneus: a review of its functional anatomy and behavioural correlates. Brain. 2006;129(pt 3):564-583. doi:10.1093/brain/awl004. 77. Schultz J, Imamizu H, Kawato M, Frith CD. Activation of the human superior temporal gyrus during observation of goal attribution by intentional objects. J Cogn Neurosci. 2004;16:1695-1705. doi:10.1162/0898929042947874. 78. Iacoboni M, Koski LM, Brass M, Bekkering H, Woods RP, Dubeau M. Reafferent copies of imitated actions in the right. Proc Natl Acad Sci USA. 2001;98:13995-13999. 79. Schultz J, Friston KJ, O’Doherty J, Wolpert DM, Frith CD. Activation in posterior superior temporal sulcus parallels parameter inducing the percept of animacy. Neuron. 2005;45:625-635. doi:10.1016/j.neuron.2004.12.052. 80. Levy CE, Nichols DS, Schmalbrock PM, Keller P, Chakeres DW. Functional MRI evidence of cortical reorganization in upper-limb stroke hemiplegia treated with constraint-induced movement therapy. Am J Phys Med Rehabil. 2001;80:4-12. 81. Liepert J, Bauder H, Miltner WHR, Taub E, Weiller C. Treatment-induced cortical reorganization after stroke in humans. Stroke. 2000;31:1210-1216. doi:10.1161/01. STR.31.6.1210. 82. Calautti C, Baron JC. Functional neuroimaging studies of motor recovery after stroke in adults: a review. Stroke. 2003;34:15531566. doi:10.1161/01.STR.0000071761.36075.A6. 83. Ward NS, Brown MM, Thompson AJ, Frackowiak RSJ. Neural correlates of motor recovery after stroke: a longitudinal fMRI study. Brain. 2003;126(pt 11):2476-2496. doi:10.1093/brain/awg245. 84. Bullmore E, Sporns O. Complex brain networks: graph theoretical analysis of structural and functional systems. Nat Rev Neurosci. 2009;10:186-198. doi:10.1038/nrn2575.

Deconinck et al 85. Feltham MG, Ledebt A, Bennett SJ, Deconinck FJA, Verheul MHG, Savelsbergh GJP. The “mirror box” illusion: effect of visual information on bimanual coordination in children with spastic hemiparetic cerebral palsy. Motor Control. 2010;14:68-82. 86. Feltham MG, Ledebt A, Deconinck FJA, Savelsbergh GJP. Mirror visual feedback induces lower neuromuscu-

361 lar activity in children with spastic hemiparetic cerebral palsy. Res Dev Disabil. 2010;31:1525-1535. doi:10.1016/ j.ridd.2010.06.004. 87. Smorenburg ARP, Ledebt A, Deconinck FJA, Savelsbergh GJP. Practicing a matching movement with a mirror in individuals with spastic hemiplegia. Res Dev Disabil. 2013;34:2507-2513.