Review
The neural network of motor imagery: An ALE meta-analysis

https://doi.org/10.1016/j.neubiorev.2013.03.017Get rights and content

Highlights

  • Motor imagery activates fronto-parietal, subcortical and cerebellar regions.

  • The motor imagery network includes regions involved during actual motor execution.

  • The primary motor cortex is not consistently activated during motor imagery.

  • Consistency of activations is modulated by the type of movements.

  • Consistency of activations is modulated by the nature of the motor imagery task.

Abstract

Motor imagery (MI) or the mental simulation of action is now increasingly being studied using neuroimaging techniques such as positron emission tomography and functional magnetic resonance imaging. The booming interest in capturing the neural underpinning of MI has provided a large amount of data which until now have never been quantitatively summarized. The aim of this activation likelihood estimation (ALE) meta-analysis was to provide a map of the brain structures involved in MI. Combining the data from 75 papers revealed that MI consistently recruits a large fronto-parietal network in addition to subcortical and cerebellar regions. Although the primary motor cortex was not shown to be consistently activated, the MI network includes several regions which are known to play a role during actual motor execution. The body part involved in the movements, the modality of MI and the nature of the MI tasks used all seem to influence the consistency of activation within the general MI network. In addition to providing the first quantitative cortical map of MI, we highlight methodological issues that should be addressed in future research.

Introduction

Remember the days when you were still a child stuck in a classroom, imagining being elsewhere, maybe fighting a blaze as a fireman, dancing on the world's biggest stages or perhaps for some, conducting intriguing science experiments. Imagining doing something is an important cognitive process that is used throughout our lifespan. Motor imagery (MI), which refers to the act of imagining a specific action without actually executing it, has fascinated scientists from a wide range of domains including sport sciences, psychology and neuroscience. However, its study has been challenging due to its covert nature: how can we measure something related to the motor domain in someone who is explicitly asked not to move? If the measure of behavioral parameters such as the temporal characteristics of imagined movements (through mental chronometry paradigms) and physiological signals provided a very interesting window into this process (see (Collet et al., 2011) for an overview), it is the development of brain imaging techniques such as positron emission tomography (PET) and functional magnetic resonance imaging (fMRI) that brought scientists ever closer to “seeing how” people are doing MI. These technological advances were paralleled by a booming interest in the study of MI as can be seen by the important increase in publications related to MI since 1995 (see Fig. 1). With these new tools, neuroscientists are now at the forefront of the scientific challenge of understanding how our brain subserves the act of imagining movements.

Motor imagery has been defined as the conscious mental simulation of actions involving our brain's motor representations in a way that is similar to when we actually perform movements (Jeannerod and Decety, 1995). This has led many to suggest that MI and motor execution rely on similar neural structures and processes (Grezes and Decety, 2001, Jeannerod, 2001, Jeannerod and Decety, 1995, Munzert et al., 2009). Numerous studies have tried to verify this claim by identifying the neural substrate of MI using neuroimaging techniques. An exhaustive overview of the results from these studies was presented more than 10 years ago by Grezes and Decety (2001) who suggested that MI recruits the primary motor cortex, premotor cortex (PMc), supplementary motor area (SMA), anterior cingulate cortex (ACC), inferior and superior parietal lobules (IPL/SPL) and the cerebellum (CB).

However, one has to be careful when trying to describe the network underlying a brain function by combining results from individual neuroimaging studies using a descriptive (as opposed to quantitative) approach such as the one used by Grezes and Decety (2001). Indeed, the statistical power of fMRI and PET studies remains relatively low, in part because of their small sample sizes (see (Yarkoni, 2009) for a commentary on this topic). Furthermore, as neuroimaging data is particularly sensitive to task and control condition selection, which often varies across studies, results obtained with a contrast approach (Task A > Control condition) are difficult to generalize. Thus, trying to get a precise and clear view of which regions are “truly” involved in the general process of MI by looking separately at data from individual studies is often impractical and potentially misleading. Large-scale synthesis methods are now available and offer new opportunities to probe and make sense of the immense data produced by the imaging community (e.g., Eickhoff et al., 2009, Wager et al., 2009, Yarkoni et al., 2011). Hence the aim of this paper is to use a meta-analytic approach to map the regions involved in MI and to assess the modulating effects of specific methodological variables that have already been topics of interest in the field of MI.

Previous work have already investigated the influence of variables such as which body parts the participants are imagining moving (Ehrsson et al., 2003, Szameitat et al., 2007) or the modality/strategy of MI (i.e., kinesthetic vs. visual MI) they use (Guillot et al., 2009). Another variable that has been less documented but could be of interest for the study of MI is the influence of the MI task the participants are instructed to complete. Indeed, if most studies have asked their participants to imagine single actions, others have instructed their participants to imagine motor sequences (often simple or complex series of finger movements). Others, instead of explicitly asking their participants to perform motor imagery, instruct them to identify the laterality of a limb shown in different orientations: a task thought to be solved using motor imagery. Since relatively few studies have looked at each of these variables, clear conclusion on how they influence brain activity during MI is still sparse. Furthermore, taken separately, each of these studies was done with relatively small samples. Meta-analyses make possible the study of these modulating variables at a much larger scale as they combine and simultaneously analyze the findings from several studies (and thus use much larger samples). Also, meta-analyses can consider data from a study even if this particular study did not directly look at the variable. For example, trying to get a sense of the influence of the modality of MI, each study that used kinesthetic MI will be compared to each study that used visual MI even if these studies did not compare the two modalities directly.

As most reviews on MI have discussed results from neuroimaging studies using a descriptive approach, there is a great need for a fresh and quantitative look at the work that has been conducted on MI using fMRI and PET. First, results from this meta-analysis should offer answers to important questions in the field. Indeed, the involvement of the primary motor cortex is still a subject of debate and the differences/similarities in the involvement of the motor system during different types of MI remains an open question. Furthermore, looking at the methodological characteristics of a large sample of studies will highlight possible shortcomings related to MI research and offer possible methodological changes that could benefit future work on MI. Lastly, as MI is increasingly proposed as a possible rehabilitation tool to train the motor system (Flor and Diers, 2009, Jackson et al., 2001, Malouin and Richards, 2010, Mulder, 2007), and even to gain access to patients in neurovegetative states (Cruse and Owen, 2010), an empirical description of the neural system supporting MI should offer clinical neuroscientists useful normative data, for instance to study neural activation during MI in clinical populations, or to assess the efficacy of training programs using MI.

Based upon the work by Grezes and Decety (2001), it is expected that studies on MI should consistently report activity in regions involved in the actual production of movements such as the SMA, the PMc, IPL, SPL and the CB, and that variations in the type of MI performed should modulate the extent of this activation.

Section snippets

Article selection

Articles were obtained through an online search of the PubMed database (http://www.ncbi.nlm.nih.gov/pubmed/) using the terms “fMRI” or “PET” and “motor imagery” during the month of September 2011. In addition, reference sections of the reviewed articles were carefully inspected to identify additional articles of interest. Selected articles had to: (1) use either fMRI or PET; (2) present data on a MI task; (3) report activations in the form of stereotaxic coordinates of foci based on either the

Number of articles and brief description

Our search resulted in 75 articles that corresponded to our inclusion criteria. The median size of their samples was 12 subjects (range: 5–60). Twelve articles used PET while the remainder used fMRI. Twenty five articles used rest as the control condition, 13 used a fixation condition (cross or scrambled image), six used a visual imagery condition (imagining a moving or static object), one used both a visual imagery and fixation condition, eight used a motor imagery of static self condition,

Discussion

By combining the results from previous fMRI and PET studies, our general ALE meta-analysis provides a quantitative map of the brain regions involved in MI. Further meta-analyses offer evidence that certain methodological variables can modulate which parts of the general MI network is recruited.

Conclusion

Up to now, the description of the neural network supporting MI could only be inferred by examining the results from neuroimaging studies individually. Our ALE meta-analysis, by greatly reducing the influence of the many confounding factors associated with the diverse methods of previous fMRI/PET studies, provides the first quantitative map of the structures activated during MI. MI seems to rely on a network involving motor related regions including fronto-parietal areas and subcortical

Acknowledgments

Supported by graduate scholarships from the Fonds de recherche du Québec – Santé (S.H.), the Canadian Institutes of Health Research (CIHR) (S.H.; M.-P.C.), the Centre interdisciplinaire de recherche en réadaptation et intégration sociale (CIRRIS) (S.H.; M.G.), and the Faculté des sciences sociales de l’Université Laval (M.G.), by postdoctoral fellowships from the CIRRIS (A.S) and the Ministère du Développement économique, de l’Innovation et de l’Exportation (MDEIE) (A.S.), and by a CIHR New

References (135)

  • F.P. de Lange et al.

    Posture influences motor imagery: an fMRI study

    Neuroimage

    (2006)
  • F.P. de Lange et al.

    Motor imagery: a window into the mechanisms and alterations of the motor system

    Cortex

    (2008)
  • R.C. deCharms et al.

    Learned regulation of spatially localized brain activation using real-time fMRI

    Neuroimage

    (2004)
  • P. Dechent et al.

    Is the human primary motor cortex involved in motor imagery?

    Brain Res. Cogn. Brain Res.

    (2004)
  • M.P. Deiber et al.

    Cerebral processes related to visuomotor imagery and generation of simple finger movements studied with positron emission tomography

    Neuroimage

    (1998)
  • S.B. Eickhoff et al.

    Assignment of functional activations to probabilistic cytoarchitectonic areas revisited

    Neuroimage

    (2007)
  • S.B. Eickhoff et al.

    A new SPM toolbox for combining probabilistic cytoarchitectonic maps and functional imaging data

    Neuroimage

    (2005)
  • L. Fogassi et al.

    Motor functions of the parietal lobe

    Curr. Opin. Neurobiol.

    (2005)
  • A.D. Fourkas et al.

    Influence of imagined posture and imagery modality on corticospinal excitability

    Behav. Brain Res.

    (2006)
  • G. Ganis et al.

    Visual imagery in cerebral visual dysfunction

    Neurol. Clin.

    (2003)
  • Q. Gao et al.

    Evaluation of effective connectivity of motor areas during motor imagery and execution using conditional Granger causality

    Neuroimage

    (2011)
  • G. Goldenberg

    Apraxia and the parietal lobes

    Neuropsychologia

    (2009)
  • M.A. Grealy et al.

    An automatic-voluntary dissociation and mental imagery disturbance following a cerebellar lesion

    Neuropsychologia

    (2011)
  • S. Grillner et al.

    Mechanisms for selection of basic motor programs – roles for the striatum and pallidum

    Trends Neurosci.

    (2005)
  • A. Guillot et al.

    Functional neuroanatomical networks associated with expertise in motor imagery

    Neuroimage

    (2008)
  • R.C. Helmich et al.

    Cerebral compensation during motor imagery in Parkinson's disease

    Neuropsychologia

    (2007)
  • S. Higuchi et al.

    Cerebellar activity evoked by common tool-use execution and imagery tasks: an fMRI study

    Cortex

    (2007)
  • E. Hoshi et al.

    Distinctions between dorsal and ventral premotor areas: anatomical connectivity and functional properties

    Curr. Opin. Neurobiol.

    (2007)
  • A. Jackson et al.

    Synchrony between neurons with similar muscle fields in monkey motor cortex

    Neuron

    (2003)
  • P.L. Jackson et al.

    Potential role of mental practice using motor imagery in neurologic rehabilitation

    Arch. Phys. Med. Rehabil.

    (2001)
  • M. Jeannerod

    Neural simulation of action: a unifying mechanism for motor cognition

    Neuroimage

    (2001)
  • M. Jeannerod et al.

    Mental motor imagery: a window into the representational stages of action

    Curr. Opin. Neurobiol.

    (1995)
  • S.H. Johnson et al.

    Selective activation of a parietofrontal circuit during implicitly imagined prehension

    Neuroimage

    (2002)
  • C.H. Kasess et al.

    The suppressive influence of SMA on M1 in motor imagery revealed by fMRI and dynamic causal modeling

    Neuroimage

    (2008)
  • C. la Fougere et al.

    Real versus imagined locomotion: a [18F]-FDG PET-fMRI comparison

    Neuroimage

    (2010)
  • M.F. Lafleur et al.

    Motor learning produces parallel dynamic functional changes during the execution and imagination of sequential foot movements

    Neuroimage

    (2002)
  • M. Lotze et al.

    Motor imagery

    J. Physiol. Paris

    (2006)
  • F. Malouin et al.

    Normal aging and motor imagery vividness: implications for mental practice training in rehabilitation

    Arch. Phys. Med. Rehabil.

    (2010)
  • J. Milton et al.

    Imaging motor imagery: methodological issues related to expertise

    Methods

    (2008)
  • J. Munzert et al.

    Cognitive motor processes: the role of motor imagery in the study of motor representations

    Brain Res. Rev.

    (2009)
  • R. Newman-Norlund et al.

    The role of inferior frontal and parietal areas in differentiating meaningful and meaningless object-directed actions

    Brain Res.

    (2010)
  • A. Parkinson et al.

    Parietal cortex coding of limb posture: in search of the body-schema

    Neuropsychologia

    (2010)
  • L.M. Parsons

    Imagined spatial transformations of one's hands and feet

    Cogn. Psychol.

    (1987)
  • L.M. Parsons

    Integrating cognitive psychology neurology and neuroimaging

    Acta Psychol. (Amst)

    (2001)
  • M. Alam et al.

    The pedunculopontine nucleus area: critical evaluation of interspecies differences relevant for its use as a target for deep brain stimulation

    Brain

    (2011)
  • F. Binkofski et al.

    Broca's region subserves imagery of motion: a combined cytoarchitectonic and fMRI study

    Hum. Brain Mapp.

    (2000)
  • G. Buccino et al.

    Action observation activates premotor and parietal areas in a somatotopic manner: an fMRI study

    Eur. J. Neurosci.

    (2001)
  • L. Cattaneo et al.

    The mirror neuron system

    Arch. Neurol.

    (2009)
  • Y. Chang et al.

    Neural correlates of motor imagery for elite archers

    NMR Biomed.

    (2011)
  • C. Collet et al.

    Measuring motor imagery using psychometric, behavioral, and psychophysiological tools

    Exerc. Sport Sci. Rev.

    (2011)
  • Cited by (626)

    View all citing articles on Scopus
    View full text