ReviewContributions of the basal ganglia and functionally related brain structures to motor learning
Introduction
Our capacity to plan, learn and retain new motor skills is essential for going through daily activities. Indeed, such ability is regularly elicited when, for example, one executes smooth eye–body coordinated actions like hitting a golf ball with a club, or when one produces multi-joint movements while reaching and grasping small objects in space. Accordingly, a great deal of research in this field has been devoted to better understand the behavioral determinants and the neural substrates that mediate this type of procedural memory. Motor learning has been experimentally studied using tasks designed to measure the incremental acquisition of sequential movements into a well-articulated behavior (motor sequence learning [MSL]), or to tests our capacity to compensate for environmental changes (motor adaptation [MA]). MSL paradigms typically require subjects to produce a sequence of movements that they either know explicitly before training (e.g., [22], [24], [29], [56], [57], [83], [103], [114]), learn implicitly through repeated practice (e.g., [3], [22], [40], [88], [93]), discover by trial and error [53], [54], [55], [89], [106], or acquire through probabilistic rules [68], [82]. Motor responses in those sequence learning tasks involve either finger-to-thumb opposition movements (e.g., [56], [57], [95]), finger presses on response boxes (e.g., [24], [39]), movements of the whole arm (e.g., [22], [40]) oculomotor sequential movements [3], [4], or varying the isometric force applied to a pressure plate to follow a repeating waveform pattern [33], [34], [40]. By contrast, MA paradigms necessitate that participants adapt to changes in visual rotations (i.e., kinematic MA measure) [51], [52], [108] or to perturbations applied by a velocity-dependent field that pushes the hand perpendicular to its direction of motion (dynamic MA measure) (e.g., [62], [96], [97], [99], [101], see [108] and [98] for reviews). Operationally, the acquisition of such motor abilities is generally measured by a reduction in reaction time and number of errors, and/or by a change in movement synergy and kinematics (see [19], [25], [61], [98] for reviews). Furthermore, these changes in performance are known to be incremental, implicit in nature, long lasting and dependent upon “on-line” and “off-line” triggered plastic changes in the brain.
Psychophysical studies have demonstrated that the incremental acquisition of motor skills follows several phases: First, an early, fast learning stage in which considerable improvement in performance can be seen within a single training session; and second, a later, slow stage in which further gains can be observed across several sessions of practice (e.g., [57]). Interestingly, an intermediate phase corresponding to the consolidation process of the motor routine has also been proposed, based on the demonstration that a motor memory trace can be disrupted by the administration of a competing task within a time window of 6–8 h, or when spontaneous performance gains are reported following a latent post-training period of more than 6 h without additional practice on the task. Once consolidated, the motor memory trace is believed to be resistant to interference [5], and to become readily retrievable despite long periods of time without additional training. Finally, motor skilled behaviors are regarded as fully automatized when actions are carried out effortlessly with little attentional resources needed for their successful completion.
Over the years, work ranging from electrophysiological and lesion experiments in animals to clinical population-based and imaging studies in humans has undoubtedly demonstrated that the basal ganglia, and the striatum in particular, play a critical role in the planning, learning, and execution of a new motor skill. The basal ganglia are composed of a series of subcortical nuclei that are organized into sensorimotor, associative and limbic territories based upon their anatomical connectivity and functions. The caudate nucleus, putamen and subthalamic nucleus constitute the input nuclei as they receive major afferent connections from the cerebral cortex, midbrain and thalamus, whereas the internal segment of the globus pallidus and the substantia nigra, pars reticulata form the output nuclei that send back treated information to frontal cortical areas via thalamic nuclei [1], [15]. Ample evidence indicates that the processing of motor information flows through a topographically organized and segregated loop linking motor-related cortical regions like the primary (M1), supplementary (SMA), premotor (PMC) and cingulate (CMA) motor areas with the sensorimotor divisions of the basal ganglia and thalamus [77]. Finally, a distinction between the anterior associative putamen region and the more posteroventral sensorimotor area of the putamen and globus pallidus has also been observed in humans based on diffusion imaging data [66], hence forming the anatomical basis for the functional dissociations seen between these putamen areas during the early learning and planning phase preceding the execution of learned motor sequences, for example [8], [21], [30], [65].
Yet, the cortico-basal ganglia circuits do not constitute the only anatomical system implicated in the acquisition and planning of skilled actions. The cerebellum and its motor-associated structures, like the somatosensory motor cortex and ventral PMC forming the cortico-cerebellar loop through the dentate nucleus and ventral-posterior lateral nucleus of the thalamus [59], have also been shown to contribute to motor learning. Dynamic brain plasticity within the striatum and cerebellum, as well as functional interactions between these two cortico-subcortical systems has been reported depending on the stage of the learning process, nature (i.e., new versus learned motor behavior) of the action being planned and the type of skill being acquired (MSL, MA). Furthermore, consistent with recent electrophysiological studies in primates [13], findings from functional magnetic resonance imaging (fMRI) studies conducted at 3.0 T (e.g., [91]) and proper correlation analyses of the BOLD changes occurring during the very early stage of the learning process [4] suggest that the hippocampus contributes also to the encoding and consolidation of motor skills.
In this review, we will thus describe the brain plasticity, i.e., the reorganization over time of brain circuits, involved in motor skill learning. We will focus on the dynamic changes that are observed within the cortico-striatal system as one is learning or planning to execute a newly learned motor behavior up to when the skill is consolidated or has become highly automatized. Imaging work from Doyon’s laboratory and from other research groups will be described. A special emphasis will be put on MSL paradigms, although results from imaging studies using MA tasks will also be reported to highlight the role that the cortico-striatal system plays in this other type of motor learning. Using standard contrast and correlation statistical analyses, as well as functional connectivity approaches, we will also discuss the functional interplay that exists between the cortico-striatal, cortico-cerebellar and limbic systems in this form of learning. Finally, these imaging results will be put into a plausible neurobiological model of motor skill learning [21], [23], [25], which proposes an integrated view of the brain plasticity mediating this form of procedural memory at different stages of the acquisition process. Due to space limitations, however, this review does not describe the work on the acquisition of arbitrary visuomotor conditional associations, normal motor prehension, object use, imitation and apraxia (see [38] for a review of the relevant literature on these related issues).
Section snippets
Motor sequence learning
From the beginning, the field of motor learning has been dominated by studies that looked only at the neural plasticity that occur in the fast learning stage where improvements in performance are most dramatic. Indeed, since early 1990s, a plethora of neuroimaging studies using positron emission tomography (PET) and fMRI have investigated the brain plasticity mediating performance changes seen during early encoding of a new motor skilled behavior. These studies have demonstrated that the
Intra-system brain plasticity associated with motor learning
In addition to the brain plasticity that occurs at the systems level, accumulating data from work in animals and humans indicate that dynamic changes in motor representations during motor learning also take place within the cortico-striatal system and cerebellum (e.g., see [21], [43], [48] for reviews). For example, in Lehéricy et al.’s study [65] described above, a gradual shift of increased activity within the putamen was observed as subjects were practicing the finger sequence task in the
Planning motor sequences
Before executing sequential behaviors that are part of our repertoire of motor skills, one regularly needs to plan individual movement elements into a properly timed and spatially ordered sequence. Such a faculty is crucial, as it allows us to “anticipate events, select movements, specify their ordering and better control actions online” [30]. Although not studied as well as the learning process itself, recent experimental work has given us some clues regarding the brain structures that could
Motor adaptation
A large body of evidence indicates that the neuronal substrate primarily responsible for the encoding, consolidation and long-term storage of adapted movements comprises the cerebellum and related structures [14], [16], [44], [101] see [19], [21], [23], [25], [61], [98] for reviews). Support to this statement comes from clinical population studies, which have demonstrated that while patients with Parkinson’s or Huntington’s disease show intact performance on paradigms designed to measure MA
Motor memory consolidation
Data accumulated so far clearly demonstrate that motor sequence learning depends initially on repeated practice, but that it also continues to develop over time after training has ended. During this latent post-learning phase, the memory of a given motor experience is thought to be transformed into a robust and enduring state, a process called “memory consolidation” [58], [110]. Motor memory consolidation possibly begins as early as after subjects have done a few practice trials, and thus after
Automatization of motor skills
To investigate the neural substrates associated with the “automatic” performance of a motor skill, researchers have used one of two main experimental designs: the first uses a dual-task paradigm to determine whether or not a secondary task can be performed with minimal interference on the motor learning (primary) task of interest. The main problem with this approach is that it is difficult, if not impossible, to make sure that performance on a motor skill has become completely automatic after
Motor skill learning: A model
The large number of studies described above have not only helped us to identify the brain systems that contribute differentially to MSL and MA, but have also provided valuable information with regards to the functional dynamic changes that occur within the cortico-striatal, cortico-cerebellar and limbic systems during the different learning stages of a motor skill (e.g., see [20], [21], [23], [25], [45], [98] for reviews). To put these results into a plausible neurobiological model, Doyon et
Acknowledgements
We wish to thank VA Nguyen for his technical assistance in preparing the manuscript. This work was supported, in part, by grants from the Canadian Institute of Health Research (CIHR), the Natural Sciences and Engineering Research Council of Canada (NSERC), and from the “Ministère du Développement Économique, de l’Innovation et de l’Exportation (MDEIE) to J. Doyon and colleagues.
References (114)
The basal ganglia: a neural network with more than motor function
Semin Pediatr Neurol
(2003)- et al.
Motor skill learning in Parkinson’s disease
J Neurol Sci
(1996) - et al.
Implicit oculomotor sequence learning in humans: time course of offline processing
Brain Res
(2006) - et al.
Both the hippocampus and striatum are involved in consolidation of motor sequence memory
Neuron
(2008) - et al.
fMRI investigation of cortical and subcortical networks in the learning of abstract and effector-specific representations of motor sequences
Neuroimage
(2006) - et al.
Identification of large-scale networks in the brain using fMRI
Neuroimage
(2006) - et al.
A role of the basal ganglia and midbrain nuclei for initiation of motor sequences
Neuroimage
(2008) - et al.
The preparation and execution of self-initiated and externally-triggered movement: a study of event-related fMRI
Neuroimage
(2002) - et al.
The preparation and readiness for voluntary movement: a high-field event-related fMRI study of the Bereitschafts-BOLD response
Neuroimage
(2003) - et al.
Premovement activity of the pre-supplementary motor area and the readiness for action: studies of time-resolved event-related functional MRI
Hum Mov Sci
(2005)
Complementary roles of basal ganglia and cerebellum in learning and motor control
Curr Opin Neurobiol
Skill learning
Int Rev Neurobiol
Reorganization and plasticity in the adult brain during learning of motor skills
Curr Opin Neurobiol
Distinct contribution of the cortico-striatal and cortico-cerebellar systems to motor skill learning
Neuropsychologia
Reconsolidation: the advantage of being refocused
Curr Opin Neurobiol
Complex spatio-temporal dynamics of fMRI BOLD: a study of motor learning
Neuroimage
From preparation to online control: reappraisal of neural circuitry mediating internally generated and externally guided actions
Neuroimage
Evidence for a distributed hierarchy of action representation in the brain
Hum Mov Sci
Learning-related fMRI activation associated with a rotational visuo-motor transformation
Brain Res Cogn Brain Res
Motor learning in man: a review of functional and clinical studies
J Physiol Paris
Central mechanisms of motor skill learning
Curr Opin Neurobiol
Towards a computational neuropsychology of action
Prog Brain Res
Distinct contribution of the striatum and cerebellum to motor learning
Brain Cogn
Exploring large-scale brain networks in functional MRI
J Physiol Paris
Somatotopically arranged inputs from putamen and subthalamic nucleus to primary motor cortex
Neurosci Res
Learned material content and acquisition level modulate cerebral reactivation during posttraining rapid-eye-movements sleep
Neuroimage
The time course of off-line motor sequence learning
Brain Res Cogn Brain Res
Awareness modifies the skill-learning benefits of sleep
Curr Biol
An FMRI study of the role of the medial temporal lobe in implicit and explicit sequence learning
Neuron
Interference to consolidation phase gains in learning a novel movement sequence by handwriting: dependence on laterality and the level of experience with the written sequence
Exp Brain Res
Striatal contribution to sleep-dependent motor consolidation
Neuroimage (Suppl.)
Learning-related coordination of striatal and hippocampal theta rhythms during acquisition of a procedural maze task
Proc Natl Acad Sci USA
Time course of changes in brain activity and functional connectivity associated with long-term adaptation to a rotational transformation
J Neurophysiol
Circuits and circuit disorders of the basal ganglia
Arch Neurol
Neural correlates of reach errors
J Neurosci
Mechanisms influencing acquisition and recall of motor memories
J Neurophysiol
Motor sequence learning and movement disorders
Curr Opin Neurol
Functional anatomy of visuomotor skill learning in human subjects examined with positron emission tomography
Eur J Neurosci
Experience-dependent changes in cerebellar contributions to motor sequence learning
Proc Natl Acad Sci USA
Functional anatomy of motor skill learning
The neurobiology of consolidations, or, how stable is the engram?
Annu Rev Psychol
To consolidate or not to consolidate: what are the questions?
Sleep forms memory for finger skills
Proc Natl Acad Sci USA
Motor memory consolidation in sleep shapes more effective neuronal representations
J Neurosci
Changing brain networks for visuomotor control with increased movement automaticity
J Neurophysiol
Distinguishable brain activation networks for short- and long-term motor skill learning
J Neurophysiol
Motor practice and neurophysiological adaptation in the cerebellum: a positron tomography study
Proc R Soc Lond B Biol Sci
Intact mirror-tracing and impaired rotary-pursuit skill learning in patients with Huntington’s disease: evidence for dissociable memory systems in skill learning
Neuropsychology
Distinct striatal regions support movement selection, preparation and execution
Neuroreport
Functional mapping of sequence learning in normal humans
J Cogn Neurosci
Cited by (548)
How do alterations of the basal ganglia affect procedural memory in Tourette syndrome?
2024, Current Opinion in Behavioral SciencesDifferential effects of acute cardiovascular exercise on explicit and implicit motor memory: The moderating effects of fitness level
2023, Neurobiology of Learning and Memory