Edited by:
Reviewed by:
*Correspondence:
This article was submitted to the journal Frontiers in Human Neuroscience.
This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) or licensor are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.
Performing accurate movements requires preparation, execution, and monitoring mechanisms. The first two are coded by the motor system, the latter by the sensory system. To provide an adaptive neural basis to overt behaviors, motor and sensory information has to be properly integrated in a reciprocal feedback loop. Abnormalities in this sensory-motor loop are involved in movement disorders such as focal dystonia, a hyperkinetic alteration affecting only a specific body part and characterized by sensory and motor deficits in the absence of basic motor impairments. Despite the fundamental impact of sensory-motor integration mechanisms on daily life, the general principles of healthy and pathological anatomic–functional organization of sensory-motor integration remain to be clarified. Based on the available data from experimental psychology, neurophysiology, and neuroimaging, we propose a bio-computational model of sensory-motor integration: the Sensory-Motor Integrative Loop for Enacting (SMILE). Aiming at direct therapeutic implementations and with the final target of implementing novel intervention protocols for motor rehabilitation, our main goal is to provide the information necessary for further validating the SMILE model. By translating neuroscientific hypotheses into empirical investigations and clinically relevant questions, the prediction based on the SMILE model can be further extended to other pathological conditions characterized by impaired sensory-motor integration.
The ability to correctly perform movements in everyday life is critical to adequately interact with the environment. Several movement disorders manifest as remarkable deficiencies in the ability to control voluntary actions and interact successfully with the surroundings. The onset of the disease may, in fact, disrupt normal functioning of fundamental neural and cognitive processes that permit efficient sampling of sensory events from the environment in order to implement adequate motor routines. Sensory-motor integration refers to this ability to coherently organize bodily sensations and motor responses. Despite its crucial role as a proficient interface with the environment, current models of sensory-motor integration lack a definitive overview on the reciprocal interplay between sensory input and motor output. The opportunity to distinguish and properly understand the relative weight of sensory and motor processes is provided by mental imagery, a cognitive task that activates sensory-motor representations without physical sensory stimulation. This experimental approach is particularly important for studying the physiology of pathological conditions characterized by sensory-motor integration deficits but without impairments in basic motor functions, such as focal dystonia.
Dystonia is a disabling movement disorder characterized by involuntary muscle contractions frequently associated with abnormal movements and/or postures (
Despite the fact that different types of dystonia have long been considered as the result of impaired motor control, the focus has progressively been switched to the integration between sensory input and motor output. Basic research in neuroscience has produced converging evidence on the tight relationship between movement generation and recalibration (see also
Behavioral data can be used to create general models, based on which it is possible to generate procedural hypotheses on the modules possibly involved in specific processes and their putative reciprocal connections in theoretical frameworks. Testing these hypotheses can contribute to individuate the origin of specific impairments and to assess early markers of a given disorder (
The SDT is a tactile task used to establish the minimal distance between two stimuli that participants can reliably discern as distinct events. Healthy subjects can detect changes in the orientation of tiny parallel grooves as thin as 1 mm when presented on the tip of the finger (
In summary, the SDT has been largely used in populations suffering from a variety of dystonic disorders showing that in these conditions there are clear impairments not only in the motor components but also in sensory processing of spatial information coming from the affected body region. Beyond this first conclusion, on the basis of the available data it could be suggested that the disorganization due to FHD is not limited to a single body part but rather extends at least to the contralateral hand. However, it is still difficult to conclude whether the spatial discrimination impairments are restricted to only the symptomatic limb or are instead bilateral.
Another frequently used task in dystonia-related research is the TDT, which identifies the minimal time interval between two stimuli that allows differentiating them as separate events. It typically involves unimodal electrical stimulation of the skin, but can be coupled, paralleled, or even replaced by visual, kinematic, or any other type of stimuli. On average, healthy subjects can discriminate two electrical stimuli on the index finger provided they are separated by at least 30 ms (
In summary, dystonic patients show evidence of abnormalities not only in spatial discrimination, but also in temporal processing. Temporal discrimination seems to be affected both at symptomatic and non-symptomatic body regions. The reviewed data suggest a critical difference in the mechanisms of FHD and generalized dystonia. FHD patients’ impairment appears to be linked to tactile processing and visual–tactile integration, whereas the generalized dystonia patients exhibit more general impairments in integration processing, including exclusively visual processing of stimuli near the hands. Overall, most studies present a coherent picture of the relationship between dystonia and TDTs, whose increase in focal dystonia is specifically selective for sensory processing but not isolated to the symptomatic limb.
It has been demonstrated that the movement induced by tonic vibration of a tendon is impaired in generalized dystonia (
Few studies investigated the structural brain organization of FHD and the available data are largely inconsistent. Some studies associated FHD with anatomical abnormalities at the cortical level (
Dystonia-related morphometric changes.
Region | Increased volume | Decreased volume |
---|---|---|
Prefrontal cortex | ||
Inferior parietal lobe | ||
Cerebellum | ||
Thalamus | ||
Putamen |
In general, one of the main reasons of such inconsistencies might be the use of different scanners, data recording sequences, and analysis procedures. One possibility for overcoming this heterogeneity may be the use of quantitative methods that are more conducive to comparisons across laboratories/scanners, such as Voxel-Based Quantification (
Using passive vibrotactile stimulation of single fingers, functional neuroimaging studies have described the disorganization of the somatosensory representations in dystonia.
Despite an initial general agreement, the level of inconsistency in terms of affected key regions between different studies increases as slightly different tasks are taken into account. Indeed if FHD patients are asked to
For several decades, the excitatory/inhibitory regulations of the central nervous system have been proposed as impaired in both generalized dystonia and FHD (
In addition to CSP, other types of inhibition features are potent markers of neural pathway mechanisms, and have been shown to present abnormalities in all types of dystonia at the level of both the central and the peripheral nervous system (
Animal studies showed that aberrant intracortical surround inhibition can lead to dystonic behaviors (
In addition to intracortical surround inhibition, also interhemispheric inhibition (the ability of a unilateral hemispheric stimulation of the motor cortex to inhibit the contralateral motor cortex given a short latency) is impaired in FHD (
Some groups focused on psychogenic dystonia, a type of dystonic disorder without a clear neurological basis and possibly associated with other psychological disorders. In this vein,
Altogether these data suggest that different types of dystonia, whether primary or secondary to psychological disorders, share basic mechanisms as well as widespread cortical and subcortical abnormalities. However, the question whether these excitatory/inhibitory regulation abnormalities common to several types of dystonias are a cause or a consequence of the disorders still needs to be resolved. Nonetheless, some studies have linked dystonic symptoms to abnormal activity of numerous modules in the basal ganglia–thalamocortical circuit (
Neural plasticity includes, but is not limited to, the ability of the brain to re-organize its neural connections as a function of both internal and external factors. Animal studies showed that over-trained repetitive movements abnormally remodels somatosensory cortical maps, leading to sensory de-differentiation between the receptive fields of neighboring digits (
Indeed, dystonia-related changes of receptive field features have been reported in sub-cortical structures such as the globus pallidus and thalamus (
The so-called “paired-associative stimulation” approach can be used to induce and identify the characteristics of neural plasticity. During paired-associative stimulation a given sensory stimulus is paired with TMS of a specific brain region, creating an artificial and relatively long-term association between an external event and the TMS pulse (
According to the defective sensory-motor learning hypothesis, the different types of dystonia would be characterized by functional alterations in the sensory-motor circuit supposed to integrate sensory input and motor output (
The aberrant sensory input would be due to the disorganization of S1 (
However, the nature of the relationship between disorganized somatosensory information and aberrant motor output is still under debate. One possible explanation is that long-lasting non-physiologic motor behavior can cause changes in somatosensory representations. Alternatively, abnormal somatosensory representations may lead to abnormal motor output explaining the particular dystonic phenotype. Consequently, one of the main focuses for future research will be to investigate movement mechanisms in FHD and other types of movement disorders, but ruling out any confounding effect due to abnormal sensory feedback.
To identify the origin of dystonic behaviors it is crucial to understand the features of sensory-motor integration mechanisms while avoiding any potential confound due to altered sensory feedback. One possibility to achieve this goal is to use an investigation tool that does not require movement execution. This would help differentiate the mechanisms related to altered sensory feedback from those related to abnormal sensory-motor representations. Mental imagery is a cognitive task with such characteristics. In healthy subjects physical execution and mental imagery of a movement – “motor imagery” – share similar temporal and kinematic properties (
In healthy controls, physical movement and motor imagery engage partially overlapping brain networks (
Combined with the manipulation of cortico-spinal excitability by means of TMS, motor imagery can be used to investigate not only the properties of cortical representations but also the characteristics of the communication between the central nervous system and the periphery. In particular, in healthy subjects top-down imagery-related mechanisms regulate the excitability of the sensory-motor pathways (
Despite the differences with respect to the typically hyperkinetic dystonic disorders, a final example of the validity of implementing motor imagery protocols to evaluate motor excitability in focal dystonia, can be individuated in a study performed by
Only few brain imaging studies investigated the neural circuits recruited by motor imagery and their task-dependent activity in FHD. Despite the clear difference between primary and secondary dystonias, and as an additional example of the implementation of motor imagery as an investigation tool in dystonia-related disorders,
At present the only available data on the neural correlates of motor imagery in primary FHD have been recently reported in two paired studies. In these studies FHD patients were asked to perform motor imagery of grasping a pencil with the purpose of either writing or sharpening it (
A straightforward way to control for and quantitatively measure motor imagery is provided by mental rotation, according to which people mentally rotate visually presented stimuli while response times and accuracy are measured. In healthy subjects the response times required to mentally align a stimulus to the vertical are a function of stimulus orientation (
Based on this evidence, at least two motivations support the implementation of mental rotation in effective experimental protocols. First, as a motor imagery task, mental rotation supplies access to sensory-motor representations without any confounding effect potentially due to sensory feedback during movement execution. Second, providing quantitative measurements of the subjects’ performance, it adds important information on an otherwise purely introspective process. Thanks to these characteristics it has been used in several clinical populations. In particular, when asked to mentally rotate hands, patients suffering from cerebral palsy show the typical modulation of response times as a function of the stimulus orientation, but doubling the latencies with respect to healthy controls (
In addition, patients who lost their dominant limb due to amputation, show longer latencies and lower accuracy in the mental rotation of images depicting the amputated hand (
In cervical dystonia – affecting the neck muscles and then altering the vestibular input (
A way to test this possibility takes into account the mechanisms of postural and proprioceptive online recalibration. In healthy subjects, congruent visuo-tactile stimulation promotes self-attribution of a fake hand as explicitly measured by self-reports (“rubber hand illusion”), but does not necessarily affect proprioceptive hand recalibration as implicitly measured by the “proprioceptive drift” procedure (
A major challenge in clinical neuroscience is building a model that can explain the causal link between dysfunctional brain networks and particular clinical phenotypes. The available computational models of sensory-motor integration (
Based on the available data, we propose that the SMILE model can represent a way to biologically situate and experimentally test previous computational models of sensory-motor integration in clinical phenomena such as dystonia. At the biological level, according to the SMILE model the movement preparation would be encoded by the premotor and supplementary motor regions (
Taking into account the hypothesized mechanisms of FHD and the possible structure described in the SMILE model, we propose that FHD is the manifestation of a breakdown in the sensory-motor loop as the result of a disorganization targeting S1 and due to over-training-related abnormal neuroplasticity, impaired cortico-subcortical dynamics, and local loss of inhibition. Based on evidence that FHD patients exhibit impairments in temporal and spatial discrimination, but not in overt motor behaviors other than the task-specific ones, a first hypothesis is that the breakdown of the sensory-motor integration happens in the high-level nodes, specifically in S1. The breakdown would determine no equivalence between the signal sent from the periphery to S1, and the one sent from S1 to M1 (calibration). When M1 sends the signal to the periphery through the brainstem, the peripheral muscle activation (through the cerebellum, basal ganglia, and thalamus) sends a feedback signal to premotor, supplementary motor and primary sensory regions, which in turn have back projections to equivalent areas in M1. We hypothesize that if the gain of the signals sent through this loop is >1, then M1 keeps increasing its firing until maximal muscle contraction occurs, that is the typical cramp of FHD (
Taking into account the possibility that the deterioration of the sensory information could happen in the low-level nodes and that the whole thalamus-basal ganglia circuit preserves somatotopic organization all along (
The SMILE model explains (1) task-specific impairments in terms of a breakdown in only some sub-components of the sensory-motor loop, (2) increasing muscle contraction resulting in cramps as a function of the unbalance between sensory input and motor output, and (3) spreading activity to agonist muscles (due to overlapping cortical representations) as a function of extremely repetitive behaviors that would cause cortical disorganization. Taking into consideration this tight association between sensory input and motor output, it is clear how crucial their dissociation is for better understanding the nature of their integration, and therefore the implementation of mental rotation as investigation tool in future experimental protocols.
Neuroimaging studies based on previous models showed the involvement of both cortical and subcortical regions, suggesting that dystonic deficits affect a broadly distributed network but leaving unsolved the issue of which different nodes of this network are specifically impaired. The inconsistencies in the available results could be due to methodological differences in experimental protocols, required tasks, scanning procedures, or the underestimation of the distorted sensory feedback as a crucial confounding factor that renders the investigation of sensory-motor processes particularly difficult. Conversely, mental rotation of body parts engages anatomically interconnected brain systems implicated in the integration of sensory-motor information and has been implemented with brain imaging for studying the properties of the sensory-motor system in movement disorders such as Parkinson’s disease. However, both neuroimaging and physiological data necessary to identify the pathophysiological characteristics of FHD are still lacking, and mental rotation is a good tool to acquire this information. This important information on brain activity and cortico-spinal communication relative to mental rotation of body parts in FHD represents an unresolved gap that could and should be filled. Finding the influence of FHD in modulating the activity of specific neural circuits, such as hyper-synchronous activity, might help not only to better understand the pathophysiology of FHD but also to develop
Rather than a conclusive definition of the pathophysiological mechanisms of the different subtypes of dystonia, the advance brought by the SMILE model is an understanding of the general mechanisms of sensory-motor integration together with the promotion of mental imagery as an experimental approach able to overcome the previous methodological limitations. Such a theoretical–experimental joint approach is essential to obtain the new data required to precisely define the pathophysiology of the different subtypes of dystonia. The lack of this combination is probably one of the reasons why, despite the comprehension of the importance of the sensory components, previous models of movement disorders have pooled together different dystonic symptomatologies (e.g.,
The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.
This work has been supported by the Swiss National Science Foundation (grant PZ00P1_148186 to Silvio Ionta and grant 320030-149982 to Micah M. Murray). Jeremie Lefebvre is supported by the Natural Sciences and Engineering Research Council of Canada.
cortical silent period
focal hand dystonia
primary motor cortex
primary sensory cortex
spatial discrimination threshold
Sensory-Motor Integrative Loop for Enacting
temporal discrimination threshold.