Accepted Manuscript Clinical studies Motor sequence learning and motor adaptation in primary cervical dystonia Petra Katschnig-Winter, Petra Schwingenschuh, Marco Davare, Anna Sadnicka, Reinhold Schmidt, John C. Rothwell, Kailash P. Bhatia, Mark J. Edwards PII: DOI: Reference:
S0967-5868(13)00553-5 http://dx.doi.org/10.1016/j.jocn.2013.08.019 YJOCN 5384
To appear in:
Journal of Clinical Neuroscience
Received Date: Accepted Date:
14 May 2013 24 August 2013
Please cite this article as: P. Katschnig-Winter, P. Schwingenschuh, M. Davare, A. Sadnicka, R. Schmidt, J.C. Rothwell, K.P. Bhatia, M.J. Edwards, Motor sequence learning and motor adaptation in primary cervical dystonia, Journal of Clinical Neuroscience (2013), doi: http://dx.doi.org/10.1016/j.jocn.2013.08.019
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
D-13-00761 Clinical Studies
Motor sequence learning and motor adaptation in primary cervical dystonia
Petra Katschnig-Wintera,b,*, Petra Schwingenschuha,b, Marco Davarea, Anna Sadnickaa, Reinhold Schmidtb, John C. Rothwella, Kailash P. Bhatiaa, Mark J. Edwardsa a
Sobell Department of Motor Neuroscience and Movement Disorders, UCL Institute of
Neurology, Queen Square, London, UK b
Department of Neurology, Medical University of Graz, Auenbruggerplatz 22, 8036 Graz,
Austria
*Corresponding author. Tel.: +43 316 3851 6437; fax: +43 316 3851 4178. E-mail address: [email protected] (P. Katschnig-Winter).
Conflicts of Interest/Disclosures The authors declare that they have no financial or other conflicts of interest in relation to this research and its publication.
Abstract Motor sequence learning and motor adaptation rely on overlapping circuits predominantly involving the basal ganglia and cerebellum. Given the importance of these brain regions to the pathophysiology of primary dystonia, and the previous finding of abnormal motor sequence learning in DYT1 gene carriers, we explored motor sequence learning and motor adaptation in patients with primary cervical dystonia. We recruited 12 patients with cervical dystonia and 11 healthy controls matched for age. Subjects used a joystick to move a cursor from a central starting point to radial targets as fast and accurately as possible. Using this device, we recorded baseline motor performance, motor sequence learning and a visuomotor adaptation task. Patients with cervical dystonia had a significantly higher peak velocity than controls. Baseline performance with random target presentation was otherwise normal. Patients and controls had similar levels of motor sequence learning and motor adaptation. Our patients had significantly higher peak velocity compared to controls, with similar movement times, implying a different performance strategy. The preservation of motor sequence learning in cervical dystonia patients contrasted with the previously observed deficit seen in patients with DYT1 gene mutations, supporting the hypothesis of differing pathophysiology in different forms of primary dystonia. Normal motor adaptation is an interesting finding. With our paradigm we did not find evidence that the previously documented cerebellar abnormalities in cervical dystonia have a behavioral correlate, and thus could be compensatory or reflect ―contamination‖ rather than being directly pathological.
Keywords: Basal ganglia; Cerebellum; Dystonia; Motor adaptation; Motor learning; Sequence learning
1.
Introduction
Historically considered a disorder of the basal ganglia, there is now evidence for a wider network of neuroanatomical structures involved in the pathophysiology of dystonia. Recent research has particularly focused on the cerebellum against the background of clinical reports of patients with cerebellar lesions presenting with dystonia and animal models of dystonia in which the cerebellum appears to play a critical pathophysiological role.1 Radiological studies have demonstrated an increase in metabolic activity in the basal ganglia, supplementary motor areas and the cerebellum in a variety of forms of primary dystonia. 2, 3, 4 Additionally, diffusion tensor imaging data have demonstrated reduced integrity of cerebellothalamic tracts in DYT1 and DYT6, which correlates with the clinical penetrance of the mutation.5 Abnormal patterns of cerebellar activation are also seen using functional MRI blood oxygen level dependent techniques in motor tasks such as tapping and eye blinking in patients with dystonia of the corresponding regions. 6, 7 Preliminary neurophysiological data lends support to the notion that the cerebellum is affected in patients with primary focal dystonia of the neck or hand, with disturbed eye blink conditioning occurring in these patients.8 The absence of overt cerebellar signs on clinical examination in dystonia, however, suggests that the experimentally observed cerebellar dysfunction may either be too mild to be expressed clinically, might simply reflect unimportant ―contamination‖ of a structure directly connected to the basal ganglia, or might represent a compensatory response to the primary pathophysiology within the basal ganglia.
Behavioural paradigms can be used to study function of neuroanatomical structures likely to be involved in dystonia. Two of these are motor sequence learning (MSL; the incremental acquisition of sequential movement patterns) and motor adaptation (MA; paradigms that test
capacity to compensate for environmental changes). Serial reaction time tasks can be used to study MSL; following training blocks with randomly presented targets, sequence learning is demonstrated by faster reaction times to sequence presentation compared with random trials.9 MA paradigms require participants to implicitly adapt to changes in the environment caused by experimental manipulations such as altered visual feedback produced by prisms, perturbations to visual feedback on a computer monitor or force fields applied to movements of a robotic limb. The anatomical substrate of implicit sequence learning is extensive but is thought to be critically dependent on the basal ganglia while MA requires an intact cerebellar circuitry.10, 11, 12 and 13
Impaired MSL has been described in both manifesting and non-manifesting patients with the DYT1 mutation, but is normal in patients with DYT6 mutations.3, 14 MA has not previously been assessed in any patients with dystonia to our knowledge. Here we hypothesized that MSL would be impaired in cervical dystonia patients due to the presumed basal ganglia dysfunction that underlies the pathophysiology of this condition, but that MA would be normal, reflecting a compensatory rather than a primary pathological role for the cerebellum in this form of primary dystonia.
2.
Methods 2.1.
Participants
We recruited 12 patients (nine women, three men; mean age 58.8 ± standard deviation 9.6 years, range 40–77 years) with idiopathic cervical dystonia from the Movement Disorder outpatient clinic at the National Hospital for Neurology and Neurosurgery, Queen Square London, UK, and 11 healthy control subjects (seven women, four men) who were matched for age (mean age 55.4
± standard deviation 9.6 years, range 43–70 years) and years in education. Disease duration ranged from 6 to 36 years. All participants were right-handed and without cognitive impairment or psychiatric disease. Patients did not suffer from head tremor and there was no segmental spread of dystonia to the hands. Mobility at the elbow and shoulder joints was unrestricted and painless in all subjects. Dystonia severity was evaluated with the Toronto Western Spasmodic Torticollis Rating Scale 15 ranging from 5 to 49 out of 85 points. All patients receiving botulinum toxin treatment had their last injections at least 3 months before the study. Informed consent was obtained and the study was approved by the local ethics committee.
2.2.
General characteristics of the motor task
Subjects were seated in front of a computer screen with a joystick secured to the table directly in front of them. Subjects were instructed to move the cursor representing the joystick position from a central starting point on the computer screen to one of eight radial targets, which were evenly spaced 45° apart and displayed as red squares (1 cm2). Subjects were asked to move the cursor into the target as fast and accurately as possible upon target appearance. They were required to retain the cursor in the target until it switched to green (1 second following entrance to the square). At this point subjects were instructed to release their grip on the joystick handle, allowing the spring-loaded device to re-center for the next trial. There were three task conditions. In the motor reference task (R), targets were presented in a pseudo-randomized order in blocks of 40 trials. We used four random blocks (R1–4) at the beginning of the experiment to familiarize subjects with the task. Further random blocks were inserted after the sequence learning (R5, 6) and adaptation tasks (R7, 8). In the MSL task, a sequence of six targets was repeated seven times per block (42 movement trials). The presence of a sequence was not mentioned to the subjects. Four identical sequence blocks (S1–4) were presented. In the motor adaptation block (MAB) the
order of target presentation was pseudo-randomized and consisted of 40 movement trials. The direction of the cursor movement on the screen was rotated clockwise by 30° relative to hand movement. Patients and controls performed 13 blocks in the following order: R1-R2-R3-R4-S1-S2-S3-S4R5-R6-MAB-R7-R8. The total time of the experiment was 1 hour with opportunities for patients to rest between random blocks. No patient reported problems with fatigue or concentration.
2.3.
Data acquisition and analysis
Targets were presented and data acquired using MATLAB (version 7; The MathWorks, Inc., Natick, MA, USA) with Cogent Toolbox, interfaced with a CED 1401 analogue digital converter device (Cambridge Electronic Design, Cambridge, UK). Data were stored in a computer for offline analysis. The following parameters were measured for each movement trial in the MSL task: onset time (OT), the time from target presentation to movement onset; movement time (MT), the time from movement onset to target hit; response time (RT), the time from target presentation to target hit (sum of OT and MT); peak velocity (pV); directional error (DE), the angle between the ideal path (straight line) and trajectory taken at pV, which was also the parameter used for MA. A directional error of 0.90 for all comparisons).
3.2.
MSL
To investigate MSL we compared the performance in block S4 with R5. Both groups successfully learned the sequence, indicated by a significant effect of BLOCK for OT (F(1,21) = 20.9, p < 0.001), MT (F(1,21) = 6.8, p = 0.016), RT (F(1,21) = 13.6, p = 0.001) and DE (F(1,21) = 4.4, p = 0.048). Figure 1 shows the mean ± standard error of each parameter for both groups plotted against block. As in the motor reference task, we found a significant GROUP effect for pV and additionally for DE. Post hoc analyses revealed this to be due to a significantly higher pV in patients (563.6 ± 75.9 mm/s) compared to controls (491.5 ± 71.9 mm/s, p = 0.030), but controls had overall lower DE compared with patients (7.1 ± 1.3º versus 9.1 ± 1.5º; p = 0.003). We did
not find any GROUP × BLOCK interaction. Thus, MSL occurred in both groups to a similar extent as evidenced by simple movement and temporal parameters (OT, MT, RT), but performance strategy was different between groups with higher pV and lower aiming accuracy in patients with an opposite pattern in controls.
3.3.
MA
To assess MA we compared the mean of DE in block R6 with the mean in every 10 movement trials in the MAB (―first 10‖, ―second 10‖, ―third 10‖, ―last 10‖). To evaluate after-effects we computed the mean of DE for every 10 movements in block R7 and compared it to the mean of DE in block R6. After-effects were considered to be present as long as there was a significant difference compared to the baseline value (Fig. 2). In line with the performance in random blocks and MSL we found a significant effect for GROUP (F(1,21) = 5.1, p = 0.035) due to higher pV in patients (477.8 ± 132.3 mm/s) compared to controls (372.3 ± 82.6 mm/s). MA was present in patients and controls to a similar extent evidenced by a significant effect of BLOCK for DE (F(4,84) = 22.8, p < 0.001) in the absence of a GROUP × BLOCK interaction. DE significantly decreased from first 10 to third and last 10 movements (p < 0.005 for all comparisons). The presence of MA was confirmed in both groups by the presence of significant after-effects in the random block performed immediately after the MAB. We did not find a GROUP effect nor an GROUP × BLOCK interaction but there was a significant effect of BLOCK for DE (F(4,84) = 12.102, p < 0.001). Post hoc analyses revealed this to be due to a significant lower aiming accuracy in the first and second 10 trials of R7 compared to R6 (p < 0.05).
4.
Discussion
In this study we found patients with cervical dystonia showed normal general motor performance on a reaction time task, except for higher pV. Sequence learning and MA were both normal, although there was evidence that patients may have employed a different strategy to controls, utilizing higher peak velocity but lower aiming accuracy. We discuss the significance of this result in the context of current pathophysiological models of dystonia.
4.1.
General motor performance
We found general motor performance to be indistinguishable between patients and controls apart from a significantly higher pV in patients. Subtle changes in velocity profiles have previously been noted in dystonia; analysis of limb trajectories in clinically affected limbs with simple upper limb movements revealed that the velocity profiles of upper limb movements were less symmetrical due to longer deceleration times.16 The otherwise normal motor performance is in keeping with some previous studies,3 but not with others which reported, for example, reduced accuracy and longer movement times with reaching movements in patients with cervical dystonia.17
4.2.
MSL
During this experiment, subjects were not informed of the presence of a repeating sequence within the task. We did not find any deficit in MSL in cervical dystonia patients as evidenced by simple movement and temporal parameters, but there was evidence for a different performance strategy between patients and controls.
This is an interesting finding as studies investigating implicit MSL suggest that the basal ganglia are integral to the performance of this task. In healthy participants, imaging data acquired during MSL tasks suggests that the basal ganglia, cerebellum, premotor and prefrontal cortex are involved, 10, 11, 18 although imaging data may not necessarily reflect the learning process but rather the basic motor implementation of the improved performance. Patients with disorders involving the basal ganglia such as Parkinson’s disease (PD) or Huntington’s disease (HD) have been shown to be impaired in acquiring new motor sequences.19,
20, 21
However, by nature of the
fact that these are disorders of motor function some argue that it is the expression of the learnt response (for example due to slow movement initiation in PD) that is disturbed rather than the actual sequence learning itself.22 In line with the proposed role for the basal ganglia in the pathophysiology of dystonia, abnormal explicit MSL with an arm-reaching version of the serial reaction time tasks has been reported in non-manifesting and manifesting DYT1 carriers,3 but of interest this does not appear to occur in non-manifesting and manifesting patients with DYT6.14 Our data, which were collected with a modified version of this task and analyzed in a different way, suggest that in cervical dystonia neither the extent nor the distribution of basal ganglia dysfunction may be sufficient to impair sequence learning. In keeping with normal MSL in DYT6 dystonia, our data also indicate that there may be important pathophysiological differences between different forms of primary dystonia.
4.3.
MA
MA is a heavily cerebellar dependent task, and to our knowledge has not previously been studied in dystonia. It is hypothesized that the cerebellum generates an internal model for movement
which predicts the sensory consequences of motor commands. This internal model then generates an error signal to a given perturbation that drives adaptation.23 Stimulation of the cerebellum using anodal transcranial direct current stimulation improves visuomotor adaptation24 and patients with lesions to the cerebellum have deficits in paradigms that require adaptation.23 It is likely that the cerebellum makes its contribution via its output to the thalamus which in turn projects to the cerebral cortex, as thalamic manipulation in patients with deep brain stimulation for essential tremor can modulate performance in this task.25 In contrast, patients with basal ganglia disorders such as HD and PD are not impaired in adaptation tasks.26,
27, 28, 29
We did not
find any deficit in MA in patients with cervical dystonia. This is in contrast to a growing literature across a range of experimental techniques that the cerebellum may have a role in the pathophysiology of dystonia. Although our task may lack sensitivity to small impairments, the data suggests that in cervical dystonia, the adaptive capacities of the cerebellum to visuomotor changes are normal. One important qualifier to this statement is that our task did require subjects to stabilize in the target, and the cursor path was visible throughout the movement. Previous studies in healthy participants and patients with cerebellar degeneration have not found additional benefit to visuomotor adaptation from the additional (visual and proprioceptive) information provided by stabilization within the target zone,30 but it is possible that this is different for patients with dystonia.
5.
Conclusion
In the current study, patients with cervical dystonia showed normal MA. Hence, our findings do not support the hypothesis that previously documented abnormalities in cerebellum-dependent
tests (for example, eye blink conditioning) reflect a primary pathological role for cerebellar dysfunction in cervical dystonia. One may speculate that abnormalities on such tests reflect compensatory cerebellar changes or reflect ―contamination‖ via the direct connection of the cerebellum with other brain structures including the basal ganglia that are abnormal. However, we cannot exclude the possibility that our paradigm was not sensitive enough to pick up mild deficits in MA. We based our sample size on previous papers on motor learning in dystonia but refrained from a power calculation. Therefore, we cannot exclude the possibility that our study was underpowered which we recognize as a limitation. It would be of considerable interest to determine whether non-visuomotor adaptation (for example split-belt treadmill adaptation) was impaired in dystonia, as it is possible that impairment could differ depending on the nature of the sensory input that drives adaptation. In combination with electrophysiological, imaging and behavioral data from specific forms of primary dystonia, these data support the hypothesis that important pathophysiological differences may exist within the broad group of conditions currently labeled as primary dystonia. Further study of well-defined groups of patients with forms of primary dystonia (and indeed dystonia-plus, secondary and degenerative dystonia) will help determine specific subtypes of dystonia that may have different neuroanatomical substrates.
References
1. Sadnicka A, Hoffland BS, Bhatia KP et al. The cerebellum in dystonia - help or hindrance? Clin Neurophysiol 2011;123:65-70. 2. Carbon M, Ghilardi MF, Argyelan M et al. Increased cerebellar activation during sequence learning in DYT1 carriers: an equiperformance study. Brain. 2008;131:146-54. 3. Ghilardi MF, Carbon M, Silvestri G et al. Impaired sequence learning in carriers of the DYT1 dystonia mutation. Ann Neurol. 2003;54:102-09. 4. Trost M, Carbon M, Edwards C et al. Primary dystonia: is abnormal functional brain architecture linked to genotype? Ann Neurol. 2002;52:853-56. 5. Argyelan M, Carbon M, Niethammer M et al. Cerebellothalamocortical connectivity regulates penetrance in dystonia. J Neurosci. 2009;29:9740-47. 6. Baker RS, Andersen AH, Morecraft RJ et al. A functional magnetic resonance imaging study in patients with benign essential blepharospasm. J Neuroophthalmol. 2003;23:11-15. 7. Wu CC, Fairhall SL, McNair NA et al. Impaired sensorimotor integration in focal hand dystonia patients in the absence of symptoms. J Neurol Neurosurg Psychiatry. 2010;81:659-65. [8] Teo JT, van de Warrenburg BP, Schneider SA et al. Neurophysiological evidence for cerebellar dysfunction in primary focal dystonia. J Neurol Neurosurg Psychiatry. 2009;80:80-83. [9] Nissen MJ, Bullemer P. Attentional requirements of learning—evidence from performance measures. Cogn Psychol 1987;19:1-32. [10] Rauch SL, Whalen PJ, Savage CR et al. Striatal recruitment during an implicit sequence learning task as measured by functional magnetic resonance imaging. Hum Brain Mapp. 1997;5:124-32. 11. Schendan HE, Searl MM, Melrose RJ et al. An FMRI study of the role of the medial temporal lobe in implicit and explicit sequence learning. Neuron. 2003;37:1013-1025. 12. Diedrichsen J, Hashambhoy Y, Rane T et al. Neural correlates of reach errors. J Neurosci. 2005;25:9919-31. 13. Ghilardi M, Ghez C, Dhawan V et al. Patterns of regional brain activation associated with different forms of motor learning. Brain Res. 2000;871:127-45. 14. Carbon M, Argyelan M, Ghilardi MF et al. Impaired sequence learning in dystonia mutation carriers: a genotypic effect. Brain. 2011;134:1416-27.
15. Consky ES, Lang AE. Clinical assessments of patients with cervical dystonia. In: Jankovic J, Hallett M, editors. Therapy with Botulinum Toxin. New York, NY: Marcel Dekker Inc 1994; 211-37. 16. Inzelberg R, Flash T, Schechtman E et al. Kinematic properties of upper limb trajectories in idiopathic torsion dystonia. J Neurol Neurosurg Psychiatry. 1995;58:312-19. 17. Pelosin E, Bove M, Marinelli L et al. Cervical dystonia affects aimed movements of nondystonic segments. Mov Disord. 2009;24:1955-61. 18. Lehericy S, Benali H, Van de Moortele PF et al. Distinct basal ganglia territories are engaged in early and advanced motor sequence learning. Proc Natl Acad Sci U S A. 2005;102:12566-71. 19. Ghilardi MF, Silvestri G, Feigin A et al. Implicit and explicit aspects of sequence learning in pre-symptomatic Huntington's disease. Parkinsonism Relat Disord. 2008;14:457-64. 20. Schneider SA, Wilkinson L, Bhatia KP et al. Abnormal explicit but normal implicit sequence learning in premanifest and early Huntington's disease. Mov Disord. 2010;25:1343-49. 21. Siegert RJ, Taylor KD, Weatherall M et al. Is implicit sequence learning impaired in Parkinson's disease? A meta-analysis. Neuropsychology. 2006;20:490-95. 22. Seidler RD, Tuite P, Ashe J. Selective impairments in implicit learning in Parkinson's disease. Brain Res. 2007;1137:104-10. 23. Shadmehr R, Krakauer JW. A computational neuroanatomy for motor control. Exp Brain Res. 2008;185:359-81. 24. Galea JM, Vazquez A, Pasricha N et al. Dissociating the roles of the cerebellum and motor cortex during adaptive learning: the motor cortex retains what the cerebellum learns. Cereb Cortex 2011;21:1761-70. 25. Chen H, Hua SE, Smith MA et al. Effects of human cerebellar thalamus disruption on adaptive control of reaching. Cereb Cortex. 2006;16:1462-73. 26 Agostino R, Sanes JN, Hallett M. Motor skill learning in Parkinson's disease. J Neurol Sci. 1996;139:218-26. 27. Smith MA, Shadmehr R. Intact ability to learn internal models of arm dynamics in Huntington's disease but not cerebellar degeneration. J Neurophysiol. 2005;93:2809-21. 28. Cavaco S, Anderson SW, Correia M et al. Task-specific contribution of the human striatum to perceptual-motor skill learning. J Clin Exp Neuropsychol. 2011;33:51-62.
29. Gabrieli JD, Stebbins GT, Singh J et al. Intact mirror-tracing and impaired rotary-pursuit skill learning in patients with Huntington's disease: evidence for dissociable memory systems in skill learning. Neuropsychology. 1997;11:272-81. 30. Tseng YW, Diedrichsen J, Krakauer JW et al. Sensory prediction errors drive cerebellumdependent adaptation of reaching. J Neurophysiol. 2007;98:54-62.
Fig. 1. Graphs showing the effect of repetitive presentation of a target sequence on performance (motor sequence learning). Data are shown for block S4 (last block of sequence presentation) and R5 (subsequent block of random target presentation). Mean ± standard error are shown for (a) onset time, movement time and response time, (b) peak velocity, and (c) directional error. DE = directional error, MT = movement time, OT = onset time, pV = peak velocity, RT = response time.
Fig. 2. Mean ± standard error of directional error for both groups as a function of trial blocks. (a) Motor adaptation from R6 (preceding block of random target presentation), MAB f10 (first 10 trials of motor adaptation), MABs10 (second 10 trials of motor adaptation), MAB t10 (third 10 trials of motor adaptation), and MAB l10 (last 10 trials of motor adaptation). (b) After-effects from R6, R7 f10 (first 10 trials of random block following motor adaptation), R7 s10 (second 10 trials of random block following motor adaptation), R7 t10 (third 10 trials of random block following motor adaptation) and R7 l10 (last 10 trials of random block following motor adaptation).
S0967-5868(13)00553-5 http://dx.doi.org/10.1016/j.jocn.2013.08.019 YJOCN 5384
To appear in:
Journal of Clinical Neuroscience
Received Date: Accepted Date:
14 May 2013 24 August 2013
Please cite this article as: P. Katschnig-Winter, P. Schwingenschuh, M. Davare, A. Sadnicka, R. Schmidt, J.C. Rothwell, K.P. Bhatia, M.J. Edwards, Motor sequence learning and motor adaptation in primary cervical dystonia, Journal of Clinical Neuroscience (2013), doi: http://dx.doi.org/10.1016/j.jocn.2013.08.019
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
D-13-00761 Clinical Studies
Motor sequence learning and motor adaptation in primary cervical dystonia
Petra Katschnig-Wintera,b,*, Petra Schwingenschuha,b, Marco Davarea, Anna Sadnickaa, Reinhold Schmidtb, John C. Rothwella, Kailash P. Bhatiaa, Mark J. Edwardsa a
Sobell Department of Motor Neuroscience and Movement Disorders, UCL Institute of
Neurology, Queen Square, London, UK b
Department of Neurology, Medical University of Graz, Auenbruggerplatz 22, 8036 Graz,
Austria
*Corresponding author. Tel.: +43 316 3851 6437; fax: +43 316 3851 4178. E-mail address: [email protected] (P. Katschnig-Winter).
Conflicts of Interest/Disclosures The authors declare that they have no financial or other conflicts of interest in relation to this research and its publication.
Abstract Motor sequence learning and motor adaptation rely on overlapping circuits predominantly involving the basal ganglia and cerebellum. Given the importance of these brain regions to the pathophysiology of primary dystonia, and the previous finding of abnormal motor sequence learning in DYT1 gene carriers, we explored motor sequence learning and motor adaptation in patients with primary cervical dystonia. We recruited 12 patients with cervical dystonia and 11 healthy controls matched for age. Subjects used a joystick to move a cursor from a central starting point to radial targets as fast and accurately as possible. Using this device, we recorded baseline motor performance, motor sequence learning and a visuomotor adaptation task. Patients with cervical dystonia had a significantly higher peak velocity than controls. Baseline performance with random target presentation was otherwise normal. Patients and controls had similar levels of motor sequence learning and motor adaptation. Our patients had significantly higher peak velocity compared to controls, with similar movement times, implying a different performance strategy. The preservation of motor sequence learning in cervical dystonia patients contrasted with the previously observed deficit seen in patients with DYT1 gene mutations, supporting the hypothesis of differing pathophysiology in different forms of primary dystonia. Normal motor adaptation is an interesting finding. With our paradigm we did not find evidence that the previously documented cerebellar abnormalities in cervical dystonia have a behavioral correlate, and thus could be compensatory or reflect ―contamination‖ rather than being directly pathological.
Keywords: Basal ganglia; Cerebellum; Dystonia; Motor adaptation; Motor learning; Sequence learning
1.
Introduction
Historically considered a disorder of the basal ganglia, there is now evidence for a wider network of neuroanatomical structures involved in the pathophysiology of dystonia. Recent research has particularly focused on the cerebellum against the background of clinical reports of patients with cerebellar lesions presenting with dystonia and animal models of dystonia in which the cerebellum appears to play a critical pathophysiological role.1 Radiological studies have demonstrated an increase in metabolic activity in the basal ganglia, supplementary motor areas and the cerebellum in a variety of forms of primary dystonia. 2, 3, 4 Additionally, diffusion tensor imaging data have demonstrated reduced integrity of cerebellothalamic tracts in DYT1 and DYT6, which correlates with the clinical penetrance of the mutation.5 Abnormal patterns of cerebellar activation are also seen using functional MRI blood oxygen level dependent techniques in motor tasks such as tapping and eye blinking in patients with dystonia of the corresponding regions. 6, 7 Preliminary neurophysiological data lends support to the notion that the cerebellum is affected in patients with primary focal dystonia of the neck or hand, with disturbed eye blink conditioning occurring in these patients.8 The absence of overt cerebellar signs on clinical examination in dystonia, however, suggests that the experimentally observed cerebellar dysfunction may either be too mild to be expressed clinically, might simply reflect unimportant ―contamination‖ of a structure directly connected to the basal ganglia, or might represent a compensatory response to the primary pathophysiology within the basal ganglia.
Behavioural paradigms can be used to study function of neuroanatomical structures likely to be involved in dystonia. Two of these are motor sequence learning (MSL; the incremental acquisition of sequential movement patterns) and motor adaptation (MA; paradigms that test
capacity to compensate for environmental changes). Serial reaction time tasks can be used to study MSL; following training blocks with randomly presented targets, sequence learning is demonstrated by faster reaction times to sequence presentation compared with random trials.9 MA paradigms require participants to implicitly adapt to changes in the environment caused by experimental manipulations such as altered visual feedback produced by prisms, perturbations to visual feedback on a computer monitor or force fields applied to movements of a robotic limb. The anatomical substrate of implicit sequence learning is extensive but is thought to be critically dependent on the basal ganglia while MA requires an intact cerebellar circuitry.10, 11, 12 and 13
Impaired MSL has been described in both manifesting and non-manifesting patients with the DYT1 mutation, but is normal in patients with DYT6 mutations.3, 14 MA has not previously been assessed in any patients with dystonia to our knowledge. Here we hypothesized that MSL would be impaired in cervical dystonia patients due to the presumed basal ganglia dysfunction that underlies the pathophysiology of this condition, but that MA would be normal, reflecting a compensatory rather than a primary pathological role for the cerebellum in this form of primary dystonia.
2.
Methods 2.1.
Participants
We recruited 12 patients (nine women, three men; mean age 58.8 ± standard deviation 9.6 years, range 40–77 years) with idiopathic cervical dystonia from the Movement Disorder outpatient clinic at the National Hospital for Neurology and Neurosurgery, Queen Square London, UK, and 11 healthy control subjects (seven women, four men) who were matched for age (mean age 55.4
± standard deviation 9.6 years, range 43–70 years) and years in education. Disease duration ranged from 6 to 36 years. All participants were right-handed and without cognitive impairment or psychiatric disease. Patients did not suffer from head tremor and there was no segmental spread of dystonia to the hands. Mobility at the elbow and shoulder joints was unrestricted and painless in all subjects. Dystonia severity was evaluated with the Toronto Western Spasmodic Torticollis Rating Scale 15 ranging from 5 to 49 out of 85 points. All patients receiving botulinum toxin treatment had their last injections at least 3 months before the study. Informed consent was obtained and the study was approved by the local ethics committee.
2.2.
General characteristics of the motor task
Subjects were seated in front of a computer screen with a joystick secured to the table directly in front of them. Subjects were instructed to move the cursor representing the joystick position from a central starting point on the computer screen to one of eight radial targets, which were evenly spaced 45° apart and displayed as red squares (1 cm2). Subjects were asked to move the cursor into the target as fast and accurately as possible upon target appearance. They were required to retain the cursor in the target until it switched to green (1 second following entrance to the square). At this point subjects were instructed to release their grip on the joystick handle, allowing the spring-loaded device to re-center for the next trial. There were three task conditions. In the motor reference task (R), targets were presented in a pseudo-randomized order in blocks of 40 trials. We used four random blocks (R1–4) at the beginning of the experiment to familiarize subjects with the task. Further random blocks were inserted after the sequence learning (R5, 6) and adaptation tasks (R7, 8). In the MSL task, a sequence of six targets was repeated seven times per block (42 movement trials). The presence of a sequence was not mentioned to the subjects. Four identical sequence blocks (S1–4) were presented. In the motor adaptation block (MAB) the
order of target presentation was pseudo-randomized and consisted of 40 movement trials. The direction of the cursor movement on the screen was rotated clockwise by 30° relative to hand movement. Patients and controls performed 13 blocks in the following order: R1-R2-R3-R4-S1-S2-S3-S4R5-R6-MAB-R7-R8. The total time of the experiment was 1 hour with opportunities for patients to rest between random blocks. No patient reported problems with fatigue or concentration.
2.3.
Data acquisition and analysis
Targets were presented and data acquired using MATLAB (version 7; The MathWorks, Inc., Natick, MA, USA) with Cogent Toolbox, interfaced with a CED 1401 analogue digital converter device (Cambridge Electronic Design, Cambridge, UK). Data were stored in a computer for offline analysis. The following parameters were measured for each movement trial in the MSL task: onset time (OT), the time from target presentation to movement onset; movement time (MT), the time from movement onset to target hit; response time (RT), the time from target presentation to target hit (sum of OT and MT); peak velocity (pV); directional error (DE), the angle between the ideal path (straight line) and trajectory taken at pV, which was also the parameter used for MA. A directional error of 0.90 for all comparisons).
3.2.
MSL
To investigate MSL we compared the performance in block S4 with R5. Both groups successfully learned the sequence, indicated by a significant effect of BLOCK for OT (F(1,21) = 20.9, p < 0.001), MT (F(1,21) = 6.8, p = 0.016), RT (F(1,21) = 13.6, p = 0.001) and DE (F(1,21) = 4.4, p = 0.048). Figure 1 shows the mean ± standard error of each parameter for both groups plotted against block. As in the motor reference task, we found a significant GROUP effect for pV and additionally for DE. Post hoc analyses revealed this to be due to a significantly higher pV in patients (563.6 ± 75.9 mm/s) compared to controls (491.5 ± 71.9 mm/s, p = 0.030), but controls had overall lower DE compared with patients (7.1 ± 1.3º versus 9.1 ± 1.5º; p = 0.003). We did
not find any GROUP × BLOCK interaction. Thus, MSL occurred in both groups to a similar extent as evidenced by simple movement and temporal parameters (OT, MT, RT), but performance strategy was different between groups with higher pV and lower aiming accuracy in patients with an opposite pattern in controls.
3.3.
MA
To assess MA we compared the mean of DE in block R6 with the mean in every 10 movement trials in the MAB (―first 10‖, ―second 10‖, ―third 10‖, ―last 10‖). To evaluate after-effects we computed the mean of DE for every 10 movements in block R7 and compared it to the mean of DE in block R6. After-effects were considered to be present as long as there was a significant difference compared to the baseline value (Fig. 2). In line with the performance in random blocks and MSL we found a significant effect for GROUP (F(1,21) = 5.1, p = 0.035) due to higher pV in patients (477.8 ± 132.3 mm/s) compared to controls (372.3 ± 82.6 mm/s). MA was present in patients and controls to a similar extent evidenced by a significant effect of BLOCK for DE (F(4,84) = 22.8, p < 0.001) in the absence of a GROUP × BLOCK interaction. DE significantly decreased from first 10 to third and last 10 movements (p < 0.005 for all comparisons). The presence of MA was confirmed in both groups by the presence of significant after-effects in the random block performed immediately after the MAB. We did not find a GROUP effect nor an GROUP × BLOCK interaction but there was a significant effect of BLOCK for DE (F(4,84) = 12.102, p < 0.001). Post hoc analyses revealed this to be due to a significant lower aiming accuracy in the first and second 10 trials of R7 compared to R6 (p < 0.05).
4.
Discussion
In this study we found patients with cervical dystonia showed normal general motor performance on a reaction time task, except for higher pV. Sequence learning and MA were both normal, although there was evidence that patients may have employed a different strategy to controls, utilizing higher peak velocity but lower aiming accuracy. We discuss the significance of this result in the context of current pathophysiological models of dystonia.
4.1.
General motor performance
We found general motor performance to be indistinguishable between patients and controls apart from a significantly higher pV in patients. Subtle changes in velocity profiles have previously been noted in dystonia; analysis of limb trajectories in clinically affected limbs with simple upper limb movements revealed that the velocity profiles of upper limb movements were less symmetrical due to longer deceleration times.16 The otherwise normal motor performance is in keeping with some previous studies,3 but not with others which reported, for example, reduced accuracy and longer movement times with reaching movements in patients with cervical dystonia.17
4.2.
MSL
During this experiment, subjects were not informed of the presence of a repeating sequence within the task. We did not find any deficit in MSL in cervical dystonia patients as evidenced by simple movement and temporal parameters, but there was evidence for a different performance strategy between patients and controls.
This is an interesting finding as studies investigating implicit MSL suggest that the basal ganglia are integral to the performance of this task. In healthy participants, imaging data acquired during MSL tasks suggests that the basal ganglia, cerebellum, premotor and prefrontal cortex are involved, 10, 11, 18 although imaging data may not necessarily reflect the learning process but rather the basic motor implementation of the improved performance. Patients with disorders involving the basal ganglia such as Parkinson’s disease (PD) or Huntington’s disease (HD) have been shown to be impaired in acquiring new motor sequences.19,
20, 21
However, by nature of the
fact that these are disorders of motor function some argue that it is the expression of the learnt response (for example due to slow movement initiation in PD) that is disturbed rather than the actual sequence learning itself.22 In line with the proposed role for the basal ganglia in the pathophysiology of dystonia, abnormal explicit MSL with an arm-reaching version of the serial reaction time tasks has been reported in non-manifesting and manifesting DYT1 carriers,3 but of interest this does not appear to occur in non-manifesting and manifesting patients with DYT6.14 Our data, which were collected with a modified version of this task and analyzed in a different way, suggest that in cervical dystonia neither the extent nor the distribution of basal ganglia dysfunction may be sufficient to impair sequence learning. In keeping with normal MSL in DYT6 dystonia, our data also indicate that there may be important pathophysiological differences between different forms of primary dystonia.
4.3.
MA
MA is a heavily cerebellar dependent task, and to our knowledge has not previously been studied in dystonia. It is hypothesized that the cerebellum generates an internal model for movement
which predicts the sensory consequences of motor commands. This internal model then generates an error signal to a given perturbation that drives adaptation.23 Stimulation of the cerebellum using anodal transcranial direct current stimulation improves visuomotor adaptation24 and patients with lesions to the cerebellum have deficits in paradigms that require adaptation.23 It is likely that the cerebellum makes its contribution via its output to the thalamus which in turn projects to the cerebral cortex, as thalamic manipulation in patients with deep brain stimulation for essential tremor can modulate performance in this task.25 In contrast, patients with basal ganglia disorders such as HD and PD are not impaired in adaptation tasks.26,
27, 28, 29
We did not
find any deficit in MA in patients with cervical dystonia. This is in contrast to a growing literature across a range of experimental techniques that the cerebellum may have a role in the pathophysiology of dystonia. Although our task may lack sensitivity to small impairments, the data suggests that in cervical dystonia, the adaptive capacities of the cerebellum to visuomotor changes are normal. One important qualifier to this statement is that our task did require subjects to stabilize in the target, and the cursor path was visible throughout the movement. Previous studies in healthy participants and patients with cerebellar degeneration have not found additional benefit to visuomotor adaptation from the additional (visual and proprioceptive) information provided by stabilization within the target zone,30 but it is possible that this is different for patients with dystonia.
5.
Conclusion
In the current study, patients with cervical dystonia showed normal MA. Hence, our findings do not support the hypothesis that previously documented abnormalities in cerebellum-dependent
tests (for example, eye blink conditioning) reflect a primary pathological role for cerebellar dysfunction in cervical dystonia. One may speculate that abnormalities on such tests reflect compensatory cerebellar changes or reflect ―contamination‖ via the direct connection of the cerebellum with other brain structures including the basal ganglia that are abnormal. However, we cannot exclude the possibility that our paradigm was not sensitive enough to pick up mild deficits in MA. We based our sample size on previous papers on motor learning in dystonia but refrained from a power calculation. Therefore, we cannot exclude the possibility that our study was underpowered which we recognize as a limitation. It would be of considerable interest to determine whether non-visuomotor adaptation (for example split-belt treadmill adaptation) was impaired in dystonia, as it is possible that impairment could differ depending on the nature of the sensory input that drives adaptation. In combination with electrophysiological, imaging and behavioral data from specific forms of primary dystonia, these data support the hypothesis that important pathophysiological differences may exist within the broad group of conditions currently labeled as primary dystonia. Further study of well-defined groups of patients with forms of primary dystonia (and indeed dystonia-plus, secondary and degenerative dystonia) will help determine specific subtypes of dystonia that may have different neuroanatomical substrates.
References
1. Sadnicka A, Hoffland BS, Bhatia KP et al. The cerebellum in dystonia - help or hindrance? Clin Neurophysiol 2011;123:65-70. 2. Carbon M, Ghilardi MF, Argyelan M et al. Increased cerebellar activation during sequence learning in DYT1 carriers: an equiperformance study. Brain. 2008;131:146-54. 3. Ghilardi MF, Carbon M, Silvestri G et al. Impaired sequence learning in carriers of the DYT1 dystonia mutation. Ann Neurol. 2003;54:102-09. 4. Trost M, Carbon M, Edwards C et al. Primary dystonia: is abnormal functional brain architecture linked to genotype? Ann Neurol. 2002;52:853-56. 5. Argyelan M, Carbon M, Niethammer M et al. Cerebellothalamocortical connectivity regulates penetrance in dystonia. J Neurosci. 2009;29:9740-47. 6. Baker RS, Andersen AH, Morecraft RJ et al. A functional magnetic resonance imaging study in patients with benign essential blepharospasm. J Neuroophthalmol. 2003;23:11-15. 7. Wu CC, Fairhall SL, McNair NA et al. Impaired sensorimotor integration in focal hand dystonia patients in the absence of symptoms. J Neurol Neurosurg Psychiatry. 2010;81:659-65. [8] Teo JT, van de Warrenburg BP, Schneider SA et al. Neurophysiological evidence for cerebellar dysfunction in primary focal dystonia. J Neurol Neurosurg Psychiatry. 2009;80:80-83. [9] Nissen MJ, Bullemer P. Attentional requirements of learning—evidence from performance measures. Cogn Psychol 1987;19:1-32. [10] Rauch SL, Whalen PJ, Savage CR et al. Striatal recruitment during an implicit sequence learning task as measured by functional magnetic resonance imaging. Hum Brain Mapp. 1997;5:124-32. 11. Schendan HE, Searl MM, Melrose RJ et al. An FMRI study of the role of the medial temporal lobe in implicit and explicit sequence learning. Neuron. 2003;37:1013-1025. 12. Diedrichsen J, Hashambhoy Y, Rane T et al. Neural correlates of reach errors. J Neurosci. 2005;25:9919-31. 13. Ghilardi M, Ghez C, Dhawan V et al. Patterns of regional brain activation associated with different forms of motor learning. Brain Res. 2000;871:127-45. 14. Carbon M, Argyelan M, Ghilardi MF et al. Impaired sequence learning in dystonia mutation carriers: a genotypic effect. Brain. 2011;134:1416-27.
15. Consky ES, Lang AE. Clinical assessments of patients with cervical dystonia. In: Jankovic J, Hallett M, editors. Therapy with Botulinum Toxin. New York, NY: Marcel Dekker Inc 1994; 211-37. 16. Inzelberg R, Flash T, Schechtman E et al. Kinematic properties of upper limb trajectories in idiopathic torsion dystonia. J Neurol Neurosurg Psychiatry. 1995;58:312-19. 17. Pelosin E, Bove M, Marinelli L et al. Cervical dystonia affects aimed movements of nondystonic segments. Mov Disord. 2009;24:1955-61. 18. Lehericy S, Benali H, Van de Moortele PF et al. Distinct basal ganglia territories are engaged in early and advanced motor sequence learning. Proc Natl Acad Sci U S A. 2005;102:12566-71. 19. Ghilardi MF, Silvestri G, Feigin A et al. Implicit and explicit aspects of sequence learning in pre-symptomatic Huntington's disease. Parkinsonism Relat Disord. 2008;14:457-64. 20. Schneider SA, Wilkinson L, Bhatia KP et al. Abnormal explicit but normal implicit sequence learning in premanifest and early Huntington's disease. Mov Disord. 2010;25:1343-49. 21. Siegert RJ, Taylor KD, Weatherall M et al. Is implicit sequence learning impaired in Parkinson's disease? A meta-analysis. Neuropsychology. 2006;20:490-95. 22. Seidler RD, Tuite P, Ashe J. Selective impairments in implicit learning in Parkinson's disease. Brain Res. 2007;1137:104-10. 23. Shadmehr R, Krakauer JW. A computational neuroanatomy for motor control. Exp Brain Res. 2008;185:359-81. 24. Galea JM, Vazquez A, Pasricha N et al. Dissociating the roles of the cerebellum and motor cortex during adaptive learning: the motor cortex retains what the cerebellum learns. Cereb Cortex 2011;21:1761-70. 25. Chen H, Hua SE, Smith MA et al. Effects of human cerebellar thalamus disruption on adaptive control of reaching. Cereb Cortex. 2006;16:1462-73. 26 Agostino R, Sanes JN, Hallett M. Motor skill learning in Parkinson's disease. J Neurol Sci. 1996;139:218-26. 27. Smith MA, Shadmehr R. Intact ability to learn internal models of arm dynamics in Huntington's disease but not cerebellar degeneration. J Neurophysiol. 2005;93:2809-21. 28. Cavaco S, Anderson SW, Correia M et al. Task-specific contribution of the human striatum to perceptual-motor skill learning. J Clin Exp Neuropsychol. 2011;33:51-62.
29. Gabrieli JD, Stebbins GT, Singh J et al. Intact mirror-tracing and impaired rotary-pursuit skill learning in patients with Huntington's disease: evidence for dissociable memory systems in skill learning. Neuropsychology. 1997;11:272-81. 30. Tseng YW, Diedrichsen J, Krakauer JW et al. Sensory prediction errors drive cerebellumdependent adaptation of reaching. J Neurophysiol. 2007;98:54-62.
Fig. 1. Graphs showing the effect of repetitive presentation of a target sequence on performance (motor sequence learning). Data are shown for block S4 (last block of sequence presentation) and R5 (subsequent block of random target presentation). Mean ± standard error are shown for (a) onset time, movement time and response time, (b) peak velocity, and (c) directional error. DE = directional error, MT = movement time, OT = onset time, pV = peak velocity, RT = response time.
Fig. 2. Mean ± standard error of directional error for both groups as a function of trial blocks. (a) Motor adaptation from R6 (preceding block of random target presentation), MAB f10 (first 10 trials of motor adaptation), MABs10 (second 10 trials of motor adaptation), MAB t10 (third 10 trials of motor adaptation), and MAB l10 (last 10 trials of motor adaptation). (b) After-effects from R6, R7 f10 (first 10 trials of random block following motor adaptation), R7 s10 (second 10 trials of random block following motor adaptation), R7 t10 (third 10 trials of random block following motor adaptation) and R7 l10 (last 10 trials of random block following motor adaptation).
