Multiple Sclerosis Resource Centre

Welcome to the Multiple Sclerosis Resource Centre. This website is intended for international healthcare professionals with an interest in Multiple Sclerosis. By clicking the link below you are declaring and confirming that you are a healthcare professional

You are here

Information processing speed impairment and cerebellar dysfunction in relapsing–remitting multiple sclerosis

Journal of the Neurological Sciences, Volume 347, Issues 1–2, 15 December 2014, Pages 246-250



The aim of this work is to study the relationship between information processing speed (IPS) impairment and motor testing that reflects cerebellar function in persons with multiple sclerosis (PwMS).


60 persons with relapsing–remitting multiple sclerosis with a mean disease duration of 4.2 ± 4 years were studied cross-sectionally. Motor cerebellar functioning was studied using the Nine-Hole Peg Test (NHPT) and the Kurtzke Functional Status Scales, and several cognitive domains were evaluated (IPS, working memory, episodic memory, attention, executive function). Correlations between the global NHPT score and neuropsychological test scores or impairment in each cognitive domain were studied using univariate and multivariate analyses.


The NHPT and a test of IPS significantly differentiated PwMS with and without cerebellar impairment. The NHPT total score was correlated with measures of IPS. Multivariate analyses showed a correlation between the NHPT and measures of IPS, but not between the NHPT and other neuropsychological tests that did not have a speed component.


In this sample of PwMS, motor cerebellar impairment assessed by the NHPT was correlated with IPS impairment.



  • Information processing speed is frequently impaired in multiple sclerosis (MS).
  • The cerebellum plays a role in cognition.
  • MS patients with cerebellar signs failed a test of information processing speed.
  • The Nine-Hole Peg Test (NHPT) correlates with measures of processing speed.
  • The NHPT does not correlate with cognitive tasks without speed component.

Keywords: Multiple sclerosis, Cerebellum, Cognition, Information processing speed, Neuropsychology.

1. Introduction

Cognitive deficits are frequently present in multiple sclerosis (MS) and could affect patients at all stages of the disease, including the very early stages[1] and [2], and all clinical phenotypes [3] . However, the pathogenesis of this impairment is still not completely understood. The most frequent feature of cognitive impairment in MS is slowness of information processing speed (IPS)[1], [2], [3], and [4]. The nature of IPS impairment suggests that it depends primarily upon the integrity of large-scale cortical integrative processes involving long-distance white matter projections that could be damaged by the disease process; however, the damage that occurs in deep and cortical grey matter and alterations to key network nodes are likely to also be involved [2] .

The cerebellum represents a major part of the central nervous system and its role in motor control and balance is well known. More recently, a role in cognition has been suspected based on anatomical, clinical and imaging studies[5], [6], and [7]. It has been suggested there is cerebellar involvement in age-related IPS impairment associated with small vessel disease [8] . The cerebellum is one of the sites that have a predilection for lesion development in MS, and extensive demyelination in the cerebellar cortex has been described [9] . However, a role for the cerebellum in the cognitive impairments observed in MS has only recently become a subject of interest. There is very little clinical data available showing a relationship between cerebellar dysfunction and cognition in MS, although an association between severe cognitive impairment (CI) and cerebellar signs has been noted by clinicians in some persons with MS (PwMS), and case reports illustrate this observation[10] and [11]. Decreased cognitive performance on measures of IPS and verbal fluency in persons with MS (PwMS) who have predominantly cerebellar symptoms compared to PwMS who do not have any cerebellar dysfunction has been reported [12] 2009).

We hypothesized that a negative relationship between cerebellar function assessed by a motor task, the Nine-Hole Peg Test (NHPT), and IPS impairment exists in PwMS.

2. Patients and methods

2.1. Patients

To study the correlation between IPS and motor testing, data from 60 persons with relapsing–remitting MS (RRMS) participating in a cross-sectional study about IPS in MS were investigated [3] . This study was approved by the ethics committee, including the institutional review board for human subject research of Bordeaux. All subjects gave written informed consent to participate in the study before their inclusion.

2.1.1. Eligibility criteria were as follows Inclusion criteria

RRMS diagnosis according to the Poser criteria, 18 years of age or older, French speaker, an elapsed time since the first MS symptoms of fewer than 10 years. Exclusion criteria

Persons with progressive MS, presence of any disease other than MS that could explain the symptoms, a history of psychiatric illness with the exception of stable depressive symptoms, starting or stopping antidepressants in the previous 2 months, alcohol, drug, or substance abuse in the previous 2 years, steroid treatment within the last 30 days, recent cognitive assessment (within less than 1 year).

2.2. Healthy controls

Normative data were obtained for the neuropsychological tests used in the MS group from a sample of 415 healthy controls (HC) divided into 20 groups according to age, sex, and education level, as previously described [4] .

2.3. Measures

All PwMS and HC were evaluated by qualified senior neuropsychologists. Six cognitive domains were evaluated: IPS, attention, working memory, verbal and visual episodic memory, and executive function ( Table 1 ).

Table 1 Cognitive domains and neuropsychological tests.

Domains IPS Attention Working memory Executive function Episodic verbal memory Episodic visual memory
Tests CSCT

Alertness (RT)

Visual scanning (RT):

- with a target

- without target

Ratio divided attention (RT):

- auditory

- visual

Flexibility (RT)
Ratio divided attention (AA):

- auditory

- visual

Visual scanning (AA):

- with a target

- without target

Numerical span test:

- forward

- backward
Flexibility (AA)

Stroop 45

WLG 90



10/36 SPART:

- Immediate recall

- Delay recall

IPS: Information Processing Speed, CSCT: Computerised Speed Cognitive Test4, RT: reaction time, in milliseconds, AA: Accurate Answers, PASAT 3 or 2: Paced Auditory Serial Addition Test 3.0 s or 2.0 s, WLG 90: Word List Generation test, SRT: Selective Reminding Test, LTS: Long-Term Storage, CLTR: Consistent Long-Term Retrieval, SRT-DR: Delay Recall, 10/36 SPART-IR: Spatial Recall Test for the immediate recall of short visuo-spatial memory, 10/36 SPART-DR: Spatial Recall Test for long-term visuo-spatial memory.

2.4. Neuropsychological evaluation

The neuropsychological (NP) assessment was previously described in detail [3] .

It included IPS evaluation using the Computerized-Speed-Cognitive-test (CSCT) [4] , a newly validated digit/symbol substitution test of IPS and reaction times values of four tasks from the Test of Attentional Performance (TAP, version 2.1) [13] : alertness, visual scanning (with and without the target), divided attention and flexibility. Psychometric properties and validity data about the CSCT have been published previously [4] . Other cognitive domains were assessed using the Paced-Auditory Serial Addition Test—3 s (PASAT 3 s), the Selective Reminding Test (the SRT and its 3 sub-scores: SRT-LTS = long-term storage, SRT-CLTR = consistent long-term retrieval, and SRT-DR = delayed recall), the 10/36 Spatial Recall Test (SPART), the delayed recall test (SPART-DR), and the Word List Generation test. Accurate answers on the computerized subtests from the TAP (alertness, visual scanning, flexibility, and visual and auditory divided attention) were measured, and the Stroop 45 second test and the numerical span test (forward and backward) were also performed.

2.5. Motor tests

Motor functioning was studied using the NHPT [14] and the timed 25 foot walk (T25FW) [15] . For both tests, two trials were performed at each session and the mean of the two trials was used. For the NHPT, a global NHPT score was defined as the mean time, in seconds, to complete the test using the dominant and non-dominant hand. Disability was measured using a French-adapted version of the Expanded Disability Status Scale (EDSS) [16] . PwMS were classified as having cerebellar impairment if their cerebellar Kurtzke Functional Status Scale (CKFSS) was ≥ 2, a score associated with a significant ataxia demonstrating clinical evidence of cerebellar impairment (score 1 is characterized by signs without disability) and having pyramidal impairment if their pyramidal Kurtzke Functional Status Scale (PKFSS) was ≥ 3, a score associated with an objective weakness of at least one limb, a score of 2 being not associated with an actual deficit demonstrating clinical evidence of pyramidal involvement. Subjective fatigue was measured, as previously described, by the UK Neurological Disability Scale fatigue score[1], [3], and [4].

Each subject answered questionnaires concerning depressive symptoms (Beck Depression Inventory II, BDI II). Subjects were considered to be free of depressive symptoms if their BDI II scores were below 13, to have mild depressive symptoms if their BDI II scores were between 14 and 19, to have moderate depressive symptoms if their scores were between 20 and 28, and to have severe depressive symptoms if their scores were greater than 29.

2.6. Statistical analysis

Statistical analyses were performed using StatView version 5.0 software for Windows. For age and disease duration, the results are shown as the means ± SDs. For the EDSS, the results are shown as the medians (ranges). For all analyses, differences were considered significant whenpvalues were less than 5%.

In the MS group, thezscores were calculated for each NP score using the following formula: (patient's score − mean value of HC group matched for age, sex, and education level) / SD of the matched HC. Z scores were calculated for each cognitive domain using the following formula: sum of the patient's NPzscores for each domain / the number ofzscores in each domain. For a given NP test, patients were considered impaired if theirzscores were below the fifth percentile for their matched HC group. CI for a given domain was defined by at least one impaired test of this domain.

Several univariate analyses were performed for prediction of the NHPT score by NP or cognitive domainzscores. Multivariate linear regressions were performed to assess the correlation between total NHPT score as the dependent variable and scores of individual NP tests. A second group of multivariate linear regressions were performed to assess the correlation between total NHPT score as the dependent variable and cognitive domainzscores.

For all multivariate models, only independent variables that had a conservative significance level ofp < 0.25 in the univariate analysis were entered simultaneously. The models included age and fatigue ifp < 0.25 in univariate analyses for these variables. Factors not significant at the 0.05 level were removed from the model by stepwise elimination. EDSS was forced in all models as a covariate.

3. Results

3.1. Demographic and clinical characteristics of PwMS

These characteristics are presented in Table 2 . The NHPTzscores did not differ significantly between PwMS with pyramidal involvement (PKFSS ≥ 3) and those without pyramidal impairment. On the contrary, the total NHPTzscore was significantly greater in patients with cerebellar impairment (CKFSS ≥ 2) (− 0.42 ± 0.65) than in those with CKFSS < 2 (0.46 ± 0.42) (p = 0.01,t = − 2.6). The EDSS also differs significantly between PwMS with and without cerebellar involvement (.p < 0.001,t = − 5.3). The following variables did not differ significantly between patients with and those without cerebellar impairment: age, sex, disease duration,zscore T25W and fatigue score. One PwMS (0.02%) and one matched HC (0.003%) counterpart had severe depressive symptoms. Nineteen PwMS (31.7%) and 21 of their matched HC (0.07%) counterparts had mild to moderate depressive symptoms.

Table 2 Demographic and clinical baseline characteristics of multiple sclerosis patients.

N 60
Women (%) 81.6%
Age a 37.3 ± 9.9
Disease duration a 4.1 ± 3.0
EDSS median (range) 1.5 (0–4.5)
NHPT global score a 21.5 ± 3.4
N of patients with PKFSS ≥ 3 2/60
N of patients with CKFSS ≥ 2 11/60

a Mean ± SD: Age and disease duration in years; Nine Hole Peg test (NHPT) in seconds; PKFSS: pyramidal Kurtzke Functional Status Scale; CKFSS: cerebellar Kurtzke Functional Status Scale.

3.2. Cognitive assessment

Only one cognitive testzscore was significantly different between PwMS with and without cerebellar impairment (zscore CSCT: − 0.57 ± 1.46 in PwMS without cerebellar impairment versus − 1.7 ± 1.66 in PwMS with cerebellar impairment (p = 0.04,t = 2.0)). The mean scores for NHPT were significantly worse in CI PwMS for IPS (mean deviation (MD) = − 4.1;p < 0.001) but did not differ between CI and cognitively unimpaired (CU) PwMS for the other domains (executive functions, working memory, attention, episodic verbal or visuospatial memory). The T25FW mean scores were slightly worse in CI PwMS for IPS (MD = − 0.6;p < 0.05) and for executive functions (MD = − 1.3;p < 0.05) than in CU PwMS. No correlation was observed between the NHPT and BDI II scores.

3.3. Correlation between cognitive z scores and NHPT and between cognitive domains and NHPT

Table 3 presents univariate and multivariate correlations between NP scores or cognitive domainzscores and the NHPT score. The only NP score that remained in the final model and was significantly correlated with the NHPT was the CSCT. IPSzscore correlated with the NHPT in univariate analysis. In multivariate analyses, episodic verbal memory and working memory were added to the model (p < 0.25 in univariate analyses). The only cognitive domain remaining in the final model was the IPS domain.

Table 3 Correlation between cognitive scores and NHPT and cognitive domainszscores and NHPT.

Dependant variable NHPT score (s) Retained independent variables (p < 0.25), NP scores R Retained independent variables (p < 0.25) Cognitive domains z scores R
  CSCTlowastlowastlowast, b − 0.36 z working memory a − 0.22
SRT-CLTRlowast a − 0.24 z verbal memory a − 0.17
Span a − 0.25 z IPSlowastlowast b − 0.36
Stroop a − 0.2 EDSS a 0.35
Visual scanning RTlowastlowast a − 0.40 Disease duration a 0.28
EDSSlowastlowast, b 0.35    
Disease durationlowast, b 0.28    

The table displays variables included in multivariate analyses but not significantly correlated with NHPT in final models (a) and variables significantly correlated with NHPT in multivariate analyses final models (b).

lowast,lowastlowast, andlowastlowastlowastindicate significantpvalues in univariate analyses:lowastindicatesp < 0.05,lowastlowastindicatesp < 0.01,lowastlowastlowastindicatesp < 0.001.

The models on the left part of the table concern individual NP scores:pand adjustedR2values of the models arep < 0.001 and 0.22

The models of the right part of the table concern cognitive domainszscores:pand adjustedR2values of the models arep < 0.05 and 0.15

See Table 1 for abbreviations of tests names. RT = reaction times; AA = accurate answers.

The T25FW correlates with the EDSS (r = 0.3), the fatigue score (r = 0.4) and visual memoryzscore (r = 0.4,p < 0.05).

4. Discussion

In this dataset, measures of IPS were consistently correlated with the NHPT scores independently of the EDSS, age and subjective fatigue in PwMS. Multivariate analyses showed that the NHPT correlated with the CSCT, a measure of IPS. When looking at correlations with cognitive domains, the NHPT correlated with IPS exclusively.

A possible limitation of this study is that NHPT is not a pure measure of cerebellar motor kinetic control and that in some patients, pyramidal arm deficits could contribute to the scores. It is very unlikely in this sample, since only two patients had a PKFSS ≥ 3. Moreover the NHPT scores were significantly worse in the group of PwMS with cerebellar impairment detected at the neurological examination (CKFSS ≥ 2). Interestingly the only NP score significantly different between patients with or without cerebellar impairment according to the CKFSS was the CSCT. Indeed, there is a large amount of literature suggesting that it is a valid measure of cerebellar functioning. The NHPT is a frequently used complex manual task [17] which is part of the MS functional composite (MSFC) [15] . After cerebellar infarction the NHPT has been shown a reliable measure of manual dexterity [18] . In MS it has been shown that the NHPT correlated well with postural tremor and may provide useful objective methods for assessing arm dexterity in tremulous PwMS [19] . Another study confirms that the NHPT could distinguish patients who have predominantly motor or cerebellar symptoms from patients who have predominantly sensory dysfunctions [20] . Indeed, the NHPT has been used as an outcome measure in a clinical trial of symptomatic therapies in patients who have cerebellar syndrome secondary to MS [21] . Using magnetic resonance imaging (MRI) it has been shown that the NHPT correlates with cerebellar cortical grey matter volume, although the T25FW, another timed measure of motor function, did not [22] . In this sample of PwMS it may therefore be hypothesized that the NHPT could be a valid measure of cerebellar motor control.

Very few studies have looked at the correlation between the NHPT and cognitive performance in MS. In one study of the reliability and validity of self-report health measures, it has been shown that PwMS who have impaired performance on the SDMT have worse NHPT scores than those who do not have impairment on this test [23] . Another study reported a correlation between the NHPT and the PASAT in PwMS, but this correlation was not maintained when the EDSS was entered in regression analysis [24] . In a more recent study focusing on inter-relationships between motor functioning and cognition in PwMS, a significant correlation was observed between the NHPT and measures of IPS and executive functions [25] . The interpretation of the authors was that motor functioning could be predicted by impairment of executive functions. This interpretation was reinforced by the fact that similar correlations were observed with the T25FW score. Interestingly, regression analyses showed that the only cognitive test remaining in the final models for NHPT in PwMS was the SDMT, which is mainly a test of IPS; although in the models for the T25FW, the SDMT and the California Verbal learning test-II, a test for episodic verbal memory, were both remaining in the final models. However, the Delis–Kaplan Executive Function System Sorting Test, which is a more specific test of executive functions, did not correlate with the NHPT in PwMS in that study. In our study, the different tests used to assess executive functions did not correlate with the NHPT. When PwMS were classified as cognitively impaired according to each cognitive domain, the NHPT was worse only in those CI for IPS, although the T25FW was worse in PwMS CI for executive functions and for IPS, but with a lower MD. In correlation analyses the T25FW did correlate with the visual memory domainzscore but not with IPS. The 25TW scores did not differ between patients with and without cerebellar involvement in this sample. This reinforces the hypothesis of a more specific link between IPS and cerebellar function.

There are several studies in the literature suggesting that cerebellar involvement could contribute to IPS impairment. In a study comparing 21 PwMS who had predominantly cerebellar symptoms and 21 PwMS who did not have any cerebellar dysfunction, IPS and verbal fluency performances were worse in those who had cerebellar symptoms than in the controls [12] . One limitation of our study is the absence of brain MRI to investigate the mechanisms underlying IPS impairment in PwMS and cerebellar damage or dysfunction. In an MRI study using lesion probability mapping of 54 PwMS, PASAT scores correlated with the amount of white matter lesions in different areas including the right cerebellum [26] . In another study of 121 PwMS using a similar method, SDMT performances correlated with widespread frontal, temporal and cerebellar hemispherical lesions correlated with SDMT performance [27] . In a 5-year longitudinal study of subjects who had clinically isolated syndromes, new lesions in the cerebellum, thalami, corpus callosum and frontal lobes after 1 year correlated with cognitive impairment at 5 years [28] . Several functional MRI (fMRI) studies using working memory tasks such as the PASAT or others have yielded various results based on patient selection (MS stage, cognitive status), sample size and task[29], [30], and [31]. Depending on the study, an increase or decrease in cerebellar activation was observed. The decrease in cerebellar activation was considered to be a failure of the cerebellum in its role facilitating rapid cognitive performances. fMRI studies of IPS have shown more consistently that the cerebellum is involved in fast cognitive processing in healthy subjects but not in PwMS[32], [33], and [34]. For instance, using a task of four successive conditions in a Go/No-go test of increasing complexity and cognitive demands in 20 healthy subjects, a significant correlation was noted between shorter reaction times and increased cerebellar activation [32] . The same paradigm was applied to 15 RRMS patients who had early-stage MS [33] . In PwMS, shorter reaction times were no longer associated with higher cerebellar activation, while compensatory activation of medial prefrontal area was observed. In PwMS, to the contrary of HC, no functional connectivity interactions were found between the prefrontal cortex and cerebellum. These results suggest that PwMS activate a substitute compensatory network instead of the typical cerebello-frontal network associated with the fastest responses in the task.

In a recent study using a Stroop task, the authors observed that the functional connectivity between cerebellar and prefrontal areas was higher in healthy subjects than in persons with RRMS or secondary-progressive MS [34] .

5. Conclusion

Altogether these studies and the results reported in the present study suggest that cerebellar impairment in MS may jeopardize cognitive functions and force the use of compensatory strategies, which are effortful and costly, that lead to IPS impairment.

Study funding

The CSCT study was supported by Bayer Healthcare, France. The sponsors did not participate in any aspects of the study design; in the collection, analysis and interpretation of data; in the writing of the report; and in the decision to submit the article for publication.

Study disclosures

Authors did not report disclosure in relation to the study.


  • [1] M.S. Deloire, E. Salort, M. Bonnet, et al. Cognitive impairment as marker of diffuse brain abnormalities in early relapsing remitting multiple sclerosis. J Neurol Neurosurg Psychiatry. 2005;76(4):519-526 Crossref
  • [2] N.D. Chiaravalloti, J. DeLuca. Cognitive impairment in multiple sclerosis. Lancet Neurol. 2008;7(12):1139-1151 Crossref
  • [3] A. Ruet, M.S. Deloire, J. Charré-Morin, et al. Cognitive impairment differs between primary progressive and relapsing–remitting MS. Neurology. 2013;80(16):1501-1508 Crossref
  • [4] A. Ruet, M.S. Deloire, J. Charré-Morin, et al. A new computerised cognitive test for the detection of information processing speed impairment in multiple sclerosis. Mult Scler J. 2013;19(12):1665-1672 Crossref
  • [5] J.D. Schmahmann. From movement to thought, anatomic substrates of the cerebellar contribution to cognitive processing. Hum Brain Mapp. 1996;4(3):174-198 Crossref
  • [6] J.D. Schmahmann. Disorders of the cerebellum, ataxia, dysmetria of thought, and the cerebellar cognitive affective syndrome. J Neuropsychiatry Clin Neurosci. 2004;16(3):367-378 Crossref
  • [7] E. Keren-Happuch, S.H. Chen, M.H. Ho, et al. A meta-analysis of cerebellar contributions to higher cognition from PET and fMRI studies. Hum Brain Mapp. 2014;35(2):593-615
  • [8] M.A. Eckert, N.I. Keren, D.R. Roberts, V.D. Calhoun, K.C. Harris. Age-related changes in processing speed, unique contributions of cerebellar and prefrontal cortex. Front Hum Neurosci. 2010;4(10):1-12
  • [9] A. Kutzelnigg, J.C. Faber-Rod, J. Bauer, et al. Widespread demyelination in the cerebellar cortex in multiple sclerosis. Brain Pathol. 2007;17(1):38-44 Crossref
  • [10] M. Zarei, S. Chandran, A. Compston, J. Hodges. Cognitive presentation of multiple sclerosis, evidence for a cortical variant. J Neurol Neurosurg Psychiatry. 2003;74(7):872-877 Crossref
  • [11] N.P. Staff, C.F. Lucchinetti, B.M. Keegan. Multiple sclerosis with predominant, severe cognitive impairment. Arch Neurol. 2009;66(9):1139-1143
  • [12] P. Valentino, A. Cerasa, C. Chiriaco, et al. Cognitive deficits in multiple sclerosis patients with cerebellar symptoms. Mult Scler. 2009;15(7):854-859 Crossref
  • [13] P. Zimmermann, B. Fimm. Testbatterie zur Aufmerksamkeitsprüfung (TAP). Version 2.1. Herzogenrath: Psytest (, 2008)
  • [14] V. Mathiowetz, G. Volland, N. Kashman, et al. Adult norms for the nine hole peg test of finger dexterity. Occup Ther J Res. 1985;5:24-38
  • [15] J.S. Fischer, R.A. Rudick, G.R. Cutter, et al. The Multiple Sclerosis Functional Composite Measure (MSFC): an integrated approach to MS clinical outcome assessment. National MS Society Clinical Outcomes Assessment Task Force. Mult Scler. 1999;5(4):244-250
  • [16] B. Brochet. Assessing incapacity at early stages of multiple sclerosis using the EDSS. Rev Neurol (Paris). 2009;165(Suppl. 4):S173-S179 Crossref
  • [17] D.E. Goodkin, D. Hertsgaard, J. Seminary. Upper extremity function in multiple sclerosis, improving assessment sensitivity with box-and-block and nine-hole peg tests. Arch Phys Med Rehabil. 1988;69(10):850-854
  • [18] J. Liepert, T. Kucinski, O. Tüscher, et al. Motor cortex excitability after cerebellar infarction. Stroke. 2004;35(11):2484-2488 Crossref
  • [19] S.H. Alusi, J. Worthington, S. Glickman, L.J. Findley, P.G. Bain. Evaluation of three different ways of assessing tremor in multiple sclerosis. J Neurol Neurosurg Psychiatry. 2000;68(6):756-760 Crossref
  • [20] L.P. Erasmus, S. Sarno, H. Albrecht, M. Schwecht, W. Pöllmann, N. König. Measurement of ataxic symptoms with a graphic tablet, standard values in controls and validity in multiple sclerosis patients. J Neurosci Methods. 2001;108(1):25-37 Crossref
  • [21] P. Feys, M.B. D'hooghe, G. Nagels, W.F. Helsen. The effect of levetiracetam on tremor severity and functionality in patients with multiple sclerosis. Mult Scler. 2009;15(3):371-378 Crossref
  • [22] V.M. Anderson, L.K. Fisniku, D.R. Altmann, et al. MRI measures show significant cerebellar gray matter volume loss in multiple sclerosis and are associated with cerebellar dysfunction. Mult Scler. 2009;15(7):811-817 Crossref
  • [23] S.M. Gold, H. Schulz, A. Mönch, et al. Cognitive impairment in multiple sclerosis does not affect reliability and validity of self-report health measures. Mult Scler. 2003;9(4):404-410 Crossref
  • [24] N. Yozbatiran, F. Baskurt, Z. Baskurt, et al. Motor assessment of upper extremity function and its relation with fatigue, cognitive function and quality of life in multiple sclerosis patients. J Neurol Sci. 2006;246(1–2):117-122 Crossref
  • [25] R.H. Benedict, R. Holtzer, R.W. Motl, et al. Upper and lower extremity motor function and cognitive impairment in multiple sclerosis. J Int Neuropsychol Soc. 2011;17(4):643-653 Crossref
  • [26] J. Sepulcre, J.C. Masdeu, M.A. Pastor, et al. Brain pathways of verbal working memory, a lesion-function correlation study. Neuroimage. 2009;47(2):773-778 Crossref
  • [27] Z.T. Kincses, S. Ropele, M. Jenkinson, et al. Lesion probability mapping to explain clinical deficits and cognitive performance in multiple sclerosis. Mult Scler J. 2011;17(6):681-689 Crossref
  • [28] D. Wybrecht, F. Reuter, W. Zaaraoui, et al. Voxelwise analysis of conventional magnetic resonance imaging to predict future disability in early relapsing–remitting multiple sclerosis. Mult Scler J. 2012;18(11):1585-1591 Crossref
  • [29] B. Audoin, D. Ibarrola, J.P. Ranjeva, et al. Compensatory cortical activation observed by fMRI during a cognitive task at the earliest stage of MS. Hum Brain Mapp. 2003;20(2):51-58 Crossref
  • [30] A. Cerasa, L. Passamonti, P. Valentino, et al. Cerebellar-parietal dysfunctions in multiple sclerosis patients with cerebellar signs. Exp Neurol. 2012;237(2):418-426 Crossref
  • [31] H.A. Wishart, A.J. Saykin, B.C. McDonald, et al. Brain activation patterns associated with working memory in relapsing–remitting MS. Neurology. 2004;62(2):234-238 Crossref
  • [32] M.C. Bonnet, B. Dilharreguy, M. Allard, et al. Differential cerebellar and cortical involvement according to various attentional load, role of educational level. Hum Brain Mapp. 2009;30(4):1133-1143 Crossref
  • [33] M.C. Bonnet, M. Allard, B. Dilharreguy, et al. Cognitive compensation failure in multiple sclerosis. Neurology. 2010;75(14):1241-1248
  • [34] M.A. Rocca, M.C. Bonnet, A. Meani, et al. Differential cerebellar functional interactions during an interference task across multiple sclerosis phenotypes. Radiology. 2012;265(3):864-873 Crossref


a Service de Neurologie, CHU de Bordeaux, F-33076 Bordeaux, France

b INSERM-CHU CIC-P 0005, CHU de Bordeaux, F-33076 Bordeaux, France

c Neurocentre Magendie, INSERM U862, Université de Bordeaux, F-33076 Bordeaux, France

d Translational Research and Advanced Imaging Laboratory (TRAIL) cluster of excellence, Université de Bordeaux, F-33076 Bordeaux, France

lowast Corresponding author at: Department of Neurology, CHU de Bordeaux, 33076 Bordeaux cedex, France. Tel.: + 33 556795521; fax: + 33 556796025.