Motor skill learning in patients with cerebellar degeneration

https://doi.org/10.1016/S0022-510X(98)00115-4Get rights and content

Abstract

To explore the role of the cerebellum in learning a complex motor task, we studied nineteen patients with cerebellar degeneration and sixteen healthy subjects who attempted to improve their performance in generating a trajectory connecting five via points on a data tablet. Multijoint arm movements were performed at a constant total movement time, and spatial error was measured. Subjects performed 100 trials at a movement time of 3.5 s (slow movements), and another 100 trials at maximum speed (fast movements). With slow movements, patients and normal subjects reduced the error over trials to the same extent, but in patients, the rate of improvement was slightly slower. With fast movements, patients showed less improvement than normal subjects. When tested 24 h later, patients demonstrated significant retention of acquired skill and tended to improve more rapidly when performing both slow and fast movements than during the first session. We conclude that patients with cerebellar degeneration can exhibit almost normal performance in skill learning with slow movements, but with fast movements, their performance improves to a lesser extent. The problem may be difficulty in the refinement of motor execution, which is more of a requirement for fast movements than for slow ones.

Introduction

Numerous studies have suggested that the cerebellum and its associated brain stem circuitry plays a role in adapting motor behavior in order to master novel motor tasks. The architecture of the cerebellar cortex, with two distinct inputs regulating the activity of the single output neuron, the cerebellar Purkinje cell, is considered an ideal substrate for implementing learning capabilities in the nervous system 1, 9, 15. There is ample experimental evidence that the cerebellum plays an important role in certain apparently simple learning processes. Lesions of the vestibulocerebellum and its target nuclei in the brain stem reduce the ability of the animal to modify the gain of simple reflex behavior, such as the vestibulo-ocular reflex 14, 22. Visual–motor adaptation to rescaled visual inputs is impaired in patients with cerebellar lesions 6, 23. The cerebellum appears to be involved in the adaptation of single joint movements to novel loads [7]or to changes in gain of a visual–motor task [4]. In addition, cerebellar structures support other forms of simple motor learning, such as the acquisition of classically conditioned responses 19, 21. It is not yet known whether or not the pathways that support such simple forms of motor learning can also mediate more complex forms of motor learning. Experimental evidence bearing on this issue is scarce.

[18]compared the performance of patients with cerebellar degeneration with that of normal subjects who were attempting to improve their speed in tracing a polygon. Whether a true increase in motor capacity occurred is unclear because both groups increased their speed only at the cost of decreasing their accuracy. This trade-off of two associated movement variables [5]represents an alteration of motor behavior that we have called motor adaptation learning 12, 18. However, in the trade-off of speed for accuracy, no new capabilities, or skills, were necessarily added to the repertoire of available motor programs. To emphasize this attribute, we operationally defined motor skill learning as the improvement of one or more naturally antagonistic movement performance variables (e.g. speed, accuracy, energy savings) without the degradation of one or more others. In a second task of [18], the subjects also traced a polygon, but while observing their hand in a mirror. The normal subjects had an improvement in both speed and accuracy, but the patients with cerebellar cortical atrophy were deficient, indicating an abnormality of complex motor skill learning. Among several components of that task, however, was a requirement to perform a difficult visual–motor adaptation that may have prevented detection of a component of skill learning.

In the present experiments, we studied the improvement in accuracy of planar multijoint arm movements as a model of complex motor skill learning in patients with cerebellar dysfunction and in healthy subjects. To prevent a trade-off of the associated movement variables of speed and accuracy, we evaluated changes in movement accuracy at constant movement times. We also evaluated the acquisition of skill under different dynamic conditions (slow movements vs. fast movements) and assessed the retention of the learned motor skill over time.

Section snippets

Patients and methods

We studied ten patients with cerebellar cortical atrophy (CCA), aged 55±13 years (mean±SD), and nine patients with olivopontocerebellar atrophy (OPCA), aged 46±17 years. Sixteen healthy subjects, aged 43±14 years, served as control subjects. The patients' clinical characteristics are summarized in Table 1. The diagnosis was CCA if patients had only upper or lower limb ataxia and/or midline ataxia, dysarthria and cerebellar oculomotor dysfunction without radiological evidence of additional

Experimental day 1

Both the normal subjects and the patients were able to adjust their arm trajectories in order to nearly connect the via points when performing slow arm movements (Fig. 1). Initially, the normal subjects' and patients' movements showed large total spatial errors, of the order of 10.8±5.5 cm in the twelve normal subjects and 16.6±5.3 cm in the sixteen patients (P=0.01, Wilcoxon). With practice, however, their trajectories became more accurate (Fig. 2).

The initial ln(spatial error), a, tended to

Discussion

Our results show that although cerebellar patients exhibited poorer performances than normal subjects, at slow speeds, they were able to improve their complex motor skill to the same relative extent, if not at the same rate, as normal subjects. They also were able to retain a substantial amount of this acquired skill for at least 24 h. On the other hand, the patients clearly showed relative difficulty in improving their skill at high speed. We may try to understand these findings in terms of,

Acknowledgements

The authors wish to thank Dr T. Zeffiro for technical assistance with the data acquisition, and Ms B.J. Hessie for skillful editing. Dr Topka was supported by the Deutsche Forschungsgemeinschaft, Germany.

References (23)

  • V.B. Brooks et al.

    Movement programming depends on understanding of behavioral requirements

    Physiol Behav

    (1983)
  • P.F. Gilbert et al.

    Purkinje cell activity during motor learning

    Brain Res

    (1977)
  • E. Watanabe

    Role of the primate flocculus in adaptation of the vestibulo-ocular reflex

    Neurosci Res

    (1985)
  • J.S.A. Albus

    A theory of cerebellar cortex

    Math Biosci

    (1971)
  • A.J. Bastian et al.

    Cerebellar ataxia: abnormal control of interaction torques across multiple joints

    J Neurophysiol

    (1996)
  • G. Deuschl et al.

    Adaptation motor learning of arm movements in patients with cerebellar disease

    J Neurol Neurosurg Psychiatry

    (1996)
  • P.M. Fitts

    The information capacity of the human motor system controlling the amplitude of movement

    J Exp Psychol

    (1954)
  • G.M. Gauthier et al.

    Visual–motor adaptation. Quantitative demonstration in patients with posterior fossa involvement

    Arch Neurol

    (1979)
  • G.L. Gottlieb et al.

    Practice improves even the simplest movements

    Exp Brain Res

    (1988)
  • S. Grossberg

    On learning of spatiotemporal patterns by network with ordered sensory and motor components. I. Excitatory components of the cerebellum

    Studies Appl Math

    (1969)
  • M. Hallett et al.

    Physiologic studies of dysmetria in patients with cerebellar deficits

    Can J Neurol Sci

    (1993)
  • Cited by (0)

    View full text