Preserved and impaired aspects of predictive grip force control in cerebellar patients

https://doi.org/10.1016/j.clinph.2005.02.015Get rights and content

Abstract

Objective: To analyze preserved and impaired aspects of feedforward grip force control during cyclic arm movements with a hand-held object after cerebellar damage.

Methods: We tested eight subjects with unilateral or bilateral cerebellar pathologies and eight healthy control subjects. Participants performed cyclic vertical arm movements with a hand held instrumented object at three different speeds.

Results: Compared to controls, patients excerted increased grip forces. The minimum force ratio between grip force and load force was constant across all movement frequencies, suggesting that patients anticipated speed-related changes in load magnitudes by adjusting the grip force. Thus the scaling of grip force level to self-generated load magnitudes was preserved. The coupling between grip and load profiles was assessed by cross correlation analysis. Patients exhibited significantly decreased maximum coefficients of cross correlation implicating impaired anticipation of inertial load fluctuations. However feedforward control could be preserved, as obvious from zero time lags of the maximum cross correlation coefficient.

Conclusions: Our findings suggest that cerebellar lesions affect the processing of predictive grip force modulation in anticipation of inertial loads. Our results add further evidence to the theoretical concept that the cerebellum implements internal feedforward models. However, preserved functions may indicate compensatory mechanisms or extracerebellar aspects of grip and load force regulation.

Significance: The observed dissociation of performance deficits may have direct clinical implication and may guide the development of individual therapeutic strategies for patients with cerebellar disorders.

Introduction

Holding and transporting objects are skilful motor processes, frequently used in daily life. The motor commands for hand and finger forces must be selected and adjusted to the objects' physical properties, such as weight, shape and surface friction as well as to dynamic movement properties such as inertia.

In repetitive vertical movements the summation of gravitational and inertial loads results in a maximum of net load force at the lower turning point of the movement and a minimum at the upper turning point of the movement. It has been found that grip force changes in parallel with load force with a maximum peak in grip force occurring at the lower turning point of the movement and a minimum grip force peak at the upper turning point of the movement (Flanagan, 1993, Flanagan and Wing, 1995). The simultaneous modulation of grip and load forces indicates an anticipatory mode of grip force adjustment to upcoming load changes during object transport (Flanagan, 1993, Johansson and Cole, 1994, Flanagan and Wing, 1995). It has been proposed that such a control mechanism is based on the use of internal models of limb and object dynamics (Flanagan and Wing, 1997, Kawato and Wolpert, 1998, Blakemore and Frith, 2001, Flanagan and Vetter, 2003, Kawato and Kuroda, 2003). In this particular context, internal models may enable grip force anticipation. The cerebellum is considered to be the anatomical correlate of such internal models (Miall and Weir, 1993, Wing and Flanagan, 1998, Kawato and Wolpert, 1998, Kawato and Kuroda, 2003). Thus the cerebellum may play an important role in the prediction of the arm movement and its impact on the object dynamics, which is reflected in the coupling between grip and load force during object manipulation. In addition, single cell recordings in monkeys and imaging studies in healthy volunteers support the idea of a central role of the cerebellum in grip force regulation during object manipulation (Espinoza and Smith, 1990, Dugas and Smith, 1992, Kawato and Kuroda, 2003).

A number of studies investigating cerebellar pathologies documented impairments of grip force control. Holmes (1917) investigated acute cerebellar lesions after gun-shot injuries and showed impairments of the maintenance of near-isometric grip forces, problems of force calibration, increased response latencies to perturbations and lower grip force amplitudes (Holmes, 1917). Mai and Bolsinger investigated grip force behaviour in chronic cerebellar disease and also found an instability of isometric force maintenance, deficits of fast repetitive force changes and impaired force tracking (Mai and Bolsinger, 1988). Maximally generated grip forces were normal. Grip force control during grasping, lifting and holding of an object in patients with cerebellar atrophy was examined by Muller and Dichgans (1994b). Patients showed longer latencies between the onset of grip and lift forces, irregular force output and reduced peak grip forces. Also, Fellows et al. found a deficit in the temporal coordination between proximal arm muscles and finger forces when cerebellar patients lifted an object (Fellows et al., 2001). In contrast to the findings of Muller and Dichgans (1994b), Fellows et al. found exaggerated grip force levels in patients with cerebellar atrophy. When patients with cerebellar atrophy performed vertical point to point movements with a hand held object they applied exaggerated grip force levels to hold and transport the object (Nowak et al., 2002a). In addition, the precision of grip force modulation with the movement induced loads was impaired in patients compared to control subjects.

Müller and Dichgans tested patients with unilateral cerebellar diseases and found similar deficits of grip and load force coordination as found in cerebellar atrophy patients, however the deficits were restricted to the affected side (Muller and Dichgans, 1994a). Fellows et al. showed that focal damage to the dentate nucleus resulted in similar deficits in the control of fingertip forces during object lifting as seen in cerebellar atrophy with lower lifting velocities and elevated grip force levels (Fellows et al., 2001). However, damage to the posterior inferior cerebellar artery (PICA) region did not have any effect on grip force coordination during object manipulation (Fellows et al., 2001).

In order to investigate how cerebellar disease affects predictive grip force adjustments to different inertial loads we examined grip force control during cyclic vertical arm movements with a hand held mass performed at three different speeds. Continuous cyclic movements impose quite vigorous demands on the motor system planning grip force. The rapid fluctuations in load force, both in time and magnitude, have to be compensated by accurate grip force adjustments planned several steps ahead. Patients were asked to move the object at three different frequencies, resulting in different inertial load fluctuations. The question is how cerebellar subjects adjust their force levels to self-induced inertial load variations. For instance, a general increase in grip force is a quite common observation during object manipulation in cerebellar subjects (Babin-Ratte and Sirigu, 1999, Fellows, 2001, Nowak, 2002). Given the results of previous findings, we hypothesize that: (1) Patients use exaggerated grip forces to compensate load force changes in cyclic movements. (2) Patients may nevertheless increase their grip force level with increasing load and keep the ratio between grip and load forces constant across different inertial loads. Such an observation might suggest that grip force scaling is processed differentially to variations in arm movement speed, despite impaired force calibration. (3) The precision of the coupling between grip force and movement induced load fluctuations may be impaired. Our data may shed some light to the understanding how the cerebellum contributes to predictive finger force control during object manipulation.

Section snippets

Subjects

Eight subjects with unilateral or bilateral cerebellar pathology (five males, three females aged between 40 and 76 years, mean age 56.7 years) took part in the experiments. Five patients presented with degenerative cerebellar disorders, two with spinocerebellar ataxia type 6 (SCA6), three with idiopathic cerebellar ataxia (IDCA). These disorders are known to affect primarily the cerebellum (Klockgether, 2000). Three patients had focal cerebellar lesions, two presented with cerebellar ischemic

Individual performance

Fig. 2 illustrates the profiles of acceleration, grip force and load force during cyclic movements performed at low, intermediate and high cycle frequencies. Approximately four movement cycles at each speed condition are shown for a control subject (Fig. 2A) and for a patient with cerebellar atrophy (Fig. 2B). Acceleration of the object resulted in load changes with a load minimum occurring at the upper turning point of the movement cycle and a load maximum at the lower turning point of the

Discussion

Here we demonstrate impairments of grip force and load force coupling during vertical cyclic movements with a hand held mass in cerebellar patients. The following major results were obtained: (1) Cerebellar patients used abnormally high grip force levels in object manipulation. (2) Patients scaled the grip force level differentially to inertial loads arising from various cycle frequencies. (3) Despite the fact that cerebellar patients retained the ability to modulate grip force with

Conclusion

The results provide further evidence that the scaling of grip force level to load magnitudes and the anticipatory modulation of grip force with load fluctuations during object manipulation are separately controlled motor processes. Thus, the calculation of the time-varying grip force profile in anticipation of the cyclic load changes and the scaling of this profiles to the actual load magnitudes may constitute different aspects of the task that may be controlled by segregated modules of an

References (32)

  • S.J. Blakemore et al.

    The cerebellum is involved in predicting the sensory consequences of action

    Neuroreport

    (2001)
  • C. Dugas et al.

    Responses of cerebellar Purkinje cells to slip of a hand-held object

    J Neurophysiol

    (1992)
  • E. Espinoza et al.

    Purkinje cell simple spike activity during grasping and lifting objects of different textures and weights

    J Neurophysiol

    (1990)
  • J.R. Flanagan et al.

    Grip-load force coupling: a general control strategy for transporting objects

    J Exp Psychol: Hum Percep Perform

    (1994)
  • J.R. Flanagan et al.

    The stability of precision grip forces during cyclic arm movements with a hand-held load

    Exp Brain Res

    (1995)
  • J.R. Flanagan et al.

    The role of internal models in motion planning and control - evidence from grip force adjustments during movements of hand-held loads

    J Neurosci

    (1997)
  • Cited by (71)

    • The cerebellum as a movement sensor

      2019, Neuroscience Letters
      Citation Excerpt :

      Such rapid modulation of force indicates that the motor system is able to predict the timing and magnitude of load variations induced by voluntary transport as well as the sensory effects of a falling object. Patients with cerebellar damage tend to generate grip forces that exceed the magnitudes required to adequately maintain hold on an object [13–15]. Furthermore, the parallel modulation of grip force with movement induced load variation is decoupled in patients with cerebellar degeneration [13,15].

    View all citing articles on Scopus
    View full text