Parameter identification of a neurological control model for the pathological head movements of cerebellar patients

Abstract

The objective of this research is to explore the role of the cerebellum in the human motor control system. The present study quantitatively compares the neurological control signals effecting fast, horizontal head rotations in normal subjects to those in patients with a cerebellar lesion. The method involves the use of a computer simulation model for one degree-of-freedom movements. A method for unconstrained global optimization, first proposed by Hans Bremermann (1970), is used to identify the timing and magnitudes of the input neurological control signals to the model, which are compared to recorded electromyograms (EMGs). Experimentally recorded kinematics from cerebellar patients and from normal subjects were used to drive the parameter search. These simulations found that cerebellar patients' neurological control signals were altered with respect to those of normal subjects, and suggest that the electromyographic activity of cerebellar patients may comprise at least five bursts of activity whereas normal subjects typically exhibit only three. The results are discussed with respect to the hypothesis that the cerebellum may be involved in both the timing and magnitudes of the neurological control signals effecting voluntary movement.