Cerebellar-dependent adaptive control of primate saccadic system

Abstract
1. The ability of the central nervous system to compensate for saccadic dysmetria was demonstrated in rhesus monkeys. The behavior of this adaptive mechanism after cerebellar ablations was examined. 2. Monkeys were trained to fixate small target lights. Eye movements were monitored while the animals were seated, with their heads fixed, in a rotating magnetic field. The horizontal recti muscles of one eye were weakened by tenectomy. Saccades made by this weakened eye were hypometric and followed by postsaccadic drift. 3. When the patch was switched so that the weak eye was viewing, the hypometric saccades made by the weak eye gradually became larger, until after 3 days they were essentially orthometric. This indicated that the central nervous system could compensate for a peripheral weakness. 4. The tenectomy operation reduced the strength of the muscles, creating hypometria, and upset the ratio of viscosity to elasticity in the orbit, creating postsaccadic drift in the weak eye. The innervation required to make a saccade has both phasic and tonic components, the so-called pulse and step. The sacccadic repair mechanism increased both the pulse and the step to compensate for the hypometria and also adjusted the ratio of the pulse to the step to eliminate postsaccadic drift. 5. Total cerebellectomies were performed on two monkeys, each of which had one tenectomized eye. These ablations created an enduring saccadic hypermetria and postsaccadic drift in the unoperated eye of both animals. The total cerebellectomy abolished all adaptive repair of the saccadic system. 6. Partial cerebellectomies were performed on two monkeys, each of which had one tenectomized eye. Lesions of the vermis and paravermis (lobes IV-IX) and the fastigial nuclei created an enduring saccadic hypermetria without postsaccadic drift in the unoperated eye of both animals. These lesions abolished adaptive control of the pulse of innervation. Adaptive changes in the step of innervation still occurred, so that postsaccadic drift was always eliminated in the experienced, viewing eye. Thus the midline cerebellum (vermis, paravermis, and fastigial nuclei) appears to be important for repair of saccadic dysmetria, but not for repair of postsaccadic drift. Additional evidence that postsaccadic retinal slip cannot be compensated for in flocculectomized monkeys suggest that the adaptive control of the step may depend on the flocculus. 7. After cerebellar lesions the monkeys were able to make saccades of all amplitudes and directions. The principal deficit in these animals seemed to be that the pulse and step of innervation were no longer appropriate to the target displacement. We conclude that the cerebellum's principal contribution to saccadic eye movements is the adjustment of the gains of the pulse- and step-generating mechanisms. Hence this study supports the hypothesis that repair of dysmetria is a general function of the cerebellum.