Distributed feedback systems for muscle control

Abstract

Responses of single neurones in motor cortex to passive stretches of the contralateral extensor digitorum communis muscle were studied in regionally anaesthetized, paralyzed cats. Parameters of stimulation could be adjusted to activate either only group Ia afferents, or a combination of group Ia, group II and group Ib afferents from muscle. Cooling of the ipsilateral cerebellar cortex or underlying nuclei was done to ascertain the role of the cerebellum in the feedback system from muscle to motor cortex. Activation of only group Ia fibres by this natural stimulation resulted in short latency (11 msec) responses in a well circumscribed region of pericruciate cortex. This region corresponded to the output region for this muscle as determined by electrostimulation in previous experiments of Asanuma and Sakata. These fast responses were usually unaffected by cerebellar cooling and were probably mediated by lemniscal pathways. Other cells in the same spatial region of motor cortex responded at a longer latency (18 msec) to higher threshold inputs from the same muscle. These latter responses were strongly influenced by cooling of the intermediate part of the anterior lobe of the cerebellar cortex or of interpositus nucleus. It was therefore concluded that this transcerebellar pathway was responsible for a significant portion of these higher threshold responses. Cells responding to the lemniscal system activation with an early response were found only in layers III and IV of motor cortex. In contrast, neurones responding to activation of the transcerebellar pathway were found in all layers of the same region of cortex, including layer V. The results are consistent with the view that proprioceptive information from single muscles controlling limb position is relayed over multiple parallel channels, including lemniscal and transcerebellar channels, to that region of motor cortex which directly controls the output of the muscle from which the feedback arises.