Origin of error signal to the cerebellum and reinforcement signal to the striatum: an hypothesis

Movement commands are learned by a network which links three subsystems together: a cerebral subsystem which can learn a goal, and two additional subsystems involved in the two concepts of control augmentation and the stabilization augmentation. The subcortico-cerebral mechanism proposed selects the target pyramidal cells of both cerebellar and striatal networks, produces the cerebellar motor error correction or the strital-origin cerebral reinforcement, reduces the cerebellar error signal or increases the dopamine reinforcing signal. Within this frame, no hardware is needed. Any olivary afference having a negative perceptual significance to the global body scheme, may be considered as a possible cerebellar error signal. Central or peripheral signals topographically project to the inferior olive and then to related cerebellar cortical beams, and further to their learned cerebral targets. During a later cerebellar learning phase, sprouting mechanism will directly select the rubral targets, sending rubrospinal commands. Any dopamine-nigral afference, issued having a positive perceptual significance, may be considered as a possible striatal reinforcing signal