Cerebellum-machine interface to understand cerebellar roles in motor learning

摘要: A cerebellum-machine interface (CMI) was developed to understand cerebellar roles in motor control and learning by testing direct causality between single unit Purkinje cell activity learning. The CMI converts simple spike firing rate into pulse-width modulation signals that drive a joint robot arm. has no adaptive capability, thus any change observed arm motion can be attributed directly activity. We employed vestibuloocular reflex (VOR) adaptation paradigm goldfish as the test for Changes eye muscle characteristics, or equivalently adding visual stimulus during head rotation require oculomotor commands calibrated minimizing image slip across retina, which acts error signal. This calibration, known VOR learning, been postulated depend on synaptic plasticity vestibulo-cerebellum, where cells are believed encode movement. If this is true, then vestibulo-cerebellar may similarly learn correct command signal when system replaced with different system, such Desired of were presented retinal slip, respectively, fish model. shown decrease gradually, but not monotonically, many cases only one direction. evidence first demonstration capable implementing control. results also suggest responsible, at least part, directional selective previously reported [6] monkeys [3].