RESEARCH _________________________ |
|
| It seems reasonable to assume (and there
is evidence suggesting) that movements are not actually planned in
terms of the requisite patterns muscle activity. However, the question
of how these high level plans are converted to muscle-like movement
command signals remains unresolved. The primary motor cortex is heavily
interconnected with the cerebellum, another brain structure long recognized
to play a critical role in limb movement. While motor cortex lesions
typically cause paralysis and an inability to move, cerebellar injury
leads to a loss of the motor learning necessary to achieve the normal
coordination and refinement of movement. Numerous models of the relation
between motor cortex and cerebellum have been proposed, but there
is no emerging consensus of opinion. We have undertaken experiments in my lab that combine dual-site cerebellar / motor cortical recording, micro-stimulation, and drug micro-injection. One advantage of this approach is that we can compare the movement-related modulation of the two sites with unparalleled precision. Equally important, we can determine the functional connections between a given pair of sites, while also studying the behavior-related signals from the same pair. This combination is vital in order to understand how signals are transformed as they propagate through the network. Micro-injection of drugs conveys the third critical component in these experiments, namely the ability to study the pharmacology of these identified networks, and to isolate the effects of different types of inputs. These experiments have led to important discoveries about the nature of the connections between the cerebellar nuclei (CN) and primary motor cortex (M1), and about the functional relation among signals recorded inM1, CN, and the cerebellar cortex. We have shown strong, spatially and temporally focused inputs from CN to M1, revealed both by the effects of CN neuron discharge and electrical micro-stimulation within CN. Connections in the other direction, from M1 to CN, are mediated both directly by mossy fibers (arising from pontine neurons which themselves receive M1 input) and indirectly via mossy fiber input to the cerebellar cortex, and thence to CN. We have demonstrated that the temporal pattern of impulses is critical to the efficacy of the direct inputs to the CN. Although these inputs are made up of relatively thin, sparse fibers, we have shown that they are responsible for a substantial portion of the CN discharge during behavior. We have also recently provided evidence that, despite their inhibitory effects on CN cells, Purkinje cells are typically positively correlated with muscle activity. These results have led us to postulate that Purkinje cells in the cerebellar cortex regulate the activity of the cerebellar nucleus predictively based on information about the prior occurrence of corrective movements, and on input from other parts of the cerebral cortex. The strong CN discharge, due in large part to the direct mossy fiber input, is apparently restrained by appropriately modulated PC inhibitory inputs. The resultant cerebellar nuclear signal is ultimately combined in the motor cortex with input from other cerebral cortical areas to produce the required muscle-like motor command. We are beginning a more extensive series of experiments to test this model using these methods while the monkey learns to adapt its limb movements to the presence of novel forces. We hypothesize that the initial neuronal adaptive response will occur in the cerebellar cortex, and be reflected in the Purkinje cell mediated component of the M1 to CN functional connection. Although this will result indirectly in altered M1 discharge, we predict that independent changes within the motor cortex will only occur, if at all, with significantly longer time course. Importantly, the picture of the cerebellar role in behavior has been broadened considerably in recent years, when it has become apparent that the type of interconnections with M1 are also maintained with a number of other cerebral areas, including the prefrontal cortex, associated with working memory and higher cognitive function. It is increasingly assumed that signal processing principles associated with the cerebellar refinement of movement may also be relevant to the more abstract role the cerebellum may play in "refining" cognitive processes. |
| The Feinberg School of Medicine, Department of Physiology, Northwestern University |
| Website modified on October 29, 2009 |