RESEARCH _________________________ |
|
| The motor cortex is the major source of
the brain's commands that control voluntary reaching and grasping
movements. As these commands descend to motor neurons in the spinal
cord, they need to be coordinated with similar commands from several
brainstem structures. In animal subjects, one can record signals from
individual neurons in these areas during movement. Since the pioneering
experiments of Ed Evarts in the mid 1960s, a mainstay of motor systems
research has been the comparison of the discharge of neurons in the
central nervous system (CNS) with various movement related signals.
Covariation of neuronal discharge with some parameter of the movement
is often taken as evidence to suggest that that neuron encodes (or
controls) that aspect of the movement. One of the foremost questions
is that of the coordinate system, or "language" of these
movement commands within any given brain area. These commands may,
for example, encode some aspect of a desired hand movement, possibly
its direction, speed, or extent. Alternatively, they might encode
the sequence and magnitude of muscle activity necessary to execute
a desired movement. Since the early 1980s, the hand-centered view of the primary motor cortex (M1) has been predominant, following a now classic series of experiments which showed that the strength of the movement-related activity depends on the direction in which an experimental monkey moves its hand. This result was interpreted to mean that the command to execute a particular hand movement is expressed by M1 neurons in the simple Cartesian coordinate system commonly used to describe positions in terms of the x, y, and z coordinates of three-dimensional space. If correct, this model requires that neurons of the spinal cord transform this simple Cartesian space command into the more complex "muscle space" command actually required to produce the movement. Research in my laboratory has cast doubt on this model. One important result of that research was the demonstration that the signals produced by single neurons in a brainstem motor area called the magnocellular red nucleus (RNm) closely resemble muscle activity, particularly the distal extensor muscles of the hand. In other words, these red nucleus signals express their movement commands in muscle space. These findings required a fundamental rethinking of the role of spinal circuitry because they strongly support the hypothesis that the spinal cord does not transform descending motor commands from Cartesian space to muscle space. Several years ago Dr. Klaus-Peter Hoffmann, Chairman of the Department of Biology and Neurobiology, of the Ruhr University in Bochum, Germany discovered a remarkable set of cells in the deep layers of the superior colliculus, in some ways analogous to the RNm cells. The superior colliculus has been known for many years to control the direction of eye and head orienting movements. However, these collicular "reach" cells appeared not to be related to eye movements, but rather to limb movements. Through a collaborative project with Dr. Hoffmanns' lab, we discovered that like RNm, the discharge of these neurons were well correlated with muscle activity. Unlike RNm, these cells were best correlated with muscles of the proximal limb and shoulder girdle. These findings strongly suggest that the superior colliculus is involved in directing the limb toward a target, in much the way the majority of its neurons direct gaze movements. This discovery was particularly important in that it identified a brainstem area that was complementary to the distal limb control provided by the more familiar red nucleus. More recently, experiments in my laboratory led to the discovery that the portions of the motor cortex which control the distal limb and hand do so by commands that appear to be expressed in a muscle space, very much like those of the red nucleus. We have also shown that it is possible to model the time-course of muscle activity using signals recorded from multiple motor cortical neurons. Most of the experiments to date have, out of technological necessity, involved a relatively modest number of simultaneously recorded neurons. However, we have begun to use a chronically implanted, state-of-the-art array of 100 electrodes, that is expected to yield simultaneous recordings of much larger numbers of neurons. Taken together, these results have demonstrated that many of the spinal projecting motor areas in both the brainstem and cerebral cortex provide a command to spinal neurons that is closely related to the time-course of muscle activity. Intriguingly however, they run counter to the theory suggesting that motor cortex neurons control the proximal arm by generating signals that encode the desired hand movement in Cartesian coordinates. Several theoretical studies from other labs have suggested that the interpretation of earlier experimental results which led to the Cartesian space model may not have been correct. We are currently running experiments focused on the proximal arm area of motor cortex in an attempt to resolve this apparent paradox. |
| The Feinberg School of Medicine, Department of Physiology, Northwestern University |
| Website modified on October 29, 2009 |