The VOR belongs to a broader class ofmotor learning termed adaptation. During adaptation the sensory consequences of movements are altered so that subjects must learn new relationships between their motor commands and the environment. In the case of the VOR, compensatory eye movements must be remapped to vestibular signals so that the same degree of head movement results in a new amount of eye movement.

The eye movements learned during VOR adaptation differ critically from many other forms of learned movements, such as reaching movements. Unlike the dynamics of the eye, the dynamics of a multijoint system, such as the arm and hand, has many degrees of freedom and is highly nonlinear. For most multijoint systems, including that composed of the wrist, arm, and shoulder, determining what muscles to activate given the desired trajectory is an ill-posed problem. Ill-posed problems are those that are insufficiently constrained so that there are multiple possible solutions. For example, there are many different sets of muscle commands that will cause the hand to move to a particular location. This situation complicates motor programming considerably because the control system must determine which solution to pursue as it attempts to optimize behavior. Therefore, attributing errors to the appropriate muscle commands becomes a complex procedure.

A common form of adaptation is prism adaptation. In this experimental paradigm, the subjects wear goggles fitted with prisms that laterally displace the visual image on the retina. Thus, if the goggles displace the image to the right, information from the left side of the visual field will fall on the fovea as the subject looks straight ahead. This apparatus alters the relationship between movements of the body and visual information. For example, in order to reach for an object that appears on the fovea when the eyes and head are pointing directly ahead, one must move the arm to a position that is displaced to the left or right of the body. This new relationship between vision and action contradicts a lifetime of experience, but particular movements can be successfully relearned in a matter of minutes.

When first worn, the prism goggles cause pointing movements to be systematically displaced from their targets, with the magnitude of the distortion approximately equal to the extent that the goggles displace the visual signal. However, performance improves quickly with practice. Within approximately 20 trials, the systematic error is eliminated and performance matches that observed before the goggles were introduced. This improvement can stem from distinct forms of behavioral adjustment that are unrelated to motor learning per se. Conscious strategies can compensate for the shifted visual image if one intentionally directs movements a few degrees to the left or right of the target to adjust for the displacement. To obtain measurements of adaptation that are uncontaminated from compensatory strategies, data are collected after adaptation has occurred and the goggles are removed. Under these conditions, subjects initially exhibit negative aftereffects of similar magnitude but in the opposite direction of the displacements observed when they first wore the goggles. Presumably, after removing the goggles subjects abandon any conscious compensatory strategies. Thus, the negative aftereffect indicates that the adaptation reflects a learned automatic process remapping visual information and motor commands.

To better understand the computational processes that are altered during adaptation, it is useful to distinguish between open-loop and closed-loop movements. Open-loop movements are programmed without reference to any sensory feedback. Thus, error signals from the visual and proprioceptive systems are unable to alter the motor programs. In contrast, closed-loop movements allow for sensory feedback to indicate adjustments in the ongoing motor command. When movements are made relatively slowly, in a closed-loop manner, the effects of the prism goggles may not be obvious. However, when movements are made ballistically, in an open-loop manner, they will terminate in a location that is displaced from the desired endpoint.

Importantly, adaptation based on interactions between visual and proprioceptive feedback does not transfer to unpracticed movements or effectors. That is, if an individual learns to make accurate rapid pointing movements exclusively with the right arm, performance with the left arm will show little benefit. This finding indicates that learning does not entail a generic remapping of visual coordinates to an internal model of space. Instead, the remapping involves specific motor commands to coordinates in egocentric space. However, although learning appears to be specific to particular sets of movements, it does generalize to other portions of the visual field. In summary, learning appears to be better characterized as motoric rather than perceptual.

Individuals with cerebellar damage demonstrate deficits in their ability to adapt to the prism goggles. Although the patients do exhibit some improvement in performance when wearing the goggles, there are no aftereffects once the goggles are removed, suggesting that the improvement was based on conscious strategy rather than true adaptation. This finding is consistent with the proposal that the cerebellum plays a critical role in coordinating motor commands with sensory information, and that such learning occurs at a procedural level.

Adaptation can also occur without manipulating visual inputs. Peter Gilbert and Thomas Thach trained monkeys to move a manipulandum with the muscles of the wrist to a specific target location. The manipulan-dum was fitted with a torque motor that could apply forces to the handle for random intervals so that the monkey had to change the amount of force exerted on the handle to return it to the target location. Once the monkeys learned to maintain the handle's location with two levels of force (one requiring activation of extensor muscles and the other requiring activation of flexor muscles), a new level of force was introduced, replacing one of the learned levels. The task is considered to measure adaptation because after training with the novel level, effects are observed on the unchanged force level.

As the monkeys performed this task, single-unit recordings were made from Purkinje cells in the cerebellar cortex. The frequency of complex spikes, driven by climbing fiber activity, increased when the novel force was introduced and returned to baseline rates as the animals adapted. The frequency of simple spikes, driven by activity along the parallel fibers, decreased as the complex spike frequency increased and remained depressed as the complex spike frequency returned to baseline. These findings indicate that the climbing fiber activity is affected by learning and that the climbing fibers may serve to modulate the strength of the Purkinje cell-parallel fiber synapses.

Understanding And Treating Autism

Understanding And Treating Autism

Whenever a doctor informs the parents that their child is suffering with Autism, the first & foremost question that is thrown over him is - How did it happen? How did my child get this disease? Well, there is no definite answer to what are the exact causes of Autism.

Get My Free Ebook

Post a comment