Strategies underlying the control of disordered movement.

The purpose of this article is fourfold. First, a theory of motor control--the dual-strategy hypothesis--is outlined. Second, the methodologies and theoretical framework that are used to develop this theory are examined. Third, motor dysfunction is discussed in the context of this theory. In particular, Down syndrome, Parkinson's disease, cardiovascular accidents, and spasticity are discussed. Finally, potential applications of the theory to physical therapy are considered. [Corcos DM. Strategies underlying the control of disordered movement. Phys Ther. 1991; 71:25-38.]

Key Words: Electromyography; kinesiology/biomechanics, general; Motor activity; Motor control theories; Motor dysfunction; Movement. Strategies Underlying the Control of Disordered Movement The field of motor behavior is concerned with understanding how movements are controlled and how they are learned. A subset of the field of motor behavior is concerned with motor dysfunction in which questions are asked concerning the characteristics and underlying causes of particular motor deficits in individuals with movement impairment. This article presents ideas primarily from one theory of motor control to suggest how they can be used to develop a rational basis for therapeutic intervention in motor dysfunction. One major assumption that underlies this approach is that theories of motor control developed from studies performed on healthy individuals can be used to form a basis for treating movement impairment. This article will suggest that this may be the case for several disorders of movement, but may not always be true. Certain disorders of movement, specifically those in which unwanted involuntary movements interact with voluntary movements, will be considered because it may well be that, in such disorders, the principles that underlie treatment should be determined functionally and not from models of motor control. This point can be illustrated by considering different possible treatments of cerebral palsy. If the primary cause of motor dysfunction is coactivation, then one can either teach an individual with cerebral palsy how to move despite the coactivation or try to first reduce the coactivation and then teach "normal" movements. The first approach will not lead to generating movements in a normal fashion, but will allow certain types of movement tasks to be accomplished. The second approach may ultimately prove to be preferable because performance in the presence of pathological coactivation will probably never reach the same levels as without this coactivation.

Theories of Motor Control

One central question of interest in the study of motor behavior is: How are movements controlled? In order for movement to take place, neural commands must activate muscle. Several theories have been advanced concerning the representation of these commands. These theories differ in terms of what is being controlled to generate a movement. The dual-strategy hypothesis of motor control, the principle theory to be developed in this article, suggests that movements are controlled by two different sets of rules fo activating muscle. [1] The first set of rules applies to tasks in which movements can be successfully made at any speed. For example, if an individual is asked to move from point A to point B, the movement is correct if the individual moves to point B. The time it takes to make this movement is unimportant for successfully completing the task. Contrast this type of movement with the task of catching a ball. If an individual does not get his or her hand to the right place at the right time, the movement will be unsuccessful. Movements of the first kind are referred to as "speed-insensitive," whereas movements of the second kind are referred to as "speed-sensitive." These two types of movements are controlled differently. In the case of speed-sensitive movements, the duration of excitation to motoneuron pools is prolonged and the muscles are activated later for longer movements or movements made against larger loads. [2] These factors lead to predictable changes in both electromyographic (EMG) activity and movement kinematics, which will be discussed in the section on methdological issues. Movements that are made at different required speeds are termed speed-sensitive and are controlled by changing the intensity of activation to motoneuron pools rather than by changing the duration of activation. [3] In the case of speed-sensitive movements, the antagonist muscle is activated earlier as movement speed increases.

Another theory of motor control is the equilibrium-point hypothesis. According to one version of this theory, movements are made by regulating the threshold of a length-sensitive reflex. In this theory, a movement occurs when there is a shift in the equilibrium point from the current limb position to a new position. Movement control can then be described in terms of two parameters: the final value of the threshold of the reflex ([lambda]) and the speed at which the threshold moves ([omega]). For isotonic movements, the final value of the reflex threshold is related to final position, and the speed at which the threshold moves is related to movement speed (ML Latash, GL Gottlieb; unpublished research). [4] This theory is entirely compatible with the dual-strategy hypothesis (see Gottlieb et al [5] for a discussion of this point) and is mentioned here because some articles on disorders of movement have been written based on this hypothesis (ML Latash, DM Corcos; unpublished research).

Another class of models suggests that movements are controlled by sets of commands to muscle. [6-8] One command is to the agonist muscle and is associated with a relatively constant EMG duration for movements that are less than 30 to 40 degrees. [9,10] A second command is to the antagonist muscle and is sometimes followed by a second command to the agonist. These three commands are often referred to as the "triphasic EMG pattern." According to this model, movements are primarily controlled by increasing the number of motor units recruited, not the duration over which they are recruited. This model has formed the basis of a large number of studies on movement dysfunction. However, it is important to note that this pattern is mainly observed during very rapid movements and has not been observed over a wide variety of movement tasks. [1]

Methodology in Motor Control

The study of motor behavior involves integrating findings from behavioral work, physiology, and biomechanics. [11] In order to fully understand the nature of a movement deficit, it is important to understand (1) which movement tasks can and cannot be performed, (2) the physiological mechanisms (both normal and abnormal) underlying the performance of the task, and (3) the type of movement trajectory that is generated.

Task Analysis

Any time a person performs a movement, he or she requires two sets of information. The first concerns objective information about what the person should do. This might be information about the position in space to which the person should move or about the load to be moved. Variables that define what a person should do have been defined as task variables. [1](p190) The second set of information refers to how the task should be performed and is usually presented in the form of verbal instructions. Verbal instructions include information about such factors as the speed and accuracy of the movement. Therefore, a task refers "to the union of the task variables and instructions: those things necessary for a subject to make a specific movement." [1](p190)

When a person performs a task, many variables can be measured that describe this performance. Examples are movement speed, movement time, and movement accuracy. These variables are called performance variables. A movement task gives rise to a movement strategy, defined as "a set of rules between a movement task and measured variables, sufficient to perform the task." [1](p192) The important point is that there are at least two sets of rules and hence two strategies for controlling movement. This point can also be inferred from the equilibrium-point hypothesis, in which one parameter ([lambda]) is related to the control of the final limb position and the other parameter ([omega]) to the speed at which the movement is performed.

Nearly all of the studies that will be discussed in this article have used tasks in which individuals made movements at only one joint. This is because such movements are much easier to analyze compared with multi-joint movements using EMG techniques. For a more comprehensive review of theories underlying the control of movements with one mechanical degree of freedom, see the article by Gottlieb et al. [1] It should, however, also be possible to apply the principles derived from the analysis of such movements to the performance of tasks that require using more than one degree of freedom, as will be discussed in the section on organizing principles. For further discussion on this issue, see the article by Corcos and colleagues. [12] For a review of studies using other tasks, see the chapter by Campbell. [13](pp220-224) See the article by Wolf et al [14](pp729-730) for a discussion of the problems inherent in relating EMG and kinematic measures to treatment efficacy.

Physiology

The electrical signal that can be measured on the surface of muscles is called the EMG signal and is shown in the bottom left section of Figure 1, which captures selected features of the motor control system. The figure depicts how a movement task determines a movement strategy that gives rise to an "excitation pulse," which is the input to the motoneuron pool, and ultimately produces the EMG signal.

The EMG is one of the principal tools for identifying mechanisms underlying motor control and motor dysfunction. In the dual-strategy hypothesis of motor control, the EMG is interpreted as a low-pass filtered version of an excitation pulse that is the input to the motoneuron pools. The low-pass filtering assumption is required because the motoneuron pool serves to smooth the many inputs that converge on it. The EMG is analyzed because the excitation pulse per se is not measurable in the intact human organism. [1] The schematic diagram is Figure 2 depicts a simple model of the rectified, filtered EMG signal. The model is a first-order differential equation with an amplitude-limiting nonlinearity. The amplitude-limiting nonlinearity is required because there are only a given number of motor units and they can only fire up to a certain frequency. The excitation pulse is assumed to be a rectangular pulse that can be modulated in height or width. If the height is altered, then the EMG signal rises more steeply. If the duration is modulated, the EMG signal rises at the same rate. A more detailed account of the filtering effects of the motoneuron pool and muscle can be found in the article by Gottlieb et al. [5]

Biomechanics

The schematic diagram in Figure 1 shows that one consequence of the action potential train is that contractile elements generate torques that may or may not lead to movement, depending on whether the contraction is isometric or isotonic. [1] The analysis of movement impairment also requires information about the motion of limbs, the forces generating the motion, and the forces generated by the motion, a very real problem for patients with central nervous system dysfunction. Such information can be in the form of forces and torques (kinetics) as well as position, velocity, and acceleration (kinematics). Such knowledge can allow one to deduce factors about the control of movement. In the study of patients with hyperactive stretch reflexes, for example, the analysis of position and velocity records can allow the researcher to deduce the presence of a movement deficit. So, for example, Corcos and colleagues [15] showed that when individuals with hyperreflexia (spasticity) make dorsiflexion movements, their movement may be impeded by the presence of a hyperactive stretch reflex that momentarily reverses movement direction. This reversal can most clearly be seen in the position record of Figure 3A. This phenomenon is somewhat difficutlt to observe without equipment to record EMG and kinematic data. However, it may sometimes be so obvious that the individual will report having noticed it upon detailed questioning. For example, the data in Figure 3B depict a plantar-flexion movement to a target. The movement slightly overshot the target. When the limb started to dorsiflex back to the target, the patient experienced clonus. Upon noticing this phenomenon, I asked the patient whether he drives a car, and to my astonishment he replied that he does! He often has to manually lift his foot off the accelerator, because he frequently experiences clonus upon making dorsiflexion movements to ease up on the accelerator.

Organizing Principles

Based on the relationship between movement tasks, physiological principles, and biomechanics, Gottlieb et al generated four organizing principles for the study of motor control, as follows:

I. Elements of a movement task lead to a strategy governing its control. We have identified two strategies to describe how single-joint movements are accomplished for a variety of tasks.

II. Strategies consist of sets of rules that determine the patterns of muscle activation.

III. Rules for muscle activation lead to patterns of muscle torques and EMGS. Well-chosen scalar measures of torque and EMG will be highly and consistently correlated irrespective of task because of their shared causal, neural activation.

IV. Muscle torques interact with limb loads to generate kinematics (angle, its derivatives, and movement intervals). Because of the role of load in determining kinematics, no general correlations between EMG and purely kinematic measures are possible. [2](pp342-343)

Based on these organizing principles, Gottlieb et al[2] demonstrated the presence of two strategies (sets of rules) for perfoming movements and also the relationship of the EMG measure to selected measures of torque. Figure 4 depicts two sets of movements that are performed by using the two different strategies. The data represented by the dark dashed lines are for one subject moving foru different distances. The inertial torques (acceleration multiplied by moment of inertia) rise at approximately the same rate and are independent of the distance moved. The same observation is true for the EMG activity. This finding should be contrasted with the plot of the data represented by the lighter dashed lines. In this case, in which the subject was asked to move one distance at four different speeds, both EMG activity and inertial torque diverge early in the trajectory. The movements are controlled in the first case by changes in the duration of the excitation pulse (speed-insensitive strategy) and in the second by changes in the intensity (speed-sensitivity strategy), as was shown in Figure 2. The differences in the two control schemes are depicted in Figure 5. It i important to note that the speed-insensitive strategy can be used with any value of intensity [16] and that the presentation in Figure 5 is not meant to imply that speed-sensitive movements have larger values of intensity than speed-insensitive movements.

The essential feature of the dual-strategy hypothesis is that at least two different sets of rules underlie the control and regulation of movement. These sets of rules are dependent on the type of task to be performed but are not limited to single-degree-of-freedom tasks. For example, Wadman et al [17] had subjects perform diverse tasks including playing simple tunes on the violin and trombone. (See Wadman et al's [17] Figure 15, which depicts trombone slide velocities during movements of different slide distances, using what appears to be a speed-insensitive strategy based on the uniform rate of rise of velocities. Their Figure 17 shows bow movements during violin playing at three different tempos, using what appears to be a speed-sensitive strategy.) Lee and colleagues [20] looked at horizontal pulls on a cable and found patterns of ground-reaction forces and hand-pull tensions that can be described using the terminology of the dual-strategy hypothesis. These data are presented in Figure 6. The dual-strategy hypothesis can, in principle, also be applied to gait. [19,20] In this gait study, subjects were asked to walk at three different speeds and ground-reaction forces were measured. The force records at the three different speeds had different rates of rise and are distinctively speed-sensitive in nature, as can be seen in Figure 7. The data in these studies all suggest that insight into the control of more complex movement tasks can be obtained through partitioning tasks into speed-insensitive and speed-sensitive categories.

From the perspective of physical therapy, the observation that there are at least two sets of rules for activating muscles implies that movement impairment can be characterized by an inability to modulate the intensity of the excitation pulse to the agonist and/or antogonist muscle, an inability to modulate the duration of the excitation pulse to the agonist and/or antagonist muscle, and inappropriate scaling of the latency of the antagonist muscle. These three possibilities are not mutually exclusive.

This description also needs to be evaluated in light of a distinction drawn between "reduced output paresis" and "subtraction paresis." [21] Certain disorders of movement can be characterized by diminished input to motoneuron pools, whereas others can be characterized by excessive output. As such, motor dysfunction can occur either because of a diminished excitation to motoneuron pools or because the timing and/or size of the excitation is excessive.

The causes of inappropriate levels of excitation to motoneuron pools can originate from either supraspinal mechanisms or spinal mechanisms, or from a combination of both. In many disorders, the actual cause of the movement deficit is difficult to identify because damage to parts of the brain can lead to changes in the delicate balance between inhibition and excitation that exist in the spinal cord as well as to long-term changes in the structure of muscle. The distinction between supraspinal and spinal mechanisms refers to the primary source of the problem and becomes important when certain types of therapeutic intervention are considered.

Application of Motor Control

Theories to Movement

Impairment

In this section, three disorders of movement will be discussed in which the primary cause is a supraspinal deficit in the ability to control movement. Down syndrome and Parkinson's disease will be discussed from the perspective of the dual-strategy hypothesis because several studies have been conducted that can easily be interpreted within this framework. Motor dysfunction arising from cerebrovascular accidents will be discussed in more general terms because no studies have been identified by the author that provide data that can be clearly interpreted within this framework. For a study to be easily interpreted within the framework of the dual-strategy hypothesis, EMG and kinematic data must be analyzed for at least two different tasks. The interpretation of the data is also greatly facilitated if data plots of the tasks are superimposed. This section will conclude with a subsection on spasticity, which occurs in a wide variety of disorders of movement.

Down Syndrome

The individual with Down syndrome can be characterized as moving and reacting slowly. Several reasons have been postulated for these characteristics, including hypotonicity and an inability to generate normal levels of force. [22,23] The key findings from the perspective of the dual-strategy hypothesis are that individuals with Down syndrome might lack the ability to modulate the intensity with which they activate motoneuron pools. This evidence comes from work by Latash and Corcos (unpublished research) and Cole et al. [24] In the study by Cole et al, [24] individuals were asked to lift a 200-g object with a sandpaper or a satin finish, which would necessitate different grip forces to lift. Individuals in the control group modulated the rate of grip force required to lift the object. In contrast, individuals with Down syndrome reached the appropriate levels of force, but did so by "increasing the duration of grip force application." [24(p754)] The data in Figures 8A and 8B illustrate this point. Latash and Corcos (unpublished research) had individuals with Down syndrome generate movements of different distances. They found that the pattern of activation of the muscles was very similar to that of control subjects (compare this pattern with the dark dashed lines of Fig. 4) but that the movements were much slower. The implication of this finding is that the intensity of activation is less in individuals with Down syndrome than in control individuals, but that individuals with Down syndrome modulate duration of activation in a normal manner to accomplish movements over different distances (Fig. 8C).

Parkinson's Disease

The movement deficits of patients with Parkinson's disease are numerous. However, one deficit that ties in with the dual-strategy hypothesis is that of an inability to modulate the rate at which different levels of force are generated in isometric tasks or an inability to accelerate at different rates in isotonic tasks. For example, Hallett and Khoshbin [25] had patients with Parkinson's disease and healthy subjects generate movements over different distances. The healthy subjects showed an early divergence of the angle trace for longer movements. This type of trajectory implies that a speed-sensitive strategy was being used and that the subjects were accelerating much more quickly for long movements than for short movements. However, the parkinsonian patients tended to follow the same trajectory for both long movements and short movements. Both sets of data are presented in Figure 9.

Although the pattern demonstrated in the parkinsonian patients is frequently observed in healthy individuals and has been termed "default," [16] the question of interest is whether parkinsonian patients have access to both strategies. This issue awaits further research.

Similar arguments can be applied to isometric contractions as studied by Stelmach and Worringham. [26] They had individuals with Parkinson's disease and healthy subjects generate isometric contractions to different levels of their maximal voluntary contraction. They observed that the major difference between the groups was not in the maximum levels of force that could be generated, but in the rate at which the force could be developed. This difference can also be seen in the work of Wing, [27] who showed lower rates of force increase and also force decrease in parkinsonian individuals.

Cerebrovascular Accidents

Cerebrovascular accidents frequently lead to "upper motoneuron syndrome." The movement disabilities associated with this syndrome include weakness, slowness, clumsiness, spasticity, and abnormal movement synergies. Berardelli et al [28] investigated how patients (aged 50-65 years) with upper motoneuron lesions of vascular origin make movements of different distances. They had healthy individuals and patients make elbow flexion movements over 10, 20, and 30 degrees. Their main finding was that the agonist and antagonist EMG activity duration of the patients were prolonged compared with those of the healthy individuals. They suggest that prolonged duration of muscle activity is a compensatory mechanism for generating sufficient force to perform the task. Again, these data are consistent with the view that these individuals do not have the capacity to generate excitation pulses of large intensities.

Spasticity

Increased muscle tone and muscle spasms are major problems facing the physical therapist in delivering appropriate therapy in many disorders that are associated with spasticity. As such, careful consideration should first be given to the use of antispastic medication in alleviating these unwanted signs of spasticity. The advent of drug-pump technology has allowed for the delivery of antispastic medications that have considerably more effect than oral medication. [29-31]

Latash and colleagues [32,33] have shown that refleces can be abolished 2 to 3 hours after a bolus injection of baclofen and that voluntary movement can be improved. This finding is shown in Figure 10A, in which voluntary dorsiflexion movements are plotted before and after the administration of baclofen, and in Figure 10B, in which elbow flexion movements are depicted.

Latash and colleagues [32,33] attribute thee dramatic changes in EMG activity, kinematics, and the consequent capacity to generate voluntary movement to changes that occur in the spinal cord and not to the central reorganization of motor commands. In terms of the dual-strategy hypothesis of motor control, this finding suggests that improved performance is due to diminished input to the motoneuron pools of antagonist muscles. Further studies on these patients and others also showed an improvement in their activities of daily living. [34]

Synopsis

Van der Kamp et al [35] have summarized several studies of rapid movements that have been performed on different patient populations. The results are presented in the Table and show that the EMG and kinematic parameters of certain disorders of movement differ from the EMG and kinematic profiles of healthy individuals. [25,35-40] However, the limited differential diagnostic value of using only one task for the assessment of dysfunction can be seen when comparisons are made among the various disorders. A clearer description of the differences between movement disorders may emerge when tasks requiring speed-sensitive and speed-insensitive strategies are used. Any deficits observed, however, should also be evaluated after individuals have had considerable opportunity to practice the task that is being used to determine the deficit. [41] One of the factors that makes individuals with movement deficits unique is the different degrees of experience they bring to performing the types of movements that are used in therapy.

There is a tendency in the literature on disorders of movement to not only fail to differentiate movement tasks, but also to group disorders into homogenous classifications. [41] For example, there are three genetically different forms of Down syndrome that are seldom alluded to in the literature on motor control. In the case of cerebral palsy, there are several types and numerous associated deficits, all of which have been poorly controlled for (or alternatively represented in insufficient numbers to allow for separate analyses) and poorly analyzed in the few studies on physical therapy efficacy that have been conducted. [42] Few assumptions, therefore, should be made about whether a deficit exists in any mechanism in any particular individual before he or she has been assessed over a wide variety of movement experiences.

Implications for

Physical Therapy

In the previous sections, it has been suggested that disorders of movement can be characterized by abnormalities in how muscles are activated. This section will discuss how extensive practice over a wide variety of movement tasks, exercise and strength training, and principles of transfer of training can be applied to improving the quality of movement in individuals with disorders of movement. These implications are either derived from or are congruent with the dual-strategy hypothesis. However, these implications are not meant to be considered in isolation. They are simply meant as another factor to consider when designing treatment programs. Consideration will also be given to situations in which the most expedient therapeutic approach might be determined functionally, because there are many other factors to consider when comparing the design of a treatment program for an elderly individual who has Parkinson's disease or is suffering from a cardiovascular accident with the design of a treatment program for a young individual with Down syndrome or cerebral palsy. These factors include economic, social, and developmental considerations.

In addition, even when motor control theory can identify causal links between movement impairment and resultant consequences, it does not address the issues that arise from indirect effects such as cardiopulmonary problems. Thus, the optimal way in which to use motor control theory is in conjunction with a model for multisystem evaluation. [43] The scheduling of treatments and the provision of feedback between and within treatment sessions are also crucial for the optimal implementation of physical therapy, and these aspects of treatment have recently been reviewed by Winstein. [44]

Practice

One of the very few agreed upon observations in the motor learning literature is that appropriate practice leads to improved motor performance. The implication for physical therapy is that patients should be encouraged wherever possible to practice making movements. As obvious as this may seem, it is not always recognized how many movements are required to obtain changes in motor performance or for how long such improvements can take place. [45] In a study of healthy individuals, Corcos et al [46] showed that performance enhancement (ad defined by the ability to move more quickly and more accurately) can take place over at least 1,400 repetitions. The causes for this improvement are primarily related to the ability to increase the intensity of activation to motoneuron pools.

In the study of mental retardation, many studies have shown that extended practice can lead to great improvements in performance such that there is little difference between the practiced group and the healthy control group. The point to be stressed in this article is that motor skills should be chosen that encourage patients to practice aspects of the movement in which they are deficient. For example, in the case of Down syndrome and Parkinson's disease, the inability to modulate the intensity of excitation to the motoneuron pools seems a potential problem. As such, the same movement task should be practiced at a wide variety of speeds.

Strength Training and Exercise

One of the points to emerge from the dual-strategy hypothesis of motor control is that substantial forces are involved in the generation of movements. This fact is demonstrated in Figure 4, in which the peak inertial torque is plotted for a healthy subject. The values of peak inertial torque are a large percentage of the person's maximal voluntary contraction, which is shown by the arrows on the inertial torque plot. Many disorders of movement are characterized by weakness that some investigators believe is related to the diminished functional capacity of the patient (for example, see article by Bohannon and accompanying commentary by Rothstein et al). [47] Strength training and exercise would therefore seem to have the potential to play an important role in facilitating improved movement capacity in disorders of movement. [48] Evidence in support of using strength training can also be derived from the findings of Colebatch et al, [49] who have shown that the strength of the elbow flexors was relatively more weakened than that of the extensors on the hemiparetic side of the individuals they studied. This finding was true for hemiparetic patients suffering from strokes as well as those with cerrebral tumors.

The following factors, however, should be considered before using exercise and strength training. The therapist should ensure that movement restraint is not caused by hyper-excitable reflexes, for example as in the date in Figure 3. If it is, any form of exercise that induces activation of the antagonist muscles could be counterproductive. The following caveat should also always be kept in mind. As pointed out by Hashimoto and Paty in discussing multiple sclerosis

Physiotherapy has an important part to play. The problem is that it can be counterproductive if pushed too hard. In a patient with limited energy resources, it does not make sense to have them squander that energy on exercises when it could be better used doing work or some other activity that is important in maintaining self-esteem. On the other hand, it is very important for a patient's self-esteem to continue to push and to see results from that striving. [50](pp579-580)

Transfer of Training

Any program of movement therapy should work on the premise that a wide variety of different movements should be practiced, because what is learned in performing one movement will not necessarily have a major effect in learning another movement. For example, Gottlieb et al [51] had healthy individuals practice movements at one distance and then generate movements to three distances including the practiced distance. Movements to the practiced distance were faster and more accurate than movements to the nonpracticed distance, suggesting only partial transfer of what is learned. These findings are in general agreement with the views of Schmidt and Young, [52] who also argue that the amount of transfer witnessed in motor tasks is low.

Theory Versus Function

The issue of how best to determine physical therapy intervention principles is far from resolved. Craik and Oatis [53] provide a detailed discussion of the appropriate times to use normal gait as a means for understanding "gait deviations" and also when to use standards derived from patient. Many of the points they address apply to determining the appropriate movement pattern desired for all types of movement, and not just gait. So far in this article, it has been argued that there are certain situations in which physical therapy intervention should be based on theories of motor control derived primarily from studies on healthy individuals. However, many times this may not be the most suitable approach. One such situation is when the disorder of movement is primarily involuntary. For example, the predominant disabling feature of athetoid cerebral palsy is excessive involuntary movements. As such, the best therapeutic approach would seem to be to identify key functional tasks that a person needs to perform and experiment with ways in which to perform such tasks so that there is minimum involvement of extraneous movements. This suggests that individuals with disorders of movement need to be observed closely and that they need different types of treatment, depending on whether the primary cause of their disability stems from the voluntary or the involuntary aspect of the disorder.

It is also important to observe whether movement impairment is related to extraneous muscle activity or whether impaired movement and extraneous activity are both signs of the underlying movement pathology. For example, some studies have shown casual relations between spasticity and movement impairment, [15,54 whereas other have not.[55] Equally, some studies have shown that treatment alleviates spasticity and improves movement, [30] whereas others have not. [56] The reason for these different findings may be that insufficient care has been taken to document the primary cause of spacticity, whether more than one mechanism may be responsible or whether the spasticity is related to the movement deficit. So, for example, McLellan et al [57] have shown that the antispastic medication tizanidine did not abolish abnormal EMG activity for one of their spastic patients. This finding led them to conclude that the abnormal patterns of muscle activation were centrally determined. Similarly, Nashner [58] has argued that spasticity is secondary to a functional loss in central and spinal programs. In such cases, treatments based on the premise of developing "normal" movements should not necessarily be expected to work because the normal mechanisms are absent. However, in cases in which the negative signs of spasticity impede movement and can be suppressed by medication or therapy based on principles of inhibition, physical therapy should be very effective and lead to even greater functional improvement by enabling more productive work on factors such as strength, endurance, and range of motion that are predicated on motor control theory. In this situation, normal mechanisms will now be able to produce coordinated movement.

Conclusion

When a person performs a task, an analysis of a wide variety of motor disorders suggests that they share many characteristics. Most disorders are characterized by (1) slow reaction times, (2) slow movement times, (3) increased variability in performance, and (4) movement trajectories with discontinuities. In addition, many disorders of movement can be characterized by a rate limitation on their ability to produce force or generate movements. ("Rate limitation" refers to the speed at which a person can generate a force or make a movement.)

This article has suggested that the dual-strategy hypothesis can be used to identify particular causes of some of these movements deficiencies. Deficiencies can result from a failure to appropriately modulate the intensity, duration, or latency of muscular activation. The therapist, therefore, needs to carefully observe how movement tasks are performed, because this performance can give insight into the underlying disorder. For example, subtle changes in instructional emphasis between speed and accuracy can reveal whether a patient has the ability to modulate the speed and precision with which movements are madE. Treatment programs can then be selected that specifically address the particular deficiency. One particular treatment principle would be to identify the range of speeds over which patients can make movements. This information can most easily be obtained by varying movement distance. Then the therapist should identify the degree to which patients can move at different speeds within that range. Once it has been established that the person cannot move quickly and cannot change speed intentionally, the next step is to identify whether the cause of this inability is limited input to the agonist muscle, excessive input to the antagonist muscle, or both. (Identification of the cause of the movement restrain is difficult without the use of electromyography.) Structured practice of different types of movements, the use of exercise programs, biofeedback and motor copy training, [14] and appropriate medication can than be chosen to increase agonist activity or to decrease antagonist and unwanted synergistic activity.

References

[1] Gottlieb GL, Corcos DM, Agarwal GC. Strategies for the control of single mechanical degree of freedom voluntary movements. Behav Brain Sci. 1989;12:189-210.

[2] Gottlieb GL, Corcos DM, Agarwal GC. Organizing principles for single joint movements, I: a speed-insensitive strategy. J Neurophysiol. 1989;62:342-357.

[3] Corcos DM, Gottlieb GL, Agarwal GC. Organizing principles for single joint movements, II: a speed-sensitive strategy. J Neurophysiol. 1989;62:358-368.

[4] Feldman AG. Once more on the equilibrium-point hypothesis ([lambda] model) for motor control. Journal of Motor Behavior. 1986;18:17-54.

[5] Gottlieb GL, Corcos DM, Latash ML, Agarwal GC. Principles underlying single joint movement strategies. In: Winters J, Woo S, eds. Multiple Muscle systems: Biomechanics and Movement Organization. New York, NY: Springer-Verlag New York Inc; 1990;236-250.

[6] Hallet M, Shahani BT, Young RR. EMG analysis of stereotyped voluntary movements in man. J Neurol Neurosurg Psychiatry. 1975;38:1154-1162.

[7] Angel RW. Electromyography during voluntary movement: the two-burst pattern. ELectroencepalogr Clin Neurol. 1974;36:493-498.

[8] Hannaford B, Stark L. Roles of the elements of the triphasic control signal. Exp Neurol. 1985;90:619-634.

[9] Berardelli A, Rothwell JC, Day BL, et al. Duration of the first agonist burst in ballistic arm movements. Brain Res. 1984;304:183-187.

[10] Brown SH, Cooke JD. Initial agonist burst duration depends on movement amplitude. Exp Brain Res. 1984;55:523-527.

[11] Kelso JAS. The process approach to understanding human behavior: and introduction. In: Kelso JAS, ed. Human Motor Behavior: An Introduction. Hillsdale, NJ: Lawrence Erlbaum Associates Inc; 1982:3-19.

[12] Corcos DM, Gottlieb GL, Agarwal GC. Does constraining movements constrain the development of movement theories? Behav Brain Sci. 1989;12:237-246.

[13] Campbell SK. Assessment of the child with CNS dysfunction. In: Rothsein JM, ed. Measurement in Physical Therapy. New York, NY: Churchill Livingstone Inc; 1985:207-228.

[14] Wolf SL, LeCraw DE, Barton LA. Comparison of motor copy and targeted biofeedback training techniques for restitution of upper extremity function among patients with neurologic disorders. Phys Ther. 1989;69:719-735.

[15] Corcos DM, Gottlieb GL, Penn RD, et al. Movement deficits caused by hyperexcitable stretch reflexes in spastic humans. Brain. 1986;109:1043-1058.

[16] Gottlieb GL, Corcos DM, Agarwal GC, Latash ML. Organizaing principles for single joint movements, III: the speed-insensitive strategy as default. J Neurophysiol. 1990;63:625-636.

[17] Wadman WJ, Denier van der Gon JJ, Geuze RH, Mol CR. Control of fast goal-directed arm movements. Journal of human Movement Studies. 1979;5:3-17.

[18] Lee WA, Michaels CF, Pai Y. The organization of torque and EMG activity during bilateral handle pulls by standing humans. Exp Brain Res. In press.

[19] Copland C. Relative Force Invariance During the Stance Phase of Human Walking at Three Different Speeds. Carbondale, Ill: Southern Illinois University; 1989. Master's thesis.

[20] Fischman MG, Copland C, De Vita P. Relative force invariance during the stance phase of human walking at three different speeds. Presented at the annual conference of the North American Society for the Physiology of Sport and Physical Activity; May 17-20, 1990; Houston, Tex.

[21] Freund HJ. The pathophysiology of central paresis. In: Struppler A, Weindl A, eds. Electromyography and Evoked Potentials. Berlin, Federal Republic of Germany: Springler-Verlag; 1985:19-20.

[22] Rarick GL, Dobbins DA, Broadhead GC. The Motor Domain and Its Correlates in Educationally Handicapped Children. Englewood Cliffs, NJ: Prentice Hall; 1976.

[23] Morris AF, Vaughan SE, Vaccaro P. Measurements of neuromuscular tone and strength in Down's syndrome children. J Ment Defic Res. 1982;26:41-46.

[24] Cole KJ, Abbs JH, Turner GS. Deficits in the production of grip force in Down Syndrome. Dev Med Child Neurol. 1988;30:752-758.

[25] Hallett M, Khoshbin S. A physiological mechanism of bradykinesia. Brain. 1980;103:301-314.

[26] Stelmach GE, Worringham CJ. The preparation and production of isometric force in Parkinson's disease. Acta Psychologica. 1988;26:93-103.

[27] Wing AM. A comparison of the rate of pinch grip force increases and decreases in Parkinsonian bradykinesia. Acta Psychologica. 1988;26:479-482.

[28] Berardelli A, Accornero N, Hallet M, et al. EMG burst duration during fast arm movements in patients with the upper motor neurone syndrome. In: Delwaide PJ, Young RR, eds. Restorative Neurology. Amsterdam, the Netherlands: Elsevier Science Publishers BV; 1985:77-82.

[29] Penn RD, Kroin JS. Continuous intrathecal baclofen for severe spasticity. Lancet. 1985;2:125-127.

[30] Penn RD, Savoy SM, Corcos DM, et al. Intrathecal baclofen for severe spinal spasticity. N Engl J Med. 1989;320:1517-1521.

[31] Ochs G, Struppler A, Meyerson BA, et al. Intrathecal baclofen for long-term treatment of spasticity: a multi-centre study. J Neurol Neurosurg Psychiatry. 1989;52:933-939.

[32] Latash ML, Penn RD, Corcos DM, Gottlieb GL. Short-term effects of intrathecal baclofen in spasticity. Exp Neurol. 1989;103:165-172.

[33] Latash ML, Penn RD, Corcos DM, et al. Effects of intrathecal baclofen on voluntary motor control in spastic paresis. J Neurosurg. 1990;72:388-392.

[34] Parke B, Penn RD, Savoy SM, Corcos DM. Functional outcome after delivery of intrathecal baclofen. Arch Phys Med Rehabil. 1989;70:30-32.

[35] van der Kamp W, Berardelli A, Rothwell JC, et al. Rapid elbow movements in patients with torsion dystonia. J Neurol Neurosurg Psychiatry. 1989;52:1043-1049.

[36] Hallett M, Alvarez N. Attempted rapid elbow flexion movements in patients with athetosis. J Neurol Neurosurg Psychiatry. 1983;46:745-750.

[37] Berardelli A, Dick JPR, Rothwell JC, et al. Scaling of the size of the first agonist EMG burst during rapid wrist movements in patients with Parkinson's disease. J Neurol Neurosurg Psychiatry. 1986;49:1273-1279.

[38] Thompson PD, Berardelli A, Rothwell JC, et al. The coexistence of bradykinesia and chorea in Huntington's disease and its implications for theories of the basal ganglia and control of movement. Brain. 1988;111:233-244.

[39] Hallett M. Ballistic elbow flexion movements in patients with amyotrophic lateral sclerosis. J Neurol Neurosurg Psychiatry. 1979;42:232-237.

[40] Fagioli S, Berardelli A, Hallett M, et al. The first agonist and antagonist burst in patients with an upper motor neuron syndrome. Movement Disorders. 1988;3:126-132.

[41] Newell KM. Down's syndrome and motor control: comments and notes. Presented at the first conference of motor control in Down syndrome; 1989; Chicago, Ill.

[42] Tirosh E, Rabino S. Physiotherapy for children with cerebral palsy. Am J Dis Child. 1989;143:552-555.

[43] Schenkman M, Butler RB. A model for multisystem evaluation, interpretation, and treatment of individuals with neurologic dysfunction. Phys Ther. 1989;69:538-547.

[44] Winstein CJ. Motor learning considerations in stroke rehabilitation. In: Duncan PW, Badke MD, eds. Stroke Rehabilitation: The Recovery of Motor Control. Chicago, Ill: Year Book Medical Publishers Inc; 1987:109-134.

[45] Kottke FJ, Halpern D, Easton JKM, et al. The training of coordination. Arch Phys Med Rehabil. 1978;59:567-572.

[46] Corcos DM, Gottlieb GL, Jaric S, et al. Organizing principles underlying motor skill acquisition. In: Winters J, woo S, eds. Multiple Muscle Systems: Biomechanics and Movement Organization. New York, NY: Springer-Verlag New York Inc; 1990:251-267.

[47] Bohannon RW. Is the measurement of muscle strength appropriate in patients with brain lesions? A special communication. Phys Ther. 1989;69:225-236.

[48] Campbell SK. Application of Exercise Principles to Treatment of Cerebral Palsy. Forum Medicum/American Physical Therapy Association Self-Study Course. Berryville, VA; 1988.

[49] Colebatch JG, Gandevia SC, Spira PJ, et al. Voluntary muscle strenth in hemiparesis: distribution of weakness at the elbow. J Neurol Neurosurg Psychiatry. 1986;49:1019-1024.

[50] Hashimoto SA, Paty DW. Multiple sclerosis. Disease-A-Month. 1986;32:523-589.

[51] Gottlieb GL, Corcos DM, Jaric S, Agarwal GC. Practice improves even the simplest movements. Exp Brain Res. 1988;73:436-440.

[52] Schmidt RA, Young De. Transfer of movement control in motor skill learning. In: Cormier SM, Hagman JD, eds. Transfer of Learning. Orlando, Fla: Academic Press Inc; 1987:47-79.

[53] Craik RL, Oatis CA. Gait assessment in the clinic: issues and approaches. In: Rothstein JM, ed. Measurement in Physical Therapy. New York, NY: Churchill Livingstone Inc; 1985:169-205.

[54] Mizrahi EM, Angel RW. Impairment of voluntary movement by spasticity. Ann Neurol. 1979;5:594-595.

[55] Sahrmann SA, Norton BJ. The relationship of voluntary movement to spasticity in the upper motor syndrome. Ann Neurol. 1977;2:460-465.

[56] Landau WM, Spasticity: What is it? What is it not? In: Feldman RG, Young RR, Koella WP, eds. Spasticity: Disordered Motor Control. Miami, Fla: Symposia Specialists Inc; 1980:17-24.

[57] McLellan DL, Hassan N, Hodgson JA. Tracking tasks in the assessment of spasticity. in: Delwaide PJ, young RR, eds. Restorative neurology. Amsterdam, the Netherlands: Elsevier Science Publishers BV; 1985:131-139.

[58] Nashner LM, A functional approach to understanding spasticity. In: Struppler a, Weindl A, eds. electromyography and Evoked Potentials. Berlin, Federal Republic of Germany: Springer-Verlag; 1985:22-29.

D Corcos, PhD, is Assistant Professor, College of Kinesiology (M/C-194), University of Illinois at Chicago, PO Box 4348, Chicago, IL 60680 (USA).

This work was partially supported by NIH FIRST Grant R29 NS23593 and the Down Syndrome Research Fund.