A primate model for studying focal dystonia and repetitive strain injury: effects on the primary somatosensory cortex.

Key Words: Area 3b, Cerebral cortex, Focal dystonia, Hand, Monkey, Repetitive motion injury, Repetitive strain injury, Somatosensory cortex.

Workers doing automated tasks under highly demanding conditions can develop discomfort in their arms, wrists, and hands. Repetitive strain injuries (RSIs), repetitive motion injuries (RMIs), and cumulative trauma disorders (CTDs) can result from excessive and repetitive end-range motions, excessive force, stretching, and awkward postures,[1-4] which can lead to chronic soft tissue inflammation, scarring, pain, fatigue, muscle spasm, and postural imbalance. Neural tissues at the cervical spine, the carpal tunnel, the cubital tunnel, or the thoracic outlet can also be compressed as a result of the swelling associated with the biomechanical microtrauma.[4] Reflex sympathetic dystrophy and occupational hand cramps (focal dystonia of the hand) can be later complications.[5-14]

Ergonomic modifications of workstations, instrumental and tool redesign, prework and postwork exercises, and the use of wrist splints have not reduced the incidence of RSIs.[15,16] The recalcitrant nature of remediation, particularly of RSIs and occupational hand cramps, suggests that there may be other undiagnosed problems interfering with recovery.

Area 3b of the primary sensory cortex contains the most highly organized representations of skin afferents in the cortex. The representations of each digit are sharply segregated and precisely differentiated.[17-19] The hand representation in area 3b is frequently targeted for studies in neuroplasticity because electrophysiological techniques exist to carefully map the area and each cortical penetration can be expected to have a precise, small, cutaneous receptive field restricted to only very limited surfaces on a single digit.[20] The preciseness of the representation of the hand in area 3b provides the sensitivity needed to guide fine-motor movements of the hand. In addition, changes in area 3b can be considered to be representative of changes in other areas of the cortex (eg, areas 3a, 4, and 1).

The tasks that lead to RSIs actually simulate the behavioral conditions that have been shown to remodel the primary somatosensory cortical areas in adult monkeys: attended, task-specific performance accuracy and repetition.[17-20] Primate studies indicate that the primary somatosensory cortical fields representing cutaneous (area 3b) and kinesthetic (area 3a and 1) inputs from the forelimb and the motor cortical fields can be remodeled by repetitive sensory inputs generated in a behavioral conditioning task.[21-23] Carefully controlled input conditions result in progressively more refined and more differentiated cortical representations of skin, muscle, joint afferents, and motor movements.[17-19,21-30] Degraded representations measured by changes in the size, distribution, and overlap of the receptive fields may occur in circumstances in which afferent activities are (1) highly repetitive, (2) spatially stereotyped, (3) synchronously engaging normally differentiated sensory feedback inputs, and (4) delivered in an attended behavior. [17-19,22-25,31,32]

Varied daily behaviors can be expected to delay the development of repetition-induced functional changes, even for the most repetitive tasks. Alternatively, the risk of an RSI could be increased when a task requires co-contractions of the agonists and antagonists (eg, writing, hammering) or nearly simultaneous contractions of adjacent digits (eg, co-stimulation of skin and muscle afferents within the 100-millisecond cortical integration time, such as that timed in a piano trill). [17-19,29,30,33,34]

In a recent primate study,[35] an attended, passive, repetitive hand-opening and hand-closing paradigm was designed to simulate the conditions surrounding RSI of the upper extremity. Two monkeys were trained to place their hands on a handpiece that rapidly and nearly simultaneously opened all adjacent digits (within 16 milliseconds) and then closed the digits from one to seven times per trial for 300 to 400 trials per day. The monkeys trained approximately 1 hour per day, 5 to 6 days per week. After 12 to 25 weeks, both monkeys developed signs that were consistent with RSI or dystonia (deterioration in performance accuracy, speed, and efficiency). Electrophysiological mapping of the primary sensory cortex revealed (1) unusually large receptive fields, (2) a loss of the normal separation of receptive fields within digit segments and between digits, and (3) multiple receptive fields extending across whole digits, located on adjacent digits, or located on both the glabrous and dorsal surfaces. In striking contrast to controls, these overlaps of single and multiple receptive fields were recorded even when there were large separations between cortical penetration sites. The findings from this study suggest that RSI is not simply a peripheral, biomechanical problem, but may have serious, central neural consequences.

The purpose of our study was to evaluate the behaviorally induced cortical plasticity effects of RSI and focal dystonia symptomatology in two owl monkeys, with one monkey performing active, repetitive hand opening and closing using a highly articulated hand-squeezing strategy and with the other monkey performing this task using a proximal arm-pulling strategy. Our study attempted to enhance previous RSI research, which focused on heavily articulated, rapid, passive opening and closing of the hand. We asked the following questions: (1) Would there be a measurable change in motor control after training? (2) Would the cortical "maps" of the skin of the trained hand be degraded after training compared with normal control maps or the map representing the untrained hand? (3) Would sensory and motor findings differ under the training conditions of a hand-squeezing strategy versus an arm-pulling strategy? and (4) Would there be measurable changes in area 3b of the contralateral primary sensory cortex representing the untrained hand?

Method

Subjects

Two adult female owl monkeys (Aotus nancimae) (OM 177, OM 410) served as subjects. Based on their weights and a full complement of adult dentition, they were judged to be young adults.[36] Both animals were neurologically intact and had not previously been used for experiments. Both monkeys had a strong right-hand preference on the pretraining food-retrieval task and the hand-training task. The hand representation of the untrained side of each trained monkey and those of a reference group of six untrained owl monkeys were used as controls in this study.

Behavioral Training

Both monkeys engaged in the behavioral training task 5 to 6 days per week over 20 weeks. The goal of the behavioral training was to create a paradigm of active, repetitive opening and closing of the hand. The monkeys were trained in a cage mounted within a soundisolated test chamber. A video system external to the cage was used to monitor the animals' behavior during training. A short cylinder mounted on the cage front guided the monkeys' reach to a handpiece that was molded to fit each monkey's vertically oriented hand. A pellet feeder was attached to a side wall of the cage. Each monkey was initially trained to reach through a cylindrical opening for slices of fruit. The monkey was then trained to reach through a similar opening to grasp a handpiece formed from two half cylinders (Fig. 1). The handpiece was driven by a spring-loaded solenoid to provide a known force against which the monkey had to squeeze or pull (80 g over a distance of 7 mm) and to provide a safe mechanism to automatically reopen the handpiece when the grip force was released. The behavioral apparatus was controlled by LabVIEW [R] virtual instruments software.(*)

[Figure 1 ILLUSTRATION OMITTED]

The monkeys initiated their tasks by placing their fingers and thumb on the handpiece. When the handpiece was successfully squeezed to 80 g, the handpiece vibrated at 250 Hz. After 1.5 seconds, the vibration stopped and a light-emitting diode (LED) signaled the completion of the holding time. The monkeys then had to complete their tasks by maintaining the closure of the handpiece and sliding it toward their body (guided by the presence of an LED). The monkeys had to maintain all fingers and the thumb in contact with the handpiece during the entire opening and closing cycle. The feeder delivered a pellet each time the monkeys successfully performed their tasks. Both monkeys performed approximately 300 to 400 trials per day (squeezing the handpiece) during a 1- to 1.5-hour-long daily training period. To achieve this target goal, OM 410 was engaged in two training sessions per day. After practice trials for approximately 3 weeks in the home cages and then approximately 4 weeks of "in-cage" training, the monkeys reached a training accuracy of 80% to 90%. Both monkeys began to perform this task using an articulated finger-squeezing strategy.

Motor Performance

Motor performance was documented at the beginning and end of the study. The efficiency and accuracy of task performance were measured (number of trials per minute and percentage of the trials completed correctly). Motor performance was also measured in terms of food-retrieval efficiency (food tray-retrieval task and feeding tube-retrieval task).

Task performance accuracy was recorded each day based on the computer analysis of rewarded trials supplemented by analysis of speed based on 5-minute videotape segments. Signs of tremors, pain, and difficulty opening and closing the hand were documented from the daily observation of the videotape records. Although focal dystonia of the hand is very specific to a target task, there is commonly some carryover of motor dysfunction to similar tasks. Thus, to complete the measurement of motor performance, food-retrieval efficency was documented. In the food tray task, the monkeys retrieved nine pieces of fruit from small individualized wells (2-16 mm in size). Each trial was rated for efficiency using an ordinal scale of -- 1 to 5 (Table). The average performance was determined across all trials. The feeding tube task consisted of nine trials with the monkeys reaching through a feeding tube to retrieve small food items (presented high, low, to the side, or in the middle of the tube), with efficiency rated on the same ordinal scale used in the food tray task and with the average performance determined across the trials.

Table.
Food-Retrieval Skills Rating Scale(a)

Grade   Definition

 5      Successful retrieval an the first attempt;
        movement smooth and efficient
 4      Successful retrieval an the first attempt;
        movement awkward and inefficient
 3      Successful retrieval an the second attempt;
        movement smooth and efficient
 2      Successful retrieval an the second attempt;
        movement awkward and inefficient
 1      Successful retrieval on the third attempt
 0      Successful retrieval on the fourth or fifth attempt
-1      Could not retrieve the food item after five
        attempts

(a) The food-retrieval scale was designed to measure the efficiency of fine-motor hand skills in two nontarget tasks: (1) taking from a tray (food tray retrieval) and (2) taking food items through tube (feeding tube retrieval).

To evaluate the reliability of the two food-retrieval scales, four independent observers who were trained in hand movement kinesiology independently viewed the videotapes, which were played in slow motion. All reviewers were blinded to the monkeys' identities and status of training. The average efficiency performance was determined for each food-retrieval task. Using the intraclass correlation coefficient (model 3), interrater reliability was .97 for the mean item score on the food tray task and .86 for the average item score on the feeding tube task. Intrarater reliability was .91 for both tasks (24 paired comparisons for each task).

Once the desired performance behavior had been reached (80%-90% accuracy), training continued until the onset of simulated occupational hand cramps (approximately 20 weeks). This condition was measured by (1) a drop in performance accuracy to 50%, (2) a 30% decrease in the rate of performance (average number of task repetitions per fixed time interval), (3) an observable decline in performance on the foodretrieval tasks, and (4) marked difficulty opening or closing the hand on the handpiece. This decline in motor function did not occur for OM 410; however, after 20 weeks of training, this monkey was also scheduled for electrophysiological mapping to determine how the two monkeys differed motorically and electrophysiologically.

Surgery and Electrophysiology

The goal of the electrophysiological study was to define the neuronal responses and the cortical representations ("maps") of the hand surfaces in area 3b in the trained and untrained cerebral hemispheres of the two trained monkeys. Anesthesia was induced with a 1.5% halothane:75% nitrous oxide:25% oxygen gas mixture to allow for placement of a venous catheter. The monkeys were subsequently anesthetized with sodium pentobarbital (initial dose: 20-30 mg/kg intravenously) and maintained at a surgical level of anesthesia by intravenous supplementation. Heart rate and blood pressure were monitored. A lactated Ringer's solution with 5% dextrose was continuously infused (6 [micro] m/mg/h) and adjusted to maintain body hydration. Core temperature was maintained at 38 [degrees] C. The bladder was evacuated at regular intervals. Animals were maintained areflexic throughout these 3- to "day-long experiments. Atropine sulfate (0.1 mg/12 h) and Cefizox [R] ([dagger]) (10-20 mg/kg/12 h) were administered at the beginning of these electrophysiological experiments. The entire experiment was performed using sterile surgical procedures to maintain preparation viability for the relatively long recording times required to derive these experimental maps.

The cortical mapping procedures used in this study have been described in several earlier reports from our laboratory. [17-20,22-25,35,37-40] The surgically anesthetized monkeys were mounted in a stereotaxic apparatus. A wide craniotomy exposed the anterior parietal cortex centered on cortical area 3b. The aura was opened and reflected. A well was created with sterile cotton gauze and filled with high-viscosity silicon oil. A Sony CCD-IRIS camera ([double dagger]) mounted on a Leica WILD M650 operating microscope ([sub sections]) was used to record a computer image of the cortical surface vasculature (x 25) and the dorsal and glabrous surfaces of the trained hand. Vascular landmarks were used as a reference in placing electrode penetrations.

Parylene-coated tungsten microelectrodes([parallel]) with 1- to 3-M [Omega] impedances (at 1 kHz) were used for recording. All microelectrode penetrations were parallel to one another and were introduced roughly perpendicular to the cortical surface. Each insertion site was marked on the image of the cortical surface by reference to vascular details. Data were collected at a depth of 700 to 800 [micro] m below the cortical surface (corresponding to deep cortical layer 3 in area 3b). Recordings were amplified, band-pass filtered, and displayed by the use of conventional electrophysiological methods.

Using a manually applied fine-tipped opaque glass probe, the receptive fields were defined for each cortical penetration. The skin stimulus criterion of "just-visible skin indentation" was used for all receptive-field definitions. Studies[26-28,39] have shown that just-visible indentations are in the range of 250 to 500 [micro] m, the middle of the dynamic range of large-fiber cutaneous mechanoreceptor afferents.[40]

After the penetration site was marked on the cortex, the receptive field(s) for the sampled neuron(s) was carefully drawn to scale on the computer-stored hand image by using a mouse cursor and MAP cortical mapping software (in-house software).[41] When receptive fields overlapped onto different functional hand surfaces, cortical representational boundaries were drawn to reflect proportional overlaps. Both monkeys were mapped over a period of 72 to 96 hours. Both the contralateral trained hemisphere and the ipsilateral untrained hemisphere were mapped. Normal reference maps of area 3b of the primary sensory cortex in untrained monkeys also served as controls for this study (Fig. 2).[20] The sizes of the receptive fields were measured, and the proportions of cortical penetrations associated with multiple receptive fields were noted (those overlapping adjacent digits on the glabrous surface and those extending to the dorsal surface).

[Figure 2 ILLUSTRATION OMITTED]

Data Analysis

Reconstruction of the cortical representations and measurements of the size of the cutaneous receptive fields and the cortical area 3b were done using MAP software.[41] The dependent variables for motor performance included task performance speed and accuracy as well as food-retrieval efficiency. The dependent variables selected to measure the differences in the hand representation in area 3b were (1) the average size of the receptive fields and (2) the percentage of receptive fields with multiple components extending across segments of a single digit, extending to adjacent digits, or extending across the dorsal and glabrous surfaces.

The paired Student's t test was used to evaluate the significance of the change in task performance speed (P [is less than] .05) for each monkey. The z test for proportional differences was used to measure each monkey's decline in accuracy on task performance and decrease in efficiency for each of the food-retrieval tasks (P [is less than] .05).

The Student's t test was used to compare the differences in receptive field size for the trained animals, the control animals, and the untrained side (with P [is less than] .0167 to control for multiple testing). The chi-square test was applied to evaluate the differences between the trained and untrained sides and the untrained control animals in terms of the proportion of cortical penetrations with multiple receptive fields (those extending across segments of individual digits, those extending to adjacent digits, or those extending to both the glabrous and dorsal surfaces). Post hoc pair-wise differences were analyzed using the z test for proportional differences, with the pair-wise contrasts of interest: (1) trained hand squeezing versus untrained controls, (2) trained arm pulling versus untrained controls, (3) trained hand squeezing versus untrained controls, (4) trained arm pulling versus untrained controls, (5) trained hand squeezing versus trained arm pulling. These contrasts were tested at P [less than] .01 to control for the multiple testing.[42]

Results

Temporary Behavioral Changes

After 3 weeks of in-cage training at the grasping task, both monkeys were able to successfully and reliably squeeze the handpiece with 80% to 90% accuracy. After 4 to 5 weeks of training, both monkeys began to have difficulty completing the task. After a week of poor behavioral performance, both monkeys adjusted to and then overcame this problem by reducing the tension of their grip on the handpiece (as noted by observing the videotapes). The monkeys then resumed normal training. OM 177 continued to use a hand-squeezing strategy, whereas OM 410 spontaneously changed from a handsqueezing strategy to a more proximal elbow and shoulder pulling strategy. Rather than squeezing with the finger flexors and extensors, the hand was functionally positioned on the handpiece, then the monkey pulled back with the shoulder and elbow muscles. The pulling was of sufficient force to both close the handpiece and bring the handpiece platform to contact the LED to complete the required task. This monkey also refused to train for more than 30 minutes in any given training session and had to be trained twice a day to achieve 300 to 400 daily trials.

Final Measurements of Motor Performance

OM 177 (using an articulated hand-squeezing strategy) began to fail more than 50% of the trials after 5 months of training, meeting the target training goal for scheduling the electrophysiological mapping studies. OM 177 also showed a 60% decline in the speed of task performance (P [is less than or equal to] .001). Despite the difficulty performing the task, OM 177 continued to work for 1 to 2 hours a day. There were no indications that there was any pain. OM 410 performed initially at a slower speed compared with OM 177. Over the training period, however, there was no decline in speed (Fig. 3).

[Figure 3 ILLUSTRATION OMITTED]

Simultaneously with a dramatic deterioration in task performance, OM 177 also exhibited deterioration in food-retrieval performance. Based on 18 trials scored by four observers (72 measurements), the food-retrieval scores decreased from 4.82 (SD = 0.38) to 3.42 (SD = 1.5) (P [is less than or equal to] .0001) on the food tray task and from 4.79 (SD = 0.41) to 2.83 (SD = 0.73) (P [is less than or equal to] .0001) on the feeding tube task (Fig. 4).

[Figure 4 ILLUSTRATION OMITTED]

OM 410 (using an arm-pulling strategy) trained for 5 months and still did not show any signs of motor difficulty,, with a performance accuracy maintained at 80% (Fig. 3). This monkey also continued to perform the food-retrieval tasks without deterioration (Fig. 4), receiving scores of 4.94 (SD = 0.25) and 4.75 (SD = 0.77) at the beginning and end of training, respectively, on the food tray task and scores of 4.50 (SD = 0.67) and 4.41 (SD = 1.78) at the beginning and end of training, respectively, on the feeding tube task. The monkey was scheduled for mapping to determine whether there had been any changes in the representation of the hand on the somatosensory cortex, even though there was no measurable deterioration of motor performance.

In summary, OM 177 met all three criteria defined for occupational hand cramps: (1) a decline in the accuracy of task performance, (2) a decrease in speed of performance, and (3) a decline in efficiency in food retrieval. There were, however, no signs of acute inflammation or pain. OM 410 did not meet any of these criteria for occupational hand cramps.

Degradation of Hand Representations in Cortical Area 3b

The representation of area 3b of the primary sensory cortex was substantially degraded in OM 177 (hand squeezing) and mildly degraded in OM 410 (arm pulling). This degradation was marked by clear abnormalities in the topographic representations of the hand in this normally highly ordered cortical zone.

Cortical representations of the skin were dedifferentiated. In normal owl monkeys, digital receptive field areas generally range from 1 to 15 [mm.sup.2], with 90% of receptive field areas falling between 2 and 9 [mm.sup.2].[43] The mean size of the digital receptive fields in this species is reported to be 8.0 [mm.sup.2] (SD = 3.0).[43] The mean size of the receptive fields for the monkey training with the hand-squeezing strategy (OM 177) was larger than that of the control animals (Fig. 5) on both the trained side (X = 110 [mm.sup.2], SD = 50.3; t = 19.61, P [is less than or equal to] .0001) and the untrained side (X = 40 [mm.sup.2], SD = 22.1; t = 11.97, P [is less than or equal to] .001). In OM 410, the average receptive field size was larger than that of the control animals for both the trained side (X = 42 [mm.sup.2], SD = 41.2; t = 11.88, P [is less than or equal to] .0001) and the untrained side (X = 41 [mm.sup.2], SD = 45.4; t = 10.92, P [is less than or equal to] .0001). The sizes of the receptive fields for the trained and untrained sides of OM 410 were not different, but the average size of the receptive fields of OM 177 was larger than the average size of the receptive fields of OM 410 (t = 13.58, P [is less than or equal to] .0001).

[Figure 5 ILLUSTRATION OMITTED]

There were differences in the presence of multiplesegment receptive fields between the untrained control animals, the untrained side of the trained monkeys, and the trained side of the trained monkeys using different training strategies ([chi square] = 46-36, P [is less than or equal to] .001). On the trained side of OM 177, 58% of the sampled neurons had receptive fields that extended across more than two glabrous segments of one finger (z = 5.69, P [is less than or equal to] .0001 compared with the control animals; z = 3.91, P [is less than or equal to] .001 compared with the ipsilateral untrained side) (Fig. 6). Seventeen percent of the receptive fields actually covered the whole hand (z = 5.41, P [is less than or equal to] .0001, compared with untrained control animals). The receptive fields involving the whole hand appeared to represent Pacinian-type responses. In OM 410, only 2% of the receptive fields covered all three segments of the digit (Fig. 7) and only one receptive field demonstrated a Pacinian-type response covering the entire hand. These values were not different from the values for the untrained control animals or the values for the untrained side.

[Figure 6-7 ILLUSTRATION OMITTED]

According to Karni,[20] the cortical area representing the digits in area 3b ranged in size from 3.2 to 5.1 [mm.sup.2] in nine normal owl monkeys.[20] The cortical hand zone of area 3b could not be accurately measured for OM 177 (hand-squeezing strategy) because of the abnormally large receptive fields, the large number of receptive fields covering multiple digits, and the large number of Pacinian-type responses covering the entire hand. The cortical area 3b in OM 410 (arm-pulling strategy) was similar to that of the untrained control animals (4.93 [mm.sup.2])

Breakdown in the normally separated representations between the different digits of the hand. There were differences in the number of multiple-digit receptive fields among the groups ([chi square] = 82.71, P [is less than or equal to] .0001). Neurons sampled in 54% of the penetrations in area 3b of the trained side of OM 177 had multiple-component receptive fields (Fig. 6A, C, D, E) with subfields on more than one digit (z = 12.33, P [is less than or equal to] .0001 compared with those of the control animals; z = 8.35, P [is less than or equal to] .0001 compared with those of the ipsilateral untrained side). On the trained side of OM 410 (Fig. 7), neurons sampled in 15% of the penetrations in area 3b represented the glabrous surfaces of the fingers with multiple-digit receptive fields (z = 5.14, P = .0001 compared with control animals where the receptive field was small and limited to one segment[20,34,44]; z = 0.196, P [is greater than or equal to] .05 compared with the ipsilateral untrained side). There was no difference between the trained and untrained sides. OM 177 had more multiple-digit receptive fields than did OM 410 (z = 3.63, P [is less than or equal to] .001).

The breakdown in the separate representations of the digits in the normally strictly topographically organized cortical area is illustrated in another way in Figures 8 and 9. The glabrous receptive fields were drawn from each individual "digit zone" in the area 3b hand map and then sorted by digit to show the extension of receptive fields to adjacent digits. Although the typical cortical representation of a digit is normally very specific to a small receptive field in one segment of a single digit (Fig. 2C), the representations of each digit had strong inputs from adjacent digits in both trained monkeys, more so in the monkey using the articulated handsqueezing strategy (OM 177) than in the monkey using a proximal arm-pulling strategy (OM 410).

[Figure 8-9 ILLUSTRATION OMITTED]

Breakdown of the normally segregated representations of glabrous and dorsal hairy skin. The proportions of two-surface receptive fields were different for the three groups ([chi square] = 46.36, P [is less than or equal to] .0001). On OM 177's trained side, 77 penetrations had dorsal receptive fields. Of these penetrations, 72% (Fig. 6B) included glabrous skin surfaces (z = 4.34, P [is less than or equal to] .001 compared with control animals[44,45]; z = 1.98, P [is less than] .05 compared with the untrained side). On the trained side of OM 410, only 15 of the cortical penetrations had a dorsal receptive field (Fig. 7). Of these, 20% included the glabrous skin surfaces (z = 1.55, P [is greater than or equal to] .05 compared with control animals; z = 0.75, P [is greater than] .05 compared with the untrained side). OM 177 had more multiple-component dorsal-glabrous receptive fields than did OM 410 (z = 4.56, P [is less than or equal to] .01).

Segmental organization of the representation of the digits. Due to the irregular and often slow shifts in receptive field location and overlap as a function of distance across the cortical map, the detailed topography of the map for OM 177 was difficult to reconstruct. The normal, orderly receptive field sequence[20,44] was replaced by a more complex sequence of multiple-digit receptive fields and unusually large single-digit receptive fields. Although this topography was grossly abnormal, a general digit 1 to digit 5, lateral to medial progression and a general digit tip to proximal digit segment, rostrocaudal progression were still recorded. The topography of the map for OM 410 was much more like that seen in normal, untrained monkeys.

Dedifferentiation of the hand representation in area 3b on the untrained side. Mild degradation of the hand representation on the somatosensory cortex was seen on the untrained hemispheres of both of the trained monkeys, and the dedifferentiation of the receptive fields on the untrained side was similar for both monkeys (Fig. 10). The average size of the receptive fields on the untrained side was about five times larger than that of control animals that had never been used in any behavioral training (t = 19.51, P [is less than or equal to] .0001 for OM 177; t = 13.42, P [is less than or equal to] .0001 for OM 410). There were also more multiple-digit receptive fields on the untrained side compared with control animals that had never been used in any behavioral training. For OM 177, 33% of the dorsal receptive fields extended onto the glabrous surface (z = 2.56, P [is less than] .05) and 12% of the glabrous receptive fields extended to multiple adjacent fingers (z = 3.14, P [is less than] .01). For OM 410, 17% of the dorsal receptive fields included the glabrous surface (z = 1.54, P [is greater than] .05) and 11 % of the glabrous receptive fields extended to multiple digits (z = 4.00, P [is less than or equal to] .0001). OM 177 had more multiplecomponent receptive fields than did OM 410 (z = 3.58, P [is less than or equal to] .001).

[Figure 10 ILLUSTRATION OMITTED]

Discussion

This study was designed to determine whether an attended, highly repetitive motor movement (ie, active opening and closing of a handpiece) was associated with an experience-induced degradation in the somatensory cortex. This association was confirmed. Repetitive, highly articulated hand squeezing was associated with a loss of motor control and a degradation of the hand representation on the somatosensory cortex. With an arm-pulling strategy, however, there were no observable signs of repetitive strain injury (eg, no indications of pain and no observable motor dysfunction), but there was mild (yet statistically significant) degradation of the hand representation on the somatosensory cortex.

The sizes of the receptive fields measured in this study were even larger than those reported following a complete local block of inhibitory inputs in area 3b in macaque monkeys.[43] For OM 177 (hand-squeezing strategy), stimulation of many small skin locations engaged neurons over a cortical area many times larger than normal (on the average, 14 times larger than normal). More than 50% of the receptive fields covered all of the segments of a digit and had multiple-field components on more than one digit, and 70% of the dorsal receptive fields had a glabrous extension. By comparison, OM 410 (slow arm-pulling strategy) maintained normal motor performance and demonstrated only modest, but still substantial, degradation of the hand representation in cortical area 3b. The receptive fields were five times larger than normal, with 2% of the glabrous receptive fields covering all segments of a single digit, 15% including multiple components that were located on adjacent digits, and 20% of the dorsal receptive fields having a glabrous component. These findings suggest that a repetitive, proximal shoulder and elbow-pulling strategy may decrease the risk of severe RSI of the hand, such as focal dystonia and occupational hand cramps. Although after about 5 weeks of training both monkeys had a period during which they refused to work, neither monkey had clinical signs of inflammation, redness, or swelling of the digits of either hand at the time of the electrophysiological mapping. The results of the electrophysiological mapping, however, suggest that the untrained side could also be at risk for the development of focal dystonia symptoms, because the receptive fields were larger than normal and some cortical penetrations had multiple receptive fields.

Induced changes are consistent with temporal-based inputs (ie, plasticity), which could arise from Hebb-like synaptic change mechanisms.[46] The motor and somatosensory changes that occurred in our study are consistent with the findings for a behavioral task in which monkeys were trained to place their hand on a handpiece that repetitively, but passively, opened and closed the hand nearly 1,000 times per day.[35] The present active task provided nearly simultaneous stimulation to adjacent digits and active closing, which is more representative of repetitive functional activities. In both the active and passive paradigms, neurons at many area 3b sites developed responses to multiple-digit and glabrousdorsal input combinations that are not normally recorded in this field. Interestingly, when repetitive closing of the handpiece did not involve alternating squeezing and extending of the fingers (the arm-pulling strategy of OM 410), the magnitude of the sensory degradation was reduced.

How Could the Changes in Area 3b Contribute to the Genesis of a Work-Induced Movement Disorder?

In several studies of primates,[47-55] the true primary sensory cortex (area 3b) has ventral parietal cortical fields, which in turn have major topographically ordered projections to areas 3a, 4, 1, 2, and 5 and SII. Moreover, area 3b has strong orderly projections to at least three other functional regions that provide strong direct projections into area 4: (1) the supplementary motor cortical areas,[40,51] (2) a strip of cortex in premotor area 6 that borders area 4,[45] and (3) the subcortical putamen.[56] With its central position as the true primary somatosensory cortical field, the substantial changes in the area 3b representation of the hand would also be expected to substantially influence cortical area 4 consequent to the behavioral training,[17,43,57-59] explaining the relationship of the sensory effects of the repetition to the loss of motor control.

The critical importance of somatosensory cortex inputs for motor skill learning has been discussed by Pavlides et al.[60] Sensory-guided (using tactile stimulation) motor skill learning in macaque monkeys required an intact primary sensory cortex. Hikosaka et al[61] found digital manipulation and grasping of small objects to be grossly degraded by temporary anesthesia of restricted sectors of the cortical area. Following the induction of microlesions in area 3b in monkeys trained to perform a task involving the retrieval of small objects, similar effects were recorded. Cortical microlesions were produced by bipolar coagulation of surface vessels following the derivation of detailed hand representation maps in trained owl monkeys. Those microlesions selectively destroyed only the area 3b representations of the fingertips, which the monkeys had learned to use to palpate and manipulate pellets. Pellet-retrieval performance was dramatically degraded by the microlesions; the monkeys behaved as if their digital surfaces were insensate. The monkeys had to peek into their partially opened hand on each behavioral trial to determine whether they had successfully grasped a pellet (they usually had not). After several additional weeks of behavioral training, the monkeys recovered from this striking functional deficit, resulting in newly emergent plastic changes in digit-tip representations, most strongly expressed in cortical areas 3a and 1 (C Xerri, MM Merzenich, and WM Jenkins; unpublished observations).

Work-induced sensory inputs in the appropriate form and generated in the right behavioral context, as in these monkeys, could profoundly, degrade cortical representational specificity, largely destroying the independent representation of digits and reducing the information-bearing capacity of the cortex by expanding the dimensions or overlaps of input-specific cortical "columns." If the stress is removed from the motor task (eg, arm-pulling strategy), however, then degradation of cortical representational specificity can be minimized.

In learning, the effectiveness of nearly simultaneous inputs are believed to be integrated by the operation of synaptic plasticity mechanisms. A relatively small, consistent separation of input events may result in the segregation--and not the integration--of inputs delivered into the cortex on a heavy behavioral schedule.[27,28,30,62] The most important conditions that appear to be necessary to avoid representational degradation include (1) inputs must be variable, (2) input activity patterns must not be rigidly stereotyped,[17] (3) although inputs may not require targeted attention in the work behavior, outside of work, sensory discrimination activities must be done on a regular basis to maintain the sharp segregation of the digits, and (4) repetitive motion events should be spaced at 100-millisecond intervals or more to minimize simultaneously engaging normally differentiated sensory inputs.

Relationship of Pain and Repetition to Dedifferentiation of the Hand Representation

Both monkeys in this study changed the strategy of their grip after several daylong periods of difficulty completing the hand-closing task. This change in strategy may have signaled an episode of pain, discomfort, or loss of voluntary grasp control. The monkey that exhibited the greatest area 3b representational changes was the monkey that used the heavily articulated hand-squeezing strategy and later developed the signs of occupational hand cramps. It is not clear whether pain was associated with the behavioral changes because the monkey could not express this feeling. Inputs from pain afferents could conceivably directly alter receptive field dimensions in cortical area 3b, as with cutaneous receptive fields in the anatomically subordinate dorsal horn system.[63,64] In contrast to spinal experiments, blocking of small fiber afferents has been found to generate enlargement of cortical receptive fields,[65,66] an effect that is also at variance with a positive pain-mediated spinal receptive field effect. Pain is probably not a necessary element for dramatic neuroplasticity effects. In other studies of monkeys,[17,18,27] large-scale representational changes have been recorded in the absence of pain.

Frequently in patients with RSI, the resolution of a painful tendinitis does not result in complete recovery. A nonpainful focal dystonia can emerge in the patient and can be sustained for long periods in the absence of frank pain. The dystonia also does not disappear when the repetitive movement is stopped. The dystonia itself likely reflects and possibly directly arises from the degradation in the cortical representations that have been induced by the "learned" repetitive inputs.[67,68] Experiments on cortical plasticity in monkeys have demonstrated major cortical representational changes in behaviors that never appeared to be painful to the animal.[17-19,21,27-30] Pain may contribute to the amplification of plasticity effects and possibly the genesis of work-induced movement disorders. Pain, however, is probably not a necessary component in their progressive development.

Whatever the role of nociceptive inputs, in our view, the striking degradation of these cortical maps must have neurobehavioral consequences. The monkey using the heavily articulated hand-squeezing strategy demonstrated motor behavior consequences. In nonhuman primates, however, it is difficult to confirm that a motor control problem is specifically a focal dystonia. This monkey had difficulty with motor retrieval activities as well as a deterioration in task performance. There was no simple way of assessing the tactile sensibilities or the haptic capacities of either of the monkeys. The sensibilities of OM 177, in our opinion, would certainly be expected to be altered in the face of major changes in feedback information guiding refined movement control.[69]

The Effect of Movement Strategy and the Learning Origins Hypothesis for RSI

Our learning origins hypothesis suggests that repetition leads to measurable changes in the somatosensory cortex, which are learned as a consequence of repetition. This hypothesis is supported not only by the current study but also by a previous study of RSI by the same authors.[35] Repetitive, nearly simultaneous inputs alter brain representations of cutaneous and deep afferent feedback signals. These alterations in brain representations change the corticifugal distributions and the strengths of inputs to sympathetic and pain centers, hypothetically contributing to the emergence of an inflammatory response or a movement disorder expressed in the arm. Serious degradation of the quality of sensory feedback could account for the uncontrolled co-contractions that commonly occur with occupational hand cramps. Human imaging studies have provided some additional evidence supporting this view.[70] We had a magnetic resonance image and a magnetic source image taken on a guitarist with focal hand dystonia to determine whether there was any difference in the somatosensory hand representation compared with the magnetic source images of asymptomatic subjects. The image for the patient showed a decrease in the density of the somatosensory map of digits 2 and 3, the involved dystonic digits of the right hand. This finding in a human subject would be consistent with the decrease in the size of the cortical representation of the hand and the large receptive fields in the trained monkey using the articulated hand-squeezing strategy.

In our study, a reduction in force and speed of the repetitive movements, making the forces more variable, in addition to interspersing regular activities between the trials of repetitive work, not only minimized the dedifferentiation of the hand representation on the somatosensory cortex, but apparently decreased the risk for losing motor control. The multiple receptive fields in OM 410 primarily extended to the adjacent digits, whereas the majority of the multiple-digit cortical penetrations in OM 177 extended to the dorsal surface. These findings were also true on the control sides. When comparing the maps of the untrained side in Figure 10, it appears that OM 410 had a greater proportion of multiple receptive fields than did OM 177. The receptive field overlaps, however, are only portrayed on the glabrous surface, not on both the glabrous and dorsal surfaces. OM 177 had more overlap of the multiple receptive fields across the glabrous and dorsal surfaces.

In our study, we did not determine which factor was most important in reducing the risk for sensory degradation and motor dysfunction: (1) the decreased force in finger squeezing, (2) the decreased speed of performance, (3) the more variable sensory input resulting from a gross motor movement pattern, or (4) the cage activities permitted between the training sessions. Moreover, we did not attempt to provide any information on how long a less-articulated repetitive hand and arm strategy could be continued without ultimately, severely compromising the sensory organization of the hand and degrading motor control. Our findings suggest that workers who use highly specific articulated movements (eg, rapid alternating extension and flexion of adjacent fingers on a keyboard with the wrist excessively extended) should seriously consider changing their strategy by slowing their speed, taking some regular breaks, and using a more proximal movement strategy. Although a more proximal arm-pulling strategy may not completely eliminate the dedifferentiation of the hand representation, the magnitude of the changes should be less severe and motor dysfunction may not occur.

The Effect of Training on the Untrained Hemisphere

The monkeys in our study trained only with the dominant hand. No training was done with the nondominant hand. Although there were no signs of tendinitis or focal hand dystonia, there were measurable changes in the somatosensory map of the untrained hemisphere in area 3b for both monkeys. This finding is consistent with Tempel and Perlmutter's finding of bilateral neural dysfunction in patients with focal dystonia.[71] This finding could explain why patients with RSI, including focal dystonia of the hand, often develop symptoms on both sides. Furthermore, it suggests that patients with RSI or focal dystonia should be treated bilaterally.

Related Observations in Patients With RSI

If the learning hypothesis of RSI genesis is correct, then it is conceivable that focal dystonia is a sensory problem rather than simply a motor problem.[68] Patients with tendinitis or focal hand dystonia may have a dysfunction in cortical sensory processing.[69,73] In one study of patients with RSI,[72] dyskinesthesia (poor dynamic joint position sense) was measured in patients with tendinitis, whereas dysgraphesthesia and astereognosis (problems of inaccuracy when interpreting tactile information through the skin) were measured in patients with focal hand dystonia. Errors in tactile interpretation could reflect the degradation in cortical fields in area 3b measured in studies of primates.

The behavioral paradigm used in our study should have resulted in the degradation of the cortical areas representing muscle afferent, joint afferent, and other proprioceptive feedback from the hand and forearm (eg, motor cortical areas; cortical areas 3a and 2; and other parietal cortical areas, including areas 5 and SII). Changes in those cortical zones will be the subject of later reports in this RSI-RMI-focal dystonia model experimental series.

Limitations of the Present Study

There are several important limitations of our study. First, only two monkeys were studied. Second, the monkeys adjusted to an early, emergent movement dysfunction by changing the force of hand gripping. Unfortunately, it was not possible to control the compensatory behavior that developed. Third, the compensatory strategy of arm pulling was associated with a decrease in speed and an unwillingness to train for more than 30 minutes at a time. This compensatory behavior resulted in two training sessions per day, which effectively provided a daily rest period or a period when other activities would break the intensive repetition. Fourth, the determination of focal dystonia in the monkeys was based on change in speed and accuracy of performance, deterioration in food-retrieval proficiency, and clinical observation of motor dysfunction. Cortical sensory testing and electromyographic studies during hand grasping could not be done, nor could the monkey provide us with descriptions of pain, fatigue, awkwardness, and subtle involuntary muscle contractions.

Although our study did not address treatment conditions for focal hand dystonia, the findings may have important implications for physical therapy management. Restoring normally ordered representations of feedback inputs from the tactile and proprioceptive receptors of the limb could block the progression of dedifferentiation and possibly reestablish a "safe zone" of movement operations. Furthermore, the bilateral nature of the somatosensory degradation, despite the unilateral motor symptoms, suggests that both upper extremities should be included in preventive as well as restorative therapy. Our study also provides support for teaching patients to use a general, more proximal, tension-free arm movement strategy wherever possible to replace heavy articulated finger movements.

The redifferentiation of sensory feedback information could be accomplished by implementing a series of repetitive, cognitively demanding, sensory-discriminative haptic and movement exercises designed to progressively reestablish cortical map order and differentiation. Even if the initial origin of the pain and inflammation characteristic of RSI is biomechanical, the repetitive inputs that generate it must also be degrading representations of movements and feedback signals from skin, muscles, and joints. The many hundreds to tens of thousands of hours of repetitive inputs that lead to RSI probably generate what are essentially learned changes in forebrain representational zones, which must be reversed by a period of sensory learning-based, highly attended exercises designed to redifferentiate and reorder degraded representations of movements and afferent feedback sources.

Conclusion

The findings from this study provide a new physiological foundation for the etiology of chronic RSI and focal hand dystonia. Further research is needed to determine whether the sensory degradation is reversible following sensory-discriminative treatment and, if so, what duration and intensity of sensory training are needed to restore the somatosensory representation.

Acknowledgments

We acknowledge the assistance of Lazlo Bocskai, Marshal Fong, and Marten Byl, who were responsible for the mechanical and electrical engineering required for this study, and Kathleen Moore-Dixon for her help in behavioral training.

(*) National Instrument Corp, 6504 Bridge Point Pkwy, Austin, TX 78730-5039.

([dagger]) Fujisawa USA Inc, Parkway North Center, 3 Parkway N, Deerfield, IL 60015-2548.

([double dagger]) Sony Japan Technical Instruments, 348 6th St, San Francisco, Ca 94143.

([sub sections]) Leica AG, Corporate Poststrasse (Rathaus), PO Box 1243, CH-90001 St Gallen, Switzerland.

([parallel]) Microprobe Inc, Box 87, Clarksburg, MO 20871.

References

[1] Cumulative Trauma Disorders in the Workplace: Costs, Prevention, and Progress. Washington, DC: Bureau of National Affairs Inc; 1991. International Standard Book No. 1-55871-220-8.

[2] Fry HJ. Overuse syndrome in musicians: prevention and management. Lancet. 1986;9:728-731.

[3] Fry HJ. Overuse syndromes in musicians--100 years ago: an historical review. Med J Aust. 1986;146:620-625.

[4] Millender LH, Louis DS, Simmons BP. Occupational Disorders of the Upper Extremity. New York, NY: Churchill Livingstone Inc; 1992.

[5] Cohen L, Hallett M. Hand cramps: clinical features and electromyographic patterns in focal dystonia. Neurology. 1988;38:1005-1012.

[6] Hochberg FH, Harris SU, Blartert TR. Occupational hand cramps: professional disorders of motor control. Hand Injury and Sports Performance Arts. 1990;6:427-428.

[7] Jancovic J, Shale H. Dystonia in musicians. Semin Neurol. 1989;9:131-135.

[8] Lederman R. Occupational cramp in instrumental musicians. Medical Problems of Performing Artists. 1988;3:45-51.

[9] Marsden CD, Sheehy MP. Writers cramp. Trends Neurosci. 1990;13: 148-153.

[10] Newman J, Hochberg F. Isolated painless manual incoordination in musicians. J Neurol Neurosurg Psychiatry. 1987;50:291-295.

[11] Sheehy MP, Marsden CD. Writer's cramp: a focal dystonia. Brain. 1982;105:461-480.

[12] Wilson FR. Coming to grips with occupational hand cramps. Presented at the Annual Meeting of the American Academy of Neurology; Boston, Mass; April 18, 1981.

[13] Wilson F. Acquisition and loss of skilled movement in musicians. Semin Neurol. 1989;9:146-151.

[14] Wilson F, Wagner C, Homberg V, Noth J. Interaction of biomechanical and training factors in musicians with occupational cramp/ focal dystonia. Neurology. 1991;4(suppl 1):292-297.

[15] Stock SR. Workplace ergonomic factors and the development of musculoskeletal disorders of the neck and upper limbs: a meta-analysis. Am J Ind Med. 1991;19:87-107.

[16] Salmon P, Shook CP, Lombart KG, Berenson G. Performance impairments, injuries, and stress in a sample of keyboard and other instrumentalists. Medical Problems of Performing Artists. 1995;10:140-146.

[17] Jenkins WM, Merzenich MM, Ochs M, et al. Functional reorganization of primary somatosensory cortex in adult owl monkeys after behaviorally controlled tactile stimulation. J Neurophysiol. 1990;63:82-103.

[18] Jenkins WM, Merzenich MM, Recanzone GH. Neocortical representational dynamics in adult primates: implications for neuropsychology. Neuropsychologia. 1990;28:573-584.

[19] Jenkins WM, Merzenich MM. Reorganization of neocortical representations after brain injury: a neurophysiological model of the bases of recovery from stroke. In: Seil FJ, Herbert E, Carlson BM, eds. Prog Brain Res. 1987;71(special issue):249-266.

[20] Karni A. When practice makes perfect. Lancet. 1995;345(8946):395. Letter.

[21] Merzenich MM, Nelson RJ, Kaas JH, et al. Variability in hand surface representations in areas 3b and 1 in adult owl and squirrel monkeys. J Comp Neurol. 1987;258:281-296.

[22] Recanzone GH, Jenkins WM, Hradek GT, Merzenich MM. Progressive improvement in discriminative abilities in adult owl monkeys performing a tactile frequency-discrimination task. J Neurophysiol. 1992; 67:1015-1030.

[23] Recanzone GH, Merzenich MM, Jenkins WM, et al. Topographic reorganization of the hand representation in cortical area 3b of owl monkeys trained in a frequency-discrimination task. J Neurophysiol. 1992;67:1031-1056.

[24] Recanzone GH, Merzenich MM, Jenkins WM. Frequency-discrimination training engaging a restricted skin surface results in an emergence of a cutaneous response zone in cortical area 3a. J Neurophysiol. 1992;67:1057-1070.

[25] Talbot WH, Parian-Smith I, Kornhuber HH, Mountcastle VB. The sense of flutter-vibration: comparison of the human capacity with response patterns of mechanoreceptive afferents from the monkey hand. J Neurophysiol. 1968;31:301-334.

[26] Wang X, Merzenich MM, Sameshima K, Jenkins WM. Afferent input integration and segregation in learning are input-timing dependent. Neuroscience Abstracts. 1994;20:1427. Abstract.

[27] Wang X, Merzenich MM, Sameshima K, Jenkins WM. Representational remodeling of hand surface map in adult cortex is input-timing dependent. Nature. 1995;378:71-75.

[28] Merzenich MM, Jenkins WM. Reorganization of cortical representation of the hand following alterations of skin inputs induced by nerve injury, skin island transfer, and experience. J Hand Ther. 1993;6:89-94.

[29] Merzenich MM, Recanzone GH, Jenkins WM, Grajski KA. Adaptive mechanisms in cortical network underlying cortical contributions to learning and nondeclarative memory. Cold Spring Harb Symp Quant Biol. 1990;55:873-887.

[30] Merzenich MM, deCharms C. Neural representations, experience, and change. In: Perdolfo L, Churchland P, eds. The Mind-Brain Continuum: Sensory Processes. Boston, Mass: The MIT Press; 1996:61-81.

[31] Ahissar M, Ahissar E, Bergman H, Vaadia E. Encoding of sound-source location and movement activity of single neurons and interactions between adjacent neurons in the monkey auditory cortex. J Neurophysiol. 1992;67:203-215.

[32] Weinberger NM, Javid R, Lepan B. Long-term retention of learning-induced receptive-field plasticity in the auditory cortex. Proc Natl Acad Sci U S A. 1992;90:2394-2398.

[33] Adams RD, Victor M. Principles of Neurology. New York, NY: McGrawHill Inc; 1977.

[34] Allard T, Clark S, Jenkins WM, Merzenich MM. Reorganization of somatosensry area 3b representations in adult owl monkeys following digital syndactyl. J Neurophysiol. 1991;66:1048-1058.

[35] Byl NN, Merzenich MM, Jenkins WM. A primate genesis model of focal dysonia and repetitive strain injury, I: learning-induced de-differentiation of the representation of the hand in the primary somatosensory cortex in adult monkeys. Neurology. 1996;47:508-520.

[36] Hershkovitz P. Living New World Monkeys (Platyrrhini). Chicago, Ill: University of Chicago Press; 1991.

[37] Merzenich MM, Kaas JH, Wall J, et al. Topographic reorganization of somatosensory cortical areas 3b and 1 in adult monkeys following restricted de-differentation. Neuroscience. 1983;8:1-33.

[38] Merzenich MM, Kaas JH, Wall JT, et al. Progression of changes following median nerve section in the cortical representation of the hand in areas 3b and 1 in adult owl and squirrel monkeys. Neuroscience. 1983;10:639-665.

[39] Merzenich MM, Nelson RJ, Stryker MP, et al. Somatosensory cortical map changes following digit amputation in adult monkeys. J Comp Neurol. 1984;224:591-605.

[40] Johnson KO. Reconstruction of population response to a vibratory stimulus in quickly adapting mechanoreceptive afferent fiber populations innervating glabrous skin of the monkey. J Neurophysiol. 1974;37: 67-72.

[41] Peterson BE, Merzenich MM. MAP: a Macintosh program for generating categorical maps applied to cortical mapping. J Neurosci Methods. 1995;57:133-144.

[42] Marachuilo LA, McSweeney M. Nonparametric and Distribution-Free Methods for the Social Sciences. Monterey, Calif: Brooks/Cole Publishing Co: 1977.

[43] Stryker MP, Jenkins WM, Merzenich MM. Anesthetic state does not affect the map of the hand representation within area 3b somatosensory cortex in owl monkeys. J Comp Neurol. 1987;258:297-303.

[44] Sur M, Merzenich MM, Kaas JH. Magnification of receptive field area and "hypercolumn" sizes in areas 3b and 1 of the somatosensory cortex in owl monkeys. J Neurophysiol. 1980;44:295-311.

[45] Merzenich MM, Recanzone GH, Jenkins WM, et al. Cortical representational plasticity. In: Raskic P, Singer W, eds. Neurobiology of Neocortex. New York, NY: John Wiley & Sons Inc; 1988:41-67.

[46] Hebb DO. The Organizational Behavior: A Neuropsychological Theory. New York, NY: John Wiley & Sons Inc; 1949.

[47] Ghosh S, Brinkman C, Porter R. A quantitative study of the distribution of neurons projecting to the precentral cortex in the monkey (M fascicularis). J Comp Neurol. 1987;259:424-444.

[48] Jones EG, Coulter JD, Hendry SHC. Intracortical connectivity of architectonic fields in the somatic sensory, motor, and parietal cortex of monkeys. J Comp Neurol. 1978;282:291-348.

[49] Krubitzer LA, Kaas JH. The organization and connections of somatosensory cortex in marmosets. J Neurol Sci. 1990;10:952-974.

[50] Kunzle H. Cortico-cortical efferents of primary motor and somatosensory regions of the cerebral cortex in Macaca fascicularis. Neuro-science. 1978;3:25-39.

[51] Jones EG. The nature of the afferent pathways conveying short-latency inputs to the primate motor cortex. Adv Neurol. 1983;39:263-285.

[52] Cusick CG, Steindler DA, Kaas JH. Corticocortical and collateral thalamocortical connections of postcentral somatosensory cortical areas in squirrel monkeys. Somatosens Mot Res. 1985;3:1-31.

[53] Pons TP, Kaas JH. Cortical connections of area 2 of somatosensory cortex in macaque monkeys: a correlative anatomical and electrophysiological study. J Comp Neurol. 1986;248:313-335.

[54] Huerta MF, Pons TP. Primary motor cortex receives input from area 3a in macaques. Brain Res. 1990;537:367-371.

[55] Kaas J. The functional organization of the somatosensory cortex in primates. Anat Anz. 1993; 175:509-518.

[56] Flaherty AW, Graybiel AM. Two input systems for body representations in the primate striatal matrix: experimental evidence in the squirrel monkey. J Neurol Sci. 1993;13:1120-1137.

[57] Aizawa H, Inase M, Mushiake H, et al. Reorganization of activity in the supplementary motor area associated with motor learning and functional recovery. Exp Brain Res. 1991;84:671-778.

[58] Mitz AR, Godschalk M, Wise SP. Learning-dependent neuronal activity in the premotor cortex: activity during the acquisition of conditional motor associations. J Neurol Sci. 1991;11:1855-1872.

[59] Aosaki T, Tsubokawa H, Ishida A, et al. Responses of tonically active neurons in the primate's striatum undergo systematic changes during behavioral sensorimotor conditioning. J Neurol Sci. 1994;14:3969-3984.

[60] Pavlides C, Miyashita E, Asanuma H. Projections from the sensory to the motor cortex is important in learning motor skills in the monkey. J Neurophysiol. 1993;70:733-741.

[61] Hikosaka O, Tanaka M, Sakamoto M, Iwamura Y. Deficits in manipulative behaviors induced by local injections of muscimol in the first somatosensory cortex of the conscious monkey. Brain Res. 1984; 325:375-380.

[62] Merzenich MM, Sameshima K. Cortical plasticity and memory. Curr Opin Neurobiol. 1993;3:187-196.

[63] Woolf CJ, King AE. Dynamic alterations in the cutaneous mechano-receptive fields of dorsal horn neurons in the rat spinal cord. J Neurol Sci. 1990;10:2717-2726.

[64] Dubner R, Ruda MA. Activity-dependent neuronal plasticity following tissue injury and inflammation. Trends Neurosci. 1992;15:96-103.

[65] Nussbaumer JC, Wall PD. Expansion of receptive fields in the mouse cortical barrelfield after administration of capsaicin to neonates or local application on the infraorbital nerve in adults. Brain Res. 1985;360(1-2):1-9.

[66] Calford MB, Tweedale R. C-fibers provide a source of masking inhibition to primary somatosensory cortex. Proc R Soc Lond B Biol Sci. 1991;243(1308):269-275.

[67] Elbert T, Pantev C, Weinbrach C, et al. Increased cortical representation of the fingers of the left hand in string players. Science. 1995;270:305-307.

[68] Hallett M. Is dystonia a sensory disorder? Ann Neurol. 1995;38:139-140.

[69] Byl NN, Hamati D, Wilson F, et al. The sensory consequences of repetitive strain injury in musicians: focal dystonia of the hand. Back and Musculoskeletal Rehabilitation. 1996;7:27-39.

[70] Flor H, Behle DJ, Birbaumer N. Assessment of pain-related cognition in chronic pain patients. Behav Res Ther. 1993;31:63-73.

[71] Tempel LW, Perlmutter JS. Abnormal cortical responses in patients with writer's cramp. Neurology. 1994;43:2252-2256.

[72] Byl NN, Scott P, Oakes A, et al. Sensory dysfunction associated with repetitive strain injuries of tendinitis and focal and dystonia: a comparative study. J Orthop Sports Phys Ther. 1996;25:234-244.

NN Byl, PhD, PT, is Professor and Director, Graduate Program in Physical Therapy, School of Medicine, 374 Parnassus Ave, Room 301, University of California at San Francisco, San Francisco, CA 94143-0376 (USA). Address all correspondence to Dr Byl.

MM Merzenich, PhD, is Professor, Coleman Laboratory and Keck Center for Integrative Neurosciences, and Departments of Otolaryngology and Physiology, University of California at San Francisco.

S Cheung, MD, is Assistant Professor, Coleman Laboratory and Keck Center for Integrative Neurosciences, and Department of Otolaryngology, University of California at San Francisco.

P Bedenbaugh, PhD, is Postdoctoral Fellow, Coleman Laboratory and Keck Center for Integrative Neurosciences, University of California at San Francisco.

SS Nagarajan, PhD, is Postdoctoral Fellow, Coleman Laboratory and Keck Center for Integrative Neurosciences, University of California at San Francisco.

WM Jenkins, PhD, is Adjunct Associate Professor, Coleman Laboratory and Keck Center for Integrative Neurosciences, and Departments of Otolaryngology and Physiology, University of California at San Francisco.

The study protocol was conducted in accordance with state of California, National Institutes of Health, and US Department of Agriculture guidelines for the care and use of laboratory animals and was approved by the University of California at San Francisco Animal Research Committee.

This research was supported by the University of California at San Francisco Research, Evaluation, and Allocation Committee and the Academic Senate.