Practice effects on the less-affected upper extremity after stroke

668

Practice Effects on the Less-Affected Upper Extremity After Stroke Patricia S. Pohl, PhD, Carolee J. Winstein, PhD ABSTRACT. Pohl PS, Winstein CJ. Practice effects on the less-affected upper extremity after stroke. Arch Phys Med Rehabil 1999;80:668-75. Objective: To test the hypotheses that (1) adults who have had a stroke, using the less affected upper extremity (UE), improve performance of an aiming task with practice, and (2) compared with control subjects, stroke patients show less improvement in a complex condition. Design: Movement time (MT) and kinematic data were collected over practice. Comparisons were made between the less-affected UE of stroke patients and the same hand of controls. Setting: A human performance laboratory. Participants: A matched sample of right-handed adults, 10 with unilateral stroke and 10 nondisabled controls. Intervention: Practice of an aiming task in an easy and complex condition as defined by target width and distance between two targets. Main Outcome Measures: MT, peak velocity, and temporal phases of the trajectory. Results: Adults who had experienced a stroke had persistently longer MTs than control subjects; however, all participants achieved faster MTs with practice in both conditions. The absolute amount of time in each temporal phase decreased without a change in the relative times. Peak velocity increased only in the easy condition. Conclusions: Adults with stroke damage can improve motor performance of the less-affected UE with practice. Further study is needed to see if practice effects are permanent and generalizable. o 1999 by the American Congress of Rehabilitation Medicine and the American Academy of Physical Medicine and Rehabilitation

I

NDIVIDUALS WHO SUFFER unilateral stroke are often left with bilateral upper extremity (UE) involvement. Although the UE contralateral to the side of the cerebral lesion is clearly more affected, the UE ipsilateral to the lesion is not normal. Movements of the less-affected UE in adults with unilateral stroke are weakerle3 and less accurate4 than the UEs of healthy control adults. In goal-directed aiming movements

From the Department of Biokinesiology and Physical Therapy, University of Southern California, Los Angeles, CA. Submitted for publication September 1, 1998. Accepted in revised form December 7, 1998. This work was completed in partial fulfillment of the requirements for the degree of Doctor of Philosophy for Dr. Pohl and was supported in part by a Doctoral Research Award from The Foundation for Physical Therapy (Dr. Pohl) and a grant from the California Physical Therapy Fund (Dr. Pohl). No commercial party having a direct financial interest in the results of the research supporting this article has or will confer a benefit upon the authors or upon any organization with which the authors are associated. Reprint requests to Dr. Patricia S. Pohl, Department of Physical Therapy Education and Center on Aging, University of Kansas Medical Center, 3901 Rainbow Boulevard, Kansas City, KS 66160-7601. 0 1999 by the American Congress of Rehabilitation Medicine and the American Academy of Physical Medicine and Rehabilitation 0003-9993/99/8006-5192$3.00/O

Arch

Phys

Med

Rehabil

Vol 80, June

1999

that require both speed and accuracy, adults who have experienced stroke demonstrate slower movements when using the less-affected UE than do adults without brain damage.5-7 Motor deficits in the less-affected UE may be partially explained by the fact that there is bilateral cortical hemisphere activity during unilateral UE movements in adults without brain damage. Winstein and colleagues8 performed a study using positron emission tomography (PET) on young adults as they performed a rapid, accurate aiming task with their dominant right hand in three conditions of task complexity. Task complexity was defined by the width of the targets and the distance between the targets.9 Of particular interest, there was an increase in activity in right dorsal premotor, bilateral supplementary motor areas (SMAs), and bilateral occipital areas with performance in the high-complexity condition. This finding matches those of other studies that have shown that the planning and execution of unimanual goal-directed tasks are accompanied by bilateral cortical activation.lO,” Thus, damage in one hemisphere may affect the motor control of both UEs. Practice is the most powerful variable in the acquisition of motor ~kil1.l~Practice of arm movements with dual goals of speed and accuracy result in faster movement times (MTs) without a loss of accuracy for both young adults13-16and older adults.15,16Participants in the aforementioned studies have been nondisabled adults. Can motor control of the less-affected UE of adults who have had a stroke be improved with practice? Most studies that have examined performance of the lessaffected UE have restricted measurements to relatively few trials. It is not clear if these motor control deficits are fixed, or if they can be ameliorated with practice. Thus, the first purpose of this study was to determine whether practice affects performance of the less-affected UE in adult stroke patients. If motor control of that extremity can be improved with practice, adults who have experienced stroke should be able to decrease MT with practice of an aiming task, as with nondisabled adults. A study of practice effects on the performance of aiming in nondisabled adults found that the kinematic parameters following practice are a function of the complexity of the task.16In a low-complexity condition, a decrease in MT is achieved by scaling (ie, increasing peak velocity) without changing the relative temporal structure (ie, the shape of the velocity profile). Equal amounts of MT are spent in acceleration and deceleration, and this does not change with practice. In a highcomplexity condition, movement times are decreasedby changing the relative temporal structure (ie, increase symmetry in the velocity profile with an increase in the relative time spent in acceleration) without changing the scaling (ie, peak velocity). Both young and older adults demonstrate these patterns of change with practice, suggesting that these strategies are related more to the complexity of the task than to age. If bilateral cortical activation is associated with performance in high-complexity conditions, it is reasonable to sttggest that persons with unilateral cortical damage may have difficulty improving performance in conditions of high task complexity. Therefore, the second purpose of this study was to examine practice effects on MT as a function of task complexity. If activation of both cortices is necessary for rapid and accurate

UPPER

EXTREMITY

PRACTICE

aiming in a high-complexity condition, adults who have experienced stroke may not be able to decrease MT with practice in a more complex task condition or they may do so using different strategies than controls. METHODS Participants Twenty self-declared right-hand-dominant adults participated in the study, including 10 with unilateral cerebral lesions and 10 gender- and age-matched neurologically healthy control subjects (table 1). All participants with stroke were at least 6 months from onset (43 i 30mo, range 6 to 84mo), thus it was assumed that the period of the greatest and most rapid natural recovery was past. The stroke survivors were no longer undergoing physical rehabilitation and none were participating in formal exercise programs. Unilateral brain damage was determined from brain imaging reports and medical records, or, in three caseswhere these were not available, by patient history. The cause of the stroke was not recorded. Using the clinical classification system described by Bamford and colleagues,17 we were able to determine that none of the participants with stroke damage had lesions involving the posterior circulation. Participants were excluded if they had a history of (1) a neurologic disorder, including a previous stroke, with the exception of a new unilateral stroke for the experimental group, (2) alcohol or drug abuse, (3) an acute medical problem, or (4) symptomatic dysfunction of the UE to be used in the task (ie, pain or limits in range or control of the UE that precluded moving rapidly between the two targets). All participants had normal or corrected-to-normal vision, were able to follow the instructions necessary to perform the task, and the stroke patients were free of uncompensated hemianopia that may have interfered with viewing the targets. The stroke patients were screened for hemineglect syndrome (table 1) with the Schenkenberg line bisection test.18 Before beginning the practice phase of the study, a clinical assessment of right and left UE function for each stroke patient was done using the motor UE section of the Fugl-Meyer Assessment.19 Comparisons using the Mann-Whitney test indicated that time since onset, line bisection scores, and Fugl-Meyer scores were not different between those with right hemisphere lesion and those with left hemisphere lesion. Instrumentation and Task The instrumentation16 and task7,*,r6 have been described elsewhere. Equipment for the aiming task included a hand-held stylus, a millisecond timer for trial length that emitted an audible tone at the end of the trial, and a tapping board with metal plates distinguishing two target regions and surrounding “error” areas with impulse counters for target hits, overshoots, and undershoots at each target. Task complexity was based on Fitts’ Law9 where MT increases as the Index of Difficulty (ID) increases. The ID is Table Group

Age (vrs)

Right hemisphere Left hemisphere

stroke stroke

Right hemisphere Left hemisphere

controls controls

60.4 59.4 61 .O 59.7

(9.3) (5.6) (7.3) (10.4)

AFTER

STROKE,

669

Pohl

calculated in bits and is equal to the Log2 (2D/W) where D is the distance between the targets and W is the width of the targets. The low-complexity condition was defined by two Scm-wide targets placed 37cm apart (ID = 3.21 bits). The highcomplexity condition was defined by two O&m-wide targets placed 18Scm apart (ID = 6.21 bits). Participants were seated at a table midline to the tapping board, such that aiming movements were made in the frontal plane. They were required to hold the stylus like a pencil, resting the tip of the stylus on the target to their left. Individuals who had experienced stroke performed the task with the less-affected UE. Controls performed with the same UE as the stroke patients to whom they were matched. The task required the participants to tap in a reciprocal fashion with the stylus, hitting each target alternately for 1Osec as quickly and as accurately as possible. The start of the trial (defined by the first target hit on the plate to the performer’s right), and each subsequent target hit, was signaled by lightemitting diodes (LED) in view of the camera but not the performer. Video recording was done with a Pulnix 120/60Hz video camera.aVideo analysis equipment included Peak Performance Technologiesb software and hardware interfaced with the video recorder. Procedure Participants signed an informed consent approved by the university’s Institutional Review Board. Participants were seated with the chair adjusted so that, with the arm next to the trunk and the shoulder relaxed, the forearm was approximately parallel to the tabletop with the elbow flexed. Participants were instructed to begin the trial any time after the experimenter indicated she was ready (reaction time was not measured) and to continue the tapping movements until they heard the tone which terminates the trial after 10sec. They were instructed to move as quickly as possible without missing the targets. No practice trials were permitted before data collection began. Participants performed the aiming movement under conditions of low and high task complexity in 10 trial practice blocks of each condition. There was approximately a liisec rest between trials, and a 2min rest between blocks when the condition changed. In addition, participants were encouraged to rest between trials if a rest was needed. Participants performed sequences of one practice block of the low-complexity condition and two practice blocks of the high-complexity condition until they had reached 500 target hits in any one condition, at which time they continued until they reached 500 target hits in the other condition. This practice block order was established from pilot data that indicated that at least twice the number of trials was needed in the high-complexity condition versus the low-complexity condition to achieve 500 target hits. Within any trial, the number of target misses could not be greater than 10% of the total number of passes from target to target. If this error criterion was not met, the trial was

1: Participants

Male/Female Ratio

5/O 4/l 5/o 411

Tme

Onset (mo)

Fugl-Meyer Score*

Line Bisection Score I%)+

56.2 (32.2) 29.2 (28.3) -

32.4 (21.0) 50.6 (19.7) -

1.9 (3.8) -4.9 (12.1) -

Measures are mean (SD). * Fugl-Meyer score of UE contralateral to side of the cerebral lesion; possible total of 66. + Normalized deviation; a positive number indicates a deviation to the right and a negative

Since

number

indicates

Arch

Phys

a deviation

Med

to the left.

Rehabil

Vol 80, June

1999

670

UPPER

EXTREMITY

PRACTICE

AFTER

STROKE,

Pohl

peak

1

A “’

horizontal

2

1 Time

Time (s)

Fig 1. Example of a 2sec time series indicated in (A); the temporal phases

of (A) horizontal (1) acceleration,

position, (B) horizontal velocity, (C) vertical (2) deceleration, and (3) dwell in (B).

unacceptable and the target hits in that trial were not counted toward the 500 required hits. Feedback was provided after each trial; participants were told the number of target hits they obtained, if the trial was acceptable, and the total number of target hits accumulated in that condition. Stylus movements were videotaped at a rate of 120Hz during all trials, using a split screen. The camera was placed orthogonal to the major plane of interest. A reflective marker, fixed lcm from the distal end of the stylus, was used during the video digitizing process to determine the trajectory of the movements. Room lights remained on for all filming. Total testing time, including screening tests, was between 2 and 3 hours. Data Analysis Digitized movement of the stylus provided horizontal (x) and vertical (y) coordinates of the trajectories sampled every 8.33msec. Position data were smoothed using a Butterworth low-pass digital filter with an 8Hz cutoff frequency supplied by the Peak Performance Motion Analysis System.b Velocities were derived from the position data using a five-point central difference algorithm. Movement time and key kinematic features were identified using DATA-PAC II” software from RUN Technologies. Target hits were defined by the lowest horizontal and vertical positions (fig 1). Movement time was the absolute time taken from lift-o.ff from one target to lift-off at the other target (ie, a cycle). Movement time was divided into three temporal phases: (1) acceleration time (ie, time from lift-off from a target to peak horizontal velocity), (2) deceleration time (ie, time from peak horizontal velocity to target hit), and (3) dwell time (ie, the length of time the stylus rested on a target, defined by the duration of zero vertical velocity). In addition to the absolute Arch

Phys

Med

Rehabil

Vol 80, June

1999

2 (s)

position,

and (D) vertical

velocity.

A cycle

is

duration of each of these temporal phases, the proportion of MT spent in each phase was calculated as a measure of relative time. Peak horizontal velocity was the maximum velocity attained in the major plane of movement. Dependent measures were extracted from each cycle of interest. Averages for each measure were computed across cycles in blocks of 10 for a total of 14 practice blocks (table 2). The number of unacceptable trials was counted for each participant in each condition and analyzed in a two-factor, 2 group (stroke, control) X 2 complexity (low, high) analysis of variance (ANOVA). No single model was appropriate for all dependent measures. Thus, we conducted separate three-factor, 2 group (stroke, control) X 2 complexity (low, high) X practice Table

2: Composition Practice

Block

of Practice

Blocks

for Analyses Cydes

1

I-IO

2

II-20

3 4 5 6 7

21-30 31-40 41-50

8 9

91-100 141-150 191-200 241-250

10 11

291-300 341-350

12 13 14

391-400 441-450 491-500

UPPER

EXTREMITY

PRACTICE

blocks (1-14) ANOVAs with repeated measures on the last factor for each other dependent measure. The GreenhouseGeisser adjusted degrees of freedom was used for the repeated measures factor of practice blocks. The level of significance was set at p < .05 for all analyses. Effect sizes were calculated in the form of eta squared (n2) using the SPSS statistical package. To gain insight into the effects of the side of the lesion, descriptive analyses were done on performance over practice for right-stroke and left-stroke patients. To avoid differences owing to the hand used for the task, the data from the stroke subjects were compared with that from control subjects using the same hand. RESULTS Task Complexity Performance markedly differed as a function of task complexity, with average MTs of 347msec and 900msec in the low- and high-complexity conditions, respectively. To accumulate the required 500 target hits in each condition, participants needed an average of 17 acceptable trials (approximately 30 hits per trial) in the low-complexity condition and 42 acceptable trials (approximately 12 hits per trial) in the high-complexity condition. Statistical analyses revealed that the effect of task complexity was robust, with a significant main effect of complexity for MT and each kinematic measure at p < .OOOl,with the exception of relative dwell time, which resulted in a main effect of complexity of p < .02. Participants had more unacceptable trials in the high-complexity condition. Indeed, the main effect of task complexity for the frequency of unacceptable trials was significant at p < .04. The results reported here concern the effects of group, practice block, and interactions with complexity. When main effects and interactions are present, only the higher order interaction is presented. Unacceptable Trials Each group, adults who have had stroke and control subjects, had five unacceptable trials in the low-complexity condition. In the high-complexity condition, the stroke subjects had a total of 69 unacceptable trials compared with a total of 53 for control subjects; however, this difference was not significant. Temporal Measures Across practice blocks, stroke patients had longer MTs compared with control subjects in each complexity condition (table 3). This was reliable as evidenced by a main effect of group p < .02 -f2 = .22. Stroke and lontrol participants decreased MT with practice in both conditions of task complexity (fig 2). The decrease in MT was reliable as indicated by a main effect of practice block, p < .OOOl,r$ = .34. The stroke-affected adults tended to spend more absolute time in each temporal phase compared with the control subjects (table 3); however, these differences were not significant. There were no group differences found in relative time of the three temporal phases (table 3). With practice, all participants decreased the absolute time spent in each temporal phase (fig 3). There were main effects of practice block for acceleration (p < .OOOl,+qz= .26); deceleration (p < .OOOl,n2 = .18); and dwell time (p < .05, q2 = .32). The relative time spent in each temporal phase did not change with practice.

AFTER

Table

STROKE,

Pohl

571

3: Mean and Standard Times and Kinematic

Movement

time

Absolute time Acceleration Deceleration Dwell Relative time

Error of the Mean for Movement Measures Across Practice

(msec)”

High Complexity

Stroke

Stroke

Control

Control

393 (9) 302 (5) 953 (14)

846 (12)

197 (5)

252 (4)

(msec) 157 (3) 266 (5) 108 (3) 456 (9) 36(l) 231 (13)

145 (4) 50(2)

453 (9) 141 (9)

(%)

Acceleration Deceleration Dwell Peak horizontal

Low Complexity

velocitv

* Group, p < .02. + Complexity X group,

(cm/set)+

51 (0)

52 (0)

2X (0)

30 (I)

37 (0)

36 (0)

48 (I)

54 (1)

13 (0) 12 (0) 173 (3) 212 (3)

24(l) 70 II)

16 (1) 73 (2)

p < ,007

Peak Horizontal Velocity Across practice blocks, stroke-affected adults had lower peak horizontal velocities compared with control subjects in the low-, but not the high-, complexity condition (table 3). This resulted in a complexity by group interaction, p < .007, q2 = .14. Both groups increased peak horizontal velocity with practice in the low-complexity condition from an average of 149cm/sec in the first practice block to 208cm/sec in the last practice block. In contrast, there was little change in peak velocity with practice in the High-complexity condition; the average peak velocity was 68cmisec in the first block and 70cmlsec in the last block. The complexity X practice block interaction was significant, p < .OOOl,q2 = .29. Right Versus Left Stroke Although not a primary question of this study and with too small a sample size to permit statistical analyses, some interesting trends emerged in the performance of individuals affected by stroke on the right versus left compared with control subjects using the same hand. For participants using the right hand, approximately 30% of MT was spent in acceleration, 53% in deceleration, and 17% in dwell time, and these percentages did not change significantly with practice. Controls using the left hand had a similar temporal distribution over practice to those

”

,I,,-./

12345

6

/

/

7

8 Practice

I

9

IO

11

12

Phys

Med

14

Block

Fig 2. Scatter plot of movement times for individuals with high-complexity condition; 0, low-complexity condition) trols (U, high-complexity condition; I, low-complexity for practice blocks.

Arch

13

Rehabil

stroke (0, and concondition)

Vol 80, June

1999

672

UPPER

EXTREMITY

PRACTICE

AFTER

STROKE,

Pohl

participants, were demonstrated by all five participants with left stroke lesion and persisted over practice. Figure 4 gives an example of the prolonged dwell times of one participant with left stroke lesion (left side series) compared with a control subject using the left hand (right side series). The time series of the participant with left stroke lesion show prolonged zero positions and velocities in the horizontal and vertical dimensions at the same point in time (eg, just prior to 2sec). As indicated in plot D, these times without movement are dwell times, ie, the times the stylus rested on a target. In contrast, the time seriesof the control show brief zero positions and velocities at target hit, as indicated in plot H. Dwell times for this participant were brief. 0%/I

12345

A

1

1

1

1

1

1

’

’

7

8

9

10

11

12

13

14

l

0

l

8

8

,

Practice

600

lo

E

300

-

'G= a :

200-

Block

.

n *e.

2 2

6

100

-

n mmm

.

a

.

n

t

1

1

1

1

1

1

1

1

6

7

8

9

10

11

12

13

14

0

0

0

0

q

0

0

l

I

I

I

o-/I

12345

:

6

f

m

Practice

:

Block

600 500

I

c g

400

-I

.-E +

300

-

= ii 13

o 000

0

200

-

q 0

O

.

0

0

q

q

0

0

I

I

I

I

.

n m3;11

a

0+-l+

1

12345

C

0

O

q 000

‘00

0

6

7

1

1

8

9

Practice

1

1

1

10

11

12

11

13

14

Block

Fig 3. Absolute time in each temporal phase for individuals with stroke (0, high-complexity condition; 0, low-complexity condition) and controls (0, high-complexity condition; n ; lowicomplexity condition): (A) acceleration; (B) deceleration; (C) dwell.

using the right hand, differing only by 1% in acceleration and deceleration and having the same relative dwell time. In contrast, individuals with left stroke lesion spent 26% of MT in acceleration, 42% in deceleration, and 32% in dwell time. These prolonged dwell times, with an absolute duration of approximately 300msec and over twice that of all other Arch

Phys

Med

Rehabil

Vol 80, June

1999

DISCUSSION The present study demonstrates that, with practice, strokeafflicted adults can achieve faster movements with the lessaffected UE, but the movements remain slower than in unafflitted control subjects. Previous work has demonstrated a slowing in the less-affected UE in adults with unilateral brain damage in tasks such as aiming between two targets? single target tapping,3 and sequencing hand postures.20,2*Extending these findings, results of the present study suggest that practice can amerliorate some of these motor performance deficits. Practice is the foundation of the physical rehabilitation of adults who have suffered stroke. There is some, albeit surprisingly little, evidence that motor performance can be improved with practice in these persons. Motor control of the paretic, contralateral UE after stroke can be improved in some individuals through rigorous practice or forced use in which the less-affected UE is physically restrained.22 Other less-rigorous practice schedules have resulted in improved performance of the paretic UE.23-26 Although motor deficits in the less-affected UE have been documented,27 investigators have not studied the effects of practice on these deficits. Some studies have measured motor recovery of the less-affected UE without including any specific intervention. Marque and colleagues2* evaluated motor function of the less-affected, left UE 20 and 90 days after onset of left stroke. Tests included hand dynamometry, elbow and wrist isokinetic testing, finger tapping, and a pegboard task. At 20 days, ipsilateral deficits were found in all but finger tapping in comparison with control subjects. At 90 days, the only ipsilatera1 UE motor function that remained impaired was performance in the pegboard task. Jones and colleagues* reported reduced grip strength and proximal strength of the less-affected UE in individuals 11 days after stroke; grip but not proximal deficits resolved at 1 year after onset. Motor control deficits of the UE on the same side as the brain lesion reflect the disrupted output from the damaged hemisphere, and this output has bilateral influences on performance. Although it is acknowledged that the relative magnitude of the motor deficit of the UE on the side of the lesion is small, control of the less-affected UE is not normal. Ipsilateral corticofugal pathways contribute to motor control of the UE; however, the extent and importance of ipsilateral pathways are still not well defined. There is evidence that ipsilateral cortical influence on UE control is greater for the left UE, compared with the right, in right-hand-dominant individuals.29 Using functional magnetic resonance imaging (fMRI), Kim and colleagues30found that the left hemisphere showed no significant difference in activity between movements of the right and left fingers. In contrast, greater right hemisphere activation was found with movements of the left, compared with the right, fingers. This finding is congruent with studies of individuals with brain damage that have shown motor deficits of the UE on the same side as the

UPPER

EXTREMITY

PRACTICE

AFTER

30

1

2

3

4

5

-I

0

STROKE,

Pohl

targethit

\

7

Time (s)

Fig 4. lime high-complexity hand.

series

of (A, condition.

E) horizontal The left-side

position, (B, F) horizontal series is from an individual

673

2

3

4

5

Time (s)

velocity, (C, G) vertical who had a left stroke; the

lesion, particularly in those with left lesions.7,20z31,32 The present study included too few subjects to address issues of laterality, however the trend toward longer dwell times in the highcomplexity condition for those with stroke (fig 4) may be related to the inclusion of subjects with left hemisphere lesions. Motor deficits of the ipsilateral UE after stroke may also reflect disruption in bilateral hemispheric activation. Studies that have used brain imaging have documented bilateral hemispheric activation with unimanual goal-directed actions.8,10,11

position, and right side from

(D, H) vertical velocity a control subject using

in the the left

Of note, ipsilateral and bilateral hemispheric activation are more prevalent in high-task-complexity conditions.8 This association led to our hypothesis that participants who had suffered stroke, compared with control participants, would not show improvement with practice in the high-complexity condition. This hypothesis was not supported by the results of this study. The stroke subjects, like the control participants, were able to decreaseMT with practice in high- and low-complexity conditions, which was shown by the lack of a three-way interaction, Arch

Phys

Med

Rehabil

Vol80,

June

1999

674

UPPER

EXTREMITY

PRACTICE

group X complexity X practice block. This, however, may have been owing to the small number of participants, resulting in a small/medium effect size (q* = .04) and an observed power of .44. Further work with a greater number of participants could elucidate whether or not this finding is robust. It is reasonable to suggest that the neural substrate responsible for the decreasein MT with practice is different for stroke versus control subjects. There is converging evidence from a variety of methodologies to suggest that motor recovery of the paretic UE is the result, at least in part, of an increase in activity in the nondamaged hemisphere.33-37Clinical improvement in the paretic UE was associated with reorganization of the contralateral motor cortex and, to a lesser extent, the ipsilateral motor cortex.38 Undamaged contralateral nuclei may also contribute to motor recovery.3g Future studies of adults with stroke damage performing with the less-affected UE that use brain imaging or magnetic transcranial stimulation during motor skill acquisition may provide some insight into the interactions of pathology, practice, and skill acquisition. In the present study, changes with practice in the kinematics of aiming were a function of the complexity of the task and were not different between stroke-afflicted individuals and control subjects. However, the observed power in this study to assessthe block X group interaction was less than SO, so this result must be interpreted with caution. In the low-complexity condition, participants increased peak velocity with practice. The decrease in MT in the low-complexity condition was accompanied by a decrease in the absolute time spent in each temporal phase, while preserving relative time. These findings replicate the changes found in healthy young and older adults practicing aiming in a low-complexity condition when each participant was required to use the left, nondominant UE.16 In the high-complexity condition, participants did not change peak velocity with practice. They did, however, decrease the absolute time spent in each temporal phase in the highcomplexity condition without altering the relative temporal structure. This is in contrast to previous results16 in which healthy young and older adults, practicing in a high-complexity condition, altered the relative temporal structure by increasing relative acceleration time and decreasing relative dwell time. Older adults, performing in the same high-complexity condition as in the present study, spent an average of 3 1% of MT in acceleration time over practice, 52% in deceleration, and 17% in dwell time. These results are similar for the participants in the present study (four of the twenty participants were the same), except for those with left stroke damage who, using their left, nondominant UE, demonstrated prolonged absolute and relative dwell times (fig 4). The prolonged dwell times of those with left stroke damage are consistent with the prolonged reversals seen in those with left-sided brain damage.7 This protracted time in which the stylus is resting on the target is thought to reflect an inability to switch or sequence components of the movement. This finding is compatible with a specialized role for the left hemisphere in timing and sequencing movements or parts of movements. *Os21 The absence of change with practice in relative dwell time in this small group of individuals with left stroke damage suggeststhat this aspect of motor control may be resistant to practice in those with left-sided brain damage. A larger group of participants is necessaryfor sufficient power to evaluate this statistically. Physical rehabilitation of stroke patients should include rehabilitation of the less-affected UE.40s41Even in individuals who are unable to regain use of the contralateral UE after stroke, functional improvements in UE tasks have been documented.40,42d4It is likely that improved control of the lessArch

Phys

Med

Rehabil

Vol 80, June

1999

AFTER

STROKE,

Pohl

affected UE was responsible at least in part for these functional gains. The results of the present study demonstrate the effectiveness of practice in improving movement speed in the less-affected UE after stroke. Limitations in the study must be acknowledged. First, generalizations to the general stroke population must be made with caution. Screening narrowed subject selection to those who had suffered unilateral hemispheric stroke, and there were only 10 stroke subjects in our sample. The exclusion of those with receptive aphasia may have resulted in a sample population of those with left stroke damage with smaller lesions. Indeed the Fugl-Meyer motor scoresof the contralateral UE of left stroke subjects tended to be higher than those of right stroke subjects (table l), suggesting that they were less involved. Future studies with larger sample sizes and more detailed descriptions of the locations and extent of stroke are needed. Second, practice was limited to a single l-day session. It is not known if additional practice or some specific intervention could lead to further improvement in motor control in the less-affected UE. Future studies are needed to evaluate the permanency of practice effects for the less-affected UE and to assessthe transfer of improvements in motor control incurred during laboratory practice to everyday functional tasks. The authorsthank PamelaDuncan, PhD, for Acknowledgment: her commentson an earlierversionof this manuscript. References

ColebatchJG, GandeviaSC. The distribution of muscularweakness in upper motor neuron lesions affecting the arm. Brain 1989;112:749-63. 2. JonesRD, DonaldsonIM, ParkinPJ.Impairment and recoveryof ipsilateral sensory-motorfunction following unilateral cerebral infarction. Brain 1989;112:113-32. 3. Robinson LM, Fitts SS, Kraft GH. Laterality of performancein fingertappingrate and grip strengthby hemisphereof strokeand gender.Arch PhysMed Rehabil 1990;71:695-8. 4. HaalandKY, Harrington D. Hemisphericcontrol of the initial and correctivecomponentsof aiming movements.Neuropsychologia 1989;27:961-9. 5. Fisk JD, GoodaleMA. The effects of unilateral brain damageon visually guidedreaching:hemisphericdifferencesin the nature of the deficit. Exn Brain Res 1988:72:425-35. 6. HaalandKY, BarringtonD, Yei R. The effectsof task complexity in left andright CVA patients.Neuropsychologia1987;25:783-94. I. Winstein CJ, Pohl PS. Effects of unilateral brain damageon the controlof goal-directedhandmovements.ExpBrain Res1995;105: 163-74. 8. WinsteinCJ, Grafton ST,Pohl PS.Motor task difficulty and brain activity: investigation of goal-directedreciprocal aiming using positronemissiontomography.J Neurophysiol1997;77:1581-94. 9. Fitts PM. The information capacityof discretemotor responses.J ExpPsycho11954;47:381-91. 10. ColebatchJG, Dieber MP, PassinghamRE, Friston KJ, Frackowiak RSJ.Regional cerebralblood flow during voluntary arm and hand movementsin humansubjects.J Neurophysiol1991;65:1392401. 11. DieberMP, PassinghamRB, ColebatchJG,FristonKJ, Nixon PD, FrackowiakRSJ.Cortical areasand the selectionof movement:A studywith positronemissiontomography.Exp Brain Res 1991;84: 1.

393-402.

12. SchmidtRA. Motor control andlearning.2nd ed. Champaign(IL): Human Kinetics; 1988. 13. Fischman MG, Lim C-L. Influence of extended practice on programmingtime, movementtime, and transferin simple targetstrikingresponses.J Mot Behav1991;23:39-50. 14. Gottlieb GL, CorcosDM, JaricS,Agarwal GC. Practiceimproves eventhe simplestmovements.Exp Brain Res 1988;73:436-40.

UPPER

EXTREMITY

PRACTICE

15. Light KE, Reilly MA, Behrman AL, Spirduso WW. Reaction times and movement times: benefits of practice to younger and older adults. J Aging Phys Activity 1996;4:27-41. 16. Pohl PS, Winstein CJ. Age-related effects on temporal strategies to speed motor performance. J Aging Phys Activity 1998;6:1-17. 17. Bamford J, Sandercock P, Dennis M, Burn J, Warlow C. Classification and natural history of clinically identifiable subtypes of cerebral infarction. Lancet 1991;337:1521-6. 18. Schenkenberg T, Bradford DC, Ajax ET. Line bisection and unilateral visual neglect in patients with neurological impairment. Neurology 1980;30:509-17. 19. Fugl-Meyer AR, Jaasko I, Leyman I, Olsson S, Steglind S. The post-stroke hemiplegic patient. A method for evaluation of physical performance.ScandJ Rehabil Med 1975;7: 13-31. 20. Haaland KY. Harrington D. Limb seauencina deficits after left but not right hemispherldamage. Brain Cogn lF94;24: 104-22. 21. Kimura D, Archibald Y. Motor functions of the left hemisphere. Brain 1974;97:337-50. 22. Taub E, Wolf SL. Constraint induced movement techniques to facilitate upper extremity use in stroke patients. Top Stroke Rehabil1997;3:38-61. 23. Butefisch C, Hummelsheim H, Denzler P, Mauritz K-H. Repetitive training of isolated movements improves the outcome of motor rehabilitation of the centrallv. paretic hand. J Neurol Sci 1995;130: _ 59-68. 24. Hanlon RE. Motor learning following unilateral stroke. Arch Phys Med Rehabil1996:77:811-5. 25. Trombly CA. Observations of improvement in reaching in five subjects with hemiparesis. J Neurol Neurosurg Psychiatry 1993;56: 40-5. 26. Trombly CA, Thayer-Nason L, Bliss G, Girard CA, Lyrist LA, Brexa-Hooson A. The effectiveness of therapy in improving finger extension in stroke patients. Am J Occup Ther 1986;40:612-7. 27. Pohl PS, Winstein CJ, Onla-or S. Sensory-motor control in the ipsilesional upper extremity after stroke. NeuroRehabilitation 1997;9:57-69. Corrigendum published in NeuroRehabilitation 1997; 9~245-49. 28. Marque P, Felez A, Puel M, Demonet JF, Guiraud-Chaumeil B, Roques CF, et al. Impairment and recovery of left motor function in patients with right hemiplegia. J Neurol Neurosurg Psychiatry 1997;62:77-81. 29. Kawashima R: Yamada K, Kinomura S, Yamaguchi T, Matsui H, Yoshioka S, et al. Regional cerebral blood flow changes of cortical motor areas and prefrontal areas in humans related to ipsilateral and contralateral hand movement. Brain Res 1993;623:33-40. 30. Kim S-G, Ashe J, Hendrich K, Ellermann JM, Merkle H, Ugurbil K, et al. Functional magnetic resonance imaging of motor cortex: hemispheric asymmetry and handedness. Science 1993;261:615-7.

AFTER

STROKE,

675

Pohl

31. Wyke M. Effect of brain lesions on the rapidity of arm movement. Neurology 1967; 17: 1113-20. 32. Giuliani CA, Purser JL, Light KE, Genova PA. Impairments in arm control in subjects with left and right hemisphere stroke. NeuroRehabilitation 1997;9:71-87. 33. Benecke R, Meyer B-U, Freund H-J. Reorganisation of descending motor pathways in patients after hemispherectomy and severe hemispheric lesions demonstrated by magnetic brain stimulation. Exp Brain Res 1991;83:419-26. 34. Brion J-P, Demeurisse G, Capon A. Evidence of cortical reorganization in hemiparetic patients. Stroke 1989;20:1079-84. 35. Chollet F. DiPeiro V. Wise RSJ. Brooks DJ. Dolan RJ. Frackowiak RSJ. The functional anatomy ‘of motor recovery after stroke in humans: a study with positron emission tomography. Ann Neurol 1991;29:63-71.

36. Silvestrini M, Caltagirone C, Cupini LM, Matteis M, Troisi E, Bernardi G. Activation of healthy hemisphere in poststroke recovery. Stroke 1993;24:1673-7. 37. Cao Y, D’Olhaberriague L, Vikingstad EM, Levine SR, Welch KM. Pilot study of functional MRI to assess cerebral activation of motor function after poststroke hemiparesis. Stroke 199829: 11222. 38. Traversa R, Cicinelli P, Bassi A, Rossini PM, Bernardi G. Mapping of motor cortical reorganization after stroke. Stroke 1997;28: 110-7. 39. Pantano P, Formisano R, Ricci M, DiPiero V, Sabatini U, DiPofi B, et al. Motor recovery after stroke: morphological and functional brain alterations. Brain 1996;119:1849-57. 40. Olsen TS. Improvement of function and motor impairment after stroke. J Net&l Rehabil 1989;3:187-92. 41. Tsai L-J. Lien I-N. The oerformance of the unaffected hand of stroke patients. J Formos Med Assoc 1982;81:705-11. 42. McDowell F, Louis S. Improvement in motor performance in paretic and paralyzed extremities following nonembolic cerebral infarction. Stroke 1971;2:395-9. 43. Stern PH, McDowell F, Miller SM, Robinson M. Factors influencing stroke rehabilitation. Stroke 1971;3:213-8. 44. Wade DT, Wood VA, Hewer RL Recovery after stroke-the first three months. J Neurol Neurosurg Psychiatry 1985;48:7-13. Suppliers a. Pulnix America Inc., 770 Lucerne Drive, Sunnyvale, CA 94086. b. Peak Performance Technologies, 7388 South Revere Parkway, Englewood, CO 80112. c. RUN Technologies, 25622 Rolling Hills Roads, Laguna Hills, CA 92653.

Arch

Phys

Med

Rehabil

Vol 80, June

1999