1992 Journal Of Neurophysiology

JOURNALOF NEUROPHYSIOLOGY Vol. 67. No. 5. May 1992. l’rirltcv’

Strength Increases From the Motor Program: Comparison of Training With Maximal Voluntary and Imagined Muscle Contractions GUANG

YUE

AND

KELLY

J. COLE

Department

of. Exercise Science, The University of Iowa, Iowa City, Iowa 52242

SUMMARY

AND

CONCLUSIONS

I. This study addressed potential neural mechanisms of the strength increase that occur before muscle hypertrophy. In particular we examined whether such strength increases may result from training-induced changes in voluntary motor programs. We compared the maximal voluntary force production after a training program of repetitive maximal isometric muscle contractions with force output after a training program that did not involve repetitive activation of muscle; that is, after mental training. 2. Subjects trained their left hypothenar muscles for 4 wk, five sessions per week. One group produced repeated maximal isometric contractions of the abductor muscles of the fifth digit’s metacarpophalangeal joint. A second group imagined producing these same, effortful isometric contractions. A third group did not train their fifth digit. Maximal abduction force, flexion/extension force and electrically evoked twitch force (abduction) of the fifth digit were measured along with maximal integrated electromyograms (EMG) of the hypothenar muscles from both hands before and after training. 3. Average abduction force of the left fifth digit increased 22% for the Imagining group and 30% for the Contraction group. The mean increase for the Control group was 3.7%. 4. The maximal abduction force of the right (untrained) fifth digit increased significantly in both the Imagining and Contraction groups after training ( 10 and 14%,respectively), but not in the Control group (2.3%). Theseresultsare consistentwith previous studiesof training effectson contralaterallimbs. 5. The abductiontwitch force evokedby supramaximalelectrical stimulationsof the ulnar nerve was unchangedin all three groupsafter training, consistentwith an absenceof musclehypertrophy. The maximal force of the left great toe extensorsfor individual subjectsremainedunchangedafter training, which argues againststrengthincreases due to generalincreasesin effort level. 6. Increasesin abduction and flexion forces of the fifth digit were poorly correlated in subjectsof both training groups. The fifth finger abduction force and the hypothenar integratedEMG increaseswere not well correlated in these subjectseither. Togethertheseresultsindicate that training-inducedchangesof synergistand antagonistmuscleactivation patternsmay havecontributed to force increasesin someof the subjects. 7. Strength increases can be achievedwithout repeatedmuscle activation. Theseforce gainsappearto resultfrom practiceeffects on central motor programming/planning.The resultsof theseexperimentsadd to existingevidencefor the neuralorigin of strength increasesthat occur before musclehypertrophy.

INTRODUCTION

Skeletal muscle strength gains from isometric strength training are believed to result from two principal factors. Whereas muscle hypertrophy occurs in the later stages of training (Enoka 1988; Komi 1986; McDonagh and Davies 1114

1984; Sale 1986), strength gains achieved during the first weeks of training reflect an increased ability to activate motoneurons and therefore appear to be neural in origin. Abundant evidence demonstrates neural changes that occur soon after training begins. Voluntary strength increases rapidly before the muscles exhibit hypertrophy (Fukunaga 1976; Ikai and Fukunaga 1970; Jones and Rutherford 1987; Liberson and Asa 1959; Moritani and de Vries 1979; Rose et al. 1957; Tesch et al. 1983) and before increases in electrically evoked tension occur (Davies and Young 1983; McDonagh et al. 1983 ) . These early voluntary strength increases are accompanied by increased integrated electromyograms (EMG) (Komi 1986; Sale 1986) and increased reflex gains (Milner-Brown et al. 1975; Sale et al. 1982, 1983a,b; Upton and Radford 1975 ). The greatest strength gains occur at the trained joint angles (Gardner 1963; Lindh 1979; Meyers 1967 ). Finally, training of one limb is associated with increased voluntary strength in the contralateral (untrained) muscles (e.g., Hellebrandt et al. 1947; Houston et al. 1983; Rose et al. 1957; see Enoka 1988; Sale 1986 for a review), even though the contralateral muscles remain virtually quiescent during training (Houston et al. 1983; Panin et al. 196 1; Yasuda and Miyamura 1983). The neural mechanisms of these strength gains are poorly understood. However, the previously noted phenomenon of increased strength from muscles contralateral to those that were trained raises the intriguing possibility that early strength gains may be induced without repetitive muscle activation; that is without repetitive activation of motoneuTons, spinal interneurons, or possibly, descending motor pathways. Instead, early strength gains may depend on changes in the central programming of a maximal voluntary contraction. If so, it may be possible to induce changes in the motor program for a maximal voluntary contraction via mental practice (that is, imagining performance of a skilled movement without overt or covert muscle activation). Research on skill acquisition has demonstrated clearly that mental practice leads to improved performance (e.g., Rawlings et al. 1972; Vandell et al. 1943; also see Corbin 1972; Fets and Landers 1983; Hall 1985; Richardson 1967 for a summary). Individuals asked to imagine themselves performing skilled, serial movements of the digits manifested increased cerebral blood flow in nonprimary motor regions of the cerebral cortex without evidence of activating the primary motor cortex (Roland et al. 1980). It is possible therefore that repeated imagined muscle contractions may alter the program for maximal torque production at a joint. These central changes may yield increased motoneu-

0022-3077/92 $2.00 Copyright 0 1992 The American Physiological Society

STRENGTH

CHANGES

AFTER

IMAGINED

ron activation and/or change the relative activation levels of synergist and antagonist muscles across a joint to yield increased torque in the desired direction. The latter phenomenon, changes in muscle “coordination,” may occur particularly for strength training involving multijoint movements or movements at a joint with multidirectional freedom (Rutherford and Jones 1986). The contribution of this improved coordination to early strength gains remains unclear. This investigation addressed whether imagined maximal muscle contraction training increases maximal voluntary contraction (MVC) force and compared any strength increases with those found in the early periods of training by the use of MVCs. Our experimental design included attempts to control or measure effort level, strength increases not specific to the trained muscle, and muscle hypertrophy. After 20 sessions, imagined maximal contraction and MVC training produced comparable levels of voluntary strength increase. METHODS

subjects The subjectswere30 healthy individuals(2 l-29 yr of age)who claimednot to have participated in regular musculartraining in the two yearsbeforethis study. Ten subjectseachwererandomly assignedto one of three groups.Subjectsin the Imagining group were trained by an imagined maximal muscle contraction method; subjectsin the Contraction group produced MVCs during their training. The Control group was instructed to refrain from strength training and sportsactivities using the hand. Informed consentwasobtainedfrom eachsubjectbeforethe experiment. Three subjects( 1in the Control group and2 in the Contraction group) failed to return for posttrainingtesting.

General aperimentul

procedures

The experiment consistedof a pretraining measurementsession, a 4-wk (20-session)training period focusedon increasing abduction force of the metacarpophalangeal joint of the left fifth digit, and a posttraining measurementsession.Abduction of the left fifth digit wasselectedbecausethe musclesof abduction are usedprincipally in taskssuchasopeningthe grip and thus are not typically engagedin sustained,high force output in mostactivities of daily living. Training effects(neural changesand musclefiber hypertrophy) may be achieved more easily. The following data were collected from each subject’sipsilateral and contralateral handsduring the pre- and posttraining measurementsessions: I) maximal isometric abduction force at the metacarpophalangeal joint of the fifth digit; 2) flexion/extension force at the metacarpophalangealjoint of the fifth digit during maximal abduction force production (measuredto indicate possiblechangesin musclecoordination): 3) twitch force (abduction) of the fifth digit generated by supramaximalelectricalstimulation of the ulnar nerve (taken asa possibleindication of musclehypertrophy; however, cf. DISCUSSION): 4) surfaceEMG during the maximal abduction force task from electrodesplacedon the hypothenar eminenceapproximately over the abductor digiti minimi (although other hypothenar muscles,or the interosseimay have contributed to this EMG during a maximal force exertion becauseof their proximity to the recordingelectrodes).In addition, the maximal extensionforce of the left greattoe servedasa control for nonspecifictraining effects suchasstrengthgainsin untrained muscles.

CONTRACTION

TRAINING

1115

Equipment Horizontal (abduction) force of the fifth digit was measured with a cantilever-beam transducer instrumented with strain gauges(Fig. 1). A secondforce transducerwasmountedorthogonally on the first to measurethe vertical ( flexion-extension) force. The transducerswereconstructedof two stainlesssteelbars2.5 x 7.0 X 0.1 cm. The crosstalk of the vertical force transducerto the horizontal transducer was 0.45%, indicating a near orthogonal relation betweenthe steelbars.The endof the beamthat measured the vertical force wassecurelyattachedto a rigid microscopestand to allow adjustmentsof the transducersystem’sposition. A metal tube ( 12mm long and 15mm diam) wassecuredto the free endof the transducerthat measuredthe abduction force. The tube could be moved from one sideof the bar to another so that the same transducerswere used for both hands(Fig. 1). A similar transducer systemwasusedto measurethe extensionforce of the left great toe.

Procedures The training programs for the Imagining and Contraction groupslasted4 wk with five training sessions per week. The subjects were trained Monday through Friday of each week in the Motor Control Laboratory at The University of Iowa. IMAGINEDCONTRACTION TRAINING. Duringeachtrainingsession, subjectsin the Imagining group were instructed to perform 15 imaginedmaximal contractions of the left abductor digiti minimi with a 20-srestafter every secondtrial. Subjectsweretold that the training involved imagining musclecontractions and that all the upper limb muscles,especiallythe trained muscle,shouldbe relaxedasmuch aspossible.In short, they wereinstructedto maximally activate the brain but not to activate the muscle.HypothenarEMG (seeEMGMEASUREMENTS) wasmonitored during most of thesetraining sessions. Subjectswere seatedcomfortably in a chair with their left arm restingon a table.The hand waskept open and relaxedwith the palm facingdownward. On a verbal signalto begin, the subjectwas instructed to concentrate on the left fifth finger and imaginethe finger pushingmaximally againstthe horizontal force transducerthat wasusedduring the pretraining test. The imaginary maximal finger abduction wasto be maintained

FIG. 1. Finger force measurement equipment and subject positioning. A: transducer for abduction force. B: transducer for extension/flexion force. C, D, and F: finger, hand, wrist, and arm restraints. E: surface EMG electrodes. G: reference electrode. Stimulation electrodes are located at the medial side of the elbow and are not visible in this drawing.

1116

G. YUE

AND

for 15 s in each trial before imagining a cessation of this pushing. During this 15-s period, the subject was instructed to keep imagining a voice shouting “harder, harder. . . .”

K. J. COLE

Theseelectrodeswere treated with an electrolyte paste,placedon the skin overlying the ulnar nerve located between the medial epicondyleand the olecranonof the ulna, and held in placewith MVC TRAINING. The subjectsin the Contraction group sat in a an elasticstrap.The stimuli consistedof rectangularvoltagepulses chair with their left armsrestingon a table. The hand wasrelaxed, 0.2 ms in duration delivered from a stimulator (GrassmodelS88 with palm facing downward and the distal part of the fifth digit with a Grassmodel SIUS stimulus isolation unit). The stimulus positionedagainsta stop that wasrigidly fixed to the table surface. intensity wasset30 V greaterthan that for a maximal M response. The subjectabductedhis/her left fifth digit asforcefully aspossi- There were 10 twitch trials (providing twitch force and M-wave ble. During each training session,subjectsin the Contraction data) in eachtest with 10 s rest betweeneachpair of trials. group were instructed to perform 15 MVCs of the left abductor TOE FORCE. Isometric force measureswere obtained from the digiti minimi with 20 s of rest separatingeachcontraction. Each left great toe of seatedsubjectsby the useof procedurescomparable with thosedescribedfor the fingerforce measures. Eachsubject trial lasted- 10s. performed five to six trials of maximal toe extension force and CONTROL GROUP. Subjects in the Control group were not restedfor at least30 s betweenevery two trials. trained, but participated in the measurementsbefore and after training. The subjectswere instructed strictly to avoid any physical exerciseor mental training of the hypothenar musclesduring Data recording and analysis the time period betweenthe pre- and posttests. The force transducer signalswere amplified and low-passfilPRE- AND POSTTRAINING FORCE MEASURES. The subjects tered at 500 Hz (Biocommunication Electronics model 205). restedtheir armson the table with the palm of the hand facing EMG signalswere differentially preamplified ( 10,000MQ input downwardand slid the fifth digit into the metal tube on the force impedance)with a gain of 100and a bandpassof 25-10,000 Hz transducer(Fig. 1). The distal edgeof the tube (the edgeaway ( BiocommunicationElectronicsmodel301) ; subsequentamplififrom the subject)wasalignedwith the baseof the fingernailduring cation occurred with a low-passcutoff of 2,500 Hz. All data were both pre- and posttests.The index, middle, and ring fingerswere recordedon an eight-channel(Vetter model D) FM tape recorder stabilizedon the tablewith a strap.An aluminum bar mounted on (DC-l, 125 Hz) and stored for later analysis.Signalswere then the tablewasplacedbetweenthe fourth and fifth digitsto maintain digitized with an analog-to-digital converter with 12-bit resoluan angleof - 5’ betweenthe digits; this ensuredroughly similar tion. The force and EMG signalsweredigitized at 1,000samples/ musclelengthsduring both pre- and posttests.Strapswere usedto s, and the M-wave and twitch responseswere digitized at 3,000 samples/s. prevent arm motion or vertical motion of the wrist (Fig. 1). For eachtrial of digitized data, the peak abduction forceand the Eachsubjectwarmedup with severalsubmaximalcontractions, then executed five or six maximal contractions of the abductor correspondingvertical force were measured( Fig. 2). The highest digiti minimi lasting -4 s with a restingperiod of at least 30 s abduction force obtained from among the recorded trials was betweentrials. Each subjectwasinstructed to exert a lateral force taken as the maximal abduction force of the fifth digit. The toe with the fifth digit againstthe force transducer.The transducers forceswere measuredsimilarly. EMG signalswererectified and integratedover a 3-speriod that werestiff ( 15 N yielded a 1-mm displacementof the unfixed end contained the peak abduction force. A normalized integrated of the steelbar), and therefore the contractions were virtually isometric becausemost subjects’ peak abduction forces were EMG value wascalculatedfrom the ratio of the highestintegrated EMG and the subject’saverageM-wave value (the standarddevia<20 N. An inherent difficulty in strengthtraining studiesisthe possibil- tion was < 1%of the meanM-wave in most of the subjects).This ity that subjectsdid not usetruly maximal efforts for producing ratio wasusedasthe dependentEMG measure. forcesin pretrainingtests.Thereforewe attemptedto increasesubjects’ motivation during the pretraining test. Each subject was Statistical analyses askedabout his/her height and weight after a few attempts at One-way analysesof variance were performedbecauseour priMVC. A horizontal trace wasthen indicated on the oscilloscope that was5- 10%higherthan the subject’smost recent force trace. mary interestwasto comparethe data for a specificvariable(e.g., The subjectwasthen told that previous researchindicated that 80%of peoplehis/ her size can reachthe indicated target. EMG MEASUREMENTS. Bipolar surface EMG of the abductor digiti minimi muscle was obtained during the maximal finger force measuresof eachhand. Before the fifth finger wasinserted into the metal tube, the skin overlying the musclewascleansed \ ABDUCTION FORCE with alcohol. Burdick Cor-Gel Electrolyte wasappliedto a pair of in vivo metric (IVM) silver silver-chloride electrodes(4 mm diam). Theseelectrodeswereappliedto the skin over the abductor digiti minimi musclebelly and oriented in a line roughly parallel to the musclefibers with - 15 mm betweenthe centersof the electrodes.The location of eachelectrodeon the skinwascarefully measuredin relation to anatomic landmarks and marked with permanentink during the pretestto facilitate electrodelocations for the posttest.The indifferent electrodewasplaced on the skin overlying the lateral epicondyleof the humerus. TWITCH FORCE. The abduction force resultingfrom supramaxima1electrical stimulation of the ulnar nerve at the elbow was measuredwith the apparatusdescribedpreviously. The stimulating electrodesweretwo chlorided silver disks, 10 mm diam, that weremountedin a perspexholder with their centers30 mm apart.

STRENGTH

CHANGES

AFTER

IMAGINED

CONTRACTION

TRAINING

1117

24

f‘ ;18 a0 0 IL

_ l2 6 t

0’ 0

1

2

3

4

5

6

7

8

9

0

10

’

’

’

’

’

’

’

1

2

3

4

5

6

7

SUBJECT

SUBJECT 40

T w

30

FIG. 3. Individual and group left 5th finger abduc8 tion force changes after training. Open squares, pretraining test values. Filled squares, posttraining test values. A : from the Imagining group. B: from the Contraction D group. C: fro m the Control group. D: mean and standard deviation of percentage increases of left 5th finger abduction force of each group.

z g

20

Z s

0’

’

’

012

’

’

’

’

’

’

3

4

5

6

7

8

-A 9

10

0 IMAGINING

CONTRACTION

CONTROL

GROUP

SUBJECT

abduction force) of the pretest with that of the posttestof each hand in eachgroup. The dependentmeasureswere maximal abduction force and its correspondingvertical force of the fifth digit, twitch force, EMG to M-wave ratio of the hypothenar musclesof two hands,and the extensionforce of the left greattoe. The probability of 0.05 wasselectedasthe level of significancefor all statistical analyses. RESULTS

Twined hand abduction .force The Imagining group’s maximal voluntary abduction force of the left fifth digit increased 22% (Fig. 3, A and D; Table 1) on average after a 4-wk, 20-session imagined contraction training period (P < 0.001). EMG monitoring during 80% of this group’s training sessions confirmed that the hypothenar muscles remained quiescent (Fig. 4). The abduction force of the left fifth digit of the Contraction group increased 29.75% (P < 0.00 1; Fig. 3, B and D). Every

subject in the two groups demonstrated a force increase (Fig. 3, A and B), but the size of the increase was more variable across subjects in the Imagining group than in the c on t raction group (Fig. 3 D). The difference in the mean abduction force increase between these two groups after training was not statistically significant (P > 0.1). In contrast, the untrained (Control) group showed only a 3.7% increase in the abduction force (P > 0.08; Fig. 3, C and 0). Contralateral

untrained hand abduction force

The untrained (right hand) fifth digit of the Imagining group demonstrated a 10.45% abduction force increase on average (P < 0.005, Fig. 5 and Table 1). Likewise, the abduction force of the right fifth finger of the Contraction group showed a 14.43% increase (P < 0.02), which is consistent with previous studies (see Enoka 1988; Sale 1986 for a summary). The MVC force changes after training between the Imagining and Contraction groups were not sig-

1. Changesin abduction.forceofthe trained and untrained hands,EMG, twitch force, and verticalforce . qfthe trained hand qfier training TABLE

Tested Variables Abduction force, trained hand Abduction force, contralateral hand EMG

Flexion force Twitch force

Groups

Pre

Imagining” Contraction b Control’ Imagining Contraction Control Imagining

9.14 + 3.38 14.01 + 3.28 12.13 AI 3.23 9.43 f 3.23 13.01 k 2.92 11.34 IL 3.36 2.08 AI 1.42 (1.75 It 1.04)9 1.62 t 0.67 1.74 + 0.45 4.36 k 1.42 7.06 f: 3.41 6.13 AI 1.52 3.18 + 1.00 4.57 + 1.92 3.7 1 + 2.25

Contraction Control Imagining Contraction Control Imagining Contraction Control

Post 11.15 18.20 12.57 10.41 14.85 11.61 2.53 (2.35 2.33 1.71 5.78 9.07 5.81 3.21 4.46 3.82

z!I 3.9 + 3.97 2 3.21 AI 3.28 + 3.41 + 3.38 f 0.91 f 0.76)8 + 0.93 I!I 0.51 of:2.30 Ifr 3.14 AI 1.57 k 1.19 + 1.68 IL 2.21

%Increase 22.03 d 29.75 d 3.68 10.45” 14.13f 2.3 21.73 34.34f 43.80f -1.63 32.2 1 28.39 -5.20 1.04 -2.48 2.91

Values are means + SD in N except values in EMG, which represent means k SD of the ratios of integrated EMG to M-wave. EMG, electromyogram. an = 10. bn = 8. ‘n = 9. dP < 0.00 1. “P < 0.0 1. fP < 0.05. gValues from the analysis when the EMG scores from subject 10 were excluded.

1118

G. YUE AND K. J. COLE

strength increases. The left hypothenar muscle EMG data are listed for the three groups in Table 1. The integrated EMG normalized to the M-wave (see METHODS) of the 500 ms trained hand in the Imagining group increased after training in 9 of the 10 subjects to an average of 2 1.73% (Fig. 6, EMGMVC Table 1). However, this EMG increase fell short of our designated level of statistical significance (P = 0.08). Further statistical analyses of the data revealed a powerful effect from the data of subject 10 (Table 1 ), whose EMG level substantially decreased with a seemingly paradoxical increase in abduction force ( 13%, Fig. 3A ), and no change in twitch force. Therefore, assuming that muscle hypertrophy did not occur ( see DISCUSSION), the strength increase in this subject must be due to I ) increased net muscle activity that FIG. 4. Comparison between raw surface EMG from an imagined maxiwas not registered electromyographically, or 2) improved mal contraction and that from an MVC. Top trace: during an imagined coordination of muscles, such as a decrease in activity of maximal muscle contraction (EMG IMMC) of the left 5th digit. Bottom antagonist muscles (see DISCUSSION). The left hypothenar trace: during a maximal abduction contraction (EMG MVC) of the same EMG of the Contraction group demonstrated a significant digit. The figure indicates that the muscle was inactive during the imagined contraction training. From subject 9 in the Imagining group. increase (43.8%) after training (P < 0.05; Fig. 6, B and 0). The posttest EMG of the same muscles of the same limb in nificantly different (P > 0.5, Fig. 5 ) . The force change ob- the Control group remained unchanged (- 1.6%, P > 0.5; served for the right hand of the Control group between the Fig. 6, C and D). In the right (untrained) hand, no signifipre- and posttests was not significant (a 2.3% increase, cant EMG change was found in any of the three groups after training even though the Imagining and Contraction P > 0.3). groups increased their strength significantly. A Pearson product-moment correlation coefficient analyStrength of the control muscles sis across subjects indicated that increases in the abduction Extensors of the left great toe were used as nontrained, force were not strongly correlated with increases in the hynonhand control muscles within each subject. The exten- pothenar EMG in either the Imagining group ( r = 0.57, P > sion force measured from the great toe increased 3 t 4.74% 0.08) or the Contraction group ( r = 0.09, P > 0.5 ). (SD) in the Imagining group and decreased 1.8 t 6.12% (SD) and 3.1 t 4.8% (SD) in the Contraction and Control Flexion force of thejjlh digit groups. These changes did not achieve statistical signifiFinger flexion forces were monitored as an indirect indicance. cation of potential changes in activation patterns across muscles because of training ( Fig. 7 ). That is, proportional Muscle activity changes increases in abduction and flexion forces are consistent A finding of increased EMG levels in the two training with increased strength in abductor digiti minimi given its groups would be consistent with increased motoneuron ac- flexion and abduction moment arms at the metacarpophativation as a mechanism for the observed abduction langeal joint (Brand 1985; Thomine 198 1). In the trained EMGIMMC

20 1

B 8 cl m O

m

l

16 0 0

0

0 l 0

01 0

1

2

3

4

5

6

7

6

9

10

0

’

’

’

’

’

’

’

1

2

3

4

5

6

7

SUBJECT

SUBJECT

20

B 16 5F

I

I 8 0

012

3

4

SUBJECT

5

6

7

6

9

IMAGINING

CONTRACTION

GROUP

CONTROL

0

FIG. 5. Individual and groun right (contralateral \ untrained) 5th finger abd&ion foice changes after training. Open squares, pretraining test values. Filled squares, posttraining test values. A-C as in Fig. 3. D: mean and standard deviation of percent increases of the right 5th finger abduction force of each group.

STRENGTH 6-

A

O5

CHANGES

AFTER

IMAGINED

CONTRACTION

TRAINING

1119

4-

0

n

F

03

B

w

n

-

m

a4CT

I

s3-

!

m

z*W

:

m 0

8

n cl

1 -

0,0

2 CT

’

’

1

’

2

’

3

’

4

li*

.

l

O

0

Cl 7’ 8’

’

5

6

Id 0

z Wl-

’

9

10

1

I

n

n

.

2

.0

I

1

l 0

1

,

3

,

6

7

8

SUB:EC:

C

0

I

0 0

SUBJECT

!

n Cl

0

cl

m 0

0

0

n

s

0

m

150

-

120

-

ifi 2

go-

5

60

-

z s

30

-

D

FIG. 6. Individual and group left hypothenar integrated EMG changes after training. Integrated EMG was normalized by measuring the ratio of integrated EMG to M-wave. Open squares, pretraining test measures. posttraining test measures. A-C . kg. -. Filled a U: - squares, ..-.. deviation as m 3. mean and standard of^percent increases of the left hypothenar integrated EMG of each group. Note a large EMG increase (subject 7) and decrease (subject 10) in the Imagining group, large increases of subjects I and 2 in the Contraction group. Also note the large variations of EMG increases among subjects in the 2 experimental groups.

0

0

12

3

4

5

6

7

8

9

IMAGINING

CONTRACTION

SUBJECT

CONTROL

GROUP

(left) hand, the metacarpophalangeal flexion force increased 32% (P > 0.1) for the Imagining group. The Contraction group showed an increase of 28.4% (P > 0.2), and the Control group showed a 5.2% decrease (P > 0.5) of the metacarpophalangeal flexion force (Table 1). Many subjects in the two training groups manifested flexion force increases that were disproportional to their abduction force increases. Moreover, some subjects showed no flexion force increase, whereas others showed decreased flexion force despite abduction force increases. These findings are summarized in Table 2 with the use of ratios of the percent flexion force increase to percent abduction force increase from each subject. A ratio of 1 indicates flexion force increases that were proportional to increases in abduction force. Pearson product-moment correlation coefficient analyses of the percent changes in abduction and flexion force were not significant for either the Imagining group ( r = 0.16, P > 0.5) or the Contraction group (r = 0.34, P > 0.3). It thus appears that in some subjects the increased abduction

strength may have resulted from improved coordination of the various muscles controlling the digit. This may represent another type of neural adaptation that occurs besides increases in net excitatory drive to the prime mover. However, unambiguous interpretations in this regard would require observation of all muscles that cross the metacarpophalangeal joint, particularly because most have moment arms in more than one plane. In the untrained hand, the Imagining group demonstrated an increase of 19.64% (P > 0.4) in the fifth finger vertical force, and the Contraction group had an increase of - 12% (P > 0.7). The vertical force of the right fifth digit of the Control group remained virtually unchanged (a 1% decrease, P > 0.9) after training. The flexion force increases were not proportional to the abduction force increases. Left fifth digit twitch abduction force The twitch forces generated by supramaximal electrical stimulations were measured, and an average of 10 trials was B

m l

0

1

2

3

4

5

6

7

8

9

10

0

1

SUBJECT

0

12

3

4

SUBJECT

5

2

3

m 0

0

6

7

SUBiEC?

6

7

8

9

IMAGINING

CONTRACTION

GROUP

CONTROL

0

8

FIG. 7. Individual and group left 5th digit flexion force changes after training. Open squares, pretraining test values. Filled squares, posttraining test values. A-C as in Fig. 3. D : mean and standard deviation of percent increases of the left 5th digit flexion force of each group. Although no group showed significant changes in flexion force, some subjects demonstrated much larger increases than the others (A-C). The Contraction and Imagining groups showed considerably larger variations in flexion force increases than those in abduction force (D).

G. YUE AND K. J. COLE

1120

2. Ratios of %increaseinflexion .force to that of ahduction&forceof’the left 5th digit of individual subjects TABLE

Groups Subjects

Imagining”

Contraction b

Control’

6.74 0.33 -0.68 -0.28 0.08 1.39

- 1.49 E” 0.83 Of 1.68 2.05 1.83 -6.46

I 2 3 4 5 6

-2.48 d 0.16 2.06 0.63 4.62 0.23

7 8

0.6 1

1.27

3.68 5.80 4.13

3.16

9 10

a 11 = 10. bag= 8. ‘n = 9. dA negative ratio indicates a decrease in the flexion force. “The abduction force increase was zero for this subject. fThe flexion force increase was zero for this subject.

used as the twitch force for each subject. No significant change was found in the twitch force after training in any of the three groups (P > 0.5, Table 1) . On average, the Imagining group increased 1.04 t 14.94% (SD), the Contraction group decreased 2.48 t 11.3% (SD), and the Control group increased 2.9 1 t 10.69% (SD). DISCUSSION

The maximal force increases in subjects who trained using effortful muscle contractions were similar to those who only imagined producing similar contractions during training. It is therefore possible that similar mechanisms were responsible for the strength gains in the two groups. If so, the increased strength that occurs early in a conventional training regimen, during the so-called “neural” phase, may not result from repetitive muscle activation. The possible origins of these strength increases may include programming/planning levels of a hierarchically organized motor system (Hasan et al. 1985; Brooks 1986). It is necessary first to consider alternative explanations to neural mechanisms that may underlie the observed strength increases. When measuring maximal voluntary contractions, one must very carefully consider psychological or emotional factors that may yield apparent strength increases. It is unlikely that the strength gains in the present study resulted from psychological or emotional factors given the Control group’s lack of strength gain and the virtual absence of strength increases of the left great toe extensors in any group. Also, sham information provided to subjects concerning their performance during the pretraining strength test appeared to be highly motivating (see METHODS) and, hopefully, reduced the likelihood of increased effort for the posttraining measures. Muscle hypertrophy also can significantly influence muscle strength after training. If the mass of a muscle is enlarged by increasing the cross-sectional area of each fiber, or the number of the fibers after training, a larger muscle strength output is expected from the increased number of parallel sarcomeres in the muscle (Edgerton et al. 1986). However, repeated production of high-intensity muscle activation is typically required for muscle hypertrophy (Atha

198 1; Goldberg et al. 1975; Goldspink and Howells 1974; MacDougall 1986). Also in some studies, muscle hypertrophy was not found even after 20 sessions of near maximal intensity strength training (Fukunaga 1976; Ikai and Fukunaga 1970). In the present study, surface EMG recorded in all subjects during the 4-wk period of imagined contraction training indicated that the trained muscle was virtually inactive during the training sessions (Fig. 4). This does not preclude the possibility that other active training may have occurred through daily living activities during the 4-wk training period despite our strict cautions against such activity. However, 1t seems highly unlikely that casual training of this type co uld ind uce mu scle hypertroph Y,Particularly after such a short period. The lack of change in evoked twitch tension in the present experiment may also be used to argue against hypertrophy (Close 1972) and is consistent with previous studies that failed to induce evoked muscle tension increases even after 5 wk of high-intensity strength training (Davies and Young 1983; McDonagh et al. 1983). Nevertheless, it is unwise to infer that unchanged or decreased evoked twitch tensions reflect an absence of muscle hypertrophy. Studies of joint immobilization have reported inconsistent changes of twitch force, or ratios of twitch force to tetanic tension, of a muscle with demonstrable atrophy (e.g., Mayer et al. 198 1; Reiser et al. 1988; Robinson et al. 199 1; Sale et al. 1982). It has been suggested that the inconsistent change in twitch force after muscle atrophy may be due to immobilization-induced changes in muscle, including modifications of mechanical properties of a muscle fiber (Sale et al. 1982 ), or alterations in speed or d uration of calcium release from the sarcoplasm iC reticulum (Reiser et al. 1988). Two possible neural mechanisms jtir strength increase

At least two mechanisms may have contributed to the increased force produced by both training groups. The hypothenar EMG increases indicate that the abduction force gain occurred from increased muscle activation (Hakkinen et al. 198 1,1985; Hakkinen and Komi 1983,1986; Komi et al. 1978; Moritani and de Vries 1979), most likely the abductor digiti minimi. Increased motoneuron activation has been acknowledged as an important source of strength increase during the first weeks of conventional strength training (Enoka 1988; Komi 1986; McDonagh and Davies 1984; Sale 1986). It also appears that some subjects developed better strategies for activating the muscles crossing the metacarpophalangeal joint, as indicated by a disproportional increase in the finger’s flexion force as compared with the abduction force increase after training. Fl exion and abduction force changes should be yoked to each other (with the relationship determ ned by the ratio of the abductor m uscles’ moment arms n each direction) 9 provided that other muscles’ activation levels remain unchanged. Rutherford and Jones ( 1986) suggested that the increased capacity to lift weights with practiced lower leg extension movements was partial1 .y due to an increase in the skill of coordinating all muscle groups involved in the movement, including those used to stabilize the body. Improved muscle coordi nation at a joint with multi ple degrees of freedom should be considered as a potential mechanism in the early so-called neural phase of strength training.

STRENGTH

CHANGES

AFTER

IMAGINED

The poor correlation between the increases in EMG and abduction force of the fifth digit may also be interpreted to indicate that some of the strength increase was due to activation of the fifth finger’s other muscles in a manner for more efficient metacarpophalangeal abduction torque production. However, such interpretations are hazardous because EMG often reflects an imperfect sampling of the neural activation of a muscle, and changes in isometric force and EMG level are not always simply related, particularly at high muscle activation levels (Howard and Enoka 199 1; for a review see Soderberg and Cook 1984). Also, it is unlikely that EMG recording during MVCs with surface electrodes on the hypothenar eminence reflect only the activity of the abductor digiti minimi. The difficulties in interpreting EMG-force relationships in the present study are illustrated by the finding that the contralateral (untrained) hand strength increases in many subjects were not accompanied by EMG increases. A definitive demonstration of the role of learning more efficient muscle activation patterns for torque production requires separately recording the activity of each muscle that crosses the joint. For abduction of the fifth digit, one must record from the fourth palmar interosseous, a primary antagonist to abduction. The size and location of this muscle makes it difficult, although not impossible to place indwelling electrodes. Strength increases ftrom changes in the motor program Because it was unlikely that the strength increases in the Imagining group were due to hypertrophy or psychological factors, the repeated imagined contractions must be considered as the factor initiating the increased motoneuron activation and improved coordination. It is reasonable to consider the nervous system levels that may have been involved in this process. At the lower levels of a hierarchically organized motor system, strength increases in general have been attributed to changes in the physiological properties of spinal motoneurons, interneurons and associated reflex pathways, and descending pathways (Hakkinen and Komi 1986; Komi et al. 1978; Milner-Brown et al. 1975; Sale et al. 1983a,b; Upton and Radford 1975) and/or the morphological properties of these neurons (Geinismann et al. 197 1; Gerchman et al. 1975; Gilliam et al. 1977; Kamen et al. 1984; Tomanek and Tipton 1967 ). The ultimate result of these changes is greater net motoneuron excitation on maximal effort contractions. However, muscle activity is expected to signal activation of these levels, and repeated executions of motor commands may be required before neural changes are induced at these lower levels of the motor system. Although powerful muscle contractions occurred during the MVC training, muscles were quiescent during the imagined training, making it unlikely that the neural changes responsible for the strength increases in the Imagining group occurred at the motor system’s executional level. With this interpretation we assume that the Imagining group did not change the frequency, duration, or level of activation of the hypothenar muscles during the 4-wk training period. In support of this assumption, we have observed similar effects in experiments involving a 5-wk period of joint immobilization that yielded measureable muscle atrophy (Yue et al. 1991).

CONTRACTION

TRAINING

1121

We suggest instead that the neural changes after the mental training occurred at programming or planning levels of the motor system, most likely involving nonprimary cerebral cortical motor areas. The altered “program” in turn may achieve strength gains via actions on spinal circuitry such as inhibition of Renshaw cells (Butler and Darling 1990), or other means of changing the excitatory and inhibitory influences on motoneurons. This interpretation is bolstered by reports that neuronal activation, presumably associated with motor programming/ planning, can occur without activating muscle (Roland et al. 1980), and that mental training of motor skills (i.e., with quiescent muscles) can improve motor performance and skill acquisition (Corbin 1972; Fetz and Landers 1983; Hall 1985; Richardson 1967). In these experiments it seems unlikely that descending pathways and spinal circuits were activated but inhibited at some point to prevent motoneuron activation. Given the similar strength increases in the Contraction and Imagining groups of the present experiment, it may be that short-term, high-intensity muscle contraction training produces only adaptations of cortical programming/ planning areas. One assumption necessary in suggesting that programming/planning changes underlie strength increases is that a maximal muscle contraction requires motor programming/planning. Although there is evidence that imagining a sequence of independent finger movements activates supplementary motor area neurons in humans (Roland et al. 1980), it is not known whether such activation occurs when performing a maximal contraction. However, Roland ( 1985) suggested that a maximal muscle contraction may require programming given the need for high effort levels. Roland and his colleagues ( 1980) reported evidence for low-level activation of the supplementary motor area during submaximal isometric contractions (Roland et al. 1980). Strength increases

Qfthe contralateral (untrained) jinger

The finding of increased strength in the hand contralatera1 to the trained hand is consistent with previous findings (cf. Enoka 1988; Sale 1986). The comparable levels of force increase in the contralateral untrained hand of the Imagining and Contraction groups prompts the speculation that changes in the motor program acquired via training of one hand may transfer to the contralateral hand. These training-induced changes may occur in areas that can influence both ipsi- and contralateral motor areas, such as the supplementary motor area. Brinkman ( 1984) observed that unilateral lesion of the supplementary motor area in monkeys produced a deficit in bimanual coordination. That is, the movements of both hands were similar even though successful task performance required very different movements of each hand. Section of the corpus callosum resulted in the disappearance of the synkinetic (“mirrored”) movements and a return to independent action of the two hands. It was suggested that in normal monkeys, the supplementary motor area may provide information to the opposite hemisphere about the intended and/or ongoing movements (Brinkman 1984).

1122

Strength increases in proximal

G. YUE AND K. J. COLE

versus distal muscles

The large strength increases observed for the Imagining group may indicate a therapeutic technique for combating the loss of strength after periods of muscle disuse from joint immobilization, peripheral nerve injuries, and the like. However, the present study focused on hand muscles, which show disproportionately large representations in the primary motor cortex. Also, the fifth digit abductor muscles may be relatively unused in the skills of daily living. Similar results must be demonstrated for proximal and often-used muscles before the potential therapeutic benefits of imaginary techniques for strength preservation can be assessed. We thank Drs. Warren G. Darling, Carl G. Kukulka, and Erich S. Luschei for valuable comments on the manuscript and James A. Hadar for clerical and editorial assistance. This research was supported by Grants YA l-9005- 1 from American Paralysis Association, 2408-95 from The University of Iowa Collegiate Association Council, and a grant in aid of research from Sigma Xi, The Scientific Research Society. Address for reprint requests: K. J. Cole, Dept. of Exercise Science, The University of Iowa, Iowa City, IA 52242. Received 20 May 199 1; accepted in final form 20 December 199 1. REFERENCES J. Strengthening muscle. Exercise Sport Sci. Rev. 9: l-73, 198 1. P. W. Clinical Mechanics qf the Hand. St. Louis, MO: Mosby, 1985. BRINKMAN, C. Supplementary motor area of the monkey’s cerebral cortex: short- and long-term deficits after unilateral ablation and the effects of subsequent callosal section. J. Neurosci. 4: 9 18-929, 1984. BROOKS, V. B. The Neural Basis qj’Motor ControZ. New York: Oxford Univ. Press, 1986. BUTLER, A. J. AND DARLING, W. G. Reflex changes accompanying isometric strength training of the contralateral limb. Sot. Neurosci. Abstr. 16: 884, 1990. CLOSE, R. I. Dynamic properties of mammalian skeletal muscles. Physiol. Rev. 52: 129-197, 1972. CORBIN, C. B. Mental practice. In: Ergogenic Aids and Muscular Performance, edited by W. P. Morgon. New York: Academic, 1972, p. 93118. DAVIES, C. T. M. AND YOUNG, K. Effects of training at 30% and 100% maximal isometric force (MVC) on the contractile properties of the triceps surae in man (Abstract). J. Physiol. Lond. 336: 3 1P, 1983. EDGERTON, V. R., ROY, R. R., GREGOR, R. J., AND RUGG, S. Morphological basis of skeletal muscle power output. In: Human Muscle Power, edited by N. L. Jones, N. McCartney, and A. J. McComas. Champaign, IL: Human Kinetics, 1986, p. 43-63. ENOKA, R. M. Muscle strength and its development. New perspectives. Sports Med. 6: 146- 168, 1988. FETZ, D. L. AND LANDERS, D. M. The effects of mental practice on motor skill learning and performance: a meta-analysis. J. Sport Psychol. 5: 25-57, 1983. FUKUNAGA, V. T. Die absolute muskelkraft und das muskelkrafttraining. Sportarzt Sportmed. 27: 255-266, 1976. GARDNER, G. Specificity of strength changes of the exercised and non-exercised limb following isometric training. Res. Q. 34: 98- 10 1, 1963. GEINISMANN, Y. Y., LARINA, V. N., AND MATS, V. N. Changes of neurones dimensions as a possible morphological correlate of their increased functional activity. Brain Res. 26: 247-257, 197 1. GERCHMAN, L. B., EDGERTON, V. R., AND CARROW, R. E. Effects of physical training on the histochemistry and morphology of ventral motor neurones. Exp. Neural. 49: 790-80 1, 1975. GILLIAM, T. B., ROY, R. R., TAYLOR, J. F., HEUSNER, W. W., AND VAN HUSS, W. D. Ventral motor neuron alterations in rat spinal cord after chronic exercise. Experientia BaseZ 33: 665-667, 1977. GOLDBERG, A. L., ETLINGER, J. D., GOLDSPINK, D. F., AND JABLECKI, C. Mechanisms of work-induced hypertrophy of skeletal muscle. Med. Sci. Sports 7: 248-26 1, 1975. ATHA, BRAND,

G. AND HOWELLS, K. F. Work-induced hypertrophy in exercised normal muscles of different ages and the reversibility of hypertrophy after cessation of exercise. J. Physiol. Land. 239: 179- 193, 1974. HAIUUNEN, K., ALEN, M., AND KOMI, P. V. Changes in isometric force and relaxation time, electromyographic and muscle fiber characteristics of human skeletal muscle during training and detraining. Acta Physiol. Stand. 125: 573-585. 1985. HAKIUNEN, K. AND KOMI, P. V. Electromyographic changes during strength training and detraining. Med. Sci. Sports Exercise 15: 455-460, 1983. HAKKINEN, K. AND KOMI, P. V. Training induced changes in neuromuscular performance under voluntary and reflex conditions. Eur. J. AppZ. Physiol. 55: 147-155, 1986. HAI&NEN, K., KOMI, P. V., AND TESCH, P. Effect of combined concentric and eccentric strength training and detraining on force time, muscle fiber and metabolic characteristics of leg extensor muscles. Stand. J. Sports Sci. 3: 50-58, 198 1. HALL, C. R. Individual differences in the mental practice and imagery of motor skill performance. Can. J. AppZ. Sport Sci. 10: 17S-2 1S, 1985. HASAN, Z., ENOKA, R. M., AND STUART, D. G. The interface between biomechanics and neurophysiology in the study of movement: some recent approaches. Exercise Sport Sci. Rev. 13: 169-234, 1985. HELLEBRANDT, F. A., PARRISH, A. M., AND HOUTZ, S. J. Cross education: the effect of unilateral exercise on the contralateral limb. Arch. Phys. Med. 28: 76-85, 1947. HOUSTON, M. E., FROESE, E. A., VALERIOTE S. P., GREEN, H. J., AND &WNEY, D. A. Muscle performance, morphology and metabolic capacity during strength training and detraining: a one leg model. Eur. J. AppZ. Physiol. 5 1: 25-35, 1983. HOWARD, J. D. AND ENOKA, R. M. Maximal bilateral contractions are modified by neurally mediated interlimb effects. J. AppZ. Physiol. 70: 306-316, 1991. IKAI, M. AND FUKUNAGA, T. A study on training effect on strength per unit cross-sectional area of muscle by means of ultrasonic measurement. Eur. J. AppZ. Physiol. 28: 173- 180, 1970. JONES, D. A. AND RUTHERFORD, 0. M. Human muscle strength training: the effects of three different training regimes and the nature of the resultant changes. J. Physiol. Lond. 39 1: l-l 1, 1987. KAMEN, G., TAYLOR, P., AND BEEHLER, P. J. Ulnar and posterior tibia1 nerve conduction velocity in athletes. Int. J. Sports Med. 5: 26-30, 1984. KOMI, P. V. Training of muscle strength and power: interaction of neuromotoric, hypertrophic, and mechanical factors. J. Sports Med. 7: lo- 15, 1986. KOMI, P. V., VITASALO, J. T., RAURAMAA, R., AND VIHKO, V. Effect of isometric strength training on mechanical, electrical and metabolic aspects of muscle function. Eur. J. AppZ. Physiol. 40: 45-55, 1978. LIBERSON, W. T., AND ASA, M. M. Further studies of brief isometric exercises. Arch. Phys. Med. 40: 330-336, 1959. LINDH, M. Increase of muscle strength from isometric quadriceps exercise at different knee angles. Stand. J. Rehahil. Med. 11: 33-36, 1979. MACDOUGALL, J. D. Morphological changes in human skeletal muscle following strength training and immobilization. In: Human Muscle Power, edited by N. L. Jones, N. McCartney, and A. J. McComas. Champaign, IL: Human Kinetics, 1986, p. 269-288. MAYER, R. F., BURKE, R. E., TOOP, J., HODGSON, J. A., KANDA, B., AND WALMSLEY, B. The effect of long-term immobilization on the motor unit population of the cat medial gastrocnemius muscle. Neuroscience 6: 725-739, 1981. MCDONAGH, M. J. N. AND DAVIES, C. T. M. Adaptive response of mammalian skeletal muscle to exercise with high loads. Eur. J. AppZ. Physiol. 52: 139-155, 1984. MCDONAGH, M. J. N., HAYWARD, C. M., AND DAVIES, C. T. M. Isometric training in human elbow flexor muscles: the effects on voluntary and electrically evoked forces. J. Bone J. Surg. Am. Vol. 65: 355-358, 1983. MEYERS, C. Effects of two isometric routines on strength, size, and endurance in exercised and non-exercised arms. Res. Q. 38: 430-440, 1967. MILNER-BROWN, H. S., STEIN, R. B., AND LEE, R. G. Synchronization of human motor units: possible roles of exercise and supraspinal reflexes. Electroencephalogr. CZin. Neurophysiol. 38: 245-254, 1975. MORITANI, T. AND DE VRIES, H. A. Neural factors versus hypertrophy in time course of muscle strength gain. Am. J. Phys. Med. 58: 115-l 30, 1979. PANIN, N., LINDENAUER, H. J., WEISS, A. A., AND EBEL, A. Electromyographic evaluation of the “cross exercise” effect.Arch. Phys. Med. RehabiZ. 42: 47-53, 196 1. GOLDSPINK,

STRENGTH

CHANGES

AFTER IMAGINED

E. I., RAWLINGS, I. L., CHEN, C. S., AND YILK, M. D. The facilitating effects of mental rehearsal in the acquisition of rotary pursuit tracking. Psychon. Sci. Sect. Hum. Exp. Psychol. 26: 7 l-73, 1972. REISER, P. J., STOKES, B. T., AND WALTERS, P. J. Effects of immobilization on the isometric contractile properties of embryonic avian skeletal muscle. Exp. Neurol. 99: 59-72, 1988. RICHARDSON, A. Mental practice: a review and discussion. Part I. Res. Q. 38: 95-107, 1967. ROBINSON, G. A., ENOKA, R. M., AND STUART, D. G. Immobilization-induced changes in motor unit force and fatigability in the cat. A4uscle Nerve 14: 563-573, 1991. ROLAND, P. E. Cortical organization of voluntary behavior in man. Hum. Neurobiol. 4: 155-167, 1985. ROLAND, P. E., LARSEN, B., LASSEN, N. A., AND SIUNHOJ, E. Supplementary motor area and other cortical areas in organization of voluntary movements in man. J. Neurophysiol. 43: 118- 136, 1980. ROSE, D. L., RADZYMINSKI, S. F., AND BEATIT, R. R. Effect of brief maximal exercise on strength of the quadriceps femoris. Arch. Phys. A4ed. 38: 157-164, 1957. RUTHERFORD, 0. M. AND JONES, D. A. The role of learning and coordination in strength training. Eur. J. Appl. Physiol. 55: 100-105, 1986. SALE, D. G. Neural adaptation in strength and power training. In: Human Muscle Power, edited by N. L. Jones, N. McCartney, and A. J. McComas. Champaign, IL: Human Kinetics, 1986, p. 289-307. SALE, D. G., MACDOUGALL, J. D., UPTON, A. R. M., AND MCCOMAS, A. J. Effect of strength training upon motoneuron excitability in man. Med. Sci. Sports Exercise 15: 57-62, 1983a. RAWLINGS,

CONTRACTION

TRAINING

1123

SALE, D. G., MCCOMAS, A. J., MACDOUGALL, J. D., AND UPTON, A. R. M. Neuromuscular adaptation in human thenar muscles following strength training and immobilization. J. Appl. Physiol. 53: 4 19-424, 1982. SALE, D. G., UPTON, A. R. M., MCCOMAS, A. J., AND MACDOUGALL, J. D. Neuromuscular function in weight-trainers. Exp. Neurol. 82: 52 l-53 1, 1983b. SODERBERG, G. L. AND COOK, T. M. Electromyography in biomechanics. Phys. Ther. 64: 18 13-l 820, 1984. TESCH, P. A., HJORT, H., AND BALLDIN, U. I. Effects of strength training on G tolerance. Aviat. Space Environ. Med. 54: 69 l-695, 1983. THOMINE, J. M. The clinical examination of the hand. In: The Hand, edited by R. Tubiana. Philadelphia, PA: Saunders, 198 1, p. 6 16-647. TOMANEK, R. J. AND TIPTON, C. M. Influence of exercise and tenecromy on the morphology of a muscle nerve. Anat. Res. 159: 105- 112, 1967. UPTON, A. R. M. AND RADFORD, P. F. Motoneuron excitability in elite sprinters. In: Biomechanics, V-A, edited by P. V. Komi. Baltimore, MD: University Park, 1975, p. 82-87. VANDELL, R. A., DAVIS, R. A., AND CLUGSTON, H. A. The function of mental practice in the acquisition of motor skills. J. Gen. Psychol. 29: 243-250, 1943. YASUDA, Y. AND MIYAMURA, M. Cross-transfer effects of muscular training on blood flow in the ipsilateral and contralateral forearms. Eur. J. Appl. Physiol. 51: 321-329, 1983. YUE, G., WILSON, S., DARLING, W. G., YUH, W. T. C., AND COLE, K. J. Effects of imaginary muscle contraction training on strength of immobilized muscle. Sot. Neurosci. Abstr. 17: 648, 199 1.