EFFECTS OF IMAGERY MOTOR TRAINING ON
TORQUE PRODUCTION OF ANKLE PLANTAR
FLEXOR MUSCLES
ABSTRACT: The aim of this study was to investigate in control subjects
the effect of imagery training on the torque of plantar-flexor muscles of the
ankle. Twenty-nine subjects were allocated to one of three groups that
performed either imagery training, low-intensity strength training, or no
training (only measurements). The low-intensity training served as an attention
control group. Plantar-flexor torques were measured before, during,
directly after, and 4 weeks after the training period. At the end of a 7-week
training program, significant differences were observed between the maximal
voluntary torque production of the imagery training group (136.3
21.8% of pretraining torque) vs. the low-intensity training group (112.9
29.0%; P 0.02) and the control group (113.6 19.2%; P 0.02). The
results of this study show that imagery training of lower leg muscles significantly
increased voluntary torque production of the ankle plantar-flexor
muscles and that the force increase was not due to nonspecific motivational
effects. Such muscle strengthening effects might be beneficial in rehabilitation
for improving or maintaining muscle torque after immobilization.
Muscle Nerve 28: 168–173, 2003
INGE ZIJDEWIND, PhD,1 SJOUKJE T. TOERING, MSc,1,2 BRAM BESSEM, MD,2
OCCO VAN DER LAAN, MD,2 and RON L. DIERCKS, PhD, MD2
Injuries of the ankle are common in sports and often
require immobilization of the ankle. Immobilization,
even for short periods of time, can result in
undesirable muscle weakness, and several experiments
(reviewed elsewhere18) have been designed to
probe strategies that could diminish the reduction in
muscle force. Recent experiments suggest that imagery
training is an interesting alternative option.15
Motor imagery is an active process during which
the representation of a specific action is internally
reproduced within the working memory, without any
motor output.7 Functional imaging studies have provided
evidence that during motor imagery similar
anatomical substrates become activated as during
motor performance,16,20,21,22 and training by motor
imagery has been shown to improve motor performance10
and muscle strength.23 To explore the use
of imagery training after ankle injures, we investigated
the effects of imagery training on torque production
of the plantar-flexor muscles in control subjects.
This study serves as a background for an
applied study in which patients with an ankle sprain
are instructed to use imagery training to diminish
plantar-flexion force loss after immobilization.
The effects of the imaginary training were compared
with two control groups; one group trained with
low-intensity contractions and one group did not train
at all. Improvement of muscle strength (hypertrophy)
by conventional strength training is known to be dependent
on the training intensity. The strength of our
low-intensity contractions was below the generally accepted
threshold needed for force improvement in the
muscle2,9,19 and, therefore, we did not expect changes
in the muscle. Thus, if the torque changes induced by
imagery training were significantly larger than the lowintensity
(control) group, this increase could not be
due to low-level muscle activation or nonspecific motivational
(e.g., attention) effects.
Abbreviations: EMG, electromyographic
Key words: imagery training; maximal voluntary muscle torque; mental training;
skeletal muscle
Correspondence to: I. Zijdewind; e-mail:
i.zijdewind@med.rug.nl
© 2003 Wiley Periodicals, Inc.
168 Imagery Training MUSCLE & NERVE August 2003
METHODS
Subjects. Twenty-nine subjects (18 women and 11
men; age range: 19–27 years) were included in this
study. Subjects were unaware of the objectives of the
study and were financially compensated for participating.
Subjects were required to have had no injuries
of their legs during the 6 months prior to starting
the experiments, and were instructed not to
change the amount of their daily physical activity
during the study. All subjects were in good health.
Five subjects were physically active less than 1 h per
week (one subject in the imaginary training group,
two subjects in the low-intensity group, and two subjects
in the control group). Nine subjects participated
for 5 or more hours per week in sports activities
(four subjects in the imaginary training group,
three subjects in the low-intensity training group,
and two subjects in the control group). Before starting
the experiments, subjects were informed about
their nature and signed a consent form to participate.
The ethics committee of the Groningen University
Hospital approved all experimental procedures.
Training Protocol. The training program lasted 7
weeks, with five sessions per week (Monday to Friday).
The imagery training was performed within the
University Hospital Groningen and was supervised
by one of the investigators. In contrast to Yue and
Cole,23 we increased the number of contractions to
50 instead of 15, but one training session still lasted
less than 25 min.
Imagery Training. The subjects (n 10, 7
women, mean age: 21.1 1.3, age range: 19–23
years) were asked to imagine that they were performing
an ankle plantar-flexion exercise. To facilitate a
mental image of the task, they were shown a video of
a person with a heavy barbell on his shoulders performing
a toe-raise exercise. The person on the
video carried a heavy barbell and moved from a
neutral standing position to standing on tiptoe; during
this movement the ankle plantar-flexors are active.
Subjects were instructed to imagine that they
were contracting their ankle flexor muscles trying to
produce the ankle flexion. The imaginary contraction
was carried out with both legs, lasted 10 s, and
was followed by a 10-s rest. The contraction was
repeated five times followed by a 30-s rest. Five series
of five contractions were followed by a 2-min rest. In
total, 50 imaginary contractions of both legs simultaneously
were performed per session. The timing of
the contractions was based on auditory instructions
and most subjects preferred to perform the contractions
with their eyes closed.
Low-IntensityTraining. This group (n 9, 6
women, mean age: 23.0 2.1, age range: 20–27
years) served as a control group. Subjects in this
group made low-intensity isometric contractions
with the same intervals as group I. Subjects sat on a
smooth pad on the floor with their knee extended
and their feet against the wall. Subjects were asked to
push against the wall; if they exerted too much force,
the pad slid away. A pilot study showed that the total
amount of force that was pushed by the subjects with
both legs simultaneously was less than 30 kg, that is
less than 50% of their body weight. Thus, per individual
leg, subjects were pushing less than 25% of
their body weight. The timing of the contractions
was given by an audiotape and was equal to the
timing of the imagery contractions. The low-intensity
training group trained at home. Before and after the
training, they had to log their heart rate and the
time of day.
Controls. Subjects (n 10, 5 women, mean age:
22.6 1.8, age range: 20–26 years) in this group did
not train, but underwent all maximal voluntary
torque measurements.
Torque Measurements. In all subjects the maximal
voluntary muscle strength of the plantar-flexors of
both legs was measured six times. The first measurement
was performed before the 7-week training program
started; 5 days after the training started, the
second measurement was performed. The following
measurements (three to five) were performed at
2-week intervals, i.e., 3, 5, and 7 weeks after the onset
of training. A final measurement was performed 4
weeks after the last training session. During the training
period, subjects had to log their physical activity.
The strength of the plantar-flexor muscles was
measured with a torque transducer (Fig. 1) as described
in detail by Hof and Van den Berg.14 In
short, a foot-plate was welded to short vertical supporters
with strain gauges. The vertical beams were
connected to horizontal beams, mounted on a base
plate. The strain gauges recorded the bending
torque in the vertical supporter. The lever arm of the
measured torque equals the distance between the
ankle axis and the metatarsophalangeal joint axis.
The subject placed one foot on the foot-plate, while
the other foot was placed on an equally high plateau.
While standing on the foot-plate, the foot of the
subjects was placed in such a way that no torque was
recorded. At this time, the malleolus was in line with
the vertical beams, and shifts in weight between the
legs did not influence the force measurements. Because
small changes in posture in the sagittal plane
Imagery Training MUSCLE & NERVE August 2003 169
could be registered as force, subjects were instructed
to stand still before the contraction. The foot was
fixed at the instep by a metal band. A piece of foam
rubber was placed between the instep and the metal
band to prevent pain during the contraction. The
band was placed at the instep as tightly as possible.
By fixating the foot, the transducer could be used as
an isometric torque recorder. However, due to the
way in which the foot was fixated, small concentric
movements could not be avoided. The torque was
measured during a maximal contraction of the ankle
plantar-flexors. Subjects were instructed to make a
maximal effort to change from the neutral position
to standing on tiptoe.
Four maximal contractions were performed. The
interval between the contractions varied between
10–30 s. The torques were recorded on paper with a
chart recorder (Rood, FB-19, Den Haag, The Netherlands),
and measured with a ruler. The highest
value was used as the maximal torque delivered by
the plantar-flexors and served as control value.
EMG Recordings. Electromyographic (EMG) recordings
of the gastrocnemius medialis were made
during the imagery training. The skin overlying the
muscle was cleaned with alcohol and two Ag–AgCl
electrodes (Graphic Controls, Buffalo, New York)
were placed on it, 3-cm apart. The reference electrode
was placed at the popliteal space. The EMG
signal was amplified (5,000 times) and filtered ( 4
kHz cutoff) and recorded on a thermic recorder
(WeKaGraph, WK-821 AX, WKK, Kaltbrunn, Switzerland).
The EMG recordings were used to monitor
muscle activity during the imagery training; if muscle
activity occurred, subjects were told to relax during
the next imagery contraction.
Analysis. The significance of the training-induced
changes between groups was assessed with a two-way
analysis (time * group) for repeated measures. Significant
interactions were further analyzed with Student
paired t-tests (between times within one
group). Significant difference between groups was
assessed with a one-way ANOVA (between groups at
a specific time), and statistical significance between
the different groups was further analyzed with a
post-hoc test according to Tukey. In the text, values
are given as means SD. A significance level of P
0.05 was used for all statistical comparisons.
RESULTS
The results of one subject in the imaginary training
group were excluded due to the development of a
foot infection during the study. All other subjects
performed all measurements of maximal torque.
None of the subjects reported missing more than
10% of the training sessions. However, in the lowintensity
training group, some training sessions were
shifted from a weekday to Saturday or Sunday.
During the imagery training sessions, some subjects
showed a little EMG activity during the training.
They were then asked to concentrate on performing
an imaginary contraction, without activating their
muscles.
No significant difference was observed, before or
after the training period, between the torque of the left
and the right ankle plantar-flexors; therefore we
pooled the data. The mean torque of the ankle plantarflexors
at the first measurement was 140.9 40.1 Nutrimuscle
for all three groups combined, similar to isometric
torque values obtained by other authors.11,17 This pretraining
measurement did not show a significant difference
between the three groups.
Figure 2 shows the change in voluntary torque of
the ankle plantar-flexors after the 7-week training
period for all subjects. Mean values for the three
groups are illustrated in Figure 3.
Statistical analysis revealed a significant interaction
between the measurements at different times
(measurement 1–6) and the three groups (P
0.001). A significant increase in plantar-flexion
torque (129.6 24.9% of pretraining torque; P
0.001) was observed in the imagery training group
after 5 training weeks, and significant differences
were found in comparison with the low-intensity
training group (111.3 24.0%; P 0.05) and the
control group (107.1 16.1%; P 0.01). At the end
of the training period, significant increases were
found in the imagery training group (136.3 21.8%
FIGURE 1. Schematic drawing of the force transducer, adapted
from Hof and Van den Berg with permission of Elsevier.14
170 Imagery Training MUSCLE & NERVE August 2003
of pretraining torque; P 0.001) and surprisingly
also in the control group (113.6 19.2% of pretraining
torque; P 0.01; low-intensity group: 112.9
29.0%). Four weeks after the training, a significant
increase in torque, compared to the pretraining
measurement, could still be observed in the imagery
training group (125.3 7.1% of pretraining torque;
P 0.001) and the control group (116.8 22.5% of
pretraining torque). A significant difference was observed
between the imagery-training group and the
low-intensity training group (103.1 30.0% of pretraining
torque; P 0.05), but no significant difference
was observed in comparison with the control
group (116.8 22.5%; P 0.03).
DISCUSSION
The results of this study showed that imagery training
may be beneficial for increasing the voluntary
force of the plantar-flexor muscles. Our results also
showed that imagery training resulted in a larger
torque increase than that of low-intensity strength
training, suggesting that the effect of the imagery
training was not due to low-level muscle activation or
nonspecific motivational training aspects.
Our control group showed a significantly smaller
but still substantial force increase compared to the
imagery training group. A force increase in a control
group has been observed also in other studies13 and
FIGURE 2. Voluntary torque production of left (open circles, n 28) and right ankle plantar-flexor muscles (crosses, n 28) before (B)
and immediately after (A) a 7-week training program. Lines connect individual pretraining and posttraining values.
FIGURE 3. Mean voluntary torque production ( SEM) as a percentage of pretraining values. Stars (*) denote significant difference
between the imagery training group vs. the low-force intensity training group. Crosses () denote significant difference between the
imagery training group and the control group.
Imagery Training MUSCLE & NERVE August 2003 171
suggests that subjects become more familiar with the
experimental arrangement after repeated measurements.
However, subjects in the low-intensity group
did not show a force increase.
Yue and Cole23 obtained similar data for the
abductor digiti minimi. They found a force increase
of about 22% after 4 weeks of imagery training. Our
data suggest a larger force increase (36%), but if we
take into account that learning effects caused our
control group also to become stronger (14%), a 20%
force increase can be attributed to the imagery training.
So far, the imagery training effects have not been
explained satisfactorily. Some theories stress the importance
of learning the cognitive components of a
motor task, whereas others propose that imagery
training generates the same central neuronal activity
patterns as the overt movement, albeit reduced in
magnitude. Small neuromuscular efferent patterns
generated during the imagery contraction would
provide kinesthetic feedback useful for optimizing
the motor pattern. If this second theory is correct, it
might be expected that some EMG activity would be
recorded during training, and indeed during the
imagery training some EMG activity was observed.
The amount of activity, however, suggested that muscles
were activated at a low intensity and the peripheral
effects of a low-intensity activation would not be
expected to differ depending on whether it is associated
with imagery activation or intended volitional
activation. The fact that our low-intensity training
groups showed a significantly smaller force increase
compared to the imagery training suggests that the
minimal activity generated does not explain all of
the force increase caused by imagery training.
Appropriate strength training results in a systematic
increase in strength of the trained muscles2,19
and there is evidence not only that the muscle fibers
themselves show an increase in force production but
also that neural mechanisms driving muscle fibers
are altered after strength training.8 Changes in the
neural control of muscles might underlie the effect
of imagery training on muscle force production, e.g.,
a change in muscle coordination or an increase in
the activation levels of the target muscles.
Muscle (Group) Coordination. An important factor
contributing to strength increase after training is the
distribution of activity between different muscles (or
muscle groups), especially between muscles with antagonist
functions. Carolan and Cafarelli5 showed
that in the early stages of training, force improvement
could result from a decline in the amount of
antagonist activation. It will be important to determine
whether our imagery training may have had
effects on the programming of agonist/antagonist
coordination. The fact that our control group
showed a significant force increase (to 116% of pretraining
values) suggests that repeated measurements
induce learning effects.
Activation Level during Maximal Voluntary Contractions.
It is difficult to activate muscles close to their
maximal evocable force by voluntary commands.12
During repeated trials in elbow flexor muscles, a
maximal force was produced in only 25% of the
contractions.1 The ease with which subjects are able
to activate a muscle maximally differs not only between
subjects but also between muscles.1,3,4 For instance,
Belanger and McComas4 showed that complete
activation of the tibialis anterior muscle was
easily achieved, whereas the plantar-flexor muscles
of the ankle were extremely difficult to activate to
near-maximal values.6 However, results with respect
to maximal activation of the plantar-flexor muscle of
the ankle are not consistent. Several studies have
revealed activation scores below 90% (sitting position4;
standing position6) whereas others have shown
activation scores of 95%.11,17 From the combined
evidence, it seems reasonable to assume that the
ankle plantar-flexor muscles are, at the very least,
difficult to activate maximally and that imagery training
could enhance the neural drive during a maximal
voluntary contraction. Unfortunately, we did not
measure voluntary neural drive during the contraction,
for example, by twitch superimposition1 before
and after training. Data from Herbert et al.13 did not
show a change in voluntary activation after imagery
or conventional training, but the activation score in
the elbow flexor muscles was already close to maximum
before training, leaving little room for improvement.
An important issue in strength training is specificity.
To induce a force increase, the training should
resemble the testing situation as closely as possible.
Care was taken that during the torque measurements
and the imagery and low-strength training,
knee and ankle angles were the same (ankle 90°
flexed, knee fully extended). However, during the
imagery contraction, subjects maintained the contraction
for 10 s, whereas during the force measurement,
subjects were allowed to relax after 2 s.
In conclusion, our results showed that imagery
training may be an effective procedure for increasing
the voluntary torque output of the plantar-flexor
muscles. This seems a promising training method for
patients unable to maintain muscle torque due, for
example, to immobilization by plaster casts24 or pain.
172 Imagery Training MUSCLE & NERVE August 2003
The significant difference between the imagery
training group and the low-intensity training group
suggest that small amounts of muscle activity, as
sometimes can be observed during imagery training,
do not explain all of the force increase seen after
such training.
The authors thank Dr. A.L. Hof for providing the force transducer.
This research was partly funded by the Junior Scientific
Masterclass of the Medical Faculty Groningen (STT), and the
School of Behavioural and Cognitive Neurosciences (Groningen,
The Netherlands).
REFERENCES
1. Allen GM, Gandevia SC, McKenzie DK. Reliability of measurements
of muscle strength and voluntary activation using
twitch interpolation. Muscle Nerve 1995;18:593– 600.
2. Atha J. Strengthening muscle. Exerc Sport Sci Rev 1981;9:1–
73.
3. Behm, DG, Whittle J, Button D, Power K. Intermuscle differences
in activation. Muscle Nerve 2002;25:236 –243.
4. Belanger, AY, McComas AJ. Extent of motor unit activation
during effort. J Appl Physiol 1981;51:1131–1135.
5. Carolan B, Cafarelli E. Adaptations in coactivation after isometric
resistance training. J Appl Physiol 1992;73:911–917.
6. Cresswell AG, Lo¨scher WN, Thorstenson A. Influence of gastrocnemius
muscle length on triceps surae torque development
and electromyographic activity in man. Exp Brain Res
1995;105:283–290.
7. Decety J, Gre`zes J. Neural mechanisms subserving the perception
of human actions. Trends Cogn Sci 1999;3:172–178.
8. Enoka RM. Neural adaptations with chronic physical activity.
J Biomech 1977;30:447– 455.
9. Enoka RM. Neuromechanical basis of kinesiology. Champaign,
IL: Human Kinetics; 1988. 303 p.
10. Feltz DL, Landers DM. The effects of mental practice on
motor skill learning and performance: a meta-analysis. J Sport
Psych 1983;5:25–57.
11. Fowles JR, Sale DG, MacDougall JD. Reduced strength after
passive stretch of the human plantarflexors. J Appl Physiol
2000;89:1179 –1188.
12. Gandevia SC. Spinal and supraspinal factors in human muscle
fatigue. Physiol Rev 2001;81:1725–1789.
13. Herbert RD, Dean C, Gandevia SC. Effects of real and imagined
training on voluntary muscle activation during maximal
isometric contractions. Acta Physiol Scand 1998;163:361–368.
14. Hof AL, Van den Berg JW. EMG to force processing III:
estimation of model parameters for the human triceps surae
muscle and assessment of the accuracy by means of a torque
plate. J Biomech 1981;14:771–785.
15. Jackson PL, Lafleur MF, Malouin F, Ricards C, Doyon J.
Potential role of mental practice using motor imagery in
neurologic rehabilitation. Arch Phys Med Rehabil 2001;82:
1133–1141.
16. Jeannerod M., Decety J. Mental motor imagery: a window into
the representational stage. Curr Opin Neurobiol 1995;5:727–
732.
17. Kawakami Y, Amemiya K, Kanehisa H, Ikegawa S, Fukunaga T.
Fatigue responses of the human triceps surae muscles during
repetitive maximal isometric contractions. J Appl Physiol
2000;88:1969 –1975.
18. Kerkhoffs GMMJ, Rowe BH, Assendelft WJJ, Kelly KD, Struijs
PAA, van Dijk CN. Immobilisation for acute ankle sprain.
Arch Orthop Trauma Surg 2001;121:462– 471.
19. Matoba H, Gollnick PD. Response of skeletal muscle to training.
Sports Med 1984;1:240 –251.
20. Porro CA, Francescato MP, Cettolo V, Diamond ME, Baraldi
P, Zuiani C, Bazzocchi M, di Prampero PE. Primary motor
and sensory cortex activation during motor performance and
motor imagery: a functional magnetic resonance imaging
study. J Neurosci 1996;16:7688 –7698.
21. Roth M, Decety J, Raybaudi M, Massarelli R, Delon-Martin C,
Segebarth C, Morand S, Gemignani A, De´corps M, Jeannerod
M. Possible involvement of primary motor cortex in mentally
simulated movement: a functional magnetic resonance imaging
study. Neuroreport 1996;7:1280 –1284.
22. Sirigu A, Cohen L, Duhamel JR, Pillon B, Dubois B, Agid Y,
Pierrot-Deseilligny C. Congruent unilateral impairment for
real and imagined hand movements. Neuroreport 1995;6:
997–1001.
23. Yue G, Cole KJ. Strength increases from the motor program:
comparison of training with maximal voluntary and imagined
muscle contraction. J Neurophysiol 1992;67:1114 –1123.
24. Yue GH, Wilson SL, Cole KJ, Darling WG, Yuh WTC. Imagined
muscle contraction training increases voluntary neural
drive to muscle. J Psychophysiol 1996;10:198 –208.
Imagery Training MUSCLE & NERVE August 2003 173
