ELECTROMYOGRAPHIC ANALYSIS OF UPPER BODY,

LOWER BODY, AND ABDOMINAL MUSCLES DURING

ADVANCED SWISS BALL EXERCISES
PAUL W.M. MARSHALL1

AND

IMTIAZ DESAI2

School of Biomedical and Health Science, University of Western Sydney, Sydney, Australia; and 2Department of Sport and
Exercise Science, University of Auckland, Auckland, New Zealand

ABSTRACT
Marshall, PWM and Desai, I. Electromyographic analysis
of upper body, lower body, and abdominal muscles during
advanced Swiss ball exercises. J Strength Cond Res 24(6):
15371545, 2010Although there is now some evidence
examining the use of a Swiss ball during core stability and
resistance exercises, this has commonly been performed using
basic or isometric exercises. There is currently no evidence
examining more advanced Swiss ball exercises. The purpose
of this study was to determine whether or not muscle activity
measured during advanced Swiss ball exercises was at an
approximate intensity recommended for strength or endurance
training in advanced, or novice individuals. After a familiarization
session, 14 recreationally active subjects performed 6 different
advanced Swiss ball exercises in a randomized order. The
primary dependent variables in this study were the activity
levels collected from anterior deltoid, pectoralis major, rectus
abdominis (RA), external obliques, lumbar erector spinae,
vastus lateralis (VL), and biceps femoris using surface electromyography. All signals were normalized to maximal voluntary
isometric contractions performed before testing for each muscle. The results of this study showed that the Swiss ball roll
elicited muscle activity in triceps brachii (72.5 6 32.4%) and VL
(83.6 6 44.2%) commensurate with the intensity recommended for strength exercises in advanced trainers. Rectus
abdominis activity was greatest during the bridge exercise
(61.3 6 28.5%, p # 0.01). This was the only exercise to elicit
RA muscle activity commensurate with a strength training
effect. The remainder of the exercises elicited abdominal activity
that would require a higher number of repetitions to be performed for an endurance training adaptation. Although this study
has provided evidence for one advanced Swiss ball exercise
providing a significant whole-body stimulus, the practical
Address correspondence to Dr. Paul W.M. Marshall, p.marshall@uws.
edu.au.
24(6)/15371545
Journal of Strength and Conditioning Research
2010 National Strength and Conditioning Association

difficulty and risks of performing these more complicated

Swiss ball exercises may outweigh potential benefits.

KEY WORDS surface electromyography, exercise intensity,

strength training, labile surface
INTRODUCTION

he Swiss ball is an unstable training device used

to increase the difficulty of various bodyweight
and traditional free-weight resistance exercises (3).
An English physiotherapist by the name of Mary
Quinton is cited as having first started to use Swiss balls with
children who had cerebral palsy in 1958 (33). A Swiss
therapist then introduced the idea of using these balls to
Dr. Susanne Klein Vogelbach, who started using them
with her physiotherapy students and clients with orthopedic
problems. Dr. Vogelbach later published a book detailing the
Swiss ball exercises she developed over her years in clinical
and teaching practice (23). Recently, research has investigated the use of Swiss balls during exercise using electromyography to quantify the activity of the various muscle
groups involved. Some of the greatest interest in the use of
Swiss balls is in the application during body-weight exercises
that involve no added external resistance. This is widely
accepted as being one of the defining modes for a type of
training known as core stability exercise. The description of
a core stability exercise relates to how a movement, usually
involving bodyweight only for resistance, provides a training
stimulus for the trunk musculature. Although evidence exists
regarding the use of the Swiss ball for increasing trunk muscle
activity during core stability exercises compared with stable
surface movements (27,42), these are commonly static or
simple tasks that do not use the ability of the ball to roll (17).
Moreover, evidence suggesting that Swiss ball exercises
are less effective than conventional resistance exercises for
trunk muscle activity only use basic movements such as
the quadruped, pelvic thrust, and back extension movements
for comparison (32). There is no evidence examining some of
the more difficult Swiss ball exercises that are observed in
the recreational gymnasium environment. Therefore, we are
unsure whether these more complex movements that take
VOLUME 24 | NUMBER 6 | JUNE 2010 |

1537

Advanced Swiss Ball Exercises

advantage of the Swiss balls design and material properties
elicit muscle activity greater than that which has been
reported for conventional resistance exercises such as the
deadlift or squat.
When surface electromyographic (EMG) recordings are
rectified and smoothed, the amplitude has been shown to
be positively and linearly related to isometric force output
(1,9,14, 25). It has been suggested that EMG activity in excess
of 60% of a maximal voluntary isometric contraction (MVIC)
is required for strength training, with endurance benefits
resulting from exercise intensities below 25% (2,5,43). The
highest recorded trunk muscle EMG levels for Swiss ball
exercises we currently have information on are generally of
a moderate intensity. Abdominal activity of approximately
50% has been measured for Swiss ball curl-ups (42) and for
bilateral isometric leg holding (28). Activity of the back
extensors while using a Swiss ball ranges between 19.5 and
45.5% (15,32). These intensities may be appropriate for
eliciting a strength training effect in novice individuals (37)
and possibly an endurance effect if more repetitions are performed by advanced individuals (12). In contrast to the Swiss
ball research we are currently aware of, trunk muscle activity
during conventional resistance exercises have been shown to
be appreciably greater.
Recent evidence has found that resistance exercises performed with moderate intensities (ca. 50% of a 1 repetition
maximum) elicit greater trunk muscle activity than performing
basic Swiss ball exercises (32). Other research investigating
variations of the deadlift technique with moderate load
(12 repetition maximum) reported trunk and leg muscle
activity in excess of 60% MVIC (18). Although this research
clearly suggests that moderately loaded resistance exercises
elicit trunk and lower body muscle activity greater than that
elicited in the current body of Swiss ball research, advanced
Swiss ball exercises have not currently been evaluated.
Therefore, the purpose of this study was to measure
normalized muscle activity using surface EMG from
abdominal, lumbar, and upper and lower body musculature
during supposedly more difficult Swiss ball exercises. This
will allow comparison of these exercises to what has been
previously measured in the literature for Swiss ball exercises,
and comparison to muscle activity levels reported for conventional resistance exercises. The hypothesis of this study
is that all of the advanced Swiss ball exercises measured
will elicit trunk muscle activity in excess of 60% MVIC and
that exercises involving an upper or lower body focus will also
elicit muscle activity in those regions commensurate with
a strength training effect.

METHODS
Experimental Approach to the Problem

A within-subject cross-sectional experiment was performed

to examine normalized muscle activity using surface EMG
during several advanced Swiss ball exercises. No comparison
to a stable surface was performed because these exercises

1538

the

TM

Journal of Strength and Conditioning Research

could only be performed using a Swiss ball. Subjects

participated in 2 sessions. The first was a familiarization
session where the subjects were trained in how to perform the
specific exercises by an experienced practitioner. Subjects
were not provided with a Swiss ball to continue practicing
these exercises, and all individuals were asked to refrain
from attempting these movements in their personal training
sessions. One week later, subjects attended the testing session
where 3 repetitions of each exercise were performed while
surface EMG was continuously recorded. Before testing,
MVICs were performed for each muscle to normalize the
measured activity during testing. The order of exercises was
randomized among subjects.
Subjects

A power calculation was performed using the results of

a previous study for the difference in triceps EMG between the
Swiss ball and a stable surface for a push-up exercise because
this most closely resembles one of the starting positions used in
this study (28). The calculation of the sample size was carried
out with a = 0.05 (5% chance of type I error), 1 2 b = 0.80
(power 80%), and a calculated effect size of d = 1.41. This
provided a sample size of n = 14 for this study.
Fourteen healthy, recreationally active subjects volunteered
to participate in this study after providing informed written
consent (7 men and 7 women, aged 24.1 6 1.7 years; height
1.74 6 0.08 m; and weight 72.9 6 13.1 kg). The University
Human Participants Ethics Committee approved all procedures used in this investigation. All subjects had been performing regular physical activity on at least 3 days per week
for the last 3 months including both aerobic and resistance
training modalities. Although some of these subjects had
experience using a Swiss ball for exercise, no subject reported
regular training on a Swiss ball or familiarity with any of
the exercises tested in this study. No subject reported current
or recent participation in a competitive, organized sports
competition. Subjects reported that they were not taking
performance-enhancing stimulants at the time of testing
and had no musculoskeletal injuries or disorders. Subjects
were instructed to refrain from any resistance or anaerobic
exercise and were required to maintain normal dietary habits
in the 24 hours before the testing session. Subjects were
required to present to testing in a 2-hour postprandial state.
Procedures

Electromyographic Measurement and Analysis. After careful

skin preparation using disposable razors to remove excess
hair, fine sandpaper, and isopropyl alcohol swabs to reduce
electrode impedance to below 5 kV (measured using a digital
multimeter), pairs of Red Dot silver/silver-chloride electrodes
(3M, St. Paul, MN, USA) with a 3-cm center-to-center
distance were applied to the following muscles on the right
hand side of the body only, aligned in a parallel arrangement
to the muscle fibers. Rectus abdominis (RA), internal obliques
(IO), erector spinae (ES), pectoralis major (PM)clavicular
placement, anterior deltoid (AD), lateral head of triceps

the

TM

Journal of Strength and Conditioning Research

| www.nsca-jscr.org

brachii (TB), vastus lateralis (VL), and biceps femoris (BF)

(13). If the measured impedance was above 5 kV, the
electrodes were removed, and preparation procedures
were performed again. The most common sites that involved
additional preparation were the abdominal sites in the male
subjects owing to not removing sufficient hair in the first
preparation. Crosstalk is an issue to address in surface EMG
recordings. To minimize crosstalk, each pair of electrodes
was placed according to previous recommendations for ideal
anatomical placement (13) to ensure that electrodes were
well within the borders of the target muscles. This included
measurements to relevant bony landmarks and other
prominences to maintain consistency of placement among
subjects. The IO site that was inferior and medial to the
anterior superior iliac spine has previously been validated
for anatomical placement (26).
Electromyographic signals were recorded using a Grass
Instruments data acquisition board (Grass-Telefactors, West
Warwick, RI, USA; common mode rejection ratio of 90 dB at
60 Hz; input impedance 100 MV, 26-dB band pass roll off at
10 and 1,000 Hz) at 2,000 Hz with 16-bit analog to digital
conversion into a Pentium IV computer. Data collection and
analysis was conducted using LabVIEW (National Instruments Corporation, Austin, TX, USA). All collected signals
were subsequently band pass filtered (between 10 and
500 Hz), then rectified and smoothed by using a root mean
square (RMS) calculation with a 50-millisecond sliding
window (19). The basis of the data analysis from the RMS
signal was identification of the greatest average 1-second
RMS from each muscle signal collected during each exercise
task. This was normalized to the MVIC measured for each
muscle before testing.
Procedures used for the MVIC for each muscle have
previously been presented (22,28). After familiarization with
the technique required, 2 MVICs were performed for each
muscle with at least 2 minutes of rest between trials (14).
The greatest RMS calculated from either trial was used as
the MVIC for the muscle. The MVIC procedures cannot be
expected to only elicit activity from the muscle of interest.
However, each movement was a prime movement for the
target muscle, performed at the midpoint of its length to
obtain the greatest amount of recruitment for normalization
of exercise intensities.

and crunch exercises, a training mat was used to beneath the

participant only. Three trials were collected for each exercise
with a 1-minute break between each trial. The normalized
RMS for the 3 collected trials was averaged for each subject
to provide the muscle activity for each site for each exercise.
For all exercises, either a 55- or 65-cm diameter Swiss ball
was used depending on the height of each subject. The ball
was chosen based on whether an individual could lie prone
with their abdomen on the ball with their hands on the
ground directly underneath the shoulders and their spine in
a relatively neutral position.

Exercise Procedures

Hold and Crunch

Six different exercises were studied (Figure 1). However,

because the starting point for the Praying Mantis exercise
represents a position commonly used for exercise, this was
also measured to provide muscle activity recorded in a static
hold position as a representation for the type of exercises
previously reported in the literature. Therefore, 7 different
exercises were analyzed. The same solid surface was used
beneath the Swiss ball for all tests. This was a laboratory
nonslip surface that allowed the ball to roll but did not slip
and potentially be unsafe for testing. For the bridge and hold

Individuals were required to lie on the floor with their legs

flexed to 90 and their shoulders flexed to be placed on the
floor behind them. The Swiss ball was then placed between
their ankle joints. Recording commenced once the ball was
placed. The individual was required to hold the ball in place
for 1 second, then crunch up with a controlled movement
while maintaining full elbow extension but extending their
shoulders to take the ball out from between their legs, then
slowly return to the starting position with the ball in their
hands and legs flexed to 90. This was one repetition.

Prone Hold and Praying Mantis

The prone hold position was a 4-second isometric contraction performed with the individual supporting themselves
on the Swiss ball with their forearms flat on the surface,
and their shoulders flexed to 90. The Praying Mantis exercise
was performed separately to the isometric contractions but
using the hold as a starting position. For the Praying Mantis
exercise, the subject was required to rotate the ball 360
clockwise, then 360 anticlockwise by moving the shoulder
girdle only (although adjustments in whole body posture
occurred these were not the focus of initiating ball movement). The movement was instructed to be performed at a
natural, self-selected speed. The ball was maintained underneath the subject throughout the movement. The position of
the hands (linked via fingers) was used to indicate the 0
starting point for the movement, and this was used to guide
the individuals eyes for the degree of rotation. One trial
represented the full rotation in each direction.
Single Leg Squat

The ball was placed so that the lowest point of contact

was just superior to the ES electrodes (L5S1 location).
Individuals were required to slowly lower themselves to 90
of knee flexion, pause for 1 second, then stand back up. The
duration of individual trials never exceeded 4 seconds.
The contralateral thigh was required to be flexed to 90,
and the shank left free although no ground contact was
allowed. Foot position was marked on the floor to ensure
consistency between trials. Foot position was based on the
individual being able to reach 90 of knee flexion at the
bottom of the squat with 90 of hip flexion. Three trials were
collected for the right and left leg squats. The greatest activity
was always measured for the right leg squat.

VOLUME 24 | NUMBER 6 | JUNE 2010 |

1539

Advanced Swiss Ball Exercises

Figure 1. Exercises performed in this study.

Bridge

Individuals were required to kneel on the floor with the ball

directly in front of them and their hands placed on the ball.
They were instructed to push the ball away from themselves
as far as possible while maintaining contact with the surface.
This involves lowering the trunk toward the ground and
flexing the shoulder until. The initial movement involved
alternating hands pushing the ball out, until both hands were
required for the final push to attain the fully extended bridge
(pivoting from the knees to allow maximal extension of
the body), which was maintained for 1-second. The hands

1540

the

TM

Journal of Strength and Conditioning Research

were maintained on the superior aspect of the ball to ensure

the subject was able to return to the starting position by
extending the shoulder downwards into the surface. This was
one trial. No trial exceeded 5 seconds in duration.
Hip Extension

Each individual was required to assume a full roll-out

position with the ventral aspect to the feet only in full contact
with the ball and the ankle plantarflexed, with the hands
placed on the ground directly beneath the shoulders. The
individual was required to extend the hip only. The distance

the

TM

Journal of Strength and Conditioning Research

of hip extension was based on limiting any torso rotation
(usually pivoting about the leg still on the ball) to aid greater
hip displacement. This ensured that only a hip movement
was tested. The hip movement was at a self-selected cadence.
At the limit of extension each individual attained, an isometric
contraction was maintained for 2 seconds, and then the leg
was lowered back onto the ball. This was one trial. No trial
exceeded 4 seconds in duration. Three repetitions were
performed for the right and left legs.
Roll

Each individual was required to lie supine on the ball with

the thoracic spine supported by the surface and their hands
linked by the fingers and placed on the xiphoid notch. This
movement involved rotation of the whole body to induce
rotation of the ball. To move the ball left, the right side of the
trunk is initially rotated into the ball, maintaining a relatively
rigid segment. The right leg is then required to be moved in
line with the trunk, and then the whole body is rotated into
a prone position with the right leg moving underneath the
supporting left leg. From this position, the individual was
required to rotate back to the starting position by initiating
movement of the right side of the body again into the ball.
Each trial consisted of moving from supine to prone to supine,
then repeating for the opposite side of the body (move the ball
rightleft side of body initiates movement into the ball). Trials
were approximately 6 seconds in duration.
Statistical Analyses

The Statistical Package for the Social Sciences (SPSS Inc, v16.1,
Chicago, IL, USA) was used for analysis. Descriptive statistics
were calculated for all exercises for each muscle. A one-way
analysis of variance (ANOVA) was used to identify differences
between exercises for each muscle. Gender was entered as

| www.nsca-jscr.org

a covariate in the analysis. If the main effect of the ANOVA was

significant, repeated contrast statistics and pairwise comparisons were used to identify where the differences were. The
level of significance for all data was p # 0.05. Unless otherwise
stated, all data are presented as mean 6 SD.

RESULTS
Gender was not a significant covariate in the analysis of
normalized EMG levels. The average normalized EMG levels
for each muscle during the exercise tasks are presented in
Tables 13.
Abdominal Muscle Activity

Rectus abdominis activity was greatest during the Swiss ball

bridge (Table 1; p # 0.01). This was identified at the terminal
point of the bridge. The praying mantis, hold and crunch,
and roll exercises were not different from each other but
had greater activity than the remaining exercises (p # 0.05).
Prone holds were different from hip extensions (p # 0.05),
which were different from the single leg squats (p # 0.001)
that had the lowest RA activity (2.1 6 2.5%) of all exercises.
Internal oblique activity was greatest during the Swiss ball
roll exercise (p # 0.001). Swiss ball bridging, praying mantis,
and hip extension IO activity were not different but were
greater than the remaining exercises (p # 0.05).
Erector spinae activity was greatest during the Swiss ball
roll exercise (p # 0.001). Activities during hip extensions and
single leg squats were not different but were significantly
greater than the remaining exercises (p # 0.05).
Upper Body Muscle Activity

Pectoralis major activity was greatest during Swiss ball roll,

praying mantis, and bridge exercises (Table 2; p # 0.001).
Activity during hip extensions was different from the remaining
exercises only (p # 0.001).
Anterior deltoid activity was
greatest during Swiss ball roll and
hip extension exercises (p #
TABLE 1. Normalized muscle activity (% maximal voluntary isometric contraction) from
0.001). Activity during the praythe abdominal muscles during the exercises performed in this study.*
ing mantis was different from the
remaining exercises only (p #
Exercise
RA
IO
ES
0.001).
Swiss ball bridge
61.3 6 28.5
20.2 6 9.3
2.8 6 1.1
Triceps brachii activity was
Swiss
40.1 6 21.3
14.0 6 8.2
6.6 6 5.8
greatest during the Swiss ball
Swiss ball hold and crunch
34.0 6 12.4
9.5 6 3.2
2.8 6 1.8
roll exercise (p # 0.001). ActivSwiss ball rolls
30.2 6 13.5
44.6 6 21.3
54.3 6 28.7
k
ity during praying mantis and
Prone hold
23.2 6 11.6
6.5 6 3.6
1.9 6 1.4
bridge exercises was different
Swiss ball hip extension
13.8 6 4.9{
15.0 6 7.8
13.7 6 6.2
Swiss ball single leg squat
2.1 6 2.5
7.3 6 5.8
10.0 6 7.3
from the remaining exercises
(p # 0.001). Activity during hip
*RA = rectus abdominis; IO = internal obliques; ES = erector spinae.
extensions and the prone hold
Values reported are mean 6 SD.
For RA, IO, and ES, the identified exercise was significantly different from all other
was greater than during the
exercises (p # 0.05).
hold and crunch and single leg
These identified exercises were not different from each other, but were different from the
squat (p # 0.001).
remaining exercises (p # 0.05).
k
For RA, this exercise was different from the remaining exercises only (p # 0.05).
{For RA, this exercise was different from the remaining exercise only (p # 0.05).

Lower Body Muscle Activity

Vastus lateralis and BF activities

were greatest during the Swiss
VOLUME 24 | NUMBER 6 | JUNE 2010 |

1541

Advanced Swiss Ball Exercises

TABLE 2. Normalized muscle activity (% maximal voluntary isometric contraction) from

the upper body muscles during the exercises performed in this study.*
Exercise

Pectoralis major

Swiss ball rolls

Swiss ball praying mantis
Swiss ball bridge
Swiss ball hip extension
Swiss ball hold and crunch
Prone hold
Swiss ball single leg squat

40.7 6 25.6
30.5 6 18.1
27.0 6 8.8
13.8 6 4.6
7.7 6 4.3
5.8 6 4.1
1.2 6 1.4

Anterior deltoid
37.7
16.7
5.5
29.0
7.4
4.1
1.1

6 16.4
6 9.4
6 4.6
6 16.3
6 5.4
6 5.2
6 1.3

Triceps brachii
72.5
27.1
28.0
14.4
1.6
12.4
1.5

6 33.4
6 15.7
6 10.1
6 3.1
6 2.1
6 5.9
6 0.8

TABLE 3. Normalized muscle activity (% maximal

voluntary isometric contraction) from the lower body
muscles during the exercises performed in this
study.*
Exercise
Swiss ball rolls
Swiss ball hip
extension
Swiss ball praying
mantis
Swiss ball single
leg squat
Prone hold
Swiss Ball hold
and crunch
Swiss ball bridge

VL

BF

83.6 6 44.2 53.6 6 27.9

27.1 6 13.2 20.6 6 12.9
25.3 6 8.5

10.6 6 4.6k

22.2 6 11.2

3.6 6 5.5

11.2 6 6.4
8.1 6 6.9

2.2 6 1.0
2.1 6 0.8

3.1 6 5.8

5.6 6 5.7

*VL = vastus lateralis; BF = biceps femoris

Values reported are mean 6 SD.
For VL and BF, the Swiss ball roll had significantly
higher activity compared to all other exercises (p # 0.05).
For VL, these exercises were not different from each
other but were greater than the remaining exercises. For
BF, this exercise was different from all remaining exercises
(p # 0.05).
k
For BF, this exercise was different from the remaining
exercises only (p # 0.05).

1542

the

TM

Journal of Strength and Conditioning Research

DISCUSSION

This study has measured the

activity of upper body, abdominal and lower body muscles
during several more advanced
Swiss ball exercises. It is important to note that no falls or
injuries were sustained during
testing despite the difficult nature of the exercises. This is probably owing to the use of
a familiarization session before testing. The speed roll exercise
was an especially difficult exercise to learn and perform owing
to the combined rotation of the body and ball. It is important
to note that these were recreationally active participants who
had a good level of physical fitness and strength. Owing to the
complicated nature of these exercises, caution should be taken
when prescribing these movements for the untrained
population.
A further consideration for the results of this study is that
the EMG for each exercise was analyzed from the greatest
1-second average activity during each trial. This may
have been during dynamic or isometric contractions for the
different exercises. This activity was then normalized
to MVICs, as is the standard procedure used in all studies
examining muscle activity during core stability exercises
(6,7,1417,21,27,28,42). This may have introduced an overestimation of the activity of some exercises where significant
concentric actions are involved. It is expected that the
greatest EMG will be measured during the concentric phase
compared with isometric or eccentric actions (14). Some
exercises, such as the praying mantis and roll, cannot incorporate a focused isometric component. It is important to note
that some of the highest EMG levels measured in this study
(VL during the roll and RA during the bridge) were measured
during relatively isometric contractions. Despite the possibility that relative activity levels are overestimated, we believe
that the measured patterns of activity accurately represent
the difficulty of each exercise.
Another consideration, which is normal for any EMG
study, is the considerably large SDs of the data. This suggests
that in the prescription of these exercises, some individuals
may find that they are more or less difficult than the average
activity indicated here. Finally, core stability of the spine is
a complex issue defined as combining the musculature, the

*Values reported are mean 6 SD.

For pectoralis major, anterior deltoid, and triceps brachii, the identified exercises were
significantly different from all other exercises only (p # 0.05).
These exercises were different from the remaining exercises only (p # 0.05).
For triceps brachii, these exercises were not different from each other but were
significantly greater than the remaining exercises (p # 0.05).

ball roll (Table 3; p # 0.001). The high level of activity is

associated with the right leg being the primary support point
with a stable contact when the left side of the body is being
rotated. For VL, activity during the hip extension, praying
mantis, and single leg squat was greater than during the
remaining exercises only (p # 0.001).

For BF, activity during hip

extension was greater than during the remaining exercises (p #
0.001). This activity was measured in the test leg that was
held in isometric extension.
Biceps femoris activity during
the praying mantis was different from the remaining exercises (p # 0.01).

the

TM

Journal of Strength and Conditioning Research

passive skeletal structures, and the central nervous system as
the control unit (34,35). This study presents the acute muscle
activity levels in an isolated session only. Therefore, individuals should not assume what the effects of performing the
type of exercise measured in this study may be on maladaptive core muscle recruitment patterns, or strength and
endurance characteristics, until appropriate training studies
are performed.
We have found limited support for our experimental
hypotheses that all the advanced Swiss ball exercises tested
here would elicit muscle activity levels commensurate with
recommendations for a strength training effect. The majority
of exercises studied do not differ appreciably to the intensity
levels described for Swiss ball exercises in previous literature,
which are of a low to moderate level. Furthermore, there was
little evidence presented in this study to suggest that advanced
Swiss ball exercises can elicit greater trunk, upper body, or
lower body muscle activity levels than moderately loaded
conventional resistance exercises such as the bench press,
squat, and deadlift (11,18,20,32). The Swiss ball roll movement clearly stands out as a significantly difficult exercise
with potential to elicit strength training effects. Activity levels
of 83.6 6 44.2% for VL and 72.5 6 33.4% for TB indicate that
this exercise is at an appropriate intensity for these muscles,
which may elicit strength effects in advanced trainers (36,37).
The highest recorded erector spinae activity (54.3 6 28.7%)
was also measured during the roll exercise. Remaining upper,
lower, and abdominal muscle activity during the roll exercise
ranged from 30 to 55%. This indicates that for most individuals, a higher number of repetitions of the Swiss ball roll
exercise might be able to elicit endurance adaptations (12),
whereas for some untrained individuals, this exercise may be
appropriate as a strength training stimulus for multiple muscles.
Resistance exercises stressing multiple muscle groups have
been shown to elicit the greatest acute metabolic response
(8,38,41). Increased metabolic demand is an important factor
for the adaptations in muscle associated with strength and
endurance. Therefore, the Swiss ball roll appears to be the
most likely exercise from the current study to apply and
investigate in clinical and research contexts for significant
muscular adaptation.
It is surprising given the widespread advocacy of Swiss ball
exercises as providing a significant abdominal stimulus that
apart from the bridge none of these advanced exercises had
levels of activity where one could reasonably justify prescribing them for a trunk or core strengthening program.
The Swiss ball bridge had RA activity at an approximate level
for strength training in untrained individuals and possibly
some more advanced trainers (61.3 6 28.5%). The results
from this study are commensurate with those of previous
research, which is yet to identify any Swiss ball exercise as
being able to elicit trunk muscle activity in the range of
a strength training stimulus (15,17,24,27,28,32). Although
there are probably more difficult Swiss ball exercises that
could be performed to elicit higher trunk muscle activity

| www.nsca-jscr.org

levels, the complexity of the movements begins to make this

exercise modality unfeasible for the majority of individuals,
especially when greater trunk muscle activity levels can be
achieved using relatively basic, moderately loaded resistance
exercises such as the squat and deadlift. Support for basic
resistance exercise movements eliciting high levels of trunk
muscle activity has also been provided with recent evidence
showing that maximal isometric shoulder movements, such
as bilateral shoulder extension, elicit trunk muscle activity
levels greater than performing maximal isolated trunk exertions (40). The Swiss ball roll, although obviously providing
the highest overall muscle activity for the body, was the
most complicated movement to teach. Although no acute
accidents happened during testing, the likelihood of an incident occurring with the use of an unstable Swiss ball during
complex whole-body exercises, or an injury from prolonged
exposure to this type of difficult movement, cannot be discounted. This raises the issue for the practicality of prescribing Swiss ball exercises, when the majority of research is
providing a case against its use as a beneficial training surface.
Recent intervention studies have not found significant
positive evidence to recommend use of the Swiss ball over
other modalities of exercise. Stanton recently found that a
6-week Swiss ball training program did not significantly
improve running fitness, economy, or posture, compared with
previous research the author cited using conventional
resistance training, which positively influences these variables
(39). A study that randomized individuals with chronic low
back pain to either supervised Swiss ball exercise or an
unsupervised control exercise advice group (no Swiss ball
exercise) found no difference between interventions for
changes in primary disability and pain outcome measures
at the long-term follow-up (30,31). Other research has
investigated the Swiss ball as a support surface during
resistance exercises and shown little practical use. Some
research has found reductions in maximal isometric force
output using a Swiss ball as a support surface (4,10), whereas
other research has found no changes in prime mover muscle
activity during the bench press (20,29). In conclusion, this
study has provided little evidence to support the hypothesis
that advanced Swiss ball exercises can elicit muscle activity
commensurate with recommendations for strength training.

PRACTICAL APPLICATIONS
This is the first study we know of to measure muscle activity
during more advanced Swiss ball exercises that are observed
in the recreational training environment. If the goal of
strength and conditioning coaches is to increase the strength
of the upper body, lower body, or trunk musculature, it seems
that advanced Swiss ball exercises will be no more beneficial
than moderately loaded resistance exercises. Moreover, the
only Swiss ball exercise to achieve a high level of muscle
activity was the most complicated movement to perform.
When compared with relatively basic to teach and perform
resistance exercises such as shoulder extensions, squats, and
VOLUME 24 | NUMBER 6 | JUNE 2010 |

1543

Advanced Swiss Ball Exercises

deadlifts, the use of a Swiss ball seems redundant. Additionally, the benefit of conventional resistance training is that
increasing the external load will increase the activity of muscle
groups of interest, while allowing for periodization and
progression of the training dose over time, whereas a Swiss
ball based program appears to require more complex and
difficult movements for ongoing progression. Advanced Swiss
ball exercises should be considered a novelty movement that
could be introduced in small amounts to alleviate staleness
with long-term training. Coaches should be advised to learn,
and then prescribe, large conventional multijoint exercises for
the greater benefits they provide rather than use complicated
circus like movements on a Swiss ball. It must be considered
that the practical difficulty and potential risks involved with
performing an advanced Swiss ball exercise, such as the roll in
this study, or standing on a Swiss ball with weights as
observed in some recreational trainers, probably outweigh
any benefits.

ACKNOWLEDGMENTS
The authors have no professional relationships with any
company or manufacturer who may benefit from the results of
the current study. The results of the current study do not
constitute endorsement of the product used by the authors or
the National Strength & Conditioning Association.

REFERENCES
1. Alkner, BA, Tesch, PA, and Berg, HE. Quadriceps EMG/force
relationship in knee extension and leg press Med Sci Sports Exerc
32: 459463, 2000.
2. Anderson, EA, Ma, Z, and Thorstensson, A. Relative EMG levels in
training exercises for abdominal and hip flexor muscles. Scand J
Rehabil Med 30: 175183, 1998.
3. Anderson, K and Behm, DG. Trunk muscle activity increases with
unstable squat movements. Can J Appl Physiol 30: 3345, 2005.
4. Anderson, KG and Behm, DG. Maintenance of EMG activity and loss
of force output with instability. J Strength Cond Res 18: 637640, 2004.
5. Arokoski, JPA, Kankaanpaa, M, Valta, T, Juvonen, I, Partanen, J,
Taimela, S, Lindgren, KA, and Airaksinen, O. Back and hip extensor
muscle function during therapeutic exercises. Arch Phys Med Rehabil
80: 842850, 1999.

Muscular adaptations in response to three different resistancetraining regimens: Specificity of repetition maximum training zones.
Eur J Appl Physiol88: 5060, 2002.
13. Cram, JR and Kasman, GS. Introduction to Surface Electromyography.
Gaithersburg: Aspen Publishers Inc., 1998.
14. De Luca, CJ. The use of surface electromyography in biomechanics.
J Appl Biomech. 13: 135163, 1997.
15. Drake, JDM, Fischer, SL, Brown, SHM, and Callaghan, JP. Do
exercise balls provide a training advantage for trunk extensor
exercises? A biomechanical evaluation. JMPT 29: 354362, 2006.
16. Ekstrom, RA, Donatelli, RA, and Carp, KC. Electromyographic
analysis of core trunk, hip, and thigh muscles during 9 rehabilitation
exercises. J Orthop Sports Phys Ther 37: 754762, 2007.
17. Ekstrom, RA, Osborn, RW, and Hauer, PL. Surface electromyographic analysis of the low back muscles during rehabilitation
exercises. J Orthop Sports Phys Ther 38: 736745, 2008.
18. Escamilla, RF, Francisco, AC, Kayes, AV, Speer, KP, and
Moorman, CT. An electromyographic analysis of sumo and
conventional style deadlifts. Med Sci Sports Exerc 34: 682688, 2002.
19. Farina, D and Merletti, R. Comparison of algorithms for estimation
of EMG variables during voluntary isometric contractions.
J Electromyogr Kinesiol 10: 337349, 2000.
20. Goodman, CA, Pearce, AJ, Nicholes, CJ, Gatt, BM, and
Fairweather, IH. No difference in 1RM strength adn muscle
activation during the barbell chest press on a stable and unstable
surface. J Strength Cond Res 22: 8894, 2008.
21. Kavcic, N, Grenier, S, and McGill, SM. Determining the stabilizing
role of individual torso muscles during rehabilitation exercises. Spine
29: 12541265, 2004.
22. Kavcic, N, Grenier, S, and McGill, SM. Quantifying tissue loads and
spine stability while performing commonly prescribed stabilization
exercises. Spine 29: 23192329, 2004.
23. Klein-Vogelbach, S. Ballgymnastik zur Funktionellen Bewegungslehre.
Berlin, Germany: Springer Verlag, 1990.
24. Lehman, GJ, Gordon, T, Langley, J, Pemrose, P, and Tregaskis, S.
Replacing a Swiss ball for an exercise bench causes variable changes
in trunk muscle activity during upper limb strength exercises. Dyn
Med 4(6), 2005. doi:10.1186/1476-5918-4-6.
25. Marras, WS and Davis, KG. A non-MVC EMG normalisation
technique for the trunk musculature: Part 1. Method development.
J Electromyogr Kinesiol 11: 19, 2001.
26. Marshall, PWM and Murphy, BA. The validity and reliability of surface
EMG to assess the neuromuscular responseof the abdominal musclesto
rapid limb movement. J Electromyogr Kinesiol 13: 477489, 2003.
27. Marshall, PWM and Murphy, BA. Core stability exercises on and off
a swiss ball. Arch Phys Med Rehabil 86: 242-249, 2005.

6. Arokoski, JP, Valta, T, Airaksinen, O, and Kankaanpaa, M. Back and

abdominal muscle function during stabilization exercises. Arch Phys
Med Rehabil 82: 10891098, 2001.

28. Marshall PWM and Murphy, BA. Changes in muscle activity and
perceived exertion during exercises performed on a swiss ball. Appl
Physiol Nutr Metab 31: 376383, 2006.

7. Arokoski, JP, Valta, T, Airaksinen, O, and Kankaanpaa, M. Back and

hip extensor muscle function during therapeutic exercises. Arch Phys
Med Rehabil 80: 842850, 1999.

29. Marshall, PWM and Murphy, BA. Increased deltoid and abdominal
muscle activity during swiss ball bench press. J Strength Cond Res
20: 745750, 2006.

8. Ballor, DL, Becque, MD, and Katch, VL. Metabolic responses during
hydraulic resistance exercise. Med Sci Sports Exerc 19: 363367, 1987.

30. Marshall, PWM and Murphy, BA. Muscle activation changes

following exercise rehabilitation for chronic low back pain. Arch Phys
Med Rehabil 891: 3051313, 2008.

9. Basmajian, JVand De Luca, CJ. Muscles Alive, Their Functions Revealed by

Electromyography. Balitmore, MD: Lippincott Williams & Wilkins, 1985.
10. Behm, DG, Anderson, K, and Curnew, RS. Muscle force and
activation under stable and unstable conditions. J Strength Cond Res
16: 416422, 2002.

31. Marshall, PWM and Murphy, BA. Self report measures best
explain changes in disability compared with physical measures
after exercise rehabilitation for chronic low back pain. Spine
33: 326338, 2008.

11. Bressel, E, Willardson, JM, Thompson, B, and Fontana, FE. Effect of

instruction, surface stability, and load intensity on trunk muscle
activity. J Electromyogr Kinesiol 19: e500e504, 2009.

32. Nuzzo, JL, McCaulley, GO, Cormie, P, Cavill, MJ, and McBride, JM.
Trunk muscle activity during stability ball and free weight exercises.
J Strength Cond Res 22: 95102, 2008.

12. Campos, GE, Luecke, TJ, Wendeln, HK, Toma, K, Hagerman, FC,
Murray, TF, Ragg, KE, Ratamess, NA, Kraemer, WJ, and Staron, RS.

33. Oetterli, S and Larsen, C. Bewegungskoordination auf dem Ball.

Physiotherapie/Fisioterapia J Swiss Fed Physiotherap 6: 2335, 1996.

1544

the

TM

Journal of Strength and Conditioning Research

the

TM

Journal of Strength and Conditioning Research

| www.nsca-jscr.org

34. Panjabi, MM. The stabilizing system of the spine. Part I. Function,
dysfunction, adaptation and enhancement. J Spinal Dis 5: 383389,
1992.

39. Stanton, R, Reaburn, PR, and Humphries, B. The effect of short-term

Swiss ball training on core stability and running economy. J Strength
Cond Res 18: 522528, 2004.

35. Panjabi, MM. The stabilizing system of the spine. Part II. Neutral
zone and instability hypothesis. J Spinal Dis 5: 390397, 1992.

40. Tarnanen, SP, Ylinen, JJ, Siekkinen, KM, Malkia, EA, Kautiainen, HJ,
and Hakkinen, AH. Effect of isometric upper-extremity exercises
on the activation of core stabilizing muscles. Arch Phys Med Rehabil
89: 513521, 2008.

36. Peterson, MD, Rhea, MR, and Alvar, BA. Applications of the doseresponse for muscular strength development: A review of metaanalytic efficacy and reliability for designing training prescription.
J Strength Cond Res 19: 950958, 2005.
37. Rhea, MR, Alvar, BA, Burkett, LN, and Ball, SD. A meta-analysis to
determine the dose resopnse for strength development. Med Sci
Sports Exerc 35: 456464, 2003.
38. Scala, D, McMillan, J, Blessing, D, Rozenek, R, and Stone, M.
Metabolic cost of a preparatory phase of training in weight
lifting: A practical observation. J Appl Sports Sci Res 1: 4852,
1987.

41. Tesch, PA. Short- and long-term histochemical and biochemical

adaptations in muscle. In: Strength and Power in Sport. Komi, PV, ed.
Boston, MA: Blackwell Scientific Publications, 1992. pp. 239248.
42. Vera-Garcia, FJ, Grenier, SG, and McGill, SM. Abdominal muscle
response during curl-ups on both stable and labile surfaces. Phys Ther
80: 564569, 2000.
43. Vezina, MJ and Hubley-Kozey, CL. Muscle activation in therapeutic
exercises to improve trunk stability. Arch Phys Med Rehabil 81: 1370
1379, 2000.

VOLUME 24 | NUMBER 6 | JUNE 2010 |

1545