Content uploaded by Eduardo Lusa Cadore
Author content
All content in this area was uploaded by Eduardo Lusa Cadore
Content may be subject to copyright.
RELATIONSHIP BETWEEN WORKLOAD AND
NEUROMUSCULAR ACTIVITY IN THE BENCH PRESS
EXERCISE
Ronei Silveira Pinto1*(A,D-G), Eduardo Lusa Cadore1 (A,B,D-F), Cleiton Silva Correa1 (A,B,D,F), Bruna Gonçalves Cor-
deiro da Silva1 (D-F), Cristine Lima Alberton1 (A-E), Cláudia Silveira Lima1 (A,B,D-F), Antonio Carlos de Moraes2 (D-F)
1Exercise Research Laboratory, Physical Education School, Federal University of Rio Grande do Sul - UFRGS,
Porto Alegre, Brazil
2Faculty of Physical Education – University of Campinas – UNICAMP, Campinas, Brazil
Abstract
Objective: To investigate the relationship between strength and electromyographic (EMG) signal in different intensities
in the bench press exercise.
Methods: Eleven healthy resistance trained men (22.8 ± 3.5) participated into the present study. Maximal isometric
strength was determined in the bench press exercise using a load cell. Muscle activation was assessed using surface elec-
tromyography (EMG) signals from the muscles pectoralis major, anterior deltoid and posterior deltoid at intensities ranging
to 60-90% of maximal voluntary contraction (MVC), in the bench press exercise. This procedure allowed the analysis of
the strength/EMG relationship.
Results: In all muscles assessed, there were significant differences in the normalized muscle activation between the
intensities of 60 and 70% of the MVC, as well as between 70 and 80% (P < 0.05), while there were no differences between
80 and 90% of MVC. In addition, there were significant correlations between strength and EMG signals for the muscles
pectoralis major (r = 0.43, P = 0.04), anterior deltoid (r = 0.52, P = 0.01), and posterior deltoid (r = 0.32, P = 0.046).
Conclusions: These results suggest that levels of muscle activation near to maximal are obtained at the intensity of
80 of MVC and no additional motor unit recruitment are achieved at 90% of MVC.
Key words: electromyography, strength training, muscle activation, bench press
Introduction
The most relevant acute variable for strength
training is the intensity, in other words, the training
workload [1-3]. The intensity determines the level of
muscle activation, in which the greatest is the work-
load, the greater will be the activation level of the
agonists muscles involved in action [1,4-6]. Consid-
ering the activation pattern of the antagonist muscle
at different training intensities, it has been observed
increases as well as decreases in the antagonist coacti-
vation at greater training intensities [7]. The decrease
in the antagonist coactivation seems to be associated
to the increase of strength development of the agonist
muscle [7,8]. On the other hand, the increase in the
antagonist coactivation gives the joint agreater stabil-
ity and integrity [7].
The muscle activation level during resistance exer-
cises has been assessed through surface electromyogra-
phy (EMG) [1,2,9-14]. The EMG shows graphically the
action potential generated in the recruited motor units
of the muscle investigated. As the overload imposed
to certain action increases, there is an increment in
the amplitude of the electromyographic signal [13-
17]. Thus, it seems that there is astrong relationship
between the force development and the EMG signal
(strength/EMG relationship) during specific muscle
actions. However, the most of the studies that inves-
tigated the association between muscle strength and
EMG signal, showed this association during exercises
for the lower limbs. Few studies, therefore, have in-
vestigated this relationship during exercises for upper
limbs, such as the bench press exercise.
The strength training intensity is often determined
using the workloads relative to the maximal load (val-
ues for 1 maximal repetition - 1RM). With regards to
the maximal strength training, it is widely suggested
that greater increases in this capacity occur at intensi-
ties ranging from 85-100% [3,18], since the recruit-
ment of the greater number of motor units would
be possible at these intensities. However, it was not
determined by the literature whether an increase in the
workload over 60% of maximal strength enhances the
EMG signal in upper-body exercises. The investigation
of the strength/EMG relationship in upper-body exer-
cise could give insights about whether perform greater
workloads is necessary to optimize the neuromuscular
activity in the muscles involved.
Given the scarce data regarding the relationship
between strength and EMG signal in the bench press
exercise, as well as the importance of investigating the
Medicina Sportiva
Med Sport 17 (1): 1-6, 2013
DOI: 10.5604/17342260.1041876
Copyright © 2013 Medicina Sportiva
ORIGINAL RESEARCH
1
2
Pinto R.S., Cadore E.L., Correa C.S., Cordeiro da Silva B.G., Alberton C.L., Lima C.S., Carlos de Moraes A. / Medicina Sportiva 17 (1): 1-6, 2013
pattern of motor units recruitment at different intensi-
ties, the purpose of the present study was investigate
the association between strength and EMG signal at
different intensities in the bench press exercise. Our
hypothesis is that there is an association between
strength and EMG signal, and as the workloads in-
creases, the greater will be the amplitude of the EMG
signals of the muscles investigated.
Methods
Participants
Eleven healthy young men (22.8 ± 3.5 years-old)
recreationally trained in resistance training, for at least
one year, participated of the present study. Calcula-
tion of the sample ‘‘n’’ was carried out using the PEPI
program (version 4.0) with astatistical power of 90%.
Each subject was informed about the methodological
procedures of the present study through the reading
of afree informed consent. This study was approved
by the University Institutional Review Board, and is
in accordance with Helsinki Declaration. Sampling
characteristics are described on Table 1.
Table 1. Physical Characteristics (n = 11)
Characteristics Mean ± SD
Age (years) 22.8 ± 3.5
Height (cm) 177.2 ± 7.8
Body Mass (kg) 76.7 ± 8.7
% Fat mass 8.7 ± 1.8
Maximal Voluntary Contraction (N) 1173 ± 284
Force and Electromyographic Acquisition and Analysis
Before start the experimental protocol, subjects
were familiarized performing specific muscle con-
tractions at submaximal effort using very light loads.
In order to obtain the maximal isometric strength,
the subjects warmed up for 5 minutes on acycle
ergometer and were, then, horizontally positioned
in the bench of the Smith machine (Sculptor, Porto
Alegre, Brazil). The bar was fitted with aload cell
with 200 kg of capacity, connected to an A/D con-
verter (Miotec, Porto Alegre, Brazil), which made it
possible to quantify the traction exerted when each
subject executed the exercise at adetermined angle.
The subjects were positioned lying in the bench with
the shoulder and the elbow at a90° angle, strapped
to the bench at the waist height. The load cell was
positioned perpendicular to the bar and the humerus
and parallel to the forearm. Subjects were instructed
to exert the maximum strength as possible when
trying to extent both elbows. Subjects had three at-
tempts to obtain their MVC, each lasting 5 seconds,
with a3-minute rest between each attempt. During
this test, verbal encouragement was provided so
that the subjects would feel motivated to develop
their maximal strength. The force-time curve was
obtained using Miograph software (Miotec), with
an acquisition rate of 2000 Hz and later analyzed
using SAD32 software. Signal processing included
filtering with aButterworth low-pass filter at acut-
off frequency of 9 Hz. Later, in order to determine
the highest MVC, a1-second slice was made in the
plateau of force, between the 2nd and 4th second of the
force-time curve. The test-retest reliability coefficient
(ICC) was 0.94 to MVC.
During the isometric strength test, the maximal
muscular agonist activation was evaluated using
surface electromyography in the pectoralis major
and anterior deltoid, and the antagonist coactivation
was determined in the posterior deltoid. Electrodes
were positioned on the muscular belly in abipolar
configuration (20mm inter-electrodes distance) in
parallel with the orientation of the muscle fibers,
according to Leis & Trapani [19]. Shaving and abra-
sion with alcohol were carried out in the muscular
belly, as previously described elsewhere, in order to
maintain the inter-electrodes resistance above 2000
Ω [14]. Reference electrode was fixed on the clavicle.
The raw EMG signal was acquired simultaneously to
MVC using a4-channel electromyography (Miotool,
Porto Alegre, Brazil), with asampling frequency of
2000 Hz per channel, connected to apersonal com-
puter (Dell Vostro 1000, São Paulo, Brazil). Follow-
ing signal acquisition, the data were exported to the
SAD32 software, in which they were filtered using the
Butterworth band-pass filter, with acut-off frequency
ranging between 20 and 500 Hz. After that, the EMG
records were sliced exactly in the 1 second when the
MVC was determined in the force-time curve and
the root mean square (RMS) values were calculated.
The RMS values of posterior deltoid were normal-
ized by the maximum RMS values of this muscle,
obtained during the MVC of horizontal extension
at 90° (Fig. 1).
After determination of maximal muscular activa-
tion, submaximal muscular activation (relative to
maximal) was randomly evaluated at different inten-
sities of MVC (60, 70, 80 and 90%). In this protocol,
subjects were oriented to maintain aspecific force
value for three seconds, receiving avisual feedback in
the computer that showed, in real-time, the strength
values. One trial was performed for each intensity, and
the resting time between trials was of 5 minutes. The
apparatus and the collection and analysis procedures
were the same used to determine the maximal EMG
signal. The submaximal RMS values were normal-
ized by the maximum RMS values obtained during
the MVC for each muscle. The test-retest reliability
coefficients (ICC values) of the EMG measurements
were over 0.85.
3
Pinto R.S., Cadore E.L., Correa C.S., Cordeiro da Silva B.G., Alberton C.L., Lima C.S., Carlos de Moraes A. / Medicina Sportiva 17 (1): 1-6, 2013
Statistical analysis
In order to analyze the collected data, descriptive
statistics were used, with the data presented as Means
± standard deviation (SD). The Shapiro-Wilk’s test
was used to verify the normal distribution of data.
Pearson product-moment linear correlation was used
for test the relationship between strength and EMG
signals for monitored muscles. ANOVA for repeated
measures was used to compare relative values of force
and EMG signal between different MVC intensities.
When applicable, LSD post-hoc was used. Significance
was accepted when P < 0.05 and the statistical power
was 90%.
Results
There were significant differences between all per-
centages of strength assessed. The values of strength
(N) corresponded to the pattern expected: 100%
(1173 ± 284 N) > 90% (1030 ± 265 N) > 80% (927 ±
242 N) > 70% (814 ± 271 N) > 60% (703 ± 279 N) of
MVC (P < 0.001). The values of the normalized EMG
signals are shown on Table 2. The pattern of activation
from the muscles pectoralis major, anterior deltoid
and posterior deltoid at the intensities assessed has
shown to be similar (Fig. 2). The normalized EMG
signal from the muscles pectoralis major, anterior
deltoid and posterior deltoid did not present signifi-
cant differences between the intensities of 80 and 90%
of MVC, however, both intensities were significantly
greater (P < 0.01) than those corresponding to 60
and 70% of MVC, which presented significant differ-
ences between them (P = 0.04). There were observed
significant correlations between strength values and
EMG signal from the muscles pectoralis major (r =
0.43; P = 0.04), anterior deltoid (r = 0.52; P = 0.01),
and posterior deltoid (r = 0.32; P = 0.046).
Table 2. Electromiography signal normalized by maximal voluntary contraction
Muscle 60% 70% 80% 90%
Pectoralis major 68.5 ± 22.2a79.9 ± 21.0b92.7 ± 23.3c97.9 ± 21.8c
Anterior deltoid 52.6 ± 16.9a62.4 ± 18.3b77.4 ± 21.8c87.3 ± 21.4c
Posterior deltoid 3.8 ± 2.1a5.0 ± 2.6b5.6 ± 2.9c7.1 ± 3.6c
%MVC: Percentage of maximal voluntary contraction. Values in mean ± SD. Different letters means significant differences, P < 0.001 (pectoralis major
and anterior deltoid) and P = 0.001 (posterior deltoid).
Fig. 1. Apparatus used for EMG record
4
Pinto R.S., Cadore E.L., Correa C.S., Cordeiro da Silva B.G., Alberton C.L., Lima C.S., Carlos de Moraes A. / Medicina Sportiva 17 (1): 1-6, 2013
Fig. 2. EMG signals normalized of pectoralis major, anterior deltoid and posterior deltoid with intensities expressed as percentages of maximal voluntary
contraction (MVC). Different letters means significant differences between intensities analyzed (P < 0.05).
Discussion
The primary findings of the present study were the
correlations observed between the isometric strength
in the bench press and the EMG signal from the mus-
cles pectoralis major, anterior deltoid and posterior
deltoid. In addition, there were significant differences
between the levels of muscle activation at 60, 70 and
80% of the MVC, with no difference observed between
80 and 90% of MVC.
Some studies which investigated the association
between strength and EMG signal have shown the
existence of alinear relationship between these param-
eters during isometric contractions [16,20]. However,
the great majority these studies have investigated this
associations in exercises for lower limbs, whereas few
studies have investigated the relationship between
strength and EMG signal in the upper limbs. In the
present study, poor to moderate correlations were
observed between strength and EMG signals of the
muscles assessed. Regarding the agonist muscles (i.e.
pectoralis major and anterior deltoid), the absence of
agreater correlation index may be explained by the
existence of other factors involved in strength produc-
tion, such as the elastic components of skeletal muscles
[17,21]. Another explanation to the results observed
may be the exercise evaluated, since in the present
study, amulti-joint exercise with broad muscle recruit-
ment was performed. Besides the muscles monitored,
the bench press involves the elbow joint, with apri-
mary contribution of the triceps brachii muscle, which
was not monitored in the present study. Furthermore,
other muscles involved in the horizontal flexion of
the shoulder and the abduction of the scapula, such
as the coracobrachialis, the pectoralis minor, and the
serratus anterior [5,22] were not monitored in the
present study.
Regarding the posterior deltoid, the poor cor-
relation observed indicates that, although there is
arelationship between the force production in the
bench press and the antagonist coactivation of the
posterior deltoid, this association is not strong, at
least when only one antagonist muscle is monitored.
Another aspect that may have influenced the existent
correlation between strength and the antagonist EMG
signal is that the subjects who participated in the
present study were strength-trained. Indeed, it has
been demonstrated that systematic strength train-
ing reduces the antagonist coactivation [5,16,18,23].
Thus, areduced pattern of antagonist coactivation
may have influenced the poor association between
force production and antagonist coactivation in the
present study.
Our results were similar to those found in the
study by Doheny et al. [18], in which was investigated
the relationship between strength and EMG signal of
agonists and antagonists muscles of the elbow flexion
(biceps brachii, brachioradialis and triceps brachii).
These authors observed coactivation values between
2 to 28% of the maximal activation (during aMVC)
of the biceps brachii during elbow extension, between
20 to 38% of the maximal activation of brachioradialis
and between 15 and 49% of the maximal activation
of the triceps brachii during the elbow flexion. In the
present study, the values of coactivation of the poste-
rior deltoid remained within arange of 3.8 to 7.12%
of MVC of this muscle.
In the present study, the comparison between the
level of activation obtained in different intensities
indicates that the difference in the amplitude of the
EMG signal, which reflects the number of motor units
recruited and the firing rate of these units, occurred
between the intensities of 60, 70 and 80%. However,
5
Pinto R.S., Cadore E.L., Correa C.S., Cordeiro da Silva B.G., Alberton C.L., Lima C.S., Carlos de Moraes A. / Medicina Sportiva 17 (1): 1-6, 2013
no difference occurred between 80 and 90%, which
suggests an absence of increase on the neuromuscular
activity of these muscles at the greater intensities. In
another study, which the lower limbs were assessed,
Suzuki et al. [24], analyzed the EMG signal of the
knee extensor muscles at five different percentages
of MVC (5, 10, 20, 30 and 50%) and they found two
ranges of intensities which the neuromuscular activ-
ity were not significantly different, between 5 and
10% as well as between 20 and 30%. Hence, the results
of the present study suggest that, to achieve amuscle
activation close to maximal in the muscles evaluated
during the bench press exercise, the intensity of 80%
of MVC is asufficient stimulus and no additional
activation occurs at 90% of MVC. However, care-
ful is necessary to extrapolating these results, since
the muscle activation was assed isometrically in the
present study. Thus, further studies investigating
the muscle activation at percentages of the dynamic
maximal strength are necessary to determine whether
performing intensities over 80% of 1RM would in-
crease the muscle activation of these muscles in the
bench press exercise.
Apossible limitation of the present study is the use
of the surface electromyography to detect the muscle
activity, since it have been extensively described that
this technique has limitations, such as crosstalk from
other muscles inquired, movement of muscle fibers,
anisotropy and inhomogeneity of the muscle, fascia,
fat and skin tissues, and the fact of the electrodes may
not reflect the activity of all motor units activated, to
cite some of them [1]. In addition, the surface EMG
underestimates the activation signal sent from the
spinal cord to muscle as aresult of the cancellation of
positive and negative phases of MU action potentials.
Therefore, the results extracted from the sEMG am-
plitude about the motor unit recruitment of the MU
may be underestimated since it is possible that there
is an amplitude cancellation. Thus, the limitations of
the technique may explain the weak associations be-
tween strength and muscle activity observed during
the bench press exercise.
Conclusion
The present results, suggests that performing 80%
of MVC is asufficient stimulus to obtain amuscle
activation close to maximal with no addition motor
units recruitment at 90% of MVC. From apractical
point of view, performing lower loads to achieve
the same pattern of motor unit activation results in
aless exigency of the joint structures (joint capsule,
ligaments and tendons). It may also be suggested that
approximately at 80% of MVC, the level of activation
of the muscles pectoralis major and anterior deltoid
is maximal, with areduction in the joint overload
and risk of injuries.
Declaration of interest
The authors report no conflicts of interest.
References
1. De Luca CJ. The use of surface electromyography in biome-
chanics. J Appl Biomech 1997; 13: 135–63.
2. Doorenbosh CAM, Harlaar J. Accuracy of practicable EMG
to force model for knee muscles: short communications.
Neurosci Lett 2004; 368: 78-81.
3. Kraemer WJ, Ratamess NA. Hormonal responses and ada-
ptations to resistance exercise and training. Sports Med 2005;
35: 339-61.
4. Alkner BA, Tesch PA, Berg HE. Quadriceps EMG/force
relationship in knee extension and leg press. Med Sci Sports
Exerc 2000; 32: 459-63.
5. Komi PV. Training of muscle strength and power: interaction
of neuromotonic, hypertrophy and mechanical factors. Inter-
national Journal of Sports Med 1986; 7: 10–5.
6. Komi PV, Linmano V, Silventoinen P, Sillanpaä M. Force
and EMG power spectrum during eccentric and concentric
actions. Med Sci Sports Exerc 2000; 32: 1757-62.
7. Gabriel DA, Kamen G, Frost G. Neural Adaptations to Resisti-
ve Exercise: Mechanisms and Recommendations for Training
Practices. Sports Med 2006; 36: 133-49.
8. Folland JP, Williams AG. The Adaptations to Strength Tra-
ining: Morphological and Neurological Contributions to
Increased Strength. Sports Med 2007; 37: 145-68.
9. Basmajian J, De Luca C. Muscles alive. Their functions re-Their functions re-
vealed by electromyography. (5th ed.) Baltimore: William &
Wilkins, 1985.
10. Ferri A, Sclaglioni G, Pousson M, et al. Strength and power
changes of the human plantar flexors and knee extensors in
response to resistance training in old age. Acta Physiol Scand
2003; 177: 69-78.
11. Lindeman E, Spaans S, Reulen JPH, et al. Surface EMG of
proximal leg muscles in neuromuscular patients and in he-
althy controls. Relations to force and fatigue. J Electromyogr
Kinesiol 1999; 9: 299-307.
12. Macdonald JH, Farina D, Marcora SM. Response of Elec-Response of Elec-
tromyography Variables during Incremental and Fatiguing
Cycling. Med Sci Sports Exerc 2008; 40: 335-44.
13. Signorile JF, Weber B, Roll B, et al. An electromyographical
comparison of the squat and knee extension exercises. J
Strength Cond Res 1994; 8: 178–83.
14. Silva EM, Brentano MA, Cadore EL, et al. Analysis of muscle
activation during different leg press exercises at submaximum
effort levels. J Strength Cond Res 2008; 22: 1059–65.
15. Ebben WP, Feldmann CR, Dayane A, et al. Muscle activation
during lower body resistance training. Int J Sports Med 2009;
30: 1-8.
16. Gordon KD, Pardo RD, Johnson JA, et al. Electromyography
activity and strength during maximum isometric pronation
and supination efforts in healthy adults. J Orthop Res 2004;
22: 208-13.
17. Rabita G, Pérot C, Lensel-Corbeil G. Differential effect of
knee extension isometric training on the different muscles
of the quadriceps femoris in humans. Eur J Appl Physiol
2000; 83: 531–8.
18. Doheny EP, Lowery ML, Fitzpatrick DP, O’malley MJ. Effect
of elbow joint angle on force–EMG relationships in human
elbow flexor and extensor muscles. J Electromyogr Kinesiol
2007; 18: 760-70.
19. Leis AA, Trapani VC. Atlas of electromyography. (1st ed.)
Oxford, NY: Oxford University Press, 2000.
20. Kellis E, Kattis A. Reliability of EMG power-spectrum and
amplitude of the semitendinosus and biceps femoris muscles
during ramp isometric contractions. J Electromyogr Kinesiol
2008; 18: 351–8.
21. Narici M, Vroi GS, Landoni L, et al. Changes in force, cross-
-sectional area and neural activation during strength training
and detraining of the human quadriceps. Eur J Appl Physiol
1989; 59: 310-9.
6
Pinto R.S., Cadore E.L., Correa C.S., Cordeiro da Silva B.G., Alberton C.L., Lima C.S., Carlos de Moraes A. / Medicina Sportiva 17 (1): 1-6, 2013
Authors’ contribution
A – Study Design
B – Data Collection
C – Statistical Analysis
D – Data Interpretation
E – Manuscript Preparation
F – Literature Search
G – Funds Collection
22. Welsch EA, Bird M, Mayhew JL. Electromyographic activity
of the pectoralis major and anterior deltoid muscles during 3
upper-body lifts. J Strength Cond Res 2005; 19, 449-52.
23. Zhou P, Rymer WZ. Factors governing the form of the rela-
tion between muscle force and the EMG: asimulation study.
J Neurophysiol 2004; 92: 2878-86.
24. Suzuki H, Conwit RA, Stashuk D, et al. Relationships between
surface-detected EMG signals and motor unit activation. Med
Sci Sports Exerc 2002; 34: 1509–17.
25. Farina D, Merletti R, Enoka RM. The extraction of neural
strategies from the surface EMG. J Applied Physiol 2004;
96:1486-95.
Accepted: March 15, 2013
Published: March 27, 2013
Address for correspondence:
Eduardo Lusa Cadore
Exercise Research Laboratory (LAPEX)
Federal University of Rio Grande do Sul (UFRGS)
Rua Felizardo, 750 – Bairro Jardim Botânico
CEP: 90690-200
Porto Alegre – RS, Brazil
Tel: + 55-51-33085894
E-mail address: edcadore@yahoo.com.br
Ronei Silveira Pinto: ronei.pinto@ufrgs.br
Cleiton Silva Correa: cleitonesef@yahoo.com.br
Bruna Gonçalves Cordeiro da Silva: brugcs@hotmail.com
Cristine Lima Alberton: tinialberton@yahoo.com.br
Cláudia Silveira Lima: claudia.lima@ufrgs.br
Antonio Carlos de Moraes: acmoraes@fef.unicamp.br
