EFFECTS OF BODY POSITION AND LOADING MODALITY

ON MUSCLE ACTIVITY AND STRENGTH IN SHOULDER

PRESSES
ATLE H. SAETERBAKKEN1 AND MARIUS S. FIMLAND2,3
1
Faculty of Teacher Education and Sport, Sogn og Fjordane University College, Sogn og Fjordane, Norway; 2Department of
Public Health and General Practice, Faculty of Medicine, Norwegian University of Science and Technology, Trondheim,
Norway; and 3Hysnes Rehabilitation Center, St. Olavs University Hospital, Trondheim, Norway

ABSTRACT the greatest stability requirement (standing and dumbbells)

Saeterbakken, AH and Fimland, MS. Effects of body position demonstrated the highest neuromuscular activity of the deltoid
and loading modality on muscle activity and strength in muscles, although this was the exercise with the lowest 1-RM
shoulder presses. J Strength Cond Res 27(7): 1824–1831, strength.
2013—Little is known about the effect of performing upper- KEY WORDS EMG, resistance exercise, instability, free
body resistance exercises with dumbbells versus barbells weights, 1-RM, shoulder press
and standing versus seated. Therefore, this study sought to
compare electromyogram activity (EMG) and one-repetition
INTRODUCTION

P
maximum (1-RM) in barbell and dumbbell shoulder presses
performed seated and standing. Fifteen healthy men volun- roponents of instability in resistance training sug-
teered for 1-RM and EMG testing with a load corresponding gest that instability enhances the stress on the
to 80% of the 1-RM. Electromyogram activity was measured in neuromuscular system compared with traditional
resistance training methods, but the scientific
the anterior, medial, and posterior deltoids and biceps and tri-
reports are inconclusive (5). In fact, the muscle activity in
ceps brachii. The following EMG differences or trends were
unstable versus more stable resistance exercises is reported
observed: For deltoid anterior: ;11% lower for seated barbell
to be higher, similar, or even lower depending on the exer-
versus dumbbell (p = 0.038), ;15% lower in standing barbell cises and muscles that are examined (2,10,18,19,23,27).
versus dumbbell (p , 0.001), ;8% lower for seated versus Several methods have been used to induce instability in
standing dumbbells (p = 0.070); For medial deltoid, ;7% resistance training, such as unstable instead of stable surfaces
lower for standing barbell versus dumbbells (p = 0.050), (4,6,16,23), free weights instead of machines (27,29,30),
;7% lower for seated versus standing barbell (p = 0.062), dumbbells instead of barbell (27,34), unilateral instead of
15% lower for seated versus standing dumbbell (p = 0.008); bilateral exercises (6,24), and standing instead of supine
For posterior deltoid: ;25% lower for seated versus standing exercises (28). Although unstable surfaces such as wobble
barbell (p , 0.001), ;24% lower for seated versus standing boards, swiss balls and BOSUÒ balls and slings have received
dumbbells (p = 0.002); For biceps, ;33% greater for seated much attention in the scientific literature and commercial
media, most daily life activities and sports are performed
barbell versus dumbbells (p = 0.002), 16% greater for stand-
on stable surfaces while moving an unstable external resis-
ing barbell versus dumbbell (p = 0.074), ;23% lower for
tance. Examples include a tennis racket, volleyball, or lifting
seated versus standing dumbbells (p , 0.001); For
a baby. According to the training principle of specificity (22),
triceps, ;39% greater for standing barbell versus dumbbells it may therefore in many cases be more relevant to include
(p , 0.001), ;20% lower for seated versus standing barbell instability components by other means than changing the
(p = 0.094). 1-RM strength for standing dumbbells was ;7% surface.
lower than standing barbell (p = 0.002) and ;10% lower than Using dumbbells instead of barbells forces those individ-
seated dumbbells (p , 0.001). In conclusion, the exercise with uals who engage in resistance training to control and balance
the weights independently and thereby could increase and/
Address correspondence to Atle H. Saeterbakken, atle.saeterbakken@ or reduce the involvement of agonists, synergists, stabilizers,
hisf.no. and antagonists (27,34). For example, both Welsch et al. (34)
27(7)/1824–1831 and Saeterbakken et al. (27) reported similar EMG activity in
Journal of Strength and Conditioning Research pectoralis major and anterior deltoid in dumbbell chest
Ó 2013 National Strength and Conditioning Association presses compared with barbell despite the fact that dumbbell
the TM

1824 Journal of Strength and Conditioning Research

Copyright © National Strength and Conditioning Association Unauthorized reproduction of this article is prohibited.
 the TM

Journal of Strength and Conditioning Research | www.nsca.com

loads were only 63–83% of the barbell loads (27,34). More- testing procedures, possible risks, and gave written consent
over, Kohler et al. (16) compared barbell and dumbbell to participate prior the testing. The participants’ stated that
shoulder presses and reported similar EMG activity in ante- they typically performed resistance training 2–4 times per
rior and medial deltoid despite that the dumbbell load was week. The 1-RM results of seated barbell press represented
only ;86% of the barbell load. ;71% of the participants’ body weight. Participants were
In contrast to the growing body of literature concerning excluded from the study if they had musculoskeletal pain,
barbells versus dumbbells (16,27,34), we are not aware of illness, injuries, or were not familiar with shoulder press
studies investigating standing instead of seated position as exercises using free weights. Before the study, ethics
an instability factor for agonists, synergists, and antagonists. approval was obtained from the local research ethics com-
Free weight resistance exercises performed standing should, mittee. The study conformed to the latest revision of the
in theory, increase the stabilizing requirements compared Declaration of Helsinki. The participants were instructed
with seated exercises (3,9). We recently demonstrated this to refrain from any additional resistance training that tar-
for the core muscles (24), but the effect of standing versus geted the upper body 72 hours before testing.
seated position on upper extremity muscles is unclear. It is
Procedures
also unclear how these different instability approaches (body
Participants were instructed to continue with their normal
position + loading modality) interact with each other with
diet throughout the study and to prepare for the sessions as
respect to muscle activity and strength.
they would for their usual training sessions. The participants
The purpose of this study was to compare the 1-RM
came to the laboratory at a time that was as similar as
strength and EMG activity in barbell and dumbbell shoulder
possible to their usual training time. All tests were executed
presses performed seated and standing. It was hypothesized
in February.
that increased stability requirement (dumbbells instead of
The participants were instructed to use moderate and
barbell and standing instead of seated) would result in similar
controlled lifting tempo in the third session, but the tempo of
EMG activity of the deltoid muscles but decrease the 1-RM
1-RM testing (sessions 1 and 2) was self-selected. A digital
strength.
watch provided feedback to the participants so that they
should use approximately 2 seconds in the eccentric phase
METHODS
and 2 seconds in the concentric phase. The use of 80% 1-RM
Experimental Approach to the Problem is typical for many training regimens and corresponds to
To examine effects of body position (seated and standing) a load between 8 and 12 RM (3,13). Five repetitions were
and loading modality (barbell and dumbbells) on 1-RM used to avoid confounding effects of fatigue on the neuro-
strength and neuromuscular activity in shoulder presses, muscular activation during the exercises (34). Before a new
a within-subjects repeated-measures design was used. The exercise, 5 repetitions of 60% of 1-RM were used as a warm-
shoulder press was chosen because it is frequently used and up to the exercise. The sets were separated by 2–4 minute
requires considerable inter- and intramuscular coordination rest periods in the third and fourth session.
of the deltoid and arm muscles, which makes it suitable as an
experimental model in the study of instability in resistance Testing Procedures
training. The participants attended 4 sessions, each separated Before a testing session, the participants performed a stan-
by 3–5 days. The first session determined 1-RM in seated dardized warm-up protocol including 10 minutes on a cycle
presses, while the second session tested standing presses. ergometer followed by 6–8 warm-up sets (60–85% of 1-RM,
A 4-minute rest period was given between each 1-RM 6–12 repetitions) seated and standing using barbell and dumb-
attempt (10,30). In the third session, the participants per- bells. After the warm-up, the load was increased to what the
formed 5 repetitions at 80% of 1-RM on each of the shoulder participants perceived as their 1-RM. Depending on whether
press exercises in randomized order (12,21). The experimen- the lift was successful or not, the load was increased or
tal (fourth) session was identical to the third but included decreased with 2.5 kg for barbells and 1.25 kg for each dumb-
EMG measurements. EMG activity was recorded from bell. The participants achieved 1-RM for the exercises within
deltoid muscles (anterior, medial, and posterior), and arm 2–4 attempts. A 4-minute rest period was given between each
muscles (biceps and triceps brachii). 1-RM attempt (10,30). Participants were vigorously encour-
aged in the 1-RM testing but not during sessions 3 and 4.
Subjects Two investigators acted as spotters and assisted the
Fifteen healthy men (age, 22 6 2 years, body mass = 79 6 14 participants to ensure safety during each lift and to stabilize
kg, stature = 1.79 6 0.06 m) participated in the study. The the weights before and after the lifts. The eccentric phase
participants had 5.0 6 2.6 years of strength-training experi- started at fully extended elbows. On audio command, the
ence but were not competitive power lifters or Olympic participants lowered the weights so that the center of
weightlifters. All participants were accustomed to the shoul- the barbell or dumbbells was at the level of the acromion.
der press exercises and performed them as part of their reg- The dumbbells were held with the thumb-side toward the
ular training program. All participants were informed of ears at all times. The barbell was lifted in front of the face. The

VOLUME 27 | NUMBER 7 | JULY 2013 | 1825

Copyright © National Strength and Conditioning Association Unauthorized reproduction of this article is prohibited.
 Comparison of Shoulder-Press Exercises

elbows were maintained in the frontal plane at all time. the same muscle. The electrode configuration had a fixed
Excessive postural sway was not allowed. Grip width of the and identical inter-electrode distance. The surface electrodes
barbell or dumbbells was self-selected. The concentric phase were positioned at biceps brachii, triceps brachii, anterior
ended when the elbows were fully extended (11). Further deltoid, medial deltoid, and posterior deltoid according to
details of the exercise technique are described by Graham the recommendations by SENIAM (14). Electrodes were
(11). The barbell or dumbbells were pressed straight up and only placed on the side of the preferred arm (6).
the lifting speed, horizontal position of the barbell, and dumb- The raw EMG signal was amplified and filtered using
bells and the body position were controlled by the 2 investi- a preamplifier located as near the pickup point as possible
gators. A standard bench (758 upward angle) was used in the to minimize noise induced from external sources. The raw
seated tests with the gluteus and upper torso in contact with EMG signals were converted to root mean square (RMS)
the bench. The legs were placed shoulder wide with a 908 signals using a hardware circuit network (frequency response
angle at the knees (11). In the standing exercises, a string 450 kHz, averaging constant 12 ms, total error 6 0.5%). The
placed in a square about 1 cm from the skin, and 3 cm above common mode rejection rate was 106 dB and the impedance
the iliac crest was used to prevent postural sway. The feet between each electrode pair was ,1012 V. The EMG signals
were placed shoulder width apart with extended knees. were sampled at a rate of 1000 Hz. Signals were band-pass
filtered with a cutoff frequency of 8 and 600 Hz, rectified and
Electromyographic Recordings integrated. The RMS converted signals were resampled at
The skin was prepared (shaved, washed with alcohol, and a rate of 100 Hz using a 16-bit A/D converter with a com-
abraded) for placement of gel-coated surface EMG electro- mon mode rejection rate of 106 dB. The stored data
des. Self-adhesive electrodes (Dri-Stick Silver circular sEMG was analyzed using commercial software Musclelab V8.13
Electrodes AE-131, NeuroDyne Medical, Cambridge, MA, (Ergotest Technology AS, Langesund, Norway).
USA) were placed in the presumed direction of the To identify the descending and ascending motion, a linear
underlying muscle fibers with a center-to-center distance encoder (Ergotest Technology AS) was attached to the barbell
of 2.0 cm and 11 mm contact diameter. Each triode and dumbbells. The linear encoder was synchronized with the
electrode configuration included an electrode pair measuring EMG recordings using Musclelab 4020e (Ergotest Technology
the electrical signal and also a reference electrode placed on AS). Position and EMG data were overlaid and marked to

Figure 1. A–D, Representative filtered EMG signals of the seated and standing barbell and dumbbell presses are presented (n = 15).

the TM

1826 Journal of Strength and Conditioning Research

Copyright © National Strength and Conditioning Association Unauthorized reproduction of this article is prohibited.
 the TM

Journal of Strength and Conditioning Research | www.nsca.com

precisely identify the beginning of the end of each repetition. position 3 exercise interaction was not significantly different
The RMS EMG was calculated as the mean activity from the (F = 1.611, p = 0.225). However, the main effects were signif-
start of the second to the end of the fifth repetition (24). icantly different for position (F = 5.053, p = 0.041) and exercise
(F = 27.109, p , 0.001). After post hoc analysis, ;11% lower
Statistical Analysis EMG activity in seated barbell versus dumbbell shoulder
To assess differences in 1-RM strength and neuromuscular presses was observed (1.016 6 0.34 vs. 1.136 6 0.36, p =
activity of the seated, standing, barbell, and dumbbell shoulder 0.038, ES = 0.33). There was ;15% lower EMG activity in
presses, we used a 2-way analysis of variance (ANOVA). The standing barbell versus standing dumbbells (1.048 6 0.34 vs.
independent variables were body position (seated and stand- 1.237 6 0.45, p , 0.001, ES = 0.47). Furthermore, there was
ing) and loading modality (barbell and dumbbells). The similar EMG activity in seated barbell versus standing shoulder
dependent variables were the EMG activity for each of the presses (1.016 6 0.34 vs. 1.048 6 0.34, p = 0.760, ES = 0.33).
muscles and 1-RM strength. When a significant interaction There was a trend for (;8%) lower EMG activity in seated
was detected, paired t-tests with Bonferroni post hoc correc- dumbbells versus standing dumbbells shoulder presses (1.136
tions were applied to determine the location of the differen- 6 0.36 vs. 1.237 6 0.45, p = 0.070, ES = 0.25).
ces. To compare the lifting time of the exercises, we used For medial deltoid (Figure 2B), the position 3 exercise inter-
a separate one-way ANOVA. All results are presented as action was significantly different (F = 10.698, p = 0.006). After
mean 6 SD and Cohen d effect size (ES). Effect size of 0.2 post hoc analyses, similar EMG activity in seated barbell
was considered small, 0.5 as medium, and 0.8 as large. Statis- versus dumbbells shoulder presses was observed (0.559 6
tical significance was accepted at p # 0.05. SPSS (version 19.0; 0.19 vs. 0.549 6 0.14, p = 0.653, ES = 0.06). There was
Inc., Chicago, IL, USA) was used for statistical analyses. ;7% lower EMG activity in standing barbell versus dumbbell
shoulder presses (0.601 6 0.18 vs. 0.647 6 0.22, p = 0.050, ES
RESULTS = 0.23). Furthermore, there was a trend for (;7%) lower EMG
Deltoid Muscle Activation activity in seated barbell versus standing barbell shoulder
Representative filtered deltoid EMG signals of the exercises are presses (0.559 6 0.19 vs. 0.601 6 0.18, p = 0.062, ES =
presented in Figure 1A–D. For anterior deltoid (Figure 2A), the 0.23). Approximately 15% lower EMG activity in seated

Figure 2. A–C, The mean and SD of the EMG activity in anterior deltoid (A), medial deltoid (B), and posterior deltoid (C), in combinations of seated and
standing barbell and dumbbells shoulder presses (n = 15). *Significant difference in EMG activity between the exercises on a 0.05 level.

VOLUME 27 | NUMBER 7 | JULY 2013 | 1827

Copyright © National Strength and Conditioning Association Unauthorized reproduction of this article is prohibited.
 Comparison of Shoulder-Press Exercises

dumbbell versus standing dumbbell shoulder presses was EMG activity in seated dumbbell versus standing dumbbell
observed (0.549 6 0.14 vs. 0.647 6 0.22, p = 0.008, ES = 0.53). shoulder presses was observed (0.405 6 0.20 vs. 0.522 6
For posterior deltoid (Figure 2C), the position 3 exercise 0.245, p , 0.001, ES = 0.52).
interaction was not significantly different (F = 0.008, p = For triceps (Figure 3B), the position 3 exercise interaction
0.930). There was a main effect for position (F = 31.521, was significantly different (F = 6.416, p = 0.024). After post
p , 0.001) but not exercise (F = 2.080, p = 0.171). After post hoc analyses, similar EMG activity in seated barbell versus
hoc analyses, ;25% lower EMG activity in seated barbell dumbbell shoulder presses was observed (0.250 6 0.12 vs.
versus standing barbell shoulder presses was observed (0.316 0.218 6 0.13, p = 0.620, ES = 0.26). There was ;39% greater
6 0.16 vs. 0.421 6 0.16, p , 0.001, ES = 0.66). Furthermore, EMG activity in standing barbell versus dumbbell shoulder
there was ;24% lower EMG activity in seated dumbbells presses (0.315 6 0.15 vs. 0.192 6 0.10, p , 0.001, ES = 1.18).
versus standing dumbbells (0.331 6 0.13 vs. 0.438 6 0.20, Furthermore, a trend for (;20%) lower EMG activity in
p = 0.002, ES = 0.63). seated barbell versus standing barbell was observed (0.250
6 0.12 vs. 0.315 6 0.15, p = 0.094, ES = 0.48). There was
Biceps and Triceps Activation
similar EMG activity in seated dumbbell versus standing
For biceps (Figure 3A), the position 3 exercise interaction
dumbbell shoulder presses (0.218 6 0.13 vs. 0.192 6 0.10,
was significantly different (F = 8.158, p = 0.013). After post
p = 0.602, ES = 0.22).
hoc analyses, ;33% greater EMG activity in seated barbell
versus dumbbell was observed (0.604 6 0.30 vs. 0.405 6 0.20, 1-RM strength
p = 0.002, ES = 0.98). There was a trend for (;16%) greater The position 3 exercise interaction was significantly different
EMG activity in standing barbell versus dumbbell shoulder for 1-RM strength (F = 14.235, p = 0.002, Figure 4). After post
presses (0.623 6 0.28 vs. 0.522 6 0.245, p = 0.074, ES = 0.41). hoc analyses, similar 1-RM strength in seated barbell versus
Furthermore, there was similar EMG activity in seated barbell dumbbell shoulder presses was observed (56.3 6 8.4 kg vs.
versus standing barbell shoulder presses (0.604 6 0.30 vs. 56.0 6 7.6 kg, p = 0.751, ES = 0.04). There was ;7% lower
0.623 6 0.28, p = 0.904, ES = 0.07). Approximately 23% lower 1-RM strength in standing dumbbell versus barbell shoulder
presses (50.7 6 5.3 kg vs. 54.7 6 6.4 kg, p = 0.002, ES = 0.68).
Furthermore, there was similar 1-RM strength in seated bar-
bell versus standing barbell shoulder presses (56.0 6 8.4 kg vs.
54.7 6 6.4 kg, p = 0.272, ES = 0.22). Approximately 10%
lower 1-RM strength in standing dumbbell versus seated
dumbbell shoulder presses was observed (50.7 6 5.3 kg vs.
56.0 6 7.61 kg, p , 0.001, ES = 0.81).
Lifting Time
There were no differences between shoulder press exercises in
total lifting time of the 4 repetitions used for analysis: seated
barbell 11.2 6 2.3 seconds, seated dumbbells 11.8 6 2.5

Figure 3. A, B, The mean and SD of the EMG activity in biceps (A) and Figure 4. Mean (SD) of 1-RM in seated barbell, seated dumbbells,
triceps (B) in combinations of seated and standing barbell and standing barbell, and standing dumbbells shoulder presses (n = 15).
dumbbells shoulder presses (n = 15). *Significant difference in EMG *Significantly different in 1-RM strength between the exercises on a 0.05
activity between the exercises on a 0.05 level. level.

the TM

1828 Journal of Strength and Conditioning Research

Copyright © National Strength and Conditioning Association Unauthorized reproduction of this article is prohibited.
 the TM

Journal of Strength and Conditioning Research | www.nsca.com

seconds, standing barbell 12.1 6 2.6 seconds, and standing In the seated position, the legs were placed shoulder wide
dumbbells 12.3 6 3.5 seconds, (F = 0.438, p = 0.727–1.000). with a bench supporting the back, creating a solid base of
support against movement in the sagittal plane. In the
DISCUSSION standing position, the increased degrees of freedom of the
In the present study, we examined effects of body position torso may have provided the participants the opportunity to
(seated and standing) and loading modality (barbell and coordinate the weight lifting in a way that provided higher
dumbbells) on 1-RM strength and neuromuscular activity in muscle activation, or the increased stability demands on the
shoulder presses, for the first time. The main finding in this deltoids may have increased motor drive. Another possibility
study is that the standing dumbbell press exercise, which is that standing instead of seated position resulted in more
was the exercise with the greatest stability requirement remote voluntary activation of leg and trunk muscles, which
(standing + dumbbells), demonstrated the highest neuro- is known to cause concurrent activation potentiation (7). In
muscular activity of the deltoid muscles, although this was addition, one could expect that a more stable base of support
the exercise with the lowest 1-RM strength. Furthermore, in the seated position could create a more favorable position
standing versus seated execution, and to some extent dumb- for lifting heavier weights that would also activate the ante-
bells versus barbell, both resulted in increased muscle acti- rior and medial deltoid muscles to a higher extent. However,
vation of the deltoid muscles. Standing instead of seated the results demonstrated that going from seated to standing
presses raises the center of the mass and also provides position negatively affected absolute strength performance
a smaller base of support as the contact points decreases only for dumbbell presses.
from 3 to 2, particularly when using a bench with a back- Greater EMG activity was generally observed for the
rest. When using a pair of dumbbells instead of a barbell, the deltoid prime mover (anterior and medial part) muscles in
main difference is that the dumbbells must be controlled dumbbell exercises compared with barbell exercises (except
independently of each other. Hence, performing shoulder for medial deltoid in the seated position). This is not in line
presses standing and with dumbbells should lead to greater with previous studies comparing EMG activity using bar-
instability. bells and dumbbells (16,27,34). Although both barbell and
Contradictory EMG results have been reported from dumbbell presses are free weight exercises, dumbbells are
studies examining exercises with various stability require- more unstable than barbells. Conversely, dumbbells instead
ments (16,23,25–27,33). However, several of the studies of barbell lifting did not seem to affect posterior deltoid
observing higher EMG activity in unstable resistance train- activity. This is probably caused by the stability inducing
ing used the same absolute and not relative resistance differences between dumbbells and barbell in overhead
(1,6,18,23,31). In those studies, it is not possible to differen- presses; both exercises are unstable in the sagittal plane,
tiate the contributions of higher relative loads and higher but only dumbbells are unstable in the coronal (frontal)
stability requirements on neuromuscular activity because plane. The posterior deltoid was probably more important
using the same absolute load usually means that a higher as a stabilizer against perturbations in the sagittal plane,
relative load is used in the unstable condition. As we explaining why there were no differences between dumbbell
matched intensity of exercises, this was not a confounding and barbell presses for this muscle.
issue in the present study. Furthermore, previous studies that There was elevated muscle activity for posterior deltoids
did match relative resistance have observed that the EMG in standing versus seated position. The central nervous
activation of the prime movers in unstable exercises have system continuously monitors the position of all body
either been lower or similar compared with more stable segments and must continuously feed the posterior deltoids
alternatives (2,4,16,25–27,34). Hence, to our knowledge, this with neural drive for appropriate coactivation during shoul-
is the first study to report higher muscle activity of prime der presses. During standing presses, postural movements
movers of common resistance training exercises with inevitably occur in the sagittal plane, which is not the case
increasing instability. for seated presses with a backrest.
Another important aspect of this investigation is that we The EMG results for biceps and triceps partly supported
compared realistic alternatives of the same exercise. That is, the hypotheses. Greater EMG activity in standing versus
shoulder presses standing or seated, with a barbell and seated dumbbell presses in biceps was observed but similar
dumbbells. In many studies, exercises with large differences when using barbell. These results are in line with Saeter-
in stability requirement have been compared (i.e., very bakken et al. (27) and Lehman et al. (17), as these authors
unstable vs. very stable) (2,4,18,19). However, comparisons reported greater antagonist coactivation with greater insta-
of exercises with major differences in stability requirement bility requirements. For triceps, a strong trend was observed
cannot be generalized to exercises with more subtle differ- for greater EMG activity in standing versus seated barbell
ences in instability requirements. Furthermore, very unstable presses but similar using dumbbells. Higher triceps activation
exercises are generally not recommended for athletes, as the during standing barbell presses compared with dumbbells is
force output may decrease to a suboptimal level for strength in line with chest press results reported by Saeterbakken
or power gains (2,4,19,20). et al. (27). Extending the arms using the triceps alone in

VOLUME 27 | NUMBER 7 | JULY 2013 | 1829

Copyright © National Strength and Conditioning Association Unauthorized reproduction of this article is prohibited.
 Comparison of Shoulder-Press Exercises

dumbbell presses would only result in moving the dumbbell (standing + dumbbells) increased deltoid muscle activation.
load further away from the shoulders in a horizontal posi- This was despite the fact that the resistance-trained volunteers
tion. However, in barbell press, both arms are locked to the lifted the least weight in this exercise. Our findings suggest
barbell making it possible to achieve greater triceps activa- that standing dumbbell presses may be more beneficial for
tion without moving the load further sideways from the muscular development of the deltoid muscles than more sta-
shoulders (32). ble alternatives, however, if mechanical power output is
In the stable seated condition, dumbbells as an instability- a higher priority, more stable alternatives may be preferable.
inducing factor did not negatively influence 1-RM strength. Furthermore, both standing instead of seated body position
In contrast, Welsch et al (34) and Saeterbakken et al. (27) and dumbbells instead of barbell as the loading modality
reported dumbbell loads of 63% (34) and 83% (27) of the increased the stability requirement compared with seated
barbell loads in chest presses. Furthermore, we observed and barbell execution, and we suggest to include these factors
similar 1-RM strength when comparing seated versus stand- in periodized resistance training programs, which may reduce
ing barbell presses. In the present study, when the 2 insta- the risk of injuries and provide variation for athletes and
bility inducing factors were included (dumbbells and others engaging in resistance training. Furthermore, as many
standing), a 7% reduction in strength was observed com- sports require standing position and independent arm move-
pared with standing barbell and a 10% reduction compared ments (pitching, smashing, and serving), we suggest, accord-
to seated dumbbells were observed. ing the principle of training specificity, that these instability-
The present study was limited by not testing EMG in the inducing variables may be beneficial. Finally, in shoulder
1-RM testing and not performing the exercises (80% of 1-RM) rehabilitation, standing shoulder presses may be a viable
to exhaustion. Fatigue has been shown to increase the EMG choice as higher muscle activation can be attained with
activity (15). Eighty percentage of 1-RM was selected because lower external load.
it is a typical load of many resistance-training programs
(;8–12 repetitions). Performing the exercises to fatigue would ACKNOWLEDGMENTS
have impaired the performance of the subsequent exercise, The authors thank Espen Krohn-Hansen and Mats Smaamo
which would have influenced the results. Alternatively, more for assistance in participant recruitment and data collection
sessions could have been used, but as it is difficult or impos- and the participants for their enthusiastic participation. This
sible to place EMG electrodes on the exact same location, we study was conducted without any funding from companies
believe that the present approach is a good compromise. Fur- or manufacturers or outside organizations.
thermore, there are inherent technical limitations with the
surface EMG, and the electrodes can only provide an estimate REFERENCES
of neuromuscular activity (8). The risks of cross talk from 1. Anderson, K and Behm, DG. Trunk muscle activity increases with
neighboring muscles are present even if a small interelectrode unstable squat movements. Can J Appl Physiol 30: 33–45, 2005.
distance was used. 2. Anderson, KG and Behm, DG. Maintenance of EMG activity and
Resistance trained young males were recruited for this loss of force output with instability. J Strength Cond Res 18: 637–640,
2004.
study, and these results cannot necessarily be generalized to
other populations. Further research should investigate the 3. Baechle, TR and Earle, RW, and National Strength and
Conditioning Association (U.S.). Essentials of Strength Training and
use of these exercises for sedentary individuals or patients Conditioning. Champaign, IL: Human Kinetics, 2000.
and investigate the neuromuscular pattern of exercises with 4. Behm, DG, Anderson, K, and Curnew, RS. Muscle force and
both greater and lower stability requirements and intensity activation under stable and unstable conditions. J Strength Cond Res
levels. 16: 416–422, 2002.
In conclusion, for the deltoids, the use of standing instead 5. Behm, DG and Anderson, KG. The role of instability with resistance
training. J Strength Cond Res 20: 716–722, 2006.
of seated body position increased the neuromuscular activ-
6. Behm, DG, Leonard, AM, Young, WB, Bonsey, WAC, and
ity, as did dumbbells instead of barbell as the loading MacKinnon, SN. Trunk muscle electromyographic activity with
modality, although not consistently for the latter. Combining unstable and unilateral exercises. J Strength Cond Res 19: 193–201,
the 2 instability approaches (standing and dumbbells) 2005.
demonstrated the highest neuromuscular activity of the 7. Ebben, WP, Petushek, EJ, Fauth, ML, and Garceau, LR. EMG
analysis of concurrent activation potentiation. Med Sci Sports Exerc
deltoid muscles, but this was also the exercise with the 42: 556–562, 2010.
largest 1-RM strength reduction (7–10%) compared with the
8. Farina, D, Merletti, R, and Enoka, RM. The extraction of neural
other exercises. strategies from the surface EMG. J Appl Physiol 96: 1486–1495,
2004.
PRACTICAL APPLICATIONS 9. Garhammer, J. Free weight equipment for the development of
We compared 4 shoulder press exercises: seated military press, athletic strength and power, part I. Strength Cond J 3: 24–26, 1981.
standing military press, seated dumbbell press, and standing 10. Goodman, CA, Pearce, AJ, Nicholes, CJ, Gatt, BM, and
Fairweather, IH. No difference in 1RM strength and muscle
dumbbell press. This study demonstrated that combining 2 activation during the barbell chest press on a stable and unstable
common instability-inducing strategies in shoulder presses surface. J Strength Cond Res 22: 88–94, 2008.
the TM

1830 Journal of Strength and Conditioning Research

Copyright © National Strength and Conditioning Association Unauthorized reproduction of this article is prohibited.
 the TM

Journal of Strength and Conditioning Research | www.nsca.com

11. Graham, JF. Exercise: Dumbbell seated shoulder press. Strength Cond 23. Norwood, JT, Anderson, GS, Gaetz, MB, and Twist, PW.
J 30: 54–55, 2008. Electromyographic activity of the trunk stabilizers during stable and
unstable bench press. J Strength Cond Res 21: 343–347, 2007.
12. Hamlyn, N, Behm, DG, and Young, WB. Trunk muscle activation
during dynamic weight-training exercises and isometric instability 24. Saeterbakken, AH and Fimland, MS. Muscle activity of the core
activities. J Strength Cond Res 21: 1108–1112, 2007. during bilateral, unilateral, seated and standing resistance exercise.
Eur J Appl Physiol 24: 1671–1678, 2012.
13. Hatfield, DL, Kraemer, WJ, Spiering, BA, Hakkinen, K, Volek, JS,
Shimano, T, Spreuwenberg, LP, Silvestre, R, Vingren, JL, 25. Saeterbakken, AH and Fimland, MS. Electromyographic activity
Fragala, MS, Gomez, AL, Fleck, SJ, Newton, RU, and Maresh, CM. and 6-RM strength in bench press on stable and unstable surfaces.
J Strength Cond Res 27: 1101–1107, 2013.
The impact of velocity of movement on performance factors in
resistance exercise. J Strength Cond Res 20: 760–766, 2006. 26. Saeterbakken, AH and Fimland, MS. Muscle force output and
electromyographic activity in squats with various unstable surfaces.
14. Hermens, HJ, Freriks, B, Disselhorst-Klug, C, and Rau, G.
J Strength Cond Res 27: 130-136, 2013.
Development of recommendations for SEMG sensors and sensor
placement procedures. J Electromyogr Kinesiol 10: 361–374, 2000. 27. Saeterbakken, AH, van den Tillaar, R, and Fimland, MS. A
comparison of muscle activity and 1-RM strength of three chest-
15. Judge, LW, Moreau, C, and Burke, JR. Neural adaptations with press exercises with different stability requirements. J Sports Sci 29:
sport-specific resistance training in highly skilled athletes. J Sports 533–538, 2011.
Sci 21: 419–427, 2003.
28. Santana, JC, Vera-Garcia, FJ, and McGill, SM. A kinetic and
16. Kohler, JM, Flanagan, SP, and Whiting, WC. Muscle activation electromyographic comparison of the standing cable press and
patterns while lifting stable and unstable loads on stable and bench press. J Strength Cond Res 21: 1271–1277, 2007.
unstable surfaces. J Strength Cond Res 24: 313–321, 2010. 29. Schick, EE, Coburn, JW, Brown, LE, Judelson, DA, Khamoui, AV,
17. Lehman, GJ, MacMillan, B, MacIntyre, I, Chivers, M, and Fluter, M. Tran, TT, and Uribe, BP. A comparison of muscle activation
Shoulder muscle EMG activity during push up variations on and off between a Smith machine and free weight bench press. J Strength
a Swiss ball. Dyn Med 5: 7, 2006. Cond Res 24: 779–784, 2010.
18. Marshall, PW and Murphy, BA. Increased deltoid and abdominal 30. Schwanbeck, S, Chilibeck, PD, and Binsted, G. A comparison of free
muscle activity during Swiss ball bench press. J Strength Cond Res 20: weight squat to Smith machine squat using electromyography.
745–750, 2006. J Strength Cond Res 23: 2588–2591, 2009.
19. McBride, JM, Cormie, P, and Deane, R. Isometric squat force output 31. Uribe, BP, Coburn, JW, Brown, LE, Judelson, DA, Khamoui, AV,
and Nguyen, D. Muscle activation when performing the chest press
and muscle activity in stable and unstable conditions. J Strength
and shoulder press on a stable bench vs. a Swiss ball. J Strength Cond
Cond Res 20: 915–918, 2006.
Res 24: 1028–1033, 2010.
20. McBride, JM, Larkin, TR, Dayne, AM, Haines, TL, and Kirby, TJ.
32. van den Tillaar, R and Saetersbakken, A. The sticking region in
Effect of absolute and relative loading on muscle activity during stable three chest-press exercises with increasing degrees of freedom.
and unstable squatting. Int J Sports Physiol Perform 5: 177–183, 2010. J Strength Cond Res 26: 2962–2969, 2012.
21. McCaw, ST and Friday, JJ. A comparison of muscle activity between 33. Wahl, MJ and Behm, DG. Not all instability training devices
a free weight and machine bench press. J Strength Cond Res 8: 259– enhance muscle activation in highly resistance-trained individuals.
264, 1994. J Strength Cond Res 22: 1360–1370, 2008.
22. Morrissey, MC, Harman, EA, and Johnson, MJ. Resistance training 34. Welsch, EA, Bird, M, and Mayhew, JL. Electromyographic activity
modes: specificity and effectiveness. Med Sci Sports Exerc 27: 648– of the pectoralis major and anterior deltoid muscles during three
660, 1995. upper-body lifts. J Strength Cond Res 19: 449–452, 2005.

VOLUME 27 | NUMBER 7 | JULY 2013 | 1831

Copyright © National Strength and Conditioning Association Unauthorized reproduction of this article is prohibited.