A number of biomechanical studies have investigated the effects of manipulating features of the squat exercise to alter muscle activity and kinetic output. These manipulations include changes to foot position (20,25), barbell position (11), stability of the surface on which the exercise is performed (1,9,16–18), mode of the resistance (19,25), and depth of movement (5,22). In addition, further biomechanical studies have compared muscle activity and kinetic output of common variations of the squat exercise. Gullet et al. (11) compared performance of the back squat and front squat using the same relative intensity of 75% 1 repetition maximum (RM) for each exercise. Results from the study demonstrated that the back squat consistently enabled heavier loads to be lifted, despite similar muscle activity measured from the quadriceps and hamstrings during both exercises. In addition, muscle activity of the erector spinae (ES) which is commonly used to provide insight into the stress placed on the lower back was also similar for both the back squat and front squat despite substantial differences in load position and magnitude (11). In a similar study comparing muscle activity during the free-weight squat and Smith machine squat, Schwanbeck et al. (26) reported greater electromyographic (EMG) values for all muscles assessed in the lower leg, upper leg, and trunk during performance of the free-weight squat. Only values recorded from the gastrocnemius (GA), vastus medialis, and biceps femoris (BF) demonstrated statistical significance in comparisons made between the 2 exercises. However, when EMG values were summed over all 7 muscles assessed, the free-weight squat was shown to produce 43% more activity than the Smith machine squat. The authors suggested that the increased muscle activity recorded was most likely because of greater stabilization required when using free-weights (26). Kinematics and kinetics were not measured in the study, so it is unclear whether any differences in position of the body and load may have also contributed to the results.
In a recent study conducted by McBride et al. (19), the authors compared EMG and kinetic variables produced during performance of the back squat and box squat by experienced powerlifters. The same absolute loads were used for the performance of both exercises, and the kinetic analyses revealed similar peak force and peak power values across all loads tested. Comparisons of the EMG data revealed a high correspondence for values produced from the muscles of the thigh and posterior trunk. The authors concluded that both exercises produced a similar stimulus despite the fact that the box squat is not as effective in using the stretch-shortening cycle (19). It is likely that the capacity to use the same absolute load in each exercise was influential in producing a similar kinetic and EMG response.
Another squatting variation that is commonly used in the preparation of athletes but has received minimal empirical study is the overhead squat. In general, the overhead squat is included in the training of athletes as it has been suggested in several publications intended for coach education (2,4,13,27,32), to be a functional movement that can enhance athletic performance. In particular, the exercise is suggested to provide an increased stimulus for the trunk musculature in comparison with exercises such as the back squat (2,4,7,13). This view is based on the hypothesis that greater activation of the trunk musculature is required to maintain the load overhead (2,4,7,13,27). Mechanistically, this could occur because of an increase in distance from the trunk to the overhead load and thus presenting a greater mechanical challenge in compassion with the load positioned across the shoulders. Additionally, it is hypothesized by practitioners that any perturbations and shift in balance during execution of the less stable overhead squat will cause an increase in activation of the trunk musculature to recover equilibrium (4,13). However, despite widespread acceptance of the effectiveness of the overhead squat to recruit and provide an appropriate training stimulus for the trunk musculature, there have been no experimental studies conducted to test these hypotheses. In addition, it is likely that the substantially lower absolute load used during the overhead squat compared with the back squat will result in a reduced kinetic stimulus. Therefore, the purpose of this research was to compare EMG activity and kinetic output of the back squat and overhead squat. A selection of popular trunk isolation exercises were included to provide additional comparisons, while the kinetic output of the back squat and overhead squat were assessed through measurement of ground reaction force. It was hypothesized that heavier loads that could be lifted during the back squat would elicit greater muscular activity in the prime movers and result in greater force output compared with the overhead squat. In contrast, it was hypothesized that reduced stability of the barbell-lifter system during the overhead squat would result in greater muscular activity in the anterior and posterior compartments of the trunk.
Experimental Approach to the Problem
A within-subjects crossover design was used to compare kinetics and EMG activity of the shoulder, trunk, and leg musculature during performance of the back squat and overhead squat (exercises are illustrated in Figure 1). A selection of popular trunk isolation exercises were also included to provide additional comparisons for trunk EMG activity. Relative resistances for the back squat and overhead squat were calculated from a predetermined 3RM test. A 3RM test was selected as it has been used effectively in previous research (29) and may be more appropriate for assessing maximum performance during a technically demanding exercise such as the overhead squat. Relative loading was primarily used to compare muscle activity as it has been proposed that absolute loading is an inaccurate method when comparing exercises that feature large differences in magnitudes of the external loads (6,18). However, to assess whether potential differences in muscle activity could be caused by differences in load positioning or load magnitude, a single absolute load comparison was also included.
Each subject was required to complete 2 sessions separated by 1 week. During the first session, subjects performed 3RM testing for the back squat and overhead squat in a randomized order. The second session comprised performance of the 2 different squat variations with 3 relative loads (60, 75, and 90% 3RM). Each subject also performed a back squat trial with the same absolute load used for the 90% 3RM condition in the overhead squat. At the conclusion of the second testing session, the subjects also performed a selection of trunk isolation exercises to provide additional EMG comparisons.
Fourteen elite male rugby union players participated in this study (age, 26 ± 7 years; stature, 182.5 ± 13.5 cm; mass, 90.5 ± 17.5 kg; training history, 9 ± 7 years; overhead squat 3RM, 72.5 ± 17.5 kg; and back squat 3RM, 137.5 ± 27.5 kg). Each subject was competent in the back squat and overhead squat, with regular performance of each exercise in their previous 6 months of training. All subjects were recruited from the same Scottish Premier 1 League team. The study was conducted in-season, and because of the elite level of the team each participant completed 4 rugby training sessions and an 80-minute match within the study period. Subjects were notified about the potential risks involved and provided written informed consent to be included in the study. Prior institutional approval was obtained from the ethics review panel at Robert Gordon University, Aberdeen, United Kingdom.
A 3-minute stationary bike warm-up followed by 10 body weight squats and dynamic stretching were performed before the 3RM testing protocol (11,29). Subjects were asked to predict a 3RM load for both squat exercises. From these estimates, warm-up loads of 3 repetitions were performed at 50, 70, and 90%. Subjects then performed a 3RM protocol comprising up to 4 maximum efforts (29). During warm-up sets, subjects were permitted a minimum of 2 minutes and a maximum of 4-minute rest (11). The final score was recorded as the maximum load that could be lifted through the proper range of motion for 3 repetitions (18,31). All squats were performed to a depth where the top of the thigh became parallel with the floor (illustrated in Figure 1), as monitored visually by the researcher (7,11,31). During maximum efforts, the upper threshold of the rest period was increased to 5 minutes (26). The 3RM protocol for the back squat and overhead squat was performed in a randomized order with 15 minutes provided to recover before repeating the process with the second variation (17). All subjects were accustomed to perform multiple maximum strength tests in the same session as part of their ongoing physical training and assessment.
Maximum Speed Testing
Throughout the session, subjects were instructed to descend in a controlled manner over 3 seconds until the upper thighs were parallel with the floor. Upon meeting the depth criteria, subjects were instructed to lift the load as fast as possible while ensuring their feet did not leave the floor. Each trial was performed twice to assess intratrial reliability. In total, 7 different squat trials were performed in a randomized order. These included trials performed with 60, 75, and 90% 3RM for both exercises and a final trial where the back squat was performed with the same absolute load used for the 90% 3RM overhead squat. After completing these trials, subjects performed a selection of trunk isolation exercises (sit-up, Swiss ball jackknife, front plank, and side plank).
The sit-up was performed on a glute-hamstring apparatus (Life Fitness, Hammer Strength, Ely, Cambridgeshire, UK), where the subject was anchored at the ankles and supported across the gluteals. Subjects were required to descend over 3 seconds until horizontal and then were given instruction to explosively return to the starting position. The Swiss ball jackknife was performed in a prone position with feet supported on the Swiss ball. Subjects were instructed to extend the legs in a controlled manner over 3 seconds and then explosively flex at the hips and knees bringing the thighs toward the chest. The front and side planks were performed on flexed elbow(s) directly underneath the shoulder, maintaining a torso position that was parallel to the floor. All sessions were monitored by an accredited strength and conditioning coach who explained and demonstrated proper execution of each exercise.
Surface EMG signals were recorded with 2 circular pre-gelled electrodes with 20 mm interelectrode spacing (Ambu Blue Sensors N-10-A). Signals were recorded from the following 8 muscles: (a) anterior deltoid (AD) on the anterior aspect of the arm 4 cm below the clavicle; (b) rectus abdominis (RA) 2 cm lateral to the umbilicus; (c) external oblique (EO) lateral to the RA above the anterior superior iliac spine; (d) ES 2 cm lateral to the L3 vertebra; (e) gluteus maximus (GM) half the distance between the trochanter and sacral vertebrae in the middle of the muscle; (f) vastus lateralis (VL) 2–5 cm above the patella just lateral to the midline of the thigh; (g) BF lateral aspect of the thigh between the trochanter and posterior of the knee; and (h) lateral GA 2 cm lateral of the midline of the knee on the belly of the muscle. All electrode placements were administered on reference of previous research (8). Electrodes were placed on the muscle belly, parallel to the orientation of the fibers and on the right side of the body (17,18,26). Skin was shaved, abraded, and cleansed with acetone to improve signal conduction (1,18).
Maximum voluntary contractions (MVCs) were recorded for each of the muscles included in the EMG analysis. All MVC attempts were performed under isometric conditions and maintained for 5 seconds (1,23). The MVC for the AD was acquired as the subject attempted to perform an anterior shoulder raise with the arms fixed at 90° angle of anterior flexion using a loaded barbell. Trunk muscle MVCs were recorded with the subject reclined in a horizontal position anchored at the ankles and supported across the hip joint on a glute-hamstring apparatus (Life Fitness, Hammer Strength). Contraction for the RA was recorded in the supine position as the researcher manually provided a downward force applied to the chest and shoulders while the subject attempted to perform a sit-up. Contraction for the EO was recorded with the subject lying on their side as manual downward force was applied at the shoulder to resist lateral spinal flexion. Assessment for the ES was made with the subject in a prone position with manual force applied across the posterior deltoids to resist spinal extension. Subjects performed an isometric squat with the knee and hip fixed at approximately 135° to record maximum EMG values for the VL. Contraction for the BF was recorded using a glute-hamstring apparatus (Life Fitness, Hammer Strength) with the subject in a prone position with the knee flexed to 135° as manual resistance was applied by the researcher across the posterior shoulders. Finally, the MVC for the GA was recorded as the subject in a seated position with the knee and ankle fixed at 90° attempted to plantar flex against a fully loaded calf-raise machine (Life Fitness, Hammer Strength). All EMG signals were received at 2,000 Hz (Zero-wire EMG; Aurion, Milan, Italy), preamplified (overall gain = 500, common mode rejection ratio 115 dB, signal to noise ratio <1 μV RMS baseline noise), and stored using a 16-bit A/D card with a ±2.5 V range. Electromyographic data were filtered (6-pole Butterworth and band pass filtered 10–500 Hz) and full-wave rectified using signal processing software (Zero-wire EMG software 2.0.2). Muscle EMG activity was analyzed for each repetition over the eccentric and concentric phases. Values were normalized and presented as an average integrated percentage of that recorded during MVCs (Acqknowledge, version 220.127.116.11; BIOPAC System, Goleta, CA, USA).
Vertical ground reaction force (VGRF) for all squat repetitions was recorded using a single AccuPower portable force platform (AMTI, Watertown, MA, USA) with dimensions 76 × 102 × 12 (length × width × height, cm). In addition to measuring kinetics, the velocity of the movement was also calculated for the lifter and external load as a single system. This was achieved by incorporating the VGRF data and using the principle that the impulse applied to the system equals its change in momentum (15,30). Briefly, trials were initiated with subjects standing erect and motionless. Changes in vertical velocity of the system center of mass (COM) were calculated by multiplying the net VGRF (VGRF recorded at the force plate minus the weight of the system) by the intersample time period divided by the mass of the system. Instantaneous velocity at the end of each sampling interval was determined by summing the previous changes in vertical velocity to the preinterval absolute velocity, which was equal to zero at the start of the movement. The position change over each interval was calculated by taking the product of absolute velocity and the intersample time period. Vertical position of the system COM was then obtained by summing the position changes.
Intratrial reliability for each variable analyzed was assessed by intraclass correlation coefficients (ICCs). The ICCs were calculated with a correction factor for the number of trials administered (n = 2) and number of repetitions used in the criterion score (n = 1) (3). Potential differences in EMG and kinetic variables measured during the overhead squat and back squat were analyzed using a 2 × 3 (squat variation × load) repeated-measures analysis of variance (ANOVA). Potential differences in EMG values measured during the heaviest relative squatting loads and trunk isolation exercises were analyzed using a 1-way repeated-measures ANOVA. Significant main effects were further analyzed with Bonferroni-adjusted pair-wise comparisons. Statistical significance was accepted at p ≤ 0.05. All statistical procedures were performed using the SPSS software package (version 17.0; SPSS, Inc., Chicago, IL, USA).
Intratrial reliability for EMG data expressed over the eccentric and concentric phases was high (ICC ≥0.92 and ≥0.91, respectively). Strong reliability was also obtained for force and velocity data (ICC ≥0.84 and ≥0.82, respectively).
The AD exhibited significantly greater muscle activity for the overhead squat during all loading conditions in both eccentric and concentric phases compared with the back squat (eccentric: exercise effect
= 0.76, p < 0.001; load effect
= 0.59, p = 0.001; concentric: exercise effect
= 0.88, p < 0.001; load effect
= 0.41, p = 0.003; Table 1). A significant main effect for squat variation was found for the RA (eccentric: exercise effect
= 0.39, p = 0.003; load effect
= 0.67, p < 0.001; concentric: exercise effect
= 0.04, p = 0.465; load effect
= 0.07, p = 0.916) and EO (eccentric: exercise effect
= 0.42, p = 0.003; load effect
= 0.68, p < 0.001; concentric: exercise effect
= 0.09, p = 0.315; load effect
= 0.52, p < 0.001), with greater values obtained during the overhead squat. The greatest differences were obtained during the eccentric phase where the RA and EO displayed significantly greater muscle activity during the overhead squat for all loads tested. In contrast, muscle activity of the ES was significantly greater during both eccentric and concentric phases of the back squat at 60 and 75% 3RM, and during the concentric phase at 90% 3RM compared with the overhead squat (eccentric: exercise effect
= 0.42, p = 0.010; load effect
= 0.38, p = 0.003; concentric: exercise effect
= 0.70, p < 0.001; load effect
= 0.29, p = 0.016).
For the relative loads, the EMG activity of the lower-body muscles was consistently greater during the back squat compared with the overhead squat (Table 2). The GM displayed significantly greater muscle activity during the back squat across the whole repetition for all loads (eccentric: exercise effect
= 0.68, p < 0.001; load effect
= 0.63, p < 0.001; concentric: exercise effect
= 0.36, p = 0.047; load effect
= 0.14, p = 0.129). A similar profile was also obtained for the VL (eccentric: exercise effect
= 0.65, p < 0.001; load effect
= 0.65, p < 0.001; concentric: exercise effect
= 0.10, p = 0.263; load effect
= 0.11, p = 0.229); however, values tended to reach significance only during the eccentric portion of the movement. In contrast, significant differences in activity of the BF (eccentric: exercise effect
= 0.28, p = 0.041; load effect
= 0.54, p < 0.001; concentric: exercise effect
= 0.74, p < 0.001; load effect
= 0.50, p = 0.001) and GA (eccentric: exercise effect
= 0.23, p = 0.072; load effect
= 0.40, p = 0.001; concentric: exercise effect
= 0.54, p < 0.001; load effect
= 0.24, p = 0.028) were generally obtained during the concentric phase of the movement only.
Comparisons of EMG activity during the back squat and overhead squat when using the same absolute load revealed less disparate results than those obtained with relative loading. As expected, AD muscle activity was significantly greater in the overhead squat across the whole repetition (exercise effect
= 0.90, p < 0.001; Table 3). For the trunk musculature, values obtained for both squats were relatively similar; however, small significant differences were obtained for the EO and ES. Significantly greater muscle activity in the EO was measured during the overhead squat across both the eccentric and concentric phases of the movement (exercise effect
= 0.62, p < 0.001). In addition, significantly greater muscle activity was measured in the ES during the eccentric phase of the overhead squat (exercise effect
= 0.24, p = 0.068). No significant main effects of exercise type were found for any of the lower-body muscles. This result was influenced by the phase of the movement, as EMG activity in all lower-body muscles was significantly greater in the overhead squat during the eccentric phase of the movement only (Table 4).
Comparison of Squat Exercises and Trunk Isolation Exercises
The greatest EMG activity in the RA muscle was recorded during the sit-up, with significantly greater values obtained relative to all other exercises measured (exercise effect
= 0.90, p < 0.001; Figure 2). The ski-tuck, front plank, and side plank also produced significantly greater EMG activity in the RA muscle compared with the back and overhead squat performed with 90% 3RM loads. Similar results were also obtained for the EO muscle, with significantly greater EMG activity recorded during all trunk isolation exercises compared with the squats (exercise effect
= 0.93, p < 0.001). In contrast, muscle activity of the ES was much larger during the heavy squat load conditions than measured during the trunk isolation exercises (exercise effect
= 0.99, p < 0.001).
Significant main effects of load and squat variation were obtained for both force and velocity values (Table 4). Significantly greater peak force values were measured during the back squat compared with the overhead squat (exercise effect
= 0.94, p < 0.001; load effect
= 0.45, p = 0.001). In contrast, significantly greater peak velocity values were measured during the overhead squat (exercise effect
= 0.25, p = 0.026; load effect
= 0.57, p = 0.011). Analysis of the kinetic data for the back squat and overhead squat using the same absolute load revealed a less disparate comparison (Table 5). No significant differences were noted in peak force production between the 2 exercises (exercise effect
= 0.04, p = 0.465). However, significantly greater peak velocity values were produced during the back squat (exercise effect
= 0.66, p < 0.001).
This is the first study to compare muscle activity and kinetics between the back squat and overhead squat. The results confirmed the hypothesis that heavier loads used with the back squat elicit a greater kinetic stimulus and increased muscular activity in the prime movers. Furthermore, the results provided limited support for the hypothesis that performance of the overhead squat creates greater muscular activity of the trunk in comparison with the back squat. Greater activity was observed from the anterior trunk muscles during the overhead squat; however, differences in magnitudes were relatively low. Comparisons made between the EMG activity of the anterior trunk muscles during both squat exercises and isolation exercises routinely performed to target the trunk musculature revealed substantially larger muscle activity in the RA and EO during performance of the isolation exercises. In contrast, muscle activity of the ES was significantly greater during squats compared with the trunk isolation exercises, with the largest values consistently obtained for the back squat. The results of this study demonstrate that the EMG and kinetic stimulus created when performing the overhead squat is different to that created when performing the back squat. The following discussion will explore these differences and their practical implications.
The results from this study confirm practitioners' suggestions (2,4,7,13) that the overhead squat elicits greater trunk muscle activation in comparison with the back squat, at least in the anterior compartment. Although the overhead squat was shown to elicit significantly greater muscle activity in the RA and EO, the magnitude of the differences observed between the squats was, in general, small. The difference in EMG activity of the RA and EO between the overhead squat and back squat ranged from 1.3 to 2.5% and from −0.7 to 7.6% (scaled to the activity produced during the MVC action; Table 1), respectively. Direct comparisons with previous research investigating trunk muscle activity while squatting is problematic due to differences in EMG analysis, loads used, and mode of exercise performed. Anderson and Behm (1) compared EMG activity of the prime movers and trunk muscles during the back squat performed in progressively more unstable conditions. The most stable condition was categorized as the Smith machine squat, followed by the free-weight squat performed on a firm surface, and finally the free-weight squat performed on labile balance discs. Results from the study demonstrated that trunk muscle activity progressively increased as the exercise was performed in less stable conditions. The authors concluded that increases in muscle activity were required to control the movement of the body and resistance in greater spatial dimensions and under more challenging circumstances. In a more recent study conducted by Comfort et al. (7), the authors investigated muscle activity of the RA during performance of the back squat, front squat, and military press with the same absolute load of 40 kg. Electromyographic activity of the RA was similar between squatting techniques but doubled when performing the overhead movement of the military press. The authors suggested that the overhead movement increased activity of the RA to assist with stabilizing the torso during a task that required constant adjustments to maintain balance (7). Collectively, the results of these previous studies and the present investigation indicate that trunk muscle activity can be augmented by manipulating an exercise to increase the mechanical challenge to maintain stability, in some exercises this challenge can be obtained by lifting the load overhead.
In contrast to the increased activity observed for the anterior trunk muscles during the overhead squat, the results from this study demonstrated that muscle activity of the ES (the primary posterior trunk muscle) was significantly greater during performance of the back squat (Table 1). A significant main effect was also found for load, with greater ES muscle activity recorded for both squat variations as the resistance increased. During the concentric phase of the 90% 3RM condition, muscle activity was close to maximum during the back squat with an average value of 94.7% obtained for the group (vs. 79.8% for the overhead squat in the associated condition). The results from this study highlight that the greater ES activity measured during the back squat is primarily influenced by differences in the loads used with each exercise. In particular, the heavier loads lifted during the back squat have the potential to create a greater resistive torque at the L5/S1 joint, thereby requiring increased muscle activity in the ES to maintain control of the trunk. In addition, when comparing the ES activity between the overhead squat and back squat using the same absolute load, EMG values were similar between exercises with significantly greater values obtained for the overhead squat during the eccentric phase of the movement (Table 3). Again, this difference most likely reflects the increased stability demands of the overhead squat, which were presumably outweighed by the effects of the load magnitude in the relative resistance comparisons.
A distinctive aspect of this study was the inclusion of popular trunk isolation exercises to compare muscle activity of the RA, EO, and ES with both squat variations. Interestingly, the results demonstrated that despite the use of near maximal external loads when squatting, anterior trunk muscle activity was significantly greater during unloaded trunk isolation exercises. Indeed, muscle activity of the RA was significantly greater in each of the trunk isolation exercises compared with the 90% 3RM squat conditions (Figure 2). Similar results were previously identified by Comfort et al. (7) who reported that RA muscle activity was significantly greater during the front plank compared with the back squat and front squat. However, a limitation of this previous research was the use of a 40 kg absolute load for the squatting movements. This study extends these results and demonstrates that even with the application of near maximum loads, RA activity during the squat is substantially less than that achieved using unloaded trunk isolation exercises. In this study, significantly greater ES muscle activity was found for all squat conditions compared with the unloaded trunk isolation exercises, a finding that is also supported by previous research (12,23). The different pattern of results obtained for the anterior and posterior trunk muscles for the loaded and unloaded exercises reflects the mechanical effects caused by the position of the external resistance. During the squatting exercises, the resistance created by the external load is anterior to the L5/S1 joint and therefore creates a trunk flexion torque that must be resisted by the ES. The external load does not have the same direct effect on the anterior trunk muscles, and therefore it appears that isolation exercises that induce greater muscle shortening or increase the demand to stabilize the pelvis can produce greater activation in these muscles.
Electrographic analysis of the lower-body muscles revealed that as expected significantly greater activity was achieved during the back squat (Table 2). The results from this study can be used to demonstrate that this is primarily a consequence of the magnitudes of the loads used for each squatting exercise. Significant differences in EMG activity of lower-body muscles were largely confined to the concentric portion of the movement where the lifter attempted to accelerate the load as fast as possible. The loads were substantially lighter during the overhead squat, and therefore significantly greater velocities were obtained with this exercise (Table 4). Previous research has demonstrated that contraction velocity exhibits a positive association with EMG activity (10). However, EMG measurements recorded during comparisons with relative resistances were significantly greater for all lower-body muscles during the concentric phase of the back squat (i.e., the phase were peak velocity was produced). In contrast, when the EMG activity of the lower-body muscles was compared during squats performed with the same absolute load, no significant differences in EMG activity of the lower-body muscles were obtained during the concentric phase of the movements. Importantly, this change in result occurred despite the fact that the reduced load for the back squat resulted in substantially greater velocities. In addition, it was observed for the same absolute load comparison that EMG values from the lower-body muscles were significantly greater for the eccentric phase during the overhead squat compared with the back squat. Collectively, the change in results between the relative and absolute load comparisons demonstrates that the greater capacity to activate the lower-body musculature during back squats in comparison with overhead squats is because of the heavier loads that can be lifted with the former exercise.
The analysis of force and velocity data produced during the back squat and overhead squat corresponds with the EMG analysis and further highlights the importance of load magnitude. Peak force values were significantly greater during the back squat (Table 4), whereas significantly greater peak velocity values were obtained during the overhead squat. Both results are readily explained by differences in the magnitude of the external load when using a percentage of the athletes' maximum performance. The heavier absolute loads used in the back squat reduced movement velocity, and as a result of the force-velocity relationship exhibited at the muscle and whole-movement level, greater force was produced (33). Comparisons made when using the same absolute load (Table 5) revealed no significant differences in peak force and that peak velocity was now significantly greater during the back squat. The finding that the same absolute load was lifted at a faster velocity during the back squat most likely reflects the more stable configuration adopted in comparison with the overhead squat.
An important limitation of this study is the use of surface EMG to assess muscular activity of a complex segment such as the trunk. In particular, with a multilayered segment, surface EMG is likely to simultaneously record adjacent muscle activity and target muscle activity (cross talk), thereby potentially generating inaccurate recordings (24). In addition, the trunk uses both global and local musculature in a synergistic manner to support and move the spine (14). As a result, it has been suggested that activity of important local musculature such as the transverse abdominis cannot be readily assumed when using surface EMG on global muscles such as the RA (21,28). However, research regarding the extent to which there is correspondence between the activity of local and global muscles of the trunk is contradictory, with the majority of studies focusing on low-intensity movements such as spinal extension/flexion and unloaded squatting (21,24,28). In contrast, the intensities used in this investigation are high, and it is unclear the extent to which EMG activity recorded from surface muscles reflects the overall stimulus received by the trunk musculature. At present, surface EMG is the most widely used method for analysis of muscle activity during resistance exercise, but the aforementioned limitations must be considered.
The results of this study clearly demonstrate that the kinetic profile and EMG stimulus created when performing the overhead squat are different from that created when performing the back squat. The findings have implications for exercise selection when designing sport-specific training programs. The results of the study suggest that where the primary goal of training is the development of strength, substituting the overhead squat for the back squat in a program is unlikely to be warranted. In addition, it does not appear prudent to substitute the overhead squat for the back squat when the goal is to increase recruitment of the muscles of the trunk. This research demonstrates that such a substitution would have minimal effect on the anterior trunk musculature and decrease recruitment of the posterior aspect of the trunk. Instead, for training the trunk musculature, it is recommended that training programs include the back squat and supplement with various trunk isolation exercises. It is important to note that there may be benefits associated with the overhead squat that are not investigated in this study. In particular, the overhead squat may be an effective exercise for developing the shoulder complex, enhancing proprioception, and improving mobility of the hip, knee, and ankle. Further research is required to establish if these potential benefits and others can be obtained through training programs featuring the overhead squat.
The results of this study do not constitute endorsement by the authors or the National Strength and Conditioning Association.
1. Anderson K, Behm DG. Trunk muscle activity increases with unstable squat movements. Can J Appl Physiol 30: 33–45, 2005.
2. Auferoth SJ, Joseph J. The overhead squat. NSCA J 10: 24–25, 1988.
3. Baumgartner TA. Reliability and error of measurement. In: Measurement Theory in Practice and Kinesiology. Wood T.M., Zhu W., eds. Champaign, IL: Human Kinetics, 2006. pp. 27–52.
4. Brown T. Core strength: Learning the overhead squat. NSCA Perform Train J 5: 21–23, 2006.
5. Caterisano A, Moss RF, Pellinger TK, Woodruff K, Lewis VC, Booth W, Khadra T. The effect of back squat depth on the EMG activity of 4 superficial hip and thigh muscles. J Strength Cond Res 16: 428–432, 2002.
6. Clarke DR, Lambert MI, Hunter A. Muscle activation in the loaded free barbell squat: A brief review. J Strength Cond Res 26: 1169–1178, 2012.
7. Comfort P, Pearson SJ, Mather D. An electromyographical comparison of trunk muscle activity during isometric trunk and dynamic strengthening exercises. J Strength Cond Res 25: 149–154, 2011.
8. Cram JR, Kasman GS. Introduction to Surface Electromyography. Gaithersburg, MD: Aspen, 1998.
9. Drinkwater EJ, Pritchett EJ, Behm DG. Effect of instability and resistance on unintentional squat-lifting kinetics. Int J Sports Physiol Perform 2: 400–413, 2007.
10. Ebersole KT, O'Conner KM, Weir AP. Mechanomyographic and electromyographic responses to repeated concentric muscle actions of the quadriceps femoris. J Electromyogr Kinesiol 16: 149–157, 2006.
11. Gullett JC, Tillman MD, Gutierrez GM, Chow JW. A biomechanical comparison of back and front squats in health trained individuals. J Strength Cond Res 23: 284–292, 2009.
12. Hamlyn N, Behm DG, Young WB. Trunk muscle activation during dynamic weight-training exercises and isometric instability activities. J Strength Cond Res 21: 1108–1112, 2007.
13. Hasegawa I. Using the overhead squat for core development. NSCA Perform Train J 3: 19–21, 2004.
14. Imai A, Kaneoka K, Okubo Y, Shiina I, Tatsumura M, Izumi S, Shiraki H. Trunk muscle activity during lumbar stabilization exercises on both a stable and unstable surface. J Orthop Sports Phys Ther 40: 369–375, 2010.
15. Kawamori N, Crum A, Blumert P, Kulik JR, Childers JT, Wood JA, Stone MH, Haff GG. Influence of different relative intensities on power output during the hang power clean: Identification of the optimal load. J Strength Cond Res 19: 698–708, 2005.
16. Kholer JM, Flanagan SP, Whiting WC. Muscle activation patterns while lifting stable and unstable loads on stable and unstable surfaces. J Strength Cond Res 24: 313–321, 2010.
17. McBride JM, Cormie P, Deane R. Isometric squat force output and muscle activity in stable and unstable conditions. J Strength Cond Res 20: 915–918, 2006.
18. McBride JM, Larkin TR, Dayne AM, Haines TL, Kirby TJ. Effect of absolute and relative loading on muscle activity during stable and unstable squatting. J Sports Physiol Perform 5: 177–183, 2010.
19. McBride JM, Skinner JW, Schafer PC, Haines TL, Kirby TJ. Comparison of kinetic variables and muscle activity during a squat vs. a box squat. J Strength Cond Res 24: 3195–3199, 2010.
20. McCaw ST, Melrose DR, Stance width and bar load effects on leg muscle activity during the parallel squat. Med Sci Sports Exerc 31: 428–436, 1999.
21. McGill S, Juker D, Kropf P. Appropriately placed surface EMG electrodes reflect deep muscle activity (psoas, quadratus lumborum, abdominal wall) in the lumbar spine. J Biomech 29: 1503–1507, 1996.
22. McKean MR, Dunn PK, Burket BJ. The lumbar and sacrum movement pattern during the back squat exercise. J Strength Cond Res 24: 2731–2741, 2010.
23. Nuzzo JL, McCaulley GO, Cormie P, Cavill MJ, McBride JM. Trunk muscle activity during stability ball and free weight exercises. J Strength Cond Res 22: 95–102, 2008.
24. Okubo YU, Kaneoka K, Imai A, Shiina I, Tatsumura M, Izumi S, Miyakawa S. Comparison of the activities of the deep trunk muscles measured using intramuscular and surface electromyography. J Mech Med Biol 10: 611–620, 2010.
25. Paoli A, Marcolin G, Petrone N. The effect of stance width on the electromyographical activity of eight superficial thigh muscles during squat with different bar loads. J Strength Cond Res 23: 246–250, 2009.
26. Schwanbeck S, Chilibeck PD, Binsted G. A comparison of a free weight squat to a smith machine squat using electromyography. J Strength Cond Res 23: 2588–2591, 2009.
27. Stevenson G. Deep overhead squat behind neck press. J UK Strength Cond Assoc 16: 18–19, 2009.
28. Stokes IA, Henry SM, Single RM. Surface EMG electrodes do not accurately record from lumbar multifidus muscles. Clin Biomech (Bristol, Avon) 18: 9–13, 2003.
29. Swank AM, Funk D, Baily C, Pinkham KK, Soldner KR. Cardiovascular and subjective responses to one- and three-repetition maximum strength testing. J Clin Exerc Physiol 4: 96–100, 2002.
30. Swinton PA, Stewart A, Agouris I, Keogh WL, Lloyd R. A biomechanical analysis of the straight and hexagonal barbell deadlifts using submaximal loads. J Strength Cond Res 25: 2000–2009, 2011.
31. Yule T. The back squat. J UK Strength Cond Assoc 1: 11–15, 2005.
32. Yule T. The overhead squat. J UK Strength Cond Assoc 5: 6–7, 2007.
33. Zink AJ, Perry CA, Robertson BL, Roach KE, Signorile JF. Peak power, ground reaction forces, and velocity during the squat exercise performed at different loads. J Strength Cond Res 20: 658–664, 2006.