Introduction
Although recruitment of the gluteus maximus (GM) and hamstring group (HG) control hip and knee stability and mobility, a primary role of these muscles is to extend the hip. Research has shown that weakness at the hip can lead to poor performance and lower extremity dysfunction and injury (16,30,36). Weight-bearing, free-weight exercise is arguably used most often to increase hip strength for sport performance and reduce the risk of injury, but the exercises that produce the greatest muscle recruitment are not clear. Variations in the type of stance and the use of single- (hip alone) and multi-joint (hip and knee) during weight-bearing exercise likely affects hip muscle recruitment levels but also needs further investigation. These modifications may alter the contribution of each hip extensor muscle that may lead to differences in strength and hip and knee injury prevention.
A key difference between the GM and HG is monoarticular vs. biarticular joint control, respectively, at the hip and knee (11). The GM reduces the risk of injury by stabilizing the sacroiliac joint, preventing hip adduction and internal rotation during weight-bearing actions, which prevents excessive knee valgus. Along with hip extension, the HG flexes the knee, prevents anterior tibia shearing, and controls knee rotations while providing medial and lateral knee support (13,38). Given these differences, it is imperative to identify the exercises that best recruit these muscle groups for the purpose of implementing resistance training programs with the goal to obtain these specific desired outcomes.
Previous research has revealed that GM and HG recruitment can be isolated primarily with single-joint actions (17,18,27). Gluteus maximus activation is relatively high during the prone, hip extension exercise (40), and while abducting the hip to 30–60° with knee flexion before hip extension (17). Commonly used as a maximum voluntary isometric contraction (MVIC), knee flexion before hip extension is suggested to minimize the contribution of the HG to hip extension by reducing the muscle length in application of the length-tension relationship. Non–weight-bearing exercises during isolated knee flexion have been shown to significantly activate the HG (18,27). However, these hip and knee exercises may have a limited effect on multi-joint, weight-bearing performance (39).
In contrast, weight-bearing, multijoint exercises (i.e., bilateral squat [BS]) can arguably be considered the most beneficial for performance and injury prevention during activities involving gait. Yet, research shows low activation levels in the HG in comparison to the quadriceps during the BS (8,32). Reciprocal inhibition from high quadriceps activation and minimal change in length after simultaneous hip and knee flexion followed by extension have been suggested to limit HG activation during the squat (41,42).
Few studies have compared the level of GM to HG activation during the squat (2,3,14,21,31,35) and other weight-bearing, multijoint exercises (1,7,20,34). The majority of the studies that have analyzed GM and HG activation have used only body weight as resistance that is typically intended as rehabilitation exercise (1,7,12,14,20). The contribution of each muscle to a joint action is likely different at higher relative intensities. In addition, few studies have included females as subjects, who have demonstrated lack of hip strength (19) and biomechanical differences in comparison to males during lower extremity exercise (4,24,29,30,43). Furthermore, GM and HG activation may differ when using an anterior-posterior stance (i.e., lunge, step-up) compared with a medial-lateral stance but direct comparison among these exercises in the same study is limited at higher relative intensities (5). Thus, the contribution of the GM and HG to hip control during these variations in stance is not clear and requires further investigation. Previous research also indicates that muscle recruitment levels at a joint can vary during single-joint vs. multi-joint actions (41). Therefore, the purpose of this study was to compare the GM and HG activation during 2 weight-bearing, multi-joint resistance exercises (BS and modified single-leg squat [MSLS]), and a single-joint, weight-bearing exercise (stiff-leg deadlift [SLDL]).
Methods
Experimental Approach to the Problem
Three commonly used weight-bearing resistance exercises were included to analyze surface electromyographic (EMG) activation of the GM and HG. These exercises were included because most sports involve weight-bearing activity. An 8 repetition maximum (RM) load was used to observe the EMG response at a relative intensity typically prescribed to increase strength and muscle size for improved sport performance and prevention of injury. The GM and HG are hip extensors but also stabilize the hip and knee during the stance phase of various activities. Evidence suggests that the activation of the GM and HG is important for the prevention of knee injury (30); therefore, females, known to have a higher rate of knee injury compared with their male counterparts, were selected to be subjects. Gluteus maximus and HG activation were compared with each exercise to determine the relative activation levels. The type of stance was a factor in the selection of the exercises. A medial-lateral stance used in the SLD and BS and the anterior-posterior stance used in the MSLS might significantly influence muscle activation levels during hip flexion and extension exercise. The effect of single- vs. multi-joint, lower-extremity motion was also considered in exercise selection.
Subjects
Eighteen women (SD ± age 20.92 ± 1.1 years, ht 165.30 ± 5.51 cm, mass 61.84 ± 6.37 kg) with previous resistance training experience (range 1–5 years) before the study completed the study. All the subjects had previous participation in a competitive sport but were not currently participating in that sport. A few of the subjects were currently active intermittently in resistance training, but none of the subjects were consistently training to maximize strength or athletic performance at the time of the study. Thus, the subjects were classified as untrained in resistance training. To be included in the study, the subjects had to complete all lifts with proper form. Subjects with current or previous lower extremity injury that would limit maximum performance on the exercises were excluded. All procedures and informed written consent form were approved by Texas State University's internal review board. All subjects volunteered and were informed of the benefits and risks of the procedures before signing the approved written informed consent form to participate.
Procedures
Surface EMG was collected using wireless Trigno IM Sensors using EMMA software (Delsys, Natick, MA). Containing a built-in triaxial accelerometer, the sensors also collected hip and knee 3-D motion, which was synchronized with the EMG data. Each sensor also contained a Butterworth filter with a high-pass corner frequency of 20 Hz (40 dB/dec) and a low-pass corner frequency of 450 Hz (80 dB/dec). Electromyographic data were collected with a sampling rate of 1,111 Hz, whereas motion was collected at 148 Hz. A base-station receiver with a 40 m range collected the data that were transferred to a laptop through USB connection. This collected data were exported to excel files followed by being imported to EMGworks Data Acquisition and Analysis software (Delsys) for further processing. Within this program, mean concentric, eccentric, and combined EMG levels across 3 repetitions were determined using root mean square calculation. Using separate data plots of motion and EMG, the software allowed curser placement to determine the specific range of motion that was synchronized to the EMG levels. Electromyographic data were normalized with conversion to a percentage of the MVIC for each muscle.
To collect hip motion data, a sensor was placed above the sacrum on L5 and on the lateral thigh approximately 6 in below and in line with the greater trochanter. For knee motion, a sensor was placed 2 inches above the lateral malleolus. For EMG analysis, one sensor was placed on the GM belly parallel to the muscle fibers. Placement of the HG sensors occurred using the positions used in recent research to represent the entire group of muscles (33). An alcohol wipe was used to clean each site, and double-sided tape was used to improve sensor application to the skin.
Familiarization Session
The initial session was used to obtain informed consent signatures and collect age, height, and mass data. Instructions were provided to successfully perform each exercise, which was practiced using the free-weight bar and light weight. Measurements were taken for placement of the support step used to elevate the trail leg on the MSLS (also called rear foot elevated split squat). The subjects were instructed to abstain from participation in resistance training during the study and given instructions to wear appropriate clothing and shoes for resistance exercise. Subjects were instructed to maintain a healthy dietary routine and maintain hydration. Testing took place mid-morning and early-afternoon hours and at the same time for each subject.
Strength Assessment
After the familiarization session and 48 hours of rest, the subjects completed a strength assessment on the SLDL (8RM) and either the BS (3RM) or MSLS (3RM) in a randomized order. A final assessment took place after 48 hours of rest to complete the MSLS or BS. A 5-minute jog and dynamic stretching protocol was completed before all strength and EMG data collection. All strength assessments included 2 warm-up sets that progressed from 40 to 50% of the subject's estimated 1RM while completing 3–5 repetitions. Based on the ease of completing the second warm-up set, the weight was increased approximately 10% for the initial set. For the SLDL, 8 repetitions were completed each set with a 3–4-minute recovery between trials. Approximately 10% was added for each successful trial. The subject had to complete all repetitions with proper form. Three repetitions were completed each set on the BS and MSLS. All strength measures occurred within 5 sets. For the SLDL, the subjects maintained knee flexion near complete knee extension and a neutral trunk. With an alternated grip at shoulder width, the subject flexed the hip until the trunk was near parallel to the floor before returning to the standing position (41). For the BS and MSLS, the lead knee was in line with the toe while completing a parallel squat position (23). A shoulder-width stance was used for the BS. For the MSLS, the trail leg was supported with the ankle in complete plantar flexion while making contact at the metatarsophalangeal joint and toes (23). The height of the step was equal to the height of the subject's tibial tuberosity. A spotter was placed behind and on both sides of the subject.
Electromyographic Data Collection
Electromyography was collected after a minimum of 48 hours of rest after the last strength assessment. Maximum voluntary isometric contraction took place after the warm-up and sensor placement. Gluteus maximus MVIC was taken with the subject prone on a lab table, whereas the hip was positioned at 0° extension and the knee was flexed at 90° (40). The subject was stabilized by spotters holding the uninvolved leg and the upper body. Manual resistance was placed at the distal femur. The subject gradually increased hip extension force until maximum effort was reached within 1–2 seconds and held for approximately 3 seconds during a 5-second trial. Two trials took place with a 1-minute rest between trials, and the highest mean EMG level over a 1-second period was recorded as the MVIC. Hamstring group MVIC occurred with subject prone and the hip and knee extended. Manual resistance was applied at the ankle as the subject attempted to flex the knee (33).
After a 5-minute rest, the subjects completed the 3 exercises in random order with 5 minutes between each exercise. Three repetitions were completed for each exercise using an 8RM load. The same technique used during strength assessments was used to complete the EMG data. The repetitions were performed with a 2-second descent and ascent.
Statistical Analyses
The dependent variables in this study were the percent of maximum EMG measures for each of the 3 lifts: (a) BS, (b) modified squat, and (c) deadlift. The 2 independent variables were (a) type of contraction (concentric vs. eccentric), and (b) muscle group (hamstring vs. gluteus). Both of these independent variables are within-subjects (repeated) variables. There were no between-subjects independent variables. Bartlett's test for equal variances and the Shapiro-Wilk test for normality were used to determine whether the EMG measures met the basic assumptions for an analysis of variance (ANOVA). The results of these tests are reported in Table 1.
;)
Table 1.: Tests for the basic assumptions of analysis of variance.*
Cronbach's alpha was used to determine the test-retest reliability across 3 trials for each site. The reliability coefficients for the EMG measures ranged from 0.85 to 0.99. Based on these high values, the EMG measures for each site were determined to be highly reliable, and suitable for analysis.
For each dependent variable, a 2-way ANOVA with repeated measures was used to determine differences between the 2 types of contractions, the 2 muscle groups, and the interaction between contraction and muscle group. Greenhouse-Geisser epsilon was used to adjust probability values for any variation in sphericity among the trials. Paired t-tests were used as post hoc comparisons between trials. Partial η2 was used to determine effect size for each statistical test. All statistical significance was defined as p ≤ 0.05.
Results
Tests for the basic assumptions of ANOVA are reported in Table 1, and indicate that the EMG measures for all data met both the assumptions of equal variances and normality. Table 2 reports the GM and HG descriptive values across trials for the combinations of concentric and eccentric contractions during the BS. Repeated measures ANOVA indicated a significantly higher concentric vs. eccentric contractions, F(1,18) = 45.4, p < 0.001, partial η2 = 0.89, a very large effect. There was also significantly greater GM EMG than HG, F(1,17) = 9.2, p = 0.008, partial η2 = 0.87, another very large effect. In addition, there was a significant 2-way interaction between type of contraction and muscle group, F(1,17) = 6.8, p = 0.018, partial η2 = 0.29, a large effect. The difference in the concentric EMG between the GM and HG was greater (21.7%) than the difference in eccentric contractions (12.4%).
Table 2.: Descriptive values for BS.*
Table 3 reports the descriptive values across trials for the MSLS. Electromyography was greater in the concentric contractions, F(1,18) = 107.9, p < 0.001, partial η2 = 0.80, a very large effect. Gluteus maximus EMG was also greater than the HG EMG, F(1,17) = 7.9, p = 0.012, partial η2 = 0.79, another very large effect. Table 3 shows that the 26.1% difference between the concentric contractions in the GM and HG was not significantly different than the 19.7% difference in eccentric contractions between the GM and HG, F(1,17) = 3.8, p = 0.069, partial η2 = 0.18.
Table 3.: Descriptive values for MSLS.*
Table 4 reports the descriptive values across trials for the SLDL. Concentric contractions were significantly higher than the eccentric, F(1,18) = 54.1, p < 0.001, partial η2 = 0.89, a very large effect. Gluteus maximus EMG was also greater than HG, F(1,17) = 4.6, p = 0.047, partial η2 = 0.70, another very large effect. Also, the 11.3% difference in the concentric contractions between the GM and HG was not greater than the 9.9% difference in the eccentric contraction between the GM and HG, F(1,17) = 0.07, p = 0.790, partial η2 = 0.004.
Table 4.: Descriptive values for SLDL.*
For the combined concentric and eccentric HG measures, a significant difference was observed among the 3 exercises, F(2,35) = 25.7, Greenhouse-Geisser epsilon = 0.665, p < 0.001, partial η2 = 0.766, a very large effect. The mean values reported in Tables 2–4 indicate that HG EMG values were the smallest for the BS (24.4 ± 10.6) and the largest for the MSLS (40.1 ± 10.8). Post hoc tests indicated that the MSLS EMG was significantly greater than the SLDL EMG, t(18) = 3.61, p = 0.001, whereas SLDL values were also significantly greater than the BS values t(17) = −2.99, p = 0.004. When the same comparisons were made with separate concentric and eccentric hamstring measures, almost identical results were observed.
For the combined concentric and eccentric GM EMG, another significant difference was observed among the 3 exercises, F(2,34) = 20.9, Greenhouse-Geisser epsilon = 0.661, p < 0.001, partial η2 = 0.825, a very large effect. The mean values reported in Tables 2–4 indicate that a significant difference in GM EMG values was observed for the MSLS (65.6 ± 15.1) and BS (40.3 ± 17.7). Post hoc tests revealed that the MSLS values were also significantly higher than the SLDL values, t(17) = 4.38, p < 0.001, but the SLDL and BS values were not significantly different, t(17) = −0.09, p = 0.466. When the same comparisons were made with separate concentric and eccentric GM measures, almost identical results were observed.
Discussion
The findings of this study revealed that the GM demonstrated significantly higher relative muscle activation in comparison to the HG across the 3 exercises. In addition, the MSLS produced significantly greater GM and HG activation than the BS and SLDL. These results were consistent across concentric, eccentric and combined. Concentric contractions were also significantly greater than eccentric contractions.
Although the GM and HG are active during hip flexion and extension in all 3 exercises, the GM seems to be the primary hip extensor. These results are in agreement with previous research that have analyzed both muscle groups during weight-bearing exercises (2,3,7,34,35). The GM is a monoarticular muscle that is also active during hip external rotation and abduction, whereas the HG is biarticular producing hip extension and providing direct knee stabilization with medial and lateral insertions across the knee. Research indicates that the primary role of the HG during the stance phase is to transfer forces across the joints of the lower extremity and increase activation for hip extension when needed (16). Using modeling techniques during gait analysis, Jonkers et al. (16) determined that the biceps femoris is able to increase activation if the GM is weak during hip extension of the stance leg. Although non–weight-bearing exercises designed to increase hip strength have not shown kinematic changes related to lower extremity injury prevention (9), weight-bearing exercise could possibly produce more effective kinematic change during the stance phase.
During the stance phase of sprinting, jumping, and sled pulling (26), GM activation was also higher than HG activation. Findings from Niinimaki et al. (25) revealed that GM cross-sectional area was greater in athletes who were involved in high impact and loading during weight-bearing activity. In addition to our results, these investigations reveal that the GM is the prime mover at the hip for a variety of weight-bearing activities. In comparison to these studies, our results demonstrated similar relative activation levels in the GM and HG during weight-bearing resistance exercises.
Body-weight resistance (1,7,12,14,20) or light loads (2) have been used in the majority of the studies that have analyzed the GM and HG during weight-bearing exercises. Compared with our results produced using an 8RM load, these studies found low activation in the GM (≤35% MVIC) and HG (≤15% MVIC). Using a higher relative intensity than these previous studies, we found higher percent MVIC levels in the combined concentric and eccentric phases for the GM (41–63%) and HG (24–40%) across the 3 exercises. The relatively low activation from the HG indicates that other types of resistance exercise may be necessary to elicit HG adaptations. Further research is needed to determine the relative activation and contribution necessary from the HG to provide sufficient knee support during various sport-related maneuvers.
Most research investigating the GM and HG at higher loads have focused on the BS (3,22,28,44). Biomechanical differences in squat patterns tend to exist between males and females but with the exception of Contreras et al. (3), these previous studies included only male subjects. Contreras et al. (3) used a 10RM load comparing the BS and hip thruster (transverse load at the hip in a bridge position) and found 37 and 16% of the MVIC in the GM and biceps femoris activation, respectively. McCaw and Melrose (22) and Paoli et al. (28) found a significant increase in GM levels with an increase in stance width, but normalization of the EMG using MVIC was not conducted in these previous studies (22,28,44); thus, comparison of relative intensity between the GM and HG was not determined and cannot be used to compare with our study.
In contrast to the BS, the MSLS is performed with a narrow, anterior-posterior stance that requires muscular control for frontal plane stability (12,23). The MSLS produced significantly higher GM and HG activation compared with BS and SLDL levels. McCurdy et al. (23) also found higher HG activity in the MSLS compared with activation in the BS, but only analyzed the biceps femoris within the HG and did not include the GM. In contrast, DeForest et al. (5) found no significant difference in muscle recruitment between the 2 exercises in untrained males, but with 50% of the BS 1RM used on the MSLS, comparisons were likely determined with different relative intensities. Jones et al. (15) also found no difference in muscle activation between the 2 types of squat in trained males. In addition to the difference in subjects included, Jones et al. (15) compared the barbell-BS to the dumbbell-MSLS that may explain the different findings compared with our study. Relative muscle activation found in our study cannot be compared with these recent studies because they did not conduct EMG normalization. Without normalization, comparison of activation across muscle groups was also not analyzed in these previous studies. With a bilateral stance in the BS and SLDL, the load is stabilized in the frontal plane that possibly minimized recruitment in the GM and HG. Our data indicate that the GM's role as a hip abductor and the HG's medial and lateral insertions provide frontal plane support (38) during the MSLS. Simenz et al. (34) also found high GM (127–240%) and HG (42–73%) activation during 4 step-up exercises performed with a 6RM load, which supports our data and suggests that the narrow medial-lateral base of support increases the GM and HG activation. Limiting comparison to our study, Simenz et al. (34) used different normalization techniques to measure GM MVIC (70° hip flexion) and HG MVIC (seated at 60° knee flexion).
Although the MSLS produced the highest activation in both muscle groups, the GM EMG was similar between the BS and SLDL. Conversely, the SLDL produced greater EMG in the HG compared with BS HG levels due to the difference in the concentric phase. Weiss et al. (37) showed further support that low HG activation occurs during the BS after finding no change in HG hypertrophy after subjects trained several weeks with the BS. In addition, Wright et al. (41) found that squats at 75% 1RM produced approximately half as much HG activity compared with the SLDL and lying leg curl. Yamishita (42) suggested that the HG may be attenuated by the simultaneous shortening at the hip and lengthening at the knee during biarticular action producing minimal change in length during the BS. Without concentric action through a significant range of motion, muscular activity may be reduced. Our data support previous research (27) demonstrating consistently greater EMG activation in the concentric phase. However, the higher EMG in the MSLS compared with the SLDL (monoarticular) does not support the suggestion that biarticular action reduces activity, which warrants further investigation. It is possible that the higher demand for frontal plane support during the MSLS may compensate for any attenuating effects of the HG's biarticular function. Although not measured in this study, reciprocal inhibition could be a plausible explanation for the low HG activity in the BS with potentially higher quadriceps activation (23).
Several limitations of this study require noting. When using a different relative intensity than the load used in this study, different results may occur. It is also important in future research to analyze EMG activation of the GM and HG comparing males and females and subjects with different training experience. Measurements of EMG collected in one session may not predict long-term training adaptations, which would require longitudinal investigation. Surface EMG may not reflect the total activation of the muscle groups analyzed, whereas cross-talk may have occurred from surrounding muscles. Timing of muscle activation and recruitment patterns are also important factors for performance and injury prevention, which were not investigated in this study.
Practical Applications
Although the GM demonstrated higher activation than the HG during all 3 exercises, training with these exercises at relatively high intensities may preferentially activate the GM leading to greater adaptation. The GM also seems to be the prime mover at the hip during various weight-bearing activities (26). Although the weight-bearing, hip extension resistance exercises analyzed are commonly implemented due to their sport specificity, strength coaches may need to include isolated hamstring exercises to achieve maximum hamstring development and benefits. The inclusion of hamstring exercises seems to be particularly needed when the BS is used as the primary lower-extremity exercise for strength as the BS produced the lowest HG activity.
The greater GM and HG activation found during the MSLS provides justification for strength coaches to preferentially and regularly include this exercise into resistance training programs in comparison to the weight-bearing exercises with a wider medial-lateral base. It seems that the narrow medial-lateral stance of the MSLS increases the recruitment of the GM and HG for frontal plane support along with their role as hip extensors. The results were found when using a relatively high intensity, thus inclusion of the MSLS during a strength phase is warranted.
Women, shown to have a higher rate of knee injury (10), may benefit most from training with the MSLS compared with the BS and SLDL. Research has shown that women have lower hip strength (19), lower hip extension moments, low GM activation (43) with higher hip adduction, internal rotation, and lower hip flexion during landing (30), which increase the risk of knee injury. Women also have been shown to produce high quadriceps and low hamstring activity during resistance exercise, landing, and cutting (6,24). Based on joint actions produced by the GM and HG contraction, these muscles are likely involved in the reduction of these risk factors of knee injury. Thus, the results of this study indicate that it would be logical to include the MSLS in a resistance training program designed to prevent knee injury.
References
1. Arokoski J, Kankaanpaa M, Valta T, Juvonen I, Partanen J, Taimela S, Lindgren K, Airaksinen O. Back and hip extensor muscle function during therapeutic exercises. Arch Phys Med Rehabil 80: 842–850, 1999.
2. Caterisano A, Moss R, Pellinger T, Woodruff K, Lewis V, Booth W, Khadra T. The effect of back squat depth on EMG activity of 4 superficial hip and thigh muscles. J Strength Cond Res 16: 428–432, 2002.
3. Contreras B, Vigotsky A, Schoenfeld B, Beardsley C, Cronin J. A comparison of gluteus maximus, biceps femoris, and vastus lateralis electromyographic activity in the back squat and barbell hip thrust exercises. J Appl Biomech 6: 452–458, 2015.
4. Decker M, Torry M, Wyland D, Sterett W, Steadman R. Gender differences in lower extremity kinematics, kinetics, and energy absorption during landing. Clin Biomech 18: 662–669, 2003.
5. Deforest B, Cantrell G, Schilling B. Muscle activity in single- versus double-leg squats. Int J Ex Sci 7: 302–310, 2014.
6. Ebben W, Fauth M, Petushek E, Garceau L, Hsu B, Lutsch B, Feldmann C. Gender-based analysis of hamstring and quadriceps muscle activation during jump landings and cutting. J Strength Cond Res 24: 408–415, 2010.
7. Ekstrom R, Donatelli R, Carp K. Electromyographic analysis of core trunk, hip, and thigh muscles during 9 rehabilitation exercises. J Orthop Sports Phys Ther 37: 754–762, 2007.
8. Gorsuch J, Long J, Miller K, Primeau K, Rutledge S, Sossong A, Durocher J. The effect of squat depth multiarticular muscle activation in collegiate cross-country runners. J Strength Cond Res 27: 2619–2625, 2013.
9. Herman D, Weinhold P, Guskiewicz K, Garrett W, Yu B, Padua D. The effects of strength training on the lower extremity biomechanics of female recreational athletes during a stop-jump task. Am J Sports Med 36: 733–740, 2008.
10. Hewett T, Ford K, Meyers G. Anterior cruciate injuries in female athletes: Part 2, a meta-analysis of neuromuscular interventions aimed at injury prevention.Am J Sports Med 34: 1–9, 2006.
11. Hof A. The force resulting from the action of mono- and biarticular muscles in a limb. J Biomech 34: 1085–1089, 2001.
12. Hollman J, Galardi C, Lin I, Voth B, Whitmarsh C. Frontal and transverse plane kinematics and gluteus maximus recruitment correlate with frontal plane knee kinematics during single-leg squat tests in women. Clin Biomech 29: 468–474, 2014.
13. Houck F. Muscle activation patterns of selected lower extremity muscles during stepping and cutting tasks. J Electromyogr Kinesiol 13: 545–554, 2003.
14. Isear J, Erickson J, Worrell T. EMG analysis of lower extremity muscle recruitment patterns during an unloaded squat. Med Sci Sports Exerc 29: 532–539, 1997.
15. Jones M, Ambegaonkar J, Nindl B, Smith J, Headley S. Effects of unilateral and bilateral lower-body heavy resistance exercise on muscle activity and testosterone responses. J Strength Cond Res 26: 1094–1100, 2012.
16. Jonkers I, Stewart C, Spaepen A. The complementary role of the plantar flexors, hamstrings, and gluteus maximus in the control of stance limb stability during gait. Gait Posture 17: 264–272, 2003.
17. Kang S, Jeon H, Kwon O, Cynn H, Choi B. Activation of the gluteus maximus and hamstring muscles during prone hip extension with knee flexion in three hip abduction positions. Man Ther 18: 303–307, 2013.
18. Kubota J, Ono T, Araki M, Torii S, Okuwaki T, Fukubayashi T. Non-uniform changes in magnetic resonance measurements of the semitendonosus muscle following intensive eccentric exercise. Eur J Appl Physiol 101: 713–720, 2007.
19. Leetun D, Ireland M. Core stability measures as risk factors for lower extremity injury in athletes. Med Sci Sports Exerc 36: 926–934, 2004.
20. Lin C, Tsai L, Press J, Ren Y, Chung S, Zhang L. Lower-limb muscle-activation patterns during off-axis elliptical compared with conventional gluteal-muscle-strengthening exercises. J Sport Rehabil 25: 164–172, 2016.
21. Lynn S, Noffal G. Lower extremity biomechanics during a regular and counterbalanced squat. J Strength Cond Res 26: 2417–2425, 2012.
22. McCaw S, Melrose D. Stance width and bar load effects on leg muscle activity during the parallel squat. Med Sci Sports Exerc 31: 428–436, 1999.
23. McCurdy K, O'Kelly E, Kutz M, Langford G, Ernest J, Torres M. Comparison of lower extremity EMG between the 2-leg squat and modified single-leg squat in female athletes. J Sport Rehabil 19: 57–70, 2010.
24. Nagano Y, Hirofumi I, Akai M, Fukubayashi T. Gender differences in knee kinematics and muscle activity during single limb drop landing. Knee 14: 218–223, 2007.
25. Niinimaki S, Harkonen L, Nikander R, Abe S, Knusel C, Sievanen H. The cross-sectional area of the gluteus maximus muscle varies according to habitual exercise loading: Implications for activity-related and evolutionary studies. Homo 67: 125–137, 2016.
26. Okkonen O, Hakkinen K. Biomechanical comparison between sprint start, sled pulling, and selected sprint-start squat exercises. J Strength Cond Res 27: 2674–2684, 2013.
27. Ono T, Okuwaki T, Fukubayashi T. Differences in activation patterns of knee flexor muscles during concentric and eccentric exercises. Res Sports Med 18: 188–198, 2010.
28. Paoli A, Marcolin G, Petrone N. The effect of stance width on the electromyographical activity of eight superficial thigh muscles during back squat with different bar loads. J Strength Cond Res 23: 246–250, 2009.
29. Pappas E, Marshall H, Ali S, Margareta N, Rose D. Biomechanical differences between unilateral and bilateral landings from a jump: Gender differences. Clin J Sport Med 17: 263–268, 2007.
30. Pollard C, Sigward S, Powers C. Gender differences in hip joint kinematics and kinetics during side-step cutting maneuvers. Clin J Sport Med 17: 38–42, 2007.
31. Robertson D, Wilson J, Pierre T. Lower extremity muscle functions during full squats. J Appl Biomech 24: 333–339, 2008.
32. Schaub P, Worrell T. EMG activity of six muscles and VMO: VL ratio determination during a maximal squat exercise. J Sports Rehabil 4: 195–202, 1995.
33. Schoenfeld B, Contreras B, Tiryaki-Sonmez G, Wilson J, Kolber M, Peterson M. Regional differences in muscle activation during hamstrings exercise. J Strength Cond Res 29: 159–164, 2015.
34. Simenz C, Garceau L, Lutsch B, Suchomel T, Ebben W. Electromyographical analysis of lower extremity muscle activation during variations of the loaded step-up exercise. J Strength Cond Res 26: 3398–3405, 2012.
35. Sugisaki N, Kurokawa S, Okada J, Kanehisa H. Difference in the recruitment of hip and knee muscles between back squat and plyometric squat jump. PLoS One 9: e101203, 2014.
36. Wagner T, Behnia N, Ancheta W, Shen R, Farrokhi S, Powers C. Strengthening and neuromuscular reeducation of the gluteus maximus in a triathlete with exercise-associated cramping of the hamstrings. J Orthop Sports Phys Ther 40: 112–119, 2010.
37. Weiss L, Coney H, Clark F. Gross measures of exercise-induced muscular hypertrophy. J Orthop Sports Phys Ther 30: 143–148, 2000.
38. Weltin E, Mornieux G, Gollhofer A. Influence of gender on trunk and lower limb biomechanics during lateral movements. Res Sports Med 23: 265–277, 2015.
39. Willy R, Davis I. The effect of a hip-strengthening program on mechanics during running and during a single-leg squat. J Orthop Sports Phys Ther 41: 625–632, 2011.
40. Worrell T, Karst G, Adamczyk D, Moore R. Influence of joint position on electromyographic and torque generation during maximal voluntary isometric contractions of the hamstrings and gluteus maximus muscles. J Orthop Sports Phys Ther 31: 730–740, 2001.
41. Wright G, Delong T, Gehlsen G. Electromyographic activity of the hamstrings during performance of the leg curl, stiff-leg dead lift, and back squat movements. J Strength Cond Res 13: 168–174, 1999.
42. Yamishita N. EMG activities in mono- and bi-articular thigh muscles in combined hip and knee extension. Eur J Appl Physiol 58: 274–277, 1988.
43. Zazulak B, Ponce P, Straub S, Medvecky M, Avedisian L, Hewett T. Gender comparison of hip muscle activity during single-leg landing. J Orthop Sports Phys Ther 35: 292–299, 2005.
44. Zink A, Whiting W, Vincent W, McLaine A. The effects of a weight belt on trunk and leg muscle activity and joint kinematics during the squat exercise. J Strength Cond Res 15: 235–240, 2001.
