A biomechanical investigation conducted by sports science researchers in Taiwan demonstrates that the snatch pull and specialized trunk exercises elicit fundamentally distinct, non-equivalent muscle excitation profiles in resistance-trained women. Published in Frontiers in Sports and Active Living, the laboratory investigation evaluated surface electromyography amplitudes across the rectus abdominis and lumbar erector spinae during the barbell snatch pull across seven graduated loading intensities, the inverted sit-up, and the horizontal hyperextension. The primary objective was to determine whether progressive loading during a compound Olympic weightlifting pull could generate muscular excitation equivalent to specialized movements targeting the anterior or posterior trunk musculature. Led by Chia-Fang Wu, Chia-Fen Wu, and Shin-Yuan Wang, the research team discovered that while the snatch pull imposed substantial, load-dependent excitation on the posterior spinal extensors, it produced comparatively low excitation across the anterior abdominal wall. Consequently, equivalence testing established that the snatch pull cannot serve as a direct neuromuscular surrogate for targeted core movements, underscoring that compound lifts and isolated core exercises play distinct, complementary roles within athletic training programs.

Core muscle strength and spinal stability are critical determinants of performance in competitive weightlifting and high-load resistance training. Trunk stabilization does not depend on a single isolated muscle group but instead relies upon the coordinated co-contraction of anterior and posterior muscular structures, forming an integrated stabilizing mechanism designed to maintain neutral spinal alignment, regulate intra-abdominal pressure, and ensure efficient mechanical force transmission between the lower extremities and the barbell. As external loading increases during dynamic lifting tasks, trunk muscles counteract substantial spinal flexion moments and shear forces by increasing muscular stiffness through co-contraction. The erector spinae contribute primarily to spinal extension and anti-flexion control, providing essential postural rigidity under heavy axial and shear loads. Conversely, the rectus abdominis assists trunk flexion and anti-extension control while contributing to the regulation of intra-abdominal pressure. Depending on the movement pattern and external resistance, the functional roles of these muscle groups shift dynamically between serving as primary movement drivers and functioning as postural stabilizers.

Surface electromyography represents one of the most widely applied non-invasive tools for evaluating muscular excitation, capturing the electrical signals generated at the skin surface to provide objective insight into myoelectric activity under varied loading conditions. To overcome inter-individual variability and differences across testing sessions, electromyographic signal amplitudes are typically processed through full-wave rectification, root-mean-square smoothing, and normalization against a maximal voluntary contraction reference value, expressing muscular excitation as a percentage of maximal voluntary contraction capacity. While surface electromyography reflects neuromuscular excitation rather than a direct mechanical measurement of absolute contractile force or internal muscle activation, it provides a standardized framework for comparing motor unit recruitment patterns. Prior investigations demonstrated that high-load free-weight movements such as squats and deadlifts generate substantial erector spinae myoelectric amplitude due to elevated stabilization demands, but the exact neuromuscular behavior of the trunk extensors and flexors during Olympic snatch pulls relative to specialized apparatus-based core exercises had remained unquantified.

In practical strength and conditioning, the snatch pull is classified as a multi-joint compound movement requiring synchronized extension of the hips, knees, and ankles while maintaining strict trunk stiffness to oppose forward flexion moments created by the external load. Due to its steep technical complexity and intense systemic loading demands, the snatch pull is frequently restricted or modified during early development stages, during transitional microcycles, or under conditions of accumulated physical fatigue. In such settings, strength coaches and weightlifters regularly implement specialized trunk exercises, such as the Roman chair hyperextension and the inverted sit-up, as supplementary or alternative training modalities. The hyperextension is widely utilized as a foundational posterior-chain drill featuring low technical complexity and targeted control of spinal extension. Conversely, the inverted sit-up uses suspension in an inverted posture to introduce an unstable training environment that generates heightened stimulation of the anterior core musculature, promoting neuromuscular control of the rectus abdominis.

To rigorously evaluate how trunk muscle excitation differs across these exercises, the investigators implemented a randomized within-subject experimental design involving twenty female university students with at least three years of documented resistance training experience. The participant cohort exhibited a mean age of 22.2 ± 5.3 years, an average height of 161.4 ± 7.2 cm, a mean body mass of 61.5 ± 10.4 kg, and an established snatch pull one-repetition maximum of 64.8 ± 30.1 kg. Inclusion criteria required all participants to be completely free from joint disorders or unresolved musculoskeletal injuries in the three months preceding data collection. To minimize movement variability and ensure lifting safety, every volunteer completed a standardized familiarization protocol consisting of three to five practice sessions across a minimum of one week prior to experimental testing. The subsequent experimental protocol spanned three weeks, during which each participant completed one designated exercise session per week, with six days of mandatory recovery between sessions to eliminate residual muscular soreness.

Standardized measurement protocols were maintained throughout all testing procedures to guarantee data integrity across participants and sessions. Bipolar silver-silver chloride surface electrodes were affixed over shaved, abraded, and alcohol-cleaned skin parallel to the underlying muscle fibers, positioned three centimeters lateral to the umbilicus for the rectus abdominis and three centimeters lateral to the L3 spinous process for the lumbar erector spinae. Muscular excitation was captured using a wireless Noraxon Ultium electromyography system sampling at 2,000 Hertz, band-pass filtered between 10 and 500 Hertz, and smoothed using a 50-millisecond root-mean-square moving window. Maximal voluntary contraction references were established at the start of each session via standardized isometric contractions against non-elastic straps, taking the highest 300-millisecond window across trials. The inverted sit-up was performed on a hanging inversion apparatus and the hyperextension was conducted on a horizontal Roman chair, both executed across three 30-second sets of approximately 15 repetitions synchronized to a metronome at 60 beats per minute with two minutes of inter-set rest. The snatch pull was performed from the floor using lifting straps across progressive intensities of 40, 50, 60, 70, 80, 90, and 100 percent of one-repetition maximum, with two to three minutes of rest between attempts.

Electromyographic amplitude analyses revealed distinct recruitment patterns across the three exercise modalities, yielding statistically significant main effects for both muscle groups. Repeated measures analysis of variance demonstrated that exercise type exerted a significant effect on rectus abdominis amplitude (F(2, 38) = 19.53, p < 0.001, partial eta squared = 0.507), with post hoc comparisons showing that the inverted sit-up generated an average amplitude of 57.66 ± 34.92 percent of maximal voluntary contraction. This abdominal excitation was significantly greater than that elicited during the snatch pull at 100 percent of one-repetition maximum, which averaged 25.52 ± 27.33 percent (p < 0.01), and the hyperextension, which produced 9.22 ± 9.52 percent (p < 0.001). Rectus abdominis excitation did not differ significantly between the maximal snatch pull and the hyperextension (p = 0.091). Furthermore, linear mixed-effects modeling showed that rectus abdominis amplitude did not increase significantly across progressive snatch pull loads, rising by merely 2.15 percent per 10 percent increase in one-repetition maximum (95 percent confidence interval: -0.09 to 4.39; p = 0.059; marginal R-squared = 0.028).

In marked contrast to the anterior core, the erector spinae exhibited substantial excitation during the barbell snatch pull that scaled directly with external load. Exercise type significantly altered erector spinae amplitude (F(1.24, 23.61) = 67.81, p < 0.001, partial eta squared = 0.781), with the 100 percent one-repetition maximum snatch pull generating 135.96 ± 57.53 percent of maximal voluntary contraction. This extensor amplitude was significantly greater than the 69.39 ± 24.35 percent recorded during the horizontal hyperextension (p < 0.001) and the 9.61 ± 4.40 percent recorded during the inverted sit-up (p < 0.001). Hyperextension also elicited significantly greater erector spinae excitation than the inverted sit-up (p < 0.001). Linear mixed-effects regression confirmed a robust load-dependent response, with erector spinae amplitude increasing by 11.14 percent of maximal voluntary contraction for every 10 percent increment in snatch pull intensity (95 percent confidence interval: 7.49 to 14.79; p < 0.001). Although load alone accounted for modest variance (marginal R-squared = 0.168), the full regression model incorporating individual participant differences explained 65.8 percent of the variance (conditional R-squared = 0.658), demonstrating meaningful inter-individual variability in extensor responsiveness.

To formally establish whether snatch pull loading could replace specialized core movements, the researchers executed paired two one-sided tests using a pragmatic equivalence bound of ±10 percent of maximal voluntary contraction. For the rectus abdominis, equivalence was rejected across all snatch pull intensities when compared against the inverted sit-up, with mean paired differences ranging between -43.53 and -29.46 percent (all equivalence p-values > 0.05). For the erector spinae, the snatch pull at 40 percent of one-repetition maximum produced an average excitation of 66.10 ± 27.42 percent, exhibiting the smallest mean paired difference relative to the hyperextension at -3.29 percent (paired t-test p = 0.695). However, its 90 percent confidence interval spanned from -17.58 to 11.00 percent, breaching the predetermined equivalence thresholds and failing the formal test (p = 0.214). At all higher intensities between 50 and 100 percent of one-repetition maximum, erector spinae excitation significantly exceeded hyperextension values by 21.54 to 66.57 percent (all p < 0.05), confirming that the exercises do not produce interchangeable myoelectric demands.

In evaluating these outcomes, the authors contextualized the physiological mechanisms underlying core recruitment during dynamic Olympic lifting. Because the barbell snatch pull does not require spinal flexion, superficial rectus abdominis excitation remains modest even at maximal loads, as trunk stabilization against heavy loads likely depends on elevated intra-abdominal pressure and deeper musculature rather than proportional recruitment of the superficial abdominal wall. Regarding the high erector spinae values exceeding 100 percent of maximal voluntary contraction—with individual trials approaching 250 percent—the authors noted that dynamic multi-joint actions frequently exceed static reference values when normalized to a single prone isometric test. This indicates that the prone test did not elicit true maximal excitation in all participants, reflecting known limitations in trunk extensor testing where no single isometric drill maximizes erector spinae output universally.

The authors outlined several methodological constraints that define the scope of their findings. The research investigated only superficial rectus abdominis and lumbar erector spinae signals, excluding deeper stabilizing structures such as the transversus abdominis and internal or external obliques. The sample consisted solely of 20 trained female university students whose one-repetition maximums varied considerably (64.8 ± 30.1 kg), meaning the neuromuscular findings may not directly generalize to male lifters, untrained cohorts, or elite international competitors who handle higher absolute and relative barbell loads. Furthermore, because the study evaluated acute myoelectric responses across single repetitions arranged in an ascending loading order with fixed two-to-three-minute rest intervals, the potential influence of cumulative fatigue cannot be entirely eliminated. The lack of a formal sample size calculation for the equivalence analysis widened the confidence intervals around low-load snatch pulls, warranting cautious interpretation until verified by larger investigations.

Ultimately, the experimental data provide concrete biomechanical evidence regarding exercise selection in strength and conditioning and weightlifting regimens. Rather than acting as functional substitutes, the snatch pull and specialized apparatus-based core exercises impose divergent physical demands on the human torso. High-intensity snatch pulls deliver intense neuromuscular overload to the posterior extensor chain that surpasses traditional hyperextensions, while offering minimal stimulus to the anterior abdominal wall. Conversely, inverted sit-ups generate robust superficial abdominal excitation that dynamic pulls cannot replicate. Consequently, athletic coaches and practitioners should program these modalities in a complementary rather than interchangeable structure, combining progressive multi-joint pulls to reinforce posterior spinal anti-flexion capacity with targeted flexion exercises to develop anterior core strength.

Subject of Research: Surface electromyography of trunk muscles during snatch pulls and specialized core training

Article Title: sEMG amplitude of core muscles under specialized core and snatch pull exercise conditions

Article References: Wu, C.-F., Wu, C.-F., & Wang, S.-Y. (2026). sEMG amplitude of core muscles under specialized core and snatch pull exercise conditions. Frontiers in Sports and Active Living, 8, Article 1927838. https://doi.org/10.3389/fspor.2026.1927838

Image Credits: AI Generated

DOI: 10.3389/fspor.2026.1927838

Keywords: surface electromyography, snatch pull, core training, erector spinae, rectus abdominis, biomechanics, weightlifting, sports science, sEMG, amplitude, core, muscles