The effects of muscle fatigue on shoulder proprioception and forehand stroke accuracy in Chinese elite table tennis athletes

Abstract

This study explores the relationship between shoulder proprioception and landing point accuracy in Chinese elite table tennis athletes, particularly under fatigue conditions. A total of 19 male athletes participated, with their shoulder proprioception tested using an ISOMED 2000 isokinetic muscle tester. The tester measured proprioception acuity in both pre-fatigue and post-fatigue conditions, alongside performance in a hitting task assessed with a high-speed serve machine. Results indicated a significant correlation between proprioception and stroke performance, especially in internal and external rotation directions (p < 0.01). After fatigue, proprioception in both the racket-holding and non-racket-holding hands significantly decreased in internal and external rotations (p < 0.01 for both hands), with no significant change in vertical extension (VE). Additionally, the performance of the hitting task after fatigue significantly declined in the racket-holding hand (p < 0.01). These findings suggest that shoulder proprioception plays a crucial role in stroke accuracy and fatigue recovery, providing valuable insights for training, rehabilitation, and performance optimization in table tennis.

Clinical trial number

Not applicable.

Highlights

Shoulder proprioception in the internal and external rotation direction decreased in both the racket-holding and non-racket-holding hands after fatigue.

The performance in the hitting task was significantly correlated with shoulder proprioception in the internal and external rotation direction, and proprioception decreased after fatigue, so did the performance in the hitting task.

We suggest that the active position reset test of shoulder should be included in the evaluation of athletic quality.

After training or during breaks in competition, athletes should actively use various methods to recover from shoulder fatigue.

Introduction

In table tennis, swinging and hitting the ball are basic actions. In the process of swinging and hitting, the shoulder, as the starting joint of the upper limb power chain, not only plays a role in transmitting power, but also the key factor in the accuracy of hitting the ball in the game [1, 2]. An athlete’s ability to hit the ball accurately is an important factor in winning the game, and it relies on good proprioception of shoulder, which is constantly adjusting the technical movements of the dynamic stereotypes by means of “automatic processing“ [3]. As the game progresses to the critical stage, athletes will be in a state of fatigue, and another key to winning the game is the performance after fatigue. Although it is generally accepted that fatigue may reduce athletic performance, there is a lack of quantitative understanding of how localized shoulder fatigue affects proprioceptive acuity and to what extent this impacts stroke accuracy. In practice, athletes and coaches often rely on subjective judgment to manage training fatigue, but objective data on proprioceptive decline under fatigue is essential for evidence-based training and recovery planning. Therefore, exploring the relationship between shoulder proprioception and sports performance and fatigue will not only help table tennis athletes to improve their technical movements, but also help to develop scientific and effective training and improve sports performance in competition.

“Ball sense” is often mentioned in table tennis. The so-called “ball sense” refers to the specialized perception ability developed by players through long-term practice. This concept is somewhat abstract, but specifically: on the one hand, the ability to accurately judge the flight path and landing point of the opponent’s ball; on the other hand, the ability to accurately control the strength and direction of the racket to hit the ball, so that the return ball can fly according to its intended target. The key here is proprioception. In competitive games, it is important to judge the spatial displacement [4]. A good athlete anticipates the speed, rotation, and landing point of the ball, etc. And the racket hand receives the signals from the central system to intercept the movement. The athlete relies on proprioception signals to adjust the position of the hand and body as well as the racket and the ball, and then to perform the hitting motion after successfully receiving the ball. Studies have shown that there is also a correlation between proprioception and high level of sports performance [5]. At present, there are more studies on the proprioception of the ankle, but their methodological choices are flawed, such as not shielding visual influences [6]. Some studies have further measured the threshold of wrist proprioception holding and touching the ball in table tennis athletes after excluding auditory and visual distractions [7], but there is still a lack of specific research on the sense of motion and position in proprioception of table tennis athletes. Moreover, most previous studies have overlooked the shoulder joint, which plays a primary role in stroke production. There remains a clear gap in understanding the specific proprioceptive changes in the shoulder under fatigue and how this directly correlates with on-table stroke performance in realistic task conditions.

We hypothesize that shoulder proprioception significantly influences stroke accuracy in Chinese elite table tennis athletes, and that fatigue causes a measurable decline in proprioception, directly contributing to reduced stroke performance. To test this hypothesis, this study investigates the relationship between active position resetting of the internal and external rotation of the shoulder [8–10] and stroke accuracy of table tennis athletes was tested before and after fatigue by using a fatigue model. By excluding visual and auditory interference, we aim to provide a more accurate evaluation of shoulder proprioception’s role in stroke performance. The results of the study will provide guidance and suggestions for the physical training, rehabilitation and fatigue recovery of Chinese table tennis athletes in the future, so as to improve sports performance as well as competition results.

Materials and methods

Subjects

Total 19 national elite athlete level table tennis players from the Chinese National Table Tennis Team were recruited into this research. All of them had participated in major competitions such as National Games of the People’s Republic of China and World Youth Championships and had more than 10 years training experience. The study was approved by the Ethics Committee of Sports Science Experiment of Beijing Sport University (2021075 H). The basic information of the athletes is shown in Table 1.

Table 1.

Basic information of the table tennis players (mean ± SD)

Subjects	Age(yrs)	Height(cm)	Body weight(kg)	Training experience(yrs)	Level of athlete
n = 19	17.87 ± 1.49	176.31 ± 4.19	66.67 ± 3.92	> 10	National Elite athlete

Experimental design

In this study, a cross-sectional survey in a descriptive study was used, and the fatigue model was established using the ISOMED 2000 system; the joint position reset method was used in this instrument to evaluate the proprioception ability of the athletes. A high-speed ball serving machine (developed independently by the Institute of Sports Science and Technology of the General Administration of Sport of China) was used to test the landing point accuracy of table tennis. The machine delivered no-spin balls at a constant velocity of 15 m/s with a fixed interval of one ball every 3 s. Ball placement precision was maintained within ± 3 cm across 100 consecutive trials, verified by 240 fps high-speed video and manual cross-checking. Although not commercially available, the device has been extensively applied in national-level training and prior experimental studies, with consistent reliability. Pre-experimental calibration and repeated test runs ensured procedural stability and minimized external variability, thereby enhancing the validity of stroke accuracy measurements under both fatigue and non-fatigue conditions.

Study procedure

The experiment requirements were explained in detail to the subjects before the experimental test. Since proprioception is susceptible to interference by many factors, a proprioception test was performed first, followed by the table tennis landing point accuracy test. The fatigue model was tested immediately after the test, that is, the shoulder proprioception test and the landing point accuracy test were performed sequentially immediately after the fatigue, and the procedure was the same as before.

Experimental equipment

• ①ISOMED 2000 isokinetic muscle tester (D. & R. Ferstl GmbH, Germany).

• ②High speed service machine made by the Institute of Sports Science and Technology of the General Administration of Sport of China.

• ③The high-speed camera (Casio EX-FH25, Japan) can record a dynamic image at a maximum rate of 1000 frames per second. In this experiment, the flight speed and landing point accuracy of the ping pong ball were captured at 240 frames per second.

Proprioception test of shoulder

The ISOMED 2000 (D. & R. Ferstl GmbH, Germany) was used to perform proprioception tests on the athletes. It has been shown to have high reliability and validity according to numerous studies [9, 11]. Many researchers have also used ISOMED 2000 to measure proprioception [12] to assess an individual’s ability to actively repeat a reference position (Fig. 1).

Fig. 1.

Isokinetic muscle strength testing of shoulder joints with ISOMED 2000

The starting position for internal and external shoulder rotation was 90 degrees of abduction, 90 degrees of elbow flexion and 30 degrees of humeral external rotation. The preset values were 40 degrees of external rotation and 20 degrees of external rotation (i.e., 10 degrees each of internal and external rotation under the starting position). This was followed by a rotational joint position reset test in sequence, with a total of seven measurements (the first three for movement learning and the last four for formal test). The subject actively moved the arm (shoulder joint) from the initial position to a predetermined target angle for 3 s, reminded the subject of this predetermined target position, and then returned to the neutral position. Then the formal experiment was carried out. The subjects were asked to move actively. When they felt the target Angle, press the pause button and record the actual Angle at this point. The shoulder proprioception ability was evaluated by comparing the difference between the actual position and the target position, and the evaluation indexes included constant error (CE), variable error (VE), and absolute error (AE). These indicators not only compare the magnitude and direction of error, but also evaluate the stability of error [13, 14]. (Table 2).

Table 2.

Calculation and significance of proprioceptive evaluation indicators of the shoulder joint

Test	Evaluation index and its calculation methods	Meaning of the indicator
	AE = (|Original error 1| +|Original error 2| +… +|Original error n|)/n	No positive or negative direction is considered. Only calculate the absolute error between the end position and the start position.
Joint position reproduction test of the shoulder joint	CE = [(Original error 1) + (Original error 2) +… + (Original error n)]/n	Taking the positives and negatives in direction into account to evaluate overall the error in the given direction of movement, reflecting whether the movement pattern overall exceeds or fails to reach the target.
	VE =√{[(Original error 1-CE)2 + (Original error 2-CE)2 +… + (Original error n-CE)2 ]/n}	Reflects the variability and consistency between the results of several position reproduction, regardless of the accuracy of the JPR.

AE: Absolute error; CE: Constant error; VE: Variable error; JPR: Joint Position Reproduction test

Table tennis landing point accuracy test

The landing point accuracy of table tennis athletes was tested using a high-speed service machine developed by the Institute of Sports Science and Technology of the General Administration of Sport of China. The service table was first divided into 15 zones [15], among which zones 7 and 9 were used as target landing areas (Fig. 2). These two zones were selected because they represent the most commonly used cross-court return areas in competitive table tennis. Specifically, zone 7 corresponds to the preferred forehand return target for right-handed players, while zone 9 corresponds to that of left-handed players. Both are tactically critical for establishing rally dominance and are frequently emphasized in national training programs.

Fig. 2.

Diagram of the landing point

The service machine was located in the center of the service side table (Fig. 3), and the no-spin ball was sent at a speed of 15 m/s. Athletes were instructed to perform standardized return strokes, aiming for the designated zone based on their dominant hand—zone 7 for right-handed players and zone 9 for left-handed players. The returned balls landed on the server’s side of the table. A high-speed camera and an observer were used to record and verify the landing coordinates. Each athlete completed three sets of 100 trials, with sufficient rest between sets. Accuracy data were recorded as the number of successful target hits versus misses.

Fig. 3.

Service Machine

Following the baseline trials, the fatigue model was implemented, and both the shoulder proprioception test and the landing point accuracy test were repeated using the same procedures.

Shoulder joint fatigue criteria

Following the proprioception assessment, participants continued to perform isokinetic concentric shoulder contractions at an angular velocity of 180°/s using the ISOMED 2000 system, to induce local muscle fatigue. This method is widely adopted in neuromuscular fatigue studies to simulate high-intensity upper-limb loading conditions [16]. During the fatigue protocol, peak torque output was continuously recorded by the system, and subjective exertion was assessed using the Borg Rating of Perceived Exertion (RPE) scale.

Fatigue was considered to be achieved when both of the following criteria were met: (1) Peak torque output dropped below 50% of the initial value for three consecutive repetitions, indicating a substantial neuromuscular decline; (2) RPE score reached 19 or higher, accompanied by visible symptoms such as pallor, labored breathing, or signs of physical exhaustion. This dual-criterion model ensured that both objective mechanical fatigue and subjective exertion were present, thereby validating the fatigue state prior to post-fatigue proprioception and stroke testing.

Statistical analyses

SPSS 20.0 (Statistical Package for Social Science, Chicago, IL, USA) software was used for statistical analysis and calculations. The Kolmogorov-Smirnov test was used to test the normality of the data. If the data met the normal distribution, the correlation between angle differences and batting performance in different directions was analyzed using a Pearson simple linear model. Otherwise, Spearman correlation analysis was used. The same statistical methods were used for angle differences and batting performance after fatigue. Paired-samples t-test or Wilcoxon signed-rank test was used to compare the differences of each index before and after fatigue. Significant differences were considered at p < 0.05 and highly significant difference were considered at p < 0.01. All indicators are expressed in mean ± SD. In this study, the correlation coefficient r was used as the effect size, and the evaluation criteria were: low correlation when the correlation coefficient effect size r ≤ 0.10; moderate correlation when 0.10 < r < 0.4; and high correlation when r ≥ 0.40.

Results

Effects of fatigue on proprioception and hitting performance

Proprioception

In the external rotation task, compared with the proprioception ability before fatigue, the AE (p < 0.01) and CE (p < 0.01) of the racket holder decreased significantly after fatigue, while there was no significant difference in VE (p > 0.05), as shown in Table 3. Under internal rotation, compared with the proprioception ability before fatigue, the AE (p < 0.01) and CE (p < 0.01) of the racket holder decreased significantly after fatigue, while there was no significant difference in VE (p > 0.05), as shown in Table 4.

Table 3.

Changes in proprioception before and after external rotation fatigue of the racket-holding hand (mean ± SD)

External rotation	Before fatigue	After fatigue	p value	t/z value
AE	6.05 ± 4.65	4.99 ± 4.42	< 0.01	-2.76
CE	4.31 ± 5.93	-1.38 ± 3.79	< 0.01	-2.80
VE	2.14 ± 1.54	3.75 ± 4.12	0.46	-0.75

Table 4.

Changes in proprioception before and after internal rotation fatigue of the racket-holding hand (mean ± SD)

Internal rotation	Before fatigue	After fatigue	p value	t value
AE	4.55 ± 2.87	7.79 ± 5.05	< 0.01	-4.94
CE	1.07 ± 4.10	1.81 ± 6.73	< 0.01	-0.43
VE	2.95 ± 2.22	4.96 ± 4.44	0.07	-1.79

Hitting performance

Compared with before fatigue, the number of errors in the batting task increased significantly after fatigue (p < 0.01), as shown in Table 5.

Table 5.

Changes in stroke accuracy before and after fatigue of the racket-holding hand (mean ± SD)

Hitting the ball	Target	Number of errors before fatigue	Number of errors after fatigue	P value	Z value
Athletes	100	7.85 ± 3.85	14.55 ± 3.60	< 0.01	-3.95

Correlation between proprioception and hitting performance

Before fatigue

Significant correlations were observed between the number of hits and proprioceptive accuracy, specifically in external rotation AE (r=–0.859, p < 0.01) and CE (r=–0.774, p < 0.01), as well as internal rotation AE (r=–0.623, p < 0.01) and VE (r=–0.615, p < 0.01), as shown in Table 6.

Table 6.

Correlation between the number of landing points and proprioception in each direction

		External rotation AE	External rotation CE	External rotation VE	Internal rotation AE	Internal rotation CE	Internal rotation VE
Number of landing points	r	0.859	0.774	-0.221	0.623	0.244	0.615
	p	< 0.01	< 0.01	0.349	< 0.01	0.299	< 0.01

After fatigue

Correlations remained significant post-fatigue in external rotation AE (r=–0.814, p < 0.01) and CE (r=–0.649, p < 0.01), and internal rotation AE (r=–0.763, p < 0.01), as shown in Table 7.

Table 7.

Correlation between the number of landing points and proprioception of corresponding direction

		External rotation AE	External rotation CE	External rotation VE	Internal rotation AE	Internal rotation CE	Internal rotation VE
Number of landing points	r	0.814	-0.411	0.649	0.763	0.202	0.332
	p	< 0.01	0.71	< 0.01	< 0.01	0.393	0.152

Discussion

This study aimed to investigate whether shoulder proprioception declines after fatigue and whether such decline impairs stroke accuracy in elite table tennis athletes. The results supported our hypothesis: both internal and external rotation proprioception were significantly reduced following fatigue, and this decline was closely associated with reduced landing point accuracy.

In this assessment of proprioception, we used of the absolute error (AE) of joint position reproduction, which quantifies an otherwise abstract sensory ability into measurable angular deviation. This method has been widely validated as a gold standard in proprioception research, as it reflects both position sense and kinesthetic awareness [8]. On the other hand, table tennis-specific skill was represented by stroke accuracy, measured through the precision of landing points in a controlled return task. By requiring athletes to consistently target zones 7 or 9, we were able to quantify their technical execution under both non-fatigued and fatigued conditions [17].

Effects of fatigue on proprioception

In this study, it was found that shoulder proprioception of the racket-holding arm significantly decreased after fatigue, particularly in internal and external rotation movements. This was reflected in increased AE and CE, indicating a reduction in the accuracy of joint position perception.

This decline can be explained by physiological effects of fatigue on neuromuscular systems [18]. Fatigue alters muscle tone and length, which directly impacts the responsiveness of proprioceptors—particularly muscle spindles—responsible for providing position and movement information. As a result, sensory feedback to the central nervous system becomes less accurate, reducing the athlete’s ability to perceive and control shoulder joint angles during motion [19].

Previous studies support these findings. Voight et al. [20] found that both active and passive position sense significantly decreased after fatigue when reassessed after establishing fatigue model on an isokinetic muscle strength tester. Siekirk et al. [21] further reported increased position reproduction errors (AE, CE) following prolonged moderate-intensity exercise. Similarly, Lee et al. [22] observed that active position resetting ability decreased after fatigue, while passive joint sense remained unaffected. This suggests that fatigue primarily impairs proprioceptive input from muscle mechanoreceptors, rather than joint mechanoreceptors. Given that table tennis strokes rely heavily on active, high-speed shoulder movements, this mechanism likely underpins the observed proprioceptive decline.

Effects of fatigue on hitting performance

This study found that stroke accuracy significantly declined after shoulder fatigue, as shown by fewer successful hits and more errors. In table tennis, where technical precision is essential during high-speed rallies, such deterioration in performance could affect match outcomes.

The decline in accuracy is likely due to reduced proprioceptive acuity, which impairs the ability to control limb position, movement timing, and joint coordination. Fatigue alters joint kinematics and disrupts the timing of upper limb movements, particularly in multi-joint tasks such as the forehand stroke. Previous studies have shown that fatigue-induced changes in joint angles and muscle control compromise coordination across the entire kinetic chain [23–26]. In this study, shoulder fatigue likely disrupted this coordination, leading to increased variability in racket movement and stroke execution.

These findings emphasize the importance of fatigue management in table tennis training, particularly for maintaining stroke accuracy during extended play.

Relationship between proprioception and stroke performance

This study revealed a strong correlation between shoulder proprioception—particularly in internal and external rotation—and stroke accuracy in elite table tennis players. These findings suggest that proprioception plays a vital role in the precise control of racket positioning, especially during rapid, repeated strokes.

Proprioceptive input enables athletes to sense and adjust joint position and movement without relying solely on visual cues. In high-speed sports like table tennis, where response times are extremely short, reliance on proprioceptive feedback is essential for maintaining stroke consistency under dynamic conditions. This is consistent with prior studies showing that proprioception is closely linked to spatial control and movement precision in complex upper-limb tasks [5, 27].

Compared with reflex-based responses, conscious proprioceptive feedback allows for fine-tuned muscle coordination and real-time movement correction [28, 29]. In this study, higher proprioceptive accuracy (lower AE) was associated with more accurate stroke execution, indicating that proprioceptive integrity is a critical factor in technical performance [30, 31].

Limitations and future directions

This study focused exclusively on forehand strokes under controlled laboratory conditions, which may not fully capture the complexity of stroke dynamics in competitive play. Future research should investigate whether similar fatigue-induced changes in proprioception and performance are observed in backhand strokes, multi-directional movements, or stroke sequences involving rapid transitions. Additionally, integrating real-match scenarios or simulated competitive environments would enhance ecological validity.

Another limitation lies in the use of a single-joint isokinetic fatigue protocol, which, while controlled, may not entirely replicate the multi-joint fatigue experienced during actual gameplay. Expanding future protocols to include compound movements or functional fatigue tasks would provide more sport-specific insight.

Finally, all participants were male elite athletes from a national training system. Future studies could examine gender differences, developmental stages, and the application of proprioceptive assessment tools in broader athletic populations. Such work could inform individualized training strategies and injury prevention frameworks in table tennis and other racket sports.

Collectively, this study contributes novel practical value by quantitatively linking specific shoulder proprioceptive impairments to declines in stroke performance under fatigue in elite table tennis players. While it is widely assumed that fatigue affects coordination, our findings provide precise, empirical evidence that can inform training decisions. By identifying measurable proprioceptive indicators associated with performance, this research offers a foundation for targeted fatigue monitoring, proprioception-based recovery protocols, and sport-specific neuromuscular training. These insights may help coaches and athletes maintain technical consistency in high-intensity, time-constrained match environments.

Conclusions

Both before and after fatigue, stroke performance and proprioception of internal and external rotation were correlated, and after fatigue, proprioception decreased significantly and stroke accuracy decreased significantly. This suggests that although fatigue cannot be avoided in long-term training or competition, we can pay more attention to the effect of shoulder proprioception on training results or competition performance. Focused training or recovery of proprioception may have an impact on training results or game performance.

In the future, researchers should focus on exploring the relationship between shoulder proprioception and sports performance, as well as the relationship between the two changes before and after fatigue, and devote themselves to guiding table tennis players to improve technical movements, helping coaches to develop scientific and effective training plans, and actively using various methods to restore shoulder fatigue after training or during game breaks, so as to improve sports performance in competitions.

Author contributions

Z.C. and T.Z. contributed equally to this study: Z.C.: proposed the thesis topic, designed the thesis framework, collect data, and wrote the thesis; T.Z.:Conceptualization; Methodology; Data curation; Formal analysis; Writing - Original Draft; Y.S.:Writing - Review & Editing; Project administration; S.F.: wrote the thesis, revise the paper, check references and translate the paper; L.S.:Conceptualization; Methodology; Supervision; All authors read and approved the final version of the manuscript.

Funding

This work is supported by: The 2019 Major Project of Philosophy and Social Sciences of Higher Education Institutions in Hubei Province [2019ZD071], titled “Screening, Classification, Assessment, and Intervention — A Study on the Path of Healthy Growth for Adolescents.” The Youth Foundation of Humanities and Social Sciences Research of the Ministry of Education of China [18YJC890032], titled “Policy, Mechanism, and Path: A Comparative Study on Promoting Adolescent Physical Activity between China and the United States” and The Program of China Scholarship Council (Grant No. 202408420281).

Data availability

The data supporting this study’s findings are available from the corresponding author upon request.

Declarations

Ethics approval and consent to participate

The study was approved by the Ethics Committee of Sports Science Experiment of Beijing Sport University (2021075 H). The study was conducted in accordance with the Declaration of Helsinki.All study participants were informed and consented to participate in the study.

Consent for publication

All participants gave written informed consent for their personal or clinical details, along with any identifying images to be published in this study.

Competing interests

The authors declare no competing interests.