Review Reports - The Effect of Music Tempo on Fatigue Perception at Different Exercise Intensities

Round 1

Reviewer 1 Report

In this study, interesting experiments were conducted on the intensity of running and the different conditions of the tempo of music. This is an interesting and practical topic; there are inaccuracies in the methodology, which should be clarified before the rest of the article review.

Abstract.

The impact of the interaction between music tempo and exercise intensity on TFP was significant (F=7.318, p=0.012). Isn't it obvious that results 1 and 2 are relevant? Isn't it worth skipping this and adding more information about the studied group of subject? In the application, it may be worth precise what tempo of music affects the lower fatigue better.

Introduction

Paragrafs 3 and 4 have a similar content, all of the introduction is quite a long, maybe can be a bit shortened by combining these paragraphs?

Methods

2.3. It is not clear how the pace of music with the intensity of gear during the experiment.

2.4.1. Only the scales of fatigue in the case of 15 have been explained, and what the rest of the Borga scale looks like?

2.4.2. There were only 2 heart rate measurements? What device was HR measured, more technical details needed.

2.4.3.

Which lower limb was examined, perhaps both?

What is MF? One can presume but the reader should be given this.

How was the skin prepared, according to what methodology was the site chosen for sticking the electrodes, what were the electrodes, what was the EMG system, how was the signal processed.... These are some of the questions for the sEMG system that need to be clarified to make the study reproducible.

2.6.1.

The study took place in a standing position, was it worth studying resting HR while standing... Does HR while standing and sitting differ?

2.6.3.

Why was this type of premeasurement used, in the literature MVC is the most common?

The figure shows that the electrode location is in the vastus lateralis muscle rather than the rectus femoris...it is difficult to hit the RF with such large electrodes.

Author Response

Open Review 1

First of all, we would like to thank you for your recognition of the research work. and thank you for your careful review of our article and for your many valuable and important comments, which we have revised one by one according to your suggestions. All changes were made in “Track Changes” function, so you can easily see the changes we made in the revised version.

Below is a description of the changes we made to each comment and our response.

Abstract

Reply:

We thank you for your valuable comments and we acknowledge the suggestions you mentioned. Having clarified the interaction between music tempo and exercise intensity, it is also necessary to go further and point out the applied significance of this study. Therefore, we precised in the abstract what tempo of music are better to bring about lower fatigue. In addition, we felt that the application implications of the results needed to be strengthened in the discussion section as well, so we added “4.4 Pratical application” to discuss the practical implications of the study, as follows.

This study aimed to determine the effect of music tempo on runners' fatigue perception at different exercise intensities, and provided a physiological and psychological explanation of the role of music tempo in resisting fatigue perception by observing time of fatigue perception, heart rate changes, and sEMG signal changes during the running experiment. The combined results of the three indicators revealed that the use of music with different tempos at different exercise intensities caused runners to exhibit different mental and physical characteristics. Specifically, runners’ TFP is influenced by the music tempo, as well as the interaction between music tempo and exercise intensity. The change in HR of runners during running is mainly influenced by the exercise intensity. Whether it is low-intensity exercise or high-intensity exercise, listening to fast music may mitigate individuals’ perception of fatigue to some extent. This was also pointed out in the study by Centala et al. The effect of listening to fast music during exercise would be better than slow music, which could regulate the mental emotions during exercise and reduce the perception of fatigue during exercise.

In addtion, this study also provides many practical implications. Running is an endurance sport, and runners will feel fatigue in different degrees during running, especially for many non-professional runners, who often do not persist in running due to insufficient exercise, limited physical strength and lack of perseverance. The results of this study can help runners to develop a more beneficial exercise program. For example, both low-intensity exercisers and high-intensity exercisers were advised to listen to fast music while exercising, which would help them maintain a better mental state, make the exercise process less tedious, and better control over the physiological state such as heart rate changes. This research could lead to better effects of music on perceived fatigue, allowing runners to have better mental and physical performance while running.

Introduction

Paragrafs 3 and 4 have a similar content, all of the introduction is quite a long, maybe can be a bit shortened by combining these paragraphs?

Reply: We have combined the two paragraphs in introduction and simplified the description as you suggested.

Methods

2.3. It is not clear how the pace of music with the intensity of gear during the experiment.

Reply:

I am sorry that our description was not clear enough and caused you confusion. In fact, we have already mentioned the music tempo and treadmill intensity settings during the experiment in section 2.6.4. Each runner ran under low-intensity and high-intensity conditions, with no music, fast music, and slow music.

In order to more clearly describe the exercise program of music tempo and treadmill intensity during the experiment, we considered it necessary to add a section in “2.3.3 Music tempo with exercise intensity” to enumerate the experimental conditions of different groups. As shown in Table 1.

Table 1 Exercise schemes

Intensity	No music	Slow music（90-100bpm）	Fast music（150-160bpm）
Low intensity（50%-60%HRR）	No music×Low intensity	Slow music×Low intensity	Fast music×Low intensity
High intensity（70%-80%HRR）	No music×High intensity	Slow music×High intensity	Fast music×High intensity

2.4.1. Only the scales of fatigue in the case of 15 have been explained, and what the rest of the Borga scale looks like?

Reply:

Thank you for your reminder. the RPE scale is detailed in Table 2, which we have supplemented in 2.4.1 section. When the RPE is 15, the exerciser will exhibit shortness of breath and develop significant muscle fatigue. When exceeding 15, there may be severe dyspnea or muscle fatigue. To ensure the safety of the experiment, we chose RPE=15 as the observation index.

Table 2 Borg's scale for rating of perceived exertion

Score	Subjective Exercise Intensity	Subjective Exercise Fatigue	Score	Subjective Exercise Intensity	Subjective Exercise Fatigue
6	No exertion at all	Not hard at all	14	-	-
7	Extremely light	Extremely relaxed	15	Hard (heavy)	Tired
8	-	-	16	-	-
9	Very light	Very relaxed	17	Very hard	Very tired
10	-	-	18	-	-
11	light	Relaxed	19	Extremely hard	Extremely tired
12	-	-	20	Maximal exertion	Try the best
13	Somewhat hard	A little tired

Reference:

Morishita S, Yamauchi S, Fujisawa C, et al. Rating of perceived exertion for quantification of the intensity of resistance exercise. Int J Phys Med Rehabil, 2013, 1(9): 1-4.

2.4.2. There were only 2 heart rate measurements? What device was HR measured, more technical details needed.

Reply:

Thank you for the careful review. I apologize that this was not clearly written in the original article. In fact, we performed 2 measurements in each set of experiments. However, it is important to emphasize that neither of the 2 measurements was an immediate measurement, but a continuous measurement.

The first measurement was taken before the run to collect the resting HR of the subject before the run. After sitting in a comfortable position for 5 minutes, the subject's HR was collected for the next 1 min and the average value was calculated as the subject's resting HR. This is described in section 2.6.1.

The second time was measured during running, where the subject was wearing a heart rate belt during running, and we would collect the HR signal during the whole running process of the subject. However, in the data processing, only the HR signal 5 seconds before the end of the run was selected, and the average of the HR signal 5 seconds before the end of the run was used as the indicator of the immediate post-run HR. The reason for intercepting the HR signal 5 seconds before the end of the run is that we want to understand the difference of the exercise HR of the subjects before and after the run, and the HR during the run is showing a regular increase, so we intercepted the average of the HR signal 5 seconds before the end of the run as the immediate HR at the end of the run.

We have added the details in section 2.4.2.

2.4.3.

Which lower limb was examined, perhaps both?

Reply:

Thank you for the reminder! The description of this detail was missing in the original article, and in fact, we chose the dominant leg for the measurement. This information has now been added in sections 2.4.3 and 2.6.3. section.

What is MF? One can presume but the reader should be given this.

Reply:

Thank you for your valuable suggestions. We have added the information of median frequency(MF) briefly in section 2.4.3.

The median frequency (MF) is one of the most commonly used indicators as a method of frequency domain analysis of sEMG signals, and is more stable for feedback of muscle fatigue state. MF refers to the median value of myofiber discharge frequency during skeletal muscle contraction. High frequency discharge is the main manifestation of excitation of fast muscle fibers, and slow muscle fibers are dominated by low frequency potential activity.

Reference:

Kupa E J, Roy S H, Kandarian S C, et al. Effects of muscle fiber type and size on EMG median frequency and conduction velocity[J]. Journal of applied physiology, 1995, 79(1): 23-32.

Reply:

Thank you for your reminder. We agree with you that describing the details clearly will make the study more reproducible. Therefore, we have supplemented the details of sEMG signal acquisition, mainly in sections 2.4.3 and 2.5. The details are as follows.

The RF and VM sEMG signals of the subject's dominant leg were collected. The subject was first asked to perform thigh flexion and extension movements to find the target muscle location. Once the target muscle was found, the location of the target muscle was marked. Subsequently, use cotton dipped in alcohol to clean the skin surface dirt, remove sweat, sebum and other impurities on the skin surface, paste the electrodes again after the skin is dry, and remove surface hairs if necessary. The purpose of taking the above measures is to reduce the impedance effect of the skin, enhance the adhesion of the electrode patch to the skin and obtain the best recording effect. Then paste the electrodes along the tissue direction of the target muscle with positive and negative inputs, and the point spacing of the electrode patch is 2-3 cm. Then the EMG signal acquisition software (Acknowledge4.2, Biopac.) was connected to the sensor, and then the subject was asked to perform flexion and extension movements, and observe whether the corresponding EMG signal in the interface of the acquisition software produced obvious changes, if the signal did not show obvious changes, or the signal showed abnormal values, the muscle selection position was further calibrated, and the electrode patch was checked with the skin whether the connection is firm. By continuously testing until the signal is stable. The above steps were followed before each acquisition of sEMG signals. The electrodes used for testing were AgCl electrocardiographic electrodes. The EMG100c EMG signal amplifier was then secured to the subject's lower leg with a strap.

The sEMG signals were collected using the MP150 telemetry physiological recorder (Biopac Inc., Goleta, CA, USA) and its accessories at a sampling rate of 1000 Hz, equipped with two EMG100c EMG signal Fish disposable ECG electrode patches (Shanghai Huyou Medical Electrode Co., Ltd.) were used to connect the target muscle to the EMG signal amplifier, with AgCl composition, and the size of the electrode patches was trimmed before use so that they could meet the electrode patch 2-3 cm spacing distance.

2.6.1.

The study took place in a standing position, was it worth studying resting HR while standing... Does HR while standing and sitting differ?

Reply:

Thank you for your careful review. For the measurement of resting HR, we considered it as follows.

(1) The resting HR before running is not the focus of this study; we are more interested in the HR difference before and after running. The purpose of measuring resting HR is only to obtain a baseline data so that we can observe the difference in HR before and after running. Resting HR was only treated as baseline data in the study, and the posture at the time of measurement did not affect the experimental results. (2) Measuring resting HR in a sitting state is one of the common ways to measure resting HR, and in many running experiments, resting heart rate is measured in a sitting state(Alansare et al. 2021;Tian et al. 2021; Chai et al. 2019).

Reference:

1.Alansare AB, Hayman J, Lee J-M, Seo M-W, Yoo D, Jung HC. The Efficacy of a Calamansi-Containing Energy Drink on Running Performance and Recovery in NCAA Division I Middle-Distance Runners: A Preliminary Study. International Journal of Environmental Research and Public Health. 2021; 18(21):11023.

2.Tian S, Mou H, Fang Q, Zhang X, Meng F, Qiu F. Comparison of the Sustainability Effects of High-Intensity Interval Exercise and Moderate-Intensity Continuous Exercise on Cognitive Flexibility. International Journal of Environmental Research and Public Health. 2021; 18(18):9631. https://doi.org/10.3390/ijerph18189631.

3.Chai G, Wang Y, Wu J, et al. Study on the recognition of exercise intensity and fatigue on runners based on subjective and objective information[C]//Healthcare. Multidisciplinary Digital Publishing Institute, 2019, 7(4): 150.

2.6.3.

Why was this type of premeasurement used, in the literature MVC is the most common?

Reply:

Thank you for your careful review. We actually took this into consideration when conducting the experimental design, but after comparing the two approaches, we finally chose this premeasurement type. In this regard of this question, we considered the following.

First, MVC describes the maximum voluntary muscle contraction, and the results are strongly influenced by the subject's subjective will, which may interfere with the results of this experiment. In Bazzucchi et al.’s study, when measuring the subject's MVC in the experiment, the test leader would need to give strong verbal encouragement to encourage the subject to "push as hard as possible " to produce maximum contraction. This also confirms that MVC can be influenced by the subject's subjective will when measured, and that exertion may be influenced when encouragement is given or when the subject develops positive psychological cues. The results of the maximum voluntary contraction (MVC) study have been criticized by some scholars in terms of reliability and validity. For example, Peacoc et al. conducted experiments using the quadriceps muscle as an example and found that when subjects received informative feedback, it resulted in an increase in maximal voluntary isometric contraction. Secondly, the study by Kankaanpää et al. showed that quantitative loading based on body weight ratios to detect muscle fatigue is a feasible method. Finally, due to objective conditions, our laboratory does not have the necessary devices for MVC measurement at the moment, and a weight-percentage-based load measurement is easier to implement.

Combining these three considerations, we believe that premeasurement type can also achieve amplitude standardization, and the results can be more accuracy. Therefore, we finally chose to that measurement type as described in the paper.

Reference:

1.Bazzucchi I, Patrizio F, Ceci R, Duranti G, Sabatini S, Sgrò P, Di Luigi L, Sacchetti M. Quercetin Supplementation Improves Neuromuscular Function Recovery from Muscle Damage. Nutrients. 2020; 12(9):2850.

2.Peacock B, Westers T, Walsh S, et al. Feedback and maximum voluntary contraction. Ergonomics, 1981, 24(3): 223-228.

3.Kankaanpää, M., Taimela, S., Webber, C. et al. Lumbar paraspinal muscle fatigability in repetitive isoinertial loading: EMG spectral indices, Borg scale and endurance time. Eur J Appl Physiol 76, 236–242 (1997).

The figure shows that the electrode location is in the vastus lateralis muscle rather than the rectus femoris...it is difficult to hit the RF with such large electrodes.

Reply:

Thank you for your careful review. First of all, regarding the inaccurate position of rectus femoris electrode patch you mentioned, we think it is necessary to clarify as follows:

From the standard muscle anatomy diagram (Figure 1-A), the position of the RF is distributed in the middle of the most anterior part of the thigh. However, considering that each individual's muscle growth position will have a certain degree of deviation, we did not determine the location of RF directly by visual observation, but rather we determined where the RF was located based on the actual site of force generated by each subject during the flexion and extension movements. This also leads to why the position of the electrodes in the paper's accompanying figure (Figure 1-B) is visually biased towards the location of the vastus lateralis muscle of the thigh.

In determining the location of the RF electrode patch, this study was conducted strictly according to the following steps: following the muscle location in the standard illustration of the RF (Figure 1-A), the subject was asked to continuously perform thigh flexion and extension movements to find the muscle. After finding the target muscle, the location of the target muscle was marked, and then the electrode patch was applied to the marked muscle location. The positive and negative electrodes were pasted along the tissue course of the muscle. Then the sEMG signal acquisition software was connected to the sensor, and then the subject was asked to perform flexion and extension movements to observe whether the corresponding sEMG signal in the interface of the acquisition software produced significant changes, and if the signal did not show significant changes, or if the signal showed abnormal values, the muscle selection position was further calibrated and the connection between the electrode patch and the skin was checked for solidity. By continuously testing until the signal is stable. The above steps were followed before each acquisition of sEMG signals.

When performing the rectus femoris position determination on the subjects in the original accompanying figure of the paper (Figure 1-B), we also followed the above rules to proceed by constant calibration in order to find the position of the rectus femoris muscle. Therefore, we are not confusing the position of the rectus femoris and the vastus lateralis muscle, but we want to determine the position of the RF by calibrating it in a more precise way.

However, your concern also led me to discover that the muscle growth position of this subject is somewhat offset compared to the standard muscle schematic. Also, in order to show the position of the vastus medialis muscle more clearly, the picture was taken at an oblique angle, which also caused the position to appear shifted. Therefore, the electrode patch image of this subject was not the most appropriate choice for illustration, which does tend to mislead the reader into misinterpreting the position of the muscle paste. In order to avoid making more readers also have such doubts, therefore, we think it is very necessary to replace this matching picture, as shown in Figure 1-C.


(A)Anatomical map of muscle	(B)the original pictures	(C)the new pictures

Figure 1 Position of the rectus femoris muscle

Secondly, regarding the electrode size you mentioned, we would like to respond as follows.

The electrode slice we use is Fish brand disposable ECG electrode(Shanghai Huyou Medical Electrode Co.). Figure 2-A shows the original size of the electrode slice, Figure 2-B shows the reverse side of the electrode sheet after trimming, and Figure 2-C shows the front side of the electrode slice after trimming. The electrode slice used in Figure 1-B of the original paper is the original size of the electrode slice, the overall diameter is 5cm (including non-woven fabric), the diameter of the adhesive media part is 2cm, and the diameter of the electrode interface is 1cm. we will overlap the electrode slice when pasting. The position of the electrode interface is precisely aligned with the marked position, and the interval between the two electrodes is kept between 2-3cm, so the electrodes are able to hit the target muscle position precisely.

However, due to the large area of the non-woven fabric, the position of the target muscle collected could not be visualized. Therefore, we trimmed the electrode slice, as shown in Figure 2-B and Figure 2-C. The overall size of the trimmed electrode slice is 3 cm. we use the trimmed electrode slice for illustration, as shown in the replacement mating figure (Figure 1-C).

Figure 2. Electrode size

Thank you again for this reminder. Your suggestion is invaluable to us, because from the author's point of view, we cannot directly detect the imprecision of the representation of these images. However, your suggestion lets us know that our representation is not precise and intuitive to the reader from the reader's point of view. In our future academic research, we will also pay attention to the representation of schematic diagrams so that both the text and the images of the article can accurately convey the true intent we want to express.

The above are the modifications we made according to your suggestions. Thank you again for your time in reviewing this paper and hope that our revisions will meet with your approval. If you are not satisfied with any of the changes we have made, or have further comments on other parts of the paper, please contact us. We are looking forward to your further response, which will make our research work even better.

Reviewer 2 Report

I would like to thank the editor for the opportunity to review the manuscript entitled "The Effect of Music Tempo on Fatigue Perception at Different Exercise Intensities". I read with great interest the manuscript and I found it really interesting and well written. The scientific design the data reporting are appropriate and the findings novel and interesting. I congratulate with the authors for the really well-done work.

I have only a minor question, related to fatigue and neuromuscolar performance. Did the authors monitored for participants' hydration status prior to any experimental session? There is evidence that it might influence neuromuscolar performance (Bigard et al., Med Sci Sports Exerc, 2001; Pallarés et al., J Int Soc Sports Nutr, 2016; Zubac et al., Eur J Sports Sci, 2020) and therefore it could be important to consider this factor when multiple testing are performed.

Author Response

Open Review 2

I have only a minor question, related to fatigue and neuromuscolar performance. Did the authors monitored for participants' hydration status prior to any experimental session? There is evidence that it might influence neuromuscolar performance (Bigard et al., Med Sci Sports Exerc, 2001; Pallarés et al., J Int Soc Sports Nutr, 2016; Zubac et al., Eur J Sports Sci, 2020) and therefore it could be important to consider this factor when multiple testing are performed.

Reply:

Thank you very much for your affirmation and recognition of our work, and for your careful review of our articles. Regarding the suggestions you mentioned, we respond as follows.

Thank you for providing us with this valuable literature, which we understand after careful reading that dehydrated states can lead to poor performance during prolonged low-intensity and high-intensity dynamic exercise (Bigard et al., Med Sci Sports Exerc, 2001). However, in our experiments, the average completion time for each group was in the range of 10 min. As we observed in our experiments, the short duration of exercise around 10 min did not result in a large amount of sweating and participants were not in a dehydrated state at the end of the run, so we did not make this a focus of the study to measure in any experimental session.

But just as you say, hydration status does affect neuromuscular performance, and this is something we took into account earlier, in the rules we asked participants to follow during the experiment. In section 2.1, we mentioned that our experiments are repeated measurement experiments and that participants are required to confirm compliance with the following requirements before each experiment: "(2) Maintain a previous habitual diet and ensure that the experiment is performed at least one hour after the meal, and avoid eating, drinking alcohol, or consuming too much drinking water within one hour before the experiment to maintain a good hydration status. "

Finally, thank you very much for this valuable advice, which shows us the importance of considering hydration status when performing neuromuscular related experiments, and we will pay more attention to this point in future related studies.

Reference: Bigard A X, Sanchez H, Claveyrolas G, et al. Effects of dehydration and rehydration on EMG changes during fatiguing contractions. Medicine and science in sports and exercise, 2001, 33(10): 1694-1700.

Reviewer 3 Report

Many thanks to the authors for this beautiful work. Each part of the article is written in detail. The details in the method section are especially great. But there are some questions that need to be answered.

-How is the applied form of the 6 experimental protocols mentioned in the method section? It should be clarified. So how is the ranking effect neutralized? The order of administration of the 6 tests should be explained. How was the sample group reached? The sample size is small, so a sample size calculation must be added.

G*power analysis should be added. -After rigorous screening mentioned in the method section? Information about this screening should be given? For example, which screening?

-Same time of the Day? Please specify the time zone.

-The decision and protocol number should be written for the approval of the ethics committee.

-In the Data Processing and Analysis section, the authors wrote "repeated-measures ANOVA". It should be corrected.

Two-way.... Why did the authors use the Paired Sample test? Bonferroni test is used mostly for pairwise comparisons. In addition, partial eta squared values and effect sizes for pairwise comparisons should be calculated and evaluated according to effect size classification. Reporting of the findings (comments under the Tables) should be made and revised according to the Two-way repeated-measures ANOVA test.

-The discussion section is weak. The research findings in the discussion part of the study are given in detail. Instead, this section should be enriched with more scientific resources. Not enough resources have been used in the discussion section. This section should be revised taking into account the research hypotheses.

-Practical application section should be added.

Author Response

Open Review 3

Below is a description of the changes we made to each comment and our response.

Reply：

First of all, we make the following responses to the application forms of the six experimental schemes:

Table 1 Exercise schemes

Intensity	No music	Slow music(90-100bpm)	Fast music(150-160bpm)
Low intensity（50%-60%HRR）	No music×Low intensity	Slow music×Low intensity	Fast music×Low intensity
High intensity（70%-80%HRR）	No music×High intensity	Slow music×High intensity	Fast music×High intensity

Secondly, for the neutralization of the ranking effect you mentioned. We respond as follows:

Because this experiment is a within-subjects experiment, each subject needs to repeat the measurement six times, so there may be ranking effects. If the rank of trials is not handled properly, additional variables may be introduced and interfere with the experimental results. In fact, we took this into account before conducting the experimental design and counterbalanced the order effects. However, we apologize for not explaining this clearly in the original manuscript and for raising your concerns. For this reason, we will elaborate on the control of experimental order here.

To prevent ranking effects from interfering, this experiment was sequenced using the Latin Square Design Method (LSD). LSD simulates the order of experiment expansion, and can balance the effects of experimental ranking by generating an independent randomized ranking table that can take into account the orthogonality and uniformity of the experimental protocols.

We numbered the exercise protocols as follows: A: no music × low intensity; B: slow tempo × low intensity; C: fast tempo × low intensity; D: no music × high intensity; E: slow tempo × high intensity; F: fast tempo × high intensity. The Latin square matrix was used to assign the experimental ranking, and the experimental order was obtained as shown in Table 2. From the experimental ranking presented in Table 2, we can see that A/B/C/D/E/F appeared equally in each order, which avoids the effect of experimental ranking on the results. Each subject performed the experiment according to the order in the Latin square matrix.

However, considering the space limitation, we did not describe the arrangement of the experimental order in detail in the text, but a brief addition is made in Section 2.3.3.

Table2 Experimental ranking

Participants number	Experimental ranking
Participants number	1	2	3	4	5	6
No.1	A	B	C	D	E	F
No.2	B	C	D	E	F	A
No.3	C	D	E	F	A	B
No.4	D	E	F	A	B	C
No.5	E	F	A	B	C	D
No.6	F	A	B	C	D	E
No.7	A	B	C	D	E	F
No.8	B	C	D	E	F	A
No.9	C	D	E	F	A	B
No.10	D	E	F	A	B	C
No.11	E	F	A	B	C	D
No.12	F	A	B	C	D	E
No.13	A	B	C	D	E	F
No.14	B	C	D	E	F	A
No.15	C	D	E	F	A	B
No.16	D	E	F	A	B	C
No.17	E	F	A	B	C	D
No.18	F	A	B	C	D	E

Reference:

1.Bradley J V. Complete counterbalancing of immediate sequential effects in a Latin square design[J]. Journal of the American Statistical Association, 1958, 53(282): 525-528.

2.Kim B G, Stein H H. A spreadsheet program for making a balanced Latin square design[J]. Revista Colombiana de Ciencias Pecuarias, 2009, 22(4): 591-596.

Finally, about the sample size you mentioned. We respond as follows.

Thank you very much for this suggestion, and we admit that we did neglect to consider it before the experimental design and only determined the sample size based on subjective assessment without performing the sample size calculation. After your reminder, we immediately conducted a priori analysis of the required sample size in the study using G*power, with the presumption of the presence of a medium effect size of f=0.25, statistical test power=0.8, and significance level α=0.05. The results of the analysis indicated that a sample size of 18 would be sufficient to achieve a medium effect size interaction effect. Therefore, we immediately supplemented the experiment with 5 additional groups to bring the sample size to 18, and the data were reorganized and analyzed. The results from the additional sample size showed that although there were changes in the descriptive values, the overall trend did not differ significantly from the previous results for the 13-group sample.

-After rigorous screening mentioned in the method section? Information about this screening should be given? For example, which screening?

Reply:

Thank you for your careful review. We added the following screening conditions:

In order to obtain more generalized findings, the experimental subjects were selected as general college students. All subjects had no significant differences in physical indicators, were between 20-30 years old, 170-180 cm in height, and 55-75 kg in weight to avoid differences in results due to age, height, and weight. No regular fitness habits, no running fitness scientific theory or method knowledge guidance; no muscle, bone, respiratory system or cardiovascular diseases. No skeletal muscle dysfunction of the lower extremities; and did not exhibit any cardiovascular or peripheral vascular disease, chronic disease or neurological or muscular dysfunction. No wounds or scratches on the thigh muscles and no allergy to alcohol components.

You can see our modifications in section 2.1.

-Same time of the Day? Please specify the time zone.

Reply:

We chose 9:00 am to 11:00 am for the experiment. There are two reasons for this: (1) Individuals are awake and non-fatigued during this time period, and choosing this time period can exclude the interference brought by the sleepiness/fatigue factor to the experiment. (2) The physiological rhythms of people are more similar during the same time period.

-The decision and protocol number should be written for the approval of the ethics committee. Reply: The programme number has been supplemented in the manustript.

-In the Data Processing and Analysis section, the authors wrote "repeated-measures ANOVA". It should be corrected.

Reply:

Thank you for your suggestion. In conjunction with your next suggestion, we have reorganized our thoughts on data processing and analysis and corrected this section. The details are as follows:

All results are presented as group means and standard deviations. The normality of the data distribution was confirmed using the Shapiro-Wilk test. To determine the effect of the intervention on the dependent variable, a two-way analysis of variance (ANOVA) for repeated measures was used to calculate between-group differences. If the interaction between music tempo and exercise intensity was significant, Bonferroni was calculated for post hoc tests. In addition, effect sizes (ES) were determined from the ANOVA output by converting the partial eta squared value to Cohen d. The criteria for classifying Cohen d were as follows: small (0 < d < 0.5), medium (0.50 ≤ d < 0.8) , and large (d ≥ 0.80). The significance level was 0.05. All statistical analyses were performed using SPSS 26.0 (SPSS Inc., Chicago, IL, USA).

Reply:

Thank you very much for your suggestion. After reorganizing our thoughts on data processing and analysis, we decided to use the Bonferroni test for pairwise comparisons. And partial eta squared values were reported and was converted to Cohen d to determine the effect size (ES) from the ANOVA output. Moreover, the comments of the study results are expanded based on the Two-way repeated-measures ANOVA test, and the original discussion of the paired samples t-test was removed.

Reply: Thank you for your careful review. We have expanded the discussion section and enhanced the comparison of the results with other studies, and added literature citations to make the results more scientific.

-Practical application section should be added.

Reply：

Thank you for your valuable suggestions. We have added a section “4.4 Pratical application” to the discussion section to discuss the practical implications of the study. The details are as follows.

Round 2

Reviewer 1 Report

Thank you for the valuable changes to the article and the work you have put into it. A few doubts remain:

1. Why did you change the number of participants to 18 and the sets to 108 in the manuscript?

2. Is the added table 3 necessary? It seems fairly obvious.

3. Should not Table 5 be referred to as a figure?

4. Is the calibration of electrode positioning as now described in the article recommended in the literature? Although such a corrected sticking of the electrodes as described in the authors' article may seem somewhat logical, it is useful to have a reference to the literature.
For the position of the electrodes on the skin of the subjects, the SENIAM methodology is usually used, as in this article - https://www.nature.com/articles/s41598-021-87001-8
It seems, now from Table 5, that perhaps this methodology has been used.

Author Response

Open Review 1

Thank you for the valuable changes to the article and the work you have put into it. A few doubts remain:

Thank you for your recognition of our revision work. Below is a description of the changes we made to each comment and our response.

Why did you change the number of participants to 18 and the sets to 108 in the manuscript?

Reply: We are sorry for not clarifying this in the previous round of responses. This is because, in the previous round of revisions, another peer reviewer mentioned that the selection of the sample size should be determined to be reliable based on the G*power sample size analysis. Subsequently, we conducted a priori power analysis of the required sample size in the study using G*power with a presumption of the presence of a medium effect size of f=0.25, statistical test power=0.8, and significance level α=0.05, and the analysis showed that a sample size of 18 would be sufficient to achieve a medium effect size interaction effect. Therefore, we immediately supplemented the experiment with five additional groups to bring the sample size to 18 and make the conclusions more convincing. After rearranging and analyzing the data, it was found that although there were changes in the descriptive values, the overall trend did not differ significantly from the previous results for the 13-group sample.

Is the added table 3 necessary? It seems fairly obvious.

Reply: Thanks to your suggestion, we have removed Table 3 and replaced it with a textual description of the experimental groups. You can see the details of the revision in section 2.3.3

Should not Table 5 be referred to as a figure?

Reply: Thank you for your careful review, we have corrected this error. The subsequent figure and table captions have been corrected accordingly.

Is the calibration of electrode positioning as now described in the article recommended in the literature? Although such a corrected sticking of the electrodes as described in the authors' article may seem somewhat logical, it is useful to have a reference to the literature.

For the position of the electrodes on the skin of the subjects, the SENIAM methodology is usually used, as in this article - https://www.nature.com/articles/s41598-021-87001-8 It seems, now from Table 5, that perhaps this methodology has been used.

Reply: Thank you for your careful review. We have added references to the electrode positioning calibration method, and the position of the electrodes on the skin of the subjects, we have followed the SENIAM method. We have also added the details of electrode sheet adhesion. You can see the details of the revision in section 2.4.3.

Additional reference:

Hermens, H. J., Freriks, B., Disselhorst-Klug, C. & Rau, G. Development of recommendations for SEMG sensors and sensor placement procedures. Electromyogr. Kinesiol. 10, 361–374 (2000).
Wolf S L. Essential considerations in the use of EMG biofeedback. Physical Therapy, 1978, 58(1): 25-31.
Pakosz P, Domaszewski P, Konieczny M, et al. Muscle activation time and free-throw effectiveness in basketball. Scientific Reports, 2021, 11(1): 1-8.
Stegeman D, Hermens H. Standards for surface electromyography: The European project Surface EMG for non-invasive assessment of muscles (SENIAM). Enschede: Roessingh Research and Development, 2007, 10: 8-12.
Hermens H, Freriks B. The state of the art on sensors and sensor placement procedures for surface electromyography: a proposal for sensor placement procedures[M]. Enschede, The Netherlands: Roessingh Research and Development, 1997.

Finally, we would like to thank you again for your guidance and help in this study and hope that our revisions can get your approval.

Reviewer 3 Report

I could not see the partial eta squared values in the comments below the table, these values should be given. It should be interpreted according to the cohen d classification specified in the statistical analysis section (small (0 < d < 0.5), medium (0.50 ≤ d < 0.8) , and large (d ≥ 0.80).

Author Response

Comments and Suggestions for Authors

Reply：Thank you for your reminder. We have explicitly listed the partial eta squared values (η²) in the table, as detailed in Table 6. In the discussion in section 4, we have also discussed η2 as an effect size value indicator and classified the effect levels of effect value musical tempo and exercise intensity with Cohen d's effect value classification criteria.

Finally, we would like to thank you again for your guidance and help in this study and hope that our revisions can get your approval.