The Influence of Forest Landscape Spaces on Physical and Mental Restoration and Preferences of Young Adults of Different Genders

Round 1

Reviewer 1 Report (Previous Reviewer 2)

The authors made significant improvements in the text and I believe it is much better and clearer now. But I still have some concerns regarding the analysis and their interpretation and related to that, with conclusions that are not in line with the results.   

GENERAL:

 

If all individual variables had normal distributions and authors performed ANOVA on them, why did authors decide to leave Kruskal Wallis test too? It is not good to have both ANOVA and Kruskal Wallis on all variables, especially since they show contradictory results. If authors decide to leave it this way, they must heavily argue why, and also describe and interpret the differences between ANOVA and Kruskal Wallis. ANOVA showed that no differences between landscapes are present, just pre-post changes, and the authors do not even mention this nor discuss it.

 

Row 165-166: I can accept the author's response and explanation on the control group, but in that manner, since there is no control group, I would remove aim 1 (The effects of forest landscape space on physiological and psychological responses in young adults of a different gender)

 

SPECIFIC COMMENTS

 

Row 317-322 I would put the first paragraph of the results in the Participants description part, including the table.

Row 327 (results of three-way ANOVA with the number): Statistical parameters are added later on table 5 which is not good. First main effects and interactions should be shown (table 5) and then after that post-hoc test (table 3).

Row 339 (Table 3 Pairwise comparison): Which post-hoc test was performed?

 

Row 353 (In addition, changes in SBP (p<0.001) and DBP (p<0.05) ): Not changes, this is the main effect of gender it only shows gender differences on DBP and SBP between two genders

 

Row 354-355 (Therefore, the following further analyzed the influence of space type and gender on the physical and mental changes of the participants.): I am not sure what are authors saying here, can you rephrase it, please?

 

Row 357 (table 5): Since there is a S*E*G interaction on HR I would expect to see detailed post-hoc tests here in order to see where interaction comes from

 

Row 365 (First, for males, their ΔSBP, ΔHR and ΔP showed): Statistical parameters should be added (test value, df, p… for global effects also, not just post-hoc significances)

 

Row 387-393: Statistical parameters should be added (test value, df, p… for all effects also, not just significances)

 

Row 405-422: Statistical parameters should be added (test value, df, p… for all effects also, not just significances)

 

Row 437(space was affected by gender): How, who had higher scores?

 

Row 446-449 (in terms of potential preference trends, the males preferred the dynamic…): We cannot say even that, there were no trends, no differences were significant

 

Row 451 (Independent sample t-tests were subsequently): No t-tests, please use post-hoc test from ANOVA

 

Row 469 (gender were significantly different in the water space and the forest space): Which test was done here, please add statistical parameters (test value, df, p, effect size…)

 

Row 486-487 (were not directly related to their preferences): Were not related, no need to claim a direct relationship since no mediation tests were performed. Also, these are not physical responses, but physiological parameters.

 

Row 497-499 (that the forest landscape had a certain restorative effect on people's physical and mental states): Without the control group we cannot claim this, we can just compare different landscapes if they differ among themselves in registered changes; authors did not test if restoration exists (no control) so this cannot be the conclusion of their study, it is not visible in the results.

 

Row 522 (not all green space environments have the same impact on people's...): I am not sure, since ANOVA did not find these differences.

 

Row 540-542 (Although there is no statistical difference, from the numerical point of view female preference for mixed forest space is also higher than male.): This is a contradiction, if something does not exist it does not exist, we cannot claim that we did not detect a difference although it surely exists (differences are not significant in this case).

 

Row 588-590 (In other words, when participants are in a highly preferred forest landscape space, their emotional state will positively change, which will benefit their cardiovascular health (Figure 9).): This conclusion cannot be noted, since no mediation or moderation effects were tested. It is beyond the results presented in this paper.

 

Row 633-638 ANOVA and Kruskal Wallis test show different results, why? This conclusion is not obvious as the authors claim at the end.

 

Row 645 (simultaneously promoted cardiovascular relief): But no, correlations of preference with physiological parameters were not significant. Also, the correlation with TMD was negative, meaning that higher preference led to smaller changes in the disturbance.

 

 

ADDITION TO AUTHORS RESPONSES ON ROUND 1 COMMENTS

 

Comment 7 – I think it would be important for authors to add this explanation in the text, not just in response to the reviewer, so readers of the paper could know it also.

 

 

Comment 9 - I think it would be important for authors to add this explanation in the text, not just in response to the reviewer, so readers of the paper could know it also. Just shortly add that the camera recorded sound too, and that HTC reproduced that audio too. 

Author Response

Dear reviewer,

Thanks for your valuable suggestions, we have made the following modifications according to your comments and suggestions.

In addition, our revised manuscript uses the "Track Changes" function. The specific modifications are as “Response letter1”

Author Response File: Author Response.docx

Reviewer 2 Report (Previous Reviewer 3)

The manuscript has been significantly improved, and I appreciate the authors' efforts to respond to reviewer comments.

They have satisfactorily addressed most of my concerns. The only remaining concerns I have with POMS are seemingly major:

POMS reflected three emotional dimensions: "interest", "relaxation" and" worry"???

How are the anxiety, interest, and relaxation scores from the POMS questionnaire calculated?

The POMS questionnaire assesses six mood subscales: tension-anxiety, depression, anger-hostility, vigor, fatigue, and confusion. 

Again, you must state in the materials and methods section how many participants took part in your study.

A week before the start of the experiment, (Reference is needed). Stop drinking tea (why, Reference).

TMD was calculated as follows: total mood disturbance = worry - (interest + relaxation) (Reference is needed).

Preference: In this study, the evaluation of participants' preference for forest landscape spaces was assessed by questionnaire. The question was "how do you like the forest landscape space. (Reference is needed).

There is no control group, which is a significant issue. To ensure that the measured benefits are due to forest landscape, it is necessary to compare it to urban space. (Limitation of the study).

Table 3. It is preferable to include the definition of SBP,……….. under the table.

Was the research protocol approved by the University Ethical Committee?

Best wishes

Comments for author File: Comments.docx

Author Response

Dear reviewer,

Thanks for your valuable suggestions, we have made the following modifications according to your comments and suggestions.

In addition, our revised manuscript uses the "Track Changes" function. The specific modifications are as “Response letter2”

Author Response File: Author Response.docx

Round 2

Reviewer 1 Report (Previous Reviewer 2)

The authors made significant improvements and now it is clear to me how results should be analyzed. 

But, my main concern is still with the analysis part. Namely, since authors clearly state that all variables had normal distributions (rows 327-328 For the same index, the data before TSST, before VR, and after VR are all normally distributed), there is no need to calculate differences (pre-post) and no need to perform non-parametric analyses (Kruskal-Wallis nor Mann-Whitney) for testing the significance of differences. On all measures, ANOVA can be applied, with post-hoc tests.

Besides that, a combination of ANOVA and t-test is not a good option due to Type 1 error inflation. In this case, 2-factorial ANOVA with post-hoc tests (LSD, Sidak, or Scheffe for example) should be performed.

 

According to previously written, authors should make changes in the following rows:

Rows 354-371 – authors did not perform correct analyses, a combination of ANOVA and paired samples t-test is not a good option due to Type 1 error inflation. As it was previously commented in this case, 2-factorial ANOVA with post-hoc tests (LSD, Sidak, or Scheffe for example) should be performed.

 

Rows 382-415 and 430-457, as I already mentioned in the previous round of review, instead of the Mann Whitney U test, post-hoc comparisons of the Kruskal Wallis test should be calculated. But, once again, I would strongly recommend abandoning these non-parametric tests and to stick with ANOVA. Now that authors explicitly mentioned that all variables had normal distributions, there is no need to calculate differences (pre-post) but instead, repeated measures ANOVA should be used, with one between-subject factor (landscape type) and one within-subject factor (pre-post, with three levels, before TSST, before VR and after VR).

 

Rows 473-486, also here, authors did not perform correct analyses, a combination of ANOVA and independent samples t-test is not a good option due to Type 1 error inflation. As it was previously commented in this case, 2-factorial ANOVA (two between-subject factors) with post-hoc tests (LSD, Sidak, or Scheffe for example) should be performed.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

This manuscript is a resubmission of an earlier submission. The following is a list of the peer review reports and author responses from that submission.

Round 1

Reviewer 1 Report

1. The abbreviation of POMS cited in this manuscript is incorrect. The authors of Citation [55] did not confirm the suitability of POMS. 2. Also, it is meaningless to discuss the suitability of the Japanese version of POMS [55] in Chinese studies. You must cite literature certifying that the Chinese version of the POMS you use reflects the characteristics of the original English version of the POMS. 3. POMS has 5 subscales (tension-anxiety, depression, anger, vitality, fatigue, confusion) and calculates TMD based on them. TMD=T+D+A-V+F+C Authors should provide a method for calculating TMD used in this manuscript and the rationale for calculating "worry", "interest", and "relaxation". TMD was calculated as: Total Mood Disorder = Worry - (Attention + Relaxation). (line 263)

Reviewer 2 Report

General remarks:

The main remark – the control group is missing. How do we know that all effects are not just due to time passing, or just due to VR (not landscapes, just VR experience per se)? Without it, we cannot relate findings to landscape viewing at all. I would say that a control group (or even 2 controls) is a necessity in this kind of research design.  

The second general remark is related to statistical analyses (Lines 310 – 321). Differences between various landscapes do not prove that change itself is significant, they only show that some change exists and that it varies but it does not prove that the effect of change (pre-post VR) is significant. So, the authors must change the analysis to do so, they should apply two factorial ANOVA, with one repeated factor in order to test this directly.

Also, gender should be included as a factor, and I propose three factorial ANOVA, with one repeated factor and two between-subject factors (landscape and gender).

Lines 420 – 428, instead of using plain correlations it would be much better if preference ratings would be used as a covariate in the proposed 3-factorial ANOVA, or ANCOVA

 

Specific remarks:

 

Line 87 (Gender is a biological determinant of human health). This is a too strong statement for existing findings, authors can say it is related to or so…, but not mark it as a determinant

Lines 118-119 (may not benefit 118 all tourists equally). Suddenly „tourists“ appear in the paper, but they are not the focus of it, findings can be related to anybody, not only tourists

Lines 146-152. What are the differences between the Shi study and this paper, what is the novelty of this paper, and authors should describe the differences from the Shi study explicitly?  

Line 195 (six types of forest landscape spaces). But only 3 are forest, 2 are water and 1 is overlooked, so it does not seem like there are 6 forest landscapes, right?

Lines 205 – 220. How was sound presented to participants, and which audio devices were used for it?

Lines 224 – 230. The number of participants is lacking in this part (although it is mentioned later)

Lines 386 – 388 (Among them, the males preferred the dynamic water landscape space (5.52)). Significance testing is lacking here, repeated factor.

Lines 403 – 411. I am not sure how was this done, and I think it is not clearly presented in the methods part. Authors should describe better how did participants estimate, or described this, and how was it assessed…

Line 530 (influence of gender on people's physical). Word “influence” is too strong, relation or effects is much better (correlation does not imply causation). 

Reviewer 3 Report

Comments

This is an interesting study and the authors have collected an interesting result using. The paper is generally well written and structured. However, in my opinion the paper has some shortcomings regarding data analyses and text.

I have made numerous comments on the text below. In several cases, I suggested citing more recent and relevant literature.

Title is appropriate

Abstract

Line 11 and is not the suitable word to connect the two sentences. Please rephrase theses sentences.

Line 11-15 less information is better to show the study gab.

Abstract is well written, and everything is right – from my perspective.

Physiological indicators (blood pressure, 19 heart rate, and pulse), mood indicators (Profile of Mood States) and preference scores of young adults. ‘’Profile of Mood States and preference scores’’ is psychological indicators.

Introductions

The introduction is relevant and theoretical. There is enough information about previous study findings to allow readers to follow the current study rationale and procedures.

Please avoid repetition ‘’Different types of forest landscapes have been shown to have differential benefits on physical and mental restoration. Different pure forest spaces have been shown to alleviate different dimensions of anxiety of college students.’’

Rephrase ‘’In particular, a maple forest environment reduced anxiety about learning, and a birch forest environment reduced anxiety about employment’’

Line 56: ‘’ Lin et al. demonstrated that exposure to the forest from the out- 56 side had the best effects on physical and mental restoration’’. Compared to …………

Lines 47-59 Please try to employ conjunctions to make the information coherent and to facilitate the reader's understanding.

Lines 62-64 Please read the paper below, which demonstrates the impact of urban trees and their physical environment on stress relief measures:

https://doi.org/10.1016/j.ufug.2019.05.007

lines 95-97 Please read the paper below, which demonstrates the physiological and psychological effects of viewing different garden styles on gander.

https://doi.org/10.1016/j.ufug.2018.11.007

lines 162-163 please rephrase the sentence

lines 163-174. It is unnecessary to reiterate the information. Please make the information as succinct as you can.

Materials ad Methods

Line 169. What is the reason of doing nonexpert evaluations.

Major. Line 207-208 Therefore, it is necessary to combine visual stimuli with sound stimuli. Does the visual stimulation include any sound? Was the sound the same or different for each landscape stimulus?

Does the auditory stimulation affect the participants’ physical and psychological restoration? How can you determine that the effect was caused by the visual stimuli only?

Line 223. How many participants participated in the experiment?

How you recruited the participants?

Line 227. A week before the start of the experiment, (Reference is needed)

Line 227. Stop drinking tea (why)

Line 251. In this study, BP and P data were collected using an Omron electronic blood pressure monitor (Japan) (Reference is needed), and HR was measured using a Lepu ECG recorder ER2 (Reference is needed).

Line 260-261 reflected four emotional dimensions: "interest", "relaxation", "worry" and "total mood disturbance" (TMD) (Reference is needed).

POMS reflected four emotional dimensions: "interest", "relaxation" and"worry"???

Line 270. Only one question was used to assess the preferences of the participants.

Data analysis contains an abundance of information that might confuse the reader.

Results

Result section is generally clear.

Lines 292-296 It is preferable to depict participants' demographic characteristics in a table.

Line 298 change ‘p’ to ‘p’

Major: Line 359 How is the POMS questionnaire used to calculate worry, interest, and relaxation?

The POMS questionnaire assesses six mood subscales: tension-anxiety, depression, anger-hostility, vigor, fatigue, and confusion. 

Discussion

Please include more recent literature in the discussion section.

Conclusion

This chapter is more like a conclusion. Please revise and strengthen this chapter and provide conclusions and recommendations based on the results of your study.

Best wishes