The Influence of Immersive Virtual Reality Systems on Online Social Application

Round 1

Reviewer 1 Report

The authors present a study aimed at evaluating the pro & cons of immersive VR interfaces on online social applications.

To achieve their goal, first of all they developed an Immersive Virtual Reality Social Application (IVRSA) enabling virtual interactions based on voice and body language. Later, they exploited this immersive environment to carry out a heterogeneous user study specifically designed to assess the advantages and disadvantages of IVRSA against traditional social interfaces.

Differently from the first version of the manuscript, the authors decided to narrow the scope of their research, focusing the paper on the evaluation of the developed application, rather than on the implementation of the same.

This choice results to be appropriate (since it gives to the work a more defined structure), but seem to be poorly reflected in the paper content, still characterized by the old setup (just note the space taken by the details of the IVRSA implementation, that is still a relevant part of the manuscript, or the related work, still addressed to describe the STAR of IVRSA implementations).

Another issue of this new approach is that, since now the evaluation stage (sections 4 & 5) became central in the paper ecosystem, it should be perfect. Instead, in my opinion, this part, although improved compared to the first submission, has still some problem (mostly it lacks an in-depth discussion of the results).

However, generally speaking, the paper can be considered better than the previous version.

Nevertheless, the points above specified (and also a good number of grammatical and stylistic issues, probably introduced together with the recent additions), make me lean to confirm my minor revision response, since in my opinion, the paper would surely benefit from any action addressed to solve the issues above mentioned.

Author Response

Dear reviewer:

 

Thank you for your letter and the reviewers’ comments on our manuscript entitled "The Influence of Immersive Virtual Reality Systems on Online Social Application". Thank you for your optimistic evaluation of our paper. At the same time, you also put forward some Suggestions on the modification of the conclusion and some details. We have made the following modifications according to your comments:

 

We have made many modifications to the conclusion:

1. Put forward the existing problems of the system and point out the direction of improvement.
2. We made a statement: The analysis of the user's age and gender in this paper is based on the data derived from the experiment, which can bind the user's age and gender information. We think this data can provide reference for this topic. This part of the analysis is only based on the experimental data in the paper, and the analysis results are not comprehensive and rigorous.
3. The evaluation of IVRSA gives a comprehensive conclusion.

 

We have supplemented our experimental hypotheses after Tables 1 and 2.

 

To distinguish between experimental system and term “IVRSA”, we renamed the IVRSA system developed in the experiment "RVM-IVRSA"."RVM" refers to the "roaming, voice communication, and motion capture" mode of function used by the system.

 

In addition, we have corrected some of the word mistakes.

 

Please refer to the manuscript for details.

 

I hope we can get your approval for the improvement of the paper this time.

 

Kind regards,

 

Zuohao Yan

E-mail: [email protected]

 

Corresponding author: Zhihan Lv

E-mail address: [email protected]

Reviewer 2 Report

I think it faithfully reflects the revisions you requested.

Thank you.

Author Response

Dear reviewer:

 

Thank you for your letter and the reviewers’ comments on our manuscript entitled "The Influence of Immersive Virtual Reality Systems on Online Social Application". Those comments are very helpful for revising and improving our paper, as well as the important guiding significance to other research. We would like to thank you for your opinion that our paper is ready for publication. But the other two reviewers suggest that this article needs to be refined for the conclusion part and some further details. So we made further modifications to the paper.

We have made many modifications to the conclusion:

1. Put forward the existing problems of the system and point out the direction of improvement.
2. We made a statement: The analysis of the user's age and gender in this paper is based on the data derived from the experiment, which can bind the user's age and gender information. We think this data can provide reference for this topic. This part of the analysis is only based on the experimental data in the paper, and the analysis results are not comprehensive and rigorous.
3. The evaluation of IVRSA gives a comprehensive conclusion.

To distinguish between experimental system and term “IVRSA”, we renamed the IVRSA system developed in the experiment "RVM-IVRSA"."RVM" refers to the "roaming, voice communication, and motion capture" mode of function used by the system.

In addition, we have corrected some of the word mistakes.

Please refer to the manuscript for details.

I hope we can get your approval for the improvement of the paper this time.

 

Kind regards,

 

Zuohao Yan

E-mail: [email protected]

 

Corresponding author: Zhihan Lv

E-mail address: [email protected]

Reviewer 3 Report

The article presents a inmersive virtual reality platform ICRSA, and a later study on the ability to recognize gestures (body language), differentiating by age and sex.

Reading the article is pleasant and well focused. But some corrections are necessary:
- Line 81. "... virtual social platform USES a more" USES must be in lowercase.
- Line 84. "...engaged [11]. development ..." should be "...engaged [11]. Development ..."
- Line 100. USES is in uppercase again.
- Line 139 appears Unity, in line 172 is referenced as unity3D. Should be referenced with the same name, normally is referenced as Unity3D.
- Line 174. "The 'Controller(left)' and 'Controller(left)' under ..." Why appears Left Controller twice? Should it be left and right?
- Line 214 THREE DIMENSIONAL in uppercase.

In section 3.6 the error is poorly explained. This is not the estimated error, but the maximum error. If you have the maximum error it is because you have the exact value, then you would not have an error (Line 282, formula (2)). It should be noted that this error can be estimated by assuming that the function is continuous and infinitely differentiable.

Box diagram could be presented for Figure 7 (time response) and Figure 8  for each gesture (time response). Only mean value is not relevant.

The conclusions have not been demonstrated with the experimentation carried out, limiting itself to showing statistical data such as the mean. In the case of the satisfaction questionnaire and social experimentation, the p-value is determined but it is not indicated what hypothesis testing has been carried out.

To demonstrate the conclusions, it is necessary to indicate which statistical hypothesis contrast has been carried out, and in the case of the conclusions on age and sex, it is necessary to carry out an ANOVA test. Since the differences may be due to the variability of the data. In any case, an ANOVA test with only 10 samples in each group is not the best option.

The results do not reach the conclusions, a better in-depth experimentation should be done, and the statistical analysis carried out. Only data are presented, in some cases only the mean is presented with the intention of drawing conclusions without any scientific rigor.

It would be necessary to make the results of the experimentation accessible so that the scientific community can verify the statistical results obtained.

Author Response

Dear reviewer:

 

Thank you for your letter and the reviewers’ comments on our manuscript entitled "The Influence of Immersive Virtual Reality Systems on Online Social Application". Those comments are very helpful for revising and improving our paper, as well as the important guiding significance to other research. The main corrections are in the manuscript and the responds to the reviewers’ comments are as follows (the replies are highlighted in blue).

 

1.Some corrections are necessary:

- Line 81. "... virtual social platform USES a more" USES must be in lowercase.

- Line 84. "...engaged [11]. development ..." should be "...engaged [11]. Development ..."

- Line 100. USES is in uppercase again.

- Line 139 appears Unity, in line 172 is referenced as unity3D. Should be referenced with the same name, normally is referenced as Unity3D.

 

We have made corrections as you requested.

 

2.- Line 174. "The 'Controller(left)' and 'Controller(right)' under ..." Why appears Left Controller twice? Should it be left and right?

 

In the SteamVR development in Unity3D, there are two child objects under the object "Camera", namely "Controller(left)" and "Controller(right)", which correspond to two controllers of HTC VIVE. I'm emphasizing the same operation on both objects.

 

3.In section 3.6 the error is poorly explained. This is not the estimated error, but the maximum error. If you have the maximum error it is because you have the exact value, then you would not have an error (Line 282, formula (2)). It should be noted that this error can be estimated by assuming that the function is continuous and infinitely differentiable.

 

We refer to this error in the paper as "computational error".

 

4.Box diagram could be presented for Figure 7 (time response) and Figure 8 for each gesture (time response). Only mean value is not relevant.

 

We have considered your question when making Figure 7 and Figure 8, but finally we chose the current form for two reasons:

1.First of all, we have tried to use side-by-side histogram, using two different colors for average time and accuracy. But this form is so bad that it is difficult for the reader to distinguish the vertical axes corresponding to the two sets of histograms, so we abandon this scheme. Later, we tried to display the two images separately, but we found that the display method was not as good as we expected. We want to put the average time and accuracy in a single graph, so that readers can sum up what actions are good and what actions are bad after considering both the average time and the accuracy. Therefore, we finally choose the method of using histogram for accuracy data and line graph for average time.

2.We inquired the use mode of line graph and came to the conclusion that the line graph can be used to clearly show the level of each group in grouping data comparison. And in Figure 8, we want to show the general trend of the data in the form of line graph, that is, the older the group is, the more time it takes to recognize the action.

 

5.The conclusions have not been demonstrated with the experimentation carried out, limiting itself to showing statistical data such as the mean. In the case of the satisfaction questionnaire and social experimentation, the p-value is determined but it is not indicated what hypothesis testing has been carried out.

 

We have supplemented our experimental hypotheses after Tables 1 and 2.

 

6.The results do not reach the conclusions, a better in-depth experimentation should be done, and the statistical analysis carried out. Only data are presented, in some cases only the mean is presented with the intention of drawing conclusions without any scientific rigor.

In addition, you also mentioned some problems in technical elaboration and details of the article, and we have also made modifications. Thank you very much for your Suggestions, which are very helpful to our work.

 

We have made many modifications to the conclusion:

1.Put forward the existing problems of the system and point out the direction of improvement.

2.We made a statement: The analysis of the user's age and gender in this paper is based on the data derived from the experiment, which can bind the user's age and gender information. We think this data can provide reference for this topic. This part of the analysis is only based on the experimental data in the paper, and the analysis results are not comprehensive and rigorous.

3.The evaluation of IVRSA gives a comprehensive conclusion.

Please refer to the manuscript for details.

 

7.It would be necessary to make the results of the experimentation accessible so that the scientific community can verify the statistical results obtained.

 

We do not think it is necessary to add this part of data to the paper, because there are so many data in this part that it would be superfluous to add it into the paper. This part of the data is complex, it is difficult to sort out. We have the right not to publish this part of data, please understand us, this is a little selfish and lazy decision.

 

Kind regards,

 

Zuohao Yan

E-mail: [email protected]

 

Corresponding author: Zhihan Lv

E-mail address: [email protected]

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

The manuscript “The Influence of Immersive Virtual Reality Systems on Online Social Application” analyses the influence of virtual reality (VR) on online social interaction and the advantages and disadvantages on the use of immersive reality (IR).

This is a relevant research topic. The authors present a good rationale for the relevance of the paper. Despite this, the manuscript has many issues and flaws, which makes the manuscript not ready for publication.

The main problems are:

1. The document needs a thorough English revision. There are too many grammatical issues (lack of definite/indefinite articles, missing words, misleading terms, poorly constructed sentences etc.) and lack of periods. This makes reading flow very difficult and it is difficult to understand the meaning of some of the sentences;
2. It is difficult to understand what the participants really assessed. The experimental design needs to be much better explained. Authors must provide, along with the manuscript, demonstration videos with examples of the social interaction scenarios. The manuscript should focus much more on the experiments (currently described on just half a page), rather than skipping immediately to the results. The underlying framework, image flowing, quality of scenarios/ graphics, social scenes, avatars, they are all very important and must be clearly presented. Overall, the authors should improve the description of the methodology;
3. One question remains: is the manuscript about the overall implementation of the VR / IR framework or about the evaluation of VR / IR for social applications? None of these parts is well described in the manuscript. If the objective is the assessment of VR / IR in social applications, why did the authors not use a well-established platform? If the objective is in the development of the platform, then the technical description should be improved.

Specific comments

• The related work section should refer to works assessing VR/IR for social applications (that are close to the author’s manuscript);
• Explain better what you mean with the sentence “Unity USES multiple lenses to observe virtual scenes to achieve 3d visual programming of the game, which is convenient for developers to observe and check the development content in real time”;
• When referring to WIMP why authors used the expression “fixed”-point devices instead of “pointer” (or pointing devices)
• The architecture presented in section 3.1. should be much better explained (very incomplete). Figure 1 is very vague and misleading, as it tries to redirect to a TCP/IP architecture. The model is certainly not supported by reference [23] as stated by the authors. The blocks “TCP + command” are meaningless;
• The way the authors explain the LED and Laser may be misleading. On the other hand, photosensors are not mentioned. The title of the subsection could be changed to “Motion tracking”, or something similar. Figure 2 is not explained in the text or contextualized;
• In section 3.1.2 is it “Wave” or “wav” file?
• There are technical explanations that could be eliminated as they refer to very well established and basic technology. For example, it makes no sense to explain the TCP/IP flow (Figure 3) as the authors are not adding anything new;
• Explain how Formula (2) was obtained, or refer to a supporting document;
• In the description of the position estimation, explain what is being done differently from other works implementing position estimation in networked games;
• In section 4.1, what is “Experimental Volunteer”?
• As already referred, section 4.2 should be greatly expanded;
• What authors mean with “During each communication, if one of the volunteers fails to communicate normally, he/she is allowed to interrupt the communication at any time and record the duration of the activity”;
• Describe the experimental conditions better, for example, how many sessions were held, how long a session lasted, etc
• How was gesture recognition implemented?;
• How did you conclude that the discomfort was caused by vertigo? From my experience with HTC Vive and Oculus Rift, vertigo and nausea are no longer a problem (at least in loco), unless of course you are placed in scenarios that cause vertigo (top of buildings, etc). Can we conclude that the vertigo effect is caused by your particular framework/platform related to network and position estimation issues?

As a final note, the overall methodology should be greatly improved. The paper focuses on too many topics, so maybe authors could focus on fewer topics but in more detail.

Author Response

Dear reviewers:

 

Thank you for your letter and the reviewers’ comments on our manuscript entitled "The Influence of Immersive Virtual Reality Systems on Online Social Application". Those comments are very helpful for revising and improving our paper, as well as the important guiding significance to other research. We have studied the comments carefully and made corrections which we hope meet with approval. The main corrections are in the manuscript and the responds to the reviewers’ comments are as follows (the replies are highlighted in blue).

 

1.The document needs a thorough English revision. There are too many grammatical issues (lack of definite/indefinite articles, missing words, misleading terms, poorly constructed sentences etc.) and lack of periods. This makes reading flow very difficult and it is difficult to understand the meaning of some of the sentences;

 

I modified some English grammar problems, and optimized the definite/indefinite articles and sentence structure. I also sorted out the paragraph structure and increased the use of periods in the hope that the new version will make it easier for readers to read.

 

2.It is difficult to understand what the participants really assessed. The experimental design needs to be much better explained. Authors must provide, along with the manuscript, demonstration videos with examples of the social interaction scenarios. The manuscript should focus much more on the experiments (currently described on just half a page), rather than skipping immediately to the results. The underlying framework, image flowing, quality of scenarios/ graphics, social scenes, avatars, they are all very important and must be clearly presented. Overall, the authors should improve the description of the methodology;

 

Studies the main content is the part of experimental data analysis, because of the lack of appropriate experimental platform, we built a simple immersive virtual reality social applications (IVRSA), because of the limited functionality of the application, our experiment at present only in immersive virtual reality scene use effect evaluation of speech communication and body language. Your feedback is very valuable, the structure of this paper really needs to be improved, we should discuss some experimental details more.

 

3.One question remains: is the manuscript about the overall implementation of the VR / IR framework or about the evaluation of VR / IR for social applications? None of these parts is well described in the manuscript. If the objective is the assessment of VR / IR in social applications, why did the authors not use a well-established platform? If the objective is in the development of the platform, then the technical description should be improved.

 

There is currently a lack of suitable applications on the market as experimental platforms. More mature immersive virtual reality social platforms have more complex functions, and their social functions are realized based on the joint function of multiple information channels. If we make a comprehensive evaluation of all IVRSA, the content of the experiment will be very complex, and the experiment needs a lot of financial support. Therefore, we decided to focus on the most basic speech and body language functions of this kind of social communication for analysis and evaluation, and also independently developed an IVRSA experimental platform with these two functions. This point is not explicitly mentioned in the paper, which is our negligence. This point has been clarified in the revised manuscript.

 

In addition, you also mentioned some problems in technical elaboration and details of the article, and we have also made modifications. Thank you very much for your Suggestions, which are very helpful to our work.

 

Figure 1 of the paper introduces the system used in the experiment from three aspects, so I don't call it architecture anymore in the revised manuscript, but explained this aspect. I added some explain about the light-sensitive sensors in the lighthouse, and change the title of this section to "Motion tracking". We added an explanation to figure 2. There is no mistake about calling a Wave file, ".wav "is an extension of a Wave file. We removed the introduction of TCP/IP technology. The reference of formula (2) is the reference of DR algorithm mentioned above. We extend the content of section 4.2 (experimental design). As for the causes of vertigo, this part is the hypothesis put forward by the researchers in combination with the subjective feedback of volunteers, which we have emphasized now. There are a lot of topics involved in the paper. The revised manuscript focuses on the research of "voice and body language expression effect". The rest of the topics are not deleted, but we reduce their length to highlight the main topic of the article. 

 

Kind regards,

 

Zuohao Yan

E-mail: [email protected]

 

Corresponding author: Zhihan Lv

E-mail address: [email protected]

 

Reviewer 2 Report

The paper presents a study aimed at evaluating the exploitation of immersive user interfaces in virtual reality socializing. For achieve their goals, the authors first developed a virtual environment (built over existing solutions) and then tested it through a heterogeneous user study.

The work, presented with good clarity of style, is interesting and deals with a trending topic.

Nevertheless, the significance of its content is not a strength. In particular, results and conclusions are quite predictable, lacks in originality, and does not offer food for discussion.

For instance, the paper misses the opportunity to contextualize the implications of a study like this in the current panorama, characterized by forced virtual interactions due to the worldwide lockdown. Or again, it does not provide any food for thought regarding the well-know issue of virtual reality sickness (such a significant problem that alone would be able to invalidate studies like this).

However, since the chosen topic is relatively unexplored, and the methodology followed by the study is consistent, the overall evaluation of the paper is still positive. But, in my opinion, the paper would surely benefit from any action addressed to solve the issues above mentioned. A minor revision in this sense would be welcome.

Author Response

Dear reviewers:

 

Thank you for your letter and the reviewers’ comments on our manuscript entitled "The Influence of Immersive Virtual Reality Systems on Online Social Application". Those comments are very helpful for revising and improving our paper, as well as the important guiding significance to other research. We have studied the comments carefully and made corrections which we hope meet with approval. The main corrections are in the manuscript and the responds to the reviewers’ comments are as follows (the replies are highlighted in blue).

 

Immersive virtual reality socializing is a nascent industry. Although our research method is not very original, due to the novelty of the research topic and content, there are few studies on the combination of immersive virtual reality technology and social effects in the world. Therefore, we think this research is quite meaningful. The results of the study are indeed similar to our predictions, but they prove that it is possible to predict some of the conditions of virtual reality based on the experience of face-to-face communication in the real world.

We mentioned some of the physiological discomforts (dizziness) caused by virtual reality technology, but these were not the main contents of the study.If we make a comprehensive evaluation of all IVRSA, the content of the experiment will be very complex, and the experiment needs a lot of financial support. Therefore, we decided to focus on the most basic speech and body language functions of this kind of social communication for analysis and evaluation, and also independently developed an IVRSA experimental platform with these two functions. This point is not explicitly mentioned in the paper, which is our negligence. This point has been clarified in the revised manuscript.

Your Suggestions have indeed had a very beneficial impact on the improvement of our thesis. Thank you very much for your help.

 

Kind regards,

 

Zuohao Yan

E-mail: [email protected]

 

Corresponding author: Zhihan Lv

E-mail address: [email protected]

Reviewer 3 Report

This study analyzed communication between users for social interaction in immersive virtual reality. To this end, IVRSA was developed and a survey experiment was conducted for participants.
I think that it deals with important issues not only about the technical approach in immersive virtual reality, but also user-centered experience and satisfaction evaluation.

However, there are some shortcomings and questions in this paper.

1. What can users do specifically through the developed IVRSA?
Only the development process of IVRSA is described, but it is not understood what social interaction the user can make in the application.
Is there only a function to communicate by gesture and voice?

 

2. In VR, there is a verified questionnaire that evaluates user experience (GEQ-social interaction), presence (PQ), and satisfaction with the interface (USE). However, I was wondering why the questionnaire was made by making questions directly with about 5 short sentences without using such questionnaires.

The questionnaire question is a factor that determines the reliability of the survey experiment, but this study seems to have no verification or concern about the questionnaire question.

 

3. Figures showing the motions used in the body language may help to improve understanding.
There are many survey experiments on the body language, but it seems to require additional explanation on what kind of motion it is taking.

And, there is a voice communication function, and I hope that there is enough explanation why the emphasis is placed on experimenting with body language.
I also feel that there is a need to compare social interactions in voice communication with body language.

 

4. Is body language the core of IVRSA's interaction?

I wonder if the social interaction presented in this study is the key to body communication.
If so, how the communication using the body actually proceeds should be expressed in detail in figures.

Author Response

Dear reviewers:

 

Thank you for your letter and the reviewers’ comments on our manuscript entitled "The Influence of Immersive Virtual Reality Systems on Online Social Application". Those comments are very helpful for revising and improving our paper, as well as the important guiding significance to other research. We have studied the comments carefully and made corrections which we hope meet with approval. The main corrections are in the manuscript and the responds to the reviewers’ comments are as follows (the replies are highlighted in blue).

 

1. What can users do specifically through the developed IVRSA?
Only the development process of IVRSA is described, but it is not understood what social interaction the user can make in the application.
Is there only a function to communicate by gesture and voice?

IVRSA is a wide range of term, applications of IVRSA type on market is not suitable for use as experimental platform, so we developed a only used to study the real-time voice and body language to express the experimental platform, the purpose is to study the user in a social situation in the immersive virtual reality scene, we believe that the real time voice and body language to express are main differences between IVRSA and traditional social software, therefore the platform only these two main function (it also has some basic functions of virtual reality systems, such as virtual roaming).

 

2. In VR, there is a verified questionnaire that evaluates user experience (GEQ-social interaction), presence (PQ), and satisfaction with the interface (USE). However, I was wondering why the questionnaire was made by making questions directly with about 5 short sentences without using such questionnaires.

The questionnaire question is a factor that determines the reliability of the survey experiment, but this study seems to have no verification or concern about the questionnaire question.

There are 10 questions in the questionnaire, the first 5 questions we hope to enable users to have a comment on the experimental platform and the interactive way it is used. The high stability is the foundation of the experimental platform, and also shows that IVRSA application development can be achieved in terms of stability compared with traditional social software. And these problems we use rather than the cause of the phrase is to give participants a guide, if you read the question 1, 3 and 4, you'll find that our participants were asked to evaluate this interaction mode, and question 2 and question 5 is focused on the performance of the system. The remaining five questions are all focused on user's social experience, with questions being more specific than phrases, and user's feedback being more detailed and targeted.

 

3. Figures showing the motions used in the body language may help to improve understanding.
There are many survey experiments on the body language, but it seems to require additional explanation on what kind of motion it is taking.

And, there is a voice communication function, and I hope that there is enough explanation why the emphasis is placed on experimenting with body language.
I also feel that there is a need to compare social interactions in voice communication with body language.

The first part of the experiment is the real-time free communication of the users. We ask the participants to use the real-time voice and body language as much as possible. We think it is not rigorous to test the function effect of real-time voice alone, because the function of real-time voice alone is not much different from that of making a phone call. The unique feature of IVRSA's real-time voice function lies in the simulation of face-to-face social contact with virtual scenes, so we did not design a separate experimental test of real-time voice.

In addition, our paper really lacks a detailed explanation of the movements in the body language test. Your opinion is very helpful to us, and we have made supplements in the revised manuscript.

 

4. Is body language the core of IVRSA's interaction?

I wonder if the social interaction presented in this study is the key to body communication.
If so, how the communication using the body actually proceeds should be expressed in detail in figures.

When we designed this study, we wanted to find the differences between VR online social networking and traditional online social networking, and we believe that real-time voice and body language communication are the main differences between them. Of course, human-scene interaction, person-to-person interaction, and so on are also important differences, but in the end, considering the limited objective conditions of our research, we decided to focus on real-time speech and body speech.

 

Your Suggestions have indeed had a very beneficial impact on the improvement of our thesis. Thank you very much for your help.

 

Kind regards,

 

Zuohao Yan

E-mail: [email protected]

 

Corresponding author: Zhihan Lv

E-mail address: [email protected]

 

Reviewer 4 Report

Authors have developed a Virtual Reality Application to evaluate the influence.
Unity is the engine used to develop the application. But, the experimentation and conclusions are not statistical correct.
First, Unity is described as tool that supports Javascript and Python. That is not correct. Javascript and Python are available to develop tools, not to deploy final application.

All the text appears the word USES in uppercase.

In page 7, third equation, is wrongly typped, should be A[t] = V[t]-V[t-a]/delta time. 

Kalman filter is used to predict the next position. To be able to use Kalman Filter it is mandatory to assume a linear system. This assumption must appear in the first paragraph of the section to be valid the rest of the formulation.

Section 4. Experimental Method must be reformulated:

• Only 20 samples are used to compare, the number of samples must be greater than 30. 
• The persons that carry out the experimentation are voluntareers. How are they choosen?
• The 20 samples are divided in four groups by age. But the number of samples is not representative, so the number of samples in each group must be greater (30 or more) to be able to extract conclusions about the difference between groups.

Section 5. Result.

• The results are not well explained, only is provided a p-value.
• The hypothesis test performed is not described, or if it was considered as paired data.
• Source data from which the statistical conclusions were drawn should be available to verify that the process is correct.
• Section 5.3 should show results with a confussion matrix, including 39 kinds of actions.
• Section 5.4 is not valid. The number of samples for each group is small, therefore the proposed conclusions cannot be reached. In addition, it could be derived from user skills and not a consequence of age.
• Section 5.5 Conclusions are detailes as a "suggestion". But again the number of samples is not significant. So, conclusions are not statistically correct. 

The conclusions must be supported by correct statistical analysis, the number of samples must be increased to carry out the differentiation between groups. And hypothesis contrast carried out must be detailed.

Author Response

Dear reviewers:

 

Thank you for your letter and the reviewers’ comments on our manuscript entitled "The Influence of Immersive Virtual Reality Systems on Online Social Application". Those comments are very helpful for revising and improving our paper, as well as the important guiding significance to other research. We have studied the comments carefully and made corrections which we hope meet with approval. The main corrections are in the manuscript and the responds to the reviewers’ comments are as follows (the replies are highlighted in blue).

 

First, Unity is described as tool that supports Javascript and Python. That is not correct. Javascript and Python are available to develop tools, not to deploy final application.

All the text appears the word USES in uppercase.

We have modified this part of the description so that it does not cause misunderstanding.

 

In page 7, third equation, is wrongly typed, should be A[t] = V[t]-V[t-a]/delta time. 

Kalman filter is used to predict the next position. To be able to use Kalman Filter it is mandatory to assume a linear system. This assumption must appear in the first paragraph of the section to be valid the rest of the formulation.

 

We have modified the formula according to your suggestion. Regarding the assumption of linear system, we have made a supplement in the paper, which is described in the Kalman filter section. What we predicted is the head position and velocity.

 

Section 4. Experimental Method must be reformulated:

Only 20 samples are used to compare, the number of samples must be greater than 30. 

The persons that carry out the experimentation are volunteers. How are they choose?

The 20 samples are divided in four groups by age. But the number of samples is not representative, so the number of samples in each group must be greater (30 or more) to be able to extract conclusions about the difference between groups.

We have done a lot of work on your modification suggestions. We temporarily conducted five more experiments, supplemented with data from 10 volunteers, bringing the number of subjects to 30.

First of all, I would like to explain why we have a small number of volunteers. There is no material reward for the volunteers in this experiment. In the test, each volunteer needs to come to laboratory in pairs for 15 different periods, and each volunteer needs to spend at least 2 hours to carry out the test. Therefore, the conditions of this experiment are relatively strict and it is not easy to recruit a large number of volunteers. In fact, after receiving your modification suggestions, we temporarily added the test data of 10 volunteers. Due to the tight schedule and the influence of covid-19, these 10 volunteers are all my family members and close friends. They didn't have the expertise about this experiment, so they had the same qualifications as the previous 20 volunteers. This is the maximum amount of data we can supplement. I hope you can understand our difficulties.

 

Section 5. Result.

The results are not well explained, only is provided a p-value.

The hypothesis test performed is not described, or if it was considered as paired data.

Source data from which the statistical conclusions were drawn should be available to verify that the process is correct.

Section 5.3 should show results with a confusion matrix, including 39 kinds of actions.

Section 5.4 is not valid. The number of samples for each group is small, therefore the proposed conclusions cannot be reached. In addition, it could be derived from user skills and not a consequence of age.

Section 5.5 Conclusions are details as a "suggestion". But again the number of samples is not significant. So, conclusions are not statistically correct. 

The conclusions must be supported by correct statistical analysis, the number of samples must be increased to carry out the differentiation between groups. And hypothesis contrast carried out must be detailed.

In this part, our main purpose is to compare IVRSA with traditional social software. Therefore, we hope to use the subjective evaluation of 30 volunteers and the P-value to illustrate which aspects IVRSA has advantages or disadvantages and which aspects are quite different from traditional social software. Hypothesis test was supplemented in the paper.

The 39 actions in section 5.3 reached 45 after the supplementary data. We don't want to enumerate every action in these 45 actions, because the test data for each action is not equal and cannot be compared at same level with the previous 5 preset actions. However, this part of the test results have a high degree of randomness, which can best illustrate real use situation. So we gave these data (75 test data) a label "Optional" and recorded it in the table as a reference. By analyzing it alone, we can better understand real use level of body language in this social mode. Combining with preset actions data, we can see impact of various movements on recognition rate. About this reason, we have supplemented the explanation in the paper.

As for the two sections of 5.4 and 5.5, because the total number of people increased, we also made data supplement. In 5.4, the data in section 5.3 were grouped according to age, which was originally divided into 4 groups and 4 people were randomly selected from each group (in order to ensure the equal amount of data in each group), now we have changed to 6 people from each group. In 5.5, the data in section 5.3 are grouped and compared by gender, it was originally divided into 2 groups and 7 people were selected for each group, but now we change to 10 people for each group.

I would like to explain some of the data and analysis in this part. Our analysis conclusion is that we found the influence of age and gender on the results of body language recognition in the experiment through grouping, but we do not deny the influence of other factors (such as occupation, culture, etc.) on the results. It is difficult to study a single factor in online social interaction. Second, about the amount of data in the experiment, the 6 people in each group or the 10 people in each group is not the amount of data, but the number of volunteers. These two sections talk about action recognition, so each person has to do five tests. The amount of data in this part does not seem small to us.

 

Your Suggestions have indeed had a very beneficial impact on the improvement of our thesis. Thank you very much for your help.

 

Kind regards,

 

Zuohao Yan

E-mail: [email protected]

 

Corresponding author: Zhihan Lv

E-mail address: [email protected]

Round 2

Reviewer 1 Report

Unfortunately, the authors did not address the raised issues and the paper is substantially the same as the first version. Moreover, the authors did not present an answer or rebuttal to each question/comment that was asked. In conclusion, the manuscript continues with the same methodological, technical and content flaws.

Reviewer 4 Report

The article lacks scientific rigor. The p-value is indicated without specifying what type of test has been performed.
I propose to detail in depth what type of statistical study has been carried out.

Otherwise it cannot be verified as correct.