Sex Differences in Spatial Activity and Anxiety Levels in the COVID-19 Pandemic from Evolutionary Perspective

Round 1

Reviewer 1 Report

The manuscript presents and interesting study that links participants spatial activity and mobility with attitudes and emotional response to covid19. Despite its novelty, some major issues are presents that should be acknowledged in the discussion and modified in the introduction, methods and results sections.

1. Authors used a single item measure of "anxiety toward covid19". Authors should report in the discussion the limits of measuring anxiety with a non-validated single item measure. Moreover, they should present clearly in the method how they define anxiety toward covid19: anxiety may be seen as an emotion, as a clinical syndrome, a neurophysiological reaction; I suspect that here, this single item could catch the dimension of worry, that is a partial aspect of what anxiety is.
2. Authors stated that "The study took place from 29 March 2020 to 27 June 2020 (Median 12-APR-2020), during the 145 virus outbreak in Russia accompanied by an obligatory strict quarantine regime". If a strict quarantine regime was present, why did participants were allowed to move outside their homes? Spatial activity during a lockdown may be influenced also by the attitude toward restrictions and fear of punishments by authorities. Authors didn't include this variable in the research design, and the focus on "health attitudes" and "anxiety", despite its relevance, didn't catch an important factor that may explain behaviour.
3. Sample descriptive statistics are presented in the legend of table 1. I would suggest to present descriptive statistics in a dedicated table (that should be table 1) before correlations, and if sex differences are one of the study focus, descriptives may compare frequencies and mean scores for male vs female and present overall score as well.
4. If I understood correctly, participants are located in Russia. It would be interesting to know something about their geographical distribution (e.g. do they live in the same region? in small or large urban areas? etc...). 
5. Figure 5 would be more appropriate for the study scope if it would presents male and female distribution of score instead of overall distribution.
6. I would suggest to submit the manuscript to a professional proofreading service to check typos and grammar.

Author Response

Authors used a single item measure of "anxiety toward covid19". Authors should report in the discussion the limits of measuring anxiety with a non-validated single item measure. Moreover, they should present clearly in the method how they define anxiety toward covid19: anxiety may be seen as an emotion, as a clinical syndrome, a neurophysiological reaction; I suspect that here, this single item could catch the dimension of worry, that is a partial aspect of what anxiety is.

1. • Thank you for the comments, we agreed with this statement and devoted to this critical drawback a certain part of the limitation section (pages 15-16)

Authors stated that "The study took place from 29 March 2020 to 27 June 2020 (Median 12-APR-2020), during the 145 virus outbreak in Russia accompanied by an obligatory strict quarantine regime". If a strict quarantine regime was present, why did participants were allowed to move outside their homes? Spatial activity during a lockdown may be influenced also by the attitude toward restrictions and fear of punishments by authorities. Authors didn't include this variable in the research design, and the focus on "health attitudes" and "anxiety", despite its relevance, didn't catch an important factor that may explain behaviour.

• • Thank you for this comment we mentioned this issue in the limitation section. And we really think that it is the potential item we missed to introduced into our survey

Sample descriptive statistics are presented in the legend of table 1. I would suggest to present descriptive statistics in a dedicated table (that should be table 1) before correlations, and if sex differences are one of the study focus, descriptives may compare frequencies and mean scores for male vs female and present overall score as well.

• • We agreed with the review and added the additional tables ( page 5)

If I understood correctly, participants are located in Russia. It would be interesting to know something about their geographical distribution (e.g. do they live in the same region? in small or large urban areas? etc...). 

• • The geographical distribution was quite broad, we included this information into methods section (lines 137-139)

Figure 5 would be more appropriate for the study scope if it would presents male and female distribution of score instead of overall distribution.

• • Accepted (page 10)

I would suggest to submit the manuscript to a professional proofreading service to check typos and grammar.      

• • We suppose to submit this paper after the procedure of improving the meaning

General information about changes in the manuscript

Authors fully agreed with the reviewers regarding the importance of better defining what anxiety was in our first manuscript; therefore, we have changed the variable name of “Anxiety” to the “level of worry.” It is a single item variable, and it mostly catches the dimension of worry, which is only a partial aspect of anxiety.

To explain the relevance of the problem, In the Abstract was added this sentence:

Despite the enforcing lockdown regime in late March 2019 in Russia, the phenomenon of the continued virus spreading highlighted the importance of studies investigating the range of biosocial attributes and spectrum of individual motivations underlying the permanent presence of the substantial level of spatial activity.

In the Abstract, the sentence at line 17 was rephrased into:

However, our data suggested that wariness largely determines adequate health attitudes; hence a higher level of warry indirectly reduced individual mobility.

 In the Introduction section, the first-to-third paragraph’s sentences were changed in their orders.

Also, lines 58-63 (see previous manuscript) were removed:

Over the past few decades, many studies have analyzed the spatial-distance aspects of human activity as a reflection of its socio-demographic characteristics [14–18]. Data from cell phone-based accelerometers were used for building predictive model for activity recognition, spatial movements and activity patterns; for instance, the most likely distances and directions of the people's daily mobility have been profoundly investigated [19–22]. Demographic attributes and socio-economic parameters are estimated and implemented in big data analyses of human spatial behavior [23].

In the Method section wee added information about the inclusion criterion(page 3, line 135-142)

Also, we expanded the statistical analyses methods procedure information (page 4, lines 179-189)

We added into Introduction section a descriptive statistics tables for males and females.

In the results section, all the diagram and model contain the variable “Worry” instead of "Anxiety”

In the Results section, we slightly reduced; in particular, this paragraph was sufficiently reduced:

Given that the full group model may mask the effects specific to each sex, the multiple group model has been implemented. We performed a multi-sample SEMs unconstrained structural models, where structural relationships are specified in all groups meaning that the model is equal across groups. Still, the coefficients in the relationships were estimated independently for each group. The results (Figures 3 and 4) demonstrated that each of the group paths estimates were in the expected direction and indicated many similarities in structural relationships with the full model presented in Table 2 and Figure 1. All paths were significant for each group except the family size (the number of inhabitants), which was not significant for females’ level of worry.

The Chi-Squared difference test between constrained and unconstrained models demonstrated a high significance p (χ2) < 0.00305. Hence, the model fit index changes were significant, indicating that metric invariance was not established, and the path estimates for males and females were different. This means that the unconstrained multiple group model should be accepted as more representative SEMs. A detected non-invariance in the model intercepts and partly non-significant pathways represent the evidence in favor of systematic sex-specific differences in the model parameters being examined. Due to this fact, in the next step, we performed two SEM’s separately for each sex. Figures 3 and 4 demonstrate the path diagrams for males’ (panel A) and for females’ sample (Panel B). (The above paragraph was rephrased into three sentences ( see lines 253-258).

In the results, figure 5 has been replaced by the same histogram data but splinted between two sexes.

In the Results section, the Paragraph under figure 6 was reduced, as this information presented in the Discussion section:

From an evolutionary perspective, the human pair-bonding has been an adaptation for offspring survival given the long period of their dependence. Under these circumstances, the sexual division of labor and the need for paternal investment were of key importance [26,59–61]. According to the parental investment theory, human males have been rarely providing direct care for their offspring. However, they are essential contributors to the family budget and mostly provided the economic sustainability of the family and offspring well-being and survival in general [62,63]. The indirect care, by definition, has a strong association with higher social network and higher spatial mobility of the male contributors. Our data suggests that females were less prone to get in contact with the dangerous environment and preferred to delegate this function to their partners.

In the Discussion section, paragraphs, 3,4,5 were reorganized to avoid repeated ideas and maintain general consistency.

The section of Limitation was added (page 15-16)

Author Response File: Author Response.pdf

Reviewer 2 Report

Comments:

-Please specify why the study focused on people who are about 35 years old? Please specify the age-range of participants included in the study.

-Please specify if the present data and questionnaire are available to other researchers. If so, please specify the repository to accomplish access to data.

-Did the cultural aspect played any role in the predisposition for mobility under adverse conditions? Please comment on this.

Please comment on the relevance of the present findings with regards to understanding the impact of stress related factors into models of covid spreading (see “Stress as a Meaningful Variable in Models of Covid-19 Spreading.” PsyArXiv. December 6. doi:10.31234/osf.io/kcpqm)

Author Response

Please specify why the study focused on people who are about 35 years old? Please specify the age-range of participants included in the study.

• Thank you for the comment. We introduced 2 tables with descriptive statistics ( page 5)

Please specify if the present data and questionnaire are available to other researchers. If so, please specify the repository to accomplish access to data.

• We are currently in the process of more profound analyses of the spatial activity level, and we are planning to investigate the exact destinations and individual aims in our participant's activity. We will focus on the singleton versus family persons to test the loneliness as a critical factor influencing the reported trips' spatial activity levels and destination. We are supposed to present an additional paper devoted to that novel investigation of mobility with this slightly different research design angle. Because several variables will be included in a new manuscript scenario, we prefer to keep the current data unpublished before submission.

Did the cultural aspect played any role in the predisposition for mobility under adverse conditions? Please comment on this.

• Thank you for this interesting question. This cultural influence could be interesting to test regarding the potential presence and the impact of traditional (and even patriarchal) family roles within a contemporary Russian family. Unfortunately, without testing this hypothesis based on a cross-cultural level, we could not be sure of more profound spatial activity in Russian males( with the critical role of the men in the resource acquisitions) mostly predisposed by the traditions and the cultural aspects.

Please comment on the relevance of the present findings with regards to understanding the impact of stress related factors into models of covid spreading (see “Stress as a Meaningful Variable in Models of Covid-19 Spreading.” PsyArXiv. December 6. doi:10.31234/osf.io/kcpqm)

• Thank you for this material. We introduced some of the issues into our new section limitations based on that paper ( pages15-16 and also, we gave a quote on this paper in line 57).

General information about changes in the manuscript

Authors fully agreed with the reviewers regarding the importance of better defining what anxiety was in our first manuscript; therefore, we have changed the variable name of “Anxiety” to the “level of worry.” It is a single item variable, and it mostly catches the dimension of worry, which is only a partial aspect of anxiety.

To explain the relevance of the problem, In the Abstract was added this sentence:

Despite the enforcing lockdown regime in late March 2019 in Russia, the phenomenon of the continued virus spreading highlighted the importance of studies investigating the range of biosocial attributes and spectrum of individual motivations underlying the permanent presence of the substantial level of spatial activity.

In the Abstract, the sentence at line 17 was rephrased into:

However, our data suggested that wariness largely determines adequate health attitudes; hence a higher level of warry indirectly reduced individual mobility.

 In the Introduction section, the first-to-third paragraph’s sentences were changed in their orders.

Also, lines 58-63 (see previous manuscript) were removed:

Over the past few decades, many studies have analyzed the spatial-distance aspects of human activity as a reflection of its socio-demographic characteristics [14–18]. Data from cell phone-based accelerometers were used for building predictive model for activity recognition, spatial movements and activity patterns; for instance, the most likely distances and directions of the people's daily mobility have been profoundly investigated [19–22]. Demographic attributes and socio-economic parameters are estimated and implemented in big data analyses of human spatial behavior [23].

In the Method section wee added information about the inclusion criterion(page 3, line 135-142)

Also, we expanded the statistical analyses methods procedure information (page 4, lines 179-189)

We added into Introduction section a descriptive statistics tables for males and females.

In the results section, all the diagram and model contain the variable “Worry” instead of "Anxiety”

In the Results section, we slightly reduced; in particular, this paragraph was sufficiently reduced:

Given that the full group model may mask the effects specific to each sex, the multiple group model has been implemented. We performed a multi-sample SEMs unconstrained structural models, where structural relationships are specified in all groups meaning that the model is equal across groups. Still, the coefficients in the relationships were estimated independently for each group. The results (Figures 3 and 4) demonstrated that each of the group paths estimates were in the expected direction and indicated many similarities in structural relationships with the full model presented in Table 2 and Figure 1. All paths were significant for each group except the family size (the number of inhabitants), which was not significant for females’ level of worry.

The Chi-Squared difference test between constrained and unconstrained models demonstrated a high significance p (χ2) < 0.00305. Hence, the model fit index changes were significant, indicating that metric invariance was not established, and the path estimates for males and females were different. This means that the unconstrained multiple group model should be accepted as more representative SEMs. A detected non-invariance in the model intercepts and partly non-significant pathways represent the evidence in favor of systematic sex-specific differences in the model parameters being examined. Due to this fact, in the next step, we performed two SEM’s separately for each sex. Figures 3 and 4 demonstrate the path diagrams for males’ (panel A) and for females’ sample (Panel B). (The above paragraph was rephrased into three sentences ( see lines 253-258).

In the results, figure 5 has been replaced by the same histogram data but splinted between two sexes.

In the Results section, the Paragraph under figure 6 was reduced, as this information presented in the Discussion section:

From an evolutionary perspective, the human pair-bonding has been an adaptation for offspring survival given the long period of their dependence. Under these circumstances, the sexual division of labor and the need for paternal investment were of key importance [26,59–61]. According to the parental investment theory, human males have been rarely providing direct care for their offspring. However, they are essential contributors to the family budget and mostly provided the economic sustainability of the family and offspring well-being and survival in general [62,63]. The indirect care, by definition, has a strong association with higher social network and higher spatial mobility of the male contributors. Our data suggests that females were less prone to get in contact with the dangerous environment and preferred to delegate this function to their partners.

In the Discussion section, paragraphs, 3,4,5 were reorganized to avoid repeated ideas and maintain general consistency.

The section of Limitation was added (page 15-16)

Author Response File: Author Response.pdf

Reviewer 3 Report

The manuscript needs improvement in aspects such as the following:

- Clarify the abstract to simplify the study and the main results.

-In the introduction, make sure the paragraphs are connected. The same ideas are presented repeatedly and there is no consistency.

- Missing information in the process.The procedure for selecting participants (inclusion and exclusion criteria) has not been clarified.

-The sample is so low, justify the reason.

- Please add information on the recruitment of participants and if language was an inclusion / exclusion criterion.

- If the Crobach's alpha is low, can the authors argue how the study is valid, reliable and replicable?

Authors can add more information on how the measures were constructed and based on what criteria.

-It is recommended to clarify the discussion section.

-A section with the conclusions of the manuscript is recommended.

Author Response

The manuscript needs improvement in aspects such as the following:

Clarify the abstract to simplify the study and the main results.

• Thank you for this comment, we made some changies in the Abstract

In the introduction, make sure the paragraphs are connected. The same ideas are presented repeatedly and there is no consistency.

• Thank you for the comment, we revised the introduction

Missing information in the process.The procedure for selecting participants (inclusion and exclusion criteria) has not been clarified.

• Added the information on page 3, line 138

The sample is so low, justify the reason.

• This is an important limitation of the study, and we added an information in the limitation section and also  discuss the consequences of the small sample size ( pages 15-16)

Please add information on the recruitment of participants and if language was an inclusion / exclusion criterion.

• Thank you for this comment, we addd this into Methods

If the Crobach's alpha is low, can the authors argue how the study is valid, reliable and replicable?

• In this study, we aimed to measure the objective behavior – the spatial activity; for that reason, we collected three types of manifested variables which are supposed to be firmly linked to the actual behavioral acts – outdoor spatial activity: (1) the most distant place out of the home within a week (2) the most distant locale out of the house within a day, (3) the number of times a person left home within a week. Their joint weights are supposed to be aggregated into the one-factor latent construct, representing the actual individual spatial behavior. By aggregating three variables into one latent construct, "Activity," we expected to have a more reliable latent variable than any of those three manifested data being separated. Here we only were trying to be as close as possible to the actual human behavior – a spatial activity level. We are sure that receiving relatively low Cronbach's alpha does not mean that our manifested variables are wrong or that the intention to aggrege them three into a higher-order latent construct is not the correct procedure as would be expected in cases where some phycological features were constructed. As we deal with accurate facts and behavioral acts, this relatively low Cronbach's Alpha means that we have a relatively small sample size. The latter is defiantly a critical limitation of our study. We are sure that by implementing repeated measures of weekly spatial activity within the same participant in a different period or the future expanding the sample size, these three manifested variables become much more correlated. In that case, the final latent construct, "Activity," is believed to achieve a much higher Cronbach's Alpha.

Authors can add more information on how the measures were constructed and based on what criteria.

• Thank you for this comment we mentioned this issue in the limitation section.

It is recommended to clarify the discussion section.

• In the Discussion section, paragraphs, 3,4,5 were reorganized to avoid repeated ideas and maintain general consistency ( pages 13-14)

A section with the conclusions of the manuscript is recommended.

• Based on the several critical issues and drawbacks we had to expanded the manuscript and introducing a sufficient limitation section, due to the fact of the increasing the article size we had to keep the word limits and focus on the revising a discussion section

General information about changes in the manuscript

Authors fully agreed with the reviewers regarding the importance of better defining what anxiety was in our first manuscript; therefore, we have changed the variable name of “Anxiety” to the “level of worry.” It is a single item variable, and it mostly catches the dimension of worry, which is only a partial aspect of anxiety.

To explain the relevance of the problem, In the Abstract was added this sentence:

Despite the enforcing lockdown regime in late March 2019 in Russia, the phenomenon of the continued virus spreading highlighted the importance of studies investigating the range of biosocial attributes and spectrum of individual motivations underlying the permanent presence of the substantial level of spatial activity.

In the Abstract, the sentence at line 17 was rephrased into:

However, our data suggested that wariness largely determines adequate health attitudes; hence a higher level of warry indirectly reduced individual mobility.

 In the Introduction section, the first-to-third paragraph’s sentences were changed in their orders.

Also, lines 58-63 (see previous manuscript) were removed:

Over the past few decades, many studies have analyzed the spatial-distance aspects of human activity as a reflection of its socio-demographic characteristics [14–18]. Data from cell phone-based accelerometers were used for building predictive model for activity recognition, spatial movements and activity patterns; for instance, the most likely distances and directions of the people's daily mobility have been profoundly investigated [19–22]. Demographic attributes and socio-economic parameters are estimated and implemented in big data analyses of human spatial behavior [23].

In the Method section wee added information about the inclusion criterion(page 3, line 135-142)

Also, we expanded the statistical analyses methods procedure information (page 4, lines 179-189)

We added into Introduction section a descriptive statistics tables for males and females.

In the results section, all the diagram and model contain the variable “Worry” instead of "Anxiety”

In the Results section, we slightly reduced; in particular, this paragraph was sufficiently reduced:

Given that the full group model may mask the effects specific to each sex, the multiple group model has been implemented. We performed a multi-sample SEMs unconstrained structural models, where structural relationships are specified in all groups meaning that the model is equal across groups. Still, the coefficients in the relationships were estimated independently for each group. The results (Figures 3 and 4) demonstrated that each of the group paths estimates were in the expected direction and indicated many similarities in structural relationships with the full model presented in Table 2 and Figure 1. All paths were significant for each group except the family size (the number of inhabitants), which was not significant for females’ level of worry.

The Chi-Squared difference test between constrained and unconstrained models demonstrated a high significance p (χ2) < 0.00305. Hence, the model fit index changes were significant, indicating that metric invariance was not established, and the path estimates for males and females were different. This means that the unconstrained multiple group model should be accepted as more representative SEMs. A detected non-invariance in the model intercepts and partly non-significant pathways represent the evidence in favor of systematic sex-specific differences in the model parameters being examined. Due to this fact, in the next step, we performed two SEM’s separately for each sex. Figures 3 and 4 demonstrate the path diagrams for males’ (panel A) and for females’ sample (Panel B). (The above paragraph was rephrased into three sentences ( see lines 253-258).

In the results, figure 5 has been replaced by the same histogram data but splinted between two sexes.

In the Results section, the Paragraph under figure 6 was reduced, as this information presented in the Discussion section:

From an evolutionary perspective, the human pair-bonding has been an adaptation for offspring survival given the long period of their dependence. Under these circumstances, the sexual division of labor and the need for paternal investment were of key importance [26,59–61]. According to the parental investment theory, human males have been rarely providing direct care for their offspring. However, they are essential contributors to the family budget and mostly provided the economic sustainability of the family and offspring well-being and survival in general [62,63]. The indirect care, by definition, has a strong association with higher social network and higher spatial mobility of the male contributors. Our data suggests that females were less prone to get in contact with the dangerous environment and preferred to delegate this function to their partners.

In the Discussion section, paragraphs, 3,4,5 were reorganized to avoid repeated ideas and maintain general consistency.

The section of Limitation was added (page 15-16)

Author Response File: Author Response.pdf

Reviewer 4 Report

This paper researches on the differences in human mobility patterns, concerning the key biosocial and demographic attributes using data of related to the COVID-19 pandemic. While the topic is interesting, I find some issues with the paper, namely:

• The abstract discusses in much detail the results that were achieved and fails to properly explain the relevance of the problem. Please note that abstracts should be standalone, so with a limited number of words it would be able to do both things, not simply discuss the results;
• Materials and Methods section uses around 1 page (in 18) and lacks in accurately explaining the methods that were employed. I.e., while the parts where the authors address the results could be slightly reduced, the Materials and Methods section should contain additional details. As it stands, it is lacking in this aspect;
• There is not a Conclusions section, which should be present so that authors present the main conclusions, limitations, implications, and future work agenda. The authors chose to end the paper with a discussion of the results, which in my view limits the ability of the paper to convey all the relevant aforementioned messages of the research to the readers;
• Sections are not uniformly formatted;
• line 127 cites reference [question mark];
• of the 90 listed references some are not used throughout the paper text.

With all this in mind, I believe there are some issues and that a resubmission is required. When preparing the revised manuscript, I suggest that authors address the concerns I listed above.

Author Response

The abstract discusses in much detail the results that were achieved and fails to properly explain the relevance of the problem. Please note that abstracts should be standalone, so with a limited number of words it would be able to do both things, not simply discuss the results;

1. • We are agreed and added line 11-14

Materials and Methods section uses around 1 page (in 18) and lacks in accurately explaining the methods that were employed. I.e., while the parts where the authors address the results could be slightly reduced, the Materials and Methods section should contain additional details. As it stands, it is lacking in this aspect;

• We added addition information about the SEM statistical methodology ( page 4, line 179 -189).The Results was slightly reduced

There is not a Conclusions section, which should be present so that authors present the main conclusions, limitations, implications, and future work agenda. The authors chose to end the paper with a discussion of the results, which in my view limits the ability of the paper to convey all the relevant aforementioned messages of the research to the readers

• • We fully agreed with the review and had introduced the additional section Limitation included the future work agenda (pages 15-16)

Sections are not uniformly formatted;

line 127 cites reference [question mark];

• • Thank you for these notes, the sign was delated

of the 90 listed references some are not used throughout the paper text.

1. • Corrected

General information about changes in the manuscript

Authors fully agreed with the reviewers regarding the importance of better defining what anxiety was in our first manuscript; therefore, we have changed the variable name of “Anxiety” to the “level of worry.” It is a single item variable, and it mostly catches the dimension of worry, which is only a partial aspect of anxiety.

To explain the relevance of the problem, In the Abstract was added this sentence:

Despite the enforcing lockdown regime in late March 2019 in Russia, the phenomenon of the continued virus spreading highlighted the importance of studies investigating the range of biosocial attributes and spectrum of individual motivations underlying the permanent presence of the substantial level of spatial activity.

In the Abstract, the sentence at line 17 was rephrased into:

However, our data suggested that wariness largely determines adequate health attitudes; hence a higher level of warry indirectly reduced individual mobility.

 In the Introduction section, the first-to-third paragraph’s sentences were changed in their orders.

Also, lines 58-63 (see previous manuscript) were removed:

Over the past few decades, many studies have analyzed the spatial-distance aspects of human activity as a reflection of its socio-demographic characteristics [14–18]. Data from cell phone-based accelerometers were used for building predictive model for activity recognition, spatial movements and activity patterns; for instance, the most likely distances and directions of the people's daily mobility have been profoundly investigated [19–22]. Demographic attributes and socio-economic parameters are estimated and implemented in big data analyses of human spatial behavior [23].

In the Method section wee added information about the inclusion criterion(page 3, line 135-142)

Also, we expanded the statistical analyses methods procedure information (page 4, lines 179-189)

We added into Introduction section a descriptive statistics tables for males and females.

In the results section, all the diagram and model contain the variable “Worry” instead of "Anxiety”

In the Results section, we slightly reduced; in particular, this paragraph was sufficiently reduced:

Given that the full group model may mask the effects specific to each sex, the multiple group model has been implemented. We performed a multi-sample SEMs unconstrained structural models, where structural relationships are specified in all groups meaning that the model is equal across groups. Still, the coefficients in the relationships were estimated independently for each group. The results (Figures 3 and 4) demonstrated that each of the group paths estimates were in the expected direction and indicated many similarities in structural relationships with the full model presented in Table 2 and Figure 1. All paths were significant for each group except the family size (the number of inhabitants), which was not significant for females’ level of worry.

The Chi-Squared difference test between constrained and unconstrained models demonstrated a high significance p (χ2) < 0.00305. Hence, the model fit index changes were significant, indicating that metric invariance was not established, and the path estimates for males and females were different. This means that the unconstrained multiple group model should be accepted as more representative SEMs. A detected non-invariance in the model intercepts and partly non-significant pathways represent the evidence in favor of systematic sex-specific differences in the model parameters being examined. Due to this fact, in the next step, we performed two SEM’s separately for each sex. Figures 3 and 4 demonstrate the path diagrams for males’ (panel A) and for females’ sample (Panel B). (The above paragraph was rephrased into three sentences ( see lines 253-258).

In the results, figure 5 has been replaced by the same histogram data but splinted between two sexes.

In the Results section, the Paragraph under figure 6 was reduced, as this information presented in the Discussion section:

From an evolutionary perspective, the human pair-bonding has been an adaptation for offspring survival given the long period of their dependence. Under these circumstances, the sexual division of labor and the need for paternal investment were of key importance [26,59–61]. According to the parental investment theory, human males have been rarely providing direct care for their offspring. However, they are essential contributors to the family budget and mostly provided the economic sustainability of the family and offspring well-being and survival in general [62,63]. The indirect care, by definition, has a strong association with higher social network and higher spatial mobility of the male contributors. Our data suggests that females were less prone to get in contact with the dangerous environment and preferred to delegate this function to their partners.

In the Discussion section, paragraphs, 3,4,5 were reorganized to avoid repeated ideas and maintain general consistency.

The section of Limitation was added (page 15-16)

Author Response File: Author Response.pdf

Round 2

Reviewer 1 Report

The Authors performed all the modifications I proposed. From my point of view, the manuscript was improved and it is ready for the publication.

Reviewer 3 Report

I consider that in the current format the manuscript has improved considerably.