A network analysis of problematic smartphone use in Japanese young adults.

BACKGROUND: We aimed to explore the overall network structure of problematic smartphone use symptoms assessed by smartphone addiction scale-short version (SAS-SV) and to identify which items could play important roles in the network.
METHODS: 487 college and university students filled out the study questionnaire, including SAS-SV. We constructed a regularized partial correlation network among the 10 items of SAS-SV. We calculated three indices of node centrality: strength, closeness, and betweenness, to quantify the importance of each SAS-SV item.
RESULTS: We identified 34 edges in the estimated network. In the given network, one item pertaining to withdrawal symptom hadthe highest strength and high closeness centrality. Additionally, one item related to preoccupation was also found to have high centrality indices.
CONCLUSION: Our results indicating the central role of one withdrawal symptom and one preoccupation symptom in the symptom network of problematic smartphone use in young adults were in line with a previous study targeting school-age children. Longitudinal study designs are required to elicit the role of these central items on the formation and maintenance of this behavioral problem.

Introduction

The penetration rate of the Internet in Japan has increased dramatically over the last two decades, from 9.1% in 1997 to 89.8% in 2019 [1]. One of the characteristic features of the latest changes in the Internet environment is the wide and rapid distribution of smartphones in all age groups, especially among young generations. In recent years, the number of internet users who access the Internet through a smartphone (63.3%) is higher than those who use the Internet via desktop and/or laptop computers (50.4%) [1].

Smartphones were launched in Japan in 2007, and three years later, the rate of smartphone users remained only 9.7% [1]. However, in accordance with the improvements in information and communication technology, the rate of smartphone users reached 83.4% in 2019. As the number of smartphone users becomes higher, problems related to smartphone overuse have become more serious.

As a historical standard, addiction is commonly referred to as, and associated with, substance abuse. However, the connotation of the term has been expanded since the 1990s to include behaviors that may lead to rewards (e.g., gambling, shopping, sex) [2]. In 2011, the American Society of Addiction Medicine (ASAM) released an official statement defining all addictions, including behavioral addictions, in terms of brain changes [3]. In the statement, ASAM provided a clear definition, namely that "Addiction is a major chronic disease of the brain’s reward, motivation, memory, and related circuits." Although the diverse positions taken by various proponents have yet to culminate in a complete consensus [4, 5], in general, problematic smartphone use is considered as a subtype of behavioral addiction in which a person is excessively engaged in challenging behaviors that are not substance-related, even though they have a negative impact on the person’s physical, social, economic and mental well-being [6, 7].

Problematic smartphone use is considered the result of excessive and ultimately destructive smartphone overuse and is characterized by clinical features of behavioral addiction; preoccupation, compulsive behavior, lack of control, functional impairment, withdrawal and tolerance [8, 9]. In response to the increase of smartphone users, studies regarding problematic smartphone use have been reported from various countries [10-12].

Several research groups developed questionnaires to screen for and assess problematic smartphone use, including the Smartphone Addiction Scale (SAS) [13-16]. The short version of SAS (SAS-SV) consists of 10 items selected by a consensus among experts in addiction psychiatry from the original version of SAS which is composed of 33 items [13, 17]. At present, SAS and SAS-SV are widely used self-rating scales to assess the severity of smartphone addiction in the world, and have been translated into several languages [18].

Historically, researchers have used a latent variable model, where the items (symptoms, signs) are manifestations of a particular underlying attribute (problematic smartphone use, for example) to understand the disease model. In this model, the observed items are independent of each other given an individual’s score on the latent variable (local independence) [19]. However, behavioral addiction is a complex human behavior phenomenon, and is composed of heterogeneous observable signs and symptoms. Therefore, treating problematic smartphone use within a latent variable model may overshadow meaningful associations existing between individual symptoms.

Although not mutually exclusive, a network model has been recently gaining attention as an alternative approach to the latent variable model in understanding the psychopathology of mental disorders [20, 21], including behavioral addiction [22, 23]. In the network model, problematic smartphone use is not an underlying latent unobservable disease entity. Rather, it is considered to be a complex network of mutually reinforcing symptoms [24]. This approach allows for identifying the overall network structure of disorders by quantifying associations among observable symptoms. Additionally, the network approach allows for identifying the most central symptom(s) that potentially play a critical role in the onset and maintenance of disorders. Two research groups have recently employed this approach to elucidate the network structure of problematic smartphone use in preadolescents and adolescents, where symptoms pertaining to withdrawal, loss of control, and continued excessive use were deemed as central symptoms [18, 25]. However, no network studies in this field have focused on individuals transitioning to adulthood, a population which is considered uniquely vulnerable to mental health disorders, including addiction problems [26]. Nearly everyone in this age group possesses smartphones with scant or no parental controls, and thus behaviors related to smartphone use in this population would not be directly influenced by family factors (parenting style, for example).

In this study, we aimed to: 1) explore the overall network structure of problematic smartphone use symptoms in Japanese college students, and 2) identify the most central symptoms in the network.

Materials and methods

Participants

The subjects of the study were 487 private college and university students in Sapporo city and its environs in Japan. Study participation rate was 81.2% (487/600). Their academic deviation scores were average or a little below the average compared to the national standard in Japan. Research collaborators for data collection were recruited through personal connections of the first author of this paper (MT). Nine teachers from three universities and six colleges agreed to voluntarily support our investigation. This study is a secondary data analysis using data obtained for a study focusing on internet addiction, smartphone addiction, and the Hikikomori trait that was conducted between July and October 2018 and was previously reported elsewhere [27]. Characteristics of the study participants are summarized in Table 1. In this study, the participants received the questionnaire sheets and filled out the questionnaire in the classroom.

Table 1

Characteristics of the study participants.

	Whole (n = 487)(Male 132, Female 355)
Age (mean ± SD)	19.6±1.5
SAS-SV (mean ± SD)	29.6±8.8
Internet use (hours) (mean ± SD)
Weekdays	4.86±3.1
Weekend	6.82±4.1
Purpose (%)
Gaming	42 (8.6)
SNS	303 (62.2)
Video-sharing	101 (20.7)
Music	23 (4.7)
Web searches	13 (2.7)
Others	5 (1.0)

SAS-SV: Smartphone Addiction Scale–Short Version, SD: Standard Deviation, SNS: Social Network Service

Measures

Smartphone addiction scale–short version

The original version of the SAS was developed as a self-report scale in South Korea [13]. The SAS includes 33 questions that assess six domains relating to smartphone overuse (continued overuse, loss of control, preoccupation, withdrawal, tolerance, and functional impairment) on a six-point Likert scale ranging from 1 (strongly disagree) to 6 (strongly agree). The Japanese-version of the SAS was developed, and its reliability and factor validity were assessed in 1,037 youth aged 15–24 years [28]. Kwon et al. also developed the short version of SAS (SAS-SV) by extracting 10 questions from the SAS to use it as a smartphone addiction screener and confirmed its validity [17]. In translating the SAS-SV into Japanese, our research team made a minor modification by inserting the term LINE [29] into the sentence of SA8 (Constantly checking my smartphone so as not to miss conversations between other people on Twitter or Facebook); however, the Japanese version of the SAS-SV has not yet been validated in Japanese youth.

The 10 items in the SAS-SV are listed in Table 2. The researchers in the present study discussed and defined the function of each SAS-SV item by referencing diagnostic criteria of gaming disorder found in the International Classification of Diseases, 11th Revision (ICD-11) and internet gaming disorder of Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition (DSM-5) to assist readers in interpreting the findings of this study. The internal consistency of the Japanese-version SAS-SV for the participants in the present study was rated good, with Cronbach’s alpha being 0.83.

Table 2

Ten items of smartphone addiction scale—Short version and mean scores in this study.

SAS-SV Items		Consensus among authors of this study	Mean scores (mean ± SD) (n = 487)
SA1	Missing planned work due to smartphone use	functional impairment (occupational)	3.5 ± 1.4
SA2	Having a hard time concentrating in class, while doing assignments, or while working due to smartphone use	functional impairment (academic)	3.3 ± 1.4
SA3	Feeling pain in the wrists or at the back of the neck while using a smartphone	functional impairment (physical)	2.3 ± 1.4
SA4	Won’t be able to stand not having a smartphone	withdrawal	3.1 ± 1.6
SA5	Feeling impatient and fretful when I am not holding my smartphone	withdrawal	2.2 ± 1.3
SA6	Having my smartphone in my mind even when I am not using it	preoccupation	2.1 ± 1.2
SA7	I will never give up using my smartphone even when my daily life is already greatly affected by it.	loss of control	3.2 ± 1.5
SA8	Constantly checking my smartphone so as not to miss conversations between other people on LINE, Twitter or Facebook	loss of control	2.9 ± 1.4
SA9	Using my smartphone longer than I had intended	continued overuse	4.5 ± 1.2
SA10	The people around me tell me that I use my smartphone too much.	continued overuse	2.5 ± 1.4

SAS-SV: Smartphone Addiction Scale–Short Version

SD: Standard Deviation

Data analyses

Network estimation

In the present study, we constructed a regularized partial correlation network using a Graphical Gaussian Model between the 10 symptoms of problematic smartphone use. In this network model, nodes represent individual items, and relationships between nodes are defined as edges. Edges are understood as partial associations between two nodes while controlling for all other nodes in the network, and their thickness or weights reflect the strength of association (an estimation of partial correlations coefficients). The lack of edge between two nodes means conditional independence relationships among the nodes. The network was estimated and visualized using the R-package ‘qgraph’ in the statistical program ‘R version 4.0.3’ [30, 31]. In order to create a more parsimonious network and minimize the likelihood of type-I errors, the graphical LASSO (Least Absolute Shrinkage and Selection Operator), reducing the edges by shrinking the smallest edges exactly to zero, in the R package qgraph. This process of regularization is coupled with best fit model selection, by minimizing an information criterion, in this case, Extended Bayesian Information Criterion (EBIC) [32]. We used the Fruchterman-Reingold algorithm for network visualization, a force-directed algorithm that encourages closely related nodes to be plotted near each other [33].

Centrality

We calculated three indices of node centrality to quantify the importance of each of the 10 symptoms in the SAS-SV network [34]: closeness (the average shortest path between a given node and the remaining nodes in the network); betweenness (the number of times that a node lies on the shortest path between two other nodes); and strength (the sum of the absolute value of its connections with other nodes in the network). Previous research showed strength to be the most robust centrality measure [35]. These centrality values are typically presented as standardized Z-scores, with higher values reflecting greater overall importance of a symptom to the network.

Network stability

To investigate the reliability and robustness of the study results, we assessed the accuracy and stability of network edges and centrality indices using the R package ‘bootnet’ [19]. More specifically, using nonparametric bootstrap methods, we estimated network stability as follows: 1) constructed a 95% bootstrapped confidence interval around the regularized edge weights, 2) computed an edge-weight difference test, and 3) estimated the correlation stability coefficient of centrality indices (via a case-dropping bootstrap procedure). Centrality indices were considered strongly stable if the values of the correlation stability coefficient were over 0.5, while values below 0.25 indicated inadequate stability [35].

Ethical issues

This study was approved by the ethics committee of Tokiwa Hospital. The study’s aim was stated on the cover page of the questionnaire sheets that requested voluntary respondents to answer all questions anonymously. Answering the questions was deemed to constitute consent.

Results

Characteristics of the study participants

Table 1 shows the characteristics of the study participants (n = 487). 355 female students (mean years of age: 19.4, standard deviation (SD): 1.4) and 132 male students (mean years of age: 20.2, SD: 1.8) completed the questionnaire. More than half of participating students used the internet primarily for social networking services (SNSs) (female: 70%, male: 40.9%). The rate of gaming was higher in males compared to that in females, 18.9% and 4.8%, respectively.

Network structure and centrality

Fig 1 depicts the network structure of relations among SA symptoms in study participants. We identified 34 edges in the estimated network, among which the weights of the edges between the items SA1 (Miss planned work due to smartphone use) and SA2 (Difficulty focusing on assignments/work due to smartphone use) and those between the items SA5 (Feel impatient and uneasy when not holding my smartphone) SA6 (Having my smartphone in my mind even when I am not using it) were large (0.49 and 0.41, respectively). Other strong associations include those between SA4 (Unable to stand not carrying a smartphone) and SA7 (Continued use of my smartphone despite its negative impact on daily life) (0.33) and between SA4 and SA5 (0.27). All edge weights of this network are listed in S1 File.

Fig 1

Regularized partial correlation network of smartphone addiction scale short-version in university and college students in Japan (n = 478).

Green lines indicate a positive association. Line thickness reflects the strength of association, controlling for all other symptom nodes in the network. SA: Smartphone Addiction Scale–Short version item SA1: Missing planned work due to smartphone use, SA2: Having a hard time concentrating in class, while doing assignments, or while working due to smartphone use, SA3: Feeling pain in the wrists or at the back of the neck while using a smartphone, SA4: Won’t be able to stand not having a smartphone, SA5: Feeling impatient and fretful when I am not holding my smartphone, SA6: Having my smartphone on my mind even when I am not using it, SA7: I will never give up using my smartphone even when my daily life is already greatly affected by it, SA8: Constantly checking my smartphone so as not to miss conversations between other people on Twitter or Facebook, SA9: Using my smartphone longer than I had intended, SA10: The people around me tell me that I use my smartphone too much.

In the given network, SA5 was considered a central node with the highest strength and high closeness (Fig 2). SA6 (Have my smartphone on my mind even when not using it) was with the highest closeness and relatively high strength in centrality indices. In particular, while this item was strongly associated with SA5, it also had the moderate association with SA10 (People around me tell me that I use my smartphone too much) that was only weakly associated with the central node (SA5). Both SA3 (Feel pain in the wrists or the neck while using a smartphone) and SA8 (Constantly check my smartphone not to miss information on SNS) were two items that notably had low centrality indices.

Fig 2

Network centrality.

Values presented as standardized Z-scores, with higher scores indicative of greater influence within the overall network. SA: Smartphone Addiction Scale—Short version item.

Network stability

The edge weights bootstrap (S2 File) showed that the 95% confidence intervals for many of the edges were overlapping. Additionally, there were few significant differences between the strong edges (S3 File). These findings indicate the most of the edges do not significantly differ and that the ranking of edge weights should be interpreted with care.

As shown in S4 File, examination of the stability of centrality indices was satisfactory for strength but not for other indices (the correlation stability coefficient of strength: 0.75, closeness: 0.44, and betweenness: 0.05).

Discussion

In the present study, we estimated the network structure of problematic smartphone use symptoms among college students in Japan and identified the central (or “core”) symptoms in the smartphone addiction psychopathology network. Although two recent studies examined the network structure of problematic smartphone use and smartphone addiction in school-age children in China [25] and Brazil [18], this is the first study that employed the network approach in understanding psychopathology of problematic smartphone use in college students in Japan who were transitioning to adulthood. This population has higher possession rates of smartphones and quite likely has scant or fewer parental controls compared to the school-age-sample previously studied [18, 25]. In fact, the survey conducted in Japan exhibited more maladaptive and excessive use of smartphones in college students compared to middle and high school students [36]. Thus, findings from the present study could further our understanding of the network structure of problematic smartphone use psychopathology at this important developmental stage and could become the foundation for future research.

Our findings that identified one withdrawal symptom (SA5) and one preoccupation symptom (SA6) as central items in the network of problematic smartphone use were in line with those of the study conducted by Andrade et al. [18]. Consistent findings in the two studies targeting different age groups suggest these two symptoms play pivotal roles in the psychopathology of problematic smartphone use for a wide range of ages (from school-age children and adolescents to young adults). Furthermore, these central items and the strong edge (connection) between these two items in our study would support the three-stage model of addiction, where three stages of binge/intoxication (overuse of smartphone), withdrawal/negative affect, and preoccupation/anticipation, feed into each other to produce the addiction cycle [37]. Although we did not identify strong centrality in items pertaining to continued overuse of the smartphone (SA9 and SA10 items), it is still possible that these symptoms contribute to either the development or maintenance of the network of problematic smartphone use via the associations between these overuse symptoms and preoccupation symptoms (edge weight 0.21 between SA6 and SA10, for example) based on the above-mentioned three-stage model of addiction. It would be meaningful to examine the directionality among symptoms in order to better understand how the symptoms spread in the network of problematic smartphone use. However, doing so is beyond the scope of this study given our cross-sectional design. Future studies may be able to address this by using advanced data collection methods (intensive longitudinal data through the use of ecological momentary assessment [38], for example) that allow us to elucidate the directionality among symptoms and symptom dynamics in problematic smartphone use.

Huang and his colleagues conducted a network analysis of problematic smartphone use, measured by using the Smartphone Addiction Proneness Scale, a self-rating scale, in grade 4 and 8 students in China [25]. They reported that loss of control and continued excessive use were the central symptoms of problematic smartphone use. Differences in the symptom centrality between the study above and our study may be attributed to the difference in the scale used for each study as the functions of the scale modified by researchers in the above-mentioned study did not have withdrawal as one of the scale dimensions. Additionally, the difference could be due to the participants’ age in the two studies given that the ability of self-control generally improves with age [39]. It is also reported that the older the age, the more people use their smartphones for social purposes, such as SNS [40]. When smartphone users communicate on SNS applications, message senders who do not receive a response sometimes perceive themselves as being neglected or ignored. Conversely, message recipients may become obsessive about checking and immediately replying to messages and keep their smartphones by their side during all their waking hours. Thus, such smartphone users could have difficulties leaving their smartphones out of reach and feel anxious without the smartphone in close proximity. Accordingly, we believe that young adult population, compared to school age population, might have withdrawal symptoms as more central in the network than overuse symptoms in the present study.

In the present study, more than half of the participants reported SNS as their primary use of the smartphone (62.2%), the percentage of which was higher than that of other reasons (gaming, for example). This raises a potential hypothesis that different smartphone usages lead to different network structures. Future studies with larger sample sizes would allow researchers to examine if the smartphone usage is an important factor contributing to network structures of problematic smartphone use.

This study has several limitations. The sample size was modest. Given that the stability of centrality indices in the network analysis is affected by the study sample size, larger sample-sized studies are desirable. However, in the present study, the stability of strength, one of the centrality indices, was satisfactory. Regarding our sampling method, we recruited study participants only from colleges in one region in Japan, and the gender ratio was far from even, affecting the generalizability of our study findings. Additionally, our cross-sectional study design prohibited us from elucidating temporal relationships among individual symptoms of problematic smartphone use. As stated above, future studies may clarify the temporality of symptoms in problematic smartphone use by employing advanced data collection methods (intensive longitudinal data, for example) and analytic methods. Further investigations should be conducted to clarify the deeper mechanisms of problematic smartphone use in order to establish future effective prevention strategies, as well as possible treatment paradigms.

Conclusions

Our results suggest that both withdrawal and preoccupation symptoms play pivotal roles in the psychopathology of problematic smartphone use in young adults. Longitudinal study designs are required to elucidate the role of these central items on the formation and maintenance of this behavioral problem.

Edge weights in the smartphone addiction network in the study participants.

(DOCX)

Click here for additional data file.

Bootstrapped 95% confidence intervals of edge weights.

The panel above presents edge weights of the estimated network of 10 Smartphone Addiction Scale items and 95% confidence intervals (CIs) calculated using bootstrapping. The red line represents the original sample values, the black dots represent the bootstrap means, and the gray areas indicate the bootstrapped CIs. Each horizontal line represents one edge of the network, ordered from the edge with the highest edge-weight to the edge with the lowest edge-weight. The y-axis labels were removed to avoid cluttering for the simplicity of the graph. Figure indicates that many edge-weights likely do not significantly differ from one-another. The generally large bootstrapped CIs imply that interpreting the order of most edges in the network should be done with care.

(DOCX)

Click here for additional data file.

Edge difference test.

In this graph, each point on the x and y axes represents a pair of edges identified in a given network. Black boxes indicate significant differences between two edges (a bootstrap stability difference test: alpha = 0.05), whereas gray boxes do not indicate any significant differences. The diagonal represents the edge strength, where white indicates weak edge strengths, while blue indicates strong edge strengths.

(DOCX)

Click here for additional data file.

Stability of network centrality.

The figure above presents the centrality stability as assessed using the case-dropping bootstrap method. Stability was assessed by re-estimating the network based on increasingly smaller subsets of the original sample.

(DOCX)

Click here for additional data file.

R code used for data analysis in the present research.

(DOCX)

Click here for additional data file.

(XLSX)

Click here for additional data file.

5 Apr 2022

A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). You should upload this letter as a separate file labeled 'Response to Reviewers'.

A marked-up copy of your manuscript that highlights changes made to the original version. You should upload this as a separate file labeled 'Revised Manuscript with Track Changes'.

An unmarked version of your revised paper without tracked changes. You should upload this as a separate file labeled 'Manuscript'.

If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter. Guidelines for resubmitting your figure files are available below the reviewer comments at the end of this letter.

If applicable, we recommend that you deposit your laboratory protocols in protocols.io to enhance the reproducibility of your results. Protocols.io assigns your protocol its own identifier (DOI) so that it can be cited independently in the future. For instructions see: https://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols. Additionally, PLOS ONE offers an option for publishing peer-reviewed Lab Protocol articles, which describe protocols hosted on protocols.io. Read more information on sharing protocols at https://plos.org/protocols?utm_medium=editorial-email&utm_source=authorletters&utm_campaign=protocols.

We look forward to receiving your revised manuscript.

Kind regards,

Roland Bouffanais, Ph.D.

Academic Editor

PLOS ONE

Journal requirements:

When submitting your revision, we need you to address these additional requirements.

1. Please ensure that your manuscript meets PLOS ONE's style requirements, including those for file naming. The PLOS ONE style templates can be found at

https://journals.plos.org/plosone/s/file?id=wjVg/PLOSOne_formatting_sample_main_body.pdf and

https://journals.plos.org/plosone/s/file?id=ba62/PLOSOne_formatting_sample_title_authors_affiliations.pdf".

2. Please provide additional details regarding participant consent. In the ethics statement in the Methods and online submission information, please ensure that you have specified what type you obtained (for instance, written or verbal, and if verbal, how it was documented and witnessed). If your study included minors, state whether you obtained consent from parents or guardians. If the need for consent was waived by the ethics committee, please include this information.

3. We note that the grant information you provided in the ‘Funding Information’ and ‘Financial Disclosure’ sections do not match. When you resubmit, please ensure that you provide the correct grant numbers for the awards you received for your study in the ‘Funding Information’ section.

4. Thank you for stating in your Funding Statement:

“This work was partially supported by the Grant-in-Aid for Scientific Research on (1) KAKENHI—the Japan Society for the Promotion of Science- JSPS (JP16H06403 & JP20H01773) and (2) The Japan Agency for Medical Research and Development - AMED (JP19dk0307073 & JP21wm0425010). The funders had no role in study design, data collection and analysis, in the decision to publish or the preparation of the manuscript.”

Please provide an amended statement that declares *all* the funding or sources of support (whether external or internal to your organization) received during this study, as detailed online in our guide for authors at http://journals.plos.org/plosone/s/submit-now.  Please also include the statement “There was no additional external funding received for this study.” in your updated Funding Statement.

Please include your amended Funding Statement within your cover letter. We will change the online submission form on your behalf.

5. Thank you for stating the following in the Funding Section of your manuscript:

“This work was partially supported by the Grant-in-Aid for Scientific Research on (1) KAKENHI—the Japan Society for the Promotion of Science- JSPS (JP16H06403 & JP20H01773) and (2) The Japan Agency for Medical Research and Development - AMED (JP19dk0307073 & JP21wm0425010). The funders had no role in study design, data collection and analysis, in the decision to publish or the preparation of the manuscript.”

We note that you have provided funding information that is not currently declared in your Funding Statement. However, funding information should not appear in the Acknowledgments section or other areas of your manuscript. We will only publish funding information present in the Funding Statement section of the online submission form.

Please remove any funding-related text from the manuscript and let us know how you would like to update your Funding Statement. Currently, your Funding Statement reads as follows:

“The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.”

Please include your amended statements within your cover letter; we will change the online submission form on your behalf.

6. In your Data Availability statement, you have not specified where the minimal data set underlying the results described in your manuscript can be found. PLOS defines a study's minimal data set as the underlying data used to reach the conclusions drawn in the manuscript and any additional data required to replicate the reported study findings in their entirety. All PLOS journals require that the minimal data set be made fully available. For more information about our data policy, please see http://journals.plos.org/plosone/s/data-availability.

Upon re-submitting your revised manuscript, please upload your study’s minimal underlying data set as either Supporting Information files or to a stable, public repository and include the relevant URLs, DOIs, or accession numbers within your revised cover letter. For a list of acceptable repositories, please see http://journals.plos.org/plosone/s/data-availability#loc-recommended-repositories. Any potentially identifying patient information must be fully anonymized.

Important: If there are ethical or legal restrictions to sharing your data publicly, please explain these restrictions in detail. Please see our guidelines for more information on what we consider unacceptable restrictions to publicly sharing data: http://journals.plos.org/plosone/s/data-availability#loc-unacceptable-data-access-restrictions. Note that it is not acceptable for the authors to be the sole named individuals responsible for ensuring data access.

We will update your Data Availability statement to reflect the information you provide in your cover letter.

7. We note that you have indicated that data from this study are available upon request. PLOS only allows data to be available upon request if there are legal or ethical restrictions on sharing data publicly. For more information on unacceptable data access restrictions, please see http://journals.plos.org/plosone/s/data-availability#loc-unacceptable-data-access-restrictions.

In your revised cover letter, please address the following prompts:

a) If there are ethical or legal restrictions on sharing a de-identified data set, please explain them in detail (e.g., data contain potentially sensitive information, data are owned by a third-party organization, etc.) and who has imposed them (e.g., an ethics committee). Please also provide contact information for a data access committee, ethics committee, or other institutional body to which data requests may be sent.

b) If there are no restrictions, please upload the minimal anonymized data set necessary to replicate your study findings as either Supporting Information files or to a stable, public repository and provide us with the relevant URLs, DOIs, or accession numbers. For a list of acceptable repositories, please see http://journals.plos.org/plosone/s/data-availability#loc-recommended-repositories.

We will update your Data Availability statement on your behalf to reflect the information you provide.

8. Please note that in order to use the direct billing option the corresponding author must be affiliated with the chosen institute. Please either amend your manuscript to change the affiliation or corresponding author, or email us at plosone@plos.org with a request to remove this option.

9. Your ethics statement should only appear in the Methods section of your manuscript. If your ethics statement is written in any section besides the Methods, please delete it from any other section.

Additional Editor Comments:

I have now heard from two reviewers and both of them agree about the importance of the topic and study. The network analysis is soundly applied. Having said that, both reviewers came up with a number of comments (some major and some minor), which I hope you'd be able to address in a revision. Specifically, I hope that you would be able to strengthen the robustness of your discussion of the results.

[Note: HTML markup is below. Please do not edit.]

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: Partly

Reviewer #2: Yes

**********

2. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: Yes

Reviewer #2: Yes

**********

3. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: No

Reviewer #2: Yes

**********

4. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: No

Reviewer #2: Yes

**********

5. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: In this article the authors perform an exploratory network analysis of the answers to the smartphone addiction scale-short version of 487 Japanese college and university students. This included estimating the network and 3 centrality indices. The authors also proceed to divide the sample according to gender and estimate two sub-sample networks, testing for their differences. They found that preoccupation and withdrawal related items are particularly important in the network and also found differences between male and female subsample networks.

The topic of the article is of great importance and the use of network analysis could be of significance to further improve the knowledge about addiction disorders. The network analysis methodology is overall well applied. The findings about the most central symptoms could indeed be of interest to the field.

However, this reviewer thinks the study’s novelty is limited and also that it lacks a robust discussion of the results.

The overall quality of the writing is modest.

Major issues

- Study’s novelty: the cited study by Andrade, 2020 also performs a network analysis of the same instrument as the current study. More so, it frames it in a validation context of the instrument, making it a sounder scientific effort than the current one. The authors claim however that the fact that they’re using a slightly older sample justifies the study’s novelty. That difference seems to this reviewer as a very small one (in the end the results did not differ much, proving that the difference was not so relevant).

- Some questions about the use of this instrument (SAS-SV) should be clarified, in particular its validation and/or the validation of the original SAS instrument in Japan. Was that work done previously? In this reviewer’s perspective that would be a fundamental first step before engaging in a study like the present one (in the cited Brazilian study, a careful validation of the SAS-SV is performed alongside network analysis)

- Regarding the sample, it represents a group of students from a specific region of Japan and with high gender distribution imbalance. This limits generalizability as the authors point. However, a more detailed description of the sample and its recruitment is necessary. For instance: when was the data collected? In which schools, from which specific region? Was there a rational for choosing this sample? Could the authors explain better the setting in which the questionnaire was filled? Were there potential participants who did not answer to the questionnaire? Simultaneously, a better sociodemographic description of these individuals should be provided – detailed school level description, age range (min and max values), and ideally some information that could inform the reader about socioeconomic status of this sample.

- Regarding the study’s aims, the reviewer believes the network comparison performed is lacking a justification. The authors should explain why they believe it is relevant to compare networks based on a gender dichotomization of the sample. Also, one problem regarding this aim is the imbalance on sub sample size. Given that the starting point is already relatively modest, the authors are left with small samples for the sub-analysis, particularly the men’s network. Was the stability of each of the subsample networks analyzed?

- As a general comment, the discussion section should be improved. In the current form it stresses too much aspects that are limited by design in the current study (eg ideas about how addiction disorders are formed or should be treated).

- The writing should be improved. A native English speaker with experience in scientific writing could help improve substantially the article (see in minor issues some passages to reconsider).

- A justification for not providing the data should be given (PLoS Data policy); This reviewer also strongly encourages the authors to make data analysis information available as supplementary material (R code)

Minor issues

- Smartphone addiction concept: the authors point the lack of consensus around this construct. The reviewer would ask the authors to consider the use of ‘Problematic Smartphone Use’ instead

- On a more technical note about the network comparison performed, the authors determine that they will perform 3 different tests in their methods section. As van Borkulo et al point out in their paper explaining this procedure, the edge strength invariance test should only be performed if there is an a priori hypothesis about specific edges to directly test between the networks, or, in case there is no a priori hypothesis, it could be done as a follow-up to a significant difference in the structural invariance test. Therefore, in this reviewer’s opinion, since there is no a priori hypothesis for the authors, in the methods section it should be specified that this test (edge strength invariance) will only be performed if a difference in structural invariance is found. Then, in the results section, the network comparison tests report seems to be a bit unclear. The authors state that there was a “significant difference for the overall network structure” but explain this as meaning that the “nodes were more densely connected overall in female students than in male students”. This explanation refers to the invariant global strength aspect of the test and not the invariant network structure. So, this reviewer is confused about what is being reported – invariant network structure or invariant global strength? The edge weight comparisons reported afterwards should only be pursued in case network structure differs. (See: Van Borkulo, C. D., Boschloo, L., Kossakowski, J., Tio, P., Schoevers, R. A., Borsboom, D., & Waldorp, L. J. (2017). Comparing network structures on three aspects: A permutation test. )

- The meaning of centrality indices should be rectified. On page 13 line 4, the way strength centrality is presented is not correct (ie, strength centrality does not translate the association of a node with “all other other nodes in the network” as stated. It is instead a measure of local influence). The meaning of betweenness and closeness centrality measures on cross-sectional data network analysis in the context of mental disorders has also been questioned. I would suggest the author to check references such as the following for this matter: Bringmann, L. F., Elmer, T., Epskamp, S., Krause, R. W., Schoch, D., Wichers, M., ... & Snippe, E. (2019). What do centrality measures measure in psychological networks?. Journal of abnormal psychology, 128(8), 892.

- In the methods section it should be mentioned which option was chosen for the network layout.

- Table 1: the asterisks in the Age and SAS-SV lines are a bit confusing and seem unnecessary to this reviewer; the readability of the table could be improved if the “(mean±SD” parts were positioned without a line break and if some hierarchical visual cue could be added to subdivisions of “internet use (hrs) and “purpose”; the acronym SNS should be explained in the table legend.

- Network estimation subsection from analytic plans: please use proper reference for qgraph package. A reference for the R statistical software is also lacking.

- Network estimation subsection from analytic plans: the sentence “leading to a sparse network with explanatory power”. What is meant by 'explanatory power'?

- Results, network structure and centrality, paragraph2: “The weights of the edges between the items SA1 (…) and SA2 (…) and the items SA5 (…) were stronger than others” – it is unclear which 2 pairs of nodes are the authors referring too. Also, the use of “stronger than others” in this context seems imprecise as this is not being tested.

- Results, network structure and centrality, paragraph3: “Inspecting the network structure” – the observations in this sentence do not follow from “inspection” of the network structure but instead of the centrality values. The visual layout of the network is rather arbitrary.

- Results, network structure and centrality, paragraph3: “strongly associated with SA5, the central node” please rephrase. The qualification “the central node” is imprecise. SA5 is the node with highest strength centrality.

- Results, network structure and centrality, paragraph3: “moderate association with SA9”. It seems that SA10 is meant here instead of SA9.

- Discussion paragraph 1: "findings from the present study could further our understanding of how smartphone addiction psychopathology is developed and maintained" - please reconsider. Cross-sectional network analysis cannot inform about dynamic aspects of mental disorders such as their development or maintenance.

- Discussion, paragraph 4: please consider rephrasing how you draw conclusions from your study to therapeutic approaches. It seems that the current formulation is too strong given the limitations of the current study design and implementation.

- Discussion, paragraph 4: There is a long description of current approaches to treat behavioral addictions. This is however disconnected from the current study’s results (and purposes) and therefore seems misplaced in the discussion.

- Discussion, paragraph 5: The explanation given about the differences between male and female subsample networks is not very sound in the opinion of this reviewer. It is unclear how “different psychological backgrounds” (a concept that is not very specific) would lead to “strong edge weights”. The remark about a supposed female ‘fear of missing out’ seems arbitrarily used. It should be discussed however if the differences in connectivity found could be attributed in the first place to female higher scores and a ‘floor-effect’ generated by lower scores on the male side.

- Conclusions: the use of “formation” of psychopathology is inaccurate because a cross-sectional network cannot inform about causality or dynamic patterns related to how psychopathology emerges.

- Conclusions: the sentence “Findings from this network analysis would provide us with deeper understandings of smartphone addiction” is too vague. In the end the reader struggles to find this study's take-away messages.

- English writing aspects – the reviewer suggests that the text is reviewed by a native English speaker. Here are however some suggestions from the reviewer:

-Introduction, second paragraph: the reviewer would suggest the authors to remove the sentence “The superb mobility and multifunction capability allows us to access the internet anytime and anywhere.

-Introduction, paragraph 3: “smartphone addiction is considered [A] behavioral addiction (A is lacking); “which a person is forced to engage – the use of “forced” seems too strong, please use a more adequate alternative;

-Introduction, paragraph 5: “including THE smartphone addiction scale” (THE is lacking)

-Introduction, paragraph 6: “behavioral addiction is complex human behaviors” should become “behavioral addiction is A complex human BEHAVIOR”

-Introduction, paragraph 7: “Rather, it is considered a ring as a complex network of mutually reinforcing symptoms” – this sentence is not clear, please rephrase, and consider changing the word “ring”.

-The section termed “analytic plans” should in this reviewer's view be renamed. Consider alternatives like “data analyses” or use “network analysis” as an umbrella that you then subdivide into subanalyses.

-Network estimation subsection: “represent individual signs (symptoms)”, consider using the term "item” as in this case the nodes represent items from an instrument.

-Centrality subsection: “sum of distances from the node to all other nodes):” use ";" instead of ":" here

-Centrality subsection: “we interpreted a symptom with the strongest strength as a central symptom in a given network” please rephrase

-Discussion, paragraph 1: “identified the important symptoms “. Rephrase paying attention to the fact that the use of THE important symptoms is inaccurate.

-Discussion, paragraph 3: “Huang and his colleagues conducted a network analysis of problematic smartphone use in grade 4 and 8 students in China using a self-rating scale, smartphone addiction proneness scale” – rewrite the end of the sentence

-“central indices” should be corrected to “centrality indices”

Reviewer #2: It is my pleasure to read this study with adequate sample and exact analysis. I have some suggestion for improving the manuscript.

1 Please include the theory or evidence for the addiction characteristics of excessive smart phone use in the introduction section.

2 The female sample is larger than male. Please explain the reason and detail for possible limitation, especially for the gender effect on smartphone use.

3 Since there is only ten item in the scale, please discuss whether the number of item will limit the analysis for the items.

4 Pleas explain the result of the figure 1 in the revsied manuscript

5 Although s4 and s5 are the central of the network, however, the score in these two item was relative lower. Please explain it.

6 The author had well demonstraed the implication of the result of the study. I agree with most of their claim. However, this content was not proved in this study. Thus, it should prevent to provide this content as it had been proved. Other reference should be provide to support their claim.

**********

6. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

Reviewer #2: No

[NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files.]

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email PLOS at figures@plos.org. Please note that Supporting Information files do not need this step.

1 Jun 2022

Our replies to reviewers’ comments

Reviewer #1: In this article the authors perform an exploratory network analysis of the answers to the smartphone addiction scale-short version of 487 Japanese college and university students. This included estimating the network and 3 centrality indices. The authors also proceed to divide the sample according to gender and estimate two sub-sample networks, testing for their differences. They found that preoccupation and withdrawal related items are particularly important in the network and also found differences between male and female subsample networks.

The topic of the article is of great importance and the use of network analysis could be of significance to further improve the knowledge about addiction disorders. The network analysis methodology is overall well applied. The findings about the most central symptoms could indeed be of interest to the field.

However, this reviewer thinks the study’s novelty is limited and also that it lacks a robust discussion of the results.

The overall quality of the writing is modest.

Major issues

- Study’s novelty: the cited study by Andrade, 2020 also performs a network analysis of the same instrument as the current study. More so, it frames it in a validation context of the instrument, making it a sounder scientific effort than the current one. The authors claim however that the fact that they’re using a slightly older sample justifies the study’s novelty. That difference seems to this reviewer as a very small one (in the end the results did not differ much, proving that the difference was not so relevant).

Our reply: Thank you for your comment and input. We acknowledge that the fact we focused on different age samples (transitional age youth) from ones in the extant study may not sound substantially novel. However, we would like to emphasize that understanding how this addictive behavior functions at different developmental age is important as the ownership of smartphones and degree of parental control and possibly school rules (for students who attend compulsory school, such as middle school and high school) can influence problematic smartphone behavior. We added a sentence and a reference below to state that maladaptive and excessive use of smartphones was identified more frequently in college students than middle and high school students in the revised manuscript to support our concept.

Discussion 1st paragraph lines 11 – 13: “In fact, the survey conducted in Japan exhibited more maladaptive and excessive use of smartphones in college students compared to middle and high school students [35]”

- Some questions about the use of this instrument (SAS-SV) should be clarified, in particular its validation and/or the validation of the original SAS instrument in Japan. Was that work done previously? In this reviewer’s perspective that would be a fundamental first step before engaging in a study like the present one (in the cited Brazilian study, a careful validation of the SAS-SV is performed alongside network analysis)

Our reply: The original instrument (SAS) was validated in Japanese youths in another study, where the authors reported the Japanese-version of SAS demonstrated good reliability and validity (DOI: 10.1007/s11469-021-00594-z). We added this statement in the revised manuscript. We acknowledge that the short-version (SAS-SV) has not validated in Japanese samples yet.

Lines 5-6 in Measures > Smartphone Addiction Scale-Short-version: “The Japanese-version of the SAS was developed, and its reliability and factor validity were assessed in 1,037 youth aged 15 – 24 years [28].”

- Regarding the sample, it represents a group of students from a specific region of Japan and with high gender distribution imbalance. This limits generalizability as the authors point. However, a more detailed description of the sample and its recruitment is necessary. For instance: when was the data collected? In which schools, from which specific region? Was there a rational for choosing this sample? Could the authors explain better the setting in which the questionnaire was filled? Were there potential participants who did not answer to the questionnaire? Simultaneously, a better sociodemographic description of these individuals should be provided – detailed school level description, age range (min and max values), and ideally some information that could inform the reader about socioeconomic status of this sample.

Our reply: We added descriptions of the study and study participants in the revised manuscript.

Participants subsection in the Material and methods section: “The subjects of the study were 487 private college and university students in Sapporo city and its environs in Japan. Study participation rate was 81.2% (487/600). Their academic deviation scores were average or a little below the average compared to the national standard in Japan. Research collaborators for data collection were recruited through personal connections of the first author of this paper (MT). Nine teachers from three universities and six colleges agreed to voluntarily support our investigation.”

We did not collect socioeconomic status or detailed characteristic information about participants except for that already reported in the manuscript.

- Regarding the study’s aims, the reviewer believes the network comparison performed is lacking a justification. The authors should explain why they believe it is relevant to compare networks based on a gender dichotomization of the sample. Also, one problem regarding this aim is the imbalance on sub sample size. Given that the starting point is already relatively modest, the authors are left with small samples for the sub-analysis, particularly the men’s network. Was the stability of each of the subsample networks analyzed?

Our reply: We agree with the reviewer’s point. We acknowledge that the comparison test was exploratory without sufficient scientific justification. Additionally, as pointed by the reviewer, even we conduct the network comparison test, the sample size differences between female and male students and the small sub-sample size of male students would result in skewed findings. Thus, the authors determined to omit this analysis and remove corresponding paragraphs from the manuscript.

- As a general comment, the discussion section should be improved. In the current form it stresses too much aspects that are limited by design in the current study (eg ideas about how addiction disorders are formed or should be treated).

Our reply: Thank you so much for your comments. We revised our manuscript, including discussion based on reviewers’ feedback.

- The writing should be improved. A native English speaker with experience in scientific writing could help improve substantially the article (see in minor issues some passages to reconsider).

Our reply: We asked a native English speaker who holds a doctor’s degree and is an acquaintance of one of the authors (MT) to proofread the manuscript.

- A justification for not providing the data should be given (PLoS Data policy); This reviewer also strongly encourages the authors to make data analysis information available as supplementary material (R code)

Our reply: We uploaded the R code use for the analysis in this study in Supplementary data 5.

Minor issues

- Smartphone addiction concept: the authors point the lack of consensus around this construct. The reviewer would ask the authors to consider the use of ‘Problematic Smartphone Use’ instead

Our reply: We agree with the authors’ suggestion. We changed the term “smartphone addiction” to “problematic smartphone” throughout the manuscript.

- On a more technical note about the network comparison performed, the authors determine that they will perform 3 different tests in their methods section. As van Borkulo et al point out in their paper explaining this procedure, the edge strength invariance test should only be performed if there is an a priori hypothesis about specific edges to directly test between the networks, or, in case there is no a priori hypothesis, it could be done as a follow-up to a significant difference in the structural invariance test. Therefore, in this reviewer’s opinion, since there is no a priori hypothesis for the authors, in the methods section it should be specified that this test (edge strength invariance) will only be performed if a difference in structural invariance is found. Then, in the results section, the network comparison tests report seems to be a bit unclear. The authors state that there was a “significant difference for the overall network structure” but explain this as meaning that the “nodes were more densely connected overall in female students than in male students”. This explanation refers to the invariant global strength aspect of the test and not the invariant network structure. So, this reviewer is confused about what is being reported – invariant network structure or invariant global strength? The edge weight comparisons reported afterwards should only be pursued in case network structure differs. (See: Van Borkulo, C. D., Boschloo, L., Kossakowski, J., Tio, P., Schoevers, R. A., Borsboom, D., & Waldorp, L. J. (2017). Comparing network structures on three aspects: A permutation test. )

Our reply: Thank you for suggested papers and your constructive comments. We agree with the reviewer’s points and concerns. In the revised manuscript, we determined not to conduct the network comparison test given no priori hypotheses driving this analysis and as the study findings from this analysis could be skewed and may lack the accuracy due to the small sub-sample size of male students and inequality of sample size between female and male students.

- The meaning of centrality indices should be rectified. On page 13 line 4, the way strength centrality is presented is not correct (ie, strength centrality does not translate the association of a node with “all other other nodes in the network” as stated. It is instead a measure of local influence). The meaning of betweenness and closeness centrality measures on cross-sectional data network analysis in the context of mental disorders has also been questioned. I would suggest the author to check references such as the following for this matter: Bringmann, L. F., Elmer, T., Epskamp, S., Krause, R. W., Schoch, D., Wichers, M., ... & Snippe, E. (2019). What do centrality measures measure in psychological networks?. Journal of abnormal psychology, 128(8), 892.

Our reply: The authors very much appreciate the reviewer’s feedback and suggested papers. We addressed this point in the revised manuscript (see the “Centrality” subsection in the “Material and methods” section).

- In the methods section it should be mentioned which option was chosen for the network layout.

Our reply: We used the Fruchterman-Reingold algorithm, which we added to the revised manuscript (the last sentence in the network estimation subsection under the Data analyses section).

- Table 1: the asterisks in the Age and SAS-SV lines are a bit confusing and seem unnecessary to this reviewer; the readability of the table could be improved if the “(mean±SD” parts were positioned without a line break and if some hierarchical visual cue could be added to subdivisions of “internet use (hrs) and “purpose”; the acronym SNS should be explained in the table legend.

Our reply: All addressed in the revised manuscript.

- Network estimation subsection from analytic plans: please use proper reference for qgraph package. A reference for the R statistical software is also lacking.

Our reply: The reference is now placed in the revised manuscript (#30). Thank you.

- Network estimation subsection from analytic plans: the sentence “leading to a sparse network with explanatory power”. What is meant by 'explanatory power'?

Our reply: We decided not to add the phrase pointed by the reviewer as the statement was vague and confusing.

- Results, network structure and centrality, paragraph2: “The weights of the edges between the items SA1 (…) and SA2 (…) and the items SA5 (…) were stronger than others” – it is unclear which 2 pairs of nodes are the authors referring too. Also, the use of “stronger than others” in this context seems imprecise as this is not being tested.

Our reply: Thank you for pointing this out. To clarify our intents, we revised the manuscript as follows (changes are highlighted in yellow): “The weights of the edges between the items SA1 (Miss planned work due to smartphone use) and SA2 (Difficulty focusing on assignments/work due to smartphone use) and those between the items SA5 (Feel impatient and uneasy when not holding my smartphone) were larger than others in the given network (0.49 and 0.41, respectively).”

- Results, network structure and centrality, paragraph3: “Inspecting the network structure” – the observations in this sentence do not follow from “inspection” of the network structure but instead of the centrality values. The visual layout of the network is rather arbitrary.

Our reply: We agree with the reviewer’s point. We deleted that phrase in the revised manuscript. Corresponding sentence in the revised manuscript is below: “SA6 (Have my smartphone on my mind even when not using it) was with the highest closeness and relatively high strength in centrality indices.”

- Results, network structure and centrality, paragraph3: “strongly associated with SA5, the central node” please rephrase. The qualification “the central node” is imprecise. SA5 is the node with highest strength centrality.

Our reply: Thank you for this feedback. We corrected the corresponding sentence as follows in the revised manuscript.

“In particular, while this item was strongly associated with SA5, it also had the moderate association with SA10 (People around me tell me that I use my smartphone too much) that was only weakly associated with the node with the highest strength centrality (SA5).”

- Results, network structure and centrality, paragraph3: “moderate association with SA9”. It seems that SA10 is meant here instead of SA9.

Our reply: It was a typo. The reviewer was right. We corrected this error in the revised manuscript.

- Discussion paragraph 1: "findings from the present study could further our understanding of how smartphone addiction psychopathology is developed and maintained" - please reconsider. Cross-sectional network analysis cannot inform about dynamic aspects of mental disorders such as their development or maintenance.

- Discussion, paragraph 4: please consider rephrasing how you draw conclusions from your study to therapeutic approaches. It seems that the current formulation is too strong given the limitations of the current study design and implementation.

- Discussion, paragraph 4: There is a long description of current approaches to treat behavioral addictions. This is however disconnected from the current study’s results (and purposes) and therefore seems misplaced in the discussion.

- Discussion, paragraph 5: The explanation given about the differences between male and female subsample networks is not very sound in the opinion of this reviewer. It is unclear how “different psychological backgrounds” (a concept that is not very specific) would lead to “strong edge weights”. The remark about a supposed female ‘fear of missing out’ seems arbitrarily used. It should be discussed however if the differences in connectivity found could be attributed in the first place to female higher scores and a ‘floor-effect’ generated by lower scores on the male side.

- Conclusions: the use of “formation” of psychopathology is inaccurate because a cross-sectional network cannot inform about causality or dynamic patterns related to how psychopathology emerges.

- Conclusions: the sentence “Findings from this network analysis would provide us with deeper understandings of smartphone addiction” is too vague. In the end the reader struggles to find this study's take-away messages.

- English writing aspects – the reviewer suggests that the text is reviewed by a native English speaker. Here are however some suggestions from the reviewer:

-Introduction, second paragraph: the reviewer would suggest the authors to remove the sentence “The superb mobility and multifunction capability allows us to access the internet anytime and anywhere.

-Introduction, paragraph 3: “smartphone addiction is considered [A] behavioral addiction (A is lacking); “which a person is forced to engage – the use of “forced” seems too strong, please use a more adequate alternative;

-Introduction, paragraph 5: “including THE smartphone addiction scale” (THE is lacking)

-Introduction, paragraph 6: “behavioral addiction is complex human behaviors” should become “behavioral addiction is A complex human BEHAVIOR”

-Introduction, paragraph 7: “Rather, it is considered a ring as a complex network of mutually reinforcing symptoms” – this sentence is not clear, please rephrase, and consider changing the word “ring”.

-The section termed “analytic plans” should in this reviewer's view be renamed. Consider alternatives like “data analyses” or use “network analysis” as an umbrella that you then subdivide into subanalyses.

-Network estimation subsection: “represent individual signs (symptoms)”, consider using the term "item” as in this case the nodes represent items from an instrument.

-Centrality subsection: “sum of distances from the node to all other nodes):” use ";" instead of ":" here

-Centrality subsection: “we interpreted a symptom with the strongest strength as a central symptom in a given network” please rephrase

-Discussion, paragraph 1: “identified the important symptoms “. Rephrase paying attention to the fact that the use of THE important symptoms is inaccurate.

-Discussion, paragraph 3: “Huang and his colleagues conducted a network analysis of problematic smartphone use in grade 4 and 8 students in China using a self-rating scale, smartphone addiction proneness scale” – rewrite the end of the sentence

-“central indices” should be corrected to “centrality indices”

Our reply: Thank you so much for these suggestions. We revised our manuscript based on these suggestions (corrected parts are highlighted in yellow in the revised manuscript). In addition, as stated above in our letter, we asked a native English speaker to proofread the revised manuscript.

Reviewer #2: It is my pleasure to read this study with adequate sample and exact analysis. I have some suggestion for improving the manuscript.

1 Please include the theory or evidence for the addiction characteristics of excessive smart phone use in the introduction section.

Our reply: As pointed out by the other reviewer, due to the lack of consensus on the constructs of smartphone addiction and no compelling theory supporting the concept of smartphone addiction at this point, we decided to use “problematic smartphone use” instead throughout the revised manuscript.

2 The female sample is larger than male. Please explain the reason and detail for possible limitation, especially for the gender effect on smartphone use.

Our reply: As answered above to respond to the other reviewer, we determined not to conduct the network comparison test as the study findings from this analysis could be skewed and may lack the accuracy due to the small sub-sample size of male students and inequality of sample size between female and male students.

3 Since there is only ten item in the scale, please discuss whether the number of item will limit the analysis for the items.

Our reply: The findings from network analysis depend on what items are entered in analysis; however, this does not mean the number of items in the scale influence accuracy and stability of network analysis findings.

4 Pleas explain the result of the figure 1 in the revsied manuscript

Our reply: Thank you for this comment. Figure 1 is only for visualization of network analytic findings. As discussed in other network analysis literature, it is recommended we discuss findings obtained from centrality analyses (figure 2) as visualization per se may be arbitrary. General description about network visualization is placed as a figure legend “Figure 1. Regularized partial correlation network of Smartphone Addiction Scale Short-Version in college and university students in Japan (N = 478). Green lines indicate a positive association. Line thickness reflects the strength of association, controlling for all other symptom nodes in the network.”

5 Although s4 and s5 are the central of the network, however, the score in these two item was relative lower. Please explain it.

Our reply: Strength centrality is a measure that reflects the degree to which each node is connected to other nodes in the network. Thus, the degree of centrality differs from the degree of item score.

6 The author had well demonstraed the implication of the result of the study. I agree with most of their claim. However, this content was not proved in this study. Thus, it should prevent to provide this content as it had been proved. Other reference should be provide to support their claim.

Our reply: Thank you for this comment. We agree with the reviewer’s point that the paragraph pertaining to the implication of the result of the study was not stated based on scientifically sound hypotheses. To avoid any confusion and to make the flow of the manuscript more coherent, we deleted this paragraph in the revised manuscript.

Submitted filename: Authors reply to reviewers_PlosOne.docx

Click here for additional data file.

30 Jun 2022

A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). You should upload this letter as a separate file labeled 'Response to Reviewers'.

A marked-up copy of your manuscript that highlights changes made to the original version. You should upload this as a separate file labeled 'Revised Manuscript with Track Changes'.

An unmarked version of your revised paper without tracked changes. You should upload this as a separate file labeled 'Manuscript'.

If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter. Guidelines for resubmitting your figure files are available below the reviewer comments at the end of this letter.

If applicable, we recommend that you deposit your laboratory protocols in protocols.io to enhance the reproducibility of your results. Protocols.io assigns your protocol its own identifier (DOI) so that it can be cited independently in the future. For instructions see: https://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols. Additionally, PLOS ONE offers an option for publishing peer-reviewed Lab Protocol articles, which describe protocols hosted on protocols.io. Read more information on sharing protocols at https://plos.org/protocols?utm_medium=editorial-email&utm_source=authorletters&utm_campaign=protocols.

We look forward to receiving your revised manuscript.

Kind regards,

Roland Bouffanais, Ph.D.

Academic Editor

PLOS ONE

Additional Editor Comments (if provided):

We have now received reports from the same two reviewers, and after careful consideration, we have decided to invite another major revision of the manuscript.

As you will see from the enclosed reports, one reviewer still raises important concerns. We find that these concerns limit the strength of the study, and therefore we ask you to address them with additional work.

If you feel that you are able to comprehensively address the reviewer’s concerns, please provide a point-by-point response to these comments along with your revision. Please show all changes in the manuscript text file with track changes or color highlighting. If you are unable to address specific reviewer requests or find any points invalid, please explain why in the point-by-point response.

[Note: HTML markup is below. Please do not edit.]

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. If the authors have adequately addressed your comments raised in a previous round of review and you feel that this manuscript is now acceptable for publication, you may indicate that here to bypass the “Comments to the Author” section, enter your conflict of interest statement in the “Confidential to Editor” section, and submit your "Accept" recommendation.

Reviewer #1: (No Response)

Reviewer #2: All comments have been addressed

**********

2. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: Yes

Reviewer #2: Yes

**********

3. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: Yes

Reviewer #2: Yes

**********

4. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: Yes

Reviewer #2: (No Response)

**********

5. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

Reviewer #2: Yes

**********

6. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: The authors conducted several changes in the manuscript that contributed to a much stronger work. Namely, the writing quality is now much improved, the analyses performed and their justification became clearer, and the availability of data and code now made this a transparent and potentially replicable scientific work.

There are some aspects though that I believe are still in need of revision in order for the manuscript to be acceptable for publication.

One general point is that, now that some unnecessary remarks were removed from the discussion, this section should be improved by adding thoughts about some of the issues the manuscript raises (see suggestions below).

(I acknowledge that some of my current minor points address issues already present in the first draft of the manuscript. Given the very high amount of changes required for the first revision, these points seemed less relevant then and/or are now in more clear need of revision given the other changes made to the article.)

Major issues

- In the abstract it is still mentioned that gender differences were explored in the network. This should be removed as that analysis is no longer performed.

- Table 1 has separate columns for male and female participants. Given that there is no longer a network comparison according to gender, could the authors state why this is still relevant? I would also like to know the authors’ opinion about 2 further aspects related to this, that might be relevant for the discussion: 1 ) why is it that ¾ of the sample are females (Is this something that mirrors the high-school/university demographics from Japan or is it a selection bias?) and how could that imbalance bias the results? 2) how could the different smartphone usages (reported in table 1) affect the results (e.g. can we consider this study as mostly representing social network use disorder? Could different usages lead to different network structures hypothetically? Would this factor be more important than gender itself?). I believe these aspects are very important to be considered in the discussion and would enrich it very much.

Minor issues

- Abstract (and results too): I believe it is confusing to state “The estimated network yielded 45 edges, among which 34 edges had non- zero weights”. What are the 0 weight edges? Maybe just state that you’ve found 34 edges.

- Intro (last sentence of the 2nd to last paragraph) “population would not arguably be influenced by family”. I suggest that you consider stating something like “would not be directly influenced by family”, given that how a family raised the subject would continue to influence him/her afterwards.

- Intro (last paragraph): Second aim “identify which items could play important roles” – to play an important role is already an extrapolation. What we can do is to identify the most central symptoms in the network (hence, as you say before, the symptoms that potentially play a critical role in the onset and maintenance of the disorder). Consider rephrasing.

- Table 1 – consider reducing the decimal places to 3, but check exactly the journal guidelines on this. P-values less than 0.001 are usually reported as p<0.001.

- Is there some information available about the nearly 200 students who did not answer the questionnaire? For instance, about gender (were they predominantly male? this could explain your sample bias).

- Methods: citation for the qgraph package should be added (in data analyses - network estimation) https://cran.r-project.org/web/packages/qgraph/citation.html#:~:text=qgraph%20citation%20info,4)%2C%201%E2%80%9318.

- Methods: please consider adding succinctly what the Fruchterman-Reingold algorithm does in terms of network layout (readers not acquainted with network methods might find it intriguing)

- Data analyses->centrality (first sentence): “We calculated several indices” – please consider rephrasing to “calculated three indices”

- Data analyses->centrality (second sentence): “a symptom with the highest value of strength was defined as a central symptom in a given network”. Consider rephrasing or removing this sentence as it is self-evident that the centrality is higher according to how high strength centrality is.

- Data analyses->centrality (last sentence): “These are typically presented as standardized Z-scores”. Consider clarifying – “these” -> centrality values

- On the first subparagraph of the results section, the text repeats what can be already found in table 1. I would strongly recommend that the authors simply refer to the table for the information that can already be found there.

- Results -> network structure and stability (first sentence): “relations among SA symptoms”. I believe you dropped the SA terminology, so please consider changing it consistently in the manuscript. This includes changing the node names in the figures. For example, instead of using SA#, consider simply using the number 1-10 (as you do in table 2).

- Results -> network structure and stability (second paragraph, first sentence): “The weights of the edges between the items SA1 (…) were larger than others in the given network“. Please consider rephrasing as this does not seem to have been specifically tested (and you mentioned there is considerable overlap in confidence intervals).

- Discussion: the term ‘smartphone addiction’ is sometimes used here in a way that is not completely harmonized within the manuscript given that there was an intention to substitute this term.

- Discussion (second paragraph, first sentence): “the Brazilian study” – please consider changing this. Would prefer either that you cite the authors name or that you refer to the “study conducted with [demographic group] in Brazil”

- Discussion (second paragraph, third sentence): “to produce [the] addiction cycle”

- Discussion (second paragraph, fourth sentence): consider enriching the discussion by explaining how you believe overuse symptoms could still “contribute to either the development or maintenance of the network” even though your results do not seem to show them as central.

- Discussion (third paragraph, third sentence): “Differences in the symptom centrality between the study above and our study may be attributed to the difference in the scale used for each study.” Please consider enriching your discussion by stating your thoughts on why that is the case. What are the differences between the instruments?

- Discussion (third paragraph, last sentence): “Thus, such smartphone users could have difficulties leaving their smartphones out of reach and feel anxious without the smartphone in close proximity.”. Good point. Consider making explicit that this is why you believe that an older population (like the one in your study) might have withdrawal symptoms as more central in the network than overuse symptoms.

- Figure 2 legend: “with higher scores indicative of greater importance within the overall network.” Same as stated above. Consider refraining from using the term “importance” as this is an extrapolation.

Reviewer #2: I appreciate the response of the authors. The manuscript had been well revised. It is fine to be published.

**********

7. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: Yes: Bernardo Melo Moura

Reviewer #2: No

**********

[NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files.]

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email PLOS at figures@plos.org. Please note that Supporting Information files do not need this step.

17 Jul 2022

Our responses to the reviewer's comments are summarized in the document and uploaded as a separate file (entitled as "Response to the reviewer_PLOSONE_R2"). Thank you.

Submitted filename: Response to the reviewer_PLOSONE_R2.docx

Click here for additional data file.

27 Jul 2022

A Network Analysis of Problematic Smartphone Use in Japanese young adults

PONE-D-21-32886R2

Dear Dr. Hirota,

We’re pleased to inform you that your manuscript has been judged scientifically suitable for publication and will be formally accepted for publication once it meets all outstanding technical requirements.

Within one week, you’ll receive an e-mail detailing the required amendments. When these have been addressed, you’ll receive a formal acceptance letter and your manuscript will be scheduled for publication.

An invoice for payment will follow shortly after the formal acceptance. To ensure an efficient process, please log into Editorial Manager at http://www.editorialmanager.com/pone/, click the 'Update My Information' link at the top of the page, and double check that your user information is up-to-date. If you have any billing related questions, please contact our Author Billing department directly at authorbilling@plos.org.

If your institution or institutions have a press office, please notify them about your upcoming paper to help maximize its impact. If they’ll be preparing press materials, please inform our press team as soon as possible -- no later than 48 hours after receiving the formal acceptance. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information, please contact onepress@plos.org.

Kind regards,

Roland Bouffanais, Ph.D.

Academic Editor

PLOS ONE

Additional Editor Comments (optional):

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. If the authors have adequately addressed your comments raised in a previous round of review and you feel that this manuscript is now acceptable for publication, you may indicate that here to bypass the “Comments to the Author” section, enter your conflict of interest statement in the “Confidential to Editor” section, and submit your "Accept" recommendation.

Reviewer #1: All comments have been addressed

Reviewer #2: All comments have been addressed

**********

2. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: Yes

Reviewer #2: Yes

**********

3. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: Yes

Reviewer #2: Yes

**********

4. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: Yes

Reviewer #2: Yes

**********

5. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

Reviewer #2: Yes

**********

6. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: The authors' revision of the manuscript contributed again to improve substantially their work.

There are 4-5 minor points I will still list below for the authors to revise. Granted that attention is paid to these points, I do not believe a further review is necessary, thus the paper is in my opinion good enough for publication.

Minor points

- Pay attention to the fact that the abstract you provide in the box section at the beginning does not exactly match the one present on the manuscript file you’ve attached

- Abstract, Results: I would suggest the following change in writing: “We identified 34 edges in the estimated network. One item pertaining to withdrawal symptoms had the highest strength and also high closeness centrality.

- Results, characteristics of the study participants: also for male students add years of age “(…) male students (mean years of age: …”

- Results, network structure and stability: Use of “SA”. Consider using “SAS-SV items” instead of “SA symptoms”

- Results, network structure and stability: I did not spot this before, but it is important to correct: “and those between the items SA5 (…)” it seems like the other node is missing, please add mention to it.

Reviewer #2: (No Response)

**********

7. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: Yes: Bernardo Melo Moura

Reviewer #2: No

**********

29 Jul 2022

PONE-D-21-32886R2

A Network Analysis of Problematic Smartphone Use in Japanese young adults

Dear Dr. Hirota:

I'm pleased to inform you that your manuscript has been deemed suitable for publication in PLOS ONE. Congratulations! Your manuscript is now with our production department.

If your institution or institutions have a press office, please let them know about your upcoming paper now to help maximize its impact. If they'll be preparing press materials, please inform our press team within the next 48 hours. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information please contact onepress@plos.org.

If we can help with anything else, please email us at plosone@plos.org.

Thank you for submitting your work to PLOS ONE and supporting open access.

Kind regards,

PLOS ONE Editorial Office Staff

on behalf of

Professor Roland Bouffanais

Academic Editor

PLOS ONE