Pub Date : 2025-12-13DOI: 10.1016/j.chbr.2025.100902
Jeahong Kim , Minwoo Jo , Sooleen Nam , Yujin Lee , Yeonji Baik
Current study examined the validity of a digitalized Person-in-the-Rain test (dPITR) in assessing stress and coping abilities using traditional univariate and machine learning methods. By analyzing drawings and comparing results with established self-reported measures (Perceived Stress Scale and Ways of Coping Questionnaire), the study identifies significant relationships between drawing features and stress and coping scores. Machine learning models (K-Nearest Neighbor, Support Vector Machine, Random Forest) revealed complex, non-linear patterns, with stress-related features showing significant predictive power for perceived stress and coping styles. The findings support dPITR as a reliable, non-invasive tool for assessing psychological constructs and highlight the value of integrating advanced analytics in psychological assessments. This digital approach offers promising applications for scalable and precise mental health evaluations.
{"title":"Decoding stress and coping ability levels through drawing features in the Person-In-The-Rain test: An exploratory study using a digital drawing device","authors":"Jeahong Kim , Minwoo Jo , Sooleen Nam , Yujin Lee , Yeonji Baik","doi":"10.1016/j.chbr.2025.100902","DOIUrl":"10.1016/j.chbr.2025.100902","url":null,"abstract":"<div><div>Current study examined the validity of a digitalized Person-in-the-Rain test (dPITR) in assessing stress and coping abilities using traditional univariate and machine learning methods. By analyzing drawings and comparing results with established self-reported measures (Perceived Stress Scale and Ways of Coping Questionnaire), the study identifies significant relationships between drawing features and stress and coping scores. Machine learning models (K-Nearest Neighbor, Support Vector Machine, Random Forest) revealed complex, non-linear patterns, with stress-related features showing significant predictive power for perceived stress and coping styles. The findings support dPITR as a reliable, non-invasive tool for assessing psychological constructs and highlight the value of integrating advanced analytics in psychological assessments. This digital approach offers promising applications for scalable and precise mental health evaluations.</div></div>","PeriodicalId":72681,"journal":{"name":"Computers in human behavior reports","volume":"21 ","pages":"Article 100902"},"PeriodicalIF":5.8,"publicationDate":"2025-12-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145737034","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-12-09DOI: 10.1016/j.chbr.2025.100903
Isaac Owusu Asante , Muhammad Ali , Feng Liu
This study examines the impact of user-generated ratings, sentiment analysis of reviews, reports of symptom improvement, and perceived satisfaction on treatment efficacy in mobile health (mHealth) platforms. Employing a mixed-methods approach, we examined text reviews and ratings from WeDoctor, a prominent Chinese website, integrating computational sentiment analysis with hierarchical regression modeling. The results indicated that comment scores exerted a more significant direct influence on treatment efficiency than sentiment scores; nevertheless, sentiment scores had a crucial indirect effect in enhancing satisfaction. The mention of symptom improvement considerably moderated the impact of comment scores, but not sentiment, indicating that clinical outcomes corroborate quantitative assessments more than emotional expressions. Using a BERT-based sentiment classification model pretrained on Chinese-language reviews, we analyzed the emotional tone of 15000 user-generated consultations. Results showed that 97 % of reviews were classified as positive, indicating a high degree of emotional satisfaction and trust in mHealth service experiences. This affective signal was then integrated with structured ratings and symptom improvement cues to model perceived treatment efficiency. The mixed-methods design leverages both structured and unstructured data, offering a multidimensional perspective that extends traditional Expectation Confirmation Theory to digital healthcare. By combining cognitive, affective, and clinical dimensions of user feedback, the study provides a multifaceted understanding of perceived efficiency. It informs the future design of emotionally intelligent and healthcare-grounded mHealth interfaces, aligning with key priorities in human-computer interaction research.
{"title":"Assessing the influence of mobile health platform ratings and sentiments on perceived treatment efficiency: A data-driven analysis","authors":"Isaac Owusu Asante , Muhammad Ali , Feng Liu","doi":"10.1016/j.chbr.2025.100903","DOIUrl":"10.1016/j.chbr.2025.100903","url":null,"abstract":"<div><div>This study examines the impact of user-generated ratings, sentiment analysis of reviews, reports of symptom improvement, and perceived satisfaction on treatment efficacy in mobile health (mHealth) platforms. Employing a mixed-methods approach, we examined text reviews and ratings from WeDoctor, a prominent Chinese website, integrating computational sentiment analysis with hierarchical regression modeling. The results indicated that comment scores exerted a more significant direct influence on treatment efficiency than sentiment scores; nevertheless, sentiment scores had a crucial indirect effect in enhancing satisfaction. The mention of symptom improvement considerably moderated the impact of comment scores, but not sentiment, indicating that clinical outcomes corroborate quantitative assessments more than emotional expressions. Using a BERT-based sentiment classification model pretrained on Chinese-language reviews, we analyzed the emotional tone of 15000 user-generated consultations. Results showed that 97 % of reviews were classified as positive, indicating a high degree of emotional satisfaction and trust in mHealth service experiences. This affective signal was then integrated with structured ratings and symptom improvement cues to model perceived treatment efficiency. The mixed-methods design leverages both structured and unstructured data, offering a multidimensional perspective that extends traditional Expectation Confirmation Theory to digital healthcare. By combining cognitive, affective, and clinical dimensions of user feedback, the study provides a multifaceted understanding of perceived efficiency. It informs the future design of emotionally intelligent and healthcare-grounded mHealth interfaces, aligning with key priorities in human-computer interaction research.</div></div>","PeriodicalId":72681,"journal":{"name":"Computers in human behavior reports","volume":"21 ","pages":"Article 100903"},"PeriodicalIF":5.8,"publicationDate":"2025-12-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145737036","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-12-06DOI: 10.1016/j.chbr.2025.100897
Tingting Gao , Sihan Lyu , Fengtong Qian , Rui Li , Yimeng Lyu , Yingying Su
Background
Prior studies examining the association between deviant peer affiliation and phubbing have predominantly relied on latent variable modeling. However, the directionality, symptom-level pathways, and sex-specific patterns of their longitudinal associations remain largely unexplored within a network-theoretical framework. To address these gaps, the present study aims to clarify how specific symptoms of deviant peer affiliation and phubbing mutually influence one another over time, while further investigating whether these temporal dynamics differ by sex.
Methods
A total of 3296 adolescents who participated in two waves of a longitudinal survey were included. Cross-lagged panel network (CLPN) analysis was conducted to clarify the temporal associations between deviant peer affiliation symptoms and phubbing symptoms across a 1.5-year follow-up.
Results
Sex-stratified contemporaneous networks revealed distinct patterns in how deviant peer affiliation and phubbing were organized among male and female adolescents. Specific deviant peer behaviors such as friends steal things (DPA4) among males and friends skip class (DPA7) among females predict subsequent increases in phubbing. In addition, friend's internet addiction (DPA5) and relieving stress by focusing on phones (GSP12) were the most influential symptoms in the male network, while friend smoking (DPA1) and friend bullying others (DPA8) played similarly central roles in the female network.
Conclusion
These findings highlight the significance of sex-sensitive interventions and valuable perspectives for addressing problematic behaviors in adolescents. By identifying symptom-level pathways linking deviant peer affiliation and phubbing, this study provides novel insights and contributes to a better understanding of the mechanisms underlying these behaviors.
{"title":"A cross-lagged prospective network analysis of deviant peer affiliation and phubbing in adolescents","authors":"Tingting Gao , Sihan Lyu , Fengtong Qian , Rui Li , Yimeng Lyu , Yingying Su","doi":"10.1016/j.chbr.2025.100897","DOIUrl":"10.1016/j.chbr.2025.100897","url":null,"abstract":"<div><h3>Background</h3><div>Prior studies examining the association between deviant peer affiliation and phubbing have predominantly relied on latent variable modeling. However, the directionality, symptom-level pathways, and sex-specific patterns of their longitudinal associations remain largely unexplored within a network-theoretical framework. To address these gaps, the present study aims to clarify how specific symptoms of deviant peer affiliation and phubbing mutually influence one another over time, while further investigating whether these temporal dynamics differ by sex.</div></div><div><h3>Methods</h3><div>A total of 3296 adolescents who participated in two waves of a longitudinal survey were included. Cross-lagged panel network (CLPN) analysis was conducted to clarify the temporal associations between deviant peer affiliation symptoms and phubbing symptoms across a 1.5-year follow-up.</div></div><div><h3>Results</h3><div>Sex-stratified contemporaneous networks revealed distinct patterns in how deviant peer affiliation and phubbing were organized among male and female adolescents. Specific deviant peer behaviors such as <em>friends steal things</em> (DPA4) among males and <em>friends skip class</em> (DPA7) among females predict subsequent increases in phubbing. In addition, <em>friend's internet addiction</em> (DPA5) and <em>relieving stress by focusing on phones</em> (GSP12) were the most influential symptoms in the male network<em>,</em> while <em>friend smoking</em> (DPA1) <em>and friend bullying others</em> (DPA8) played similarly central roles in the female network.</div></div><div><h3>Conclusion</h3><div>These findings highlight the significance of sex-sensitive interventions and valuable perspectives for addressing problematic behaviors in adolescents. By identifying symptom-level pathways linking deviant peer affiliation and phubbing, this study provides novel insights and contributes to a better understanding of the mechanisms underlying these behaviors.</div></div>","PeriodicalId":72681,"journal":{"name":"Computers in human behavior reports","volume":"21 ","pages":"Article 100897"},"PeriodicalIF":5.8,"publicationDate":"2025-12-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145685533","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-12-05DOI: 10.1016/j.chbr.2025.100901
Bongsu Kang , Jundong Kim , Taerim Yun , Hyojin Bae , Chang-Eop Kim
This study quantitatively examines which features of AI-generated text lead humans to perceive subjective consciousness in large language model (LLM)-based AI systems. Drawing on 99 passages from conversations with AI and focusing on eight features—Metacognitive Self-reflection, Logical Reasoning, Empathy, Emotionality, Knowledge, Fluency, Unexpectedness, and Subjective Expressiveness—we surveyed with 123 participants. Using regression and clustering analyses, we investigated how these features influence participants' perceptions of AI consciousness. The results reveal that metacognitive self-reflection and the AI's expression of its own emotions significantly increased perceived consciousness, while a heavy emphasis on knowledge reduced it. Participants clustered into subgroups, each showing distinct feature-weighting patterns. Additionally, higher prior knowledge of LLMs and more frequent usage of LLM-based chatbots were associated with greater overall likelihood assessments of AI consciousness. This study underscores the multidimensional and individualized nature of perceived AI consciousness and provides a foundation for a better understanding of the psychosocial implications of human-AI interaction.
{"title":"Identifying features that shape perceived consciousness in LLM-based AI: A quantitative study of human responses","authors":"Bongsu Kang , Jundong Kim , Taerim Yun , Hyojin Bae , Chang-Eop Kim","doi":"10.1016/j.chbr.2025.100901","DOIUrl":"10.1016/j.chbr.2025.100901","url":null,"abstract":"<div><div>This study quantitatively examines which features of AI-generated text lead humans to perceive subjective consciousness in large language model (LLM)-based AI systems. Drawing on 99 passages from conversations with AI and focusing on eight features—Metacognitive Self-reflection, Logical Reasoning, Empathy, Emotionality, Knowledge, Fluency, Unexpectedness, and Subjective Expressiveness—we surveyed with 123 participants. Using regression and clustering analyses, we investigated how these features influence participants' perceptions of AI consciousness. The results reveal that metacognitive self-reflection and the AI's expression of its own emotions significantly increased perceived consciousness, while a heavy emphasis on knowledge reduced it. Participants clustered into subgroups, each showing distinct feature-weighting patterns. Additionally, higher prior knowledge of LLMs and more frequent usage of LLM-based chatbots were associated with greater overall likelihood assessments of AI consciousness. This study underscores the multidimensional and individualized nature of perceived AI consciousness and provides a foundation for a better understanding of the psychosocial implications of human-AI interaction.</div></div>","PeriodicalId":72681,"journal":{"name":"Computers in human behavior reports","volume":"21 ","pages":"Article 100901"},"PeriodicalIF":5.8,"publicationDate":"2025-12-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145737032","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-12-05DOI: 10.1016/j.chbr.2025.100899
Mochammad Hannats Hanafi Ichsan , Cecilia Sik-Lanyi , Tibor Guzsvinecz , Aisshah Roesiana Dewi
This study examines the usability, user experience, and cognitive performance of an IntuNav that incorporates a Multi-Browser Virtual Environment (VE), a user-friendly desktop virtual reality (VR) system. The evaluation included three user groups that varied in characteristics: mainstream users, older adults, and students with neurodivergent conditions (Autistic Spectrum Disorders and Attention Deficit Hyperactivity Disorder). Fourteen hypotheses were developed to investigate differences in objective (5Q score, task time, error count, and perplexity) and subjective (SUS, IPQ, and NASA-TLX) metrics using a between-subjects experimental design. Statistical analyses indicated no significant differences in core performance metrics (5Q scores, error count) among groups, implying the system's overall usability. Significant variations in task time and perplexity were observed between older adults and neurodivergent users compared to mainstream users, highlighting the impact of cognitive and generational factors on navigational complexity. Older adults exhibited the highest subjective usability and presence scores, whereas cognitive load levels were elevated among older and neurodiverse users. The results indicate that the IntuNav navigation model and Multi-Browser VE provide inclusive and accessible desktop VR interaction for a diverse user base. This demonstrates the system's practical applicability in contexts necessitating multi-window VR interaction, including education, research, and digital productivity. Design recommendations are presented to enhance inclusivity, minimize cognitive demands, and improve adaptive navigation in future VR systems. An anonymized dataset and complete evaluation scripts are publicly accessible (OSF: 10.17605/OSF.IO/HU478), along with implementation resources (GitHub), which allows for reproducibility.
{"title":"User-centered evaluation of an IntuNav in multi-browser virtual reality across diverse cognitive user profiles","authors":"Mochammad Hannats Hanafi Ichsan , Cecilia Sik-Lanyi , Tibor Guzsvinecz , Aisshah Roesiana Dewi","doi":"10.1016/j.chbr.2025.100899","DOIUrl":"10.1016/j.chbr.2025.100899","url":null,"abstract":"<div><div>This study examines the usability, user experience, and cognitive performance of an IntuNav that incorporates a Multi-Browser Virtual Environment (VE), a user-friendly desktop virtual reality (VR) system. The evaluation included three user groups that varied in characteristics: mainstream users, older adults, and students with neurodivergent conditions (Autistic Spectrum Disorders and Attention Deficit Hyperactivity Disorder). Fourteen hypotheses were developed to investigate differences in objective (5Q score, task time, error count, and perplexity) and subjective (SUS, IPQ, and NASA-TLX) metrics using a between-subjects experimental design. Statistical analyses indicated no significant differences in core performance metrics (5Q scores, error count) among groups, implying the system's overall usability. Significant variations in task time and perplexity were observed between older adults and neurodivergent users compared to mainstream users, highlighting the impact of cognitive and generational factors on navigational complexity. Older adults exhibited the highest subjective usability and presence scores, whereas cognitive load levels were elevated among older and neurodiverse users. The results indicate that the IntuNav navigation model and Multi-Browser VE provide inclusive and accessible desktop VR interaction for a diverse user base. This demonstrates the system's practical applicability in contexts necessitating multi-window VR interaction, including education, research, and digital productivity. Design recommendations are presented to enhance inclusivity, minimize cognitive demands, and improve adaptive navigation in future VR systems. An anonymized dataset and complete evaluation scripts are publicly accessible (OSF: 10.17605/OSF.IO/HU478), along with implementation resources (GitHub), which allows for reproducibility.</div></div>","PeriodicalId":72681,"journal":{"name":"Computers in human behavior reports","volume":"21 ","pages":"Article 100899"},"PeriodicalIF":5.8,"publicationDate":"2025-12-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145685534","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-12-05DOI: 10.1016/j.chbr.2025.100900
Lazaros Michailidis, Jesus Lucas Barcias
The prolific consumption of video games has inspired a rich body of research to explain their ability to induce satisfying experiences. The flow experience is often defined as a state that encapsulates the essential qualities of engaging gameplay. However, there is limited research addressing the detection of flow, and existing approaches often involve intricate equipment requirements that compromise commercial applicability. In this study, we review the challenges behind flow's conceptualization and discuss how they leak into approaches that employ machine learning techniques. We propose a novel, multidimensional representation of flow that combines subjective and objective data – game controller interaction and player performance on a secondary oddball task – whilst treating flow detection as a regression problem. Contrary to previous studies, the game difficulty is divorced from the model's training, with the aim of improving generalizability across game titles. The results indicate high predictive accuracy with an average composite loss rate of 0.0681 (±0.0036) that significantly outperformed two baseline models. This finding suggests an effective mapping of the objective data onto self-reported flow and reinforces the viability of game controllers as a means for flow detection. In addition, we identified that the main drivers behind the model's predictions were of inertial origin. We conclude that these insights offer practical considerations for the future of interactive applications.
{"title":"Hold me tight: Towards the detection of the flow experience using a commercial video game controller","authors":"Lazaros Michailidis, Jesus Lucas Barcias","doi":"10.1016/j.chbr.2025.100900","DOIUrl":"10.1016/j.chbr.2025.100900","url":null,"abstract":"<div><div>The prolific consumption of video games has inspired a rich body of research to explain their ability to induce satisfying experiences. The flow experience is often defined as a state that encapsulates the essential qualities of engaging gameplay. However, there is limited research addressing the detection of flow, and existing approaches often involve intricate equipment requirements that compromise commercial applicability. In this study, we review the challenges behind flow's conceptualization and discuss how they leak into approaches that employ machine learning techniques. We propose a novel, multidimensional representation of flow that combines subjective and objective data – game controller interaction and player performance on a secondary oddball task – whilst treating flow detection as a regression problem. Contrary to previous studies, the game difficulty is divorced from the model's training, with the aim of improving generalizability across game titles. The results indicate high predictive accuracy with an average composite loss rate of 0.0681 (±0.0036) that significantly outperformed two baseline models. This finding suggests an effective mapping of the objective data onto self-reported flow and reinforces the viability of game controllers as a means for flow detection. In addition, we identified that the main drivers behind the model's predictions were of inertial origin. We conclude that these insights offer practical considerations for the future of interactive applications.</div></div>","PeriodicalId":72681,"journal":{"name":"Computers in human behavior reports","volume":"21 ","pages":"Article 100900"},"PeriodicalIF":5.8,"publicationDate":"2025-12-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145737033","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-12-04DOI: 10.1016/j.chbr.2025.100898
Xue Yang , Xu Chen , Qian Li , Yilun Huang , Winnie W.S. Mak
Background
Adolescent Internet gaming disorder (IGD), depressive symptoms, and suicidal ideation are significant public health concerns. However, their temporal dynamics and symptom-level interactions in adolescents remain unclear. This study aims to explore the longitudinal relationships and network interactions between IGD symptoms, depressive symptoms, and suicidal ideation among Hong Kong adolescents.
Methods
We conducted a one-year, two-wave longitudinal study with 7581 Hong Kong adolescents. The participants completed an anonymous, self-administered questionnaire in their classrooms. Two trained research assistants administered the surveys, which took approximately 20 min to complete. Cross-lagged panel network analysis was used to examine contemporaneous and temporal associations, with subgroup analyses by gender.
Results
Both the contemporaneous and temporal network analyses demonstrated distinct patterns when stratified by gender. In contemporaneous networks, significant differences in the overall structures of male and female adolescents’ networks were observed (M = 0.10, p < 0.01). “Giving up other activities due to gaming” (IGD5) was a central symptom for males' network, whereas suicidal ideation was the most central symptom for females' network. Longitudinally, depressive symptoms were strongly correlated with subsequent suicidal ideation in both genders.
Conclusion
Our findings emphasize important gender-specific differences in these dynamic relationships and offer direction for developing tailored transdiagnostic interventions aimed at addressing the complex interplay between behavioral addictions and emotional disorders among adolescents.
青少年网络游戏障碍（IGD）、抑郁症状和自杀意念是重要的公共卫生问题。然而，它们在青少年中的时间动态和症状水平的相互作用尚不清楚。本研究旨在探讨香港青少年IGD症状、抑郁症状与自杀意念之间的纵向关系及网络相互作用。方法对香港7581名青少年进行为期一年的两波纵向研究。参与者在教室里完成了一份匿名的、自我管理的问卷。两名训练有素的研究助理负责调查，大约需要20分钟才能完成。交叉滞后面板网络分析用于检查同期和时间关联，并按性别进行亚组分析。结果在按性别分层时，同期和时间网络分析都显示出不同的模式。在同期网络中，男性和女性青少年网络的整体结构存在显著差异（M = 0.10, p < 0.01）。“因游戏而放弃其他活动”（IGD5）是男性网络的中心症状，而自杀意念是女性网络的中心症状。纵向上，抑郁症状与随后的自杀意念在两性中都有很强的相关性。我们的研究结果强调了这些动态关系中重要的性别差异，并为制定针对性的跨诊断干预措施提供了方向，旨在解决青少年行为成瘾和情绪障碍之间复杂的相互作用。
{"title":"The network and temporal changes among internet gaming disorder symptoms, suicidal ideation, and depressive symptoms among adolescents in Hong Kong","authors":"Xue Yang , Xu Chen , Qian Li , Yilun Huang , Winnie W.S. Mak","doi":"10.1016/j.chbr.2025.100898","DOIUrl":"10.1016/j.chbr.2025.100898","url":null,"abstract":"<div><h3>Background</h3><div>Adolescent Internet gaming disorder (IGD), depressive symptoms, and suicidal ideation are significant public health concerns. However, their temporal dynamics and symptom-level interactions in adolescents remain unclear. This study aims to explore the longitudinal relationships and network interactions between IGD symptoms, depressive symptoms, and suicidal ideation among Hong Kong adolescents.</div></div><div><h3>Methods</h3><div>We conducted a one-year, two-wave longitudinal study with 7581 Hong Kong adolescents. The participants completed an anonymous, self-administered questionnaire in their classrooms. Two trained research assistants administered the surveys, which took approximately 20 min to complete. Cross-lagged panel network analysis was used to examine contemporaneous and temporal associations, with subgroup analyses by gender.</div></div><div><h3>Results</h3><div>Both the contemporaneous and temporal network analyses demonstrated distinct patterns when stratified by gender. In contemporaneous networks, significant differences in the overall structures of male and female adolescents’ networks were observed (<em>M</em> = 0.10, <em>p</em> < 0.01). “Giving up other activities due to gaming” (IGD5) was a central symptom for males' network, whereas suicidal ideation was the most central symptom for females' network. Longitudinally, depressive symptoms were strongly correlated with subsequent suicidal ideation in both genders.</div></div><div><h3>Conclusion</h3><div>Our findings emphasize important gender-specific differences in these dynamic relationships and offer direction for developing tailored transdiagnostic interventions aimed at addressing the complex interplay between behavioral addictions and emotional disorders among adolescents.</div></div>","PeriodicalId":72681,"journal":{"name":"Computers in human behavior reports","volume":"21 ","pages":"Article 100898"},"PeriodicalIF":5.8,"publicationDate":"2025-12-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145685538","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-12-04DOI: 10.1016/j.chbr.2025.100880
Larry Auyeung , Winnie W.S. Mak , Ella Zoe Tsang
{"title":"Beyond efficacy: Eliciting preference for face-to-face and digital psychological interventions among people with depression using discrete choice experiment","authors":"Larry Auyeung , Winnie W.S. Mak , Ella Zoe Tsang","doi":"10.1016/j.chbr.2025.100880","DOIUrl":"10.1016/j.chbr.2025.100880","url":null,"abstract":"","PeriodicalId":72681,"journal":{"name":"Computers in human behavior reports","volume":"21 ","pages":"Article 100880"},"PeriodicalIF":5.8,"publicationDate":"2025-12-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145685536","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-12-03DOI: 10.1016/j.chbr.2025.100896
Nitesh Kumar Jha , Meng-Jung Tsai
This study employs machine learning approaches to examine how socio-demographic, student-related, and school-related variables predict the computational thinking (CT) performance of 5211 Taiwanese eighth graders in the ICILS 2023 study (Fraillon, 2024). It further aims to identify the key predictors of Taiwanese students' CT scores in this international evaluation project. The study used seven trained models: Multinomial Logistic Regression, Random Forest, AdaBoost, XGBoost, LightGBM, Gradient Boosting classifier, and Stacking Ensemble to identify and rank the variables that affect CT scores. The CT performance score was used as a binary variable with two classes: below and above average score. Findings showed that XGBoost and Stacking Ensemble performed best when classifying below and average CT scores respectively in terms of precision, recall and F1 score. In addition, among the variables, student-related variables had the highest impact on students' CT skills followed by school-related and socio-demographic. Among student-related variables, CT disposition was the most significant variable followed by ICT self-efficacy and academic multitasking. Further, among school-related factor, learning special applications in class had significant impact followed by a low impact of socio-demographic variables such as home literacy and parents' education. This study offers practical implications for educators, policymakers, and curriculum designers by underscoring the role of CT disposition and recommending targeted support for enhancing students’ digital self-efficacy. Additionally, the study shows the potential of ML for creating adaptive learning environments and guiding data-informed decisions in educational policy and practice.
{"title":"Using machine learning approaches to predict Taiwanese eighth graders' computational thinking performance in ICILS 2023 study","authors":"Nitesh Kumar Jha , Meng-Jung Tsai","doi":"10.1016/j.chbr.2025.100896","DOIUrl":"10.1016/j.chbr.2025.100896","url":null,"abstract":"<div><div>This study employs machine learning approaches to examine how socio-demographic, student-related, and school-related variables predict the computational thinking (CT) performance of 5211 Taiwanese eighth graders in the ICILS 2023 study (Fraillon, 2024). It further aims to identify the key predictors of Taiwanese students' CT scores in this international evaluation project. The study used seven trained models: Multinomial Logistic Regression, Random Forest, AdaBoost, XGBoost, LightGBM, Gradient Boosting classifier, and Stacking Ensemble to identify and rank the variables that affect CT scores. The CT performance score was used as a binary variable with two classes: below and above average score. Findings showed that XGBoost and Stacking Ensemble performed best when classifying below and average CT scores respectively in terms of precision, recall and F1 score. In addition, among the variables, student-related variables had the highest impact on students' CT skills followed by school-related and socio-demographic. Among student-related variables, CT disposition was the most significant variable followed by ICT self-efficacy and academic multitasking. Further, among school-related factor, learning special applications in class had significant impact followed by a low impact of socio-demographic variables such as home literacy and parents' education. This study offers practical implications for educators, policymakers, and curriculum designers by underscoring the role of CT disposition and recommending targeted support for enhancing students’ digital self-efficacy. Additionally, the study shows the potential of ML for creating adaptive learning environments and guiding data-informed decisions in educational policy and practice.</div></div>","PeriodicalId":72681,"journal":{"name":"Computers in human behavior reports","volume":"21 ","pages":"Article 100896"},"PeriodicalIF":5.8,"publicationDate":"2025-12-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145685532","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-12-02DOI: 10.1016/j.chbr.2025.100893
Kilhoe Na
Virality metrics—such as the number of likes, shares, and comments—are often conceptualized in digital media research as heuristic or normative cues that help individuals assess credibility, popularity, or social approval. This study proposes instead that virality metrics function as interpretive cues that shape how users emotionally and cognitively engage with conflicting health information on social media. These metrics may influence how people perceive social disagreement, ambiguity, or controversy rather than merely signaling popularity. To examine this idea, an online experiment (N = 876) exposed participants to mock blog posts containing contradictory health claims—about either artificial sweeteners or processed meat—paired with either high or low virality metrics. The results supported a serial mediation model: virality metrics affected sharing intention by increasing perceived controversy, which in turn heightened anxiety and led to greater motivation to share the post. These findings challenge traditional heuristic and normative frameworks, instead positioning virality metrics as emotionally charged and interpretively meaningful cues that influence how users make sense of ambiguous information. The present study contributes to a more psychologically nuanced understanding of digital behavior in complex informational environments. It also raises practical considerations for platform design and health communication, where visible metrics may inadvertently amplify uncertainty, emotional discomfort, and the spread of controversial or ambiguous content—even in the absence of clear credibility cues.
{"title":"Social media virality metrics as interpretive cues: Affective pathways to sharing conflicting health information","authors":"Kilhoe Na","doi":"10.1016/j.chbr.2025.100893","DOIUrl":"10.1016/j.chbr.2025.100893","url":null,"abstract":"<div><div>Virality metrics—such as the number of likes, shares, and comments—are often conceptualized in digital media research as heuristic or normative cues that help individuals assess credibility, popularity, or social approval. This study proposes instead that virality metrics function as interpretive cues that shape how users emotionally and cognitively engage with conflicting health information on social media. These metrics may influence how people perceive social disagreement, ambiguity, or controversy rather than merely signaling popularity. To examine this idea, an online experiment (<em>N</em> = 876) exposed participants to mock blog posts containing contradictory health claims—about either artificial sweeteners or processed meat—paired with either high or low virality metrics. The results supported a serial mediation model: virality metrics affected sharing intention by increasing perceived controversy, which in turn heightened anxiety and led to greater motivation to share the post. These findings challenge traditional heuristic and normative frameworks, instead positioning virality metrics as emotionally charged and interpretively meaningful cues that influence how users make sense of ambiguous information. The present study contributes to a more psychologically nuanced understanding of digital behavior in complex informational environments. It also raises practical considerations for platform design and health communication, where visible metrics may inadvertently amplify uncertainty, emotional discomfort, and the spread of controversial or ambiguous content—even in the absence of clear credibility cues.</div></div>","PeriodicalId":72681,"journal":{"name":"Computers in human behavior reports","volume":"21 ","pages":"Article 100893"},"PeriodicalIF":5.8,"publicationDate":"2025-12-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145685535","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
扫码关注我们
求助内容：
应助结果提醒方式：
京公网安备 11010802042870号