Pub Date : 2024-05-15DOI: 10.1016/j.chb.2024.108287
Andrea Stašek , Lukas Blinka , Vasileios Stavropoulos
Despite the official inclusion of Gaming Disorder (GD) in the International Classification of Diseases, there is still an ongoing debate over its conceptualization and assessment. Several necessary steps have been recommended, including exploring the structure and the relationships of the GD symptoms, whilst considering how gaming may satisfy/meet the gamers’ needs. To address this aim, the responses of a large sample of active/dedicated adult gamers (N = 3895; Mage = 26.17; SDage = 6.48; 82.5% men; gaming hours per week M = 26.06) were analyzed using a network analysis in the present study. GD symptoms were assessed with AICA-S and needs satisfaction, both outside and inside the game world, with the Balanced Measure of Psychological Needs and Player Experience of Needs Satisfaction, respectively. The GD network was revealed to be composed of Time-Related, Cognitive-Emotional (with Craving and Tolerance most central), and Behavioral-Consequential (with Continuation Despite Consequences most central) symptom clusters. Escapism was shown to be the bridge between real-life needs, in-game needs, and GD symptoms. The results highlight the necessity to reconsider the structure of GD symptoms and their differential roles. Diagnostic, assessment, and treatment implications are illustrated.
{"title":"Disentangling the Net of Needs Satisfaction and Gaming Disorder Symptoms in Adult Gamers","authors":"Andrea Stašek , Lukas Blinka , Vasileios Stavropoulos","doi":"10.1016/j.chb.2024.108287","DOIUrl":"10.1016/j.chb.2024.108287","url":null,"abstract":"<div><p>Despite the official inclusion of Gaming Disorder (GD) in the International Classification of Diseases, there is still an ongoing debate over its conceptualization and assessment. Several necessary steps have been recommended, including exploring the structure and the relationships of the GD symptoms, whilst considering how gaming may satisfy/meet the gamers’ needs. To address this aim, the responses of a large sample of active/dedicated adult gamers (N = 3895; M<sub>age</sub> = 26.17; SD<sub>age</sub> = 6.48; 82.5% men; gaming hours per week M = 26.06) were analyzed using a network analysis in the present study. GD symptoms were assessed with AICA-S and needs satisfaction, both outside and inside the game world, with the Balanced Measure of Psychological Needs and Player Experience of Needs Satisfaction, respectively. The GD network was revealed to be composed of Time-Related, Cognitive-Emotional (with Craving and Tolerance most central), and Behavioral-Consequential (with Continuation Despite Consequences most central) symptom clusters. Escapism was shown to be the bridge between real-life needs, in-game needs, and GD symptoms. The results highlight the necessity to reconsider the structure of GD symptoms and their differential roles. Diagnostic, assessment, and treatment implications are illustrated.</p></div>","PeriodicalId":48471,"journal":{"name":"Computers in Human Behavior","volume":null,"pages":null},"PeriodicalIF":9.9,"publicationDate":"2024-05-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141023390","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"心理学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-05-15DOI: 10.1016/j.chb.2024.108300
Yang Song , Litong Wang , Zhiyuan Zhang , Lubica Hikkerova
Digital presence is increasing on social platforms, and increasing numbers of companies have begun to invite virtual influencers to endorse their products. However, the question of whether AI endorsers can completely replace real humans is a challenging one. Therefore, in this study, we attempt to identify the differences in consumer purchase intention between consumers of products promoted by AI endorsers versus real human endorsers. In two studies, we found that AI endorsers effectively stimulated consumers' purchase intentions when recommending search products. For experience products, the marketing effect of a real human celebrity endorser is better, however, and consumers' purchase intention is stronger. Perceptions of congruency mediate the interaction of endorser and product type on consumers' purchase intention; self-image congruency mediates the influence of AI endorsers on consumers' purchase intention for search products. Moreover, functional congruency mediates the influence of celebrity endorsers on consumers' purchase intention for experience products. This paper is helpful for companies to encourage them to consider the role of different product attributes and adopt more appropriate strategies to maximize the effect of endorsement marketing strategies.
{"title":"AI or human: How endorser shapes online purchase intention?","authors":"Yang Song , Litong Wang , Zhiyuan Zhang , Lubica Hikkerova","doi":"10.1016/j.chb.2024.108300","DOIUrl":"10.1016/j.chb.2024.108300","url":null,"abstract":"<div><p>Digital presence is increasing on social platforms, and increasing numbers of companies have begun to invite virtual influencers to endorse their products. However, the question of whether AI endorsers can completely replace real humans is a challenging one. Therefore, in this study, we attempt to identify the differences in consumer purchase intention between consumers of products promoted by AI endorsers versus real human endorsers. In two studies, we found that AI endorsers effectively stimulated consumers' purchase intentions when recommending search products. For experience products, the marketing effect of a real human celebrity endorser is better, however, and consumers' purchase intention is stronger. Perceptions of congruency mediate the interaction of endorser and product type on consumers' purchase intention; self-image congruency mediates the influence of AI endorsers on consumers' purchase intention for search products. Moreover, functional congruency mediates the influence of celebrity endorsers on consumers' purchase intention for experience products. This paper is helpful for companies to encourage them to consider the role of different product attributes and adopt more appropriate strategies to maximize the effect of endorsement marketing strategies.</p></div>","PeriodicalId":48471,"journal":{"name":"Computers in Human Behavior","volume":null,"pages":null},"PeriodicalIF":9.9,"publicationDate":"2024-05-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141042319","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"心理学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-05-15DOI: 10.1016/j.chb.2024.108303
Daniel Amo-Filva , David Fonseca , Francisco José García-Peñalvo , Marc Alier Forment , Maria José Casany Guerrero , Guillem Godoy
The study systematically reviews the integration of Fog and Edge Computing within Learning Analytics to enhance data privacy and security in educational settings that use cloud computing. Employing the PRISMA methodology, we analyze current literature from Web of Science and Scopus databases to examine how these decentralized computing technologies can mitigate the risks associated with centralized cloud storage by processing data closer to its source. Our findings highlight the significant potential of Fog and Edge Computing to transform Learning Analytics by enabling real-time, context-aware data analysis that supports personalized learning while ensuring stringent data privacy. This approach challenges conventional data management practices, advocating for privacy by design in developing new strategies and frameworks. The research underscores the need for collaborative efforts in establishing standards and guidelines for secure and effective technology use in education, pointing towards the necessity of addressing technical, operational, and ethical challenges to maximize the benefits of fog and edge computing in Learning Analytics.
本研究系统地回顾了学习分析中雾计算和边缘计算的整合情况,以提高使用云计算的教育环境中的数据隐私和安全性。采用 PRISMA 方法,我们分析了来自 Web of Science 和 Scopus 数据库的现有文献,以研究这些分散式计算技术如何通过在更靠近数据源的地方处理数据来降低与集中式云存储相关的风险。我们的研究结果凸显了雾计算和边缘计算在改变学习分析方面的巨大潜力,它们可以实现实时、上下文感知的数据分析,从而支持个性化学习,同时确保严格的数据隐私。这种方法对传统的数据管理实践提出了挑战,主张在制定新战略和框架时通过设计来保护隐私。这项研究强调,有必要通力合作,为在教育领域安全、有效地使用技术制定标准和准则,并指出有必要解决技术、操作和道德方面的挑战,以最大限度地发挥雾计算和边缘计算在学习分析中的优势。
{"title":"Exploring the landscape of learning analytics privacy in fog and edge computing: A systematic literature review","authors":"Daniel Amo-Filva , David Fonseca , Francisco José García-Peñalvo , Marc Alier Forment , Maria José Casany Guerrero , Guillem Godoy","doi":"10.1016/j.chb.2024.108303","DOIUrl":"https://doi.org/10.1016/j.chb.2024.108303","url":null,"abstract":"<div><p>The study systematically reviews the integration of Fog and Edge Computing within Learning Analytics to enhance data privacy and security in educational settings that use cloud computing. Employing the PRISMA methodology, we analyze current literature from Web of Science and Scopus databases to examine how these decentralized computing technologies can mitigate the risks associated with centralized cloud storage by processing data closer to its source. Our findings highlight the significant potential of Fog and Edge Computing to transform Learning Analytics by enabling real-time, context-aware data analysis that supports personalized learning while ensuring stringent data privacy. This approach challenges conventional data management practices, advocating for privacy by design in developing new strategies and frameworks. The research underscores the need for collaborative efforts in establishing standards and guidelines for secure and effective technology use in education, pointing towards the necessity of addressing technical, operational, and ethical challenges to maximize the benefits of fog and edge computing in Learning Analytics.</p></div>","PeriodicalId":48471,"journal":{"name":"Computers in Human Behavior","volume":null,"pages":null},"PeriodicalIF":9.9,"publicationDate":"2024-05-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141067631","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"心理学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-05-14DOI: 10.1016/j.chb.2024.108297
Marcos Lerma, Rory A. Pfund, James P. Whelan
Social media has provided gambling operators with access to millions of individuals and novel ways to promote gambling. Research has suggested that exposure to gambling advertisements on social media platforms is associated with increased gambling in individuals at-risk for problem gambling. These findings bring into question whether social media platforms are sensitive to differences in user account activity (e.g., tweets, likes, accounts visited) when displaying promoted advertisements and gambling-related content. To assess for these differences, four Twitter/X accounts were created and assigned to send out tweets containing pro-gambling or safe-gambling messages. Additionally, each account was assigned to interact with Twitter/X profiles associated with gambling operators or responsible gambling. Accounts were assessed daily for promoted advertisement traffic and gambling-related content from January to March 2022. The study included three phases that implemented changes in privacy settings, websites visited, and gambling-related tweets observed. To assess for between-phase differences, Tau-U analyses were performed using R. Gambling-related content observed was dependent on assigned account activity. Accounts that interacted with gambling operators’ profiles were only displayed pro-gambling content. Conversely, accounts that interacted with responsible gambling profiles were only displayed safe-gambling content. No promoted gambling advertisements were observed throughout the study. Findings suggest that Twitter/X is sensitive to differences in account activity, and user activity may influence gambling content displayed on Twitter/X. Nevertheless, gambling operators should adopt a conservative approach on social media to ensure protection of consumers. Consumers should be given autonomy to engage with gambling content without being drawn in involuntarily.
{"title":"Does user activity promote gambling-related content on Twitter/X?","authors":"Marcos Lerma, Rory A. Pfund, James P. Whelan","doi":"10.1016/j.chb.2024.108297","DOIUrl":"10.1016/j.chb.2024.108297","url":null,"abstract":"<div><p>Social media has provided gambling operators with access to millions of individuals and novel ways to promote gambling. Research has suggested that exposure to gambling advertisements on social media platforms is associated with increased gambling in individuals at-risk for problem gambling. These findings bring into question whether social media platforms are sensitive to differences in user account activity (e.g., tweets, likes, accounts visited) when displaying promoted advertisements and gambling-related content. To assess for these differences, four Twitter/X accounts were created and assigned to send out tweets containing pro-gambling or safe-gambling messages. Additionally, each account was assigned to interact with Twitter/X profiles associated with gambling operators or responsible gambling. Accounts were assessed daily for promoted advertisement traffic and gambling-related content from January to March 2022. The study included three phases that implemented changes in privacy settings, websites visited, and gambling-related tweets observed. To assess for between-phase differences, Tau-<em>U</em> analyses were performed using R. Gambling-related content observed was dependent on assigned account activity. Accounts that interacted with gambling operators’ profiles were only displayed pro-gambling content. Conversely, accounts that interacted with responsible gambling profiles were only displayed safe-gambling content. No promoted gambling advertisements were observed throughout the study. Findings suggest that Twitter/X is sensitive to differences in account activity, and user activity may influence gambling content displayed on Twitter/X. Nevertheless, gambling operators should adopt a conservative approach on social media to ensure protection of consumers. Consumers should be given autonomy to engage with gambling content without being drawn in involuntarily.</p></div>","PeriodicalId":48471,"journal":{"name":"Computers in Human Behavior","volume":null,"pages":null},"PeriodicalIF":9.9,"publicationDate":"2024-05-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141058376","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"心理学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-05-14DOI: 10.1016/j.chb.2024.108298
Nicola Henry , Rebecca Umbach
The growing threat of sexual extortion (”sextortion”) has garnered significant attention in the news and by law enforcement agencies around the world. Foundational knowledge of prevalence and risk factors, however, is still nascent. The present study surveyed 16,693 respondents, distributed equally across 10 different countries, to assess prevalence of victimization and perpetration of threatening to disseminate intimate images. Weighted by gender, age, region, and population, 14.5% of respondents indicated at least one experience of victimization, while 4.8% of respondents indicated perpetration of the same. Demographic risk factors for perpetration and victimization were also assessed. Consistent with findings from other studies, men (15.7%) were 1.17 times more likely to report being victimized compared to women (13.2%), and 1.43 times more likely to report perpetration. LGBTQ+ respondents were 2.07 times more likely to report victimization compared to non-LGBTQ+ respondents, and 2.51 times more likely to report offending behaviors. Age was significantly associated, with younger participants more likely to report both victimization and perpetration experiences. The most common type of perpetrator, as reported by victims, was a former or current partner. Despite the strong likelihood of under-reporting given the topic area, the study found that experiencing threats to distribute intimate content is a relatively commonplace occurrence, impacting 1 in 7 adults. Implications for potential mitigation are discussed.
{"title":"Sextortion: Prevalence and correlates in 10 countries","authors":"Nicola Henry , Rebecca Umbach","doi":"10.1016/j.chb.2024.108298","DOIUrl":"10.1016/j.chb.2024.108298","url":null,"abstract":"<div><p>The growing threat of sexual extortion (”sextortion”) has garnered significant attention in the news and by law enforcement agencies around the world. Foundational knowledge of prevalence and risk factors, however, is still nascent. The present study surveyed 16,693 respondents, distributed equally across 10 different countries, to assess prevalence of victimization and perpetration of threatening to disseminate intimate images. Weighted by gender, age, region, and population, 14.5% of respondents indicated at least one experience of victimization, while 4.8% of respondents indicated perpetration of the same. Demographic risk factors for perpetration and victimization were also assessed. Consistent with findings from other studies, men (15.7%) were 1.17 times more likely to report being victimized compared to women (13.2%), and 1.43 times more likely to report perpetration. LGBTQ+ respondents were 2.07 times more likely to report victimization compared to non-LGBTQ+ respondents, and 2.51 times more likely to report offending behaviors. Age was significantly associated, with younger participants more likely to report both victimization and perpetration experiences. The most common type of perpetrator, as reported by victims, was a former or current partner. Despite the strong likelihood of under-reporting given the topic area, the study found that experiencing threats to distribute intimate content is a relatively commonplace occurrence, impacting 1 in 7 adults. Implications for potential mitigation are discussed.</p></div>","PeriodicalId":48471,"journal":{"name":"Computers in Human Behavior","volume":null,"pages":null},"PeriodicalIF":9.9,"publicationDate":"2024-05-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.sciencedirect.com/science/article/pii/S0747563224001663/pdfft?md5=f2c2ac00c8316f0ae60834ee44d48d81&pid=1-s2.0-S0747563224001663-main.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141029220","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"心理学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-05-13DOI: 10.1016/j.chb.2024.108301
Muhammad Shoaib , Nasir Sayed , Jaiteg Singh , Jana Shafi , Shakir Khan , Farman Ali
Campus Management Systems (CMSs) are vital tools in managing educational institutions, handling tasks like student enrollment, scheduling, and resource allocation. The increasing adoption of CMS for online and mixed-learning environments highlights their importance. However, inherent limitations in conventional CMS platforms hinder personalized student guidance and effective identification of academic challenges. Addressing this crucial gap, our study introduces an AI Student Success Predictor empowered by advanced machine learning algorithms, capable of automating grading processes, predicting student risks, and forecasting retention or dropout outcomes. Central to our approach is the creation of a standardized dataset, meticulously curated by integrating student information from diverse relational databases. A Convolutional Neural Network (CNN) feature learning block is developed the extract the hidden patterns in the student data. This classification model stands as an ensemble masterpiece, incorporating SVM, Random Forest, and KNN classifiers, subsequently refined by a Bayesian averaging model. The proposed ensemble model shows the ability to predict the student grades, retention, and risk levels of dropout. The accuracy achieved by the proposed model is assessed using test data, culminating in a commendable 93% accuracy for student grade prediction and student risk prediction, and a solid 92% accuracy for the complex domain of retention and dropout forecasting. The proposed AI system seamlessly integrates with existing CMS infrastructure, enabling real-time data retrieval and swift, accurate predictions, enhancing academic decision-making efficiency. Our study's pioneering AI Student Success Predictor bridges the chasm between traditional CMS limitations and the growing demands of modern education.
{"title":"AI student success predictor: Enhancing personalized learning in campus management systems","authors":"Muhammad Shoaib , Nasir Sayed , Jaiteg Singh , Jana Shafi , Shakir Khan , Farman Ali","doi":"10.1016/j.chb.2024.108301","DOIUrl":"10.1016/j.chb.2024.108301","url":null,"abstract":"<div><p>Campus Management Systems (CMSs) are vital tools in managing educational institutions, handling tasks like student enrollment, scheduling, and resource allocation. The increasing adoption of CMS for online and mixed-learning environments highlights their importance. However, inherent limitations in conventional CMS platforms hinder personalized student guidance and effective identification of academic challenges. Addressing this crucial gap, our study introduces an AI Student Success Predictor empowered by advanced machine learning algorithms, capable of automating grading processes, predicting student risks, and forecasting retention or dropout outcomes. Central to our approach is the creation of a standardized dataset, meticulously curated by integrating student information from diverse relational databases. A Convolutional Neural Network (CNN) feature learning block is developed the extract the hidden patterns in the student data. This classification model stands as an ensemble masterpiece, incorporating SVM, Random Forest, and KNN classifiers, subsequently refined by a Bayesian averaging model. The proposed ensemble model shows the ability to predict the student grades, retention, and risk levels of dropout. The accuracy achieved by the proposed model is assessed using test data, culminating in a commendable 93% accuracy for student grade prediction and student risk prediction, and a solid 92% accuracy for the complex domain of retention and dropout forecasting. The proposed AI system seamlessly integrates with existing CMS infrastructure, enabling real-time data retrieval and swift, accurate predictions, enhancing academic decision-making efficiency. Our study's pioneering AI Student Success Predictor bridges the chasm between traditional CMS limitations and the growing demands of modern education.</p></div>","PeriodicalId":48471,"journal":{"name":"Computers in Human Behavior","volume":null,"pages":null},"PeriodicalIF":9.9,"publicationDate":"2024-05-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141027572","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"心理学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-05-13DOI: 10.1016/j.chb.2024.108299
Min Wu, Zhaotong Li, Kum Fai Yuen
To address the issue of self-serving attributional bias linked to automation technology and artificial intelligence (AI), this study recruited a gender-balanced sample to investigate the positive effects of perceived anthropomorphism and hierarchical status of computer systems on users’ self-accountability and use intention. The specific research context was intelligent vehicles, a common form of automation and AI. Findings showed that anthropomorphism can significantly boost user self-accountability for both successes and failures, while hierarchical status primarily affected accountability for failures. Besides, the positive effects of anthropomorphism on self-accountability and use intention were mediated by cognitive empowerment and social rewards, respectively. This study also found that educational background amplified the impact of anthropomorphism, whereas ethnic differences moderated the effects of hierarchical status on self-accountability. Furthermore, incident experiences were found to positively moderate the relationship between hierarchical status and use intention, which indicates the need for more safety-focused strategies for human-computer interaction (HCI). In general, this study presented a promising strategy for academia and industry in designing human-like interactions to balance self-serving bias and foster self-accountability, which can potentially result in more inclusive and effective HCI experiences.
{"title":"Effect of anthropomorphic design and hierarchical status on balancing self-serving bias: Accounting for education, ethnicity, and experience","authors":"Min Wu, Zhaotong Li, Kum Fai Yuen","doi":"10.1016/j.chb.2024.108299","DOIUrl":"10.1016/j.chb.2024.108299","url":null,"abstract":"<div><p>To address the issue of self-serving attributional bias linked to automation technology and artificial intelligence (AI), this study recruited a gender-balanced sample to investigate the positive effects of perceived anthropomorphism and hierarchical status of computer systems on users’ self-accountability and use intention. The specific research context was intelligent vehicles, a common form of automation and AI. Findings showed that anthropomorphism can significantly boost user self-accountability for both successes and failures, while hierarchical status primarily affected accountability for failures. Besides, the positive effects of anthropomorphism on self-accountability and use intention were mediated by cognitive empowerment and social rewards, respectively. This study also found that educational background amplified the impact of anthropomorphism, whereas ethnic differences moderated the effects of hierarchical status on self-accountability. Furthermore, incident experiences were found to positively moderate the relationship between hierarchical status and use intention, which indicates the need for more safety-focused strategies for human-computer interaction (HCI). In general, this study presented a promising strategy for academia and industry in designing human-like interactions to balance self-serving bias and foster self-accountability, which can potentially result in more inclusive and effective HCI experiences.</p></div>","PeriodicalId":48471,"journal":{"name":"Computers in Human Behavior","volume":null,"pages":null},"PeriodicalIF":9.9,"publicationDate":"2024-05-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141057136","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"心理学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-05-09DOI: 10.1016/j.chb.2024.108286
Guandong Gao , Ke Xiao , Hui Li , Shengzun Song
Criminal cases often exhibit imbalance and cannot be extended by data augmentation when classified into attribution types. To solve the problem of unbalance data in offenders’ attribution classification, this paper proposes a criminal psychological attribution assessment model by an improved Balanced TF-Distinguishing IDF method (B-TF-dIDF) and constructed a hybrid network with attention method to fuse numerical and text features for improving the accuracy. First, as a statistical method, B-TF-dIDF is presented to reduce the impact of class-imbalance for extraction of numerical features, in which a balanced element is added to reduce the effects of incorrect type keywords on classification, and a distinguishing element is added to discriminate the types of keywords. Then, an improved hybrid network model composed of Long Short-Term Memory (LSTM) and Convolutional Neural Network (CNN) is constructed to balance the influence of different lengths of text samples for extracting the semantic features of criminal texts. For evaluating different feature weights by their importance, Spatial Attention is used to improve CNN in the feature maps. Moreover, the self-attention is also performed to re-evaluate the mixed features. Finally, the softmax classifier provides a scientific basis for developing a hierarchical treatment mechanism further. Additionally, we build a criminal data set with labels from real cases for testing. The experiment proved that the proposed model is better than other related methods in various evaluation indicators, including the micro and macro scopes. Moreover, the F1 of minority samples has increased by 6%–8%, indicating that the proposed method can reduce the impact of class-imbalance.
犯罪案件在归因类型划分时往往表现出不平衡性,无法通过数据扩充进行扩展。为了解决罪犯归因分类中数据不平衡的问题,本文通过改进的平衡 TF-Distinguishing IDF 方法(B-TF-dIDF)提出了一种犯罪心理归因评估模型,并利用注意力方法构建了一个混合网络,将数字特征和文本特征进行融合,以提高评估的准确性。首先,作为一种统计方法,B-TF-dIDF 被提出来用于减少抽取数字特征时类别不平衡的影响,其中添加了一个平衡元素来减少错误类型关键词对分类的影响,同时添加了一个区分元素来区分关键词的类型。然后,构建一个由长短期记忆(LSTM)和卷积神经网络(CNN)组成的改进型混合网络模型,以平衡不同长度文本样本对提取犯罪文本语义特征的影响。为了根据不同特征的重要性评估其权重,使用了空间注意力来改进特征图中的 CNN。此外,还使用自我关注来重新评估混合特征。最后,softmax 分类器为进一步开发分层处理机制提供了科学依据。此外,我们还建立了一个带有真实案件标签的犯罪数据集进行测试。实验证明,所提出的模型在微观和宏观等多个评价指标上都优于其他相关方法。此外,少数样本的 F1 提高了 6%-8%,这表明提出的方法可以减少类不平衡的影响。
{"title":"An intelligent assessment method of criminal psychological attribution based on unbalance data","authors":"Guandong Gao , Ke Xiao , Hui Li , Shengzun Song","doi":"10.1016/j.chb.2024.108286","DOIUrl":"https://doi.org/10.1016/j.chb.2024.108286","url":null,"abstract":"<div><p>Criminal cases often exhibit imbalance and cannot be extended by data augmentation when classified into attribution types. To solve the problem of unbalance data in offenders’ attribution classification, this paper proposes a criminal psychological attribution assessment model by an improved Balanced TF-Distinguishing IDF method (B-TF-dIDF) and constructed a hybrid network with attention method to fuse numerical and text features for improving the accuracy. First, as a statistical method, B-TF-dIDF is presented to reduce the impact of class-imbalance for extraction of numerical features, in which a balanced element is added to reduce the effects of incorrect type keywords on classification, and a distinguishing element is added to discriminate the types of keywords. Then, an improved hybrid network model composed of Long Short-Term Memory (LSTM) and Convolutional Neural Network (CNN) is constructed to balance the influence of different lengths of text samples for extracting the semantic features of criminal texts. For evaluating different feature weights by their importance, Spatial Attention is used to improve CNN in the feature maps. Moreover, the self-attention is also performed to re-evaluate the mixed features. Finally, the softmax classifier provides a scientific basis for developing a hierarchical treatment mechanism further. Additionally, we build a criminal data set with labels from real cases for testing. The experiment proved that the proposed model is better than other related methods in various evaluation indicators, including the micro and macro scopes. Moreover, the F1 of minority samples has increased by 6%–8%, indicating that the proposed method can reduce the impact of class-imbalance.</p></div>","PeriodicalId":48471,"journal":{"name":"Computers in Human Behavior","volume":null,"pages":null},"PeriodicalIF":9.9,"publicationDate":"2024-05-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140913832","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"心理学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-05-07DOI: 10.1016/j.chb.2024.108271
Yanqing Sun, Juan Xie
An important step in developing effective interventions to stop the spread of misinformation is to develop a comprehensive understanding of the characteristics of people who are likely to share misinformation on social media. Accordingly, we conducted a meta-analysis of 60 articles on the individual traits of people who share misinformation. Results showed that the passing time motivation (r = 0.524) had the strongest positive relationship with misinformation sharing, whereas trust in information (r = 0.437) and the socialization motivation (r = 0.350) had a large positive effect. The motivations for entertainment (r = 0.276) and altruism (r = 0.220), and trust in social media (r = 0.219) all had medium positive effects. Prior exposure to misinformation (r = 0.191) and political conservatism (r = 0.119) both had a small positive association with misinformation sharing, whereas the personality trait of agreeableness (r = −0.094) had a weak negative association with misinformation sharing. Information literacy (r = −0.229) exerted a medium negative effect, but new media literacy was not significantly related to misinformation sharing. Finally, contextual and methodological factors emerged as important moderators. Older people and women were more likely to spread health misinformation, whereas younger people and men were more likely to share political misinformation. Overall, this study indicates that misinformation sharing is more strongly related to psychological traits than to personality and demographic traits. Moreover, the uses and gratifications theory, theories related to trust and credibility, and the illusory truth effect may well explain misinformation sharing in the online space.
{"title":"Who shares misinformation on social media? A meta-analysis of individual traits related to misinformation sharing","authors":"Yanqing Sun, Juan Xie","doi":"10.1016/j.chb.2024.108271","DOIUrl":"10.1016/j.chb.2024.108271","url":null,"abstract":"<div><p>An important step in developing effective interventions to stop the spread of misinformation is to develop a comprehensive understanding of the characteristics of people who are likely to share misinformation on social media. Accordingly, we conducted a meta-analysis of 60 articles on the individual traits of people who share misinformation. Results showed that the passing time motivation (r = 0.524) had the strongest positive relationship with misinformation sharing, whereas trust in information (r = 0.437) and the socialization motivation (r = 0.350) had a large positive effect. The motivations for entertainment (r = 0.276) and altruism (r = 0.220), and trust in social media (r = 0.219) all had medium positive effects. Prior exposure to misinformation (r = 0.191) and political conservatism (r = 0.119) both had a small positive association with misinformation sharing, whereas the personality trait of agreeableness (r = −0.094) had a weak negative association with misinformation sharing. Information literacy (r = −0.229) exerted a medium negative effect, but new media literacy was not significantly related to misinformation sharing. Finally, contextual and methodological factors emerged as important moderators. Older people and women were more likely to spread health misinformation, whereas younger people and men were more likely to share political misinformation. Overall, this study indicates that misinformation sharing is more strongly related to psychological traits than to personality and demographic traits. Moreover, the uses and gratifications theory, theories related to trust and credibility, and the illusory truth effect may well explain misinformation sharing in the online space.</p></div>","PeriodicalId":48471,"journal":{"name":"Computers in Human Behavior","volume":null,"pages":null},"PeriodicalIF":9.9,"publicationDate":"2024-05-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141053140","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"心理学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-05-07DOI: 10.1016/j.chb.2024.108285
Yujie Zheng , Baojun Ma , Xiwen Zhou , Benjiang Lu
This study explored how the features of consumer-generated images (CGIs) influence consumers' attention and purchase intention in both browsing and buying stages of online shopping, as well as the mediation of these effects. We consider the common features of image reviews (e.g. brightness, clarity, product displaying proportion and consistency) as heuristic cues evaluated by consumers. We posit that image brightness, clarity and product displaying proportion are product irrelevant cues associated with CGI attractiveness in the browsing stage, whereas product consistency is a product relevant cue associated with CGI attractiveness and purchase intention during the buying stage. Eye-tracking experiments with 127 undergraduates using Parka products support our hypotheses. The results indicate a positive correlation between the quality of product-irrelevant cues and CGI attractiveness in browsing, and a similar positive association with product-relevant cues during buying. The results also show that both product relevant and irrelevant cues are positively associated with consumers’ purchase intention, mediated by eliciting emotional arousal rather than visual attention. This study extends the literature by shifting the focus from assessing the overall aesthetic quality of CGIs to the importance of specific features in different online shopping stages. The study provides important implications for e-commerce platforms to strategically encourage users to submit CGIs that maintain consistency with the merchant-provided images and exhibit high image quality attributes such as brightness and clarity. Future research should explore CGIs across different product types to understand their varying roles.
{"title":"Not all consumer-generated images are attractive and persuasive: A heuristic cue perspective","authors":"Yujie Zheng , Baojun Ma , Xiwen Zhou , Benjiang Lu","doi":"10.1016/j.chb.2024.108285","DOIUrl":"https://doi.org/10.1016/j.chb.2024.108285","url":null,"abstract":"<div><p>This study explored how the features of consumer-generated images (CGIs) influence consumers' attention and purchase intention in both browsing and buying stages of online shopping, as well as the mediation of these effects. We consider the common features of image reviews (e.g. brightness, clarity, product displaying proportion and consistency) as heuristic cues evaluated by consumers. We posit that image brightness, clarity and product displaying proportion are product irrelevant cues associated with CGI attractiveness in the browsing stage, whereas product consistency is a product relevant cue associated with CGI attractiveness and purchase intention during the buying stage. Eye-tracking experiments with 127 undergraduates using Parka products support our hypotheses. The results indicate a positive correlation between the quality of product-irrelevant cues and CGI attractiveness in browsing, and a similar positive association with product-relevant cues during buying. The results also show that both product relevant and irrelevant cues are positively associated with consumers’ purchase intention, mediated by eliciting emotional arousal rather than visual attention. This study extends the literature by shifting the focus from assessing the overall aesthetic quality of CGIs to the importance of specific features in different online shopping stages. The study provides important implications for e-commerce platforms to strategically encourage users to submit CGIs that maintain consistency with the merchant-provided images and exhibit high image quality attributes such as brightness and clarity. Future research should explore CGIs across different product types to understand their varying roles.</p></div>","PeriodicalId":48471,"journal":{"name":"Computers in Human Behavior","volume":null,"pages":null},"PeriodicalIF":9.9,"publicationDate":"2024-05-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140905405","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"心理学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}