Computers in Human Behavior最新文献

From iron to diamond: Collaborative behavior development across competitive tiers in League of Legends 从铁到钻石：《英雄联盟》中跨竞争层级的合作行为发展

IF 8.9 1区心理学 Q1 PSYCHOLOGY, EXPERIMENTAL

Computers in Human Behavior

Pub Date : 2026-01-27 DOI: 10.1016/j.chb.2026.108928

Jimoon Kang, Seongcheol Kim

Understanding how individuals develop collaborative competencies in digital environments has become critical as work, education, and recreation increasingly occur through screens. This mixed-methods study examines collaborative behaviors across competitive skill tiers in League of Legends, analyzing 560,000 matches and conducting expert interviews. Quantitative findings reveal that collaborative behaviors—vision control, strategic communication, and assists—differentiate tiers far more strongly than individual performance. Vision control increased over 200 %, strategic communication increased 169 % with forward-looking signals showing the largest effects, and disruptive behaviors declined five-fold across tiers, while individual combat effectiveness remained flat. Hierarchical regression demonstrated that these dimensions developed as integrated systems, together predicting 87.3 % of tier variance. Expert interviews identified bidirectional selection-socialization processes where collaborative predisposition facilitates advancement and higher-tier environments further reinforce capabilities through stricter norms, peer modeling, and reputational consequences, with threshold effects around mid-tiers indicating qualitative environmental shifts. These findings provide large-scale behavioral evidence supporting Social Identity Theory, team cognition theory, and status characteristics theory in naturalistic digital contexts, demonstrate that teamwork theories require adaptation for fluid-membership environments with standardized communication tools, and reveal collaborative competence as integrated systems requiring holistic development.

随着工作、教育和娱乐越来越多地通过屏幕进行，了解个人如何在数字环境中发展协作能力变得至关重要。这项混合方法的研究分析了《英雄联盟》中不同竞技水平的合作行为，分析了56万场比赛并进行了专家访谈。定量研究结果显示，合作行为——视觉控制、战略沟通和协助——比个人表现更能区分层级。视觉控制增加了200%以上，战略沟通增加了169%，前瞻性信号显示出最大的影响，破坏行为在不同层次上下降了五倍，而个人战斗力保持不变。层次回归表明，这些维度是作为一个综合系统发展起来的，共同预测了87.3%的层次方差。专家访谈确定了双向选择社会化过程，其中协作倾向促进了进步，高层环境通过更严格的规范、同伴建模和声誉后果进一步加强了能力，中层周围的阈值效应表明了质量环境的转变。这些研究结果为社会认同理论、团队认知理论和身份特征理论在自然数字环境下的应用提供了大规模的行为证据，表明团队合作理论需要使用标准化的沟通工具来适应流动性成员环境，并揭示了协作能力是一个需要整体发展的集成系统。

{"title":"From iron to diamond: Collaborative behavior development across competitive tiers in League of Legends","authors":"Jimoon Kang, Seongcheol Kim","doi":"10.1016/j.chb.2026.108928","DOIUrl":"10.1016/j.chb.2026.108928","url":null,"abstract":"<div><div>Understanding how individuals develop collaborative competencies in digital environments has become critical as work, education, and recreation increasingly occur through screens. This mixed-methods study examines collaborative behaviors across competitive skill tiers in <em>League of Legends,</em> analyzing 560,000 matches and conducting expert interviews. Quantitative findings reveal that collaborative behaviors—vision control, strategic communication, and assists—differentiate tiers far more strongly than individual performance. Vision control increased over 200 %, strategic communication increased 169 % with forward-looking signals showing the largest effects, and disruptive behaviors declined five-fold across tiers, while individual combat effectiveness remained flat. Hierarchical regression demonstrated that these dimensions developed as integrated systems, together predicting 87.3 % of tier variance. Expert interviews identified bidirectional selection-socialization processes where collaborative predisposition facilitates advancement and higher-tier environments further reinforce capabilities through stricter norms, peer modeling, and reputational consequences, with threshold effects around mid-tiers indicating qualitative environmental shifts. These findings provide large-scale behavioral evidence supporting Social Identity Theory, team cognition theory, and status characteristics theory in naturalistic digital contexts, demonstrate that teamwork theories require adaptation for fluid-membership environments with standardized communication tools, and reveal collaborative competence as integrated systems requiring holistic development.</div></div>","PeriodicalId":48471,"journal":{"name":"Computers in Human Behavior","volume":"179 ","pages":"Article 108928"},"PeriodicalIF":8.9,"publicationDate":"2026-01-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146081671","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}

引用次数: 0

User cognitive fit in human-AI interaction: Exploring the link between input representation and generative output complexity 人机交互中的用户认知契合：探索输入表示与生成输出复杂性之间的联系

IF 8.9 1区心理学 Q1 PSYCHOLOGY, EXPERIMENTAL

Computers in Human Behavior

Pub Date : 2026-01-21 DOI: 10.1016/j.chb.2026.108927

Hechang Cai , Jinlai Zhou

Generative artificial intelligence (AI) has transformed knowledge production into an interactive process in which user inputs actively influence generative outputs. Yet, little is known about how different forms of input representation affect the complexity of generative. Building on cognitive fit theory, this study examines how the characteristics of user input jointly determine output complexity in human–AI interaction. Using a large corpus of over sixty thousand conversations, we adopt a multimethod approach that combines fixed-effects modeling with interpretable machine learning (XGBoost, SHAP, and time-sensitive pattern analysis). The results reveal a nonlinear and dynamic relationship between input representation fit and output complexity, indicating that both low-fit and high-fit inputs can increase output complexity through distinct mechanisms. We further demonstrate that these effects differ according to interaction depth and temporal context, suggesting that representational alignment is not static but dynamically recalibrated during iterative exchanges. The study enhances the cognitive fit theory by conceptualizing fit as an adaptive alignment process in generative environments instead of a fixed match between human cognition and system representation. The findings highlight the necessity for AI systems that support adaptive prompting and context-aware feedback to enhance collaborative generative improvement.

生成式人工智能（AI）将知识生产转变为用户输入积极影响生成式输出的交互过程。然而，人们对不同形式的输入表示如何影响生成的复杂性知之甚少。在认知契合理论的基础上，本研究考察了用户输入的特征如何共同决定人机交互中的输出复杂性。使用超过6万个对话的大型语料库，我们采用了一种多方法方法，将固定效应建模与可解释的机器学习（XGBoost、SHAP和时间敏感模式分析）相结合。研究结果揭示了输入表示拟合与输出复杂度之间的非线性动态关系，表明低拟合和高拟合输入都可以通过不同的机制增加输出复杂度。我们进一步证明，这些影响根据交互深度和时间背景而不同，这表明表征对齐不是静态的，而是在迭代交换中动态重新校准的。本研究将认知契合定义为生成环境中的自适应对齐过程，而不是人类认知与系统表征之间的固定匹配，从而增强了认知契合理论。研究结果强调了支持自适应提示和上下文感知反馈的人工智能系统的必要性，以加强协作生成改进。

{"title":"User cognitive fit in human-AI interaction: Exploring the link between input representation and generative output complexity","authors":"Hechang Cai , Jinlai Zhou","doi":"10.1016/j.chb.2026.108927","DOIUrl":"10.1016/j.chb.2026.108927","url":null,"abstract":"<div><div>Generative artificial intelligence (AI) has transformed knowledge production into an interactive process in which user inputs actively influence generative outputs. Yet, little is known about how different forms of input representation affect the complexity of generative. Building on cognitive fit theory, this study examines how the characteristics of user input jointly determine output complexity in human–AI interaction. Using a large corpus of over sixty thousand conversations, we adopt a multimethod approach that combines fixed-effects modeling with interpretable machine learning (XGBoost, SHAP, and time-sensitive pattern analysis). The results reveal a nonlinear and dynamic relationship between input representation fit and output complexity, indicating that both low-fit and high-fit inputs can increase output complexity through distinct mechanisms. We further demonstrate that these effects differ according to interaction depth and temporal context, suggesting that representational alignment is not static but dynamically recalibrated during iterative exchanges. The study enhances the cognitive fit theory by conceptualizing fit as an adaptive alignment process in generative environments instead of a fixed match between human cognition and system representation. The findings highlight the necessity for AI systems that support adaptive prompting and context-aware feedback to enhance collaborative generative improvement.</div></div>","PeriodicalId":48471,"journal":{"name":"Computers in Human Behavior","volume":"179 ","pages":"Article 108927"},"PeriodicalIF":8.9,"publicationDate":"2026-01-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146081670","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}

引用次数: 0

Training and oversight of algorithms in social decision-making: Algorithms with prescribed selfish defaults breed selfish decisions 社会决策中算法的训练和监督：带有预设自私默认值的算法会产生自私的决策

IF 8.9 1区心理学 Q1 PSYCHOLOGY, EXPERIMENTAL

Computers in Human Behavior

Pub Date : 2026-01-20 DOI: 10.1016/j.chb.2026.108924

Terence D. Dores Cruz , Mateus A.M. de Lucena

Human social preferences increasingly shape oversight or training data for Artificial Intelligence (AI) social decisions that affect human–human interactions. We test how algorithms with and without prescribed social preferences shape social decision-making and delegation. In an incentivised online experiment (n = 1290), participants completed a Social Value Orientation (SVO) measure as input to a decision-making algorithm, revealing their preferences for outcomes favouring oneself or an anonymous other. We manipulated whether participants (1) provided training data to an algorithm without prescribed preferences by answering the SVO without defaults or (2) oversaw algorithms with prescribed preferences by including proself/prosocial pre-selected defaults for each item. When decisions involved an algorithm, defaults were labelled as algorithmic; in a control condition, identical defaults were unlabelled. Participants’ social preferences were not significantly impacted by providing input to an algorithm without prescribed preferences (vs no defaults) nor by oversight of the algorithm with prescribed prosocial preferences (vs identical unlabelled defaults and vs the algorithm without prescribed preferences). Only providing oversight of the algorithm with prescribed proself preferences resulted in more selfish social preferences (vs the algorithm without prescribed preferences and vs the algorithm with prescribed prosocial preferences), even though participants perceived feeling less influenced by proself than prosocial defaults. Most participants delegated a second social decision-making task to the algorithm they encountered. These findings tentatively suggest that human-in-the-loop oversight, where humans can alter algorithmic suggestions, might alone fall short to address algorithmic biases, as individuals acted more selfishly when exposed to pre-existing selfish tendencies in algorithms.

人类的社会偏好越来越多地影响着人工智能（AI）社会决策的监督或训练数据，这些决策会影响人与人之间的互动。我们测试了有和没有规定的社会偏好的算法如何塑造社会决策和授权。在一项激励在线实验中（n = 1290），参与者完成了一项社会价值取向（SVO）测量，作为决策算法的输入，揭示了他们对自己或匿名他人的偏好。我们操纵了参与者(1)通过回答没有预设值的SVO来为没有预设偏好的算法提供训练数据，还是(2)通过为每个项目包括自我/亲社会预设值来监督具有预设偏好的算法。当决策涉及算法时，默认情况被标记为算法；在控制条件下，相同的默认值没有标记。参与者的社会偏好不会因为向没有预设偏好的算法提供输入而受到显著影响（相对于没有预设），也不会因为对具有预设亲社会偏好的算法进行监督而受到显著影响（相对于相同的未标记的预设和没有预设偏好的算法）。尽管参与者感觉受到自我的影响比受到亲社会的影响要小，但只提供对带有预设自我偏好的算法的监督会导致更自私的社会偏好（与没有预设偏好的算法和带有预设亲社会偏好的算法相比）。大多数参与者将第二项社会决策任务委托给他们遇到的算法。这些发现初步表明，人类在循环中的监督（人类可以改变算法建议）可能不足以解决算法偏见，因为当个人暴露于算法中存在的自私倾向时，他们会表现得更自私。

{"title":"Training and oversight of algorithms in social decision-making: Algorithms with prescribed selfish defaults breed selfish decisions","authors":"Terence D. Dores Cruz , Mateus A.M. de Lucena","doi":"10.1016/j.chb.2026.108924","DOIUrl":"10.1016/j.chb.2026.108924","url":null,"abstract":"<div><div>Human social preferences increasingly shape oversight or training data for Artificial Intelligence (AI) social decisions that affect human–human interactions. We test how algorithms with and without prescribed social preferences shape social decision-making and delegation. In an incentivised online experiment (n = 1290), participants completed a Social Value Orientation (SVO) measure as input to a decision-making algorithm, revealing their preferences for outcomes favouring oneself or an anonymous other. We manipulated whether participants (1) provided training data to an algorithm without prescribed preferences by answering the SVO without defaults or (2) oversaw algorithms with prescribed preferences by including proself/prosocial pre-selected defaults for each item. When decisions involved an algorithm, defaults were labelled as algorithmic; in a control condition, identical defaults were unlabelled. Participants’ social preferences were not significantly impacted by providing input to an algorithm without prescribed preferences (vs no defaults) nor by oversight of the algorithm with prescribed prosocial preferences (vs identical unlabelled defaults and vs the algorithm without prescribed preferences). Only providing oversight of the algorithm with prescribed proself preferences resulted in more selfish social preferences (vs the algorithm without prescribed preferences and vs the algorithm with prescribed prosocial preferences), even though participants perceived feeling less influenced by proself than prosocial defaults. Most participants delegated a second social decision-making task to the algorithm they encountered. These findings tentatively suggest that human-in-the-loop oversight, where humans can alter algorithmic suggestions, might alone fall short to address algorithmic biases, as individuals acted more selfishly when exposed to pre-existing selfish tendencies in algorithms.</div></div>","PeriodicalId":48471,"journal":{"name":"Computers in Human Behavior","volume":"179 ","pages":"Article 108924"},"PeriodicalIF":8.9,"publicationDate":"2026-01-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146045287","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}

引用次数: 0

Evaluating human-based and large language model-based content analysis: A case study on negative word-of-mouth classification in social media 评估基于人类和基于大语言模型的内容分析：社交媒体负面口碑分类的案例研究

IF 8.9 1区心理学 Q1 PSYCHOLOGY, EXPERIMENTAL

Computers in Human Behavior

Pub Date : 2026-01-20 DOI: 10.1016/j.chb.2026.108904

Yolande Yunhsiou Yang , Chih-Chien Wang , Pau-Lin Chou , Chieh-Yu Tien , Jia-Ci Wen , Nien-Hsin Chen

This main purpose of this study is to examine the effectiveness of human annotation versus large language model (LLM)-based automated content analysis in classifying negative word-of-mouth (NWOM) on social media. Using ChatGPT as the representative LLM, we evaluated its performance in categorizing comments into five sentiment types: positive, neutral, negative, irony, and boycott. Through three experiments, we compared ChatGPT's classification accuracy, precision, recall, and F1 score with human annotators under varying conditions, including the provision of classification definitions and keywords, as well as parameter adjustments such as temperature and Top_p. The findings reveal that while ChatGPT demonstrated strong recall in detecting negative comments, it struggles with nuanced classifications such as irony and boycott, where human annotators exhibit higher precision. Incorporating explicit definitions and keywords significantly enhanced ChatGPT's classification accuracy and consistency, particularly in distinguishing boycott-related comments. However, adjustments to temperature and Top_p yield only marginal improvements. These results highlight the potential of integrating LLMs with human annotation to optimize large-scale content analysis while addressing the limitations of automated sentiment classification. The study provides theoretical and practical insights into AI-assisted content analysis, with implications for social media monitoring, brand crisis management, and influencer engagement strategies.

本研究的主要目的是检验人工注释与基于大语言模型（LLM）的自动内容分析在对社交媒体上的负面口碑（NWOM）进行分类方面的有效性。使用ChatGPT作为代表性LLM，我们评估了它在将评论分为五种情绪类型方面的表现：积极，中立，消极，讽刺和抵制。通过三个实验，我们比较了ChatGPT在不同条件下的分类准确率、精密度、召回率和F1分数与人类注释器的比较，包括提供分类定义和关键词，以及温度和Top_p等参数的调整。研究结果表明，虽然ChatGPT在检测负面评论方面表现出很强的召回能力，但它在诸如讽刺和抵制等细微分类方面表现得很差，而人类注释者在这些方面表现出更高的准确性。结合明确的定义和关键词显著提高了ChatGPT的分类准确性和一致性，特别是在区分抵制相关评论方面。然而，温度和Top_p的调整只产生了微小的改善。这些结果突出了将llm与人类注释集成在一起以优化大规模内容分析的潜力，同时解决了自动情感分类的局限性。该研究为人工智能辅助内容分析提供了理论和实践见解，对社交媒体监控、品牌危机管理和网红参与策略具有重要意义。

{"title":"Evaluating human-based and large language model-based content analysis: A case study on negative word-of-mouth classification in social media","authors":"Yolande Yunhsiou Yang , Chih-Chien Wang , Pau-Lin Chou , Chieh-Yu Tien , Jia-Ci Wen , Nien-Hsin Chen","doi":"10.1016/j.chb.2026.108904","DOIUrl":"10.1016/j.chb.2026.108904","url":null,"abstract":"<div><div>This main purpose of this study is to examine the effectiveness of human annotation versus large language model (LLM)-based automated content analysis in classifying negative word-of-mouth (NWOM) on social media. Using ChatGPT as the representative LLM, we evaluated its performance in categorizing comments into five sentiment types: positive, neutral, negative, irony, and boycott. Through three experiments, we compared ChatGPT's classification accuracy, precision, recall, and F1 score with human annotators under varying conditions, including the provision of classification definitions and keywords, as well as parameter adjustments such as temperature and Top_p. The findings reveal that while ChatGPT demonstrated strong recall in detecting negative comments, it struggles with nuanced classifications such as irony and boycott, where human annotators exhibit higher precision. Incorporating explicit definitions and keywords significantly enhanced ChatGPT's classification accuracy and consistency, particularly in distinguishing boycott-related comments. However, adjustments to temperature and Top_p yield only marginal improvements. These results highlight the potential of integrating LLMs with human annotation to optimize large-scale content analysis while addressing the limitations of automated sentiment classification. The study provides theoretical and practical insights into AI-assisted content analysis, with implications for social media monitoring, brand crisis management, and influencer engagement strategies.</div></div>","PeriodicalId":48471,"journal":{"name":"Computers in Human Behavior","volume":"179 ","pages":"Article 108904"},"PeriodicalIF":8.9,"publicationDate":"2026-01-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146081669","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}

引用次数: 0

From content consumers to content creators: Farmers using TikTok in northern Vietnam's mountainous regions 从内容消费者到内容创造者：越南北部山区使用TikTok的农民

IF 8.9 1区心理学 Q1 PSYCHOLOGY, EXPERIMENTAL

Computers in Human Behavior

Pub Date : 2026-01-19 DOI: 10.1016/j.chb.2026.108925

Nguyen Ngoc Quynh , Nguyen Khanh Doanh

Farmers in Vietnam's Northern mountainous region are increasingly exposed to social media platforms like TikTok, yet little examine their transition from content consumers to active content creators (TikTokers). To explore this phenomenon, we integrate the Uses and Gratifications Theory (U&G), the Technology Acceptance Model (TAM), and the concept of normative barriers. In a structured study, we examined farmers' intentions to become TikTokers based on (i) perceived usefulness and ease of use, (ii) non-utilitarian gratifications from being a viewer, and (iii) the influence of traditional norms. Our findings reveal three key patterns. First, perceived usefulness and ease of use significantly increased farmers' intention to become TikTokers. Second, gratifications such as entertainment, social connection, and information seeking gained in the viewer role enhanced perceptions of TikTok's utility and usability in the creator role. Third, normative barriers rooted in cultural traditions inhibited intention, but were mitigated by satisfaction with information and strong viewer engagement. These results suggest that while socio-cultural constraints persist, they are not insurmountable. Encouraging content creation among farmers thus requires not only intuitive platform design and accessible information, but also a sustained focus on fulfilling users' emotional and social needs.

越南北部山区的农民越来越多地接触到TikTok等社交媒体平台，但很少有人审视他们从内容消费者到活跃内容创造者（TikTok）的转变。为了探索这一现象，我们整合了使用与满足理论（U&；G）、技术接受模型（TAM）和规范障碍的概念。在一项结构化研究中，我们基于(i)感知到的有用性和易用性，（ii）作为观众的非功利满足，以及（iii）传统规范的影响，研究了农民成为TikTokers的意图。我们的发现揭示了三个关键模式。首先，感知到的有用性和易用性显著增加了农民成为tiktok用户的意愿。其次，在观众角色中获得的娱乐、社会联系和信息寻求等满足感增强了TikTok在创作者角色中的实用性和可用性。第三，植根于文化传统的规范障碍抑制了意图，但对信息的满意和强烈的观众参与减轻了这一障碍。这些结果表明，虽然社会文化制约因素仍然存在，但它们并非不可克服。因此，鼓励农民进行内容创作不仅需要直观的平台设计和可访问的信息，还需要持续关注满足用户的情感和社交需求。

{"title":"From content consumers to content creators: Farmers using TikTok in northern Vietnam's mountainous regions","authors":"Nguyen Ngoc Quynh , Nguyen Khanh Doanh","doi":"10.1016/j.chb.2026.108925","DOIUrl":"10.1016/j.chb.2026.108925","url":null,"abstract":"<div><div>Farmers in Vietnam's Northern mountainous region are increasingly exposed to social media platforms like TikTok, yet little examine their transition from content consumers to active content creators (TikTokers). To explore this phenomenon, we integrate the Uses and Gratifications Theory (U&G), the Technology Acceptance Model (TAM), and the concept of normative barriers. In a structured study, we examined farmers' intentions to become TikTokers based on (i) perceived usefulness and ease of use, (ii) non-utilitarian gratifications from being a viewer, and (iii) the influence of traditional norms. Our findings reveal three key patterns. First, perceived usefulness and ease of use significantly increased farmers' intention to become TikTokers. Second, gratifications such as entertainment, social connection, and information seeking gained in the viewer role enhanced perceptions of TikTok's utility and usability in the creator role. Third, normative barriers rooted in cultural traditions inhibited intention, but were mitigated by satisfaction with information and strong viewer engagement. These results suggest that while socio-cultural constraints persist, they are not insurmountable. Encouraging content creation among farmers thus requires not only intuitive platform design and accessible information, but also a sustained focus on fulfilling users' emotional and social needs.</div></div>","PeriodicalId":48471,"journal":{"name":"Computers in Human Behavior","volume":"178 ","pages":"Article 108925"},"PeriodicalIF":8.9,"publicationDate":"2026-01-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146024718","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}

引用次数: 0

Personal cancer worry and Systemic Cancer Concern: Pathways to health behaviors via social media and emotional well-being 个人癌症担忧和系统癌症关注：通过社交媒体和情感健康的健康行为途径

IF 8.9 1区心理学 Q1 PSYCHOLOGY, EXPERIMENTAL

Computers in Human Behavior

Pub Date : 2026-01-17 DOI: 10.1016/j.chb.2026.108918

Chi-Chin Hsiao , Hsuan-Wei Lee

Cancer worry and concern are associated with how individuals engage with and interpret digital health environments, yet existing research has not distinguished between personal worry about individual risk and systemic concerns regarding institutional responses. This study examines how personal worry and systemic cancer concern relate to digital health behaviors through distinct affective and trust-related pathways. Drawing on an integrated theoretical framework, we analyze nationally representative U.S. data (HINTS 6; N = 6252) using structural equation modeling. Personal cancer worry is associated with increased social media engagement and reduced emotional well-being, with emotional vulnerability in particular aligning with greater involvement in harmful behaviors (e.g., smoking), whereas social media engagement shows more selective behavioral associations. In contrast, systemic cancer concern, marked by institutional distrust, corresponds to limited or passive digital engagement and diminished emotional well-being, patterns that relate to lower participation in positive health behaviors. Social media engagement is associated with perceiving less health misinformation, showing modest behavioral associations. Overall, the findings distinguish two forms of cancer-related psychological responses: a pattern of active but emotionally vulnerable engagement associated with personal worry, and a pattern of trust-sensitive, low-engagement interpretation associated with systemic concern. Findings inform psychologically responsive digital health design. Emotion-sensitive features may support users with heightened vulnerability, whereas credibility cues may address institutional distrust.

癌症的担忧和关注与个人如何参与和解释数字健康环境有关，然而现有的研究并没有区分个人对个人风险的担忧和对机构反应的系统性担忧。本研究考察了个人担忧和系统性癌症担忧如何通过不同的情感和信任相关途径与数字健康行为相关。利用一个完整的理论框架，我们使用结构方程模型分析了具有全国代表性的美国数据（HINTS 6; N = 6252）。个人癌症担忧与社交媒体参与度的增加和情绪幸福感的降低有关，尤其是情绪脆弱性与更多的有害行为（如吸烟）有关，而社交媒体参与度则显示出更多的选择性行为关联。相比之下，以机构不信任为特征的系统性癌症关注，对应于有限或被动的数字参与和情绪幸福感的减少，这些模式与积极健康行为的参与度较低有关。社交媒体参与与感知较少的健康错误信息有关，表现出适度的行为关联。总体而言，研究结果区分了两种与癌症相关的心理反应：一种是与个人担忧相关的积极但情感上脆弱的投入模式，另一种是与系统担忧相关的信任敏感、低投入解释模式。研究结果为心理响应型数字健康设计提供了依据。情绪敏感的特征可能支持具有高度脆弱性的用户，而可信度线索可能解决机构不信任。

{"title":"Personal cancer worry and Systemic Cancer Concern: Pathways to health behaviors via social media and emotional well-being","authors":"Chi-Chin Hsiao , Hsuan-Wei Lee","doi":"10.1016/j.chb.2026.108918","DOIUrl":"10.1016/j.chb.2026.108918","url":null,"abstract":"<div><div>Cancer worry and concern are associated with how individuals engage with and interpret digital health environments, yet existing research has not distinguished between personal worry about individual risk and systemic concerns regarding institutional responses. This study examines how personal worry and systemic cancer concern relate to digital health behaviors through distinct affective and trust-related pathways. Drawing on an integrated theoretical framework, we analyze nationally representative U.S. data (HINTS 6; N = 6252) using structural equation modeling. Personal cancer worry is associated with increased social media engagement and reduced emotional well-being, with emotional vulnerability in particular aligning with greater involvement in harmful behaviors (e.g., smoking), whereas social media engagement shows more selective behavioral associations. In contrast, systemic cancer concern, marked by institutional distrust, corresponds to limited or passive digital engagement and diminished emotional well-being, patterns that relate to lower participation in positive health behaviors. Social media engagement is associated with perceiving less health misinformation, showing modest behavioral associations. Overall, the findings distinguish two forms of cancer-related psychological responses: a pattern of active but emotionally vulnerable engagement associated with personal worry, and a pattern of trust-sensitive, low-engagement interpretation associated with systemic concern. Findings inform psychologically responsive digital health design. Emotion-sensitive features may support users with heightened vulnerability, whereas credibility cues may address institutional distrust.</div></div>","PeriodicalId":48471,"journal":{"name":"Computers in Human Behavior","volume":"178 ","pages":"Article 108918"},"PeriodicalIF":8.9,"publicationDate":"2026-01-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146024716","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}

引用次数: 0

Social bonding through expressed needs: Insights from an identity-shape matching task 通过表达需求的社会联系：来自身份形状匹配任务的见解

IF 8.9 1区心理学 Q1 PSYCHOLOGY, EXPERIMENTAL

Computers in Human Behavior

Pub Date : 2026-01-17 DOI: 10.1016/j.chb.2026.108917

Su-Ling Yeh , Ti-Fan Hung , Te-Yi Hsieh , Chia-Huei Tseng

This study advances understanding of human-robot social bonds using an identity-shape matching task, where participants associated geometric shapes (circle, square, triangle) with identities (best friend, robot partner, stranger) and response times for matched pairs indicated processing advantages. Across three sets of experiments (1A–1C, 2A–2B, 3A–3B), we tested whether partner-advantage effects generalize to human–robot interaction (1A–1C), whether robots' functional assistance enhances human–robot partnership and elicits friend-like processing advantages (2A–2B), and whether a robot's need for human help further strengthens social connectedness (3A–3B). Experiments 1A–1C demonstrated that brief encounters with a single robot, without collaboration or physical presence, elicited partner-advantage effects, though less pronounced than for best friends. This effect was maintained in Experiments 2A–2B when the robot provided functional assistance, irrespective of its physical presence. Uniquely, Experiments 3A–3B revealed that the robot expressing needs without companionship was perceived as socially distant, like a stranger, but when expressing minor needs with reciprocal companionship, it achieved perceptual prioritization comparable to best friends. These findings validate the identity-shape matching task as a robust method to quantify human-robot relationships and highlight the unique role of expressed needs with companionship in fostering friend-like social bonds.

本研究通过身份-形状匹配任务推进了对人类-机器人社会纽带的理解，在该任务中，参与者将几何形状（圆形、正方形、三角形）与身份（最好的朋友、机器人伙伴、陌生人）联系起来，匹配对的反应时间表明处理优势。通过三组实验（1A-1C, 2A-2B, 3A-3B），我们测试了伙伴优势效应是否会推广到人机交互（1A-1C），机器人的功能协助是否会增强人机伙伴关系并引发类似朋友的加工优势（2A-2B），以及机器人对人类帮助的需求是否会进一步加强社会联系（3A-3B）。实验1A-1C表明，与单个机器人的短暂接触，在没有合作或实际存在的情况下，会产生伴侣优势效应，尽管没有最好的朋友那么明显。在实验2A-2B中，当机器人提供功能性协助时，无论其物理存在与否，这种效果都保持不变。独特的是，实验3A-3B显示，在没有陪伴的情况下表达需求的机器人被认为是社交疏远的，就像一个陌生人，但当表达次要需求时，有相互的陪伴，它实现了与最好的朋友相当的感知优先级。这些发现证实了身份-形状匹配任务是一种量化人类与机器人关系的强大方法，并强调了表达需求与陪伴在培养朋友般的社会纽带方面的独特作用。

{"title":"Social bonding through expressed needs: Insights from an identity-shape matching task","authors":"Su-Ling Yeh , Ti-Fan Hung , Te-Yi Hsieh , Chia-Huei Tseng","doi":"10.1016/j.chb.2026.108917","DOIUrl":"10.1016/j.chb.2026.108917","url":null,"abstract":"<div><div>This study advances understanding of human-robot social bonds using an identity-shape matching task, where participants associated geometric shapes (circle, square, triangle) with identities (best friend, robot partner, stranger) and response times for matched pairs indicated processing advantages. Across three sets of experiments (1A–1C, 2A–2B, 3A–3B), we tested whether partner-advantage effects generalize to human–robot interaction (1A–1C), whether robots' functional assistance enhances human–robot partnership and elicits friend-like processing advantages (2A–2B), and whether a robot's need for human help further strengthens social connectedness (3A–3B). Experiments 1A–1C demonstrated that brief encounters with a single robot, without collaboration or physical presence, elicited partner-advantage effects, though less pronounced than for best friends. This effect was maintained in Experiments 2A–2B when the robot provided functional assistance, irrespective of its physical presence. Uniquely, Experiments 3A–3B revealed that the robot expressing needs without companionship was perceived as socially distant, like a stranger, but when expressing minor needs with reciprocal companionship, it achieved perceptual prioritization comparable to best friends. These findings validate the identity-shape matching task as a robust method to quantify human-robot relationships and highlight the unique role of expressed needs with companionship in fostering friend-like social bonds.</div></div>","PeriodicalId":48471,"journal":{"name":"Computers in Human Behavior","volume":"178 ","pages":"Article 108917"},"PeriodicalIF":8.9,"publicationDate":"2026-01-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146024680","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}

引用次数: 0

Bounty design in online communities: Uneven effects on prosocial behavior across user groups 在线社区的赏金设计：不同用户群体对亲社会行为的不均匀影响

IF 8.9 1区心理学 Q1 PSYCHOLOGY, EXPERIMENTAL

Computers in Human Behavior

Pub Date : 2026-01-14 DOI: 10.1016/j.chb.2026.108922

Jing Xu , Jianwei Liu , Kee-Hung Lai , Xu Gao , Yahe Yu , Dong Jing

User-led bounty design has emerged as a novel mechanism in online communities, enabling users to offer monetary incentives to boost engagement. Nevertheless, as an extrinsic stimulus, bounties can exert complex influences on users' prosocial behavior, affecting the motivation for user participation on online platforms and thereby generating spillover or crowding-out effects. This study employs a difference-in-differences (DID) model to examine how bounty design affects prosocial behavior across different user groups in online communities. The results reveal a significant decline in overall prosocial behavior, particularly among users with high extrinsic motivation. Conversely, users with high intrinsic motivation exhibit increased prosocial behavior. Additionally, bounty amount and historical comment sentiment significantly moderate these effects. Innovatively, this study focuses on the heterogeneous outcomes of bounty incentives on groups with different motivational levels, indicating that motivational level is a crucial factor influencing the effectiveness of user incentives—users with different motivational levels may even exhibit completely opposite responses to bounty incentives. This study provides insights into the complex mechanisms of bounty design and emphasizes the importance of accounting for different user groups in online community design. These findings may also help inform the optimization of incentive mechanisms.

用户主导的赏金设计已经成为在线社区的一种新机制，允许用户提供金钱奖励来提高参与度。然而，作为一种外在刺激，奖励会对用户的亲社会行为产生复杂的影响，影响用户参与网络平台的动机，从而产生溢出效应或挤出效应。本研究采用差异中的差异（DID）模型来研究赏金设计如何影响网络社区中不同用户群体的亲社会行为。结果显示，整体亲社会行为显著下降，特别是在具有高外在动机的用户中。相反，具有高内在动机的用户表现出更多的亲社会行为。此外，赏金数额和历史评论情绪显著调节了这些影响。创新之处在于，本研究关注赏金激励对不同动机水平群体的异质性结果，表明动机水平是影响用户激励有效性的关键因素，不同动机水平的用户甚至可能对赏金激励表现出完全相反的反应。这项研究为赏金设计的复杂机制提供了见解，并强调了在在线社区设计中考虑不同用户群体的重要性。这些发现也有助于激励机制的优化。

{"title":"Bounty design in online communities: Uneven effects on prosocial behavior across user groups","authors":"Jing Xu , Jianwei Liu , Kee-Hung Lai , Xu Gao , Yahe Yu , Dong Jing","doi":"10.1016/j.chb.2026.108922","DOIUrl":"10.1016/j.chb.2026.108922","url":null,"abstract":"<div><div>User-led bounty design has emerged as a novel mechanism in online communities, enabling users to offer monetary incentives to boost engagement. Nevertheless, as an extrinsic stimulus, bounties can exert complex influences on users' prosocial behavior, affecting the motivation for user participation on online platforms and thereby generating spillover or crowding-out effects. This study employs a difference-in-differences (DID) model to examine how bounty design affects prosocial behavior across different user groups in online communities. The results reveal a significant decline in overall prosocial behavior, particularly among users with high extrinsic motivation. Conversely, users with high intrinsic motivation exhibit increased prosocial behavior. Additionally, bounty amount and historical comment sentiment significantly moderate these effects. Innovatively, this study focuses on the heterogeneous outcomes of bounty incentives on groups with different motivational levels, indicating that motivational level is a crucial factor influencing the effectiveness of user incentives—users with different motivational levels may even exhibit completely opposite responses to bounty incentives. This study provides insights into the complex mechanisms of bounty design and emphasizes the importance of accounting for different user groups in online community design. These findings may also help inform the optimization of incentive mechanisms.</div></div>","PeriodicalId":48471,"journal":{"name":"Computers in Human Behavior","volume":"178 ","pages":"Article 108922"},"PeriodicalIF":8.9,"publicationDate":"2026-01-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146024717","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}

引用次数: 0

The impact of generative artificial intelligence usage on job performance through job crafting and work engagement: Does digital competence matter? 生成式人工智能的使用对工作绩效的影响：数字能力重要吗？

IF 8.9 1区心理学 Q1 PSYCHOLOGY, EXPERIMENTAL

Computers in Human Behavior

Pub Date : 2026-01-14 DOI: 10.1016/j.chb.2026.108921

Bui Nhat Vuong

The objective of this study is to investigate the impact of generative artificial intelligence (Gen-AI) usage on job performance (JP), focusing on the mediating roles of job crafting (JC) and work engagement (WE). Additionally, the study examines the moderating effect of digital competence (DC). Data were collected through surveys from 758 employees working in small and medium-sized enterprises (SMEs) in Vietnam, and analyzed using the partial least squares structural equation modeling (PLS-SEM) technique. The research findings indicate that Gen-AI usage positively influences job performance. This relationship is partially mediated by job crafting and work engagement. Furthermore, the effects of Gen-AI usage on job crafting, work engagement, and job performance are amplified among employees with strong digital competence. Finally, the study presents several management implications for SME managers in Vietnam, highlighting the urgent need to promote Gen-AI usage within their organizations and to enhance digital capabilities to improve employee job performance.

本研究的目的是探讨生成式人工智能（Gen-AI）的使用对工作绩效（JP）的影响，重点关注工作制作（JC）和工作投入（WE）的中介作用。此外，研究还考察了数字能力（DC）的调节作用。通过对越南中小企业（sme） 758名员工的调查收集数据，并使用偏最小二乘结构方程模型（PLS-SEM）技术进行分析。研究结果表明，使用Gen-AI会对工作绩效产生积极影响。这种关系部分由工作塑造和工作投入来调节。此外，在拥有强大数字能力的员工中，使用Gen-AI对工作制定、工作投入和工作绩效的影响会被放大。最后，该研究为越南的中小企业管理者提出了几点管理启示，强调了在其组织内促进Gen-AI使用并增强数字化能力以提高员工工作绩效的迫切需要。

引用次数: 0

Investigating perceived trust and utility of balanced news chatbots among individuals with varying conspiracy beliefs 调查具有不同阴谋信仰的个人对平衡新闻聊天机器人的感知信任和效用

IF 8.9 1区心理学 Q1 PSYCHOLOGY, EXPERIMENTAL

Computers in Human Behavior

Pub Date : 2026-01-13 DOI: 10.1016/j.chb.2026.108920

Shreya Dubey , Paul E. Ketelaar , Tilman Dingler , Hannah K. Peetz , Hein T. van Schie

In the current media landscape, various ideas and narratives gain traction, influenced by the dynamics of selective exposure and a decline in trust in traditional information sources. This trend holds the potential to cultivate polarisation of perspectives, as individuals actively seek information that resonates with their existing attitudes. Hence, diversifying information that is available online can encourage users to engage with multiple perspectives, especially when provided by a trustworthy source. This paper presents findings from two studies which compared individuals with a higher belief in generic conspiracy theories (Study 1; n = 84) and specific conspiracy beliefs on climate change (Study 2; n = 23) to those with lower conspiracy beliefs (n_{study 1} = 93; n_{study 2} = 35) on perceived trustworthiness and usefulness of the so called ‘balanced news chatbots’. These chatbots present a selection of opposing alternative and mainstream perspectives on topics of societal divide like climate change. We found that participants from both groups responded positively to the balanced news chatbot. Trust and perceived usefulness were identified to be key indicators of a positive attitude towards and high intentions of using such a chatbot, corroborating the acceptance of balanced news chatbots as a potential tool to reduce polarisation and conflict, piercing existing information bubbles. In both studies we also found that participants with higher conspiratorial beliefs responded even more positively to the balanced news chatbot than individuals with lower conspiratorial beliefs. We conclude that balanced chatbots are promising as a trusted source of diversified information for individuals with varying levels of conspiracy beliefs.

在当前的媒体环境中，受选择性曝光的动态和对传统信息来源信任度下降的影响，各种想法和叙述获得了吸引力。这种趋势有可能培养观点的两极分化，因为个人积极寻求与他们现有态度产生共鸣的信息。因此，在线提供多样化的信息可以鼓励用户从多个角度进行接触，尤其是在可靠来源提供信息的情况下。本文介绍了两项研究的结果，这两项研究比较了在所谓的“平衡新闻聊天机器人”的感知可信度和有用性方面，对一般阴谋论（研究1，n = 84）和对气候变化的特定阴谋论（研究2，n = 23）持较高信念的个体与对阴谋论持较低信念的个体（研究1 = 93；研究2 = 35）。这些聊天机器人就气候变化等社会分歧话题提出了一系列对立的另类观点和主流观点。我们发现，两组参与者对平衡新闻聊天机器人的反应都是积极的。信任和感知有用性被确定为对使用这种聊天机器人的积极态度和高度意图的关键指标，证实了平衡新闻聊天机器人作为减少两极分化和冲突的潜在工具的接受，穿透现有的信息泡沫。在这两项研究中，我们还发现，与阴谋论信念较低的人相比，阴谋论信念较高的参与者对平衡新闻聊天机器人的反应更为积极。我们的结论是，平衡的聊天机器人有望成为具有不同程度阴谋信仰的个人的多样化信息的可靠来源。

{"title":"Investigating perceived trust and utility of balanced news chatbots among individuals with varying conspiracy beliefs","authors":"Shreya Dubey , Paul E. Ketelaar , Tilman Dingler , Hannah K. Peetz , Hein T. van Schie","doi":"10.1016/j.chb.2026.108920","DOIUrl":"10.1016/j.chb.2026.108920","url":null,"abstract":"<div><div>In the current media landscape, various ideas and narratives gain traction, influenced by the dynamics of selective exposure and a decline in trust in traditional information sources. This trend holds the potential to cultivate polarisation of perspectives, as individuals actively seek information that resonates with their existing attitudes. Hence, diversifying information that is available online can encourage users to engage with multiple perspectives, especially when provided by a trustworthy source. This paper presents findings from two studies which compared individuals with a higher belief in generic conspiracy theories (Study 1; <em>n</em> = 84) and specific conspiracy beliefs on climate change (Study 2; <em>n</em> = 23) to those with lower conspiracy beliefs (<em>n</em><sub><em>study 1</em></sub> = 93; <em>n</em><sub><em>study 2</em></sub> = 35) on perceived trustworthiness and usefulness of the so called ‘balanced news chatbots’. These chatbots present a selection of opposing alternative and mainstream perspectives on topics of societal divide like climate change. We found that participants from both groups responded positively to the balanced news chatbot. Trust and perceived usefulness were identified to be key indicators of a positive attitude towards and high intentions of using such a chatbot, corroborating the acceptance of balanced news chatbots as a potential tool to reduce polarisation and conflict, piercing existing information bubbles. In both studies we also found that participants with higher conspiratorial beliefs responded even more positively to the balanced news chatbot than individuals with lower conspiratorial beliefs. We conclude that balanced chatbots are promising as a trusted source of diversified information for individuals with varying levels of conspiracy beliefs.</div></div>","PeriodicalId":48471,"journal":{"name":"Computers in Human Behavior","volume":"178 ","pages":"Article 108920"},"PeriodicalIF":8.9,"publicationDate":"2026-01-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146024714","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}

引用次数: 0