Computers in Human Behavior最新文献_第3页

Adolescent social media use profiles: A longitudinal study of friendship quality and socio-motivational factors 青少年社交媒体使用概况：友谊质量和社会动机因素的纵向研究

IF 8.9 1区心理学 Q1 PSYCHOLOGY, EXPERIMENTAL

Computers in Human Behavior

Pub Date : 2025-12-04 DOI: 10.1016/j.chb.2025.108880

Federica Angelini , Ina M. Koning , Gianluca Gini , Claudia Marino , Regina J.J.M. van den Eijnden

Early adolescence is a crucial period for social and emotional development, with relationships with close friends shaping individual and social well-being. Today, these interactions occur both offline and online, requiring nuanced theoretical models to capture their complexities. Using a three-wave longitudinal, person-centered approach, this study aims to: (1) identify distinct social media use profiles based on specific online activities, (2) examine socio-motivational factors predicting profile membership, and (3) test the role of profile membership in the longitudinal stability of perceived quality of close friendship.

A sample of 1211 Dutch adolescents (M_age = 13.1, SD = 0.80; 53.8 % males) participated in three waves of data collection between 2016 and 2018. Four profiles of social media users emerged (1 = Low users, 2 = High self-disclosing users; 3 = High self-oriented users, 4 = All-round users), each explained by different socio-motivational factors. In particular, a low quality of perceived close friendships at T1 increased the odds of being Low social media users, characterized by lower social media activity at T2. Furthermore, both Low and All-round social media users showed stable perceived friendship quality over time, maintaining consistently low and moderate-to-high levels, respectively. These findings highlight the foundational role of peer interaction in shaping online behaviors and social experiences. While social media can reinforce friendships, its benefits largely rely on pre-existing relationships. Adolescents with weaker friendships may struggle to use social media to build meaningful relationships, missing opportunities to enhance their connections and potentially further deepening social difficulties.

青少年早期是社会和情感发展的关键时期，与亲密朋友的关系塑造了个人和社会福祉。今天，这些互动发生在线下和线上，需要细致的理论模型来捕捉它们的复杂性。本研究采用三波纵向、以人为本的方法，旨在：(1)根据特定的在线活动识别不同的社交媒体使用概况；(2)研究预测概况成员的社会动机因素；(3)测试概况成员在亲密友谊感知质量纵向稳定性中的作用。1211名荷兰青少年（Mage = 13.1, SD = 0.80, 53.8%为男性）参与了2016年至2018年的三波数据收集。出现了四种社交媒体用户类型（1 =低用户，2 =高自我披露用户，3 =高自我导向用户，4 =全方位用户），每种类型都有不同的社会动机因素解释。特别是，在T1阶段，低质量的亲密友谊增加了成为低社交媒体用户的几率，其特征是在T2阶段社交媒体活动较低。此外，随着时间的推移，低社交媒体用户和全方位社交媒体用户都表现出稳定的感知友谊质量，分别保持在低水平和中高水平。这些发现强调了同伴互动在塑造在线行为和社交体验方面的基础作用。虽然社交媒体可以加强友谊，但它的好处主要依赖于已有的关系。友谊较弱的青少年可能很难利用社交媒体建立有意义的关系，失去了加强联系的机会，并可能进一步加深社交困难。

{"title":"Adolescent social media use profiles: A longitudinal study of friendship quality and socio-motivational factors","authors":"Federica Angelini , Ina M. Koning , Gianluca Gini , Claudia Marino , Regina J.J.M. van den Eijnden","doi":"10.1016/j.chb.2025.108880","DOIUrl":"10.1016/j.chb.2025.108880","url":null,"abstract":"<div><div>Early adolescence is a crucial period for social and emotional development, with relationships with close friends shaping individual and social well-being. Today, these interactions occur both offline and online, requiring nuanced theoretical models to capture their complexities. Using a three-wave longitudinal, person-centered approach, this study aims to: (1) identify distinct social media use profiles based on specific online activities, (2) examine socio-motivational factors predicting profile membership, and (3) test the role of profile membership in the longitudinal stability of perceived quality of close friendship.</div><div>A sample of 1211 Dutch adolescents (<em>M</em><sub><em>age</em></sub> = 13.1, <em>SD</em> = 0.80; 53.8 % males) participated in three waves of data collection between 2016 and 2018. Four profiles of social media users emerged (1 = Low users, 2 = High self-disclosing users; 3 = High self-oriented users, 4 = All-round users), each explained by different socio-motivational factors. In particular, a low quality of perceived close friendships at T1 increased the odds of being Low social media users, characterized by lower social media activity at T2. Furthermore, both Low and All-round social media users showed stable perceived friendship quality over time, maintaining consistently low and moderate-to-high levels, respectively. These findings highlight the foundational role of peer interaction in shaping online behaviors and social experiences. While social media can reinforce friendships, its benefits largely rely on pre-existing relationships. Adolescents with weaker friendships may struggle to use social media to build meaningful relationships, missing opportunities to enhance their connections and potentially further deepening social difficulties.</div></div>","PeriodicalId":48471,"journal":{"name":"Computers in Human Behavior","volume":"177 ","pages":"Article 108880"},"PeriodicalIF":8.9,"publicationDate":"2025-12-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145685068","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

Science communication in social Media: Analysis of success on TikTok, Instagram, and YouTube across scientific disciplines 社交媒体中的科学传播：TikTok、Instagram和YouTube跨科学领域的成功分析

IF 8.9 1区心理学 Q1 PSYCHOLOGY, EXPERIMENTAL

Computers in Human Behavior

Pub Date : 2025-12-04 DOI: 10.1016/j.chb.2025.108866

Montserrat Aiger, Carmen Elboj, Raquel Lozano-Blasco, Marian Acero-Ferrero

引用次数: 0

Game on or gone too far? Executive functioning and implicit sequence learning in problematic vs. recreational gamers 游戏开始了还是走得太远了？问题型和休闲型玩家的执行功能和内隐序列学习

IF 8.9 1区心理学 Q1 PSYCHOLOGY, EXPERIMENTAL

Computers in Human Behavior

Pub Date : 2025-12-03 DOI: 10.1016/j.chb.2025.108878

Krisztina Berta , Zsuzsanna Viktória Pesthy , Teodóra Vékony , Bence Csaba Farkas , Orsolya Király , Zsolt Demetrovics , Dezső Németh , Bernadette Kun

Video gaming often sparks controversy, though negative effects are mainly linked to gaming disorder, not gaming itself. Research shows that gaming disorder correlates with reduced executive functioning and greater reliance on habitual processes, while recreational gaming may relate to enhanced cognitive functions. However, comprehensive comparisons of cognitive profiles across gaming behaviors remain scarce. We aimed to compare the cognitive functioning of non-gamers (NG), recreational gamers (RG), and gaming disorder risk individuals (GDR). Based on the Ten-Item Internet Gaming Disorder Test scores, 114 participants were classified into these three groups. Executive functions were assessed with the Go/No-Go, Counting Span, Digit Span, Card Sorting, 1-back, and 2-back tasks; implicit sequence learning with the Alternating Serial Reaction Time task. Group effects were adjusted for weekly gaming time. The GDR group showed reduced working memory, performing worse on the Digit Span task than the NG group, and worse on the Counting Span task than both NG and RG groups. Conversely, the RG group displayed enhanced attention-related performance compared to the NG group. No group differences emerged in implicit sequence learning. Interaction analyses revealed a negative relationship between implicit sequence learning and inhibitory control/updating across groups, while a positive link between working memory and implicit sequence learning in NG and GDR groups suggests possible compensatory mechanisms. This study underscores that cognitive impairments are linked to gaming disorder rather than gaming time itself, while recreational gaming may offer cognitive benefits. These findings provide insights into the distinct cognitive profiles of recreational gaming and gaming disorder.

电子游戏经常引发争议，尽管负面影响主要与游戏障碍有关，而不是游戏本身。研究表明，游戏障碍与执行功能下降和对习惯过程的更多依赖有关，而休闲游戏可能与增强认知功能有关。然而，对于不同游戏行为的认知特征的综合比较仍然很少。我们的目的是比较非游戏玩家（NG）、休闲游戏玩家（RG）和游戏障碍风险个体（GDR）的认知功能。根据十项网络游戏障碍测试分数，114名参与者被分为这三组。执行功能通过Go/No-Go、计数广度、数字广度、卡片分类、1-back和2-back任务进行评估；交替序列反应时间任务的内隐序列学习。小组效应根据每周游戏时间进行调整。GDR组表现出工作记忆减少，在数字跨度任务上的表现比NG组差，在计数跨度任务上的表现比NG组和RG组都差。相反，与NG组相比，RG组表现出更强的注意力相关表现。内隐序列学习不存在组间差异。相互作用分析表明，内隐序列学习与抑制控制/更新之间存在负相关关系，而工作记忆与内隐序列学习在NG和GDR组之间存在正相关关系，表明可能存在代偿机制。这项研究强调，认知障碍与游戏障碍有关，而不是与游戏时间本身有关，而娱乐性游戏可能对认知有好处。这些发现为休闲游戏和游戏障碍的不同认知特征提供了见解。

{"title":"Game on or gone too far? Executive functioning and implicit sequence learning in problematic vs. recreational gamers","authors":"Krisztina Berta , Zsuzsanna Viktória Pesthy , Teodóra Vékony , Bence Csaba Farkas , Orsolya Király , Zsolt Demetrovics , Dezső Németh , Bernadette Kun","doi":"10.1016/j.chb.2025.108878","DOIUrl":"10.1016/j.chb.2025.108878","url":null,"abstract":"<div><div>Video gaming often sparks controversy, though negative effects are mainly linked to gaming disorder, not gaming itself. Research shows that gaming disorder correlates with reduced executive functioning and greater reliance on habitual processes, while recreational gaming may relate to enhanced cognitive functions. However, comprehensive comparisons of cognitive profiles across gaming behaviors remain scarce. We aimed to compare the cognitive functioning of non-gamers (NG), recreational gamers (RG), and gaming disorder risk individuals (GDR). Based on the Ten-Item Internet Gaming Disorder Test scores, 114 participants were classified into these three groups. Executive functions were assessed with the Go/No-Go, Counting Span, Digit Span, Card Sorting, 1-back, and 2-back tasks; implicit sequence learning with the Alternating Serial Reaction Time task. Group effects were adjusted for weekly gaming time. The GDR group showed reduced working memory, performing worse on the Digit Span task than the NG group, and worse on the Counting Span task than both NG and RG groups. Conversely, the RG group displayed enhanced attention-related performance compared to the NG group. No group differences emerged in implicit sequence learning. Interaction analyses revealed a negative relationship between implicit sequence learning and inhibitory control/updating across groups, while a positive link between working memory and implicit sequence learning in NG and GDR groups suggests possible compensatory mechanisms. This study underscores that cognitive impairments are linked to gaming disorder rather than gaming time itself, while recreational gaming may offer cognitive benefits. These findings provide insights into the distinct cognitive profiles of recreational gaming and gaming disorder.</div></div>","PeriodicalId":48471,"journal":{"name":"Computers in Human Behavior","volume":"177 ","pages":"Article 108878"},"PeriodicalIF":8.9,"publicationDate":"2025-12-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145737241","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

Content camouflage: How diversified posting patterns influence human detection of AI-enabled social bots 内容伪装：多样化的发布模式如何影响人类对人工智能社交机器人的检测

IF 8.9 1区心理学 Q1 PSYCHOLOGY, EXPERIMENTAL

Computers in Human Behavior

Pub Date : 2025-12-03 DOI: 10.1016/j.chb.2025.108881

C.J. Saucier , M. Wack , D. Linvill , A. Okoronkwo , G. Tatineni , A. Sezgin

AI-enabled social bots pose a growing threat to information integrity by exploiting users’ inability to distinguish between automated and authentic human accounts—a vulnerability that sophisticated bots can leverage through strategic content camouflage. This vulnerability raises concerns for information integrity, particularly as bots adopt more sophisticated strategies to appear socially authentic. We examine one such tactic, known as content camouflage that involves blending political posts with non-political material, as a psychological mechanism that may influence how users judge account authenticity. Drawing on real bot profiles deployed during the 2024 U.S. election, which relied on sports content to mask accompanying political content, we conducted two experiments (N = 565 undergraduates; N = 601 adults) comparing reactions to profiles posting political-only content versus mixed political-and-sports content. Across both, content diversification did not directly improve bot detection but reliably increased perceived trustworthiness, which in turn made accounts more likely to be classified as human. In the adult sample, this trustworthiness pathway was moderated by political ideology. Together, the findings identify a subtle credibility mechanism through which AI-generated bots exploit users’ reliance on heuristic cues, highlighting the need for interventions that account for new AI-enabled evasion tactics, such as strategic content blending.

通过利用用户无法区分自动账户和真实的人类账户，人工智能社交机器人对信息完整性构成了越来越大的威胁——复杂的机器人可以通过战略内容伪装来利用这一漏洞。这一漏洞引发了人们对信息完整性的担忧，尤其是当机器人采用更复杂的策略来表现出社交真实性时。我们研究了一种被称为内容伪装的策略，它涉及将政治帖子与非政治材料混合在一起，作为一种可能影响用户如何判断账户真实性的心理机制。利用2024年美国大选期间部署的真实机器人档案，我们进行了两个实验（N = 565名本科生；N = 601名成年人），比较了人们对只发布政治内容的档案和混合政治和体育内容的档案的反应。在这两种情况下，内容多样化并没有直接提高机器人的检测能力，但却可靠地增加了人们对可信度的感知，这反过来又使账户更有可能被归类为人类。在成人样本中，这一信任路径受到政治意识形态的调节。总之，研究结果确定了一种微妙的可信度机制，通过这种机制，人工智能生成的机器人利用用户对启发式线索的依赖，强调了对新的人工智能规避策略（如战略内容混合）进行干预的必要性。

{"title":"Content camouflage: How diversified posting patterns influence human detection of AI-enabled social bots","authors":"C.J. Saucier , M. Wack , D. Linvill , A. Okoronkwo , G. Tatineni , A. Sezgin","doi":"10.1016/j.chb.2025.108881","DOIUrl":"10.1016/j.chb.2025.108881","url":null,"abstract":"<div><div>AI-enabled social bots pose a growing threat to information integrity by exploiting users’ inability to distinguish between automated and authentic human accounts—a vulnerability that sophisticated bots can leverage through strategic content camouflage. This vulnerability raises concerns for information integrity, particularly as bots adopt more sophisticated strategies to appear socially authentic. We examine one such tactic, known as <em>content camouflage</em> <em>that</em> involves blending political posts with non-political material, as a psychological mechanism that may influence how users judge account authenticity. Drawing on real bot profiles deployed during the 2024 U.S. election, which relied on sports content to mask accompanying political content, we conducted two experiments (N = 565 undergraduates; N = 601 adults) comparing reactions to profiles posting political-only content versus mixed political-and-sports content. Across both, content diversification did not directly improve bot detection but reliably increased perceived trustworthiness, which in turn made accounts more likely to be classified as human. In the adult sample, this trustworthiness pathway was moderated by political ideology. Together, the findings identify a subtle credibility mechanism through which AI-generated bots exploit users’ reliance on heuristic cues, highlighting the need for interventions that account for new AI-enabled evasion tactics, such as strategic content blending.</div></div>","PeriodicalId":48471,"journal":{"name":"Computers in Human Behavior","volume":"177 ","pages":"Article 108881"},"PeriodicalIF":8.9,"publicationDate":"2025-12-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145685066","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

Dating app use, psychological health, and psychological well-being: A systematic review and quantitative meta-analysis 约会软件的使用、心理健康和心理健康：一项系统回顾和定量荟萃分析

IF 8.9 1区心理学 Q1 PSYCHOLOGY, EXPERIMENTAL

Computers in Human Behavior

Pub Date : 2025-12-01 DOI: 10.1016/j.chb.2025.108879

Liesel L. Sharabi , Lihong Ou , Paige A. Von Feldt , Thomas D. Parsons

This pre-registered study reports the results of a systematic review and quantitative meta-analysis of the clinical (e.g., psychological health and well-being) outcomes of dating app use. We conducted a keyword search of the literature for studies published between 2007 and 2024 using seven databases in communication, psychology, and the biomedical and life sciences. Our sample included 23 studies (N = 26,068) and captured the period of time since dating apps were first brought to market to the present. We found that dating app users reported significantly worse psychological health and well-being (i.e., depression, anxiety, affective dysregulation, loneliness, and psychological distress) than dating app non-users. We also performed a subgroup analysis to explore the significant heterogeneity across studies. The effect of dating app use versus non-use on psychological health and well-being differed by platform type, relationship status, sexual orientation, and cultural context, but these differences were not significant. Overall, our synthesis of 17 years of research coheres with prior trends and underscores the need for targeted interventions to improve the outcomes of dating app users.

这项预先注册的研究报告了使用约会应用程序对临床（如心理健康和幸福）结果的系统回顾和定量荟萃分析的结果。我们使用传播学、心理学、生物医学和生命科学等7个数据库，对2007年至2024年间发表的研究文献进行了关键词搜索。我们的样本包括23项研究（N = 26,068），并捕获了从约会应用程序首次进入市场到现在的一段时间。我们发现，与非约会应用用户相比，约会应用用户的心理健康和幸福感明显更差（即抑郁、焦虑、情感失调、孤独和心理困扰）。我们还进行了亚组分析，以探索研究之间的显著异质性。使用与不使用约会应用对心理健康和幸福的影响因平台类型、关系状态、性取向和文化背景而异，但这些差异并不显著。总的来说，我们对17年研究的综合结果与之前的趋势一致，并强调了有针对性的干预措施的必要性，以改善约会应用程序用户的结果。

{"title":"Dating app use, psychological health, and psychological well-being: A systematic review and quantitative meta-analysis","authors":"Liesel L. Sharabi , Lihong Ou , Paige A. Von Feldt , Thomas D. Parsons","doi":"10.1016/j.chb.2025.108879","DOIUrl":"10.1016/j.chb.2025.108879","url":null,"abstract":"<div><div>This pre-registered study reports the results of a systematic review and quantitative meta-analysis of the clinical (e.g., psychological health and well-being) outcomes of dating app use. We conducted a keyword search of the literature for studies published between 2007 and 2024 using seven databases in communication, psychology, and the biomedical and life sciences. Our sample included 23 studies (<em>N</em> = 26,068) and captured the period of time since dating apps were first brought to market to the present. We found that dating app users reported significantly worse psychological health and well-being (i.e., depression, anxiety, affective dysregulation, loneliness, and psychological distress) than dating app non-users. We also performed a subgroup analysis to explore the significant heterogeneity across studies. The effect of dating app use versus non-use on psychological health and well-being differed by platform type, relationship status, sexual orientation, and cultural context, but these differences were not significant. Overall, our synthesis of 17 years of research coheres with prior trends and underscores the need for targeted interventions to improve the outcomes of dating app users.</div></div>","PeriodicalId":48471,"journal":{"name":"Computers in Human Behavior","volume":"177 ","pages":"Article 108879"},"PeriodicalIF":8.9,"publicationDate":"2025-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145685064","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 Trusted Partner for financial decision making: Romantic partner or AI? 金融决策的可靠伙伴：浪漫伴侣还是人工智能？

IF 8.9 1区心理学 Q1 PSYCHOLOGY, EXPERIMENTAL

Computers in Human Behavior

Pub Date : 2025-11-27 DOI: 10.1016/j.chb.2025.108877

Erik Hermann , Max Alberhasky

Consumers increasingly turn to financial advisory services driven by artificial intelligence (AI)—financial robo advisors—to seek financial advice, yet financial and investment decisions are often made within romantic partnerships. We investigate whether financial advice is more likely to be accepted when it comes from one's romantic partner versus AI as the advice source and explore the underlying processes and AI design features shaping willingness to follow financial advice. Across four experiments, we find robust evidence of AI aversion: consumers prefer financial advice from their romantic partner over AI advice. The relationship between the source of advice and willingness to follow the advice is mediated by cognitive and affective trust, with affective trust being the stronger mediator. Study 2 shows that anthropomorphizing AI reduces AI aversion—advice from an anthropomorphized robo advisor is accepted at rates similar to partner advice. Studies 3 A and 3 B demonstrate that advice from a romantic partner using AI assistance is accepted at similarly high rates as romantic partner advice but significantly more than standalone AI. These findings highlight the critical role of affective trust in high-stakes financial decisions and show that anthropomorphism and human-AI collaboration can mitigate consumer resistance to AI in relational financial contexts.

消费者越来越多地转向由人工智能（AI）驱动的金融咨询服务——金融机器人顾问——寻求金融建议，但金融和投资决策往往是在恋爱关系中做出的。我们调查了来自浪漫伴侣的财务建议是否更容易被接受，而不是人工智能作为建议来源，并探索了潜在的过程和人工智能设计特征，形成了遵循财务建议的意愿。在四项实验中，我们发现了厌恶人工智能的有力证据：与人工智能建议相比，消费者更喜欢伴侣提供的财务建议。建议来源与建议采纳意愿之间存在认知信任和情感信任的中介作用，其中情感信任的中介作用更强。研究2表明，拟人化人工智能减少了对人工智能的厌恶——拟人化机器人顾问的建议被接受的比率与合作伙伴的建议相似。研究3a和3b表明，使用人工智能辅助的浪漫伴侣的建议被接受的比例与浪漫伴侣的建议相似，但明显高于独立的人工智能。这些发现强调了情感信任在高风险金融决策中的关键作用，并表明拟人化和人类-人工智能协作可以减轻消费者在关系金融环境中对人工智能的抵制。

{"title":"The Trusted Partner for financial decision making: Romantic partner or AI?","authors":"Erik Hermann , Max Alberhasky","doi":"10.1016/j.chb.2025.108877","DOIUrl":"10.1016/j.chb.2025.108877","url":null,"abstract":"<div><div>Consumers increasingly turn to financial advisory services driven by artificial intelligence (AI)—financial robo advisors—to seek financial advice, yet financial and investment decisions are often made within romantic partnerships. We investigate whether financial advice is more likely to be accepted when it comes from one's romantic partner versus AI as the advice source and explore the underlying processes and AI design features shaping willingness to follow financial advice. Across four experiments, we find robust evidence of AI aversion: consumers prefer financial advice from their romantic partner over AI advice. The relationship between the source of advice and willingness to follow the advice is mediated by cognitive and affective trust, with affective trust being the stronger mediator. Study 2 shows that anthropomorphizing AI reduces AI aversion—advice from an anthropomorphized robo advisor is accepted at rates similar to partner advice. Studies 3 A and 3 B demonstrate that advice from a romantic partner using AI assistance is accepted at similarly high rates as romantic partner advice but significantly more than standalone AI. These findings highlight the critical role of affective trust in high-stakes financial decisions and show that anthropomorphism and human-AI collaboration can mitigate consumer resistance to AI in relational financial contexts.</div></div>","PeriodicalId":48471,"journal":{"name":"Computers in Human Behavior","volume":"177 ","pages":"Article 108877"},"PeriodicalIF":8.9,"publicationDate":"2025-11-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145685069","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 illusion of reality: How AI-generated images (AIGIs) are fooling social media users 现实的幻觉：人工智能生成的图像（AIGIs）如何愚弄社交媒体用户

IF 8.9 1区心理学 Q1 PSYCHOLOGY, EXPERIMENTAL

Computers in Human Behavior

Pub Date : 2025-11-24 DOI: 10.1016/j.chb.2025.108876

Márk Miskolczi

This study investigates the deceptive impact of AI-generated images [AIGIs] on social media, focusing on content features, user engagement, and underlying psychological mechanisms. By analyzing Facebook pages and user reactions [n = 8922], this research addressing a significantly underexplored research area, identifies key AIGI attributes and examines the most common user reactions. Findings reveal that cognitive biases play a remarkable role in amplifying the deceptive power of AIGIs, fostering users’ trust in fabricated content. The study explains a novel method for detecting AIGIs and AI bots in social media environment, and with the extension of Grounded Theory [GT] coding process, a robust framework for future research is also presented. Highlighting the growing threat of AIGIs in suppressing reality perception, this study calls for rapid regulatory measures, increased digital literacy and systematic monitoring to tackle the spread of the phenomenon.

本研究调查了人工智能生成的图像在社交媒体上的欺骗性影响，重点关注内容特征、用户参与度和潜在的心理机制。通过分析Facebook页面和用户反应[n = 8922]，本研究解决了一个明显未被开发的研究领域，确定了关键的AIGI属性并检查了最常见的用户反应。研究结果表明，认知偏差在放大aigi的欺骗能力、培养用户对捏造内容的信任方面发挥了显著作用。该研究解释了一种在社交媒体环境中检测人工智能和人工智能机器人的新方法，并随着扎根理论[GT]编码过程的扩展，为未来的研究提供了一个强大的框架。该研究强调了aigi在抑制现实感知方面日益增长的威胁，呼吁采取迅速的监管措施，提高数字素养和系统监测，以解决这一现象的蔓延。

引用次数: 0

Does transparency matter when an AI system meets performance expectations? An experiment with an online dating site 当人工智能系统达到性能预期时，透明度是否重要？一个在线约会网站的实验

IF 8.9 1区心理学 Q1 PSYCHOLOGY, EXPERIMENTAL

Computers in Human Behavior

Pub Date : 2025-11-23 DOI: 10.1016/j.chb.2025.108875

Yuan Sun , Mengqi (Maggie) Liao , S. Shyam Sundar , Joseph B. Walther

Powered by sophisticated algorithms, artificial intelligence (AI) systems are notorious for being opaque, prompting demands for more algorithmic transparency. However, empirical evidence reveals inconsistent effects of transparency on user trust, suggesting critical boundary conditions. Drawing on the Expectation Confirmation Model and Heuristic-Systematic Model, we propose that the effectiveness of providing algorithmic transparency depends on whether the AI system’s performance meets, exceeds, or falls short of users’ expectations. We tested this through a 3 (Algorithmic transparency: high vs. low vs. absent) × 3 (Expectation confirmation: positive vs. simple vs. negative) between-subjects online experiment (N = 227), in which participants interacted with an online dating site driven by date-matching algorithms. Results revealed that detailed algorithmic explanations enhanced perceived understanding and trust when system performance either exceeded or fell short of expectations, while concise explanations proved more effective when performance met expectations. These findings advance our theoretical understanding of algorithmic transparency and provide practical implications for designing adaptive explainable AI systems that adjust transparency levels based on performance outcomes.

人工智能（AI）系统由复杂的算法驱动，因不透明而臭名昭著，这促使人们要求提高算法的透明度。然而，经验证据显示透明度对用户信任的影响不一致，这表明了关键的边界条件。根据期望确认模型和启发式系统模型，我们提出提供算法透明度的有效性取决于人工智能系统的性能是否满足、超过或低于用户的期望。我们通过3（算法透明度：高、低、缺席）× 3（期望确认：积极、简单、消极）受试者之间的在线实验（N = 227）来验证这一点，在该实验中，参与者与一个由日期匹配算法驱动的在线约会网站进行互动。结果显示，当系统性能超过或低于预期时，详细的算法解释增强了感知的理解和信任，而当性能达到预期时，简明的解释证明更有效。这些发现促进了我们对算法透明度的理论理解，并为设计基于性能结果调整透明度水平的自适应可解释人工智能系统提供了实际意义。

{"title":"Does transparency matter when an AI system meets performance expectations? An experiment with an online dating site","authors":"Yuan Sun , Mengqi (Maggie) Liao , S. Shyam Sundar , Joseph B. Walther","doi":"10.1016/j.chb.2025.108875","DOIUrl":"10.1016/j.chb.2025.108875","url":null,"abstract":"<div><div>Powered by sophisticated algorithms, artificial intelligence (AI) systems are notorious for being opaque, prompting demands for more algorithmic transparency. However, empirical evidence reveals inconsistent effects of transparency on user trust, suggesting critical boundary conditions. Drawing on the Expectation Confirmation Model and Heuristic-Systematic Model, we propose that the effectiveness of providing algorithmic transparency depends on whether the AI system’s performance meets, exceeds, or falls short of users’ expectations. We tested this through a 3 (Algorithmic transparency: high vs. low vs. absent) × 3 (Expectation confirmation: positive vs. simple vs. negative) between-subjects online experiment (<em>N</em> = 227), in which participants interacted with an online dating site driven by date-matching algorithms. Results revealed that detailed algorithmic explanations enhanced perceived understanding and trust when system performance either exceeded or fell short of expectations, while concise explanations proved more effective when performance met expectations. These findings advance our theoretical understanding of algorithmic transparency and provide practical implications for designing adaptive explainable AI systems that adjust transparency levels based on performance outcomes.</div></div>","PeriodicalId":48471,"journal":{"name":"Computers in Human Behavior","volume":"177 ","pages":"Article 108875"},"PeriodicalIF":8.9,"publicationDate":"2025-11-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145610124","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

Gamified feedback in adaptive retrieval practice: Points and progress-bars enhance motivation but not learning 适应性检索练习中的游戏化反馈：积分和进度条增强了动机，但不能增强学习

IF 8.9 1区心理学 Q1 PSYCHOLOGY, EXPERIMENTAL

Computers in Human Behavior

Pub Date : 2025-11-22 DOI: 10.1016/j.chb.2025.108862

Gesa S. E. van den Broek , Stefanie Scholten , Bente van Thuil , Hedderik van Rijn , Tamara van Gog , Maarten van der Velde

Retrieval practice is an effective but effortful learning strategy during which learners bring information from memory to mind (e.g., when self-testing with flashcard software). Because it is effortful, learners often find it hard to maintain during self-study. We therefore investigated if gamified feedback could make retrieval practice more attractive to learners, while maintaining or increasing its benefits for learning. In an online experiment, 166 adult participants studied foreign vocabulary (i.e., practiced retrieval of the word meaning) with an adaptive fact learning system. Each participant practiced once in a control condition, and once in a gamified condition: either a points condition which showed a score that increased for every correct answer, or a points-and-progress condition, that additionally displayed on two progress bars how the score increased. Compared to the control condition, the gamified conditions enhanced self-reported feelings of competence, enjoyment, and task value as well as experienced goal setting and the desire to perform well. Differences between points and points-and-progress condition were only found for goal-setting (which was higher for points-and-progress). Gamified feedback did not affect learning behaviors during practice, nor recall on a delayed posttest. However, the response speed, score achieved, number of practiced words, and number of trials completed with gamified feedback were lower when participants first practiced with gamified feedback and then without, compared to when they first completed the control condition and then received gamified feedback. Taken together, results suggest that gamified feedback may be a useful addition to enhance motivation for learning with retrieval practice software.

检索练习是一种有效但费力的学习策略，在此过程中，学习者将记忆中的信息带入脑海（例如，当使用抽认卡软件进行自我测试时）。因为它是费力的，学习者在自学过程中经常发现很难保持。因此，我们调查了游戏化反馈是否可以使检索练习对学习者更具吸引力，同时保持或增加其对学习的益处。在一项在线实验中，166名成年参与者使用自适应事实学习系统学习外语词汇（即练习单词含义检索）。每个参与者在控制条件下练习一次，在游戏化条件下练习一次：在分数条件下，每个正确答案的分数都会增加，或者在分数和进度条件下，在两个进度条上额外显示分数的增加情况。与控制条件相比，游戏化条件增强了自我报告的能力、享受和任务价值感，以及经验目标设定和表现良好的愿望。分数和分数加进步条件之间的差异只存在于目标设定上（分数加进步条件的差异更大）。游戏化反馈不会影响练习中的学习行为，也不会影响延迟后测的回忆。然而，与先完成控制条件后接受游戏化反馈的参与者相比，先接受游戏化反馈后再接受游戏化反馈的参与者的反应速度、得分、练习单词的数量和完成游戏化反馈的试验次数都要低。综上所述，结果表明，游戏化反馈可能是一个有用的补充，以提高学习动机与检索练习软件。

{"title":"Gamified feedback in adaptive retrieval practice: Points and progress-bars enhance motivation but not learning","authors":"Gesa S. E. van den Broek , Stefanie Scholten , Bente van Thuil , Hedderik van Rijn , Tamara van Gog , Maarten van der Velde","doi":"10.1016/j.chb.2025.108862","DOIUrl":"10.1016/j.chb.2025.108862","url":null,"abstract":"<div><div>Retrieval practice is an effective but effortful learning strategy during which learners bring information from memory to mind (e.g., when self-testing with flashcard software). Because it is effortful, learners often find it hard to maintain during self-study. We therefore investigated if gamified feedback could make retrieval practice more attractive to learners, while maintaining or increasing its benefits for learning. In an online experiment, 166 adult participants studied foreign vocabulary (i.e., practiced retrieval of the word meaning) with an adaptive fact learning system. Each participant practiced once in a control condition, and once in a gamified condition: either a <em>points</em> condition which showed a score that increased for every correct answer, or a <em>points-and-progress</em> condition, that additionally displayed on two progress bars how the score increased. Compared to the control condition, the gamified conditions enhanced self-reported feelings of competence, enjoyment, and task value as well as experienced goal setting and the desire to perform well. Differences between points and points-and-progress condition were only found for goal-setting (which was higher for points-and-progress). Gamified feedback did not affect learning behaviors during practice, nor recall on a delayed posttest. However, the response speed, score achieved, number of practiced words, and number of trials completed with gamified feedback were lower when participants first practiced with gamified feedback and then without, compared to when they first completed the control condition and then received gamified feedback. Taken together, results suggest that gamified feedback may be a useful addition to enhance motivation for learning with retrieval practice software.</div></div>","PeriodicalId":48471,"journal":{"name":"Computers in Human Behavior","volume":"177 ","pages":"Article 108862"},"PeriodicalIF":8.9,"publicationDate":"2025-11-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145685063","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

Know thyself through data: Improving WhatsApp interaction awareness with data-driven visualizations 通过数据了解自己：通过数据驱动的可视化提高WhatsApp的交互意识

IF 8.9 1区心理学 Q1 PSYCHOLOGY, EXPERIMENTAL

Computers in Human Behavior

Pub Date : 2025-11-21 DOI: 10.1016/j.chb.2025.108867

Olya Hakobyan, Hanna Drimalla

Digital behaviors are often quantified through screen time or frequency metrics, providing insights into activity patterns and their alignment with subjective perceptions. However, these metrics alone fall short of capturing texting, a behavior involving complex interpersonal dynamics. Research on messaging self-perception is scarce — a critical gap since biased perceptions can strain relationships and affect well-being. In one of the first empirical studies on this topic, we conducted an online study with 68 participants, who donated meta-data (message lengths and times) from over 300 WhatsApp chats. Participants answered questions about their messaging behavior, such as whether they write more or respond faster than their contacts. They then received personalized visualizations based on donated data and reassessed their perceptions, allowing us to examine whether data-driven insights improved self-report accuracy. The results show a significant improvement in the alignment between self-reports and objective metrics for absolute response speed and participants’ contribution within a chat. In contrast, assessments of peak activity times and response speed relative to contacts were already fairly accurate and did not change. Finally, although self-report scores changed for two subjective estimates – the even distribution of messages across chats and texting throughout the day – their alignment with objective metrics did not improve significantly. To account for the potential emotional impact of challenging participants’ existing self-perceptions, we assessed their mood and did not find significant changes in positive or negative affect. To conclude, personalized visualizations can enhance the alignment between self-perceptions and objective behavior, potentially guiding individuals toward more intentional texting behavior.

数字行为通常通过屏幕时间或频率指标来量化，从而深入了解活动模式及其与主观感知的一致性。然而，这些指标本身不足以捕捉到短信，这是一种涉及复杂人际动态的行为。关于信息自我认知的研究很少——这是一个关键的缺口，因为偏见的认知会使人际关系紧张，影响幸福感。在这一主题的首批实证研究之一中，我们对68名参与者进行了一项在线研究，他们捐赠了来自300多个WhatsApp聊天的元数据（消息长度和时间）。参与者回答了关于他们的信息传递行为的问题，比如他们是否比他们的联系人写得更多或回复得更快。然后，他们收到了基于捐赠数据的个性化可视化，并重新评估了他们的看法，使我们能够检查数据驱动的见解是否提高了自我报告的准确性。结果显示，在自我报告和绝对响应速度和参与者在聊天中的贡献的客观指标之间的一致性方面有了显著的改善。相比之下，对高峰活动时间和相对于接触的反应速度的评估已经相当准确，并且没有改变。最后，尽管自我报告的得分在两项主观评估中发生了变化——全天聊天和发短信的信息均匀分布——但它们与客观指标的一致性并没有显著提高。为了解释挑战参与者现有自我认知的潜在情绪影响，我们评估了他们的情绪，并没有发现积极或消极影响的显著变化。总之，个性化的可视化可以增强自我认知和客观行为之间的一致性，潜在地引导个人更有意识地发短信。

{"title":"Know thyself through data: Improving WhatsApp interaction awareness with data-driven visualizations","authors":"Olya Hakobyan, Hanna Drimalla","doi":"10.1016/j.chb.2025.108867","DOIUrl":"10.1016/j.chb.2025.108867","url":null,"abstract":"<div><div>Digital behaviors are often quantified through screen time or frequency metrics, providing insights into activity patterns and their alignment with subjective perceptions. However, these metrics alone fall short of capturing texting, a behavior involving complex interpersonal dynamics. Research on messaging self-perception is scarce — a critical gap since biased perceptions can strain relationships and affect well-being. In one of the first empirical studies on this topic, we conducted an online study with 68 participants, who donated meta-data (message lengths and times) from over 300 WhatsApp chats. Participants answered questions about their messaging behavior, such as whether they write more or respond faster than their contacts. They then received personalized visualizations based on donated data and reassessed their perceptions, allowing us to examine whether data-driven insights improved self-report accuracy. The results show a significant improvement in the alignment between self-reports and objective metrics for absolute response speed and participants’ contribution within a chat. In contrast, assessments of peak activity times and response speed relative to contacts were already fairly accurate and did not change. Finally, although self-report scores changed for two subjective estimates – the even distribution of messages across chats and texting throughout the day – their alignment with objective metrics did not improve significantly. To account for the potential emotional impact of challenging participants’ existing self-perceptions, we assessed their mood and did not find significant changes in positive or negative affect. To conclude, personalized visualizations can enhance the alignment between self-perceptions and objective behavior, potentially guiding individuals toward more intentional texting behavior.</div></div>","PeriodicalId":48471,"journal":{"name":"Computers in Human Behavior","volume":"177 ","pages":"Article 108867"},"PeriodicalIF":8.9,"publicationDate":"2025-11-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145737244","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