Civically active people online are usually more prone to careless information sharing. Thus, although usually framed positively, the youth civic engagement can also result in the sharing of potentially harmful content. However, the evidence on underage social media users is limited. This paper focuses on two factors that can potentially influence online behavior: one through well-being, and the second through information navigation and processing digital literacy. A three-wave random intercept cross-lagged panel model was employed, and the analyses were conducted on adolescents (Mage = 14.3; N = 2660) from Estonia, Germany, Italy, Poland, and Portugal. The data was collected between Spring 2021 and Spring 2023. Between-person associations between online civic engagement and well-being, and between online civic engagement and information navigation and processing literacy, were supported. This means more engaged adolescents are also more dissatisfied and have higher digital literacy, corroborating the existing literature. There were only limited within-person associations between careless sharing and information navigation and processing literacy; however, none of them supported the proposed associations.
Incident reporting behaviour is a critical yet underexplored human factor in organisational cyber resilience, shaping how organisations learn from incidents and adapt to evolving digital threats. This systematic literature review synthesises evidence from 106 peer-reviewed studies across cyber resilience, healthcare, and aviation to examine how cultural, organisational, technological, and psychological factors influence employees’ reporting behaviour. The review identifies five thematic clusters of antecedents—organisational, psychological, technical, social, and motivational—and shows how their interaction can either facilitate or hinder incident reporting behaviour. Building on these findings, we propose a novel, integrative multilevel conceptual model that extends the Cybersecurity Culture–Behaviour (CSCB) framework of Sutton and Tompson (2024) by incorporating a widely used behaviour-change framework, perceived consequences of reporting, and the moderating roles of incident characteristics and reporter attributes. The results highlight the importance of multilevel, human-centred interventions that foster psychological safety, provide timely and meaningful feedback, and align managerial practices with principles of a just culture. This study advances understanding of incident reporting behaviour as a socio-technical behaviour and offers a unifying framework for designing interventions to strengthen cyber resilience through improved human reporting practices.
Conversational AI bots are increasingly integrated into writing instruction. However, little is known about how these tools may function as multimodal learning partners in writing activities, especially among younger learners. This study examined changes in primary school students' critical thinking during and after a conversational AI bot–supported multimodal writing practice and explored students' experiences to contextualize the observed changes. A mixed-methods design was employed with 60 Grade 5 students in a Chinese primary school. Over an eight-week intervention, students transformed their narratives into multimodal representations using a conversational AI bot and created short videos combining images and narration. Quantitative data were collected at three time points (pre-intervention, post-intervention, and one-month follow-up). Semi-structured interviews with nine students explored their perceptions and experiences. Quantitative results showed significant increases in interpretation, analysis, evaluation, and explanation from pre-intervention to post-intervention and follow-up. Inference showed no significant increase, while self-regulation showed a short-term increase that was not sustained at follow-up. Qualitative findings suggest that conversational AI bot-supported multimodal composing helped students to externalize ideas through visual representations, which they associated with processes such as idea verification, gap detection, and revision. However, explicit multimodal representations may have reduced the need for inferential reasoning, while peer collaboration appeared to create complementary occasions for students to infer others' interpretations and reconsider their own texts. These findings suggest the potential of conversational AI bots as multimodal learning partners, while highlighting the need for continued instructional scaffolding to help sustain observed increases in students’ critical thinking.
The rapid integration of artificial intelligence (AI) into workplace processes has transformed organizational communication, yet little is known about how AI reshapes employee help-seeking behaviors. Leveraging on the social exchange perspective, this research examines how help source (human vs. AI colleague) and help context type (autonomy-oriented vs. dependency-oriented) influence employees' mechanisms to seek help in working contexts. Across two experiments, Study 1 investigates the psychological mechanisms underlying help-seeking from AI versus human, focusing on perceived autonomy and perceived indebtedness as mediators. Study 2 extends this framework by testing the moderating role of help context type on the proposed direct and indirect effects. Results indicated that employees perceived higher autonomy and lower indebtedness when receiving dependency-oriented help from AI compared to human colleagues, which in turn increased their willingness to seek help. The findings advance understandings of AI-mediated workplace interactions, offering theoretical and practical insights into the “help-seeker's dilemma”.
The rising popularity of fitness apps underscores the need to understand how their use translates into workout intention. However, a paucity of research has examined this issue from a user-centric perspective. Grounded in Uses and Gratifications (U&G) theory and the adapted Cognitive Mediation Model (CMM), this study develops and tests a process-oriented framework to explicate how gratifications obtained from fitness app use are related to workout intention through cognitive and affective mechanisms. Structural equation modeling based on survey data (N = 795) from Chinese fitness app users shows that all five examined gratifications are positively associated with cognitive elaboration, whereas four (excluding social interaction gratification) are positively related to the emotion of inspiration. Cognitive elaboration, in turn, is positively tied to both objective and subjective fitness knowledge. Inspiration is positively associated only with subjective knowledge. Importantly, although both forms of knowledge are positively linked to workout intention, subjective knowledge exhibits a significantly stronger association. By integrating motivational, cognitive, and affective perspectives, this study extends the CMM through the incorporation of context-specific gratifications and emotional components. The findings illuminate the distinct roles of objective and subjective knowledge and underscore the dynamic interplay between emotion and cognition in shaping health-related behavioral intentions. These results contribute to a more nuanced, user-centered understanding of human-fitness technology interaction and offer implications for the design and optimization of digital health interventions.
In online communities, user participation behaviors (e.g., comments and likes) are the source of sustainable development for these communities. Community platforms are actively exploring measures to enhance user participation. Benefiting from generative AI technologies, online communities have begun to deploy intelligent chatbots. However, the content generated by these bots often suffers from the problem of AI hallucination. Prior studies indicate that AI hallucination exerts negative impacts in fields such as healthcare and tourism, yet there has been limited focus on its influence on user participation behaviors in online communities. This study employs a total of 161,130 chatbot records from Sina Weibo over one year. Based on the theory of prosocial behavior, we examine the impact of AI hallucination on user participation behaviors in online communities, as well as the moderating effects of day type and active period. The study reveals that AI hallucination exhibits a “brick-to-jade effect”, increasing comments from users while simultaneously reducing their likes. Additionally, the study also shows that the positive correlation between AI hallucination and users' comments is stronger during weekends and active period. Theoretically, this study examines the effect of AI hallucination from the perspective of users, expanding the scope of research on the effect of AI hallucination and proving that AI hallucination has a positive role on the comments of online community users. This breaks the common perception in existing research that AI hallucination is always negative. Practically, it has significant reference value for community managers to stimulate user participation behaviors and enhance their activity levels.
This study investigates how language mutations affect the persistent diffusion of COVID-19 conspiracy theories on social media. Drawing on a three-year dataset of conspiracy-related posts from X, and applying computational linguistic analysis alongside survival modelling, we find that conspiracy claims with greater semantic mutations have substantially longer lifespans. Mutations in psycholinguistic properties, including pronouns, social reference words, cognitive process terms, risk- and health-related vocabularies, are associated with extended lifespans. Mutations in actor, action and target (AAT) categories are associated with longer lifespans as well. Qualitative analysis identifies two predominant mutation patterns: simplification and assimilation, at both linguistic and AAT structural levels. Taken together, the results advance our understanding of the association between language mutations and conspiracy persistence online and shed lights on longitudinal content moderation strategies. We argue that content moderation should consider the mutability of conspiracy claims and focus on the core claims that can address their potential variations.
This research aimed to develop an integrated theoretical model to explain the factors influencing verification behavior regarding social media disinformation among young adults in Indonesia. The model combined the Stimulus–Organism–Response (SOR) framework with the Norm Activation Model (NAM) and the Social Identity Theory (SIT) to examine the collective effects of social, cognitive, and moral processes in shaping responsible information behavior. An online cross-sectional survey was conducted with 746 respondents, who actively used social networking sites to obtain and share information. The results showed that verification behavior was primarily driven by information skepticism and personal norms, emphasizing the importance of critical thinking and moral obligation for responsible engagement. In the Organism stage, awareness of fake information, perceived deception, and critical consumption enhanced moral sensitivity and analytical reasoning. At the social level, factors such as collective memory, parasocial interaction, and status-seeking were reported to be significant identity-based stimuli in shaping cognitive and moral responses, with gender found to moderate these effects. Thematic analysis suggested that most young adults verified information through cross-checking and peer consultation, but were influenced by social validation. Theoretically, this research contributed to disinformation research by framing verification as a cognitive-normative process rather than a reactive behavior. Different initiatives were recommended by educational institutions, governmental bodies, and community organizations to strengthen moral reasoning, digital literacy, and civic responsibility in combating disinformation.
Smart technologies are increasingly introduced in workplaces, yet employees do not always translate positive evaluations into technology adoption. This study examines how different forms of employee knowledge shape persuasion toward smart technologies and how employee engagement conditions the relationship between persuasion and technology adoption, an area that has received little attention. Drawing on the Diffusion of Innovation (DOI) framework, employee knowledge is conceptualised as a broader domain comprising four theoretically related but empirically distinct first-order dimensions: implicit, explicit, perceptual, and generational knowledge. Survey data from 279 employees in the Sri Lankan e-waste sector were analysed using covariance-based structural equation modelling (CB-SEM) with latent interaction modelling. The results show that implicit, explicit, and perceptual knowledge strengthen employee persuasion, whereas generational knowledge weakens it. However, persuasion alone does not directly lead to the adoption of smart technology in the e-waste sector businesses. Instead, employee engagement positively moderates the persuasion–smart technology adoption relationship, suggesting that engagement helps explain when favourable evaluations are more likely to translate into adoption behaviour. Therefore, mediation analysis indicates a more complex pattern: persuasion showed a negative direct association with smart technology adoption, and the indirect effects suggest a possible context-specific discontinuity between persuasion and reported adoption behaviour rather than a straightforward knowledge-persuasion-smart technology adoption pathway. These findings highlight that employee responses to smart technologies depend not only on knowledge but also on behavioural engagement and implementation conditions, providing new insight into human–technology interaction in workplace digitalisation.
As generative artificial intelligence (AI) becomes increasingly involved in digital content creation, disclosure messages play an important role in shaping how users interpret human–AI collaboration. Drawing on cue utilization theory, this study examines how two dimensions of AI disclosure—authorship dominance (human-centered vs. AI-centered) and AI role framing (logical, emotional, or integrity enhancement)—influence the perceived authenticity of online reviews across search and experience goods. A 2 × 3 × 2 mixed-factorial experiment was conducted with 272 Korean online shoppers. Linear mixed-effects analyses showed that human-centered disclosure produced higher perceived authenticity than AI-centered disclosure overall. However, AI role framing did not have a uniform main effect; instead, its influence depended on authorship dominance and product type. For search goods, human-centered disclosure enhanced perceived authenticity when AI was framed as supporting integrity, whereas for experience goods, this advantage emerged when AI was framed as supporting logical organization. No significant authorship-dominance difference was found under emotional-expression enhancement. Product-specific Bayesian mediation analyses further showed that disclosure configurations influenced purchase intention through perceived authenticity and platform trust, although the relative serial indirect effects differed across product types. For search goods, all comparison conditions produced lower serial indirect effects than the human-centered integrity-enhancement condition, whereas for experience goods, a significant difference emerged only for the AI-centered logical-enhancement condition. These findings demonstrate that AI disclosure functions as a set of interdependent interpretive cues through which users infer human agency, the nature of AI involvement, and the authenticity of AI-assisted reviews.
Digital mindfulness (DM) refers to the capacity to engage with digital technologies in a deliberate, attentive, and non-reactive manner. As digital work intensifies, DM has shown early promise as a protective factor for employee well-being. However, a domain-specific mindfulness measure grounded in the secular mindfulness tradition (Kabat-Zinn, 1994) and designed to capture a resource that is protective of well-being in digitally mediated work contexts is currently lacking. This study reports the development and validation of a 15-item Digital Mindfulness Scale (DMS) grounded in Shapiro et al.’s (2006) Intention–Attention–Attitude (IAA) framework. Scale development involved expert panel review (N = 12; 10 retained for analysis) of a theoretically developed item pool (N = 30) followed by exploratory factor analysis (EFA; N = 398) and confirmatory factor analysis (CFA; N = 336). Results supported a stable three-factor structure reflecting intention, attention, and attitude. The DMS demonstrated excellent internal consistency (α = .90–.91), good test–retest reliability (ICC = .77), and evidence of convergent and criterion validity. Using longitudinal data, DM predicted lower technostress and exhaustion and higher well-being over a month, over and above demographic variables, technology use, and extant mindfulness measures. These findings support the DMS as a psychometrically robust, context-specific instrument for assessing mindful engagement with digital work and contribute to theory and practice by distinguishing dispositional mindfulness from its behavioural expression in digitally mediated environments.
People use multiple social network sites (SNSs) to interact and present different facets of themselves. However, existing studies often focus on a single platform, lacking a holistic view of users' behavior across multiple SNSs. Drawing on signaling theory, an affordance-based social media usage framework is proposed, incorporating three affordances: network structure, anonymity, and visual presentation. The framework contrasts honest signals with conventional signals, theorized as ego-centric versus interest-based networks, low versus high anonymity, and self-included versus non-self-included visual presentations. These six affordances-based behaviors were operationalized using a national representative sample of 1549 participants from the Taiwan Communication Survey. The latent class analysis (LCA) identified six latent user styles: ego-centric networkers, anonymous hobbyists, public social butterflies, cross-platform masqueraders, relationship maintainers, and lurkers. The strategic use of both honest and conventional signals across platforms leads to the greatest social benefits, such as close-tie networks, social capital, and well-being, particularly among cross-platform masqueraders.
Digital misogyny is increasingly pervasive, making it vital to understand how users perceive and respond to it. Drawing on distinctions between incivility, intolerance, and threats and on the elaboration likelihood model as a framework for understanding when peripheral cues such as perpetrator attractiveness shape moral judgment, this preregistered online experiment (N = 884) examined how hate type (incivility, intolerance, threat) and perpetrator attractiveness (high, medium, low) shape users' cognitive, affective, attitudinal, and behavioral responses to misogynist hate. Results show that users’ responses broadly followed a normative continuum from incivility to intolerance to threat, with both incivility and threat reliably distinguished across all response dimensions, while intolerance proved more nuanced and outcome-dependent. Threats and intolerance were judged as less acceptable and more sexist than incivility, though cognitive evaluations plateaued between the two more severe forms, suggesting a ceiling in moral judgment once digital hate crosses a clear normative threshold. Negative emotions continuously intensified with hate type. Threats elicited stronger indirect intervention intentions and greater support for content moderation than incivility and intolerance. Readiness to intervene directly remained limited, pointing to a fundamental asymmetry between moral condemnation and visible action. Perpetrator attractiveness did not influence user responses across any outcome or hate type, marking an important theoretical boundary condition for appearance-based biases in moral judgment, where in morally unambiguous contexts, such as misogynist digital hate, content-based reasoning appears to override heuristic shortcuts such as the pretty privilege bias.
This study examined whether counterarguments generated by large language models (LLMs) influence the moral judgments of younger and older adults, and whether these effects vary by dilemma type, cognitive functioning, trust in AI, and prior LLM experience. Using the switch and footbridge trolley dilemmas, 130 participants (56 younger adults and 74 older adults) were presented with ChatGPT-generated counterarguments that opposed their initial judgments. More than 30% of participants reversed their judgments in both dilemmas (32.31% in the switch dilemma and 36.92% in the footbridge dilemma). Older adults tended to be more likely than younger adults to reverse their judgments and showed a significantly greater degree of judgment change in the switch dilemma. In the emotionally aversive footbridge dilemma, older adults with lower cognitive functioning were significantly more likely to align with the LLM-generated counterargument. General trust in AI and prior LLM experience did not predict judgment reversal, whereas lower initial confidence and higher perceived task difficulty were associated with greater susceptibility to LLM influence. These findings suggest that LLMs may support cognitive offloading but increase susceptibility among individuals with limited cognitive resources. The ecological generalizability of these findings to everyday dilemma situations remains to be examined in future research.
Creativity is fundamentally a collaborative process. Yet as generative AI becomes increasingly integrated into creative work, understanding how AI reshapes collaboration has become critical. This pre-registered study directly compares human-human and human-AI collaboration dynamics across two creative tasks: the Alternative Uses Task (AUT) and creative short story writing. Participants were randomly assigned to pairs in either human-human (N = 68 pairs) or human-AI (GPT-4o; N = 72 pairs) conditions, with partners alternating turns as first responders to examine how initiation order shapes the creative process over time. Our findings reveal that the apparent "AI advantage" in creative collaboration is illusory, driven primarily by increased AI verbosity rather than enhanced creativity. Critically, collaboration with AI partners negatively impacted humans' own creative responses compared to human-human partnerships, with human-AI collaboration failing to enhance idea originality or diversity relative to human-human collaboration. Human partners demonstrated superior collaborative effectiveness that strengthened over time, indicating that current generative AI systems, while producing more verbose outputs, do not replicate the collective creativity characteristic of human-human collaboration. These results challenge assumptions about AI’s creative potential, with direct implications for AI system design and collaborative creative practice.
Concerns over the potential over-pathologization of generative AI (GenAI) use and the lack of conceptual clarity surrounding GenAI addiction call for empirical tools and theoretical refinement. This study developed and validated the PUGenAIS-9 (Problematic Use of Generative Artificial Intelligence Scale-9 items) and examined whether PUGenAIS reflects addiction-like patterns under the Internet Gaming Disorder (IGD) framework. Using samples from China and the United States (N = 1508), we conducted confirmatory factor analysis and identified a robust 31-item structure across nine IGD-based dimensions. We then derived the PUGenAIS-9 by selecting the highest-loading items from each dimension and validated its structure in an independent sample (N = 1426). Measurement invariance tests confirmed its stability across nationality and gender. Person-centered (latent profile analysis) and variable-centered (network analysis) approaches revealed a 5–10% prevalence rate, a symptom network structure similar to IGD, and predictive factors related to psychological distress and functional impairment. These findings indicate that PUGenAI shares features of the emotionally vulnerable subtype of IGD rather than the competence-based type. These results support using PUGenAIS-9 to identify problematic GenAI use and show the need to rethink digital addiction with an IDC (Infrastructure-Device-Content) model. This keeps addiction research responsive to new media while avoiding over-pathologizing.
