To overcome the inability of existing approaches to detect weak, multimodal and emerging early signs of virality, this study proposes a novel recognition framework. By integrating morphological and signal-to-noise separation (SNS) features, the framework accurately identifies authentic viral signals and predicts their growth trajectories. The proposed framework constructs content signal morphology by combining multimodal sentiment analysis based on BERT and ViT, CRITIC-based influence weighting and HDBSCAN clustering. Three SNS features, frequency orderliness, temporal dependency and semantic novelty, are extracted to characterize the separation between emerging viral signals and background noise. An LSTM model is then employed for dynamic classification of early-stage virality signals. Experiments conducted on a large-scale Sina Weibo dataset demonstrate that the integrated framework significantly outperforms traditional approaches and single-feature baselines in terms of precision, recall and overall accuracy. The results confirm the complementary and synergistic effects of morphological and SNS features in distinguishing authentic early-stage virality signals from noisy content streams. This study advances early content virality recognition by integrating signal morphology and SNS features within a unified analytical framework. It shifts the analytical focus from reactive viral event detection to proactive discovery of latent virality signals embedded in noisy, multimodal social media streams. The proposed framework offers a theoretically grounded and empirically validated solution for real-time content virality detection and early warning systems on social media platforms.
Purpose With the advances in AI-driven audiovisual synthesis, multimodal disinformation has become prevalent in social media, where user engagement may complicate its evaluation. It remains unclear why individuals are susceptible to such content, especially in contexts featuring social engagement cues. Therefore, this study aims to investigate the roles of multimodality, online comments and virality metrics in disinformation evaluation. Design/methodology/approach Guided by the heuristic–systematic model, multimodal framing theory, and social proof theory, we carried out a mixed-methods study with two phases. In Phase I, a between-subjects experiment (N = 433) examined how various cues independently and interactively influenced credibility judgments of disinformation. In Phase II, follow-up interviews (N = 24) uncovered the causes driving the observed effects. Findings Text-with-video disinformation was perceived more credibly than text-with-image and text-only types. Participants generally considered videos more difficult to forge. This view served as a heuristic that triggered less critical evaluation. Additionally, online comments could correct or mislead credibility judgments, whereas virality metrics showed no such effect. Interestingly, the impact of online comments could be attenuated by multimodality, making participants more susceptible to disinformation. Because audiovisual cues offer sufficient intuitive evidence, participants relied less on social proof for judgment-making. Originality/value Synthesizing the abovementioned theories, this study advances the understanding of individual susceptibility to multimodal disinformation. This susceptibility stems not only from multimodality per se, but also from its tendency to overshadow the efficacy of social proof in correcting falsehoods. From the information processing perspective, this phenomenon is driven by intuitive sufficiency of audiovisual realism, which satisfies judgmental confidence and renders social proof cues less significant.
Purpose Online academic medical communities (OAMCs) have become central channels for physicians to exchange medical knowledge. Yet a substantial share of physician-generated medical content on these platforms fails to elicit meaningful peer engagement, constraining the dissemination of medical knowledge. Given that the intrinsic features of content in OAMCs are likely key drivers of peer engagement, understanding what makes content appealing is essential. However, the relationship remains unclear. This study aims to address this gap by examining how distinct content features influence different forms of peer engagement in OAMCs. Design/methodology/approach Leveraging data from a leading Chinese OAMC, we operationalize six content features (complexity, subjectivity, professionalism, completeness, informativeness, multimodal information presentation) via natural language processing and examine their effects on physicians' collecting and interacting behaviors and the moderating role of disease severity. Findings Peer physicians prioritize information-rich content for collecting, while favoring discussion-oriented content for interacting. Multimodal, comprehensive content is welcomed, whereas overly sophisticated, highly specialized content is less effective. Disease severity moderates these relationships, with physicians exhibiting greater selectivity in collecting but more interactive under high-severity conditions. Originality/value This study is among the first to uncover the impact of content features on physicians' engagement behaviors while considering disease severity as the moderating variable. By incorporating natural language processing techniques, it extends the information adoption model to online physician platforms and offers actionable guidance for optimizing physician online engagement.
Purpose-In human-generative AI collaboration (HGAIC), employees can adopt different role orientations to collaborate with GenAI, but it remains unclear how employee role orientations affect their innovation and how they are shaped. Based on social information processing theory (SIPT), this study examines the effects of passive and co-production role orientations on GenAI-enabled employee innovation, the multiple antecedents of these orientations, and the moderating role of employee opportunism. Design/methodology/approach-This study used partial least squares structural equation modeling (PLS-SEM) and fuzzy-set qualitative comparative analysis (fsQCA) to analyze survey data from 409 employees. Findings-The results indicate that passive and co-production role orientations positively affect GenAI-enabled employee innovation. GenAI intellectual stimulation and task urgency positively influence passive role orientation, whereas task innovativeness negatively influences it. GenAI feedback richness, GenAI intellectual stimulation, and task innovativeness positively affect co-production role orientation. Opportunism only amplifies the positive effect of passive role orientation on GenAI-enabled employee innovation. Additionally, the fsQCA results reveal three configurations that lead to high levels of GenAI-enabled employee innovation. Originality/value-First, this study reveals the relationship between different role orientations in HGAIC and GenAI-enabled employee innovation. Second, this study identifies the antecedents of passive and co-production role orientations from the perspectives of GenAI feedback and task requirement cues. Finally, this study reveals a counterintuitive finding: employee opportunism can act as a positive catalyst for innovation under specific HGAIC conditions.
Purpose-Physicians' participation in online medical teams (OMTs) often extends beyond workload sharing or financial gain, encompassing opportunities to expand clinical knowledge, and refine professional skills. Yet it remains unclear whether such knowledge-related benefits are realized and how they might translate into tangible personal outcomes. From a knowledge-based perspective, this study aims to examine whether team knowledge attributes influence physicians' financial and reputational returns, and whether physicians' self-efficacy mediates these effects. Design/methodology/approach-Unbalanced panel data were obtained from China's online health communities (OHCs) covering the period from June 2021 to April 2022. The final dataset comprised 552 teams with a total of 1,347 physicians. To test the proposed hypotheses, we employed an empirical analysis based on a time fixed-effects model. Findings-The findings indicate that the knowledge attributes of OMTs exert a positive impact on physicians' financial and reputational returns, with physicians' self-efficacy serving as a key mediating mechanism. Furthermore, physicians' teamwork engagement moderates the relationship between knowledge attributes and self-efficacy in an inverted U-shaped manner. Originality/value-This study advances OMT research by showing how team-level knowledge attributes generate individual-level value for physicians. It distinguishes an external signaling route from an internal developmental route, thereby clarifying how OMTs shape both market-facing outcomes and physicians' self-efficacy. By identifying the inverted U-shaped boundary role of teamwork engagement, this study further shows that engagement in virtual medical teams should be optimized rather than maximized, offering implications for team design, knowledge presentation, and physician workload management in OHCs.
