2026-01-27
Maria Rosaria Nappa, Sara Pabian, Mara Morelli, Angela Costabile, Elena Cattelino, Roberto Baiocco
Parasocial bonds with Social Media Influencers (SMIs) can significantly influence followers’ behaviors, attitudes, and well-being, especially during adolescence and young adulthood. This study examined how diverse dimensions of connectedness with SMIs (i.e., escape, imitation, modeling, and aspiration) relate to perceived social connectedness in offline, combined offline-online social connectedness and online contexts, as well as to social self-efficacy and problematic social media use, age and sex. A total of 554 respondents (346 from Italy and 308 from the Netherlands) aged 18 to 35 years ( M age = 23.85; SD age = 3.63; 67.6% female) completed an online survey. Structural equation modeling was employed to test the hypothesized associations, followed by multigroup analysis to assess cross-cultural differences. Findings revealed that offline social connectedness was negatively associated with the escape dimension, while offline-online connectedness showed negative associations with escape and imitation. In contrast, online connectedness was positively associated with these same dimensions. Thus, the form in which social connectedness is experienced seems to relate to more immediate and superficial forms of parasocial engagement. Social self-efficacy was positively associated only with aspiration, highlighting a selective, identity-driven engagement with SMIs. Problematic social media use was positively linked to all connectedness dimensions, suggesting its broad influence. Younger respondents were more inclined to imitation, modeling and aspiration compared to older respondents, while male respondents reported higher scores than females across all dimensions. Finally, multigroup analysis revealed significant differences between Italian and Dutch respondents, emphasizing the role of cultural context in shaping parasocial dynamics with SMs.
2026-01-27
Jing Niu, Bilal Mazhar, Yilin Ren, Inam Ul Haq
As generative artificial intelligence becomes more prevalent, understanding the factors that influence users' willingness to disclose personal information to these platforms is crucial. This study explores how Perceived Autonomy and Perceived Empathy in generative AI platforms influence users’ Behavioral Intent to Disclose personal information. Using a sample of 1,034 generative AI users in China, Structural Equation Modeling (SEM) was employed to analyze the relationships between autonomy, empathy, trust, privacy concerns, and disclosure intentions. The results show that both Perceived Autonomy and Perceived Empathy positively influence Behavioral Intent to Disclose personal information on generative AI platforms. These relationships are positively mediated by Trust in Generative AI and negatively mediated by Privacy Concerns, highlighting the role of trust in enhancing disclosure and privacy concerns in reducing it. These findings align with the Extended Technology Acceptance Model, which emphasizes the role of trust and privacy as external factors in shaping user behavior. This study provides valuable insights into how AI system attributes influence disclosure decisions, offering guidance for the design of more trustworthy and privacy-sensitive generative AI platforms.
2026-01-27
Matilde Buattini, Donald Barjami, Lorenza Paponetti, Dalila Torres, Rosita Borlimi, Gabriele Caselli
This exploratory qualitative study investigates ChatGPT-4’s capacity to apply the LIBETcase formulation model by analyzing its feedback on anonymized interview transcripts. The study aimed to assess whether ChatGPT-4’s outputs reflectedaccurate identification and interpretation of two key psychological constructs—lifethemes and semi-adaptive plans—while adhering to theoretical principles, and toexplore recurring errors and limitations in its clinical reasoning. Ten non-clinicalparticipants underwent semi-structured interviews, and a custom-configured versionof ChatGPT-4 was provided with structured instructions and theoretical material.Reflexive thematic analysis revealed four overarching themes: (1) limitations inabstraction and interpretative barriers, (2) consistent structure and contentorganization, (3) hypothesis-driven reasoning with cautious language, and (4) partialadherence to LIBET theory through appropriate terminology. While ChatGPT’sstructured reasoning and alignment with theoretical vocabulary suggest its potentialas a reflective support tool—particularly in training or supervision—it also showeddifficulties in distinguishing emotional vulnerabilities from coping strategies, and ininterpreting abstract, relational constructs such as life themes. Findings support theimportance of improving prompt design, expanding training on psychologicalconstructs, and developing rigorous validation pipelines. Future research shouldaddress these limitations before deploying LLMs as assistive tools in clinical reasoningand decision-making.
2026-01-27
Cheng-Hong Liu, Fa-Chung Chiu, Po-Sheng Huang, Ting-Huan Lin, Chia-Lin Chang
Empathy is a key predictor of bystanders’ helping behaviors, yet its role in cyberbullying contexts has yielded inconsistent findings. This study investigated whether accuracy motivation, the drive to form correct and well-reasoned judgments, moderates the relationship between dispositional empathy and helping tendencies among cyberbullying bystanders. In Study 1, 164 Taiwanese undergraduates (61 males, 100 females, 3 unknown; M age = 22.96, SD = 14.25) read a cyberbullying scenario and completed measures of empathy, helping tendencies, and need for cognition (NFC), a personality trait reflecting enjoyment of cognitively effortful tasks, widely used as a proxy for accuracy motivation. Results showed that the positive association between empathy and helping tendencies was weaker among participants with higher NFC scores (i.e., higher accuracy motivation). In Study 2, 180 undergraduates (67 males, 113 females; M age = 20.96, SD = 5.44) were randomly assigned to either control or high accuracy motivation group. Findings revealed that, for participants in the control group, empathy predicted helping through increased feelings of responsibility. This mediational pathway was not observed in the high accuracy motivation condition. These findings suggest that accuracy motivation may attenuate the influence of empathy on helping behavior in cyberbullying contexts. Theoretically, this provides a possible explanation for the mixed results observed in prior research. Practically, it highlights the importance of encouraging individuals to prioritize providing timely support in cyberbullying incidents, rather than becoming overly focused on verifying the authenticity of online messages.