The use of artificial intelligence (AI) and the emergence of AI-generated virtual influencers have made it necessary to examine their effects on societal dynamics, particularly in the daily lives of young people. This study focuses on the representation of gender in AI virtual influencers and their impact on perceptions of reality, aiming to identify the expectations of generation Z (Gen Z) university students in Turkey regarding AI influencers. The data obtained in the study were analyzed through logistic regression. The results of a survey conducted with 406 university students in Eskisehir revealed that members of Gen Z prefer a more inclusive understanding of AI influencers rather than a gender-based approach. The findings indicate that AI influencers are not strongly associated with traditional gender roles. This study demonstrates that the representations of AI influencers blur the boundaries between reality and simulation, resonating with Baudrillard’s (1994) concept of simulacra. Gen Z possesses an evolving perspective on gender norms and prefers a more inclusive representation within AI influencer culture.
Academic writing remains a significant challenge for many first-year students in open distance and e-learning contexts within large enrolment modules. This study investigates the underlying causes of these challenges and evaluates the effectiveness of targeted, technology-mediated support interventions. Following the assessment and moderation of the first assignment in a first-year academic writing module, key areas of difficulty were identified, prompting the use of an online evaluation questionnaire to explore students’ experiences. Guided by transactional distance theory, the study aims to reduce students’ sense of isolation and enhance engagement through the integration of podcasts and vodcasts. A qualitative research design was employed, drawing on data from the questionnaire and document analysis of second assignment and take-home examination scripts. The findings indicate measurable improvements in students’ academic writing skills, including the construction of logical arguments, effective use of discourse markers for cohesion, structured paragraphing, paraphrasing, and the development of an academic voice. Improvements in citation practices, referencing accuracy, and language use were also observed. However, persistent challenges such as plagiarism and limited utilization of learning resources remained. This study contributes to digital pedagogy by demonstrating the value of multimedia interventions in enhancing academic literacy in online distance environments.
The widespread adoption of digital technologies has profoundly influenced the lives of children, youths, and adolescents, raising growing concerns about the psychological and social impacts of online engagement. In response, this study presents a comprehensive bibliometric analysis of the research landscape in this field, highlighting publication trends, key contributors, emerging research themes, and patterns of collaboration. Using data retrieved from the Scopus database, the analysis examined publication growth, identified the most influential authors, journals, and countries, explored co-authorship relationships among authors and nations, and assessed knowledge exchange through bibliographic coupling of countries. The results indicate a steady increase in publication output over the past decade, with a pronounced surge in recent years, reflecting heightened scholarly interest in this area. Several authors, institutions, and countries emerged as major contributors to the literature. Keyword co-occurrence analysis revealed prominent research themes, including cyberbullying, internet addiction, social media use, and their implications for mental health. In addition, the study examined collaboration patterns among researchers and journals and identified clusters of countries with strong collaborative ties. Overall, these findings provide valuable insights into the current state of research on the psychological and social effects of digital technology use on children’s and adolescents’ well-being in online environments, highlighting research gaps, collaboration opportunities, and future research directions. The bibliometric approach employed offers a distinctive perspective on the evolving knowledge landscape and supports efforts to strengthen collaboration and enhance research impact in this critical field.
Climate change communication (CCC) is an important area of research that examines how various strategies are used to raise awareness, educate, and promote action on the impacts of climate change. Although CCC holds a significant place in the environmental communication literature, there is limited research identifying the development and key topics in this field over the past 20 years. Therefore, this paper aims to provide a comprehensive overview of research on CCC worldwide from January 2006 to December 2025. To address the research question, we used bibliometric mapping to systematically assess and visually represent the development of this research field using VOSviewer to summarize and visualize research topics and the evolution of the frontiers in CCC literature. For data collection, 825 articles in the Scopus database were analyzed. The results showed that most journals with the highest number of publications are classified as Q1 in Scopus, and co-cited documents covered the field, which began with psychological behavioral theories, developed through media ethics and framing strategies, and has evolved into a multidimensional discipline that now encompasses political and economic constraints. In addition, the results demonstrated that emerging research themes included public perceptions, psychological and demographic dynamics, policy and risk management, public opinion and public health, social media, and geographic focus. Furthermore, the results highlight a significant research gap, indicating that future studies should examine CCC in diverse cultural environments.
Generative artificial intelligence (GenAI) refers to systems capable of producing original content from large volumes of data and deep learning models. There has been an increasing adoption of GenAI in higher education as cognitive support, a tool for academic production, and a resource for personalized learning. However, the extant empirical evidence demonstrates a lack of consensus and, in certain instances, a paucity of concordance, impeding a holistic comprehension of its educational impact. In this context, the objective of this research is to estimate the overall relationship between GenAI and university educational outcomes through a correlational meta-analysis. The study employs a quantitative approach and is conducted in accordance with the PRISMA 2020 guidelines as an international reporting standard. This ensures transparency, reproducibility, and rigor in the identification, selection, evaluation, and synthesis of evidence. The findings confirm that GenAI constitutes a relevant educational phenomenon whose impact cannot be interpreted from simplistic or deterministic perspectives. Whilst the relationship with educational outcomes is consistent and methodologically stable, it is strongly influenced by context, pedagogical decisions and usage patterns. The absence of homogeneous patterns that would permit direct generalizations is a salient finding, underscoring the necessity for nuanced and critical analyses. This demonstrates that the educational transformation associated with GenAI is contingent not on the technology itself, but rather on its pedagogical and institutional integration.
This scoping review maps the emerging evidence on how artificial intelligence (AI) technologies shape the psychology of sustainable consumption. Guided by Arksey and O’Malley’s (2005) framework and PRISMA-ScR guidelines, a comprehensive search was conducted on June 25, 2025, using Scopus and Web of Science, combining the keywords artificial intelligence, psychology, and sustainability. There were 1,561 publications when the corpus was limited to English-language papers published between 2020 and 2025. After removing duplicates, screening titles and abstracts, and reviewing full texts, 19 research satisfied the requirements for inclusion. Using descriptive charts and inductive thematic analysis, four main things were found: (1) anthropomorphic chatbots are the most popular AI touch-point and consistently increase people’s desire to buy green products; (2) AI effects are mediated by psychological states, such as social presence, hedonic motivation, perceived usefulness, and green satisfaction, rather than by technology alone; (3) algorithmic advice can be less effective than human guidance when moral or reputational stakes are high; and (4) theory building is heavily biased toward technology-acceptance models, leaving value-based and affective mechanisms under-explored. These findings highlight both the promise and the boundary conditions of AI-enabled persuasion and chart a research agenda that integrates richer motivational theories, hybrid human–AI designs, and longitudinal real-world evaluations.
Political microtargeting (PMT) has become a powerful tool for political campaigns, shaping voters’ exposure to tailored messages. This study investigates the impact of PMT exposure on political interest and engagement during an election campaign, distinguishing between within-person and between-person associations. Using a longitudinal design across three waves and applying a random intercept cross-lagged panel model, we examined how exposure to PMT influences individual-level changes in political outcomes over time. The results show strong positive associations at the between-person level, indicating that people who are more often exposed to PMT also report greater political interest and engagement. However, no significant within-person associations were found. Although these within-person relationships were not significant, they showed an interesting trend that was in line with the theoretical expectation that a sudden surge in PMT exposure (e.g., during an election campaign) might negatively impact political outcomes. The potential reasons for this non-significance are further explored in the manuscript. These findings contribute to the understanding of PMT’s role in shaping political attitudes and provide important directions for future research.
Solutions journalism (SoJo) has been widely promoted as a response to declining trust in news and growing audience disengagement, yet empirical evidence regarding its effectiveness remains fragmented across methods, contexts, and levels of analysis. This systematic review synthesizes 41 peer-reviewed studies (2016-2025) identified through preferred reporting items for systematic reviews and meta-analyses 2000 (PRISMA 2020)-guided searches to determine when, how, and under what conditions SoJo produces measurable outcomes. Integrating findings across contextual environments, narrative forms, audience processes, and institutional dynamics, the review shows that SoJo consistently improves affective responses and perceived efficacy but yields uneven behavioral effects. Outcomes vary systematically depending on narrative structure, audience relevance, and organizational feasibility, while structural constraints such as resource limitations, platform logics, and newsroom norms restrict scalability. By specifying how narrative design, media context, and institutional implementation interact in shaping civic outcomes, the study advances journalism communication theory and provides evidence-based guidance for practitioners. The findings reframe SoJo not as a universal remedy for journalism’s crisis but as a conditional innovation whose impact depends on identifiable mechanisms and contextual contingencies.
This study aims to examine the impact of university students’ motivations for social media use (information seeking, socialization, entertainment, and identity formation) on their intentions to use artificial intelligence (AI)-powered learning tools such as ChatGPT, Gemini, and Copilot. Designed within the framework of the technology acceptance model (TAM), the research addresses the mediating role of perceived utility and the moderating role of digital literacy. The study aims to contribute to the literature by understanding how social media habits evolve into academic technology adoption processes. This research, based on a sample of 370 university students, investigates the relationship between various social media motivations (social connection/FOMO, popularity/identity formation, appearance/impression management, and civic/advocacy) and behavioral intention (BI) to use AI learning tools. Perceived usefulness (PU) is included as a mediating variable in this relationship. Results from regression-based mediation analyses (PROCESS Model 4 equivalent) indicate that social media use motivations significantly predict both PU and BI. The indirect effect through PU was statistically significant (ab = 0.248, SE = 0.062, z = 3.988, p < .001), supporting a partial mediation model. Civic/advocacy motivations demonstrated the strongest relationship with PU and BI among subscales. These findings advance understanding of technology adoption in educational contexts and highlight the role of social media usage patterns in shaping AI tool adoption.
The study aims to provide a comprehensive bibliometric analysis of the academic literature on artificial intelligence (AI) ethics and social dynamics. Publications between 2020 and 2025 from Web of Science and Scopus databases were examined. The study aims to reveal expose the evolution, patterns, geographic distribution, and multidisciplinary structure of the subject of AI ethics. With an increase of over 100% in both databases, particularly between 2023 and 2024, the results show that the field has displayed fast expansion recently. Though there are variations in production and influence, the USA, the UK, and China dominate the field. Journal analysis shows that the journal “AI & Society” is the most influential publication in both databases. Keyword and thematic analyses show that while “AI”, “ethics” and “machine learning” remain central, new themes such as “ChatGPT” and “generative AI” are on the rise. Author collaboration networks reveal the multidisciplinary nature of the field and the existence of diverse research groups. Differences in coverage between databases suggest that Scopus better represents health sciences and current technological developments, while WoS better represents ethics. This study emphasizes that the research agenda in the field of AI ethics should be more inclusive and based on interdisciplinary collaboration and provides recommendations for future research directions.
This study examines the relationship between social media competence (SMC) and self-efficacy for artificial intelligence (AI) learning in university students on the basis of social cognitive theory. A quantitative and cross-sectional survey design was used in the study. In the data collection process, SMC scale and AI learning self-efficacy measures were used. The data obtained from 451 undergraduate students studying at Sechenov University in Russia were analyzed using partial least squares structural equation modeling (PLS-SEM). The findings showed that the content generation dimension (beta = .302, p <.001)and the technical usabilitydimension (beta = .273, p < .001) were the strongest positive predictors of AI learning self-efficacy. The anticipatory reflection dimension showed a weak but statistically significant negative effect(beta =-0.094, p = .034). The content interpretation dimension was found to have a weakly positive but statistically insignificant effect (beta = .073, p = .151). The model explained approximately 25 percent of the variance in AI learning self-efficacy (R-2 = .251). These findings suggest that competencies developed through active content production and technical use on social media platforms can increase students' confidence in learning AI. This result is in line with social cognitive theory's approach that emphasizes direct experience of success as the main source of self-efficacy. The study contributes to understanding how digital competencies transfer to learning confidence in new technological domains and provides practical implications for how social media-based strategies can be integrated into AI education.
International comparative studies show that adults’ data literacy is considerably lower than that of younger generations. Many older adults do not engage in continuing education and instead depend on journalists, functioning as knowledge brokers, to interpret and contextualize data. The news media have traditionally carried both an informative and educational function for their audiences. This article examines journalists’ own data literacy, given their potential to shape the public’s capacity to navigate data-rich environments. The study innovates by distinguishing data literacy as comprising numeracy, statistical literacy, and sociological data knowledge, and tests data literacy across various journalistic contexts. Drawing on a data literacy test conducted among journalists in Estonia and Turkey (respectively 10 and 20 interviews), the study identifies notable gaps in these three data literacy components. Whereas general math skills of journalists align with the numeracy level of the general population, knowledge of statistics and sociological data depends more on the particular educational contexts. The study reveals the necessity of additional data training for journalists to respond to the challenges facing in datafying societies. The study argues for targeted educational initiatives to strengthen data literacy among journalists and the wider population.
This study aims to examine the relationships among university students’ artificial intelligence (AI) literacy, AI ethical awareness, and technology use, and to determine the mediating role of AI ethical awareness in this relationship. The sample of the study consisted of 438 university students in Kazakhstan (233 female, 205 male). Data were collected using the AI literacy scale, AI ethical awareness scale, and technology use scale. Pearson correlation analysis, independent samples t-test, one-way analysis of variance, and mediation analysis with PROCESS macro (version 4.2) were employed for data analysis. The findings revealed that male students scored significantly higher than female students in AI ethical awareness and technology use according to the gender variable. Significant differences were found among age groups in terms of AI ethical awareness and technology use, with students aged 27 and above obtaining the highest scores. Regarding the field of study variable, social sciences students had the highest means in AI ethical awareness and technology use, whereas health sciences students demonstrated the lowest scores. The results indicated positive and significant relationships among AI literacy, AI ethical awareness, and technology use. Mediation analysis results revealed that AI ethical awareness played a partial mediating role in the effect of technology use on AI literacy. Technology use had both direct and indirect effects on AI literacy through AI ethical awareness. In conclusion, this study demonstrated that technology use influences AI literacy both directly and indirectly through the development of ethical awareness. The findings suggest that AI literacy education in higher education institutions should be designed with holistic approaches that incorporate ethical dimensions alongside technical content.
This study examines how institutional climate communication campaigns associated with 2023 United Nations Climate Change Conference (COP28) strategically deploy multimedia and artificial intelligence (AI)-assisted tools across social media platforms to shape public engagement. Adopting a mixed-methods research design, the study combines quantitative analysis of platform engagement metrics using PLS-SEM modeling with a complementary qualitative thematic reading of publicly available audience comments on Facebook, Instagram, YouTube, X (Twitter), and LinkedIn. The quantitative findings indicate that semantic-rich video content is positively associated with higher audience engagement (beta = 0.50, p = 0.007), AI-enabled content personalization is associated with improved campaign effectiveness measured through click-through and conversion indicators (beta = 0.42, p = 0.020), and interactive multimedia features are associated with stronger user interaction and retention (beta = 0.57, p = 0.004). These results are interpreted through media richness and social presence perspectives, alongside platform-based communication logics that emphasize the role of algorithmic visibility, data-driven optimization, and sociotechnical affordances in shaping institutional publics online. The study contributes to social media scholarship by extending empirical evidence on AI-mediated multimedia engagement to the domain of institutional climate communication, while highlighting implications for responsible personalization, transparency, and public trust within platform environments.
Education 5.0 is driving a paradigmatic shift toward human-centered, artificial intelligence (AI)-enabled learning ecologies in which Agentic AI autonomously orchestrates creative media production, content management, and instructional design. This study offers a mixed-methods systematic review of Agentic AI-driven creative media management in mass communication education, integrating bibliometric and qualitative evidence to address the current fragmentation of the field. Drawing on Scopus and Web of Science, the review combines bibliometric analysis using VOSviewer and Bibliometrix to map intellectual structures, collaboration networks, and thematic clusters, with inductive content analysis to synthesize strategic management dimensions, core components, and tools deployed in practice. The results reveal dominant thematic clusters, highly influential authors and sources, and evolving keyword trajectories that collectively delineate the research landscape of Agentic AI in Education 5.0. Content-level synthesis identifies key functional roles of Agentic AI in resource optimization, personalized content delivery, and autonomous workflow orchestration, while also cataloguing the specific platforms and architectures used. The review culminates in a comprehensive knowledge map that links macro-level research patterns with micro-level design features, and it foregrounds critical research gaps around AI governance, ethical frameworks, and contextual implementation. These insights provide a foundation for future scholarly work and offer actionable guidance for educational leaders and media managers seeking to responsibly integrate Agentic AI into creative media communication management.
This bibliometric study examines the evolution of scholarly research on fashion and gender over a thirty-year period, with the aim of identifying key trends, intellectual structures, and emerging research hotspots in the field. Accordingly, 1,940 documents on fashion and gender were indexed in the Web of Science database between 1990 and 2022 were analyzed. Network visualization and in-depth bibliometric analyses were conducted using the CiteSpace software. The results indicate that the USA, the UK, Canada, and Australia are not only the most prolific contributors to the literature butalso serve as central hubs connecting other countries engaged in fashion and gender research. The most highly cited journals include Journal of Consumer Research, Fashion Theory, Sex Roles, and Journal of Personality and Social Psychology, highlighting the interdisciplinary character of the field, which spans fashion studies, consumer research, social psychology, business, marketing, and management. Citation burst analysis reveals that authors such as D. Crane, R. Lewis, and B. Barry have recently gained increased scholarly attention. Furthermore, keyword burst analysis demonstrates a clear shift in research focus toward themes such as social media, politics, and antecedents, reflecting broader sociocultural and technological transformations. Overall, the findings provide valuable insights for researchers by illustrating how fashion, gender, sustainability, and social media are increasingly interconnected, and how these intersections contribute to ongoing debates on inclusion and within the fashion
This study investigates how glucagon-like peptide-1 receptor agonists (GLP-1 RAs) are discursively constructed on TikTok through a corpus-assisted multimodal discourse analysis of 100 English-language videos tagged #glp1, comprising 137,032 words of transcribed speech and 254 minutes of content. Integrating keyword and collocation analysis, stance evaluation, MIPVUbased metaphor identification, and systematic multimodal annotation, the study examines dominant linguistic patterns, evaluative strategies, and semiotic resources shaping GLP-1 RA narratives. Results suggest a predominantly weight-normative and biomedical framing, with positive stance appearing prevalent across both verbal and non-verbal modes. Metaphorical framing appeared to center on four recurrent domains: transformation, struggle/battle, journey/progress, and cravings as noise, potentially reinforcing pharmacological intervention as simultaneously medical and personal. Multimodal features largely appeared to align with evaluative stance across semiotic layers, though descriptive patterns tentatively suggest variation by creator type. The study points to how TikTok's communicative affordances may shape public understandings of pharmaceutical intervention, potentially privileging emotionally resonant narratives over clinically balanced discourse. It contributes to multimodal communication research by offering indicative evidence of how platform-specific semiotic resources may collaboratively shape health-related meaning-making, with possible implications for applied linguistics and digital health communication.
Online game addiction (OGA) has emerged as a significant psychological issue among university students, although the influence of interpersonal communication competency on this phenomenon is inadequately elucidated. This research investigated the correlations between interpersonal communication competence (ICC) and OGA among university students in Russia. The study involved 384 students from a state university and employed a convenience sample method predicated on voluntary participation. Data were gathered using an online questionnaire and examined utilizing partial least squares structural equation modeling (PLS-SEM). The results revealed that the environmental control dimension had significant negative effects on tolerance and withdrawal, while the interaction management dimension showed significant positive effects on significance and problems. Additional group analyses revealed that various relationships varied according to gender and daily Internet usage duration. Overall, ICC failed to successfully explain why people became addicted to online games. The study contributed to the research literature by demonstrating that ICC is not a consistent protective factor against OGA and that its effects are specific to certain dimensions and contexts. These findings suggest that communication-related issues should be examined alongside broader psychosocial variables in research and intervention efforts addressing problematic gaming among university students.
The study investigates how the media shape perceptions of security in the European Union's border regions, focusing on Latvia and Estonia, both of which share a border with Russia. Drawing on media dependency theory, it examines the relationship between media use and perceived threats under conditions of social instability and uncertainty. Two research questions guide the analysis: (1) how risk perception relates to goals of mass media use and (2) how border residents choose different information sources regarding potential military threats. The focus group discussions indicated that respondents' goals of media use closely aligned with dependency relations under conditions of perceived risk and uncertainty: maintaining a sense of safety and being informed, protecting family members and property, reducing uncertainty, distinguishing credible information from misinformation, and sustaining psychological stability. The risk perception and media use were mutually reinforcing: the media not only provided information but also regulated emotions, fostering both preparedness and fatigue. Differences between Latvian- and Russian-speaking audiences in Latvia and Estonia highlighted how competing narratives influenced stress levels and perceptions of security. The findings suggest that border residents having opportunity to live in different media spaces (or systems), lay on the media content they trust more, and have opportunity to compare sources and adopt protective behavior. In this context, media dependency extends beyond information-seeking to include emotional management and action-oriented responses during times of crisis
The aim ofthis study is to develop a valid and reliable scale evaluating students'views on artificial intelligence (AI) supported academic communication. Knowing how students view these tools is crucial considering AI's increasing presence in the classroom. We applied a thorough approach comprising contentvalidation, pilot research and factor analysis in addition to literature review. Initially, 40 items were created and reduced to 37 items after expert evaluation. As a result of the analysis of the data obtained from 580 participants, itwas determined that the scale showed a two-factor structure as "positive dimension" and "negative dimension". Factor analyses both exploratory and confirmatory helped to establish the scale's construct validity. The scale's internal consistency reliability came out to be really strong. Based on gender and age, Bayesian statistical studies revealed no appreciable variation in students' opinions of academic communication supported by AI. The developed scale offers academics and teachers a consistent instrument to evaluate students' perception of AI technologies. This scale will help to shape plans for better integration of AI into learning environments.
