Abstract: Abstract Using virtual reality (VR) to revisit the past is a pervasive theme in philosophy, fiction, and futurism. We conducted a longitudinal exploratory study (N = 136) investigating how collectively revisiting shared real-world social interactions differs from envisioning plausible interactions in VR, and how experiencing self-built virtual environments or those built by others influences psychological and behavioral outcomes. Participants who revisited, compared to those who envisioned, reported greater increases in group cohesion, as well as higher collective psychological ownership and liking of virtual environments. They also used more first-person plural pronouns and language related to affiliation and social behavior. Participants in self-built virtual environments, compared to those in other-built environments, stood closer to others and reported greater increases in group cohesion and higher collective psychological ownership of the virtual environment. We discuss theoretical advances and practical implications for using VR to revisit memories.

Abstract: Abstract Beyond digital news consumption, users may express themselves in relation to the consumed news—for example, through commenting, sharing, or reacting. They may also limit what is termed here news engagement visibility, the extent to which a user’s involvement with news content can be seen by others. Drawing on privacy calculus theory and literature about engagement, avoidance, and relationship management, this study examines—through 50 in-depth interviews in Canada—how and why news users limit their news engagement visibility. It presents three ways for limiting this visibility, including lower expression volume, more private expression space, or more closed-ended expression type. Furthermore, it introduces the four P’s framework—protection, pointlessness, personality, and particularity—to explain the reasons behind limited news engagement visibility. The study advances the understanding of news engagement by suggesting that it involves the management of visibility boundaries and by elucidating barriers to public expression. The implications of these contributions are discussed.

Abstract: Abstract Building on the idea that subjective evaluations of social media use (SMU) may shape well-being effects, this study investigated how two “social media mindsets”—agency (perceived control over SMU) and valence (perceived effects of SMU)—relate to self-reported and logged SMU, and four well-being indicators (depression, anxiety, stress, and life satisfaction) in a South African sample of young adults (N = 1,858; M age = 21.01). Agency mindsets were negatively correlated with SMU and positively with well-being, while valence mindsets were associated with higher SMU but not with well-being. Logged SMU showed weaker associations with mindsets and well-being than self-reported SMU, suggesting differences between perceived and actual behavior. The findings replicate, extend, and in some cases contradict prior research, emphasizing how users’ sense of control is more important for well-being than SMU duration or perceptions of its effects, and that more work is needed to understand whether the “mindsets” concept is fruitful.

Abstract: Abstract Individuals rely on messaging platforms to form and maintain intimate relationships, trusting shared information will remain within intended digital confines. However, the screenshot feature allows people to capture and store pieces of private conversations as a separate file on their device, rendering them shareable with third parties. While usage of this feature can be benign, this study focuses on its ability to breach privacy expectations within messaging platforms, termed within communication privacy management theory as privacy turbulence. This study recognizes the power of both interpersonal dynamics and platform affordances in constraining existing norms around screenshot collection and sharing others’ private messages. Experimental results (n = 302) suggest obscuring received messages upon use of the screenshot feature and stating an explicit privacy rule significantly reduce screenshot collection and sharing, respectively. Implications for communication theory and the future of messaging platform design will be discussed.

Abstract: Abstract Digital inequality research shows that certain sociodemographic groups (e.g., older persons) experience fewer benefits from using digital technologies (e.g., not finding useful information). While disadvantage is typically conceptualized as not experiencing benefits, users also experience online harms (e.g., unauthorized access to personal information), which have been largely ignored. Better understanding of systematic differences in online experiences can contribute to understanding mistrust in online companies and chilling effects (i.e., inhibiting certain digital behaviors). Using Bayesian inference to analyze data of a preregistered survey-based study (N = 1,410), we find confirming and contradicting results for digital inequalities. While older and less educated persons experience fewer online benefits and women and older persons tend to inhibit their digital behaviors more, no inequalities regarding harmful privacy experiences were found (older persons even reported fewer harmful experiences). Generally, mistrust and self-inhibition are higher when people experience online harms and lower when they experience online benefits. We discuss these findings in relation to structural disadvantages, intersectionality, and potential solutions.

Abstract: Abstract Adolescents perceive that they have digital well-being when smartphone use benefits outweigh the drawbacks in the social, cognitive, and emotional domains. Perceptions of digital well-being play a role in digital media effects, yet have received little research attention. This 1-year, three-wave panel study among 1,081 Slovenian adolescents investigated the reciprocal relationships between perceived digital well-being and self-esteem, with gender, parental education, and smartphone screen time as moderators. Random intercept cross-lagged panel models demonstrated a significant positive between-person relationship between perceived digital well-being in the emotional domain and self-esteem, but not for the social and cognitive domains. A positive, inconsistent within-person, cross-lagged relationship occurred between self-esteem and perceived digital well-being in the cognitive domain. Unstable differences occurred in the links between gender and the social domain and between smartphone screen time and the cognitive domain. These findings offer new insights into the debate on the effects of smartphone use.

Abstract: Abstract For those with marginalized identities, social media may offer opportunities for identity exploration and affirmation but may also pose challenges of context collapse, wherein the presence of multiple audiences makes identity-related posting awkward or dangerous. To probe short-term longitudinal patterns, 337 LGBTQ+ 15- to 22-year-olds completed six weekly surveys during Pride Month. As hypothesized, identity exploration was associated with self-reported frequency, positivity, and authenticity of youths’ LGBTQ+ self-presentations online, primarily among those who, on average, rated their online audiences as less heterogeneous (i.e., perceived less context collapse). Consistent with the self-effects model, positivity of self-presentations predicted increases in self-acceptance and self-esteem, with the latter association stronger for those who, on average, received more supportive feedback. In contrast, frequency and authenticity of self-presentations did not predict self-acceptance and self-esteem.

Abstract: Abstract Unintended encounters with news on social media can foster a reliance on incidental news exposure and the perception that “news finds me.” Because the “news-finds-me” (NFM) perception has negative consequences for informed and engaged citizenship, explaining it has become a priority in recent years. This two-wave survey of adults in the USA incorporates scholarship on social media habits and social media mindsets for an expanded view of the antecedent psychological conditions of the “news-finds-me” perception. Path analyses indicate habitual use of social media to have direct effects and core beliefs about social media and algorithmic control to have indirect effects through habitual use. These patterns hold for two of the three sub-dimensions of the NFM perception, indicating habit and mindset to be fruitful theoretical directions for future research.

Abstract: Abstract The rise of large language models (LLMs) expands opportunities for creating interpersonal messages. Building on artificial intelligent (AI)-mediated communication (AI-MC), this study examines how people use LLM-based chatbots to generate support messages and how patterns of human–AI collaboration shape message features and influence message evaluations of helpfulness and authenticity. We propose the process-adoption model, categorizing message generation into four patterns: human-only, AI-only, modified-AI, and AI-guided. Results showed that AI-only and modified-AI messages were more likely than human-only messages to include informational and emotional support, which in turn, enhanced viewers’ evaluations of message helpfulness and authenticity. AI-guided messages were more likely to provide reciprocal self-disclosure than AI-only messages, which enhanced perceived authenticity. Lastly, AI-guided messages were rated as more authentic than AI-only messages even after accounting for the mediating effects of message features. These findings provide a nuanced understanding of the AI-MC spectrum, and discussions are provided about human–AI collaboration in support provision.

Abstract: Abstract Social media technologies have been criticized as ineffective sources of information because users seem to increase their subjective but not their objective knowledge. While different factors such as multitasking have been brought up as explanations, direct and fair comparisons of learning outcomes between social and traditional news media are missing. This two-wave field experiment (N1 = 902, N2 = 663) shows that whereas participants receiving information on social media did learn less than participants receiving the same information through an email newsletter, this difference disappeared when cognitive elaboration on social media was triggered through elaboration questions. Correlational analyses further suggest that multitasking is one factor which hinders successful knowledge acquisition on social media. Our findings suggest that learning through social media is possible if social media contents invite cognitively elaborate processing.

Abstract: Abstract Social media browsing has been linked to both declines and improvements in affective well-being, with recent research suggesting its effects depend on key factors. This experience sampling study among emerging adults (N = 108, Mage = 22.29, 61 female) examines three such factors: levels of analysis, underlying reasons, and content valence. Results reveal no significant between-person associations, but a small average within-person association, suggesting that, on average, social media browsing slightly reduces positive affect. However, this effect was only statistically significant in one model, and person-specific analyses showed no significant individual-level effects, underscoring the high statistical uncertainty. Additionally, when examining the underlying factors, browsing was modestly linked to lower affective well-being when driven by habit and when individuals encountered positive content. This study contributes to a deeper understanding of the complex relationship between social media browsing and well-being, emphasizing the importance of considering key underlying factors when interpreting these effects.

Abstract: Abstract Mobile and social media are often identified as a means of enhancing social connection for those who wish to connect. In line with research on relational maintenance and the communicate bond belong theory, a need for connection can drive individuals to reach out to relational partners through mobile and social media, which might boost overall well-being. Using 3 longitudinal datasets that differ by time period (i.e., 6 years, 1 year, 1 month) and context (i.e., older adults’ pandemic-era communication, college student communication with friends, daily diary study of social media use), we test whether change in sociality moderates the association between frequency of mobile and social media use and well-being. In contrast to our hypothesis, when experiencing declines in sociality, the use of mobile and social media was either unrelated to well-being or associated with a decline in well-being. These findings challenge long-held assumptions about relational maintenance through media.

Abstract: Abstract Personalized marketing driven by AI-powered algorithms on social media has introduced significant privacy challenges, prompting users to adopt various privacy management strategies. This study employs latent profile analysis (LPA) to identify distinct user profiles based on privacy management patterns and explore the protection motivation factors that predict them. Using a cross-national survey in the United States and Singapore (N = 2,078), we identified four latent user profiles in both countries. “Privacy-benefit Maximizers” actively engaged in all privacy management strategies, including information withdrawal, disclosure management, and avoidance. The “Privacy Unnerved” group relied predominantly on avoidance strategies while showing low engagement in other privacy practices. “Balanced Guardians” adopted a moderate approach across all strategies. The fourth group, unique to each country, exhibited distinct patterns in their reliance on privacy management strategies. Among privacy motivation factors, privacy self-efficacy emerged as a key predictor of profile membership across different user groups in both countries.

Abstract: Abstract While the spread of toxicity on social media has received ample attention, studies offer conflicting expectations and mixed evidence about its contagious nature. This study advances understanding of how toxicity becomes contagious by distinguishing between exposure to toxicity by ingroup versus outgroup members and separating toxicity into impolite style and intolerant substance. We focus on Israel during 2023, a period marked by intense political polarization. Using Twitter panel data from original and replication datasets, we analyze ∼1M tweets from 12,481 users and ∼6M tweets from the 713,231 accounts they follow. We find that exposure to toxic behavior by ingroup members is the primary driver of contagious toxicity, compared to smaller, less consistent associations with outgroup toxicity. Moreover, seemingly less harmful forms of toxicity (impoliteness) by the ingroup are not only associated with increased users’ impoliteness but also with increased intolerance toward political groups—which may carry troubling implications for democracies.