Wenbo Li

This study examined whether basic science knowledge helps explain education-based gaps in domain-specific science knowledge and whether intentional and incidental exposure on social media condition those gaps. A national survey of American adults focused on two domains: climate change and algorithms. Results showed that basic science knowledge mediated the relationship between education and domain-specific knowledge across both domains. These indirect relationship were attenuated at higher levels of intentional social media exposure. The findings suggest that strengthening foundational science literacy and encouraging more deliberate engagement with science content on social media may help reduce knowledge disparities.

This study investigated how scientific consensus messaging influences public support for socially good artificial intelligence (AI), drawing on the gateway belief model. Two preregistered experiments found that consensus messages increased support for AI research and willingness to raise awareness indirectly through increasing perceived consensus, which enhanced favorable beliefs and hope. These effects varied by collective scientific efficacy—perceived ability of scientists to work together effectively. Belief-based pathways were stronger under low efficacy, while hope-based pathways were stronger under high efficacy. Findings underscore the value of combining consensus and efficacy cues in science communication to promote engagement with emerging technologies like AI.

This study synthesized research from 4 theoretical frameworks that explicitly conceptualize media and communication as dynamic processes over time, emphasizing reciprocal and self-causing effects. A meta-analysis of 223 effect sizes from 103 studies supported the presence of both reciprocal and self-causing relationships, suggesting their robustness as communication phenomena. Time lag moderated self-causing effects, such that longer intervals weakened these effects. Longer lags attenuating the influence of media/communication on subsequent outcomes approached but did not reach statistical significance, and did not significantly alter the reverse pathway. Exploratory moderation analyses showed that effect sizes varied by audience age, analytic methods, topic areas, and media types. Findings highlight the importance of time lag specification in studying dynamic media effects.

This study investigates individuals’ lay definitions—naïve mental representations—of artificial intelligence (AI). Two national surveys in the United States explored lay definitions of AI in the workplace (Study 1) and in everyday life (Study 2) using both open- and closed-ended questions. Open-ended responses were analyzed with natural language processing, and quantitative survey data identified factors associated with these definitions. Results show that conceptions of AI differed by context: workers emphasized efficiency and automation in the workplace, while the general public linked AI to diverse everyday technologies. Across both groups, conceptions remained nuanced yet limited. Sociodemographic factors and personality traits were related to sentiments expressed in definitions, and greater trust in AI predicted more positive sentiments. These findings underscore the need for targeted training and education to foster a more comprehensive public understanding of what AI is and what it can do across different contexts.

We conducted a comprehensive meta-analysis of 160 media literacy interventions from 1983 to 2023, finding overall positive effects. Effects were larger for media-over behavior-relevant outcomes and for social over health topics. Stronger effects were found for knowledge outcomes at both immediate-post and delayed follow-up assessments. Attitudes and critical beliefs predicted behavior. No substantial decay in effects was observed between immediate-post and follow-up assessments for multiple outcomes but not for behavior. Interventions targeting social media showed smaller effects for some outcomes compared to those that did not. Online delivery had smaller effects for various outcomes compared to in-person delivery. These findings suggest challenges and opportunities regarding digital and social media for the interventions. Dose was inconsistently linked to outcomes. No significant relationship was found between study year and effect size. The heterogeneity of effects observed across variables suggests a need for more parsimonious frameworks in media literacy research.

Past studies have investigated the variability in how people engage with their personal networks, yet less is known about how people perceive their energy allocation to different ties. Drawing on an online survey sample (N = 906), we tested whether subjective perceptions of energy allocation conform to so-called Dunbar’s Number(s). In addition, we evaluated the predictive roles of Big Five personality traits and self-esteem while controlling for differences in network structure. Results revealed significant heterogeneity in perceived energy allocation to different layers of personal networks (i.e., inner 5 vs. middle 15 vs. outer 150 relationships). In contrast to expectations, extraversion was not associated with perceived energy allocation, whereas self-esteem was associated with greater energy allocation to the middle (vs. inner) network layer. Our findings add to our knowledge of how people perceive relationship maintenance across their personal networks, along with the links to key psychological traits. More broadly, the findings suggest that more attention should be paid to psychological implications of the middle layer of personal networks. To conclude, we discuss the importance of studying individual differences in how people prioritize – and reflect on – different relationships in their networks.

Employing a US national sample (N = 1041), this project investigates the relationship between media usage and vaccine hesitancy, as well as the potential variations in this relationship based on vaccine types and individuals’ racial backgrounds. Findings revealed that participants held different levels of acceptance of different vaccines. Compared to other racial groups, African Americans held a lower level of positive attitudes toward different types of vaccines; however, racial groups did not differ significantly in terms of intention to receive vaccines. Individuals’ exposure to information emphasizing various aspects of vaccination either mitigate or enhance their vaccine hesitancy. In addition, the number of social media platforms individuals regularly use is negatively associated with their vaccine hesitancy. The study offers valuable insights for addressing vaccine hesitancy in diverse populations. Theoretical and practical implications were discussed.