Designing social media to foster user engagement in challenging misinformation: a cross-cultural comparison between the UK and Arab countries

The rapid spread of misinformation on social media platforms poses a significant challenge, impacting various domains from politics to health. Cognitive biases like "truth by repetition" and anchoring exacerbate the issue. While algorithm-based approaches are being developed to combat misinformation, social correction, where users challenge misinformation directly, offers a complementary strategy. However, user participation in social correction remains limited. This study leverages persuasive system design (PSD) principles and communication theories to explore design techniques that could foster prosocial behavior in challenging misinformation. Previous research, often limited to WEIRD (Western, Educated, Industrialized, Rich, and Democratic) samples, has identified several barriers to user engagement in correcting misinformation. This study addresses this limitation by comparing two distinct cultural contexts: the UK (a low-context culture) and Arab countries (high-context cultures). The PSD model provides a framework for analyzing persuasive principles categorized into primary task support, dialogue support, system credibility support, and social support. Each category offers specific principles, such as reduction, tunnelling, self-monitoring, liking, suggestion, verifiability, normative influence, and recognition. Prior research in the UK has identified seven potential design techniques to encourage social correction of misinformation on social media; this study will investigate the effectiveness of these techniques in different cultural contexts. It is hypothesized that the impact of these design techniques will vary depending on individual factors (personality traits, age, empathy, perspective-taking) and cultural context.

Existing literature highlights the challenges in combating misinformation, focusing on inoculation, automated detection and correction, and social media user interventions. Social correction shows promise, but user reluctance persists. Studies have explored reasons for this reluctance, including technical, content-related, self-related, other-related, individual characteristic, and relationship factors. Cognitive psychology theories offer insights into user reactions and correction attempts. Persuasive design techniques, using the PSD model, have proven effective in behavior change across various contexts. The PSD model systematically categorizes persuasive principles into four categories to influence user behaviors and attitudes. This study builds upon previous research that used the PSD model to develop seven specific design techniques for social media interfaces aimed at encouraging users to confront misinformation. However, this previous work was primarily focused on the UK. Cross-cultural studies on online behavior have often overlooked non-WEIRD cultures. Therefore, this research extends existing knowledge by examining the role of cultural factors in the effectiveness of persuasive techniques for combating misinformation.

This study employed a quantitative approach using an online survey distributed through Prolific and Cint platforms. A total of 462 participants (250 from the UK and 212 from Arab countries) participated, meeting specified criteria including age (18+), fluency in English (UK) or Arabic (Arab), active Facebook accounts with real identities, and prior experience encountering misinformation on Facebook. The questionnaire, available on the Open Science Framework (OSF), included sections on demographics (gender, age, education), willingness to challenge misinformation, and the perceived persuasiveness of seven design techniques. The techniques were presented via high-fidelity prototypes based on a misinformation scenario. Participants rated their likelihood of challenging the misinformation and the impact of each technique on a seven-point Likert scale. Personality traits were assessed using the 10-item Big Five Inventory (BFI-10), empathy and perspective-taking using the Interpersonal Reactivity Index (IRI). Data analysis involved descriptive statistics, t-tests, Mann-Whitney U tests, Wilcoxon signed-rank tests, and binomial logistic regression to compare the two cultural groups, assess the effectiveness of each technique compared to a standard comment box, and identify the influence of demographic and personality factors on the perceived persuasiveness of the techniques.

The study revealed a significantly higher willingness to challenge misinformation among Arab participants (M = 4.70, SD = 1.80) compared to UK participants (M = 3.30, SD = 1.94), t(458) = 8.01, p < 0.001, d = 0.75. Except for private commenting, all design techniques were perceived as significantly more persuasive in the Arab context than in the UK context. Predefined question stickers, thinking face reaction, sentence openers, fact checker badge, social norm messages, and tone detector all showed significantly higher persuasiveness scores in the Arab sample compared to the UK sample. The Mann-Whitney U test showed significant cross-cultural differences. Comparing each technique to a standard comment box using the Wilcoxon signed-rank test, predefined question stickers, thinking face reaction, and private commenting were significantly more persuasive than the standard comment box in both samples. However, sentence openers were significantly less persuasive than the standard comment box in the UK sample. In the Arab sample, the fact checker badge and social norm message were also significantly more persuasive than the standard comment box. Binomial logistic regression revealed various significant predictors for each technique's persuasiveness. For predefined question stickers, extraversion (Arab) and age and perspective-taking (UK) were positive predictors. For thinking face reaction, agreeableness (Arab) and openness (UK) were negative predictors. Age negatively predicted private commenting in the UK. Openness positively predicted social norm messages in the Arab context. No significant predictors were found for sentence openers and fact checker badge.

The findings suggest that cultural communication styles significantly influence the effectiveness of persuasive design techniques. The higher effectiveness of indirect techniques in the Arab context aligns with high-context communication, where nonverbal cues and context are crucial. In contrast, the UK's low-context preference for direct communication might explain the lower effectiveness of some indirect techniques. The similar effectiveness of private commenting across both cultures may be attributed to its balance between direct and indirect communication. The difference in effectiveness between the standard comment box and the other techniques highlights the potential benefit of providing structured options for challenging misinformation. Individual factors further moderate the effectiveness of these techniques, with age, personality traits, and perspective-taking playing significant roles across both cultural contexts. The varying effectiveness of the techniques across cultures highlights the importance of tailoring design interventions to cultural communication styles and individual differences.

This study demonstrates that persuasive design techniques can be effective in encouraging social media users to challenge misinformation, but their effectiveness varies significantly across cultures and individual characteristics. The study highlights the importance of considering both cultural communication styles (high vs. low context) and individual differences (personality, age, perspective-taking) when designing interventions to combat misinformation. Future research should explore the effectiveness of these techniques in other cultural contexts, use more direct behavioral measures, and investigate the role of other individual factors, such as risk aversion and cognitive ability.

This study relies on self-reported data, which might be subject to bias. The focus on Facebook acquaintances and a single time point limits generalizability. Within-culture variations (regional, socioeconomic) were not considered. The study focuses solely on social media; future studies could explore other communication contexts. Future research should employ experimental designs to directly measure behavioral changes in response to the design interventions. Future studies could also explore the influence of other relevant individual characteristics.