Decriminalizing same-sex relationships and its impact on social media discussions in Asia

nature human behaviour Article https://doi.org/10.1038/s41562-026-02485-6 Decriminalizing same-sex relationships and its impact on social media discussions in Asia Received: 5 March 2024 Kokil Jaidka 1 , Preetika Verma 2 & Yiting Chen3 Accepted: 23 April 2026 Published online: xx xx xxxx Check for updates While legal reforms may coincide with shifts in public discourse, broader normative change often remains uneven and context dependent. Since 2018, four Asian countries—India, Taiwan, Thailand and Singapore— have decriminalized same-sex relationships, providing an opportunity to examine how such institutional changes are associated with public discussions of LGBTQ+ topics. Here, using supervised and unsupervised natural language processing methods, we analysed 136,255 Facebook news posts and 65,969 public Instagram captions within a 100-day window before and after each reform. To contextualize online discourse with offline sentiment, we incorporated nationally representative survey data from the Gallup World Poll (N = 46,200). Following decriminalization, online discussions were characterized by reduced moralized and affectively negative language, particularly in contexts with greater freedom of expression. By contrast, offline perceptions of neighbourhood safety for same-sex couples shifted more modestly and unevenly across age groups and countries. Nearly 35% of countries continue to criminalize same-sex sexual activity1; however, under increasing international pressure, many have taken steps towards greater inclusion of same-sex relationships2. For instance, in 2018, India repealed the 377A law and legalized consensual same-sex intercourse between adults3. Four years later, Singapore repealed a similar 377A law4. Meanwhile, Taiwan legalized same-sex marriage in 20195. In 2022, Thailand announced a future date after which same-sex marriages would be legalized6. However, the normative acceptance of same-sex relationships goes beyond legislative action and is intertwined with social and government infrastructural support. Especially in conservative societies, individuals who violate societal ‘moral contracts’ often face hate speech7–9. To further explain the dynamics between legislative change and social attitudes, Structuration Theory10 offers a valuable lens, as it posits how social practices are shaped by both individual agency (the actions of individuals) and structures (institutional frameworks). Legislative changes, such as the decriminalization of LGBTQ+ identities, can act as catalysts to changes in social norms and attitudes. When governments formally recognize LGBTQ+ rights, they help reinforce societal values of inclusion, which can then influence public discourse, both offline and online. Therefore, the actions of institutions (such as the decriminalization of same-sex relationships) and individual expressions on social media are interconnected, with both contributing to shifts in public opinion and acceptance of LGBTQ+ individuals. For instance, a lack of representation in government policies may reflect in how LGBTQ+ people are underrepresented or vilified in the media and online or portrayed through stereotypes11,12. This may also explain why individuals were more likely to express their opinions and provide their rationale on LGBTQ+ issues when news emphasizes LGBTQ+ issues13. Most studies on public opinion towards LGBTQ+ people rely on national-level surveys, which may not capture online realities and are often affected by government censorship and social desirability biases, particularly in conservative societies. Yet, the debate over whether online spaces are more inclusive for marginalized groups, such as LGBTQ+ individuals, or more vilifying compared with offline realities remains unresolved. Few opportunities exist to explore how external changes, such as the lifting of laws criminalizing same-sex Department of Communications and New Media, National University of Singapore, Singapore, Singapore. 2School of Computer Science, Carnegie Mellon University, Pittsburgh, PA, USA. 3School of Journalism and Mass Communication, University of Wisconsin Madison, Madison, WI, USA. e-mail: 1 Nature Human Behaviour Article relationships, shape public discourse and media attention to LGBTQ+ issues. This motivates our first research question: RQ1: What are the effects of decriminalizing same-sex relationships on (a) news coverage and (b) social media attention to LGBTQ+ topics? Empirical findings present a mixed picture of whether social media helps LGBTQ+ individuals bridge identity differences with broader society14. On the one hand, some research highlights social media’s role in promoting acceptance and advancing LGBTQ+ rights15,16, while on the other, studies suggest that platform design and anonymity can reinforce social norms that restrict open expression, particularly for queer individuals17–19. In conservative societies, individuals who violate dominant ‘moral contracts’ often face hate speech and backlash7–9. For instance, a UK survey found that over 60% of LGBTQ+ respondents experienced online harassment20, and such exposure has been linked to psychological distress and social distrust21–24. Beyond measuring incidence, however, we lack a fine-grained understanding of how users morally reason about LGBTQ+ topics across different cultural and platform contexts. Prior research shows that moral language often structures public discourse25, and that moral values shape how news events and marginalized groups are framed26–29. LGBTQ+ issues are particularly likely to trigger moral intuitions that influence stereotypes, judgements and public evaluations of who deserves moral consideration30. To examine these dynamics, we use Moral Foundations Theory (MFT) as a measurement framework. MFT organizes moral reasoning into two key clusters: individualizing foundations (care/ harm and fairness/cheating), which emphasize justice and protection of individuals; and binding foundations (loyalty/betrayal, authority/subversion and sanctity/degradation), which emphasize group cohesion, tradition and purity31. These foundations provide a structured way to measure how moral stereotypes manifest in discourse about LGBTQ+ identities, particularly in response to major institutional shifts. While MFT has been widely applied to political ideology and moral framing, few studies have leveraged it to systematically analyse public discourse about LGBTQ+ issues across multiple social media platforms and national contexts. Recent work has shown that moralized language is a key component of online toxicity32, highlighting the need for a framework that can detect shifts in moral reasoning over time. Given the dual legal and moral nature of decriminalization, our next research question asks: RQ2: How does the legal acceptance of same-sex relationships affect (a) affective negativity and (b) moral stereotypes in LGBTQ+ social media discourse? Structuration theory would also e (...truncated)