In recent years, the proliferation of misinformation across online platforms has drawn considerable concern from elected officials, political analysts, and nonprofit organizations worldwide. Much of the discourse has centered on mainstream social media platforms dominant in Western countries, such as Twitter (now known as X) and Facebook. However, this focus often overlooks the distinctive ecosystems through which misinformation spreads in other parts of the world, notably in the Global South. In these regions, social messaging applications like WhatsApp play a crucial role in information dissemination, yet their influence remains understudied and less understood.
Unlike traditional social media feeds characterized by public timelines and algorithm-driven content, WhatsApp operates as an encrypted, private messaging platform where content is shared directly between individuals or within closed groups. This feature has facilitated the rapid spread of multimedia misinformation—particularly videos and images—that are highly shareable and can easily traverse social networks without the oversight mechanisms typical of public platforms. These dynamics have profound implications for democratic processes in countries reliant on WhatsApp as a primary communication tool.
Recognizing this gap, a team of researchers based at New York University’s Center for Social Media and Politics (CSMaP) undertook an empirical investigation into the effects of WhatsApp usage on misinformation during Brazil’s 2022 presidential election. Their experimental study uniquely tackled the challenge of isolating the impact of multimedia content by instructing participants to deactivate automatic downloads of images and videos on the app for a sustained period during the electoral campaign. This method allowed the team to analyze differences in misinformation recall and belief without the confounds of uncontrolled content consumption.
The findings illuminated a nuanced relationship between exposure to multimedia content on WhatsApp and the recall of false information. Participants who restricted multimedia downloads exhibited a significant decline in their ability to recall misinformation that circulated widely in the lead-up to the election, suggesting that limiting access to viral videos and images can dampen the spread of false rumors. This behavioral intervention offers a promising avenue for mitigating misinformation in contexts where direct content moderation or algorithmic intervention is infeasible due to the encrypted and private nature of the platform.
Nevertheless, the experiment also revealed trade-offs. The reduction in multimedia exposure correspondingly decreased participants’ recall of true news stories, albeit to a markedly lesser degree than the reduction observed for misinformation. This asymmetric effect raises critical questions about how information quality and truthfulness interact with content formats and user behavior in digital environments, indicating that efforts to curb misinformation must account for potential impacts on information diversity and citizen knowledge.
Perhaps most strikingly, despite the reduced recall of false headlines, no significant difference emerged between the treatment and control groups in terms of participants’ belief in false news. This outcome challenges assumptions that decreased exposure will straightforwardly translate into diminished acceptance of misinformation. Furthermore, the study detected no measurable shifts in political polarization levels, suggesting that diminished exposure to potentially divisive political content on WhatsApp may not suffice to counteract entrenched political divides.
Tiago Ventura, the study’s lead author and former postdoctoral researcher at CSMaP, emphasized the scarcity of research on WhatsApp’s political effects despite its centrality in many Global South countries. Ventura noted that the impact of social messaging platforms can be more limited and variable than commonly presumed, underscoring the need to investigate context-specific dynamics rather than extrapolating findings from Western-centric studies. He also highlighted the platform’s dual role, referencing WhatsApp’s documented function as a tool for political coordination and mobilization following Brazil’s 2022 election.
Importantly, Ventura pointed to heterogeneity in users’ ability to discern misinformation: individuals who encountered political content on WhatsApp multiple times daily showed an improved capacity to identify false rumors, whereas those with infrequent exposure were less proficient. This differential suggests that habitual interaction with politically oriented content may enhance critical evaluative skills, warranting further exploration into how repeated exposure shapes discernment and contributes to resilience against misinformation.
The study was methodologically rigorous, involving around 800 Brazilian participants recruited during the volatile electoral period from September 15 to October 5, 2022. Subjects were randomly assigned to a treatment condition requiring the deactivation of automatic multimedia downloads and abstention from consuming WhatsApp media for three weeks, contrasted with a control group instructed to avoid WhatsApp media only for three days. Compliance measures leveraged participants’ self-reported WhatsApp data storage metrics, providing an innovative proxy for multimedia consumption levels.
At the conclusion of the intervention, participants were assessed via a series of news summaries related to the election campaign. They were asked to recall which stories they had encountered previously on social media, encompassing both truthful and false information. By correlating recall accuracy with WhatsApp usage patterns, the researchers could delineate the influence of platform mechanics on misinformation dissemination and retention.
The broader implications of this research resonate beyond the specific Brazilian case, urging a recalibration of how scholars and policymakers approach the mitigation of online misinformation globally. Joshua A. Tucker, co-director of CSMaP and co-author of the study, emphasized the necessity to "move beyond what we know about the impact of social media usage on politics in the United States." Given that the majority of social media users worldwide reside outside the U.S., cross-contextual analyses are imperative to forge more effective strategies against misinformation worldwide.
This study also punctuates the challenges of addressing misinformation in end-to-end encrypted environments where conventional moderation tools are limited. The experiment’s results advocate for interventions targeting user behavior—such as disabling automatic downloads of multimedia—as a viable complement to platform-level policies. However, they simultaneously caution against simplistic expectations that reducing exposure will automatically diminish misinformation belief or political polarization.
As misinformation continues to threaten democratic norms and social cohesion, especially in democracies with volatile political landscapes, this research contributes critical empirical evidence to the ongoing dialogue. It sheds light on the complex interplay between platform architecture, user engagement, and information ecosystems in the Global South, painting a more intricate picture of how digital tools shape political cognition and public discourse in diverse contexts.
Future research must further explore mechanisms that enhance users’ media literacy and critical consumption skills, particularly among less frequent consumers of political content who appear more vulnerable to misinformation. Building resilience through education, community-based fact-checking initiatives, and tailored interventions could complement technical adjustments such as multimedia deactivation.
The interdisciplinary collaboration exemplified by the team at NYU—comprising experts in political science, communications, and data science—demonstrates the importance of multi-faceted approaches to untangling the complex problem of misinformation. Supported by philanthropic foundations, this work underscores how academic inquiry, combined with practical experimentation, can generate actionable insights to address one of the defining challenges of the digital age.
Subject of Research: People
Article Title: Misinformation Beyond Traditional Feeds: Evidence from a WhatsApp Deactivation Experiment in Brazil
News Publication Date: 6-Jun-2025
Web References:
- Journal Article DOI
- Brazilian Capitol Attack and Social Media Interaction
References: The Journal of Politics, Vol. Issue (2025)
Keywords: Social media, Misinformation, WhatsApp, Political communication, Multimedia content, Experimental study
