Mapping Coordination Across Borders
The papers gathered here share a conviction that coordinated behaviour and problematic information are best understood not as isolated national pathologies but as patterns that recur—sometimes in strikingly similar forms—across electoral and health-crisis contexts worldwide. The topic’s title announces its method as much as its subject: these are comparative, often collaborative, studies that use the apparatus of computational social science (coordination-detection algorithms, network analysis, cross-platform scraping) to trace how narrative contests, inauthentic amplification, and populist communication travel across Germany, Brazil, Nigeria, India, Serbia, Korea, the Philippines, and beyond.
Brazil as a Recurring Laboratory
Brazil anchors a disproportionate share of this literature, serving almost as a natural experiment in democratic stress. Rodarte2026-dk shows how parliamentarians contested epistemic authority during the Manaus COVID collapse, with government critics borrowing journalistic form while Bolsonaro allies attacked the press directly—establishing a template of elite-driven narrative warfare that recurs across the corpus. Rossini2026-jn and Rossini2026-mj extend this into the 2022 election itself, using panel surveys to show that belief in electoral fraud claims both increases political intolerance and erodes trust in results, while mainstream news use (even among Bolsonaro voters) bolsters democratic resilience—a finding that complicates any simple “social media causes polarization” story by isolating informational pathways rather than platform use per se. Kansaon2025-id and Ventura2025-sw push into WhatsApp’s closed ecosystem, showing coordinated accounts seeding misinformation that catalyzed real-world mobilization (including the January 8 attacks), and that deactivating the platform measurably shifts belief and knowledge—together arguing that feed-based platform research has been looking in the wrong place for the Global South. Ventura2026-yc closes this arc with the 2024 X ban, demonstrating that platform-level interventions ostensibly aimed at curbing misinformation can produce durable partisan “sorting ratchets,” an unintended consequence that resonates with this topic’s broader skepticism about platform governance as a clean fix. Inacio-da-Silva2026-zf rounds out the Brazilian cluster methodologically, showing that volunteer-based ad auditing can surface irregularities platforms themselves do not disclose.
Detecting Coordination: Method as Argument
A second thread treats coordination detection itself as the object of methodological innovation and theoretical refinement. Nizzoli2020-cf reframes coordination as a continuum rather than a binary, showing—via the 2019 UK election—that the most tightly coordinated clusters are small activist and antagonist groups, and that coordination is orthogonal to automation. This conceptual move underwrites later work: Kuznetsova2025-nu finds analogous forwarding-based coordination among pro-government Telegram channels in Russia and Belarus, while Kulichkina2026-zk documents coordinated multilingual activity during China’s COVID protests, showing the same infrastructure enabling both dissent and repression. Jovanovic-Harrington2026-ze pushes the detection problem further by arguing that Serbia’s SNS-linked networks evade standard algorithmic signatures by mimicking authentic grassroots support, a warning that detection frameworks built on centralized bot farms may already be obsolete. Gerard2025-br’s CANE framework responds to this evolving arms race by modeling users through shared narrative participation rather than interaction signals, identifying small “bridge user” populations that disproportionately seed cross-platform narrative migration between Truth Social and X—a structural reconceptualization of brokerage that dovetails with Gaw2025-ru’s ethnographic account of Philippine influence operations as a form of political brokerage, mediating between clients, platforms, and voters through durable political-economic infrastructures rather than ad hoc propaganda.
State Power, Propaganda, and the Authoritarian Register
Several papers extend the coordination lens into explicitly state-linked and authoritarian contexts. Kim2026-br offers a two-decade, 112-million-comment analysis of suspected state-linked trolling in Korea, finding that condemnation—rather than praise—is the dominant and most algorithmically rewarded rhetorical strategy, aimed at amplifying polarization across the ideological spectrum. Makeev2026-ma theorizes a “dual-objective” propaganda model distinguishing Russia’s domestic wartime messaging from its foreign-facing narratives, while Beacken2026-zb broadens the aperture to six democratically weakened states (including Nigeria and India) to argue against technologically deterministic accounts of generative AI’s political effects, insisting instead on structurally grounded, context-sensitive comparison. Eady2023-xg serves as an important counterweight within this cluster: its careful linkage of survey and Twitter-timeline data on the Russian IRA campaign finds exposure heavily concentrated among already-partisan Republicans and dwarfed by domestic political content, with no measurable effect on attitudes or votes—a finding in productive tension with Kim2026-wg, whose direct-exposure tracking of 2016 US voter-suppression ads finds real, geo-racially targeted turnout effects, suggesting that aggregate null findings can mask substantial harms concentrated in targeted subpopulations.
Radical-Right Networks, Visibility, and Transnational Diffusion
A substantial cluster turns from covert coordination to the more visible machinery of radical-right digital campaigning, particularly around the 2024 European Parliament elections. Tornberg2025-ir finds across 26 countries that misinformation sharing by parliamentarians is driven specifically by the conjunction of populism and right-wing ideology, not either alone. Darius2026-xl and Iris2026-pg corroborate and extend this at the level of platform engagement and media visibility respectively: populist radical-right parties receive disproportionate audience support (especially on TikTok) and disproportionate news coverage relative to their electoral weight, a visibility premium that Gattermann2025-yx links back to voters’ own disinformation concerns. Bouchafra2026-ts and Klinger2023-qu provide granular case studies of how this amplification is engineered—the Sweden Democrats tailoring securitizing visual rhetoric platform-by-platform, and the AfD “laundering” Identitarian Movement content through intermediary superspreader networks into mainstream legitimacy. Kalsnes2025-zb shows the underlying emotional mechanics at a Scandinavian scale: populist parties excel at generating “Angry” reactions that convert disproportionately into shares. At the transnational level, Heft2021-ky and Groebner2026-pc trace how right-wing news ecologies hyperlink and audience-overlap across borders, with US content and audiences functioning as a persistent structural hub—evidence that radical-right digital communication increasingly operates as a connected transnational system rather than a set of parallel national phenomena.
Infrastructure, Access, and the Limits of Comparison
A final, reflexive thread interrogates the very possibility of comparative research in this domain. Humprecht2020-gd supplies the field’s foundational comparative framework, clustering eighteen Western democracies by structural “resilience” to disinformation and isolating the US as a singular outlier—a warning against overgeneralizing from any one national case that implicitly licenses the cross-national designs found throughout this topic. Nenno2025-xa operationalizes a parallel warning empirically, showing that news-value signatures of flagged misinformation differ between WEIRD and non-WEIRD countries, with the US and Brazil again emerging as distinct outliers. Lukito2026-nb and Schulte2026-df close the topic by turning methodological frustration into argument: both show that the central bottleneck in multi-platform election research is no longer theoretical but infrastructural—unstable, unevenly distributed, and WEIRD-centric data access regimes that systematically determine which democracies, platforms, and harms can be studied at all. Read together, these two papers reframe everything preceding them: the comparative claims this topic makes about Germany, Brazil, Nigeria, India, and beyond are only as strong as the fragile, contested infrastructures of observability that made them possible.