-
A study by Newtral and the Mind, Brain and Behavior Research Center (CIMCYC) at the University of Granada reveals that the majority of posts regarding the 2025 Torre-Pacheco incidents utilized specific psychological mechanisms previously associated with misinformation and the spread of prejudiced content.
-
The study identifies a structural "rhetorical package"—group identity, distortion, unjustified causal inferences and emotional charge—that is repeated in an almost inseparable manner across X/Twitter, TikTok and Telegram.
- Incitement to action (whether aggressive or otherwise) never appears in isolation: in 99% to 100% of cases, it is accompanied by derogatory assessments or insults.
MADRID / GRANADA – Why do certain hoaxes spread like wildfire while others go unnoticed? A multidisciplinary team made up of fact-checkers from Newtral and researchers from the Mind, Brain and Behavior Research Center (CIMCYC) at the University of Granada has explored the psychological processes involved in viral social media events.
After analyzing nearly 290,000 posts related to the July 2025 riots in Torre-Pacheco (Murcia), the study reveals the existence of a "rhetorical package" featuring a combination of psychological mechanisms that previous research has linked to the spread of misinformation and prejudice.
The discovery: A structural "rhetorical package"
The core finding of the research is that the analyzed posts exhibit specific psychological mechanisms that consistently appear together in these interactions. There is a coordinated combination of four psychological mechanisms that appear together between 86% and 99% of the time, regardless of the platform.
This formula, which has been termed a "rhetorical package", combines the following mechanisms:
-
Group identity: Language that emphasizes "us" versus "them".
-
Distorted information: Biased presentation of facts to fit a narrative.
-
Unjustified causal inferences: Simplistic attribution of cause (a "scapegoat") or conspiratorial framing.
-
Use of emotions: Language loaded with emotion, primarily anger, fear or disgust.
This pattern is structural; it was detected with the same consistency in texts on X, in TikTok video descriptions and in Telegram channels, suggesting that the combination of these mechanisms is systematic rather than anecdotal and is, to a certain extent, independent of the format.
The bridge between words and action
The research provides insight into the patterns that accompany aggressiveness in digital environments, revealing that the transition from words to action (sometimes violent) may follow a systematic rather than a random process. One of the most striking findings is that posts expressing behavioral tendencies or calls to action appear very frequently alongside insults or derogatory language, directed either at other social media users or specific groups. The data confirm an almost perfect correlation—between 99% and 100% across all analyzed platforms—between calls to violent action and this type of derogatory language.
In practice, this means that incitement to action (often aggressive) practically never occurs in isolation; instead, it is accompanied by a pejorative representation of the recipients (often other users or social groups).
This phenomenon is further amplified by the significant power of appeals to authority within the mechanics of virality. The study shows that content invoking arguments of credibility or authority is not only likely to be more persuasive but also spreads with much greater momentum than posts without this element: on X/Twitter, these posts reach up to five times higher engagement than the platform's baseline.
Finally, the research debunks the idea that this discourse is a spontaneous, majority-driven phenomenon, showing an extreme concentration of posts containing these analyzed mechanisms in the hands of a very small number of accounts. Far from being a massive social outburst, just 12.4% of X/Twitter users posting about Torre-Pacheco during this period were responsible for generating half of the posts containing derogatory interactions between users or toward other groups.
This centralization is even more pronounced on Telegram, where just 14 specific channels were sufficient to produce 50% of all toxic content detected during the Torre-Pacheco crisis.
An unprecedented alliance: Science and fact-checking
This project, funded by the Journalism Science Alliance, marks a milestone by combining journalistic expertise with scientific rigor. While Newtral provided the technical data extraction capabilities and field knowledge, CIMCYC at the University of Granada provided the theoretical framework to identify cognitive biases. Together, they developed a robust, validated AI model that aims to replicate the expert criteria of scientists.
Access the study here: https://www.newtral.es/paquete-retorico-odio-desinformacion-torre-pacheco/
Contact:
Marilín Gonzalo Airut, periodista (Newtral) @email
Fernando Blanco Bregón, PhD (Departamento de Psicología Social, Universidad de Granada) @email
Team
Fernando Blanco Bregón - CIMCYC
Felisa González - CIMCYC
Rocío Martínez - CIMCYC
Jose César Perales - CIMCYC
Marilín Gonzalo - Newtral
Miguel Ros - Newtral
About Newtral
Newtral is a journalism and technology company working to combat misinformation through fact-checking, technological innovation and media literacy. As an organization certified by the International Fact-Checking Network (IFCN) and the European Fact-Checking Standards Network (EFCSN), it drives national and international projects to strengthen access to rigorous, high-quality information.
About CIMCYC
The Center for Brain, Mind and Behavior Research (CIMCYC) at the University of Granada is a multidisciplinary research center dedicated to studying human behavior, cognition and the brain. It promotes research excellence and knowledge transfer in fields such as psychology, neuroscience and health, collaborating on high-impact national and international projects.