Researchers from the University of Vermont and the Santa Fe Institute have developed a mathematical model that explains why some posts, videos, or memes on the internet go viral while others are quickly forgotten. The study, published in Physical Review Letters (PRL) on August 23, 2025, shows that the key to virality lies in the ability of content to evolve as it spreads.

How Does Virality Work?
Traditional models of information spread compared it to a branching tree: one person shares an idea with two others, those two share it with new people, and so on. In these models, the idea itself was assumed to remain unchanged. However, the new model accounts for the fact that information online changes: a joke may take on an unexpected twist, a rumor may gain a convincing detail, and a news story may acquire an emotional tone that makes it more appealing.

“We were inspired by forest fires,” explains Sid Redner, professor at the Santa Fe Institute. “A fire burns more intensely in a dense forest and weakens in open clearings. The same happens with information: in some conditions, it strengthens; in others, it dies out.”

The model shows that at each “step” of spreading, content may either lose its appeal or become stronger. If an idea grows less interesting, the chain of transmission breaks. But even a small improvement — for example, a successful edit or the addition of emotion — sharply increases its chances of spreading like an avalanche.

What Makes Content Viral?
According to the study, virality does not require a special “critical state” of the system, as previously thought. It is enough for the “quality” of content to be able to change during transmission. This explains why most posts online quickly lose popularity, but some unexpectedly go viral.

“Previously it was believed that this required a special critical state of the system. But our work shows that it is enough for the ‘quality’ of an idea to change in the course of its spread,” noted lead author Laurent Hébert-Dufresne of the University of Vermont.

The model produces a statistical picture similar to real life: most publications disappear within hours or days, but rare posts that receive a “successful boost” spread like wildfire.

Applications of the Model
The new model has a wide range of applications:

  • Understanding social processes: It explains how beliefs form, misinformation spreads, or how “social contagions” — mass trends and hypes — emerge.
  • Marketing and advertising: Companies can better predict which content will go viral and optimize their campaigns.
  • Combating disinformation: The model can help identify mechanisms behind the spread of fake news and develop strategies to contain it.

The researchers plan to test the model on real data from social networks such as X and TikTok to confirm its predictions. This may involve analyzing viral videos, memes, or news posts to understand what kinds of changes make content more contagious.

Why Does It Matter?
Understanding the mechanisms of virality changes the way we study information flows online. In an era when social networks shape public opinion, politics, and culture, the ability to predict which content will become popular is a key tool. The model also emphasizes that the internet is not a static system, but a dynamic environment where ideas are constantly evolving.

In Brief…
Mathematicians from the University of Vermont and the Santa Fe Institute have developed a model explaining online virality. Unlike classical approaches, it accounts for the evolution of information: jokes, rumors, or posts can strengthen or weaken as they spread. The study, published in Physical Review Letters, shows that even small improvements to content sharply increase its chances of viral spread. The model opens new opportunities for analyzing social trends, marketing, and fighting disinformation, with testing on real-world data being the next step.