It’s no coincidence that the word “viral” is used to describe how ideas spread on social media. Researchers have long used infectious disease models to understand how information — and even misinformation — spreads quickly. However, researchers believe these models struggle to accurately describe how rumors spread.

"Most infectious disease models treat the spread of rumors as a process of passive acceptance of infection, thereby ignoring people's behavioral and psychological changes in the real world, as well as the impact of external events on the spread of rumors," explained Zheng Wenrong of Shandong Normal University.

To remedy this shortcoming, Zheng worked with Liu Fengming and Sun Yingping to create a rumor spreading model inspired by the chain reaction of nuclear fission (atom splitting). The process begins with the spontaneous splitting of a uranium nucleus into two smaller nuclei and several neutrons. If these neutrons are absorbed by other uranium nuclei, those uranium nuclei are more likely to split, triggering a fission chain reaction.

The two most common isotopes of uranium are uranium-238 and uranium-235. The former must absorb multiple neutrons to split, while the latter will split after absorbing only one neutron.

In the trio's model, a neutron passing through a piece of uranium is the rumor. Uranium-235 is a person who spreads rumors as soon as he receives them. Uranium-238 is a person who has to receive rumors many times before they can spread.

"When individuals encounter rumors, they will be affected by personal interests and decide whether to spread them or whether they need repeated exposure before spreading them," Zheng explained. "Based on different considerations of uranium fission thresholds, dividing individuals into different groups based on the impact of their own self-interest thresholds, and fully considering individual behaviors and differences, is more in line with the actual situation."

The researchers concluded that their model outperformed some infectious disease models in simulating the spread of real-life rumors. They also believe that rumors tend to spread slowly at first, which may mean the spread of misinformation can be countered.

Their model is described in AIPAdvances.