Greater exposure to false messages, but similar willingness to share them
With more connections among like-minded people in artificial networks, over 70% saw congenial false messages. In integrated networks, just over 35% did.
The key idea
In these networks, congenial false messages reached more people even though no difference in willingness to share them was detected.
The study, in plain language
A false news story can gain reach because more people choose to share it, or because the network structure puts it in front of more people. To distinguish these possibilities, the team built artificial networks with either more integrated connections or more concentrated connections among people with similar political views.
In more segregated networks, more participants were exposed to false messages aligned with their political views. Individual willingness to share them showed no detectable difference between the two network types. Even so, false messages made up a larger share of everything that was shared.
Simulations reproduced the pattern in other configurations that were tested. The clearest difference was not a new willingness to share, but the route that brought certain messages to more people.
What they found
- False messages aligned with each person’s political views reached more people in more segregated networks.
- Willingness to share those messages was similar in segregated and integrated networks.
- In segregated networks, false messages made up a larger share of everything shared.
How they studied it
The team combined a model, simulations, and an online experiment with 1,536 participants in the United States. They circulated 32 messages in 16 artificial networks of 96 people each, measuring how many people saw false messages and how many shared them.
What the study cannot show
- Simulations depend on their assumptions and configurations.
- Exposure does not explain why a person decides whether or not to share content.
Something to think about
When a questionable message reaches you, where do its connections come from? In these artificial networks, who was connected to whom affected reach, even though willingness to share was similar.
The original study
Network segregation and the propagation of misinformation
Source under CC BY 4.0. English adaptation and explanation by PaperFlow, with AI assistance. The text has been reorganized for a general audience; it is not the complete paper and does not imply endorsement by its authors.
One study alone does not represent a field's consensus. Educational content does not replace professional advice. Report a possible error.
