What the study found
Most users tend to conform to local normative environments, meaning they align their toxicity with the toxicity levels of the chats where they participate.
Why the authors say this matters
The authors conclude that these findings deepen understanding of toxic online behavior and highlight the importance of contextualized approaches to content moderation.
What the researchers tested
The researchers analyzed toxic user behavior across multiple Telegram chats and languages using six large-scale datasets with over 500 million messages collected between 2015 and 2024.
They introduced a methodological framework based on a conformity index to classify behavior as conformist, anti-conformist, or independent.
What worked and what didn't
The pattern of conformity was consistent across datasets and languages, showing a strong association between community norms and user behavior online.
Higher levels of user participation in chats were associated with a stronger tendency toward conformity with the surrounding social context.
The abstract does not describe any specific findings that failed or any null results.
What to keep in mind
The abstract does not describe detailed limitations beyond the scope of the datasets and platform studied.
The findings are based on Telegram messages and the authors' conformity framework, so the summary is limited to that context.
Key points
- Most users aligned their toxicity with the toxicity levels of the chats they joined.
- The study used six large-scale Telegram datasets with more than 500 million messages from 2015 to 2024.
- A conformity index was used to label behavior as conformist, anti-conformist, or independent.
- The conformity pattern was consistent across datasets and languages.
- Greater participation in chats was associated with stronger conformity to local social context.
Disclosure
- Research title:
- Telegram users generally align toxicity with local chat norms
- Authors:
- Lorenzo Alvisi, Victor Popa, Guglielmo Cola, Serena Tardelli, Maurizio Tesconi
- Institutions:
- IMT School for Advanced Studies Lucca, Institute of Informatics and Telematics, National Research Council, University of Pisa
- Publication date:
- 2026-04-24
- OpenAlex record:
- View
Get the weekly research newsletter
Stay current with peer-reviewed research without reading academic papers — one filtered digest, every Friday.