After studying 120 million re-tweets in a year’s time, a team from the Indiana U. School of Informatics and Computing found that whether a tweet goes viral doesn’t necessarily depend on its message or the user who posted it.

Instead, the study shows it has more to do with the fact that Twitter users have limited attention to devote to a massive amount of information transmitted daily on the Twitter network.

The 2012 study behind the information, “Competition Among Memes in a World with Limited Attention,” was conducted by third-year doctoral student Lilian Weng, along with informatics professors Alessandro Flammini, Alessandro Vespignani and Filippo Menczer.

The study, Weng said, is the first to scientifically show how social network users’ attention spans affect popularity of posts.

Weng said a tweet’s survival can be measured in the number of times the post is re-tweeted and does not necessarily depend on the post’s message.

She said two hashtags grouping tweets about singer Justin Bieber illustrate her point.

“#BieberFact and #Bieberthing — they both exist and are about the same object,” she said. “They try to represent the same thing, but one of them is extremely popular, and the other was re-tweeted less than 15 times.”

In fact, #BieberFact was re-tweeted 139,760 times during the course of the study, while #Bieberthing was re-tweeted only three times.

According to the report, factors like the tweet’s exposure to media and its relation to world events can affect the popularity and longevity of posts.

The reason for the discrepancy, Weng said, was because of the large number of tweets on Twitter competing for users’ attention.

Competition in the Twitter world works in a similar way to competition in nature, she said, where memes are like species fighting for limited space in users’
memories.

“Imagine in an ecosystem, you have various species, and they’re fighting with each other to get limited resources,” she said.

“In order to get resources to survive and reproduce, they have to compete with each other. You can think of attention as a limited resource in the system.”

In the grand scheme of things, Weng said, most tweets don’t go viral.

Instead, most stop circulating soon after they are posted.

Weng said the structure of the microblogging network, complete with hashtags and the ability to re-tweet, contributes to the distribution of tweet popularity.

Twitter users looking to craft a viral tweet might not find the secret in the new study, however.

Weng said the research looked at aggregate values and the larger context of tweet popularity as opposed to individual cases.

“In our paper, we studied the heterogeneity of meme popularity at a very aggregated, average level,” she said.

“But if you look at individual cases, it’s very hard to predict whether it will be successful. Sometimes it’s just luck.”