【Uschi Karnat】

Over the weekend,Uschi Karnat Twitter told 677,775 people via email that they interacted with Russian propaganda during the 2016 presidential election. But according to Kate Starbird, an assistant professor at the University of Washington, that's far from enough.

In a Twitter thread on Saturday, Starbird shared her recent research into social networks and the "echo chambers" they create. Her team successfully identified that polarized tweets are one way people, including Russian trolls, can effectively manipulate conversations.

In an academic paper titled "Examining Trolls and Polarization with a Retweet Network," the researchers used previous research on Twitter discourse about the #BlackLivesMatter movement and police-related shootings in 2016 to analyze how Russian troll accounts played a part.

Starbird highlighted a figure from the paper that shows retweets of troll accounts:

This Tweet is currently unavailable. It might be loading or has been removed.

Using Twitter's publicly available data, the researchers collected related tweets from Dec. 31, 2015 to Oct. 5, 2016. They searched for keywords “shooting," “shooter," “gun shot," and “gun man" and then narrowed that set of nearly 59 million tweets to 248,719 that also contained either “#BlackLivesMatter,” “#BlueLivesMatter,” or “#AllLivesMatter."

After Twitter released a list of accounts that were tied to Russian troll factories, the researchers cross-referenced and incorporated them into the data set.

"We observe that retweets of troll accounts are largely contained within each cluster, suggesting that the RU-IRA trolls participated in distinct information flow networks," the paper reads. "We also note that retweets of troll accounts appear more pervasive in the left-leaning cluster than on the right-leaning cluster, suggesting greater infiltration with the left-leaning side of the conversation."

As the research showed, the troll accounts retweet each other.

This Tweet is currently unavailable. It might be loading or has been removed.

Twitter shared in its email to users that it was committed to identifying and taking action against "coordinated malicious activity." But as Starbird and her team of researchers at the University of Washington showed, most of the damage has already been done.

This Tweet is currently unavailable. It might be loading or has been removed.

On Monday, Facebook executives and outside experts released their thoughts on the negative effects social media has had on democracy.

"From the Arab Spring to robust elections around the globe, social media seemed like a positive," wrote Katie Harbath,a global politics and government outreach director at Facebook. "The last US presidential campaign changed that, with foreign interference that Facebook should have been quicker to identify to the rise of 'fake news' and echo chambers."

Twitter's Public Policy blog has kept users up to date — well, belatedly, by U.S. lawmakers' standards — on the investigation into the use of its platform during 2016 election. But the company hasn't dived quite as deep into the psychological consequences, at least not publicly. Twitter declined to comment on the University of Washington study.

For now, Starbird and other researchers can continue investigating how our digital conversations take shape — and they'll most likely use the same platforms they're studying to disseminate their insights.

(Editor: {typename type="name"/})