Of all the techniques that trolls use amplify misinformation, the one I find the most fascinating involves using 2 different personas to sow discord. Here’s how it works:
🔸You post something that gets their attention
🔸Persona 1 antagonizes you
🔸Persona 2 defends you

It’s like good cop/bad cop. The goal is to get you to engage— to break up the fight, so to speak. When that happens, your followers get notified and they see the misinformation the first troll (Persona 1) is putting out there

They do this over and over and over again across hundreds of accounts with a large or influential following. Once they get the flywheel spinning, Twitter’s algorithms take over

The end result is that you see the same misinformation over and over again. Even if it was obviously untrue to begin with, it starts to seem more credible via reinforcement

All that to say: I don’t always correct misinformation in replies b/c I don’t want to breathe life into the problem. Instead, I’ll just report it and move on

“A top State Department official said Thursday that Russia is behind “swarms of online, false personas” that sought to spread misinformation about coronavirus on social media sites, stressing the “entire ecosystem of Russian disinformation is at play.”

“We demonstrate how anti- and provaccination IRA messages were used strategically to enhance the credibility of manufactured IRA “personas” in ways that could benefit the organization’s attempts to […] increase the partisan polarization around vaccines.”

Here’s a visualization of the cross-talk between 9 personas used by the Internet Research Agency (IRA), a known Russian troll farm

Note how fake pro-**ump and BLM accounts were designed to expressed anti-vax sentiment at higher rates

src: ajph.aphapublications.org/doi/10.2105/AJ…

• • •

Missing some Tweet in this thread? You can try to force a refresh

Keep Current with Jorge A. Caballero, MD

Jorge A. Caballero, MD Profile picture

Stay in touch and get notified when new unrolls are available from this author!

Read all threads

This Thread may be Removed Anytime!


Twitter may remove this content at anytime! Save it as PDF for later use!

Try unrolling a thread yourself!

how to unroll video
  1. Follow @ThreadReaderApp to mention us!

  2. From a Twitter thread mention us with a keyword "unroll"
@threadreaderapp unroll

Practice here first or read more on our help page!

More from @DataDrivenMD

21 Jul
⚠️ This thread does not apply to #DeltaVariant. This was applicable between January and early June. It no longer applies because the assumptions made by the author are invalid.
🔸 hospitalization rates vary
🔸 vaccination rates vary

Get #vaccinated and #WearAMask
If you’re a verified account that amplified this, I urge you to spend be next few days catching up on the data from Scotland, England, Israel, and Australia. *Look at the actual data*

The headlines may be a little off, but they’re directionally accurate
The same can be said about a pre-print study on pediatric deaths due to COVID. If you amplified the Marty Makary post, spend some time reading through the critiques in this thread. The upshot is the same: *look at the actual data* before you amplify whatever aligns w/ your priors
Read 9 tweets
20 Jul
Here’s what worries me most about #DeltaVariant: folks like me, the ones willing to speak truth-to-power, are burning out

Politicians are making decisions based on poll numbers, and a lot of that has to do with public perception. We keep them honest, and influence policy.

Academics have a million things going on, they tend to be reactive. They’ll respond to specific prompts or comments. Whether they agree or disagree, we help keep them engaged up to speed on the data.

Since Biden took over, it’s been a struggle to steer politicians in the right direction. The honeymoon period made it harder to shift public sentiment, and the current White House is far better at leveraging access to help deflect bad press

Read 16 tweets
20 Jul
⚠️ J&J vaccine update: the upshot of the study is that, in the setting of #DeltaVariant, ”the data underscore the importance of surveillance for breakthrough infections that result in severe COVID-19”

primary source: biorxiv.org/content/10.110…
Looking through the plots in the pre-print paper, and I must say: I'm struggling to square their results with real-world experience with the J&J vaccine. Nevertheless, I do agree that it’s worth improving our surveillance for breakthrough infections.

biorxiv.org/content/10.110… Image
I disagree w/ the NYT headline that “one dose of J&J vaccine is ineffective against Delta, study suggests.” There is, however, enough to warrant improved surveillance testing for breakthrough cases across the board.*

*Just my opinion; I defer to @angie_rasmussen & @PeterHotez
Read 5 tweets
20 Jul
If you’re policing terminology (again) and wondering why the public has tuned you out, then look in the mirror. 280+ million adults aren’t the problem. You are

Please spare us a redux of the airborne debates

PS— #VaccinesSaveLives *and* #DeltaVariant requires that we #WearAMask
I’m *really* close to calling people out by name rn

If you want to reinforce your priors, go right ahead, but stop amplifying academic esoterica that is not grounded in *what we know is happening in the real-world*

This ⤵️👀 is textbook. #DeltaVariant is real life. Wake up. Image
Read 4 tweets
20 Jul
Knowing what I know about this (a lot)— this isn’t going to help patients, but it will help commercial insurers during contract negotiations

#priorities 🤷🏻‍♂️
Medical billing is based on a system of codes (like UPCs, but for medical services) — hospitals negotiate w/ insurers to determine how much each code is worth

Example: These are some of the codes used for COVID-19 testing at urgent care facilities by UnitedHealthcare
Note the language about when these codes can/cannot be used— this is typical: each insurer has specific guidelines like this

Read 11 tweets
19 Jul
To anyone who feels reassured that the hospitalization rate for fully vaccinated persons is 40% instead of 60%: get *fully* vaccinated and #WearAMask

To everyone else: get *fully* #vaccinated and #WearAMask
Dear people replying/DM’ing to explain why I’m wrong: #DeltaVariant ain’t going to argue w/ you on Twitter, and neither am I

Yes, I’m aware of nuance in the statement. No, it’s not substantive (see CDC data below)

Get *fully* #vaccinated and #WearAMask
3/ We know from a Scottish study that individuals who are *fully* vaccinated do end up in the hospital at significantly higher rates *due to #DeltaVariant* as compared to Alpha

screenshot: ed.ac.uk/news/2021/delt…
primary source: thelancet.com/journals/lance…
Read 5 tweets

Did Thread Reader help you today?

Support us! We are indie developers!

This site is made by just two indie developers on a laptop doing marketing, support and development! Read more about the story.

Become a Premium Member ($3/month or $30/year) and get exclusive features!

Become Premium

Too expensive? Make a small donation by buying us coffee ($5) or help with server cost ($10)

Donate via Paypal Become our Patreon

Thank you for your support!

Follow Us on Twitter!