Overview: Psychology

many of the initial conclusions ...about the scope of fake news consumption, and its effects on our politics, were exaggerated or incorrect. Relatively few people consumed this form of content directly during the 2016 campaign, and even fewer did so before the 2018 election. Fake news consumption is concentrated among a narrow subset of Americans …

it will rank news organizations by credibility based on user feedback... trust rankings will emerge from surveys ... Google announced it would cancel a two-month-old experiment, called Knowledge Panel, that informed its users that a news article had been disputed by independent fact-checking organizations... Facebook learned the wrong lesson fro…

a major new Yale University study finds that fact-checking and then tagging inaccurate news stories on social media doesn’t work... “disputed” tags made participants just 3.7 percentage points more likely to correctly judge headlines as false... Trump supporters and adults under 26... could actually end up increasing the likelihood that users wil…

if publishers disagree with their labels they’ll have to take it up with the third-parties... won’t receive any payment from Facebook, but may get a traffic and branding boost from the debunk post links... Facebook is still working on better classifiers to automatically detect fake news, and preventing fake news from appearing as “Related Article…

Mantzarlis expects the partnership will result in a huge increase of applications to sign on to the IFCN’s code.... leads to a surge in genuine fact-checking projects, so much the better... Facebook decides which fact-checkers to include; the IFCN code is just the ‘minimum condition...

Users will be able to mark stories as fake...algorithm will look at whether a large number of people are reporting a particular article, whether or not the article is going viral, and whether the article has a high rate of shares... create an algorithm-vetted set of links that then goes on to a team of researchers within Facebook... links are sent…

A work in progress from an upcoming eponymous post. Another experiment with the enewsletter format: some initial thoughts on this seemingly intractable problem, with some of the source materials I’m studying.

Perfectly open communities always go sour. You need filters. Every functional community has them. And that’s where machine learning comes in... If you can detect trolls, you can protect the people they’re trolling by muting or putting a warning over the trolls’ posts... Twitter... already have a way of screening out porn. Why don’t they do the sam…

As Facebook attempted to capture the fast-moving energy of the news cycle from Twitter... it built a petri dish for confirmation bias... Here’s how... ‘Share’ Button ... encouraging people to share quickly and without much thought... “original sharing,” where people post their own photos, text updates... was declining..., content from celebritie…

Ironically, with the widening of (national) news choices that the Internet has spawned, we’re depending on fewer pipelines of news. It’s a narrowing of the filter funnel...t as troubling as the filter bubbles that used to occupy our concerns, but likely more potent. As those pipelines narrow, necessarily, the decision on what is news, and what is …