A special set of editorials published in today’s edition of the journal Science argues that social media in its current form can be fundamentally broken for the purposes of presenting and disseminating facts and reasons. Algorithms are running the show now, they argue, and systems priorities are unfortunately lagging behind.
In an incisive (and free-to-read) opinion piece by Dominique Brossard and Dietram Scheufele of the University of Wisconsin-Madison, the basic disconnect between what scientists need and what social media platforms provide is convincingly presented.
“The rules of scientific discourse and the systematic, objective, and transparent assessment of evidence are fundamentally at odds with the reality of debates in most online spaces,” they write. “It is debatable whether social media platforms designed to monetize outrage and disagreement among users are the most productive channel for convincing the skeptical public that the established science on climate change or vaccines is not up for debate.”
The most elementary feature of social media that reduces the communication effect of scientists is pervasive classification and recommendation mechanisms. This produces what Brossard and Scheufele call “homophilic self-classification” – those who receive this content are those who are already familiar with it. In other words, they are preaching to the choir.
“The same profit-driven algorithmic tools that bring curious, science-friendly followers to scientists’ Twitter feeds and YouTube channels will increasingly disconnect scientists from the audience they most urgently need to connect with.” , they write. And there is no obvious solution: “The cause is a tectonic shift in the balance of power in scientific information ecologies. Social media platforms and their underlying algorithms are designed to overcome the scientific audience’s ability to sift through rapidly growing streams of information and capitalize on their emotional and cognitive weaknesses in doing so. No one should be surprised when that happens.”
“But it’s a good way for Facebook to make money,” said Holden Thorp, editor-in-chief of the Science family of journals.
Thorp, who also wrote an editorial on the topictold me that there are at least two distinct problems with the way scientists and social media interact these days.
“One is that, especially with Twitter, scientists like to use it to openly debate and air ideas, support them or tear them down — the things they used to do on a blackboard or at a conference,” he said. . “It was happening before the pandemic, but now it has become an important way for this kind of exchange to happen. The problem with that, of course, is that there is now a permanent and lasting record of it. And some of the hypotheses that are made and turned out to be wrong, overturned in the normal course of science, are handpicked by people who are trying to undermine what we are doing.”
“The second is the naivety towards algorithms, especially those on Facebook, which place a high value on disagreement and informal posts that spread disagreement. You know, ‘my uncle wore a mask to church and got COVID anyway’ – this is going to trump official information every time,” he continued.
As Brossard and Scheufele point out, the combination of these things puts scientists “at a distinct disadvantage…
Unfortunately, there isn’t much anyone can do on the science side. Arguably, the more they participate in the system, the more they reinforce the silos around them. No one is arguing that we should give up – but we really need to recognize that the problem isn’t just a matter of the scientific community being less effective communicators on social media than disinformation sellers.
Thorp also acknowledged that this is just the latest phase of growing antifactual tendencies and politicization that stretches back decades.
“I think people tend to get a little more emotional about it without acknowledging that it’s a very simple thing: political parties are not going to take the same position – and when one of those positions is scientifically rigorous, the other is going to be against science.” , he explained. It is true that the Democratic Party is more often on the side of science, but it has also been on the other side with GMOs and nuclear energy, she pointed out. The important thing is not who is for what, but that the two parties define themselves by the opposition.
“This is a political party coming to the conclusion that it was politically more useful to be against science than to be for it,” he said. “So that’s another thing that scientists are naive about, saying ‘we’re not getting our message across!’ But you are facing this political machine that now has the power of Facebook behind it.”
Brossard and Scheufele draw a final parallel in Garry Kasparov’s defeat by Deep Blue—afterward, no one asked for special training to outperform supercomputers, and no one blamed Kasparov for not playing well enough. After the shock wore off, it became clear to everyone that we had turned the corner not just in chess, but in the possibilities of computation and algorithms. (Kasparov’s own views also evolved, as he told me some time ago.)
“The same understanding is now here for scientists,” they write. “It’s a new era to inform public debates with facts and evidence, and some realities have changed forever.”