Adrift in a Sea of Similar Science Stories
For six months in 2025, I spent my days working from a small office on the bottom floor at the Heidelberg Institute for Theoretical Studies. My window was almost roof to ceiling, looked out onto the Institute's amazing grounds, and I could open it and step straight into a garden bed, if I so desired. I didn't.
I'd been selected as the 13th Journalist in Residence, and I'd come to Heidelberg to work on tools that (I hope) will help science journalists report on scientific misconduct and research integrity breaches (more on that in the future). In between that work, there was a lot of time to think.
When you get six months to percolate on your profession and your contributions to it, you spend a lot of time contemplating how its cogs turn, how it's operating on a day-to-day basis, where it falls down. And once you're off the hamster wheel, you see clearly how squeaky its axle is.
What I kept noticing was the abundance of press release science journalism. The stuff where journalists grab a new piece of research, from a major journal, and drop some 800-odd words about it. Often, these reports did not live up to the goals of science journalism: reporting without fear or favour, holding science to account, and empowering the public to make decisions based on evidence. The pieces had a tendency to misinform audiences, not intentionally, but through acts of omission regarding uncertainty and robustness, and even how studies were funded or supported.
The types of stories that were frustrating me had similar features: They focused on a single study, and did little more than take a uni press release, slap in some quotes, and occasionally add a few lines from the paper itself. Often, it seemed the journalists didn't even read the paper they were reporting on. They leant into hype and exaggeration. And it was plain to see this kind of work was susceptible to AI replacement. [In fact, it has already come for this kind of work].
With this ever-fastening cataclysm in mind, I wrote "What The **** Is The Point Of Us Then?" towards the end of my tenure in southwest Germany, as I was about to emerge from my science journalism chrysalis. I'd been particularly annoyed that a study about Chronic Fatigue Syndrome had received uncritical coverage in major publications with all sorts of 'breakthrough' and 'groundbreaking' adjectives thrown around. [the study has, today, a pretty extensive thread on PubPeer detailing some issues with its data collection and analysis]
Jackson Ryan
The thesis of my piece was that the vast majority of today's science journalism can not survive if we accept this standard of coverage, particularly from mainstream outlets, the places where this kind of work is (at least scarcely) resourced. It's a threat to the survival of this profession – a profession I treasure, a profession I have made efforts to strengthen, a profession I am constantly yapping about.
So it was with delight that, in May this year, I saw Chris Mooney, a US climate journalist, explore the potential problem AI poses to this type of reporting, too. He did so in an eloquent, cautiously optimistic, less defeated way (the kind of way befitting a Pulitzer Prize winner, I guess!). At his Substack, he wrote a piece headlined:
"The scientific "study story" has a questionable future. That may be a good thing"
It's a piece that helps pull apart some of the challenges science journalism faces, and the single study story is its reference point. Mooney spoke to several long-term science journalists, explained the amount of science being produced and how it doesn't reflect the coverage you typically see because the field is biased towards covering particular journals' outputs and studies. He also added a section on the importance of single study stories to young, early career journalists, trying to get their start as a reason the study story might stick around for a while yet.
Like Mooney, I've written a lot of single study stories in my time. Often, the reason to write them was to just have some content for any given day; bound by the newsroom's insatiable desire for NEW THINGS PEOPLE CLICK ON. Science is very good at providing that – the biggest journals in the world, such as Science and Nature are dropping press releases about new discoveries on basically every day of the week. Then there's the health journals like Lancet and NEJM that can make a splash, too. As do Cell, PNAS and the PLOS papers. If you're a researcher and your paper is in one of these journals, it's likely your institute is also putting out a press release.
Single study stories routinely see a large number of readers. The biggest single study stories on my CV did millions and millions of views. But they're sort of... hollow wins in a newsroom, in hindsight. I look back on mine and think they're rather forgettable, built for algorithmic success. The better ones are somewhat educational or surprising, at least.
In a digital newsroom, your brain gets hijacked by the KPIs, the pressures and the peaks that come with this style of reporting, including those millions of readers suggesting you're doing something right. There's one story I wrote that explicates this, about a Google Doodle for christsakessss, that did like 10 million views or something ridiculous. Those views meant little and had little impact. I was barely doing journalism in that story; I was just filling space. The incentives were to hit big numbers, not to make the best decisions for our readers.
Mooney's followed up his excellent May post in a recent piece, again on his Substack, titled
"How ‘mega-science’ helps explain the plight of science communicators right now"
In this piece, Mooney asks "How do we communicate about – or as journalists, cover – 3.5 million scientific studies per year?" and (again, in a really well written piece) goes a bit deeper on the rise and rise of science publications alongside the precipitous fall of journalist jobs. Scientific papers have increased from around 1.2 million papers a year some 30 years ago, to around 3.5 million papers per year today. That, he suggests, could be a problem as he writes:
"...in a world of mega-science, there will likely be mega-problems in science communication. Some haystack needles of science will make their way into the media in a very selective, and sometimes misrepresentative, way. But the majority of the information won’t."
It's a valid point, and in this newsletter, I wanted to return to the opening question Mooney had: "How do we communicate about – or as journalists, cover – 3.5 million scientific studies per year?"
I'm interested in the incentive structures that are influencing how science moves from bench to broadsheet. These incentives remain the clearest obstruction to covering science effectively and for the public interest (no matter how much of it is produced). Mooney's question may not be a literal one, but it's a useful one to frame the situation we're in.
[Note: I very much am approaching this from a written, digital journalism point of view, with both experience as a newsroom journalist and a freelancer.]
There are two ways I want to answer, or maybe one answer in two parts: We don't (derogatory) and we don't (complimentary).