Science Fictions

This is a free preview of a paid episode. To hear more, visit sciencefictionspod.substack.comAnd now… following last week’s episode on ECT, here’s part two of our double episode on depression treatments. This time we’re looking at antidepressants. You’ll be delighted to hear that we immediately encounter our favourite thing—dueling meta-analyses.To hear the whole episode and read the show notes, become a paying subscriber at www.sciencefictionspod.com/subscribe.

Open up some scientific papers, and you’ll hear electroconvulsive therapy described as the most effective treatment for depression (especially very severe depression). But open up others, and you’ll see it described as completely useless—and a sad indictment on a medical establishment who’ve completely failed to provide proper evidence on it. Not only that, but they’ve exposed patients to serious side effects, like memory loss, for no good reason.Who’s right? In this episode, we look into the most controversial psychiatric treatment since lobotomy.NEXT WEEK: we’ll follow this with an episode on another controversial psychiatric treament: antidepressants.On this week’s episode we discussed the article “The Perks of Being a Mole Rat”, from our sponsor, Works in Progress magazine. As ever, we’re very grateful for their support. You can find many more excellent articles at worksinprogress.co.Show notes* 1937 article by Egas Moniz, lobotomy Nobel Prize-winner* Weird 1998 article defending him on the Nobel Prize website* Megan McArdle on Walter Freeman* The ECT scene in One Flew Over the Cuckoo’s Nest* 2024 article discussing the possible mechanisms of ECT’s effect* 2010 review about sham ECT studies* 2019 review of each individual sham ECT study and the meta-analyses that include them* 2022 response to the review* Response to the response* Contemporary news article about the controversy* 2021 article in defense of ECT* The parachute RCT* 2010 meta-analysis on cognitive effects* 2025 meta-analysis on autobiographical memory lossCreditsThe Science Fictions podcast is produced by Julian Mayers at Yada Yada Productions. This is a public episode. If you'd like to discuss this with other subscribers or get access to bonus episodes, visit sciencefictionspod.substack.com/subscribe

This episode is dedicated to Justin Eldridge.We like to think that, in often hamfisted ways, we’re applying critical thinking on this show. But what even is “critical thinking”? Can you measure it? Can you teach it to kids—or for that matter, to anyone? Can teaching critical thinking help people defend themselves against misinformation and disinformation? It would be very ironic if “critical thinking” had become a buzzword in the world of education—a buzzword that people used, er, uncritically…The Science Fictions podcast is brought to you by the marvellous Works in Progress magazine. The article on Swiss vs. Japanese watches that we mention in the episode can be found at this link, and all the other Works in Progress articles can be found at worksinprogress.co.Show notes* NY Times article on schools teaching critical thinking* UK Government Curriculum Review from 2025* Daisy Christodoulou on teaching students to spot misinformation* Daniel Willingham’s 2007 article on critical thinking* His book Why Don’t Students Like School?* The Pacific Northwest Tree Octopus study* The tree octopus website* 2017 Dutch replication study* Two bigger studies in 2016 and 2019* 2015 meta-analysis in Review of Educational Research* Woodworth and Thorndike (1901) - a psychological classic* Herbert Simon on “problem isomorphs”* The Stanford Civic Online Reasoning programmeCreditsWe’re very grateful to Daisy Christodoulou for talking to us for this episode. Any mistakes are, of course, our own. The Science Fictions podcast is produced by Julian Mayers at Yada Yada Productions. This is a public episode. If you'd like to discuss this with other subscribers or get access to bonus episodes, visit sciencefictionspod.substack.com/subscribe

Here’s another episode that revists a topic we’ve covered before. A while back, we did an episode on the downsides of cannabis (for example, the risk of psychosis). But of course, a lot of people claim there are medical benefits, too! Not least among them is Donald Trump, who recently re-scheduled cannabis so that it can be studied more for medical purposes.That research is sorely needed. In this episode, we discuss the very uncertain state of our knowledge on medical marijuana.The Science Fictions podcast is brought to you by Works in Progress magazine, where you can find Samuel Hughes’s excellent new article on “Urban expansion in the age of liberalism”, as discussed in today’s episode. That’s at worksinprogress.co—which is packed full of other great articles on science, tech, and human progress.Show notes* “Trump expands access to cannabis” (December 2025)* Info from the DEA on drug schedules* The new JAMA review on medical marijuana* Reason article on the 2011 RAND study; Retraction Watch article on the same* Studies on the crime impacts of medical marijuana (increase; neutral; decrease)* Useful 2021 review of the wider societal effects of medical marijuana legalisation* 2020 Arizona “natural experiment” study* Washington Post article on the poor resemblance of “real” marijuana compared to what’s allowed for research studies* Survey on the use of cannabis for medical purposesCreditsThe Science Fictions podcast is produced by Julian Mayers at Yada Yada Productions. This is a public episode. If you'd like to discuss this with other subscribers or get access to bonus episodes, visit sciencefictionspod.substack.com/subscribe

This is a free preview of a paid episode. To hear more, visit sciencefictionspod.substack.comIt’s rare that we return to a topic, but it’s also nice to have been right. In 2024 we did an episode on microplastics, and cast a lot of doubt on some of the more outrageous claims about them filling up your brain, your arteries, and (for the fellas) your testicles. Since then a lot more flaws in the science have been found—and at least one of them is utterly devastaing. Become a paid subscriber to listen to the full episode and read the show notes.

Here’s a cheery one for our first episode of the year. Guess what happens when you give several sets of scientists the same dataset and ask them to answer the same question? Well, they all find the same results, right? Right!?Sadly not. This “Many Analysts” problem has been analysed and debated in multiple different scientific fields and across several papers. We cover them in this episode. What does it tell us about the objectivity of science if different teams draw different conclusions from the exact same data?The Science Fictions podcast is brought to you by Works in Progress magazine. Their excellent new article on how we’re living in “the golden age of vaccine development”, as discussed on the show, can be found (along with the rest of their articles on science, history, and technology), at worksinprogress.co. We’re very grateful that they support the podcast.Show notes* 2015 Nature commentary article on “crowdsourced research” (on racism in football)* And the full 2018 writeup titled “Many Analysts, One Data Set”* Gelman and Loken on the “Garden of Forking Paths”* 2020 many-analysts neuroscience (fMRI) paper* And the plan for the similar study on EEG* 2022 PNAS many-analysts paper on the “hidden universe of uncertainty”* 2026 critique on ideological bias from George Borjas* 2023 critique on effect sizes vs. statistical significance* 2025 ecology & evolution many-analysts paper on blue tits and eucalyptus* 2025 economics many-analysts paper with results on data cleaning* 2024 PNAS critique of many-analysts research* Julia Rohrer’s critique of multiverse analysisCreditsThe Science Fictions podcast is produced by Julian Mayers at Yada Yada Productions. This is a public episode. If you'd like to discuss this with other subscribers or get access to bonus episodes, visit sciencefictionspod.substack.com/subscribe

We’ve covered a lot of bad science stories over the year. Here are a few more. But in the optimistic spirit of the “holiday season”, the last one has a happy ending. Thanks for listening—especially if you’re a subscriber! See you in 2026.Stuart & TomShow notes* A surge of low-quality AI papers on public datasets* A surge of low-quality AI letters to the editor* Retraction Watch story on the Dana Farber scandal* NY Times story on the papers being retracted or corrected* The settlement in the case CreditsThe Science Fictions podcast is produced by Julian Mayers at Yada Yada Productions. This is a public episode. If you'd like to discuss this with other subscribers or get access to bonus episodes, visit sciencefictionspod.substack.com/subscribe

STOP PRESS: a beloved 20th Century populariser of psychology who wrote massively successful books has been shown to be full of crap. Actually… don’t stop press. Just put it on the pile with all the others.This time it’s Oliver Sacks, the neurologist who wrote The Man Who Mistook his Wife for a Hat, Awakenings, and many other books. An article in The New Yorker has shown that a lot of his case studies were, well… let’s say they’re not what they seem. In this episode we discuss the new article and Oliver Sacks’s career more generally, and ask: should we have known?The Science Fictions podcast is brought to you by Works in Progress magazine. The article we discussed on today’s show is about the tragically low South Korean birth rate, and why it got that way. Find that, and so many more articles about human progress, science, and technology, at worksinprogress.co.Show notes* Rachel Aviv’s December 2025 New Yorker article on Oliver Sacks* Journal of Autism and Developmental Disorders letter about “questionable aspects” of the autistic savant twins story, by Makoto Yamaguchi * Follow-up article by the same author* Response letter by Allan Snyder* Medical Humanities article on 10 years since Sacks’s death* Paul McHugh’s 1995 bad review of Sacks’s work* Science isn’t storytellingCreditsThe Science Fictions podcast is produced by Julian Mayers at Yada Yada Productions. This is a public episode. If you'd like to discuss this with other subscribers or get access to bonus episodes, visit sciencefictionspod.substack.com/subscribe

This is a free preview of a paid episode. To hear more, visit sciencefictionspod.substack.comDoes the evidence support the use of hormone replacement therapy (HRT)? Depends on when you asked the question. At one point the consensus was “yes”; more recently it reversed. But should it have?It also depends on what symptoms you’re talking about. Is HRT just all about hot flushes, or can it also treat mood and cognitive problems too? In this paid-only episode, we look at the evidence.To listen to the full episode and read the show notes, please become a paid subscriber to the Science Fictions podcast.

Hello everyone! We weren’t able to record a podcast this week, because 1) Stuart was busy and 2) it’s Tom’s birthday. So by way of apology we’re re-releasing this one about some drama last year between Jon Haidt, sworn enemy of smartphones, and some guys who like meta-analyses. Hope you enjoy it!A while back, The Studies Show covered the question of whether smartphones and social media cause mental health problems. Amazingly, that podcast didn’t settle the issue, and the debate has continued—and continued rather acrimoniously.Psychologists—most notably Jonathan Haidt—are currently laying into each other, analysing, re-analysing, and meta-analysing datasets to try and work out whether “it’s the phones”. In this paid-only episode of The Studies Show, Tom and Stuart explain the story so far, and in the process get very disappointed by their heroes.If you want to hear the whole episode and read the show notes, it’s easy to become a paid subscriber at thestudiesshowpod.com.Show notes* The summary of Jonathan Haidt’s upcoming book, Life After Babel* The Google Doc on social media effects maintained by Haidt, Twenge, and Rausch* Christopher Ferguson’s meta-analysis of causal social media effects studies* Very useful online calculator to interpret effect sizes* Study on the (non-)relation between reported and measured phone use* Haidt & Rausch’s first article criticising the Ferguson meta-analysis and re-calculating the effects* Anne Scheel’s critical tweet* Matt Jané’s first article responding to Haidt & Rausch* Haidt & Rausch respond to Jané (and criticise Ferguson again)* Jané responds to Haidt & Rausch, again* Haidt & Rausch’s second (or is it third?) article criticising the Ferguson meta-analysis (this is the one where they note the more basic errors)* Article by Mike Males making the point that, whoever is right, the effects are all very smallCredits* The Studies Show is produced by Julian Mayers at Yada Yada Productions. We’re very grateful to Malte Elson, Pete Etchells, and Matt Jané for talking to us for this episode—but any errors are our own. This is a public episode. If you'd like to discuss this with other subscribers or get access to bonus episodes, visit sciencefictionspod.substack.com/subscribe

Everyone has read Entangled Life, the wonderfully-written book about fungi that took the world by storm about 5 years ago. Among many other things, it popularised the “wood wide web”—the idea that trees can communicate with one another through networks of fungi at their roots.But is the wood wide web real? It turns out scientists have some major questions. We air them on this episode.And just to be completely clear, there are no personal vendettas here! Everyone recording this podcast is 100% free of “beefs” of all kinds. Even the co-host who was beaten in a book contest by the aforementioned mushroom book.The Science Fictions podcast is brought to you by Works in Progress magazine. Their most recent article is about the wonderful invention (and history) of the dishwasher, one of several incredible labour-saving devices that have made so many lives just a bit less dull. Read this, and so many more stories about human progress, at worksinprogress.co.Show notes* Entangled Life by Merlin Sheldrake* Winner of the 2021 Royal Society book award* Rupert Sheldrake and the concept of “morphic resonance”* Suzanne Simard’s TED talk about “how trees talk to each other”* Her 1997 paper on “net transfer of carbon”* 2023 paper by Karst et al.: “Positive citation bias and overinterpreted results lead to misinformation on common mycorrhizal networks in forests”* Nature piece following the 2023 paper* 2015 paper on “stress signaling” via fungal networks* 2023 paper on tree proximity* Simard’s response to Karst et al.CreditsThe Science Fictions podcast is produced by Julian Mayers at Yada Yada Productions. This is a public episode. If you'd like to discuss this with other subscribers or get access to bonus episodes, visit sciencefictionspod.substack.com/subscribe

It has happened again. A new paper, based on a tranche of unsealed historical documents, casts serious doubt on a piece of social psychology research from the mid-20th Century. Shocker!This time it’s about some of the fundamental inspirations for the idea of cognitive dissonance—the idea that holding contradictory views in one’s head creates discomfort and a need to change one of the beliefs. So what does the new historical research say? What about all the studies that claim to find evidence for cognitive dissonance—surely the whole thing isn’t a load of nonsense? Listen to this episode to find out.The Science Fictions podcast is brought to you by Works in Progress magazine. They’ve recently been publishing a whole host of podcasts, including the one we mentioned this week, on “the economics of the baby bust” (that’s the opposite of a baby boom, by the way). You can find it and many other podcasts at podcast.worksinprogress.co. Show notes* The new paper, “Debunking When Prophecy Fails”* And the related paper “Failed Prophecies are Fatal”* The lobotomy article in the Washington Post* Scott Alexander on using facts to persuade* Dan Engber on the same* Matti Heino on the original Festinger & Carlsmith (1959) paper* The paper itself* The GRIM test (with an online tool to do it yourself)* 2024 multi-lab replication attempt on cognitive dissonance* 1983 study that was replicatedCreditsThe Science Fictions podcast is produced by Julian Mayers at Yada Yada Productions. This is a public episode. If you'd like to discuss this with other subscribers or get access to bonus episodes, visit sciencefictionspod.substack.com/subscribe

This is a free preview of a paid episode. To hear more, visit sciencefictionspod.substack.comCreatine is the supplement of the moment, but both of us had vaguely heard that this one might actually not be total garbage.On the other hand: there are a lot of surprising claims made about it! If proponents are to be believed, it doesn’t just boost muscle mass – it reduces depression, prevents cancer, and improves your cognitive function.How much of this should we believe, and how much is it just a big load of crap? We thought we would take a look.

Around these parts, we have a tradition to do a paranormal episode at Halloween. We’ve done psychic powers, ghosts, and now… reincarnation. What are we to make of the stories—sometimes told in NYT-bestselling books—of children who appear to remember details of their past lives? What about the many peer-reviewed scientific papers that claim that something supernatural is going on here? In this EXTRA-SPOOKY episode, we find out.🎃 Happy Halloween! 🎃The Science Fictions podcast is brought to you by WoooOOOOOooooOOOOorks in ProoooooOOOOOooooooOOOOOooogress magazine, which is bursting with historical stories, policy ideas, and well-written scientific explanations, all focused on the topic of progress. You can find all of Works in Progress—essays, shorter pieces, podcasts, and even the chance to subscribe to the print edition—at worksinprogress.co.Show notes* The University of Virginia’s Department of Perceptual Studies* And the University of Edinburgh’s Koestler Parapsychology Unit* Jim Tucker, child psychologist and reincarnation researcher* Tucker’s 2025 paper on the >2,500 reported cases of reincarnation* 2024 review of cases on the “reincarnation birthmark” issue* Michael Sudduth’s 94-page debunking of the James Leininger case* Response from Jim Tucker on Leininger* Reply from Sudduth* Sudduth’s blog post on the same issue* Steven Novella on memory, children’s learning, and supposed reincarnationCreditsThe Science Fictions podcast is produced by Julian Mayers at Yada Yada Productions. This is a public episode. If you'd like to discuss this with other subscribers or get access to bonus episodes, visit sciencefictionspod.substack.com/subscribe

Maybe it’s the most important thing any scientist can study: what makes people happy? The trouble is, despite the importance, a lot of the science on “wellbeing” tends to be very rickety.But did you know that even one of the best-known findings of wellbeing research—the midlife crisis, or “inverted U shape” of happiness over the lifespan—has been questioned? In this episode we discuss the controversy.The Science Fictions podcast is brought to you by Works in Progress magazine. At worksinprogress.co you can read issue after issue of fascinating articles bursting with ideas on how humans made scientific and technological progress in the past, and how we can keep it going. You can also check out their selection of other podcasts at podcast.worksinprogress.co. Show notes* The American Psychologist paper that claimed to reveal the fluid dynamics of human happiness* Nick Brown and Alan Sokal’s devastating rebuttal* And coverage in The Guardian at the time* David Blanchflower’s original work on the inverted U-shape of happiness* And subsequent work that backs it up…* …and subsequent work that does not back it up* New paper that tries to work out why there are differing results* Afghanistan reporting the lowest wellbeing in recorded history* Our previous episode on the weird phenomenon of collider biasCreditsWe’re grateful to Dr. Julia Rohrer of Leipzig University for talking to us for this episode (though as usual, if there are mistakes, they’re ours and not hers). The Science Fictions podcast is produced by Julian Mayers at Yada Yada Productions. This is a public episode. If you'd like to discuss this with other subscribers or get access to bonus episodes, visit sciencefictionspod.substack.com/subscribe

You requested; we delivered. Lots of Science Fictions listeners have asked us to take a look into Donald Trump and RFK, Jr.’s recent claims about Tylenol (that is, paracetamol or acetaminophen—all the same thing). Does it cause autism?It turns out there’s more to this than you might’ve thought—regardless of all the recent hype, a lot of very reputable scientists take the idea seriously. But should they? In this emergency podcast, we go through all the relevant studies.The Science Fictions podcast is brought to you by Works in Progress magazine. In the ad this week we mentioned “The Death Rays that Guard Life”, an article from Issue 20 of the magazine about far-UVC light and how—with a lot more research—it might be the next big thing for reducing the spread of germs in hospitals and classrooms. Find that and many other articles and podcasts at worksinprogress.co.Show notes* The FDA’s September 2025 announcement on Tylenol and autism* The UK’s Department of Health and Social Care announcement the same day* “The phrase ‘no evidence’ is a read flag for bad science communication”, by Scott Alexander* 2003 theoretical paper with speculation about paracetamol and neurodevelopmental disorders* 2013 sibling control study in the International Journal of Epidemiology* “Ecological” study in Environmental Health from 2013 about circumcision rates, paracetamol, and autism* 2015 Danish seven-year follow-up study* 2019 cord blood study in JAMA Psychiatry* 2021 “consensus statement” on paracetamol and neurodevelopment* 2025 Japanese sibling-control study* 2024 very large Swedish sibling-control study* Study that sparked the current debate: the “Navigation Guide” review from Environmental Health* Description of what “Navigation Guide” is* STAT News on the evidence for a paracetamol-autism link; and on the controversy about the Dean of the Harvard School of Public Health* White House statement defending the existence of the link* BMJ article summing up the controversyCreditsThe Science Fictions podcast is produced by Julian Mayers at Yada Yada Productions. This is a public episode. If you'd like to discuss this with other subscribers or get access to bonus episodes, visit sciencefictionspod.substack.com/subscribe

This is a free preview of a paid episode. To hear more, visit sciencefictionspod.substack.comWarning: As you can tell from the title, this podcast covers a potentially distressing topic.Recent events have had us wondering about “copycat” violence. If people see violence reported in the media, are thRey more likely to do the same thing themselves? Does this apply to homicide, or suicide too? We start with an episode on suicide—one on homicide is coming soon. What’s the evidence for suicide contagion? What does this mean for how we should portray suicide in the news, and in fiction? As ever, there’s a scientific controversy behind every question.This is paid-only episode of the Science Fictions podcast: become a paid subscriber to hear the whole thing and read the show notes.

How can both of the following be true? (1) The world has record crop harvests this year; (2) climate change is ruining crop harvests and threatening food security. Does that make sense? Is it even really a contradiction? We look into how climate change is affecting crop yields, whether positively or negatively, and try to answer the biggest question of all: do we actually have to hand it to climate change deniers who say “CO2 is plant food”?The Science Fictions podcast is brought to you by Works in Progress magazine. You can now hear the editors of Works in Progress on their own podcast, talking to interesting people from the worlds of science, policy, technology, and history. Their most recent episode, on how traffic has ruined cities, is available at podcast.worksinprogress.co.Show notes* Hannah Ritchie’s new book, Clearing the Air* Her article on record harvests in 2025* An example of Matt Ridley making the argument that “CO2 is plant food”* Our World in Data on crop yields* Paper on the slower growth in crop yields due to climate change* Nature Plants paper on trees in the Amazon getting bigger over time* 2016 paper on the effects of climate change on crops and weeds* EarthArxiv preprint on the balance of the effects of temperature and CO2 on crop yields* The World Bank on fertilizer use per hectare* And on cereal yields* China’s fertiliser use peaking in around 2015* Less good news from Sub-Saharan Africa* Our older episode on climate sensitivity* Global per capita dietary data on calories consumed per day* Emissions from different kinds of food transportationCreditsWe’re very grateful to Dr. Hannah Ritchie from the University of Oxford and Our World in Data for talking to us for this episode. Any errors are ours, not hers. The Science Fictions podcast is produced by Julian Mayers at Yada Yada Productions. This is a public episode. If you'd like to discuss this with other subscribers or get access to bonus episodes, visit sciencefictionspod.substack.com/subscribe

What is that? What is that!? What is it!? Augh! No! Not the bees! NOT THE BEES! AAAAAUGGGHHHHH MY EYES! MY EYES! AAAGGHHHH!!!!This episode is about bees. And all insects, actually—are they in the process of being wiped out, in what’s become known as “insectageddon”? It’s one of those scientific beliefs that a lot of people kinda-sorta hold without knowing much about the evidence, so we thought we’d look into it.Tune in to hear how, in exactly the same way as they affect our episodes on psychology and psychiatry and medicine, really difficult issues of measurement are at the bottom of the entire debate.And once you’ve listened, feel free to buzz off.We’ve been talking about Works in Progress magazine on this podcast for years, but it’s only ever been an online magazine. Now, you can become a subscriber and get a beautiful IRL physical copy of the magazine delivered to you: a gorgeously-produced item that’ll make you desperate to read the excellent articles on science, technology, and human progress within. Find out more at www.worksinprogress.co/print.Show notes* The Rothamstead insect survey* The famous 2017 Krefeld paper in PLOS ONE* An earlier 2012 survey study from the Zoological Society of London* George Monbiot’s article where he coins the term “insectageddon”* The 2019 meta-analysis finding 40% of insect species are at risk of extinction* Response letter from Andrew Bladon et al.* Improved 2020 meta-analysis from van Klink et al.* Response letter* New paper on the improvements to the Chicago River* Two papers from the GLiTRS projectCreditsWe’re very grateful to Dr. Andrew Bladon from the University of Reading for talking to us for this episode. As ever, all errors are ours. The Science Fictions podcast is produced by Julian Mayers at Yada Yada Productions. This is a public episode. If you'd like to discuss this with other subscribers or get access to bonus episodes, visit sciencefictionspod.substack.com/subscribe

5—6—3—4—3—1—7—2In the first episode under our new podcast name (it’s now the Science Fictions podcast!), we ask whatever happened to all those games that claimed to tell you your “brain age”—games that turned into a whole scientific literature on brain training. We discuss: the still-unresolved question of whether training one specific cognitive ability makes you generally smarter; seemingly endless contrasting meta-analyses; and the small matter of what brain training might tell us about the nature of intelligence.(If you can repeat the list of numbers from the top in reverse order then you have the brain of a 25-year-old. If you’re 25 or younger, then I don’t know what to tell you.)We’re now an official part of the Works in Progress podcast world. You can find their other podcasts, including Hard Drugs, the one we talked about on today’s episode (about the remarkable development of a drug for HIV), at podcast.worksinprogress.co. Show notes* The 2008 PNAS paper that started the craze for working memory training* The under-discussed rebuttal* 2013 meta-analysis concluding there’s no evidence for far transfer* 2015 meta-analysis concluding there is no convincing evidence brain training is effective* 2016 meta-analysis saying there is no convincing evidence brain training is NOT effective* Very useful and detailed 2016 review of the evidence and the methodological issues inherent in brain training (including active vs. passive control groups)* 2020 meta-meta-analysis arguing that the active-passive distinction doesn’t matter* 2023 review criticising the meta-meta-analysis* And the authors’ own 2020 meta-analysis* 2022 meta-analysis of commercial brain training in older peopleCreditsThe Science Fictions podcast is produced by Julian Mayers at Yada Yada Productions. This is a public episode. If you'd like to discuss this with other subscribers or get access to bonus episodes, visit sciencefictionspod.substack.com/subscribe