You say potato, I say leprosy

This is the quarterly links ‘n’ updates post, a collection of things I’ve been reading and doing for the past few months. As late as 1813, some parts of the European medical establishment believed that potatoes cause leprosy. (Don’t even get ‘em started on scrofula!) Potato historian Salaman Redcliffe suggests that people were skeptical because potatoes look kinda weird, they grow in the ground, and you plant them as tubers rather than seeds, which are all extremely suspicious things for a food to do. You may remember the Spurious Correlations website, which dredges up random datasets and finds correlations between them—for instance, Lululemon’s stock price and the popularity of the first name Stevie. Now thanks to AI, each one of those correlations can be instantly turned into a full academic paper, like: LULU-LEMONADE: A STATISTICAL STUDY OF THE STEVIE-NIZED MARKET. Sadly, this technology makes many academic departments completely redundant. Via : there’s a good chance that 2025 will have the fewest murders ever recorded in the US. (We only have reliable data going back to 1960). I just added a new entry to my list of all-time great blog posts: Ask not why would you work in biology, but rather: why wouldn’t you?, by of . An excerpt: Yes, biology is very interesting, yes, biology is very hard to do well. Yet, it remains the only field that could do something of the utmost importance: prevent a urinary catheter from being shunted inside you in the upcoming future. Before World War I, the US government had basically no cryptographic capacity. So when the war broke out and suddenly they needed people to do code-breaking, where did they turn? To the Riverbank Institute, which had been set up by “Colonel” George Fabyan to decode the most important cipher of all: the one that supposedly proved the works of Shakespeare were written by Francis Bacon. Elizebeth and William Friedman, the couple who broke that cipher while working at Riverside, went on to become the first cryptologists at the precursor to the National Security Agency.12 One of the wildest blog posts I’ve read this year is about an American guy going to fight in the Ukraine war. Honestly, it sounds like a huge bummer: you squat in a trench and pretend to shoot at Russians and hope to not be killed by a drone. When I saw ’s post called How Pen Caps Work, I was like, “what do you mean? Pen caps work by...being....caps for pens”. Apparently not: for fountain pens, anyway, pen caps work through vacuum power. Putting the cap on and taking it off causes a tiny amount of suction that draws the ink into the nib. , who was one of the winners of my 2025 blog post competition, has a great series on why neuroscientists still can’t simulate a worm: I told [my mom] that this is what my job feels like—each animal has a different kind of radio in its head and/or body, and neuroscientists are trying to figure out things about them. Some neuroscientists want to fix radios; some want to build better radios. Others, like me, are just trying to understand them. To which Li’s mom responded: Other nematode fun facts from Li’s piece: they use static electricity to teleport themselves onto bumblebees as a way of getting around: And...nematodes survived the Space Shuttle Columbia explosion?? More great work from a 2025 Blog Extravaganza honoree: has a terrific article in Works in Progress: Why Isn’t AI Replacing Radiologists? Radiology was supposed to be the first medical speciality to be rendered obsolete by AI. Instead, radiology jobs are more numerous and salaries are higher than ever. The Polarization Dashboard is a useful sanity check against current events. Whenever something big happens in politics, people are like “WOW OUR VERY SOULS HAVE PERMANENTLY CHANGED” when in fact people almost always have the same opinions that they did yesterday. Here’s the change in support for murdering members of the opposite political party over time. Currently, <2% of Democrats and Republicans support it. (h/t ) See also: You’re Probably Wrong About How Things Have Changed. Some beginner researchers successfully replicated my Things Could Be Better paper without any expert help. I’m really proud of this! , who ran the workshop, writes: I did not help replicate this study because the group replicating Measures of Anchoring in Estimation Tasks [the other study being replicated] needed help understanding the language the paper was written in. In contrast the group replicating Things Could Be Better started their own replication within 15 minutes of being handed the paper and did not have any followup questions for me before they began the replication. Two years ago a Harvard Business School professor named Francesca Gino was fired for faking her data. (I wrote about the debacle here.) She sued the bloggers who outed her, but that lawsuit was thrown out. She also sued Harvard, claiming discrimination. Now Harvard is suing Gino back, alleging that when Gino submitted data to prove prove that her original data wasn’t fake...the new data was fake, too. and his friends ask: What if the NIH had been 40% smaller?. I appreciate how circumspect the authors are, but the short answer seems to be, “We would be significantly worse off, because many important medicines rely on research that would not have happened under a smaller budget”. This is further evidence of just how important it is to invest in science: even when we do it in a totally boneheaded way, it somehow still pays off. Recently, the Financial Times set the internet alight with these graphs: re-analyzed the data and claims the changes in conscientiousness are minimal, if they exist at all. I’m inclined to trust Ferguson’s account on this one: it’s super weird to see such huge changes in such small amounts of time on basically any psychological variables. Speaking of personality, ClearerThinking now has one mega personality test that will give you all your results from a bunch of different tests at once. I previously cited their work showing that supposedly scientific personality tests do not obviously outperform the bullshit ones, which continues to boggle people’s minds whenever I bring it up. One of the most famous psychology experiments of all time is Festinger and Carlsmith (1959), the classic demonstration of cognitive dissonance. The psychologist Matti Heino points out, though, that the main results literally don’t add up. In the table below, the circled means are impossible given the reported sample size—there’s no way to get an average of 3.08, for instance, if you have 20 people giving ratings on a 0-10 scale. I used to be one of those people who was like “well cognitive dissonance has been replicated in hundreds of studies” but it’s not like I ever actually read those studies. It just kinda seems like, c’mon! It’s cognitive dissonance! Everybody knows cognitive dissonance! Anyway, when a bunch of labs tried to replicate another classic demonstration of cognitive dissonance, they found no effect.3 And a new paper claims that When Prophecy Fails, a landmark book that documented the effects of cognitive dissonance in a doomsday cult, may have been greatly embellished or deliberately orchestrated by the authors.4 None of this means that cognitive dissonance doesn’t or can’t exist, but it does make me feel a whole lot of, uh, dissonance. According to this study, losing a Michelin star improves TripAdvisor ratings. This is probably because getting a star invites harsher judgment (“It’s good, but...is it Michelin good?”). The perfect restaurant is one that raises your expectations high enough to get you to come in, but leaves your expectations low enough that you can still be wowed. This is a hard target to hit: I’ve only ever seen two restaurants that have a 4.9 on Google after receiving thousands of ratings. Both of them are Thai restaurants that are nice but not fancy and cool but not trendy—you go expecting good Thai food and you get fantastic Thai food, and everybody goes…