Ancient Chinese alchemists wanted to find an elixir of immortality, none of which seemed to work. When you realize how much lead and mercury were in the elixirs, that makes sense. Unfortunately, there aren’t good records about when people actually died from testing their own elixirs, because it’s tough to differentiate someone who’s left their earthly body behind from somebody with, well, mercury poisoning.
The author also speculates that this seriously slowed down scientific progress in China at the time. This isn’t because the alchemists’ methods were insufficiently rigorous — quite the opposite: “the elixir mania may have acted as a real inhibiting factor for the progress of chemical knowledge during the medieval Chinese centuries, for often no doubt the most experienced or industrious experimenters were the most enthusiastic believers and in the end the surest victims.”
Hosting this one myself to spare others the insane amount of hoops I had to jump through to find this pdf.
Insane story of cocaine as local anesthetic for eye surgeries. Skip to page 25 of the pdf (332 by page numberings) for the good stuff:
‘Now it was necessary to go one step further and to repeat the experiment upon a human being. We trickled the solution under the upraised lids of each other’s eyes. Then we put a mirror before us, took a pin in hand, and tried to touch the cornea with its head. Almost simultaneously we could joyously assure ourselves, “I can’t feel a thing”. We could make a dent in the cornea without the slightest awareness of the touch, let alone any unpleasant sensation or reaction. With that the dis- covery of local anesthesia was completed. I rejoice that I was the first to congratulate Dr. Koller as a benefactor of man- kind’ (10).
Key thesis?
The three key ways Bell Labs nudged its basic researchers toward the right problems were:
- Granting researchers what I’ll call a “long leash, but a narrow fence” in which to conduct their explorations.
- Facilitating very regular interactions between the basic researchers and Bell’s fundamental development researchers, engineers, manufacturing facilities, and implementation staff.
- To top it off, Bell had a corps of what they called systems engineers who ensured that the integration of its best researchers and most pressing problems was not left to chance.
My biggest takeaway from the article? Holy shit systems engineers are badass. Want to know why I think that? Read the article.
detailed neuropsychological testing showed that the disparity between A.V.’s preserved intellectual functioning (Full Scale IQ: 115) and severe memory deficit (Delayed Memory Index: 42) is one of the largest on record. Despite this deficit, A.V. has regained a higher level of functioning and autonomy compared to previously documented amnesic cases with major bilateral MTL lesions. As a millennial, one advantage which A.V. has over prior amnesic cases is fluency with digital technology – particularly the smartphone. The analysis of his phone and specific app usage showed a pattern that is consistent with the strategy to offload cognitive tasks that would normally be supported by the MTL.
This guy can’t form new memories, and he drives! Safely! Super interesting case, and cool to see how much of modern cognition we can offload onto devices.
99% of the population averages 8.06 hours of sleep a night.
1% of the population averages 6.25 hours of sleep a night.
Why?
Three known cellular mutations:
The craziest part is that people with these mutations don’t seem to get the negatives that come with sleep deprivation for the rest of us.
Soooo… can we get those benefits for everybody? if you’re alive for 80 years and sleep for 8 of your 24 hours, you’re spending 20 years asleep. But if we could cut it down to 6, that’s 10 years of life re-gained. Huge! The TAM is literally everybody.
Rough ideas from the author:
- AAV-based therapies to restore orexin signaling (which DEC2 inhibits) in narcoleptic mice already work, increasing wakeful time by 13% and decreasing wake bouts by a fifth (more single stretches wakefulness) [21]
- The actual approach could end up using this kind of AAV delivery and/or plasmids carrying a base editing system, decreasing the efficacy of DEC2 suppressing orexin in the brain by making the protein less functional. Or directly upregulating orexin.
- Dual orexin antagonists (blocking both OX1R and OX2R) are in human trials and potentially promising for chronic insomnia [22]
- So what about small-molecule orexin agonists in the morning? Yes, that’s plausible:
- The first direct orexin agonists are running human trials, such as TAK-861 [23]
- Farther-out idea: What about automatic orexin peptide administration in sync with your personal circadian rhythm (think of insulin pumps)?
- Small molecules are also so much easier to get FDA-approval for.
- Weirder ideas:
- Semi-permanent (months?) silencing the mRNA transcripts, e.g. with siRNA, short hairpin RNA, etc
- Flagging DEC2 or other inhibitor proteins for faster intracellular degradation
- Destabilizing the protein, snipping in variants, etc.
- Insert extra copies of the wakefulness mechanisms
There’s a lot of room to explore here: We’re already given at least 6 mutations whose effects to potentially emulate; we could target inhibitors and/or the wakefulness mechanisms themselves; we could try lots of different paths (molecules, antibodies, editing, RNA, simulation, …). That makes it such an interesting problem to work on, and de-risks the project a little. Keep in mind: intra-brain editing is currently hard™️. But people are feverishly working on it.
One of the people working feverishly is Helena Rosengarten.
She’s got ideas.
Not flawless — how do we deliver it effectively (they have to be small, but the changes we want to make might be big)? How do we only target the cells we want to target (Helena suggests checking for RNA that’s only expressed in the cells we want to edit)? How do we get them through the blood-brain barrier?
Unlike the above, less invasive and reversible. Tougher, though.
There are some proteins that bring about tiredness and that the short sleepers lack. What if we just… got rid of them? There’s autophagic precedent.
Maybe degrading is a bit much — what if we inhibit?
Regardless: > Most important to remember: This phenotype already exists. It works and is safe. The only question is how to mimic this in the easiest way. It’s not about if, just how.
I can say that I, personally, would take the short-sleeping pill if I could. Coffee, but instead of increasing wakefulness, it decreases the amount of sleep you actually need? Sounds like treating a cause rather than symptoms.
Until the eighteenth century, accurate offshore navigation was an impossible dream. There was no method or technology to determine longitude precisely in the open sea. The longitude puzzle was finally solved with the marine chronometer, “one of the most important inventions of the era of the Industrial Revolution” on a par with the spinning jenny and the steam engine.
To the best of our knowledge no one has studied the causal impact of this innovation on navigation and urbanization in a systematic econometric framework.
In principle, the longitude of a ship in the open sea can be inferred by comparing the time in a fixed location, say Greenwich, with the local time, since the earth rotates on its axis 15 degrees every hour.
The first solution to this longitude puzzle, the lunar distance method, came in the 1750s and was based on the observation of the angular distance between the moon and a celestial body. This method, which was quickly adopted by all the major navies of the period, had however an important drawback: it would produce a reliable measure of the longitude at sea only when the stars were perfectly visible at night.
The final solution came some decades after with the marine chronometer, a precise clock, that could keep the Greenwich time with minimal error.
We show that: (1) there was an exceptional increase in the relative speed of vessels navigating in cloudy regions – where the lunar method could not be applied – compared to vessels navigating in clear-sky regions in the first half of the nineteenth century; (2) this relative change in speed is fully explained by open sea routes, while we do not observe it for coastal navigation.
We document this phenomenon with a second set of estimates, highlighting two significant changes in sailing routes attributable to the advantages offered by the chronometer. First, we show a relative increase in the number of observations of ships navigating under cloudy skies compared to clear skies, starting from the 1810s. Second, we find that the rise in crossings through cloudy regions in the open sea is largely due to instances of non-parallel navigation.
In the second part of our empirical investigation, we study how the chronometer reshaped the geography of the world.
In the third part of the empirical analysis, we study the impact of this change in geography induced by the chronometer on urbanization.
Our difference-in-differences estimates corroborate the message coming from these raw data. When looking at coastal grid cells outside of Europe, we find that a one percent reduction in the time required for a round trip from Europe enabled by the chronometer is associated with an increase in urban population of approximately 1.7 percent in 1850 and 4.7 percent in 1900.
We find no impact of the chronometer on urban population in the inland grid cells.
We provide some suggestive evidence that the contribution of the chronometer to urbanization is likely to be, at least partially, a result of its impact on the European colonial expansion in the nineteenth century.
We show that the chronometer was likely guiding the expansion of the British Empire, in the first decades of the nineteenth century, in Asia, Africa and Oceania.
The invention of the chronometer had large but geographically uneven effects on navigation and led to massive shifts in the distribution of cities and urban population.
Our estimates suggest that the invention of the chronometer accounts for more than a third of the increase in the urban population living along the coast outside of Europe during the nineteenth century.
Ozy’s book review of Merchants of Doubt. Whenever I recommend the book to someone, I follow up that they should probably just read this review instead.
There’s a lot of hype around ketamine as a depression treatment. Ketamine is really tough to run blinded trials for, though, since you can generally tell whether you’re on ketamine.
A 2023 study from Lii et al. found a cool way around this — find depressed people who were undergoing unrelated surgeries, then give them ketamine while they’re under anesthesia. That way, they won’t be able to tell whether they’re part of the treatment group or the control group! Super cool study idea!
They found that the blinding worked, and participants couldn’t tell whether they got actual ketamine, but also found no differences in outcomes across groups — not in depression scores, response rates, or opioid consumption.
This blogpost argues that Lii et al.’s results probably aren’t a true negative, though — “the most obvious way to miss a true effect,” after all, “is through the sample size being too small.” Would you trust an antidepressant study with 20 participants per group? Probably not. Well, Lii et al.’s study had 40 participants — 20 treatment, 20 control. We normally call this “statistical power,” and
A power calculation generally requires three parameters: (i) the minimum effect size we wish to detect; (ii) the alpha level (this is the threshold for finding a false positive result and by convention is set to 0.05, meaning that on 5% of occasions we would expect to find a positive result purely by chance; (iii) the power we require to detect a true effect (by convention this is usual set at 80%, meaning if a true effect exists between treatments we would expect to detect it 80% of the time).
The authors don’t state the expected effect size, only that they will detect a 30% change of depressive symptoms from baseline. This doesn’t make sense to me - a power calculation for a placebo controlled trial is explicitly to test the difference between groups, not just the change in symptoms from baseline. They calculated 15 participants per group (remarkably low). However, they then aim to recruit an additional five participants per group to account for patient dropout. I find it difficult to believe they were concerned that 1/3 of their sample would drop out, given the follow-up for the primary outcome was three days (indeed in the actual study no participants dropped out).
Ultimately, the super cool anesthesia study probably shouldn’t update you much in either direction — both I and the author of the linked post would be excited to see this in a study with a larger sample size, though.
Hyeongsoo’s experience was similar to that of many researchers in North Korea: absurd, sycophantic, isolated, and stressful. And yet scientists in this country still manage to maintain a leading nuclear weapons program and make other strides, including developing an artificial knee joint, ultrasound machine, and CT scanner. Nuclear and chemical weapons get priority in North Korea, Hyeongsoo said, followed by research into the health of the ruling family. Isolation means North Korean scientists have a hard time getting access to the global knowledge base, much less specialized or expensive equipment.
My favorite excerpt: > The incentives for fraud in business academia are significant. If you can meet the standards for hiring, promotion, and tenure at an R1 university (something that is much easier once you fabricate your data), you will get: > > - A 6-figure salary with full benefits until you retire > - Complete job security > - A flexible work environment (no boss, remote work…) > - The social status and reputational benefits that go with the “Professor/Dr.” title > - Opportunities to do book deals, TED talks, to teach in executive education, to conduct corporate workshops… > > The benefits of fraud must be balanced with the risks of course. Are the risks of being caught for faking data high enough? I don’t think so: > > The peer review process, as it exists today, makes it extremely difficult to catch fraud. Only a minority of journals require making data and materials available to reviewers, who are overworked and under-qualified to detect fraud. Even fewer journals require posting data and materials to the public. If fraud isn’t detected during peer-review, who will be able to catch it? > > The bar to accuse someone of fraud is extremely high. Failing to replicate the effect? Not enough. Non-sensical effect sizes? Not enough. Anomalies in data? Not enough. Unless you can invest the resources to identify anomalous patterns of fraud across multiple papers, THEN drum up enough support from journals or universities to consider your suspicions, THEN hold their feet to the fire when they are unwilling to act… the probability that the person will never face consequences for fabricating data is very high. > > The incentives to investigate and call out fraud are non-existent. In fact, the opposite is true: If you find something fishy in a paper, your mentor, colleagues, and friends will most likely suggest that you keep quiet and move on (or as I have learned the hard way, they might even try to bully you into silence). If you are crazy enough to ignore this advice, you are facing a Sisyphean task: Emailing authors to share data (which they do want not to), begging universities to investigate (which they do not want to), convincing journals to retract (which they do not want to), waiting months or years for them to share their findings with the public (if it ever happens)…
First, super apt name for a guy writing about medicine.
‘Between 1999 and 2015, rates for the top three causes of death in the United States fell sharply,’ writes Charles Piller in his new book Doctored. ‘Alzheimer’s rates and fatalities went in the opposite direction.’
Piller, an investigative journalist and a beneficiary of nominative determinism, was instrumental in uncovering part of why that is, in an article in Science in 2022.
Lesné, who published many more papers on Aβ*56, was almost the only researcher ever to detect the substance: Only a handful of other groups have published anything on it, and other researchers have tried and failed to find it. Certainly, no other group has found any indication that it affects memory.
A typical ADHD diagnosis in America is done by a paediatrician or a family doctor in an office visit as brief as 15 minutes. The norm in Europe is an hours-long assessment by a psychiatrist. The difference has to do with America’s scant health-insurance coverage for mental health care and lax rules about who can diagnose ADHD. Lots of corners are cut in those 15 minutes. A full ADHD assessment that follows psychiatric guidelines entails cognitive tests of the child, detailed interviews of parents and teachers and even observing the child in the classroom. This helps to rule out other causes for ADHD-like behaviour. Often, it is a child’s normal reaction to a chaotic home or classroom.
Particularly damningly, > ADHD diagnoses in America spike by 14% on Halloween compared to the ten weekdays before or after that.