Links 1/16: Shaolink
The McCullough Effect is an optical illusion where after you stare at one picture, another picture looks like it has some different colors. But unlike normal retinal-satiety optical illusions which last a couple of seconds, the McCullough Effect can last hours to months depending on how carefully you prime it. Try it now.
Reddit: What would the person who named walkie-talkies have named other things? Defibrillators are “heartie-starties”; cruise missiles are “zoomie-boomies”. H/t Kaj)
Including ethnic studies classes in secondary school will increase school attendance by 21% and GPA by 1.4 grade points, is apparently by far the best educational intervention ever discovered, and can singlehandedly save the school system. Or possibly there’s a problem with the methodology. The paper’s paywalled, so who knows?
That star with the weird brightness fluctuations that some people thought might be an alien Dyson sphere has even weirder brightness fluctuations than we thought. And cometary clouds have been pretty much ruled out. [EDIT: or maybe not?]
IQ scientist and Intelligence: All That Matters author Stuart Ritchie reviews Garrett Jones’ Hive Mind
for Intelligence. Brings up some of the same points as in my review; I feel vindicated! Also: Jones interviewed by AEI.
A lot of people have sent me this article where Carol Dweck says growth mindset is being misused. Certainly we shouldn’t misuse things, but I want to reiterate that my disagreement is prior to any misuse and I still am not sure that even Dweck-approved correctly-used growth mindset is as effective as generally believed.
Stumbling and Mumbling on capitalism vs. markets.
It looks like Greg Cochran no longer believes mutational load is a crucial determinant of IQ. As always, ability to change one’s mind is to be praised and celebrated as a rare but powerful talent. Also canalization.
New study finds that cannabis use does not affect IQ, apparently more authoritative than all the past studies that found it did. Why are cigarettes such an important confounder? Do they cause cognitive issues?
Mike Hearn says that Bitcoin is doomed. Bram Cohen says that Mike Hearn is a whiner who is ragequitting Bitcoin because nobody wanted to let him take it over. The Economist explains the controversy using the phrase “forking hell”. Other people roll up their sleeves and come up with a temporary but mutually agreeable solution to the supposedly Bitcoin-killing problem. We are all helpfully reminded that Bitcoin has been declared doomed 89 times so far, yet continues to exist.
In theory, an infinite number of monkeys could produce the complete works of Shakespeare. In practice, “[we] concluded that monkeys are not random generators“.
Conservatives always say, kind of as preemptive schadenfreude, that nobody would ever hire spoiled student protesters. But this article in the Financial Post is the first time I’ve seen an apparently apolitical, practical-minded discussion in a business context of how to avoid them; it suggests for example searching people’s social media for telltale signs [by employment lawyer; may be self-promotional]. I’m very split; on the one hand I believe in freedom of association and if somebody is clearly going to be trouble you shouldn’t force people to throw out that information and place themselves in a position to depend on that person anyway. On the other hand, I also think that part of meaningful freedom of opinion is that expressing your opinion won’t prevent you from getting a job twenty years down the road. I guess maybe look at the things people do as part of protests (eg if they burn something down or trash buildings) but don’t necessarily judge them for protesting something even if you don’t agree with their position? But I hope this reinforces what I’ve been saying about how getting good meta-level rules about not punishing people for speaking their mind is a common cause of all sides of political debates.
Sort of related: University of Missouri rumored to have declining application rate due to bad publicity from protests last fall.
Fossil words are antiquated words that only survive as part of an expression, like the “eke” in “eke out” or the “beck” in “beck and call”. Related: linguistic Siamese twins are two-word phrases that have to be in a specific order, like “hammer and sickle” or “salt and pepper”.
All-cause mortality over the course of a year rises with proximity to New Years’ Day, which is the deadliest day of the year. Nobody knows why, and it doesn’t seem to have to do with drunk driving, weather, or hospital schedules.
An easy way to fund some kind of important or charitable project you have going on: get a government grant. Related: the government grant process is a terrible confusopoly, which is mostly bad but can be good if you learn to navigate terrible confusopolies and don’t want too many competitors.
The Church Of The Flying Spaghetti Monster has been protesting people being allowed to wear hijabs in drivers license photos by demanding the same “faith-based” right to wear colanders on their head. Unfortunately one of them tried this in Russia and got his comeuppance: he is allowed to drive while wearing a colander on his head, and only while wearing a colander on his head.
Hitler is a rock star in South Asia. Not literally. Literally he’s a retired plumber in Argentina.
Over the past ten years, there’s been an almost 50% increase in deaths due to legal intervention (ie shot by police).
I’ve been saying this for a while, but I’m glad to have backup: pregnant women should supplement with iodine even in developed countries.
More neat methodologies: mental rotation is often used as a proxy for mathematical ability. Boys are usually better at mental rotation than girls, but it’s hard to tell whether this is biological or cultural. But girls who have a male twin get exposed to lots of testosterone in the uterus and probably have more male-like brains. So you might be able to distinguish biological from cultural effects by comparing mental rotation performance of girls who had male vs. female twins.
Related: straight men do better than gay men (and gay women better than straight women) on rotation tasks. Was Turing just a gigantic outlier, or what?
Related: why are gay men shorter than straight men?
While civilized countries debate how many new immigrants to let in, Britain is planning to deport all legal residents who have lived in the UK for more than five years unless they can meet an income threshold which is actually significantly higher than the average UK income. Is there anyone who thinks deporting upper-middle-class people who have been in Britain for decades and have houses and families there is vitally important important to national security? Especially bad because it’s a new law, so these people planned their lives in Britain around people not doing this.
Women whose resume suggests that they’re lesbian get 30% fewer calls for job interviews.
Charter schools in Boston get better test scores than public schools in Boston. Some argue this is because they teach to the test more than public schools do. A new study tries an interesting methodology: see whether these schools have a greater advantage on the higher-stakes, more traditional, more easily-gamed tests that teach-to-the-test schools would be more likely to be teaching to.
Related: state takeovers help failing schools. Public schools in Louisiana outperform voucher schools. I don’t really care that much about takeovers or vouchers, but results like this drag me out of my skepticism and force me to admit there’s some effect of how well schools are run on test scores. Whether that matters for real life applications ten years later is a harder question.
Not as related as it sounds: doubling teacher salary had no effect on any educational parameter in Indonesia. But they just kept all the same teachers and paid them more for no reason, so this doesn’t prove that increasing teacher salaries in the way people usually mean (ie in order to attract better teachers) wouldn’t be a good idea. And a new study does show pay for performance has improved DC’s public schools.
Thank whatever God you believe in that you’re not a junior doctor in the United Kingdom (I was a medical student in Ireland, which was close enough to inspire me to flee across the Atlantic). Proving that it is always good for making things worse, the UK government accuses doctors of killing people by occasionally having days off, but the evidence isn’t enough to support their claim.
Unfortunately-named consequentialist Max Harms has written a sci-fi book about the Singularity, Crystal Society. Haven’t read it yet but people I trust including Brienne Yudkowsky and Kaj Sotala say it’s good. Also: island exploration computer game The Witness (by the author of Braid) is donating 10% of sales to Against Malaria Foundation.
Ultra-premium water is on the rise. I didn’t even know “water sommelier” was a real profession.
Lots of people are warning against the alt-right these days, but needless to say Xenosystems’ warning is a little different. “For the Alt-Right, generally speaking, fascism is basically a great idea; for NRx, fascism is a late-stage leftist aberration made peculiarly toxic by its comparative practicality. There’s no real room for a meeting of minds on this point. From the NRx perspective, the Alt-Right is to be appreciated for helping to clean us up. They’re most welcome to take whoever they can, especially if they shut the door on the way out.”
The Dictionary Of Ancient Magic Words And Spells is a pretty good resource for all of the interesting things our ancestors thought you could do with garbled Latin and a copious supply of newt eyes.
Why does Donald Trump play Phantom of the Opera at all his campaign rallies? Does he just really like Phantom of the Opera? Sort of related: Developing And Testing A Scale To Measure Need For Drama.
The Empirics Of Free Speech (warning: long). What does free speech actually do or not do, according to the evidence? Does it let corporations buy elections? Does it result in heavily biased media? Can people use it to incite violence? Do people actually call “FIRE” in crowded theaters just to laugh as everyone tramples each other? This post will tell you much more than you wanted to know about all of these questions.
I’ve seen this idea floating a few places before under the name “proxy democracy” – a government that’s a direct democracy, but you can delegate your vote to anyone you like, be it a professional senator or just your friend who knows more about politics than you do. Now Google is calling it liquid democracy and testing it for some forms of corporate decision-making.
Jerry Coyne (Why Evolution Is True) reports on a controlled experiment on Facebook – make two otherwise identical Facebook groups, one anti-Palestine and the other anti-Israel. Sure enough, the anti-Palestine one gets banned and the anti-Israel one is left up [though the experiment itself is done by a pro-Israel group I do not trust as much as I trust Coyne]. And Marc Randazza of Popehat says that he’s tried a similar experiment and found that social-justice-branded accounts on Twitter can harass as much as they want up to and including death threats without consequences, but conservative-branded accounts are cracked down upon for slight offenses [though he does not post proof of this experiment]. Overall not likely to convince the not-already-convinced, but matches the anecdotal evidence I hear. Although private companies have the right to monitor their own customers as they see fit, I think FIRE’s philosophy – hold organizations to their stated principles and rules, but criticize them when they fall short or enforce them selectively – is fair. A Facebook that said outright “We’ll ban you for criticizing Palestine, but criticize Israel as much as you want” would have every right to go through with its policies as written – but also might not have too many users.
Studies traditionally show that immigrants do not “steal” native jobs or harm the native economy in any way. The major study to contradict this wisdom was Borjas on the Mariel boatlift of refugees from Cuba, but more recently reanalyses of the data by other economists (or as we now call them, “research parasites”) have cast doubt on that conclusion and the entire field has become an impenetrable quagmire. RealClearPolicy has an excellent and unbiased summary of the debate and of how people are getting such different results.
Philosophers experience silly cognitive biases even on exactly the kinds of problems where they should be most philosophical.
Quantifying Gains In The War On Cancer: ” We estimate that 3-year cancer-related mortality of cancer patients fell 16.7% from 1997 to 2007. Overall, advances in treatment reduced mortality rates by approximately 12.2% while advances in early detection reduced mortality rates by 4.5%.”
The New York Times has a hit piece perfectly nice article on the Center For Applied Rationality, a Less Wrong-affiliated self-help workshop group in Berkeley.
Who would Chinese people vote for in the US presidential election? Spoiler: Donald Trump, but only until somebody tells them what Donald Trump believes.
Related: @DPRK_News (parody North Korean Twitter account actually run by Popehat) covers the Democratic debate.
A pretty comprehensible explanation of what’s going on with Flint’s water. But I worry it might be too quick to exonerate politicians based on them not necessarily making bad water treatment decisions, when the things people are really angry about is them covering it up / not reacting fast enough.
Razib Khan predicts ISIS’ ideology will become more popular.
Dr. David Ludwig debates Dr. Stephan Guyenet on the calorie hypothesis vs. the insulin hypothesis of obesity. They’re both really smart and excellent communicators and this is a great demonstration of the level that this kind of debate should be held it.
This is a really neat new study: The impact of having a father who went to Vietnam. Since whether or not a 1960s American man went to Vietnam was partially determined by the draft lottery, you should be able to factor out all the other reasons someone might or might not go to Vietnam and get what was basically a randomized experiment sending people into war zones. The research finds that the children of people with bad draft numbers (more likely to have gone to Vietnam) make about $200 – $500 less fifty years later than the children of similar people who were less likely to have gone to Vietnam. That doesn’t seem like much, but since only about 10% of the people in the bad-draft-number category actually ended up in Nam and this is the effect that must be driving the average difference, it might be that those children are making $2000 – $5000 less, which actually is a lot (note that this was during a period when very few Americans died in Vietnam, so not much selection effect; the paper also adjusted for all the other reasonable objections I can think of). This is really weird. It’s unclear exactly how the father’s military service hurts the kid, but good guesses would be something like PTSD making the father less effective as a parent, or the father’s military service preventing them from getting as good a job. But that would be a shared environmental effect, which shouldn’t happen, and nongenetic intergenerational transfer of human capital, which also shouldn’t happen! Very interesting.
I’d never heard this before and it sounds fascinating: A Drug To Cure Fear. Apparently you can insta-cure a phobia by taking propranalol (a common drug that blocks some of the bodily effects of emotion) and then exposing yourself to the phobic trigger. Sounds plausible – you’re habituating yourself by “proving” to your brain that it doesn’t scare you – but the drug is so common I’d be surprised nobody noticed before. Anybody with a phobia and access to propranalol want to try this and tell me how it works?