introduction to why I actually laughed out loud when I saw this en.wikipedia.org/wiki/Funnel_pl… x.com/literalbanana/…
I have two suspicions (detailed in that thread):

1. there are sometimes two distributions, one for honest studies and one for sketchy ones

2. if there's a clear peak, eyeballing the peak is often more informative than doing math alibis to get an effect size far from the peak

idk much about 1776 but I've been spending a lot of time in 1690s new england and I'm so happy there's firewood discourse because there's a subtle way that firewood issues were a proximate cause of the salem witch panic

https://twitter.com/DKThomp/status/1941894499141115992

Robert Calef mentions in his 1700 book that the minister of Salem Village, Samuel Parris, a major villain of the witch panic, had been a failed merchant before taking up the ministry (Calef himself was a successful merchant so I think it was kind of an own)

welcome to my TED talk today we will take a look at one tiny little Fact, how it was constructed, and the layers of context necessary to evaluate the evidentiary value of a study: the Fact we will be looking at is from the Sapolsky book we've been enjoying, Determined: Life Without Free Will - "exogenous oxytocin administration increases third-party punishment"

I'm reading Robert Sapolsky's Determined: Life Without Free Will as a mild form of self harm and here are two perfect pages - ego depletion, surgeon's birthdays, Francesca Gino/Dan Ariely study, implicit bias, hungry judges - all on the same page!!!

don't worry - in the notes he explains it's still all real! it's 2010 again!

interesting interpretation - the disastrous retracted study claiming that rigorous open-science practices produced high replicability implied backwards causation (like in the study in the QT)

https://twitter.com/literalbanana/status/1768441768708809182

That is, the original studies under replication were not clearly using rigorous open-science practices - only the later replication attempts were. Yet these somehow cause a high rate of replicability?? (sources joebakcoleman.com/blog/2024/prot… / users.eecs.northwestern.edu/~jhullman/Hull…)

I have spent several days bathing in the porta potty tank of science and I wanted to share my journey with all of you in a characteristically brief thread sciencedirect.com/science/articl… The basic subject matter here is the psychological science that says that inducing disgust (showing a picture of a poop, touching a plastic poop, smelling fart spray etc.) makes people more morally judgmental

One Fact that I thought was relatively solid ground was that crime is a function of youth - that it was a human universal that it's mostly young males maybe 15-22 who do most crimes and then get tired of it Apparently this is not so universal - for example, in the US and UK, the youth crime thing seems to be a weird phenomenon of the 1980s and 90s (less youth-concentrated before and after)

Something I've been interested in lately is how often skepticism of the underlying theory motivates the discovery of fake scientific results (fraud or accidental fakery) That certainly seems to be the case with Jorge Hirsch's discovery of fake results in superconductor science - he is a skeptic of BCS theory, the currently-accepted theory, and that seems to have motivated his investigations

https://x.com/literalbanana/status/1776429847788884050

In 1955, Henry Beecher published one of the most disastrous documents in the history of science, The Powerful Placebo dcscience.net/beecher-placeb…
In the 1990s, two scholars, Gunver Kienle and Helmut Kiene, pulled a banana and actually looked at the studies, and found that the 1955 paper was made of misquotation, sloppiness, and lies: web.archive.org/web/2023041104…

Is the placebo effect real? A brief (lol) thread of some of the points from my unfortunately novella-length post about why I think the placebo effect is fake: Placebos have a very real and powerful use: as a blind. They can show when apparent pre-post healing effects are caused by a treatment, or to something not related to the treatment. Placebos can show if a hyped treatment is fake.

I really wondered if this or Stardew 1.6 would come out first

https://x.com/stephaniemlee/status/1768406323488821332

study 1 of the infamous "sign at the top" study claims to test how much signing at the top or bottom of a form affects cheating at a math task...but apparently the subjects didn't even see the form they signed until AFTER the cheating task??

I am skeptical of the "exercise cures depression" meta-analysis because these are often ways of laundering lots of bad studies that may or may not have ever happened into Serious Science Facts

https://x.com/SteveStuWill/status/1758276648506671283

Of course I do not think it would be hard to demonstrate superiority to antidepressants given how low their efficacy seems to be - but we know how poor it is from lots of analyses of published and unpublished data carcinisation.com/2022/09/07/the…

I've been trying to figure out why writing for a public audience seems less and less rewarding to people lately One hypothesis I have is to oversimplify into three distinct periods of time - before mass media, mass media era, and present day

The author of this is awesome for compiling it (working link in first reply) - I would love to see the opposite of this, a list of real/true findings in psychology

https://twitter.com/SteveStuWill/status/1475496701393784841

It's natural to look at a list like this and imagine that the failures to replicate are rare, and that there's a strong body of genuinely explanatory and predictive findings in the field - I don't think that's true, but I would love a list of purported ones!

One of the problems with the executive function/self-regulation/ego depletion literature is that "self-regulation" is a really complicated phenomenon that only pretends to be simple (definition from baumeister et al.) It seems obvious to me that there's a major suffering dimension involved - for instance, if you naturally have low appetite, you'll be able to stay thin without suffering, but if you have a high appetite, it will involve a lot of suffering to avoid eating too much

Why, in general, are old things beautiful and new things ugly? This is my descriptive answer, but how it got that way is more complicated carcinisation.com/2015/04/01/the… Mathematical art can achieve a high degree of life - levels of scale and deep interlock are particularly strong here, I imagine they’re baked into the math!

https://twitter.com/graveolens/status/1438809071419895808

Have you seen this chart? I’ve been seeing it a lot lately and wondered what it was based on Other than the obvious y-axis shenanigans, one author’s criticism of this chart was that the “sugar” line tracks estimates of the CONSUMPTION of certain types of sugar, but the way this was calculated from AVAILABLE sugar changed medium.com/@robertagreer/…

I was excited to read this formalization of Andrew Gelman’s interesting and in my opinion profound “piranha principle” stat.columbia.edu/~gelman/resear… so of course I have to talk about the one tiny thing I found to disagree with in it

This was micro-popular enough that it made its way into my life - I wanted to explain how it works because I think this will become more common as a method

https://twitter.com/margrev/status/1372978689328091137

The tool used is called the “Evaluative Lexicon” - this comes in a checklist form of 42 words, and a software form of around 1500 words.

a brief thread of visual patterns I’m trying to get a handle on

1. the BZ reaction 1. lace agate youtube.com/c/MichiganRocks

I’m volunteer hostile fact checking Nudge because I hate myself and I thought it would be a bunch of n=30 fMRI studies and priming stuff but the first reference is just a blog post from 1997 (third image is commentary) sometimes it will be a reference to a popular book with no page number and you have to dig hard to figure out that it’s an n=30 fMRI study birc.jaredjustus.com/assets/publica…