Psychologists have cataloged a list of common cognitive biases. As far as I know most of what Kahneman/Tversky’s work still holds up. If a person can overcome some of these biases then they should be able to get an advantage in many situations.

A more controversial point:
I think one plausible way to get advantage is to exploit the fact that many people are very unwilling to break certain laws. Even when its pretty safe. Take modafinil, most people agree its effective and since its schedule 4 its not a big risk to buy it online. But you can presumably get an advantage from being in the 30-50% of people who are wlling to take prescription drugs to be more productive.

Report comment

I hypothesize that quite such an extreme differential is largely driven by buyers-to-let, who pay disproportionate attention to postcode-scoped statistics as a quick way of summarizing information about an area.

So I’m profiting from the worst argument in the world, I guess.

Report comment

TLDR: Bacognize good. Himalaya not as good, but under the limits. Nootrabiolabs bad.

Report comment

Report comment

Report comment

That’s a really good heuristic.

Report comment

Report comment

Like, if I tried to get all other men to believe that they shouldn’t care about hygene or social skills or money or whatever, despite my having naturally neutral smell or wit or inheritance. Am I fighting against my privilege, or encouraging others not to try to reduce my natural advantage in areas where the tastes of women are unlikely to change?

Report comment

Lithium is a very different story. There are a handful of suggestive studies about lithium deficiency being associated with mental illness, but there’s no consensus and no generally agreed on deficiency syndrome. A lot of people pushing the lithium deficiency concept are supplement pushers and quacks. On the other hand, there are a damn ton of unpleasant side effects to lithium at therapeutic levels, including obesity, sedation, and cognitive blunting…Lithium offers poorly-documented benefits, and extensive and well-understood harms.

There’s a damn ton of side effects – at doses as much as 6000x larger (the ceiling of natural levels tends to be 150 micrograms/l, therapeutic doses seem to be often around 900mg or 900000mcg). This is a point that has been made repeatedly.

More importantly, you are attacking a strawman here. Neither me nor Scott nor anyone else is calling for running out and dumping lithium in local reservoirs. We certainly do agree that the data, while interesting, is very far from definitive. Our criticisms are about the absence of experimentation: we are calling for experiments to be done to furnish precisely the sort of information which would show lithium to be useful at all.

After all if the correlations turn out to not be causal (as is one’s strong prior for any correlational result of this sort), then there is nothing further to debate! Obviously. It will just have been some boring confounding from insufficient controlling for rural populations or the success of the local harvest being caused by additional rain or something else we’ll never figure out, and of no general interest.

The question is, if the causality is verified, whether lithium should then join the ranks of other public health interventions such as iodization, iron fortification, vitamin D fortification, fluoridization, vaccines, remineralization of overly-purified water (eg from revere osmosis), and if not, what clear moral principle bars its use.

My response is that you can tell the difference based on what happens when you administer it. If you administer it, it causes some unarguably beneficial changes, and then the changes stop, then it’s a nutrient, and the point at which the changes stop is the baseline level at which the body functions normally. If you administer it, and it keeps causing changes all the way up to the point of toxicity without there being any level that you can call baseline, and there are harmful changes as well as beneficial ones or the changes are only arguably beneficial, then it’s a drug.

That is completely ridiculous. All ‘drugs’ and ‘nutrients’ have different dose-response curves and ranges. Does ‘water’ pass your criterion? What about hormones? Consider the case of melatonin: after 10mg or so, it stops having effects on sleep and instead starts having different effects (for example, being a contraceptive); is melatonin a ‘nutrient’ or a ‘drug’? By your definition, apparently it is a ‘nutrient’ because it has ‘some unarguably beneficial changes’ (sleep is good) and then changes stop (past 10mg). What about drugs that change their effects at different doses, like nicotine switching from stimulant to depressant? Or let’s take iodine: iodization can kill people in previously-deficient populations because of thyroid shock; was it a ‘drug’ for some people and a ‘nutrient’ for others…?

This is not a distinction which cleaves reality at the joints. Status quo bias indeed.

Report comment

A nutrient is something which it is necessary to consume for normal functioning of the body. Lack of a nutrient is a deficiency.

A drug is a substance which alters the normal functioning of the body. Lack of a drug is not a deficiency.

Alexander tried to blur the distinction on the grounds that in both cases you add a substance to the body and cause changes, and that “normal functioning” is arbitrary, so there is no difference between a nutrient (iodine) and a drug (lithium).

My response is that you can tell the difference based on what happens when you administer it. If you administer it, it causes some unarguably beneficial changes, and then the changes stop, then it’s a nutrient, and the point at which the changes stop is the baseline level at which the body functions normally. If you administer it, and it keeps causing changes all the way up to the point of toxicity without there being any level that you can call baseline, and there are harmful changes as well as beneficial ones or the changes are only arguably beneficial, then it’s a drug.

And the fact that nutrients and drugs behave in these different manners is *why* we can trust untrustworthy people to supply nutrients more than we can trust them to supply drugs.

St. Rev: Or in other words, yes it’s about the benefit curves, but nutrients and drugs inherently have different benefit curves.

Report comment