[guzey]

I'm curious - what do you believe I should've done? How do you discredit a p-hacked small-n experiment, aside from noting that it should not be trusted a priori due to the extreme bias of the results expected from such kind of studies?

[RandomThrow321]

Is all sleep science based p-hacked small-n experiments?

To your point - I probably would have tried to rely on research that supports your points (which you've got me interested in now). I probably wouldn't have included a handful of anecdotes in the the appendix (including your own and Elon Musk's!), as well as 8 replies from a reddit thread as a supporting section. I probably also wouldn't equate "a person with an ADRB1 mutation can sleep less" and "a single individual that underwent brain surgery can sleep less" with "Decreasing sleep by 1-2 hours a night in the long-term has no negative health effects". I wouldn't include any arguments that say modern sleep is "unnatural", which doesn't have any real meaning or basis in reality (is modern medicine natural? what about sanitation?). The analogy to hunger is a justification rather than any type of proof, and taking the analogy further, it would suggest I should go back to sleep in the morning since I usually wake up sleepy, just as I would eat more when I'm hungry. I would be careful about saying sleep duration is a cause of depression/mania rather than considering both might be driven by a confounding variable (stimulants will certainly cause both mania and wakefulness!). I also probably wouldn't make claims like:

> Convincing a million 20-year-olds to sleep an unnecessary hour a day is equivalent, in terms of their hours of wakefulness, to killing 62,500 of them.

Without considering that you might be wrong about lifespan (not to mention healthspan) since you might very well be convincing others to effect a behavior change with your post.

[guzey]

>I probably also wouldn't equate "a person with an ADRB1 mutation can sleep less" and "a single individual that underwent brain surgery can sleep less" with "Decreasing sleep by 1-2 hours a night in the long-term has no negative health effects".

I made 5 points in that section (https://guzey.com/theses-on-sleep/#decreasing-sleep-by-1-2-h...):

> 1. A sleep researcher who trains sailors to sleep efficiently in order to maximize their race performance believes that 4.5-5.5 hours of sleep is fine.

> 2. 70% of 84 hunter-gatherers studied in 2013 slept less than 7 hours per day, with 46% sleeping less than 6 hours.

> 3. A single-point mutation can decrease the amount of required sleep by 2 hours, with no negative side-effects.

> 4. A brain surgery can decrease the amount of sleep required by 3 hours, with no negative-side effects.

> 5. Sleep is not required for memory consolidation.

You cited (3) and (4) but ignored (1), (2), and (5) all of which are based on studying dozens and hundreds of people.

[jafitc]

“This guy fucks”