Since LLLT was so cheap, seemed safe, was interesting, just trying it would involve minimal effort, and it would be a favor to lostfalco, I decided to try it. I purchased off eBay a $13 48 LED illuminator light IR Infrared Night Vision+Power Supply For CCTV. Auto Power-On Sensor, only turn-on when the surrounding is dark. IR LED wavelength: 850nm. Powered by DC 12V 500mA adaptor. It arrived in 4 days, on 7 September 2013. It fits handily in my palm. My cellphone camera verified it worked and emitted infrared - important because there’s no visible light at all (except in complete darkness I can make out a faint red light), no noise, no apparent heat (it took about 30 minutes before the lens or body warmed up noticeably when I left it on a table). This was good since I worried that there would be heat or noise which made blinding impossible; all I had to do was figure out how to randomly turn the power on and I could run blinded self-experiments with it.
Power times prior times benefit minus cost of experimentation: (0.20 \times 0.30 \times 540) - 41 = -9. So the VoI is negative: because my default is that fish oil works and I am taking it, weak information that it doesn’t work isn’t enough. If the power calculation were giving us 40% reliable information, then the chance of learning I should drop fish oil is improved enough to make the experiment worthwhile (going from 20% to 40% switches the value from -$9 to +$23.8).
Remembering what Wedrifid told me, I decided to start with a quarter of a piece (~1mg). The gum was pretty tasteless, which ought to make blinding easier. The effects were noticeable around 10 minutes - greater energy verging on jitteriness, much faster typing, and apparent general quickening of thought. Like a more pleasant caffeine. While testing my typing speed in Amphetype, my speed seemed to go up >=5 WPM, even after the time penalties for correcting the increased mistakes; I also did twice the usual number without feeling especially tired. A second dose was similar, and the third dose was at 10 PM before playing Ninja Gaiden II seemed to stop the usual exhaustion I feel after playing through a level or so. (It’s a tough game, which I have yet to master like Ninja Gaiden Black.) Returning to the previous concern about sleep problems, though I went to bed at 11:45 PM, it still took 28 minutes to fall sleep (compared to my more usual 10-20 minute range); the next day I use 2mg from 7-8PM while driving, going to bed at midnight, where my sleep latency is a more reasonable 14 minutes. I then skipped for 3 days to see whether any cravings would pop up (they didn’t). I subsequently used 1mg every few days for driving or Ninja Gaiden II, and while there were no cravings or other side-effects, the stimulation definitely seemed to get weaker - benefits seemed to still exist, but I could no longer describe any considerable energy or jitteriness.
Is 200 enough? There are no canned power functions for the ordinal logistic regression I would be using, so the standard advice is to estimate power by simulation: generating thousands of new datasets where we know by construction that the binary magnesium variable increases MP by 0.27 (such as by bootstrapping the original Noopept experiment’s data), and seeing how often in this collection the cutoff of statistical-significance is passed when the usual analysis is done (background: CrossValidated or Power Analysis and Sample Size Estimation using Bootstrap). In this case, we leave alpha at 0.05, reuse the Noopept experiment’s data with its Magtein correlation, and ask for the power when n=200
Nootropics (/noʊ.əˈtrɒpɪks/ noh-ə-TROP-iks) (colloquial: smart drugs and cognitive enhancers) are drugs, supplements, and other substances that may improve cognitive function, particularly executive functions, memory, creativity, or motivation, in healthy individuals. While many substances are purported to improve cognition, research is at a preliminary stage as of 2018, and the effects of the majority of these agents are not fully determined.