70 pairs is 140 blocks; we can drop to 36 pairs or 72 blocks if we accept a power of 0.5/50% chance of reaching significance. (Or we could economize by hoping that the effect size is not 3.5 but maybe twice the pessimistic guess; a d=0.5 at 50% power requires only 12 pairs of 24 blocks.) 70 pairs of blocks of 2 weeks, with 2 pills a day requires (70 \times 2) \times (2 \times 7) \times 2 = 3920 pills. I don’t even have that many empty pills! I have <500; 500 would supply 250 days, which would yield 18 2-week blocks which could give 9 pairs. 9 pairs would give me a power of:
Whole pill at 3 AM. I spend the entire morning and afternoon typing up a transcript of Earth in My Window. I tried taking a nap around 10 AM, but during the hour I was down, I had <5m of light sleep, the Zeo said. After I finished the transcript (~16,600 words with formatting), I was completely pooped and watched a bunch of Mobile Suit Gundam episodes, then I did Mnemosyne. The rest of the night was nothing to write home about either - some reading, movie watching, etc. Next time I will go back to split-doses and avoid typing up 110kB of text. On the positive side, this is the first trial I had available the average daily grade Mnemosyne 2.0 plugin. The daily averages all are 3-point-something (peaking at 3.89 and flooring at 3.59), so just graphing the past 2 weeks, the modafinil day, and recovery days: ▅█▅▆▄▆▄▃▅▄▁▄▄ ▁ ▂▄▄█. Not an impressive performance but there was a previous non-modafinil day just as bad, and I’m not too sure how important a metric this is; I must see whether future trials show similar underperformance. Nights: 11:29; 9:22; 8:25; 8:41.
One thing to notice is that the default case matters a lot. This asymmetry is because you switch decisions in different possible worlds - when you would take Adderall but stop you’re in the world where Adderall doesn’t work, and when you wouldn’t take Adderall but do you’re in the world where Adderall does work (in the perfect information case, at least). One of the ways you can visualize this is that you don’t penalize tests for giving you true negative information, and you reward them for giving you true positive information. (This might be worth a post by itself, and is very Litany of Gendlin.)
These are quite abstract concepts, though. There is a large gap, a grey area in between these concepts and our knowledge of how the brain functions physiologically – and it’s in this grey area that cognitive enhancer development has to operate. Amy Arnsten, Professor of Neurobiology at Yale Medical School, is investigating how the cells in the brain work together to produce our higher cognition and executive function, which she describes as “being able to think about things that aren’t currently stimulating your senses, the fundamentals of abstraction. This involves mental representations of our goals for the future, even if it’s the future in just a few seconds.”
Enhanced learning was also observed in two studies that involved multiple repeated encoding opportunities. Camp-Bruno and Herting (1994) found MPH enhanced summed recall in the Buschke Selective Reminding Test (Buschke, 1973; Buschke & Fuld, 1974) when 1-hr and 2-hr delays were combined, although individually only the 2-hr delay approached significance. Likewise, de Wit, Enggasser, and Richards (2002) found no effect of d-AMP on the Hopkins Verbal Learning Test (Brandt, 1991) after a 25-min delay. Willett (1962) tested rote learning of nonsense syllables with repeated presentations, and his results indicate that d-AMP decreased the number of trials needed to reach criterion.
That it is somewhat valuable is clear if we consider it under another guise. Imagine you received the same salary you do, but paid every day. Accounting systems would incur considerable costs handling daily payments, since they would be making so many more and so much smaller payments, and they would have to know instantly whether you showed up to work that day and all sorts of other details, and the recipients themselves would waste time dealing with all these checks or looking through all the deposits to their account, and any errors would be that much harder to track down. (And conversely, expensive payday loans are strong evidence that for poor people, a bi-weekly payment is much too infrequent.) One might draw a comparison to batching or buffers in computers: by letting data pile up in buffers, the computer can then deal with them in one batch, amortizing overhead over many items rather than incurring the overhead again and again. The downside, of course, is that latency will suffer and performance may drop based on that or the items becoming outdated & useless. The right trade-off will depend on the specifics; one would not expect random buffer-sizes to be optimal, but one would have to test and see what works best.
Chocolate or cocoa powder (Examine.com), contains the stimulants caffeine and the caffeine metabolite theobromine, so it’s not necessarily surprising if cocoa powder was a weak stimulant. It’s also a witch’s brew of chemicals such as polyphenols and flavonoids some of which have been fingered as helpful10, which all adds up to an unclear impact on health (once you control for eating a lot of sugar).

Yes, according to a new policy at Duke University, which says that the “unauthorized use of prescription medicine to enhance academic performance” should be treated as cheating.” And no, according to law professor Nita Farahany, herself based at Duke University, who has called the policy “ill-conceived,” arguing that “banning smart drugs disempowers students from making educated choices for themselves.”
Flow diagram of epidemiology literature search completed July 1, 2010. Search terms were nonmedical use, nonmedical use, misuse, or illicit use, and prescription stimulants, dextroamphetamine, methylphenidate, Ritalin, or Adderall. Stages of subsequent review used the information contained in the titles, abstracts, and articles to determine whether articles reported studies of the extent of nonmedical prescription stimulant use by students and related questions addressed in the present article including students’ motives and frequency of use.
Phenotropil is an over-the-counter supplement similar in structure to Piracetam (and Noopept). This synthetic smart drug has been used to treat stroke, epilepsy and trauma recovery. A 2005 research paper also demonstrated that patients diagnosed with natural lesions or brain tumours see improvements in cognition. Phenylpiracetam intake can also result in minimised feelings of anxiety and depression. This is one of the more powerful unscheduled Nootropics available.
Panax ginseng – A review by the Cochrane Collaboration concluded that "there is a lack of convincing evidence to show a cognitive enhancing effect of Panax ginseng in healthy participants and no high quality evidence about its efficacy in patients with dementia."[36] According to the National Center for Complementary and Integrative Health, "[a]lthough Asian ginseng has been widely studied for a variety of uses, research results to date do not conclusively support health claims associated with the herb."[37]