One curious thing that leaps out looking at the graphs is that the estimated underlying standard deviations differ: the nicotine days have a strikingly large standard deviation, indicating greater variability in scores - both higher and lower, since the means weren’t very different. The difference in standard deviations is just 6.6% below 0, so the difference almost reaches our usual frequentist levels of confidence too, which we can verify by testing:
This is not something you notice when you talk to Seltzer. And though our memory is probably at its peak in our early 20s, few 30-year-olds are aware of a deficit. But Seltzer considers himself a transhumanist, in the mould of the Oxford philosopher Nick Bostrom and the futuristic writer and inventor Ray Kurzweil. Transhumanists are interested in robots, cryogenics and living a really, really long time; they consider biological limitations that the rest of us might accept, or even appreciate, as creaky obstacles to be aggressively surmounted. On the ImmInst (Immortality Institute) forums, Seltzer and other members discuss life-extension strategies and the potential benefits of cognitive enhancers. Some members, Seltzer among them, use a drug called piracetam, which was first marketed by a Belgian pharmaceutical company in 1972 and in recent years has become available in the US from retailers that sell supplements. Although not approved for any use by the FDA, piracetam has been used experimentally on stroke patients - to little effect - and on patients with a rare neurological condition called progressive myoclonus epilepsy, for whom it proved helpful in alleviating muscle spasms. Data on piracetam's benefits for healthy people is virtually nonexistent, but many users believe that the drug increases blood flow to the brain.
A fancier method of imputation would be multiple imputation using, for example, the R library mice (Multivariate Imputation by Chained Equations) (guide), which will try to impute all missing values in a way which mimicks the internal structure of the data and provide several possible datasets to give us an idea of what the underlying data might have looked like, so we can see how our estimates improve with no missingness & how much of the estimate is now due to the imputation:
It looks like the overall picture is that nicotine is absorbed well in the intestines and the colon, but not so well in the stomach; this might be the explanation for the lack of effect, except on the other hand, the specific estimates I see are that 10-20% of the nicotine will be bioavailable in the stomach (as compared to 50%+ for mouth or lungs)… so any of my doses of >5ml should have overcome the poorer bioavailability! But on the gripping hand, these papers are mentioning something about the liver metabolizing nicotine when absorbed through the stomach, so…
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:
Starting from the studies in my meta-analysis, we can try to estimate an upper bound on how big any effect would be, if it actually existed. One of the most promising null results, Southon et al 1994, turns out to be not very informative: if we punch in the number of kids, we find that they needed a large effect size (d=0.81) before they could see anything:
You have probably heard and you already love the term “soul food.” You should know that there’s “brain food” too. Natural supplements are the best way to express your gratitude for all the hard work your brain does for you around the clock. These products aren’t reserved only for the elderly users. On the contrary, if you start using them while you’re still young and sharp, you can ensure the proper protection against all those age-related mental deterioration processes.
Since each 400mg pill takes up 2 00 pills, that’s 4 gel caps a day to reach 800mg magnesium citrate (ie. 136mg elemental magnesium), or 224 gel caps (2x120) for the first batch of Solgar magnesium pills. Turning the Solgar tablets into gel capsules was difficult enough that I switched to NOW Food’s 227g magnesium citrate powder for the second batch.
Seltzer's decision to take piracetam was based on his own online reading, which included medical-journal abstracts. He hadn't consulted a doctor. Since settling on a daily regime of supplements, he had sensed an improvement in his intellectual work and his ability to engage in stimulating conversation. He continued: "I feel I'm better able to articulate my thoughts. I'm sure you've been in the zone - you're having a really exciting debate with somebody, your brain feels alive. I feel that more. But I don't want to say that it's this profound change."
In general, I feel a little bit less alert, but still close to normal. By 6PM, I have a mild headache, but I try out 30 rounds of gbrainy (haven’t played it in months) and am surprised to find that I reach an all-time high; no idea whether this is due to DNB or not, since Gbrainy is very heavily crystallized (half the challenge disappears as you learn how the problems work), but it does indicate I’m not deluding myself about mental ability. (To give a figure: my last score well before I did any DNB was 64, and I was doing well that day; on modafinil, I had a 77.) I figure the headache might be food related, eat, and by 7:30 the headache is pretty much gone and I’m fine up to midnight.
If you have spent any time shopping for memory enhancer pills, you have noticed dozens of products on the market. Each product is advertised to improve memory, concentration, and focus. However, choosing the first product promising results may not produce the desired improvements. Taking the time to research your options and compare products will improve your chances of finding a supplement that works.
Attention Deficit and Hyperactivity Disorder (ADHD) is a condition that relates to a collection of behavioural symptoms such as hyperactivity, impulsiveness and inattentiveness. It is most commonly diagnosed in childhood between the ages of 6 and 12 when disruptive behaviour begins to show, however, due to a growing awareness of the condition, it is also becoming common among adults. According to the thinktank Demos, the cost of undiagnosed ADHD in adults in the UK who are unable to work or hold down a full-time job are estimated to cost billions of pounds to the nation. They warn that too many may be going through life struggling, unable to access the support ot diagnosis they need, which means there could be a huge amount of wasted talent.
I have elsewhere remarked on the apparent lack of benefit to taking multivitamins and the possible harm; so one might well wonder about a specific vitamin like vitamin D. However, a multivitamin is not vitamin D, so it’s no surprise that they might do different things. If a multivitamin had no vitamin D in it, or if it had vitamin D in different doses, or if it had substances which interacted with vitamin D (such as calcium), or if it had substances which had negative effects which outweigh the positive (such as vitamin A?), we could well expect differing results. In this case, all of those are true to varying extents. Some multivitamins I’ve had contained no vitamin D. The last multivitamin I was taking both contains vitamins used in the negative trials and also some calcium; the listed vitamin D dosage was a trivial ~400IU, while I take >10x as much now (5000IU).
Between midnight and 1:36 AM, I do four rounds of n-back: 50/39/30/55%. I then take 1/4th of the pill and have some tea. At roughly 1:30 AM, AngryParsley linked a SF anthology/novel, Fine Structure, which sucked me in for the next 3-4 hours until I finally finished the whole thing. At 5:20 AM, circumstances forced me to go to bed, still having only taken 1/4th of the pill and that determines this particular experiment of sleep; I quickly do some n-back: 29/20/20/54/42. I fall asleep in 13 minutes and sleep for 2:48, for a ZQ of 28 (a full night being ~100). I did not notice anything from that possible modafinil+caffeine interaction. Subjectively upon awakening: I don’t feel great, but I don’t feel like 2-3 hours of sleep either. N-back at 10 AM after breakfast: 25/54/44/38/33. These are not very impressive, but seem normal despite taking the last armodafinil ~9 hours ago; perhaps the 3 hours were enough. Later that day, at 11:30 PM (just before bed): 26/56/47.
Participants (n=205) [young adults aged 18-30 years] were recruited between July 2010 and January 2011, and were randomized to receive either a daily 150 µg (0.15mg) iodine supplement or daily placebo supplement for 32 weeks…After adjusting for baseline cognitive test score, examiner, age, sex, income, and ethnicity, iodine supplementation did not significantly predict 32 week cognitive test scores for Block Design (p=0.385), Digit Span Backward (p=0.474), Matrix Reasoning (p=0.885), Symbol Search (p=0.844), Visual Puzzles (p=0.675), Coding (p=0.858), and Letter-Number Sequencing (p=0.408).
I posted a link to the survey on my Google+ account, and inserted the link at the top of all gwern.net pages; 51 people completed all 11 binary choices (most of them coming from North America & Europe), which seems adequate since the 11 questions are all asking the same question, and 561 responses to one question is quite a few. A few different statistical tests seem applicable: a chi-squared test whether there’s a difference between all the answers, a two-sample test on the averages, and most meaningfully, summing up the responses as a single pair of numbers and doing a binomial test:
Analgesics Anesthetics General Local Anorectics Anti-ADHD agents Antiaddictives Anticonvulsants Antidementia agents Antidepressants Antimigraine agents Antiparkinson agents Antipsychotics Anxiolytics Depressants Entactogens Entheogens Euphoriants Hallucinogens Psychedelics Dissociatives Deliriants Hypnotics/Sedatives Mood Stabilizers Neuroprotectives Nootropics Neurotoxins Orexigenics Serenics Stimulants Wakefulness-promoting agents