Compared with those reporting no use, subjects drinking >4 cups/day of decaffeinated coffee were at increased risk of RA [rheumatoid arthritis] (RR 2.58, 95% CI 1.63-4.06). In contrast, women consuming >3 cups/day of tea displayed a decreased risk of RA (RR 0.39, 95% CI 0.16-0.97) compared with women who never drank tea. Caffeinated coffee and daily caffeine intake were not associated with the development of RA.
My worry about the MP variable is that, plausible or not, it does seem relatively weak against manipulation; other variables I could look at, like arbtt window-tracking of how I spend my computer time, # or size of edits to my files, or spaced repetition performance, would be harder to manipulate. If it’s all due to MP, then if I remove the MP and LLLT variables, and summarize all the other variables with factor analysis into 2 or 3 variables, then I should see no increases in them when I put LLLT back in and look for a correlation between the factors & LLLT with a multivariate regression.
The compound is one of the best brain enhancement supplements that includes memory enhancement and protection against brain aging. Some studies suggest that the compound is an effective treatment for disorders like vascular dementia, Alzheimer’s, brain stroke, anxiety, and depression. However, there are some side effects associated with Alpha GPC, like a headache, heartburn, dizziness, skin rashes, insomnia, and confusion.
One item always of interest to me is sleep; a stimulant is no good if it damages my sleep (unless that’s what it is supposed to do, like modafinil) - anecdotes and research suggest that it does. Over the past few days, my Zeo sleep scores continued to look normal. But that was while not taking nicotine much later than 5 PM. In lieu of a different ml measurer to test my theory that my syringe is misleading me, I decide to more directly test nicotine’s effect on sleep by taking 2ml at 10:30 PM, and go to bed at 12:20; I get a decent ZQ of 94 and I fall asleep in 16 minutes, a bit below my weekly average of 19 minutes. The next day, I take 1ml directly before going to sleep at 12:20; the ZQ is 95 and time to sleep is 14 minutes.
The important factors seem to be: #1/MR6 (Creativity.self.rating, Time.Bitcoin, Time.Backups, Time.Blackmarkets, Gwern.net.linecount.log), #2/MR1 (Time.PDF, Time.Stats), #7/MR7 (Time.Writing, Time.Sysadmin, Time.Programming, Gwern.net.patches.log), and #8/MR8 (Time.States, Time.SRS, Time.Sysadmin, Time.Backups, Time.Blackmarkets). The rest seem to be time-wasting or reflect dual n-back/DNB usage (which is not relevant in the LLLT time period).
A randomized non-blind self-experiment of LLLT 2014-2015 yields a causal effect which is several times smaller than a correlative analysis and non-statistically-significant/very weak Bayesian evidence for a positive effect. This suggests that the earlier result had been driven primarily by reverse causation, and that my LLLT usage has little or no benefits.
What worries me about amphetamine is its addictive potential, and the fact that it can cause stress and anxiety. Research says it’s only slightly likely to cause addiction in people with ADHD,  but we don’t know much about its addictive potential in healthy adults. We all know the addictive potential of methamphetamine, and amphetamine is closely related enough to make me nervous about so many people giving it to their children. Amphetamines cause withdrawal symptoms, so the potential for addiction is there.
Amphetamine – systematic reviews and meta-analyses report that low-dose amphetamine improved cognitive functions (e.g., inhibitory control, episodic memory, working memory, and aspects of attention) in healthy people and in individuals with ADHD. A 2014 systematic review noted that low doses of amphetamine also improved memory consolidation, in turn leading to improved recall of information in non-ADHD youth. It also improves task saliency (motivation to perform a task) and performance on tedious tasks that required a high degree of effort.