Whether you're a Tim Ferriss fan or not, he puts out some solid empirical data, particularly from tests he conducts on himself. This one is in regards to sleep, something I know a lot of you are looking to improve.
For me, the problem with improving sleep was being unable to measure it. I could record the time when I got into bed and when I woke up, but I couldn't pinpoint when I fell asleep, much less what happened while I was asleep.
The problem with testing these in a proper sleep lab (the test is called a polysomnogram) is that you generally have at least 22 wires attached to you to measure brain activity (EEG), eye movements (EOG), skeletal muscle activation (EMG), heart rhythm (ECG), respiration, and sometimes peripheral pulse oximetry.
Guess what? No one can sleep in a weird lab with 22 wires attached to them on the first night. So the data are terrible. Then they come in the second night after an effective all-nighter and crash like heroin addicts. Double bad data.
Alas, I would need a pocket-sized sleep lab to test them under realistic sleeping conditions, and I was able to do this recently using the Zeo brain-tracking device, video recording of sleep movements, accelerometers, and more.
Here are some of the most important initial findings, ranking sleep from 1-10, 10 being most restful:
1. 8-10 sleep was most dependent on the ratio of REM-to-total sleep, not total REM duration.
The higher the percentage of REM sleep, the more restful the sleep. The higher the REM percent, the better the recall of skills or data acquired in the previous 24 hours. Higher percent REM sleep also correlated to lower average pulse and temperature upon waking. Based on available studies, I expected deep wave to affect the latter two, but the correlation was erratic.
2. I could increase REM percent by extending total sleep time past 9 hours or waking for 5 minutes at approximately 4.5 hours after sleep onset.
Short wakings of 5-10 minutes, particularly one additional waking approximately 6.5 hours after sleep onset, dramatically increased REM percent. It turns out that waking is not necessarily a bad thing, at least when intentional.
3. 200mcg (micrograms) of huperzine-A 30 minutes pre-bed can increase total REM by 20-30 percent
Huperzine-A, an extract of huperzia serrata, slows the breakdown of the neurotransmitter acetylcholine ( 1). It is a popular nootropic (smart drug), and I have used it in the past to accelerate learning and increase the incidence of lucid dreaming. The increased REM seemingly caused by huperzine-A could explain the increased retention some experience with it. I no longer use huperzine-A except for the first several weeks of language acquisition, no more than three days per week to avoid side-effects, and I do not recommend using it unless you do your homework. Inhibition in the human body usually triggers compensation -- and often delayed side-effects -- somewhere else. The brain is a sensitive instrument, and this drug is contraindicated with a fair number of medications.
4. More than two glasses of wine within four hours of sleep onset decreases deep wave sleep 20-50 percent.
Even four glasses six hours beforehand did not appear to have this effect. Conversely, taking 15+ drops of California Poppy extract appeared to increase deep wave sleep up to 20 percent.
5. Having two tablespoons of organic almond butter (or peanut butter) on celery sticks before bed eliminated at least 50 percent of "feel like shit" 1-3 awakenings.
Ever wonder how you can sleep 8-10 hours and feel tired? Often the culprit is low blood sugar. Make a pre-bed snack part of your nutritional program.
1-2 tablespoons of flaxseed oil (120-240 calories) can be used in combination with the above to further increase cell repair during sleep and thus decrease fatigue. It tastes like a mixture of cat urine and asparagus, so I recommend pinching your nose while consuming it per Dr. Seth Roberts.
Turning Off Monkey Mind
The most important thing, of course, is getting to sleep in the first place. No matter how theoretically restful my sleep should be based on Zeo results, more than 30 minutes of onset insomnia negated it all.
Here are the changes and tools that had the largest and most predictable effects. Some will no doubt be more convenient than others. I excluded drugs from testing to avoid both side-effects and dependencies:
6. 67-70-degree temperature
This was the variable I most experimented with while in Nicaragua for my medical tourism adventure (another story -- I used one hospital trip to pay for a beach vacation), and it was also the variable that had the most consistent effects. Specifically, using a single bed sheet, 67-70-degrees Fahrenheit produced the fastest time to sleep. Warmer temperatures never worked, but as low as 65 would work equally well if I wore socks to keep my feet warm. If you can't control the ambient temperature, testing socks of different thicknesses is the easiest variable to change for tweaking heat loss. No joke.
Ideal temperature is highly individual and a narrow range, so experiment with precise controls.