Questioning REM and Other Thoughts on Sleep

“Awesome, another 87%.” I was way too excited to share my ‘sleep score’ with my girlfriend Natalie, who rolled her eyes and probably thought about how long I’d be raving about this new gadget.

The app I was talking about was Sense, one of a myriad of new technology trackers that use accelerometers to track how well we slept the night before. Whether in the new Fitbit or in Kickstarter projects like Sense, there is a clear trend to get more efficient at sleeping by analyzing the data that today costs a fraction of what it did just 5 years ago.

IMG_0090 IMG_0091

While these apps have different grading schemes and metrics for judging sleep, their main determinant of a good night sleep is how much ‘deep sleep’ we get. The thought is that in human sleep cycles, we enter different periods of sleep, and the most ‘rejuvenating’ of them is REM (rapid eye movement) sleep. REM sleep is also when we toss and turn the least

The Quest for REM

Being active in the Quantified Self movement, I began to hear the same stories over and over. How can we maximize REM sleep? I began taking a teaspoon of honey before bed, based on the Seth Roberts theory that the simple glucose would provide the brain with nutrients to sustain better REM sleep. I began taking 15 drops of California Poppy Seed extract before bed because Tim Ferriss found it increased his REM sleep by 20%. The obsession was clear with everyone I spoke to: minimize tossing and turning and focus on the deep REM sleep. But then I came across a study* that shook my foundation. The study, done in 2004, replicated motor learning and studied the effect that sleep had on how humans form memories. What they found was very contradictory to what most had expected: the light, non-rem stages of sleep were actually the most important to consolidating memories. Therefore, the ‘tossing and turning’ ‘kind of half-awake’ sleep was actually some of the most important sleep to get. Damn!


The first lesson learned is to realize how early we are in the quantifying-sleep movement. Most apps today use heart rate and accelerometers to show you effectively how much you moved around during the night. But, as science is beginning to show, just because you were, still doesn’t really mean you slept better. It’s just too subjective. With how much we still don’t know about the complex beast that is sleep, the only safe bet is to assume that the more sleep cycles we go through, the better our memories will form and the better we will feel.

There is no magic number to how many hours you need. As someone who sets his own schedule, I’ve found that I need between 9-10.5 hours to function at 100%. Is that the best news? No. I’d love to think I could get away with 5-6 hours and become 2x productive while everyone else is sleeping. But it’s the truth. I’ve also come to learn just how much the amount of sleep you need varies based on the stressors you put on your body. I require about an extra hour of sleep the night after I’ve done some vigorous activity (weightlifting, hiking, etc).

Another takeaway is how valuable sleep is to memory formation. The single best piece of advice on learning and sleep: Practice or study your new craft before bed each night and the memory will compound in strength. When learning the piano, I had a keyboard next to my bed that I would jam on for just 10 minutes before bed each night. When I woke up, the motions were effortless and my brain had done all the heavy lifting. Must have been from all the tossing and turning…

* Stickgold, R. & Walker, M.P. (2004). Sleep-dependent learning and memory consolidation. Neuron, 44, 121-133.

Leave A Comment

Your email address will not be published. Required fields are marked *