Smash Notes
The Science & Practice of Perfecting Your Sleep | Dr. Matt Walker
In this episode, my guest is Dr. Matt Walker, Ph.D., Professor of Neuroscience and Psychology and the Founder & Director of the Center for Human Sleep Science at the University of California, Berkeley. He is also the author of the international best-selling book Why We Sleep and the host of "The Matt Walker Podcast." We discuss the biology of sleep…
People
Notes
-
On Huberman Lab, involuntary muscles stay unparalyzed during rapid eye movement sleep, which is one reason men have erections and women have vaginal lubrication. The other reason is autonomic activity. Autonomic storms activate the body enough for erections and vaginal discharge, and paralysis continues through them. Only two voluntary muscle groups are spared: the extraocular muscles and the inner ear muscle.
The autonomic nervous system controls automatic behaviors, including aspects of reproductive function. During a storm, heart rate decelerates and drops and blood pressure goes down. Then, utterly at random and with no regular pattern, heart rate accelerates dramatically as the sympathetic nervous system, the fight or flight branch, fires up and shuts down again. The extraocular muscles stay free so rapid eye movements can still happen. The inner ear muscle was later found to be spared as well. There is no good understanding of why the two groups escape paralysis. A cranial nerve explanation is judged unlikely, with a lean toward something more sensory. Some people argue that eyeballs left inactive for long periods may develop oxygen issues in the aqueous or vitreous humor, so the eyeballs have to keep moving.
-
On Huberman Lab, a night that begins at 3 a.m. rather than at 10:30 p.m. does not follow a fixed program that always opens with the first couple of hours of deep sleep. Sleep mostly moves into rapid eye movement phases and stays significantly deficient in deep sleep.
Some deep sleep still arrives at the start, because a longer time awake creates significantly greater pressure for deep sleep. The same split is an experimental technique. First-half-of-the-night deprivation, then sleep in the second half, leaves a person mostly deep-sleep deprived while they still get mostly all of their rapid eye movement sleep. Restricting sleep to the first four hours yields mostly deep non-rapid eye movement sleep and almost no rapid eye movement sleep.
-
On Huberman Lab, a face-value reading treats sleep as profoundly detrimental. During sleep a person is not finding a mate, not reproducing, not foraging for food, not caring for the young, and is vulnerable to predation. Dr. Matt Walker says sleep probably should have been selected against on any single one among lost mating, lost reproduction, lost foraging, neglected young, or predation risk. Sleep was not selected against.
Sleep has fought its way through every step along the evolutionary path. Dr. Matt Walker says every sleep stage has also survived, as best he can tell, and states outright that the stages are non-negotiable. He is sure vast evolutionary benefits would have followed if Mother Nature had thin-sliced some of the sleep away. He says it looks as though she has not, and he is usually in favor of her wisdom after 3.6 million years.
Each stage stays non-negotiable because the stages survived the evolutionary path. He draws that conclusion from survival of the stages, while the missing thin slice of sleep remains his reading of the record rather than a settled removal.
-
On Huberman Lab, four hours of sleep leave you impaired the next day even when the quality is brilliant, and eight hours of sleep leave you impaired when the quality is very poor. Quantity and quality both have to be present for a beneficial next day. One without the other fails.
Waking up and spending some time awake is normal and natural. The concern begins when you are consciously aware of waking six, seven, or eight times through the night, so sleep is fragmented, or when an awake stretch runs upwards of 25 minutes. Sleep science in the past five or ten years has established that quantity is important and quality is just as important.
Infrequent, short waking does not call for alarm. Frequent waking all through the night, or long stretches upwards of 25 minutes, is the point to look into the pattern.
-
Matt Walker starts workouts around 7:45 or 8 in the morning on a spin bike facing an east window so exercise and daylight arrive together, even though the time he would rather exercise is around 1 in the afternoon. Daylight and exercise are both cues that align the circadian rhythm and reset it each day.
Walker is neither a strong morning type nor a strong evening type. His preference is a workout sometime in the middle of the day, around 1 in the afternoon. He still works out most days at a gym he chose for huge amounts of window exposure facing east, and he starts with cardio on a spin bike facing a window. In California, sunlight usually comes through. On a cloudy day the intensity of the light coming through is still splendid. He matches the session to circadian light exposure rather than to the hour he would pick just to crush a workout, because he also wants to get on with the working day.
The earlier hour stacks movement and morning light, two wakefulness cues, into one session.
-
On Huberman Lab, caffeine latches onto adenosine receptors and blocks them. Caffeine does not activate the receptors. Activating the receptors would dial up sleepiness. Caffeine does the opposite.
Subjective sleepiness usually follows a linear process, or probably sits closer to an exponential. Sampling the brain every couple of minutes and asking how sleepy someone feels has not been done in humans the way animal studies siphon samples. Caffeine comes into the system, competes with adenosine, forces adenosine out of the way, hijacks the receptor by latching onto it, and blocks the receptor. Caffeine does not inactivate the receptor and does not activate the receptor. It functionally inactivates the receptor by taking the receptor out of the game for adenosine.
-
On the show, Dr. Matt Walker refuses to state that CBD is beneficial for sleep, because he says nobody can make that statement now. He still lays out three candidate mechanisms he has been exploring in case CBD ends up potentially beneficial: a thermoregulatory effect, an anxiolytic effect, and a change in adenosine signaling. He calls the mechanisms tentative, says any one of them is viable, and says all three are viable together.
The first is thermoregulatory. In some animal models, CBD created a profile of hypothermia: it cools core body temperature down, and a lower core body temperature is known to be good for sleep. The second is anxiolytic, a reduction in anxiety. He says the anxiety data is quite strong, including recent functional imaging in which the amygdala, an epicenter of emotion deep within the brain, is quietened down with CBD. He calls the anxiety route a second, non-mutually exclusive possibility. The third rests on recent data suggesting CBD can alter the signaling of adenosine. Producing more adenosine is not required. He says CBD can perhaps modulate the sensitivity of the brain so the weight of the same adenosine is weightier as a brain signal and therefore creates stronger pressure for sleep.
The three routes can operate at once: cooling the body, quieting anxiety at the amygdala, and increasing the sleep pressure carried by adenosine.
-
On the show, melatonin usually starts to rise as dusk approaches, hours before sleep. When the system is working in the correct way, melatonin is already on the march for some hours before sleep itself, and the rising peak usually arrives an hour or two later, around the time of sleep, though timing varies from person to person.
Light through the eyes normally acts like a hard brake pedal that stamps down and prevents the release and production of melatonin. Hunter-gatherer tribes untouched by electricity show the unlit pattern: as dusk approaches, the brake fades, the spigot of melatonin opens, and melatonin starts getting released. The rise continues, and the peak usually lands an hour or two later, around sleep.
Melatonin has usually already kicked in before the eyes shut.
-
On Huberman Lab, one speaker is concerned about high melatonin because the hormone has effects on the reproductive axis. He sees people take a hundred, ten, five hundred, sometimes even ten thousand times the amount the body normally releases. He does not want anyone, including himself, to take a hormone known to be androgen suppressive, and he says the concern is not nested in any one specific human study.
The other speaker calls the point a good one and says melatonin is pretty safe on lethality. The first speaker answers that you can take a lot before you die. The second speaker rejects survival as the yardstick: health, not whether the pill will kill you, is the decision for taking it. He then recalls a study he thinks the first speaker mentioned too. He thinks the study looked at at least over 20 brands of melatonin supplements, and he thinks the capsules ranged from 83 percent less than the bottle to 478 percent more than the bottle.
The 10 milligram figure is a separate what-if, not the dose in the brand comparison. If a pill is 10 milligrams and the capsule holds 478 percent more than 10 milligrams, he says 10 milligrams is already many tens of times more than a physiological dose and sits in supraphysiological territory. His conclusion is that people need to be a bit thoughtful.
-
On Huberman Lab, fifty milligrams of apigenin about thirty minutes before sleep went with a subjectively better night. Apigenin is a derivative of chamomile, taken in supplement form. Sleep was not measured, and the person taking it is not a sleep tracker.
Papers on apigenin include control groups, but they are not published in blue ribbon journals, and the show calls the papers somewhat interesting. Chamomile has long been thought of as a mild sedative. The reply on the show is no to chamomile tea and no to taking apigenin. After a look at the sleep data, the show says the evidence is mostly subjective rather than objective, hard sleep measures, so the result is unclear and is not dismissed. Absence of evidence is not evidence of absence.
A personal test on the show would record a supplement-free baseline for a month, then a month on the supplement, using objective metrics from an Oura Ring and subjective notes written in the morning. Both kinds of measures count in sleep research. The negative experiment is to come off for another month and see whether things get worse.
-
On Huberman Lab, sleeping pills are a short-term solution to certain forms of insomnia and are not recommended for the long term. Cognitive behavioral therapy for insomnia, a non-drug psychological approach, is just as effective as sleeping pills and more effective in the long term.
Clinical practice still leaves a possible time and a place for sleeping pills. The long-term record favors cognitive behavioral therapy for insomnia. A recent study found that after working with a therapist, some of the benefits lasted almost a decade. Stopping sleeping pills usually brings rebound insomnia, and sleep goes back to being just as bad, if not worse.
The split is what happens after the intervention ends. Benefits from cognitive behavioral therapy for insomnia can remain for almost a decade after the work with a therapist, while stopping sleeping pills commonly returns sleep to the same poor level or worse.
-
Matt Walker says REM sleep is the only time in the 24-hour period when serotonin and noradrenaline, also called norepinephrine, shut off completely. Both must shut down to produce REM sleep. Serotonin is the same chemical as 5-HT. Acetylcholine ramps up to produce REM sleep. The three neuromodulators run a reciprocal push-pull that generates a natural architecture of sleep through the night.
Raising serotonin dramatically at night, and keeping that rise in place while REM sleep should start, could artificially fragment REM sleep. Matt Walker says he does not know the data and does not think anyone has really got good data. If the data arrived in two years, he would look first at disruption of REM sleep, non-REM sleep, and reciprocal regulation, because serotonin needs to be up at one time and down at another.
On the show, the host agrees and says he is personally never taking tryptophan or serotonin again unless a clinical reason requires it.
-
Matt Walker says NASA pioneered short naps back in the 1990s. During the missions, naps of as little as 26 minutes improved mission performance by 34 percent and improved daytime alertness by 50 percent. The finding birthed the NASA nap culture throughout all terrestrial NASA staff during that time period.
Astronauts tried the naps while missions were underway. A nap of 26 minutes improved mission performance by 34 percent. Daytime alertness improved by 50 percent from naps of that short length. Terrestrial NASA staff then carried the NASA nap culture through the rest of that time period.
Walker adds that the benefit of naps has long been known. The 1990s mission data put a size on the gain: 34 percent better mission performance and 50 percent better daytime alertness from 26 minutes of sleep.
-
Seven to nine hours is the sweet spot for sleep and all-cause mortality. Below seven hours, shorter sleep predicts a shorter life. Past nine hours the mortality curve stops going down, and at ten or eleven hours it hooks back up, almost like a J shape tilted over and reversed.
Sleep science has at least two explanations, and they are not mutually exclusive. The first is that the illness which took their lives was too much for sleep to deal with. When people get sick, an inflammatory, cytokine-mediated response makes them want to sleep more and curl up in bed. They were calling on sleep, the Swiss Army knife of health, but the illness was too powerful for sleep to overcome. The pattern then looks as though more sleep leads to a higher risk of death, while sleep was responding to the mortality risk and lost the battle.
-
Difficulty and anxiety at night in the dark is not the same difficulty and anxiety in the light of day. Thoughts at night come with a magnitude of rumination and catastrophization disproportionate to the description given while awake. Keeping one journal decreased the time it takes to fall asleep by 50 percent, five zero, well on par with any pharmaceutical agent.
The page is for vomiting concerns out and zipping them up before sleep. The host, speaking separately, ties a 3 a.m. waking to writing down things forgotten and still to do, and describes the same list as much more tractable and reasonable after the second half of the night. The journal idea first sounded like hooey, very Berkeley, holding hands and walking home at the end of the day. The speaker does not know what is going on with the brain and thought and emotion at night and has a bunch of theories as to why. Really good studies from good people then found the 50 percent drop.
Being well on par with any pharmaceutical agent is how large the journal effect is. The stated reason for the page is the extra rumination and catastrophization in the dark, so the list goes down before sleep rather than at 3 a.m.
-
On Huberman Lab, the guidance is to remove all clock faces. On a tough night, knowing the time is 3:22 in the morning or 4:48 in the morning does not help in the slightest, and the reading only makes matters worse than better.
The sequence is direct. A person is already having a tough night, then learns the exact minute, either 3:22 or 4:48. The information fails to help at all. It pushes the night in the worse direction rather than the better one. The practical step is to try to remove all clock faces, and some people have found the step helpful. The tip is described as slightly unconventional, set apart from the stock fare of five tips for sleep hygiene.
Taking the faces out of the room removes the moment when a sleeper reads 3:22 or 4:48 and the night gets worse.
Related notes
- How Does Sleep Deprivation Impact Our Immune System
- What Does Caffeine Do to Your Body
- Why Wait Ninety Minutes Before Morning Caffeine?
- Why Is No Caffeine Best for Sleeping
- Why Does Sleeping Less Than Six Hours Not Buy More Useful Time?
Related episodes
- #164 – Andrew Huberman: Sleep, Dreams, Creativity & the Limits of the Human Mind
- #521: Dr. Andrew Huberman — a Neurobiologist on Optimizing Sleep, Performance, and Testosterone
- Exercise, Nutrition, Hormones for Vitality & Longevity | Dr. Peter Attia
- GUEST SERIES | Dr. Andy Galpin: How to Assess & Improve All Aspects of Your Fitness
- Matt Walker: I did whatever I wanted to do
- Sleep is your superpower by Matt Walker
- Elon Musk on DOGE, AI, & Are we in a Simulation? | KMP Ep.18
- What's on Your Mind in That Brain of Yours?