Your Top Health Questions Answered

This episode is a solo installment with no guest, but it is full of outside voices. The host, who introduces himself as Andrew Huberman, a professor of neurobiology and ophthalmology at Stanford School of Medicine, explains that listeners were invited on social media to record questions. He says he is hearing each question for the first time as it plays and answering in real time. The focus is science and health, with topics he expects to span fitness, mental health and supplementation for women, men and people of all ages.

He notes that far more questions came in than one episode can hold. A question left out today may still appear in a future Q&A, and he says the full set of submissions shows what listeners care about and will shape future guest episodes, solo episodes and Q&As.

The first caller, who gives his name as Anthony Richards, trains late because of work and usually goes to bed about an hour and a half after finishing. He asks for the most effective pre-workout to support strength training without impairing sleep. The answer starts by reframing the problem. Strength and muscle-size training is likely to be moderate to high intensity, so it will raise cortisol well above its late-day baseline. The host acknowledges that people react to that with alarm about a "stress hormone," but answers "Yes and no." Cortisol should ideally peak in the morning, but for someone who cannot train early, the real question is how to get enough alertness and autonomic arousal to produce the intensity that drives the cortisol rise.

We don't want to avoid that high cortisol level during these workouts. That's part of the stimulus that's going to induce the adaptations that we want.

Caffeine is ruled out. The host repeats his position that caffeine within the eight to 10 hours before sleep disrupts sleep, even if a person can still fall asleep. He then names what he calls perhaps the only option he knows of that raises alertness and focus for training while also helping sleep: raising acetylcholine. The most direct and safest over-the-counter route he is aware of is Alpha-GPC, at 600 to 900 milligrams taken with water.

His case for Alpha-GPC rests on its relationship to REM sleep. He describes it as one of the few supplements he knows of that can increase REM sleep. He ties REM to consolidating learning from preceding days and to removing the emotional load of several days of memories. Too little REM, he says, tends to leave people feeling more emotional. His claim is measured: Alpha-GPC "has been shown to modestly increase REM sleep," and even if it does not do that for a given person, he says it will not reduce REM or slow-wave deep sleep.

He hesitates to call it a stimulant. Caffeine and other stimulants that raise the catecholamines dopamine, epinephrine and norepinephrine are sympathomimetics, meaning they increase activity in the sympathetic arm of the autonomic nervous system. Alpha-GPC instead works by increasing output of the acetylcholine pathway, which is involved in focus and to some extent energy. He says it either does not seem to hinder REM or can slightly increase the share of REM, which makes it a strong tool, though "it's not a perfect tool."

The limitation he raises is cardiovascular. Raising choline can, after several biochemical steps, increase circulating TMAO. Fish, meat and anything else that raises choline can do this, but he says Alpha-GPC does it fairly potently. Consistently high TMAO over time carries some cardiovascular risk. He cites one paper from a few years earlier that suggested people taking Alpha-GPC regularly at high doses, 900 milligrams several times a day over long periods, may have a higher risk of stroke and other cardiovascular insults. He frames this as something to pay attention to, not as settled.

His mitigation is garlic. He says some literature suggests taking 600 milligrams of garlic within the same 24-hour period as Alpha-GPC, ideally together. The rationale is that garlic contains allicin, which can inhibit some TMAO production, much of which happens in the gut. He sets aside the full production pathway for lack of time. He adds that garlic capsules are inexpensive and do not cause garlicky breath, and that some regular garlic intake is probably a good idea anyway. His suggestion is to take 600 milligrams with every Alpha-GPC dose, or within the following 12 to 24 hours if that is not possible. He notes he does not know how many days a week the caller trains.

He also expects the effect to fade with frequent use, such as daily or more than three or four times a week. So he recommends saving Alpha-GPC for once or twice a week, three at most. The goal is to reach the needed intensity in most sessions without any supplement.

The non-supplement route starts with arriving at the session relatively rested, which means proper sleep the night before. The host argues that people "vastly underestimate" how much getting into the right frame of mind supports intense training. He contrasts that with the common habit of going to the gym and looking at a phone. His advice is to make workouts your own and put everything you can into them mentally and physically. Alpha-GPC on top of that is "doubling up these tools."

He later calls this "the main one." He describes training in Spain with six-time Mr. Olympia Dorian Yates, who had him ride a stationary bike in the corner for three or four minutes before the session. In the host's account, that warm-up was as much about warming up the mind and focusing on the day's work as about raising core body temperature. He prefers the word "training" to "working out" because each session is practice aimed at getting better, and he notes this preparation "doesn't require taking anything at all."

The second half of the answer covers sleeping well after a late workout, whatever was taken beforehand. The first step is a hot shower, not necessarily right away. The mechanism he gives is that warming the body's surface triggers neurons in the medial preoptic area of the hypothalamus, which act as a thermostat for core temperature and help drive it down. He calls a falling core temperature a prerequisite for sleep. He contrasts this with a cold shower, which he calls stimulating.

The second step is long-exhale breathing, such as the physiological sigh, a double inhale followed by a long exhale, during or around the shower. He says a deliberate, drawn-out exhale activates a descending portion of the vagus nerve. By his description, about 85% of the vagus pathway carries sensory information up from the body. The other 10% to 15%, a figure he says no one knows exactly, runs from the brain down to the organs, including the heart, and exhaling slows the heart rate. Three or four such breaths after training are meant to lower autonomic activation, and he says cortisol levels follow. His claim is that even after a hard session triples or quadruples cortisol, it can be brought down deliberately.

The third step pushes back on a rule he says he has repeated himself: avoid eating two or three hours before sleep because it raises heart rate and nighttime and morning glucose. Again his answer is "yes and no." Waking up hungry after a hard workout also disrupts sleep. He describes doing this himself, waking at 3:00 a.m. to eat oatmeal before falling back asleep. For late moderate-to-high-intensity sessions, he recommends eating something afterward. Options he lists include slower-digesting milk proteins such as casein, Greek yogurt, casein protein powder, or oatmeal and other starch to replenish glycogen, even in the final hour before sleep. He says this will not "completely nuke" growth hormone during sleep, since the workout already raised it sharply. For people who are not training late, he still favors a buffer between the last full meal and sleep, while saying that skipping it occasionally is "not the end of the world."

He closes this answer by lowering the bar. No one does all of these things perfectly week after week, and doing some of them should still make a significant difference. He presents these as the tools he knows to be most potent for this question.