CEOInterviews.AI
Start App
Matt Walker
Professor of Neuroscience and Psychology, University of California, Berkeley

The BEST Scientific Sleep Advice Doctors Don't Tell You

📅 Nov 28, 2025 The Diary Of A CEO Clips 21 MIN 597305 VIEWS 66 SEGMENTS · 2 SPEAKERS
World-leading sleep expert Dr Matthew Walker reveals the real causes of modern sleep problems and why millions are unknowingly making their nights worse. After scanning thousands of brains and studying sleep for more than 20 years, Dr Walker breaks down the science behind insomnia, digital habits, melatonin, stress, and the surprising mistakes that quietly destroy your sleep quality. Discover: • The number one behaviour ruining your sleep without you realising • Why most people misuse melatonin and the safe dose you should know • The real effect of phones, screens, and social media at night •...

Questions asked in this interview

12
  1. 0:00Sleep is so important, isn't it?
  2. 0:33But what is the new information or advice that we can give them that will help them once and for all sleep eight hours a night and get their life back?
  3. 1:09Digital detox. Okay, what does that mean?
  4. 1:33But it's also good to like and subscribe, would you say?
  5. 2:44When was that? A while ago?
  6. 5:18So what dose of melatonin should I be taking?
  7. 5:26And do you advise melatonin for people?
  8. 7:33Is there — there's not a way to test, is there?
  9. 8:16Are you concerned that melatonin is becoming more and more popular as a way to solve lifestyle issues that have caused sleep impairment?
  10. 12:04There's always a trade-off, and some of my friends often like talk to me about like these miracle things or this thing or take metformin and everything will be fine and dandy, but what's the trade-off?
  11. 19:06Did evolution not give parents any leeway or anything when they have kids that suddenly their brain changes and now they can survive with less sleep?
  12. 20:02Did you see that documentary about the octopus?
Steven Bartlett 0:00 ↗
Sleep is so important, isn't it? And I think everybody now knows that because of people like yourself, or really, you know, you were the real pioneer in pushing the subject into the world. I remember when you first went on Joe Rogan's show, the amount of my friends that sent me that episode and then sent me your book was just staggering. It was like that was the catalyst moment, I think, in society for the mainstream to really start talking and caring about sleep. However, there's still so many people struggling because of all the factors that you talked about when we're talking about group two who have those lifestyle factors, the stress, the psychological factors, the trauma, etc., etc.
What can we give them, those people, the people that really, really struggle? Maybe they haven't got a disorder. We'll come to the people with sort of sleep disorders later and we'll talk about optimizers. But what is the new information or advice that we can give them that will help them once and for all sleep eight hours a night and get their life back?
Matt Walker 0:54 ↗
If you were to push me to say what are the three most impactful things that you can start doing tonight to start sleeping better, it would be the following: digital detox.
Steven Bartlett 1:09 ↗
Okay, let me write this down. Digital detox. Okay, what does that mean?
Matt Walker 1:12 ↗
One hour before bed, try to limit activating social media engagement, email, and text messages that are going to trigger you.
It's fine to listen to a podcast, you know.
Steven Bartlett 1:28 ↗
Thank God for that.
Matt Walker 1:30 ↗
You heard it here, folks.
Steven Bartlett 1:32 ↗
Right from that book.
But it's also good to like and subscribe, would you say? I would say that, you know, like and subscribe and then just click here so that you get the latest notification, you get that bell, just click on the bell icon so that you...
Matt Walker 1:46 ↗
So it's not a problem of blue light. A quick aside: we've been taught this myth of the blue light effect from devices, and it really is a myth because an incredible Australian researcher, a guy called Michael Gradazar, has almost single-handedly changed what I think of as being the zeitgeist for a while after a very influential paper, which is a great paper. And what they showed was that one hour of iPad reading before bed ended up impairing your melatonin, it disrupted sleep, it reduced the amount of dream sleep, and even after they stopped reading the iPad, the blast radius impact on your dream sleep lasted a week. It's almost like the drug needed to get washed out the system, that even when you'd stop using technology for an hour before bed, the impact of that technology, even though you'd stopped using it, could still be seen in the echo of sleep disruption for a week later. It was a very influential study in a very prestigious journal.
Steven Bartlett 2:44 ↗
When was that? A while ago?
Matt Walker 2:45 ↗
That was probably about 10 years ago. But then Michael Gradazar, this incredible Australian researcher, started to say, 'Well, I can't replicate these findings.' And what he was discovering is that it's not the blue light that's the problem. Now, the blue light will change aspects of your melatonin. And melatonin is a hormone. It simply tells your brain and your body when it's nighttime, when it's time to fall asleep. It doesn't participate in the generation of sleep. Melatonin is like the starting official at the 100-meter race. It brings all of the racers to the line and it begins the timing of the race, but it doesn't participate in the creation of the race itself. That's a different set of chemicals.
It doesn't make you go to sleep. It doesn't make you... And if you look at what we call meta-analyses, where we gather together all of the individual studies on a topic and we put them in a big statistical bucket, what they found is that melatonin will only improve the speed with which you fall asleep by about 3.4 minutes, and it will only increase the efficiency of your sleep by about 2.2%. So not much more than placebo. So melatonin is... it's now... the placebo effect is the most reliable effect in all of pharmacology. So, you know, maybe no harm, no foul, I would say. And don't forget, I will come back to the three, but when it comes to melatonin, be careful. More isn't always better. And you run the risk of confusing your morning brain into a dense nighttime fog. And what I mean by that is 10 milligrams or 20 milligrams of melatonin is what we call a supraphysiological dose, which again is just a fancy medical term to say it's a size of magnitude of melatonin that your body would never naturally release. It's far greater. So melatonin is the signal of darkness. And normally by the morning hours, our natural release of melatonin has stopped and you're down to zero levels again...
Steven Bartlett 4:39 ↗
In the morning.
Matt Walker 4:39 ↗
In the morning. So you wake up and your body no longer has the signal of melatonin saying it's night, it's night, it's night. But if you've dosed yourself with 10 milligrams or 20 milligrams of melatonin, you run the risk of saying, 'Well, yes, I know it quote-unquote knocks me out.' But the problem is in the first, you know, three or four hours of the waking morning, you're struggling because you're in this fog of a hormonal melatonin, a hormonal signal saying it's still pitch black. No, it's not. It's bright light outside, but your body is fooled into thinking it's pitch black because you've dosed yourself too high. And no wonder you're reaching for two or three cups of coffee in the morning.
Steven Bartlett 5:18 ↗
So what dose of melatonin should I be taking?
Matt Walker 5:21 ↗
Somewhere between probably about 0.1 to 3 milligrams.
Steven Bartlett 5:26 ↗
And do you advise melatonin for people?
Matt Walker 5:29 ↗
Yes, in two conditions — or circumstances, I should say. The first is when you're going through jet lag. Wonderfully helpful. But timing is critical. You need to create that sort of artificial signal of night because let's say that you and I here in Los Angeles, we're both going to fly back to London tomorrow and London is eight hours ahead. So we fly overnight, we arrive in London, and then that first night, let's say we decide to go to bed — or I decide to go to bed in the hotel at midnight. The problem is here in Los Angeles, in my body clock, it's still 4:00 p.m. because London's eight hours ahead. So my melatonin is not going to rise for probably another six or seven hours. So I need to artificially hijack my melatonin system and tell my brain, 'Oh no, it's not 4:00 p.m., it's instead midnight.' And so there, under conditions of jet lag, very helpful. The second is if you have a circadian rhythm disorder. Let's say that you're someone who has an advanced circadian phase. What that means is you're someone who really can't get sleepy until three or four in the morning and you would prefer to be sleeping throughout most of the day. So you're almost nocturnal.
Steven Bartlett 6:45 ↗
Is that a genetic disorder?
Matt Walker 6:46 ↗
And yes, it's a genetic disorder in part.
Steven Bartlett 6:49 ↗
How many people have that?
Matt Walker 6:51 ↗
Probably one to two percent of the population have a very severe, what we call an advanced circadian phase disorder. But their melatonin can also be helpful because, once again, part of their problem is that their melatonin is very delayed, so they don't get the signal of 'Oh, it's night' until maybe 4:00 in the morning. You and I, we start to get our signal of melatonin, depending on our chronotype, by somewhere between about 9, 10, or 11 p.m. They may be delayed by five hours. So if we can give them melatonin, we can artificially try to fool their brain into thinking it's actually earlier in their biological rhythm, so they sleep earlier and they're more in sync with the rest of society.
Steven Bartlett 7:33 ↗
How does someone know if they're that type of person, if they have that disorder? Is there — there's not a way to test, is there?
Matt Walker 7:38 ↗
Well, what we typically do is we will bring you into a laboratory and we will measure your innate level. So we will shut out sort of all windows, all clock faces are gone, and we just let you run your natural rhythm.
And they have the same rise and fall in melatonin just like you and I do, except where it's doing that rise and fall of melatonin on the 24-hour clock face is radically different.
Steven Bartlett 8:02 ↗
Okay.
Matt Walker 8:02 ↗
For you and I, it's, you know, 9, 10 p.m., 11:00 p.m. at night when we're starting that melatonin crescendo.
For them, it's 4:00 in the morning. So we can measure it. It's not in their minds. It's not their choice. It's a biological edict.
Steven Bartlett 8:16 ↗
Are you concerned that melatonin is becoming more and more popular as a way to solve lifestyle issues that have caused sleep impairment? Because I'm seeing loads of — you know, I'm an investor, so I see lots of companies now pitching me different products that have melatonin in them as a sort of day-to-day sleeping supplement.
Matt Walker 8:37 ↗
I think I'm really torn. I've been on both sides of this argument and I'm cautious about it for two reasons. The first is in pediatric populations here in the United States.
Steven Bartlett 8:48 ↗
Pediatric.

28 more exchanges in this transcript

Sign in free to read the rest of this interview. No card required.

Sign in to read the full transcript

Cite this transcript

APA, MLA, BibTeX
APA

Walker, M. (2025, November 28). The BEST Scientific Sleep Advice Doctors Don't Tell You [Interview transcript]. The Diary Of A CEO Clips. CEOInterviews.AI. https://ceointerviews.ai/interview/3277260/

MLA

Matt Walker. "The BEST Scientific Sleep Advice Doctors Don't Tell You." The Diary Of A CEO Clips, 28 Nov. 2025. Transcript, CEOInterviews.AI, https://ceointerviews.ai/interview/3277260/.

BibTeX
@misc{walker2025_3277260,
  author       = {Matt Walker},
  title        = {The BEST Scientific Sleep Advice Doctors Don't Tell You},
  howpublished = {Interview transcript, The Diary Of A CEO Clips. CEOInterviews.AI},
  year         = {2025},
  month        = {nov},
  url          = {https://ceointerviews.ai/interview/3277260/},
  note         = {Speaker-attributed transcript with timestamps}
}