Hacker News Reader

Comment by 000ooo000 | Hacker News Reader

Andrew Huberman of Huberman Lab podcast discusses this in the first 4-5 episodes of the podcast. The takeaway is: it's late-night bright light, rather than blue light, that has effects on the circadian rhythm. It's an appeal to authority but I think I'll go with the professor of neurobiology and opthalmology on this one.

View on HN

(Post author here, huh, it hit HN...)

This is not consistent with the studies I've read. It's a "total lumineous flux in the 460-490nm range" concern. The deep reds, up past 600nm, don't seem to have any impact on melatonin, nor do violets below ~420nm - though they can have other impacts on the retina and such.

Bright light will, all other things being equal, be worse than dim light, but "dim blue light" is still really bad for melatonin levels and sleep.

Some of the older studies did find an impact up in the greens, but they were for fairly short duration exposures - 30 minutes or so. The impact of the longer wavelengths, over a longer term, seems to be muted by [mechanisms somewhere in the system], leaving the blues with the most impact.

You're welcome to trust Huberman, but I'd encourage you to read the papers yourself (I've read dozens and simply picked the most relevant here). There's quite a bit of work on this matter from the past 20-30 years, and they all seem to conclude that "blues matter." Greens and above don't have nearly the same impact, and they don't have the longer term impact. I'll also suggest being careful about conclusions drawn purely from the absorption spectrum of the molecules involved, as the body doesn't seem to react directly to them, but to some skewed version. Unit tests are nice, but end to end tests cover all the details in the process.

Since you seem super duper into this, you could try what I do on many evenings even just as an experiment. I just strap on a very nice LED headlamp that has a pure red LED that I use for "night mode".

That it uses way less power is actually just a side effect of me wanting all the blue lights off at night. I know it's not for everybody, but I find it really relaxing. The trick is to get one that's not a joke lamp from china with a fat markup but a nice headlamp with powerful LEDs. Like this: https://www.blackdiamondequipment.com/en_US/product/spot-400...

The relevant bit being: "Settings include proximity and distance modes, dimming, strobe, red LED night-vision and lock mode"

I also have a couple of 'hurricane' lanterns and I really enjoy the fire-based lighting but even with "clean" kerosene the smell can be unpleasant indoors.

Somewhat likewise I set my Hue lights to red or a dark orange at night; this seems to work well, at least for me.

There's even a home assistant blueprint that is quite nice for matching the external colour temperature for LED lights that support it, it's basically flux for your LED Lights.

https://gist.github.com/HarvsG/cc86381dc99169159d6d4cd870c36...

This plus dimming plus red lights in the bedroom and flux for devices I have found works pretty well.

Nice, thanks!

Ooh the post author! I have a question. When I buy light bulbs they always list the color temperature but not the lumineous flux, what's the relationship?

I buy dim bulbs that have a temperature of 2700K for evening applications precisely because of this belief that brightness and color impacts sleep quality, am I right or is it just placebo? It seems to help...

I think the temperature is describing, in a way, the dominant wavelength (think the peak of a normal-ish curve on the EM spectrum). A lightbulb that appears less blue will inevitably be emitting less blue. But that’s not to say it isn’t emitting across a broad swath of the spectrum and you’re still getting the wavelengths that are being argued as problematic to sleep. You’d need more data or tools to determine just how much of the “offending” wavelengths are making it into your peepers.

Anyways, hopefully a real expert chimes in. :)

Temperature describes a distribution of wavelengths. Higher is bluer, lower is yellower. Both brightness and color matters. If you want to be as safe as possible, buy pure red LEDs for use 1-2 hours before bed and don't expose yourself to any non-red light. The difference is (no pun intended) night and day.

"When I buy light bulbs they always list the color temperature but not the lumineous flux, what's the relationship?"

Hi! LED and LASER module design and manufacturing tech here!

There is a small correlation with LEDs and luminous flux - the ones with higher green peaks (2700K-~4000K CCT) will have a higher luminous flux at full power as green (~552nm) is where the lumen is weighted and those tend to have more output in that converted color range versus the cool white (which have a little less green and a bit more blue.) This really only applies to phosphor-converted LEDs, RGB combo LEDs are narrow-spectrum in emission, and you can't really describe them by luminous flux, rather photon flux or radiant flux per output channel, as the luminous flux will always be highest from the green LED, no matter what, due to the weighting of the lumen in the green wavelengths.

Keyword to search is melanopic spectral efficiency.

This is covered by Huberman. Our night vision is ~3x more sensitive to blue light than red, probably because starlight has more blue. Blue light is ~2.5 worse for keeping us up but would actually be better if turned it way down like the 3x implies we can.

At normal levels the 3x gets lost in the exponential to much light we get at night.

I’m surprised to see him championed here. He pops up on my Twitter feed sometimes and as a scientist myself, he comes off as an absolute quack. Dr. Oz for millennials and zoomers that think they can fix everything with the right supplements, red light therapy, and daily affirmations, with very little real evidence behind it.

A good summary here-

https://www.reddit.com/r/podcasts/comments/rpu1b4/any_opinio...

Not sure I'd qualify my comment as 'championing'. I refer only to his podcast episodes where he discusses topics that one would reasonably expect a professor of neurobiology and opthalmology to have an expert opinion on. It would seem ill-advised to accept as fact his opinions on things like fitness supplements, psychology, or anything else outside his wheelhouse, and while he may be producing content on that now, it seems largely irrelevant to me here.

I don't agree with that assessment at all. I have listened to several hours of his episodes and he discusses supplements, but rarely have I heard him actually recommend any supplement particularly strongly, except maybe Magnesium Threonate if you want to take Magnesium, but that's hardly quack level supplementation.

The impression that I get is that he discusses supplements because the audience wants to know about supplements, and because he experiments with them himself, but I've gotten several excellent and useful pieces of information about diet and sleep regulation and they all involved zero supplements.

Comparing him to Dr Oz is wildly inaccurate, even based on the top comment on the Reddit link you posted (which I still think is inaccurate).

He also did an interview with a person from the NIMH who personally only uses candle light past dark. It makes sense when you think about it.

Also apparently overhead lights are worse- the angle makes a difference.

Candles also supposedly awful for PM 2.5

They are. Any sort of "open flame burner" is likely to emit an awful lot of PM 2.5.

However, most of the lanterns (think your standard "hurricane lantern" or "cold blast lantern") are quite clean burning. I've been using lanterns for a while now, and I have the equipment to measure PM2.5 and PM10 (post author here, you can find reviews of my lanterns and my particulate sensors elsewhere on my blog).

The lanterns are "basically zero" in terms of PM2.5 and PM10 in the exhaust stream, and when you dig into papers talking about kerosene lighting and particulates, they come to the same conclusions - most of the particulate is from the open flame kerosene burners, not the actual lanterns.

I use lanterns in my office in the winter for some extra heat (I'm the "solar shed" guy, winter sucks during inversions), and I see a fairly rapid rise in CO2 levels (about like another 2-3 people in my space), but I don't see high PM2.5 unless I've got an actual open flame candle going.

How about the fake candles?

I went through a period where as soon as I got home and out of the shower I’d go dark. I had a battery powered lamp that I used exclusively. I’d wear an eye mask with something to listen to playing. It definitely helped. I can definitely feel blue light waking me up much more easily than red. (Worked swing shift until midnight at that point.)

Isn't a fake candle just a normal light in a pretty container?

What about pilot lights on a stove/oven? I have an old-school unit in my house that has three pilot lights that are always on... the put off a fair bit of heat. I don't love the idea of them burning constantly from an energy standpoint, but I didn't consider the health effects...

Not nearly as bad as a candle, so long as the flame is burning blue.

Probably quite bad unfortunately. Look into NO2 exposure from stoves.

also increased co2 in closed environment is bad too

(Post author)

Yes, there's an angle component that's a bit concerning as well. It seems that our eyes mostly respond to the blues "above the horizon" - which makes sense given how our world works. Blue on the ground isn't interesting, nor bright, and our bodies don't tend to waste energy building things where they're not useful.

It does imply that our overhead lighting (most homes have overhead lighting in some forms or another) is more concerning than we've broadly considered - screens in the line of sight are important if bright, but the overhead lighting is hitting more of the relevant cells.

It makes more sense to use pure red LEDs if you check the melanopic spectral function, where you see red has practically no effect. It's also less likely to emit PM2.5, CO2, or start a house fire.

Candles produce soot which has much more significant data to indicate that it is detrimental to your health.

Blue light(instead of just lumens) has much less evidence.

Indeed, even after reading several of the studies cited here and elsewhere, this whole idea of blue light > specialized photoreceptors > melatonin production > sleep (not to mention cancer) seems like just correlations stacked on top of each other.

I understand that some people have problems sleeping, but saying blue wavelengths are a general hazard to sleep (not to mention cancer) is absurd.

What's your problem with the links in that chain? We know (or, at least, papers argue...) that all of the linkages are true, and I (post author) link some studies that look at the end-to-end results - blue light vs sleep. With what I consider to be fairly straightforward results.

I write having spent the last few months testing exactly this sort of thing in my life. Nights spent with LED lighting, or LED lit screens, I take a long time to fall asleep. Nights spent on incandescent lighting (lanterns or bulbs), I fall asleep quickly. It matches the results of the papers I've linked, and I'm pretty sure "Reading a book on paper by lantern light" vs "reading the same book on Kobo by frontlight" is the same mental activity.

> "Reading a book on paper by lantern light" vs "reading the same book on Kobo by frontlight" is the same mental activity.

How many times brighter was the light emitted from the ebook compared to plain paper?

We know that not getting sleep is bad, we know that bright lights affect sleep.

For this study to be convincing we'd need a number of things:

1) more than 12 people to take part.

2) Spectra controlled but the same overall lux level in the room. Both booth and ebook emitting/reflecting the same number of lumens.

Yeah those were good episodes, the key takeaway for me was early morning and late everning light, low angle in the sky, the "netflix innoculation".

BTW you're allowed to appeal to authority. The problem is "Appeal to false authority" which means listening to a UFC commentator about vaccination.

Denialists use the term "appeal to authority" when they want to stop you from citing any source, no matter how relevant, to refute their baseless opinions. It's obvious that we must appeal to authority sometimes, because otherwise specialisation couldn't exist and we'd all be back to sewing our own trousers.

Reply on news.ycombinator.com