Hacker News new | past | comments | ask | show | jobs | submit login
Optimal Light Exposure for Rapidly Correcting Circadian Misalignment (ploscompbiol.org)
76 points by timr on April 25, 2014 | hide | past | favorite | 15 comments



Anyone with sleep problems owes it to themselves to read Chronotherapy by Terman and McMahan (2012).

Basically, for manipulating your circadian rhythm, light therapy is recommended as the first line of treatment (possibly supplemented with low-dose melatonin). The timing and intensity of light relative to when your internal clock thinks it is in the day determines the amount of slowing or lengthening of your cycle. So bright light during when your body thinks it is 'morning' will slow down the cycle, and bright light at 'night' will lengthen it. That's why many people use f.lux. But how much, how long?

Your cycle's natural free-running cycle length (without zeitgebers, or cues, to entrain it) appears to be determined by genetic factors. This is hard to measure without doing some tracking and, ideally, actually free-running. Pretty inconvenient.

This paper, which I look forward to reading in-depth, claims to have figured out optimal ways [took another look: general principles] of calculating the correct exposure and absence of light to allow for rapid recovery from jet lag. If true, it will probably make all light therapy much easier to do and calculate.


I'd be careful with melatonin. It appears to act like a stress hormone and has a bunch of significant effects. It boosts estrogen and lowers metabolism.


Light therapy is the main thing (seems daylight simulation and negative ions may also work, but this is less researched). The melatonin option is indicated at the level of micrograms, and as far as I know is quite safe. I'm mostly drawing from gwern's analysis and the opinions of several doctors, though, for the safety profile: http://www.gwern.net/Melatonin

Care to share some references? It would be useful to add to the page above.


There's a bunch of easily found stuff on animal experiments where it shrinks gonads and brains. It clearly turns off fertility in animals. It's well understood as the "winter hibernation" hormone. It's not the "good circadian rhythm night of deep sleep" hormone like people seem to think. It rises to adapt the body to the stress of darkness of winter, it doesn't rise to make you get better rest. It makes the retina more sensitive to light and lightens skin. It's high in seasonally depressed people. Taking a sunny vacation lowers it, and makes you feel better.

Nobody has a melatonin deficiency; you don't need melatonin levels to get restful sleep. If stress hormone levels are low you can fall asleep under bright lights with no melatonin.

I can't say taking 5mg every now and then would be harmful to humans, but when you understand melatonin's essential role as the winter hibernation hormone, and consider the animal experiments, I think you might want to prefer other means to get to sleep. Personally, I find diphenhydramine (benadryl) pretty good for occasional help nodding off. And that's very clearly safe for occasional use.


I couldn't find any statements backing up your claim that Melatonin shrinks brains; in fact, I found allegations to the opposite: Melatonin appears to have neuroprotective properties.[0] I suggest you cite your sources, especially when making sweeping claims like "it's well understood as...", "easily found stuff", etc.

[0] http://www.lef.org/magazine/mag2013/sep2013_Melatonin-The-Br...


If "stuff" is so easily found, why don't you reference it?


I am a tad perturbed that a quick skim didn't present to my eyes any statements along the lines of "But we never tried it, at all, so take it with a grain of salt."

Either that or I somehow completely missed the table showing the effect of the schedule on subjects vs. controls, or any other sort of experimental confirmation whatsoever.

To be clear I don't object to the analysis or its methods, I object that it's not being clearly confessed to be an analysis in bold letters.


This just a numerical simulation on two (simple) models of the human sleep cycle system that tries to find a more efficient way of shifting the cycle of those methods.

I've only spend 40 minutes reading the paper so far and only have a superficial understanding of it, but to me it looks like they are using a somewhat oscillator like model and looking to introduce a phase shift in the response by driving it with different input functions.

There is no experimental verification whatsoever. This is why this was in the computational biology section of plosone.


I've been meaning to ask if you've read up on chronotherapeutics; when I read the book (that I cited above, in top-level) a few months ago it seemed to explain why the melatonin procedure that MetaMed gave you worked - a full biological basis. I didn't follow that up, though, since I was trying to fix my own DSPS.


This is quite interesting. As someone who has delayed sleep phase syndrome, I wish I'd known about the mathematical models they mention a few years ago. I had to figure out my own light therapy schedule by trial and error.

The next step, of course, is to actually try it out. It appears there's an iOS app based on this: http://entrain.math.lsa.umich.edu/


I think the important thing is to get as much natural daylight exposure as possible. Sunlight lowers stress hormones, lowers depression, increases metabolism, and so forth. So I'm skeptical about "apps" and algorithms that are about micromanaging how much dim artificial light you get. The much more important thing is getting out in the midday sun for an hour or more every day.


I'm a definite fan of getting mid-day sun, but my understanding is while it'll help maintain sleep rhythm, it won't correct it if it's already out of whack.

For that, I'd suggest getting some sun (or failing that, some light therapy lamp exposure—a special-purpose lamp that puts out a blinding amount of light) first thing when you wake up. Of course, you could optimize further too, like this paper suggests.

Source is admittedly Wikipedia: http://en.wikipedia.org/wiki/Delayed_sleep_phase_disorder#Ma..., and http://en.wikipedia.org/wiki/File:PRC-Light%2BMel.png in particular.


I have occasional problems with sleep disruption in the winter months (October to March, 49 degrees latitude). The problem is that natural sunlight simply isn't strong enough to reset your body clock.

When it happens I take a melatonin at night, and it seems to reset my sleep cycle and gives me a really good night's sleep. I don't normally take it more than once a week at most, just in case there are any negative effects. The delayed release tablets seem to work best (Jamieson, I believe it is), as they slowly release the melatonin over a period of 8 hours.


Sunlight is great, but once your circadian rhythm gets messed up it's not enough. Blue light from artificial sources does suppress melatonin production quite well, which can change your circadian rhythm for better or worse.


In case the main link is not working as it did for me, I spotted an alternate source for the publication [1].

[1] http://www.ncbi.nlm.nih.gov/pmc/articles/PMC3983044/pdf/pcbi...




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: