Melatonin & me

published by Gwern Branwen on Thu, 06/16/2011 - 11:42
English

Hi everyone. I've posted screenshots of my graphs and the CSV for the last 5 months or so; I spent much of it alternating days using 1.5mg of melatonin. The results seem to be that it decreased my sleep need by ~1 hour, didn't affect length of deep sleep, and reduced length of REM sleep. See http://www.gwern.net/Zeo#melat.....n-analysis

 

If anyone wants to do a better analysis with the CSV or share their own data series, that'd be interesting.

Hypnagog

Fri, 06/17/2011 - 09:11

I've found that with varying doses of melatonin that I wake up earlier than I would normally too.

An evening dose of anything from .25mg to 3mg will have me waking up around 4am, reaching for another dose to stay asleep until a decent time.

...and like you I also found a decrease in REM sleep, contrary to what a lot of chatter on various forums says.

 

The one thing that it did do for me (which was a big factor in trying it) was to increase the frequency of sleep paralysis.

Gwern Branwen

Mon, 06/20/2011 - 11:49

It *increased* sleep paralysis? But that sounds like a reason not to try it...

Tneshen

Wed, 08/08/2012 - 01:50

We are currently working on a college paper and we have been exploring your blog for a few hours. Thank you for your post it proved helpful for us. coding jobs

Gwern Branwen

Wed, 02/29/2012 - 19:17

Update: I recently learned how to use R for very simple statistical analyses, and I've added basic p-values for all effects and an example of how to correct for multiple comparisons.

lacespace

Tue, 03/06/2012 - 17:28

I take a very small dose and have always had very vivid dreams, mostly unpleasant ones. So my mother bought me some melatonin and it makes me feel groggy in the morning. Sounds like this might be common. Also instead of drugging myself to sleep I would rather be able to control my dreams. I feel fortunate that I remember my dreams but for now it is discouraging because they are always negative. I need to learn more.

Hypnagog

Mon, 06/20/2011 - 23:41

gwern0 said:

It *increased* sleep paralysis? But that sounds like a reason not to try it...

For me, that's the reason to try it.

I have sleep paralysis spontaneously a couple of times a month. I have done for years. It's only recently since I've begun exploring it and not trying to "escape" from it I have been able to turn the episodes into lucid dreams and sometimes to OOBEs.

 

Although I use Zeo most nights, I don't always record the live data, so I want to increase the chances of an episode occurring on the nights that I am recording, also giving me more episodes to explore. It's a fascinating "world".

Heather Stoker

Sat, 06/25/2011 - 09:46

I didn't read through your entire link so you may have come to this conclusion on your own.  Most REM sleep is in the later morning hours of sleep.  If you wake earlier on melatonin that is the reason your REM numbers came down.  It's really that simple.  Also, just because melatonin causes you to wake earlier, it doesn't mean that you need less sleep.  It just means that you are waking up earlier and possibly losing out on some good REM sleep by taking it.  That said, you are probably a better judge of your mood and well-being and I suppose only you can decide if that reading is accurate or not.

Paradise

Sat, 06/25/2011 - 17:14

Read your link, gwern0, and found it fascinating. I can't take melatonin as it gives me what I describe as a physical depression the next morning...overwhelming, physically heavy exhaustion and have to go back to bed 'til it's out of my system. I'm so envious of the amount of your deep and REM sleep, melatonin or not! I seem to have better luck in keeping the headband on, tho'. It's come off only twice in 5 months.

Gwern Branwen

Sun, 06/26/2011 - 06:14

spazzyjanet said:

Most REM sleep is in the later morning hours of sleep.  If you wake earlier on melatonin that is the reason your REM numbers came down.  It's really that simple.

 

Well, maybe. My total fell by 12% (456 minutes to 402 minutes), while my REM fell 20% (145 to 116 minutes). Not sure how one would investigate that in more detail.

 

Paradise said:

Read your link, gwern0, and found it fascinating. I can't take melatonin as it gives me what I describe as a physical depression the next morning...overwhelming, physically heavy exhaustion and have to go back to bed 'til it's out of my system. I'm so envious of the amount of your deep and REM sleep, melatonin or not! I seem to have better luck in keeping the headband on, tho'. It's come off only twice in 5 months. 

That reminds me of the time I took a big melatonin overdose; doses can range quite a bit - some people seem to do best on a dose 1/15th mine.

As for keeping the headband on, I continue to have problems with that. I just lost a night of data (running the standing-on-one-leg experiment now) to it slipping off again.

Paradise

Sun, 06/26/2011 - 22:36

gwern0 said:

That reminds me of the time I took a big melatonin overdose; doses can range quite a bit - some people seem to do best on a dose 1/15th mine.

As for keeping the headband on, I continue to have problems with that. I just lost a night of data (running the standing-on-one-leg experiment now) to it slipping off again.

If you don't mind me asking, how much was that overdose and what's your normal dose? I tried taking 1 mg!

A friend of mine just returned her Zeo because she couldn't keep the headband on. I honestly don't know why it hasn't been a problem for me especially since I also wear a sleep mask and I don't let it overlap the headband. I look like I'm ready for Halloween every night!

I'll look forward to hearing the results of your one leg experiment.

Gwern Branwen

Mon, 06/27/2011 - 03:47

My normal is 1.5mg, and no more than 3mg; the overdose was 9mg.

Heather Stoker

Tue, 06/28/2011 - 00:33

If you go into the Zeo cause and effect tab you can compare "time in REM" to "rise time" and it will show you the minutes of REM based on the time you woke up.  For me, REM almost always is lower with earlier wake times and goes up with each later wake block.  I know this doesn't guarantee a direct correlation (especially if you believe both factors are caused directly by melatonin use) but it's pretty telling.

Gwern Branwen

Tue, 06/28/2011 - 05:39

Besides melatonin, there's an obvious confound there: the alarm clock. If you wake up to it and then go back to sleep, then by the usual effect exploited in WILD lucid dreaming, you're going to have disproportionately much REM sleep. And this is true at any time after dawn where light might wake you up temporarily or at least begin reducing melatonin levels.

Adam Bell

Mon, 07/04/2011 - 07:57

Gwern, I love the experiments you are doing!  I would loveto see the results of all your proposed experiments.

I'm going to try you melatonin techique, and see if it reduces sleep need for me as well. 

How do you tell the difference between needing less sleep and just waking up early? 

Maybe you could use the 'your day feel' in the sleep journal to compare gong with 1 hour less sleep for 1 week with melatonin and then for a week without?

 

Keep up the interesting experiments

Gwern Branwen

Mon, 07/04/2011 - 08:50

agbell said:

        How do you tell the difference between needing less sleep and just waking up early? 

Maybe you could use the 'your day feel' in the sleep journal to compare gong with 1 hour less sleep for 1 week with melatonin and then for a week without?

 

Keep up the interesting experiments

I am not entirely sure how I would distinguish. The alarm part might make that a bit hard. I guess one could look at the data and see whether melatonin use correlates with the alarm clock going off early (I use smartwake and right now I have it set to 9 AM with a 30 minute prior period - so you would see how often it goes off before 9 AM). I don't know whether the data includes this. It's probably buried in the CSV.

Your second question prompted me to look at other metrics, and I see that while I never used 'day feel', I have recorded 'morning feel', and the values are 2.9 on non-melatonin days and 3.1 on the melatonin days. This is only suggestive like the rest of my data (one could argue that I have become 'addicted' to melatonin and 3.1 is my normal value and 2.9 is withdrawal/disruption value), but is interesting. Better add that in to the page.