Hacker News new | past | comments | ask | show | jobs | submit login
Direct Evidence Of Role Of Sleep In Memory Formation Is Uncovered (sciencedaily.com)
76 points by codeodor on Sept 17, 2009 | hide | past | favorite | 31 comments



Original article: http://www.nature.com/neuro/journal/vaop/ncurrent/abs/nn.238...

I was inclined to be skeptical---breaking news, shocking the brain during sleep impairs performance---but they used three different groups of rats:

  ...we tested the role of SPW-Rs on memory consolidation.
  Three groups of rats (test group, n = 7; stimulated 
  controls, n = 7; unimplanted controls, n = 12) were 
  trained to find food rewards ... . 

  During post-training rest and sleep, all of the online-
  detected ripples were suppressed by commissural 
  stimulations in test rats (average online detection rate 
  was 86.0 ± 1.3% (s.e.m.) of post hoc detected SPW-Rs;
  ...). Stimulated control rats underwent the same 
  protocol, except that a random delay (80–120 ms)
  was introduced between SPW-R detection and stimulation, 
  ensuring that the stimulations occurred mainly outside of 
  the ripple episodes.
Now check this out:

  Thus, these control rats received the same number of 
  stimulations as test rats, but their hippocampal
  ripples were left largely intact. The global architecture 
  of sleep and the local field potential power in distinct 
  sleep stages were not modified by the suppression of 
  SPW-Rs ... .  As stimulation outside SPW-Rs had
  no detectable effect on task performance ..., the two 
  control groups were pooled and compared with test rats. 
  Performance of the test rats was significantly impaired 
  ... .
Cool experiment design with an intriguing result.


Thanks for that. I was going to include the link to that, but when I clicked for the full text it asked me to subscribe.

Now that I've gone back, I was able to grab the PDF.


In doing research for another comment I posted here, I came across an article that does a very good job of summarizing the current state of research into the role of sleep in memory consolidation.

http://www.nature.com/nature/journal/v437/n7063/abs/nature04...


"Sharp wave ripples are intense, compressed oscillations that occur in the hippocampus when the hippocampus is working off-line, most often during stage four sleep, which, along with stage three, is the deepest level of sleep."

I'm not an expert on the subject, but how can two stages both be the "deepest"?


Both stages have delta waves in the same range but induce different effects on the brain. They're both characterized by the same measure for depth (of sleep).

To clarify with more, specific information: http://en.wikipedia.org/wiki/Slow-wave_sleep


"Sharp wave ripples are intense, compressed oscillations that occur in the hippocampus when the hippocampus is working “off-line,” most often during stage four sleep, which, along with stage three, is the deepest level of sleep."

So do people who practice polyphasic sleep have awful memory?


Polyphasic sleepers are a myth. The only truth is a handful of people who tried it for a couple of months and then went back to a conventional sleeping pattern. (However, many of them, for some bizarre reason, are still strong proponents of how great it is.)

Edit: Downmodder - Give me one documented exception please.


The way I understand it, Polyphasic sleep is designed to trick your body into skipping the first two stages of sleep, so you spend all the time in the last two, where the biggest benefits are.


Is there any proof of this or is this just what polyphasic sleepers tell themselves?


You're right, I can't find any evidence other than several polyphasic sleepers' blogs. I can find a handful of articles on polyphasic sleep, but all of them are behind pay-walls and I can't get to any actual data.

The only actual data I could find related to sleep periods evening out to a similar daily distribution after 45 days.


I could probobly help you access the journals if you want to investigate further, I'd be interested in hearing the answer.


At this point, I don't think there is much in the way of real work behind it. From what I can tell there has been one paper written about it (http://www.informaworld.com/smpp/ftinterface?content=a782431...), which is a "field study", not a laboratory test.

All I seem to find are pseudo-science and anecdotes supporting both sides of the argument, with no actual evidence.


That seems a bit off, trick your body into skipping the first two stages by...staying awake?


Not quite, you try and sleep at set times and after some time you adjust to actually sleep during those times. Since the sleep period is typically very short (~30 minutes), the idea is that your body adjusts. From what I've read, people who've tried it (for longer than a month or two, ie they actually adjusted properly) found they were just as rested as everyone else.

Having said that, who knows what negative effects it may have on them...


I've read multiple accounts of people who tried it and claimed they were just as rested. But I emphasize try because they no longer do it, and I bet the reason is they were not just as rested.

edit: polyphasic sleep discussion: http://news.ycombinator.com/item?id=652650


http://www.supermemo.com/help/faq/polyphasic.htm This was a great find, thanks for the link. Pretty much proved all my thoughts true


Interesting read.

There was that guy (though I cannot remember who it was) who claimed to have sustained polyphasic sleep for 9 months and only stopped it because he found it hard to interact with people and socialise because his schedule was too different from everyone elses. But, I have no evidence, so.. that FAQ seems reasonable though.


Steve Pavlina?

http://www.stevepavlina.com/blog/2007/03/polyphasic-sleep-on...

he has a ton of articles on his blog describing his experience


Yeah, that was probably who I was thinking of. Thanks for the link.


I tried it for six weeks, but never got it to settle down into a reasonable schedule. I'd have a few days of pretty good sleep, followed by a hard crash, then a struggle to get back onto the schedule etc. Especially falling asleep quickly was very difficult for me (whereas I fall asleep very easily on a normal schedule).


From what I read, it took quite a long time. Between one and three months, depending on the person. I guess three months would be pretty painful. I've never tried it personally, so...


After pulling an allnighter, I fell asleep during a math class and woke up so many times that I would start dreaming before my eyes closed.

It is extremely annoying not knowing if what you just experienced was true or a dream, so I ended up going home.


Unless you know something I don't, there has been zero research into this topic, and anything you read on blogs is a wild guess.

Please prove me wrong by citing some research. I just did a search (terms: "polyphasic", "sleep", "sleep architecture", "REM", "SWS"), that didn't turn up anything.


Some Claudio Stampi made some research on this. His findings are summarized in a hard-to-find, expensive book:

"Why We Nap: Evolution, Chronobiology, and Functions of Polyphasic and Ultrashort Sleep" (1992)

http://tinyurl.com/n86q3d


According to that article, this guy is "founder, director and sole proprietor of the Chronobiology Research Institute which he runs from his home". Forgive me, but my quack-o-meter just redlined.

Can you point me to something that has been peer reviewed?


http://scholar.google.com/scholar?hl=en&q=Claudio+Stampi...

But hey, don't let me interfere with your blind application of heuristics.


Thanks, I did that search too, and yes, the guy's published a paper or two.

What I meant was, show me something that's been peer reviewed and relevant to this topic... an EEG study of people following polyphasic schedules, for example.


You certainly look too hard at it then. You apparently missed: http://www.informaworld.com/smpp/ftinterface?content=a782431..., the article that I most want to read from it. Sure, it isn't an EEG of the sailors, but it sounds (from what I've heard second hand about it) like it may have some relevancy.


I did see that article before, but did not consider it relevant to your question. While this study shows that in a "highly demanding [continuous work] situation" people can adapt easily and maintain high performance under polyphasic schedules, it says nothing about how their sleep architecture (the arrangement of their sleep stages in a 24h period) changes.

Incidentally, I think it covers what most polyphasic bloggers get wrong while adjusting: they do not create continuous work demands, and so at some point they get bored, fall asleep, and revert to their original sleep schedule. If you want to do the Uberman right, follow the Pomodoro technique in cycles of 8 pomodoros, sleeping one and working seven. Eat something after each nap.


Those rats that selectively had all ripple events eliminated by electrical stimulation were impeded in their ability to learn from the training, as compressed information was unable to leave the hippocampus and transfer to the neocortex.

Probably remembered being electrically stimulated instead.


I think the experiment design eliminates that possibility. Why would this group of rats do worse than the ones that were electrically stimulated, but not in a way that disrupted the ripple events?




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

Search: