Difference between revisions of "Talk:1345: Answers"

Explain xkcd: It's 'cause you're dumb.
Jump to: navigation, search
(observation)
 
Line 36: Line 36:
 
It's long been a fundamental technique of artificial neural network learning, to alternate between "learning" and "sleep" modes.  I've heard (but cannot find the citation, sigh) that when running neural networks, it turns out that they lose the ability to learn after running a long time.  But you can avoid this effect if you periodically bathe the neural network with completely random input.  [[User:Jorgbrown|Jorgbrown]] ([[User talk:Jorgbrown|talk]]) 07:35, 23 January 2015 (UTC)
 
It's long been a fundamental technique of artificial neural network learning, to alternate between "learning" and "sleep" modes.  I've heard (but cannot find the citation, sigh) that when running neural networks, it turns out that they lose the ability to learn after running a long time.  But you can avoid this effect if you periodically bathe the neural network with completely random input.  [[User:Jorgbrown|Jorgbrown]] ([[User talk:Jorgbrown|talk]]) 07:35, 23 January 2015 (UTC)
 
:Makes sense, and reminds me of some optimization algorithms. Interesting! [[Special:Contributions/108.162.219.125|108.162.219.125]] 03:01, 4 March 2015 (UTC)
 
:Makes sense, and reminds me of some optimization algorithms. Interesting! [[Special:Contributions/108.162.219.125|108.162.219.125]] 03:01, 4 March 2015 (UTC)
 +
 +
It seems to me, (in a completely speculative way) that sleep is evolutionary debt from older species. In other species sleep is used to conserve energy at times when is is not necessary, or not possible to do anything beneficial for the individual. It is a trait that evolved very early, and due to it's timing, many other physical and mental processes are tied to it so they can happen while the body is not exerting itself. This ties together a huge number of unrelated things into the process of sleep, which is inherently an energy saving mode, and makes it necessary to trigger all the other critical processes. In modern day first world countries, energy savings is not as nescessary, and artificial lighting, intelligence, and social structures give us useful stuff we could be doing instead, and if the extra processes were dissociated from sleep, it seems that we would be able to function as a 24 hour being (and not just like dolphins or whales where half their brain sleeps at a time, but true 24 hour consciousness), but separating all the myriad of functions tied to the circadian rhythm and the process of sleep is likely to prove difficult or impossible. --[[Special:Contributions/188.114.106.41|188.114.106.41]] 17:35, 11 May 2015 (UTC)

Latest revision as of 17:35, 11 May 2015

Not true. We know that sleep is important for storing memories and cleaning out toxins. http://www.nih.gov/news/health/oct2013/ninds-17.htm 108.162.222.89 11:06, 21 March 2014 (UTC)

That report is entitled "Brain may flush out toxins during sleep". Note the "may". Add it to the list of hypotheses. Jim E (talk) 15:49, 21 March 2014 (UTC)

The idea is that we do not know why we evolved to need sleep, when microorganisms do not sleep. If we had evolved without developing the need to sleep we'd also have evolved another way to retain memories and flush toxins. There is no truly biological reasons why a species would evolve the need to sleep when the option to be alert all the time is more obvious, as it means there is less risk of attack and danger.108.162.219.16 20:00, 21 March 2014 (UTC)

Apparently, one of the thing brain or body does while sleeping - reindexing memories, flush toxins, self-repairs, I think that the list of hypotheses is long - give us so big advantage when awake that the limited awareness (note that some noises can still wake us up) in period of sleep is worth it. Or, perhaps the alternative solutions are too hard to evolve. Remember that even while sleeping, humans are much more active that most microorganisms: it is probable that need for sleep evolved at the same time as the brain itself. -- Hkmaly (talk) 12:54, 22 March 2014 (UTC)

Stupid personalized jokes and the like in this explanation... 173.245.53.174 11:19, 21 March 2014 (UTC)

I'm not sure this is the correct explanation. The paradox of being confronted daily with a mystery and not trying to solve it is inconsistent with the title text. So this explanation doesn't sound right to me. I think it's more about defining humanity as seeking for answers, while spending a huge amount of time closing off from the world for apparently no reason. In other words, IMHO, it's not about "[not being] distracted by this mystery", but about "not being able to investigate any mystery during 1/3 of our life even if we want to".

Also, with my explanation, the original puchline "touché" works better than the the current explanation's suggestion "Which is why it keeps me awake all night". -- Shirluban@108.162.229.36 12:28, 21 March 2014 (UTC)

I disagree with the above poster (and agree with the explanation) on the basis of the boldface text... "And nobody knows why". Every human sleeps, so if humans were really curious, someone should have figured out why by now.NikoNarf (talk) 14:23, 21 March 2014 (UTC)

While we do not know why we sleep we do know what happens during sleep. All the studies and hypotheses cited here do show information about the states of our brain during sleep. We are well aware of most of the biological processes that happen in the brain during sleep. Its not that sleep is some utterly mysterious thing our bodies do. Sleep research has not been as rigorously studied as other subjects in science and due to the very nature of science and scientific study to consolidate all this data, test hypotheses and develop theories takes a lot of time, findings must be checked, rechecked, and verified, and then there's the time it takes to use the facts gathered to actually come up with a working theory which, if refuted by testing, the whole process has to be done over again. The point is science takes a lot of time, money and manpower. We already know we need to sleep, we know how not sleeping affects us, there are other important questions in science which we are more driven to find answers to. So, our curiosity is considerable and day by day we continue to discover new things, but not many scientists are interested in a field of study which will get you way less money and recognition than breakthroughs in genetic engineering. Lackadaisical (talk) 15:57, 28 March 2014 (UTC)

I agree with the original poster. It is interesting how people spend their entire careers studying a life event that they may never experience (consider a man studying the act of giving birth), yet most of us simply take sleep for granted. Now if we could only make sleep more efficient! I think we could spare a couple months worth of study to this. http://www.explainxkcd.com/wiki/index.php/1205 Puck0687 (talk) 14:53, 21 March 2014 (UTC)

We could spare quite a lot more than a couple months on this. First, 1205 talks about the benefit over five years, and for us the benefit would be over an entire lifetime. Furthermore, far more people don't study sleep than study it, so the "couple months" you talk about can be multiplied by the total population of people who benefit (both alive today and yet to live), and divided by the population of people studying sleep. That gets you quite a lot more than two months. 108.162.219.11 16:18, 21 March 2014 (UTC)

I've thought for a while that the reason we sleep is primarily due to the accumulation of adenosine in the brain (?) - who really knows... Brettpeirce (talk) 16:02, 21 March 2014 (UTC)

The title text is referring to a statement William Dement (Stanford University) actually said. Source: http://ngm.nationalgeographic.com/2010/05/sleep/max-text 108.162.245.117 17:35, 21 March 2014 (UTC)

I disagree with the notion that we are no longer curious while we are sleeping (implied perhaps only by me?). I have awoken from sleep with answers to questions I went to bed with (or at least possible explanations to investigate). Brain activity has not stopped while we are asleep. I believe we have at least correlated benefits to sleep (or adverse consequences to the lack of sleep) but we don't know how much further down the root cause tree we still need to go - e.g why does sleep help with memory and weight loss and muscle repair. Ghaller825 (talk) 19:23, 21 March 2014 (UTC)

just becuase we are hungry does not mean we are fed. 108.162.249.231 07:52, 22 March 2014 (UTC)

I guess we can know what sleep is for by looking at what happens if we, or other animals, are prevented from sleeping. "Sleepiness" isn't just an urge, but an imperative which is torture if not obeyed. Hallucinations, irritability, and eventually death due to immune system degradation occurs, (at least in rats, dunno about humans) and this happens more quickly than starvation. Seems to me that what we don't know is not the "why" of sleep, but the exact pathways by which these malfunctions are caused, 108.162.245.117 (talk) (please sign your comments with ~~~~)

The referenced national geographic article has an interesting section on "fatal familial insomnia" in humans. Nealmcb (talk) 14:33, 24 March 2014 (UTC)

It's long been a fundamental technique of artificial neural network learning, to alternate between "learning" and "sleep" modes. I've heard (but cannot find the citation, sigh) that when running neural networks, it turns out that they lose the ability to learn after running a long time. But you can avoid this effect if you periodically bathe the neural network with completely random input. Jorgbrown (talk) 07:35, 23 January 2015 (UTC)

Makes sense, and reminds me of some optimization algorithms. Interesting! 108.162.219.125 03:01, 4 March 2015 (UTC)

It seems to me, (in a completely speculative way) that sleep is evolutionary debt from older species. In other species sleep is used to conserve energy at times when is is not necessary, or not possible to do anything beneficial for the individual. It is a trait that evolved very early, and due to it's timing, many other physical and mental processes are tied to it so they can happen while the body is not exerting itself. This ties together a huge number of unrelated things into the process of sleep, which is inherently an energy saving mode, and makes it necessary to trigger all the other critical processes. In modern day first world countries, energy savings is not as nescessary, and artificial lighting, intelligence, and social structures give us useful stuff we could be doing instead, and if the extra processes were dissociated from sleep, it seems that we would be able to function as a 24 hour being (and not just like dolphins or whales where half their brain sleeps at a time, but true 24 hour consciousness), but separating all the myriad of functions tied to the circadian rhythm and the process of sleep is likely to prove difficult or impossible. --188.114.106.41 17:35, 11 May 2015 (UTC)