91 comments

[ 3.0 ms ] story [ 170 ms ] thread
I will read this article now and then take a (long) break…
Not knowing how long a read it is, I'm going for the break first, just in case.
(comment deleted)
Don't forget to take breaks short breaks while reading this article, which itself is a long break from doing other things.
I wonder what are the best ways to take breaks when you're on the computer for 7 hours or more a day without losing your concentration when you have to go back to the screen (NB: I often have several client projects to manage in the same day, my brain goes crazy)
The Pomodoro method jumps to mind.

But I’m with you - if the work at hand isn’t stimulating, any break will take a long while to “recover” from.

mental burnout is a thing. The best way to take breaks ... Is to take breaks! :D

Benefits of taking breaks are many. Such as getting up and moving, increasing blood circulation. Also relieving eye strain by focusing on far away things, rather than your close-up monitor.

I try to practice the 90m on, 30m off regimen that Huberman advocates, and it's worked well. I can do that all day without burnout.
> "An experiment began when a subject was shown the code “41234” on a screen and asked to type it out as many times as possible for 10 seconds and then take a 10 second break. Subjects were asked to repeat this cycle of alternating practice and rest sessions a total of 35 times."

I was hoping this would be about taking a few days or weeks off from whatever you are learning and have that break strengthen that skill. It's worked for me for several things in the past when I get bored or burnt out. When I get back to the task after about a week or so, I instantly understand the task I was learning. Also, some things that weren't clear before suddenly make sense.

I have experienced this as well. Similarly:

- Learning controls in a new-to-me video game can be tough (e.g. Rocket League, Quake Live, games with lots of mechanics.) Playing daily can slowly improve the ability. However, taking a few days off can sometimes result in a dramatic improvement upon return.

- Entering a flow state of code is great for churning out a lot of lines of code quickly. However, sometimes I see that while I am objectively achieving my goal, the code feels messy and hard to follow. At this point, I aim to just get it working by the end of the day, then stop. In the morning, I usually have a much better sense of how to refactor the code.

Same experience with Rocket League. A few years ago, I had a couple months between jobs, and my brother was in a similar spot, so we'd hang out and play Rocket League a fair amount. I slowly got better, but I only improved significantly after some time away from the game. Now, we might play once every few weeks or once every few months. Takes a bit to knock off the rust, but once we get going, I almost always notice some unexpected improvement - dribbling, positioning, and transitions from ball to car cam are the areas that seem to improve the most. Love the game, but it's not high on my list of areas I want to improve, and I'd definitely improve more with shorter breaks (days instead of weeks or months) and more time playing.
>"An experiment began when a subject was shown the code “41234” on a screen and asked to type it out as many times as possible for 10 seconds and then take a 10 second break. Subjects were asked to repeat this cycle of alternating practice and rest sessions a total of 35 times."

To me this doesn't sound like it's testing a skill at all. It's testing short term memory, maybe. 10 seconds on 10 seconds off to test your ability to memorize a string of numbers and type them out isn't the same thing as testing a skill.

I recognized this is how I best learn a long time ago. It meant that I could never cram for tests. But, once I learn something I usually really learn it. You're right though that days or weeks is usually my timeline.

For example, if I want to learn some new tech I will jump in and tinker until I get to some spot that is beyond what I understand. Take a few days or weeks, and when I come back the base has solidified making it almost trivial to move on.

I also just noticed the same thing with training Jiu-Jitsu. I had to take a long break because of covid. While I was obviously rusty on my return, I had a bunch of skills solidify while I was gone leading me to see many more details than before I had to take a break.

Learning is a lot like weight lifting. It's the breaks between sessions where the strength gains occur.

I made an adage some years ago:

Practice doesn't necessarily make perfect, it can also make you habitually wrong.

---

I've trained, I don't, 50 apprentices. They all learn faster of they can do a short stint on a new task, say 10 to 29 mins, then come back to it after at least one sleep.

They almost always develop bad habits if you make then do the new task all day because they'll get either overly focused on some inconsequential aspect of it, it lose interest because they're not good at it, or sone such.

I've heard this stated as "practice makes permanent ", especially in the context of physical arts, and also safety culture.

Once you start getting sloppy, it's time to context switch.

Very applicable to musical instruments and arts as well. At least to me
This is also very true for marksmanship.
Yes, also similarly is "neurons that fire together wire together".
> They all learn faster of they can do a short stint on a new task, say 10 to 29 mins, then come back to it after at least one sleep.

I have this experience playing GTA. After failing a hard mission 6+ times, I take a break, and usually pass it on the first attempt after the break. I also had not problem following the train, but the RC copters were a nightmare.

Yes!! It's like SRS but I've noticed it works for much more than flashcards/learning language type learnings - even works for physical activities like you mentioned.

I first noticed this for myself wayy back when I was learning how to ride a bike. I remember falling and falling and then giving up, coming back a few weeks later and nailing it on the first try!

Then again I noticed this as a youth while playing video game campaigns. I remember not being able to defeat Bowser in Luigi's Mansion or something, giving up, then returning in a week to handily defeat him.

I think it's after noticing this in video games that I began to internalize it as a learning strategy.

From then on I never really crammed for tests (which also benefits you on test day as you're not sleep deprived or stressed) or stressed out over pushing past any walls I would hit like you mention. Instead I'd focus on gradual learning with LONG breaks to absorb (not actively, just put your attention to something else and let the information "download asynchronously/on background" so to speak).

More recently I've begun to speak up about this learning style and, as I've begun mentoring junior employees or teaching friends technical skills, I've realized this is a foreign concept to many. So fascinating :)

I recently read the 80/20 book and it had a good section about giving your subconscious something to chew on.

Our subconscious churns through a lot without us really realizing it. In the book, he suggests thinking/working on something and stepping away with intent when you get stuck to let your mind wander while your subconscious gets to work.

I think this is a similar concept. You don't necessarily need to be stuck, but while you're away from something your mind puts all the "files and folders" away neatly.

I have a feeling there are different tiers to this phenomenon. Take a 15 minute break, and you'll be better at that piano piece you've been practicing intensely for the last hour. Take a week long break, and you'll be better at painting those landscapes you've been studying for the last month.
In sports training this is called super compensation. Usually you tax yourself for some period of training and then take a short break. Your body will recover to the initial level of fitness you had + you make some new gains. When your return to training your stronger/faster than before.
I recently started piano, and noticed this. I’d find a set of finger movements impossibly frustrating. Work on it for 30 minutes then give up. Then sit down the next day or two later and just be able to do it.

It’s given me a whole new perspective on how I can learn complex things over time.

I’m most productive on my 10-minute walk during the work day when I rubber ducky to myself about the problem I’m working on.

I haven’t kept exact track, but I’m definitely hitting over .500 with that specific technique.

Looks like this gives credence to the efficacy of the pomodoro technique.

Similarly, exercise (including walking) plays a role in learning, memory consolidation, and novel thought process formation.

Taking a 10-20 minute break to practice non-sleep deep rest immediately following a period of intense focus (~90 minutes or so) has been shown to increase rates of learning of whatever you were doing during the period of focus.
Is this why I retain so much of my high school and early college education? I deep rested so hard my face hit the desk once per lecture or so, but I can still tell you about what the professor was teaching.
> "An experiment began when a subject was shown the code “41234” on a screen and asked to type it out as many times as possible for 10 seconds and then take a 10 second break. Subjects were asked to repeat this cycle of alternating practice and rest sessions a total of 35 times."

This makes me suspicious of finding. Doesn't it decrease our focus/attention span with so many breaks? Isn't it equivalent of checking Twitter/FB every minute and not able to deliver anything?

When results don't match your hypothesis, it's usually your hypothesis that was wrong, not the results.
totally agree, not saying my hypothesis is correct, but it is already proven that checking social network frequently is bad for mental health, especially notifications which is related to dopamine generation. not apples to apples comparison, but something is common on these things
There's a big difference between checking Twitter and sitting in silence.
Depends on the amount of noise in the domain that you're studying. The more noise and the more implausible the alternative is, the stronger the evidence you need to be shifted off the null.
You're suspicious because the results don't match up with your bias?

The ten second break doesn't really provide the ability to change focus to something else. It's not very long and you are limited to what you can focus on, so the ten seconds will more likely be spent on determining how to improve performance.

Some practical advice on how this can be put to use from Barbara Oakley: https://youtu.be/vd2dtkMINIw
I highly recommend her book too. Combining what I learned from her book with spaced repetition software has boosted my capacity to learn significantly.
Which book? She has quite a few published.
A Mind for Numbers is the one I read. It also goes well with her Coursera course Learning How to Learn.
Is there one or two particularly high impact things you could share here?
There are high focus and low focus modes of thinking that serve different purposes in learning something.

Chunking is the encapsulation of several steps or processes into an overarching process. Think of all the steps required to put on a shirt. You need to know how to hold the shirt, have an intuitive understanding of cloth dynamics, pull it in a certain way so that you can put your arms in, etc. But because you do it every day, it has become second nature and all you have to think to complete the task is "put on shirt." It has been chunked.

Chunking requires 3 things: focus, understanding, and practice. Math is largely about chunking underlying concepts so you can learn more complex concepts.

Chunking is a high focus activity, and spaced repetition is best as a low-focus activity that can be used to efficiently memorize jargon and other things that involve rote memorization.

There's a lot more in the book, but those are the pieces that I remember best and that have had the biggest impact for me. I review flash cards every day, and once every few days, I do a high-focus learning activity. Even though I don't put much effort in, it feels like I am learning as much as I did in university.

Does she cite any scientific evidence that her method works?
Yes, there's an entire references section in the end of the book.

The process I use isn't everything the book said to do, it's just what I chose to adopt to strengthen my weaknesses. The book covers much more.

Hmm, but I wonder if not doing enough at once continuously makes focus and productivity lower, up to a point, by doing activities like alternating between work and video games/glowing screen notification checking.

I also would like to know what is the neurological basis for the painful post-exam / difficult work "hurty brain." I can only imagine what it must feel like after taking the JEE.

I suspect that alternating to another activity might interfere with the process by using up the bandwidth of the channel for replay.
This why I quit my job every couple years and take a sabbatical
As I’ve been working on my own projects, on my own schedule, at home, I’ve taken to having breaks; sometimes, even short naps.

I don’t really have a fixed schedule. I generally start the day with some goals, and usually take breaks between work on each goal. Sometimes, they are an hour or more. I may do errands. I tend to start the day quite early, and often work until bed.

If I am stymied by a nasty problem, I will often take a break, and maybe watch a part of a movie, or something, then go back to it. Bugs never last more than a day, with me.

If it’s a big context switch, I may have a “take it easy” day, where I take care of administrative crap.

I enjoy more productivity than I have in my entire life.

That's how I organized my work in grad school. I basically worked when I felt like it. Some weeks that was 80 hours, others just 20, following the natural flow of my motivation. I went for errands in the middle of the day, took time to cook, parties, but also worked late nights and weekends when it felt right. I slept on a shitty futon and ate a lot of ramen noodles, but it sure felt nice to be able to work at my own pace and sleep as much as I needed, yet still manage to publish more than anyone else in my lab.
Which is why I support smoke brakes in the office
This is how I first learned programming at 25, then leetcoding five years later.
Its not just short breaks. Effect of a good night's sleep are almost magical in this respect.

Why We Sleep by Matthew Walker is filled with examples on how sleep plays an outsized role in our lives.

One relevant example from the book that comes to mind is an experiment where they instructed some college students to type out a particular sequence of characters on a keyboard and then measured their performance across two days. The group which had a good night's sleep had dramatic improvement in their typing coordination overnight.

Apparently the author got an insight to pursue deeper into cognitive effects of sleep after a chance encounter with a pianist after a speech he gave on benefits of sleep, where the pianist told him how he struggled with new compositions on evenings and then magically gets them right after a good night's sleep.

I can list a bunch of trivial reasons why we sleep, about memory and so on. I can also give trivial explanations like "it's because you're tired and you get rest".

But I wonder if anyone has explained why typing coordination would be worse without sleep. This is a mechanical task where analytical thinking and so on are not involved, just simple motor memory patterns.

Playing piano doesn't require analytical thinking either. Motor memory is just another set of patterns, and rest helps establish the patterns in our mind.
I feel we need to flesh that out more. Because if we just say sleep establishes the patterns, it's almost as if we say a pianist who has been playing for decades forgets how to be a pianist without a day of sleep, and then while sleeping he remembers how to be a pianist.

I feel I need a more detailed explanation here.

Apparently something gets depleted from the brain during the day, that we get back during sleep. Maybe it's structural, not chemical, but something gets depleted.

In the aforementioned anecdote, the pianist was struggling with a new composition, not general piano skills. Which would make sleep a catalyst for making new neural pathways for that specific composition, not how to be a pianist in general.

That's my take as a layman.

Learning a new piece usually requires learning novel movements, stuff you've never done before. Maybe you've played a C major before, but you don't always use the same fingers depending on where you're coming from and where you're going, maybe the cadence or intensity is different, and so on.
Still, I can witness myself if I haven't slept well not only my ability to figure out complex programming code suffers, but the mere ability to type fast without a ton of typos.

Otherwise I agree after sleep, what we learn seems more readily available.

I'm not sure that's entirely true. There are many ways to approach playing piano, although they all require muscle memory there's a huge cognitive portion to improvisation for example. Some do that by feel, and some do that by actively thinking about the theory they're playing.
The book Brain Rules by John Medina is worth checking out for some suggestions on possible answers. I don't think science has definitive answers yet but if I remember correctly it's thought that the circuits involved in what the brain learned during the day are active during certain sleep stages which helps with consolidation of memory due to strengthening of connections involved. It's 10 years since I've read into this but I think the biology involved was that oligodendrocytes lay down thicker myelin on axons that are active, resulting in faster/stronger connections in the circuits involved. There's also inhibitory actions of established networks which have to be overcome by new circuits which takes sustained 'attack' over time. If the inhibition didn't happen the previously learned activities would turn noisy very quickly so there's a tradeoff between learning speed and maintenance of high fidelity signal for existing pathways in the brain. I guess the brain has evolved to accept new networks from actions that are repeated and thought about repeatedly because these generally are important whereas stuff that is thought about only once probably isn't.
Do you happen to know what good sleep means?

I've done polysomnography and they told me nothing except I don't have sleep apnea, and got a lot of microawakenings -- to me it felt I didn't sleep at all with all the wires and noise by other patients in other rooms snoring, nurse re-attaching wires, nurse walking on hallways to check on other patients and so on.

My smartwatch tells me I spend 60% in light sleep, 20% in REM and 20% in deep sleep, but I get very little continous deep sleep.

I struggle a bit with perception of how much I've slept. It can often feel like I've been awake the entire night, aware, thinking, I swear I must dream of lying in bed awake.
Plus not sleeping in your own bed/room also has an effect..
Good sleep means waking up without an alarm refreshed and ready to face the world. Good sleep means waking up with less physical pain then when you lied down. Good sleep usually means there was a vivid, but unimportant, dream. Good sleep means I stayed on my back the whole night, and woke up in the same position I went to sleep in.

YMMV Of course

(comment deleted)
I can't sleep on my back and I usually toss and turn all night and when I wake up, I feel like I need just 3 more hours of sleep.

My mileage does value indeed :))

When I was younger I forced myself to learn to sleep on my back. I can only do it with a contour pillow, not necessarily the brand name one, but any with that contour in it. I can't always do it, but when I do my sleep quality is significantly better.
"waking up without an alarm refreshed and ready to face the world"

Does that actually happen outside of stories?

Hey, it happened plenty for me during the holidays between school and university terms.
I dont use an alarm. Haven’t it years. Whether or not I’m refreshed depends on how late I went to bed.
Yep, it is the rule rather than the exception for many who are "natural" deep sleepers. I used to be one of those until I got overweight.
yes, it happens to me most days. Except the ready to face the world part, It's usually more like waking up without an alarm refreshed, and ready to browse HN and watch youtube.

edit: the waking up part didn't happen until I started working from home. I was always over-sleeping when I had to get showered, dressed, and commute every morning.

Kind of.

In general, getting good sleep means that you fall asleep fairly quickly (30 minutes or less), don't wake up often (once or twice), and you get around the average sleep for your age (even if it varies a bit). You should feel pretty well-rested through the day, and not overly tired if you are in good health.

I'd wager that you don't get good sleep, and you are suffering from us not really being able to "fix" sleep very well - in no small part because we are really in the infancy of sleep science. The sleep study, for you, really just ruled out some things, but couldn't provide answers.

> you are suffering from us not really being able to "fix" sleep very well

Yes, the only drug that made me feel like a baby waking up in the morning was trazodone.

But it also made me wake up with painful boners in the middle of the night, so that's out. I'm so envious of a female acquaintance who swears by trazodone.

Perhaps she's pairing the trazodone with something else and not telling you. Like a Magic Wand.
I just had a sleep study with EEG and visual/audio recording. They said sleep efficiency of 95%, yet I almost always wake-up very tired, with a dull/stabbing headache, and sweating around the neck.

I also have physiological anxiety, treatment-resistant depression, attention/concentration/alertness/consciousness/memory deficits, tachycardia >100, rapid shallow breathing, high blood pressure (diastolic), veinous insufficiency, cold hands (but normal perfusion), almost no sensation anywhere, excessive sweating w exertion, lower blood volume, swelling neck/fingers/toes, and flushed forehead/ears/cheeks.

I wonder if it's a lack of exercise, chronic blood-vessel narrowing, and/or chronic low blood volume.

Live in a different environment for a few days to a week to rule that out.
That sounds exactly like dysautonomia (autonomic nervous system dysfunction/disorder), and specifically POTS (postural orthostatic tachycardia syndrome).

I have a rare form of dysautonomia. It is a very rare immune-mediated neurological disease affecting my peripheral nervous system, called autoimmune autonomic ganglionopathy. It is in pharmaceutical remission, which means that it hopefully stays in remission, as long as I take medicine for it, for life.

Anyways, I can help you out with all of this, and really show you the way. I literally have all of the above symptoms. Check out my profile for my email. I will respond to you with advice.

You may want to start here: DYNA (Dysautonomia Youth Network of America) is an awesome organization, and they have a phone line that they answer during the day. You may want to call them first for advice. Adults call the line all the time. I mean, I am a member of it, and participate there regularly, and I am in my early 30s. Anyways, the DYNA office phone number is: 301-705-6995

I strongly encourage you to call them. Also, please contact me. I also promise that I will help you.

Here are some good references:

Dysautonomia Symptoms: http://dynainc.org/dysautonomia/symptoms

About Dysautonomia: http://dynainc.org/dysautonomia

Living With Dysautonomia: http://dynainc.org/living

Finding the Right Doctor: http://dynainc.org/living/finding-right-doctor

For Newly Diagnosed Patients: http://dynainc.org/living/new-patients

There is now a special board certification/subspecialty in neurology in the United States for Autonomic Disorders (my neurologist has this board subspecialty certification also has an autonomic disorder--so I am tremendously lucky to have him). A handful (about 50) doctors in the US are certified in this subspecialty. But, you really need to get an EMG test, since you are having sensory issues. A neurologist does this test. You really need to be seeing a neurologist regularly if you are having these types of symptoms.

Anyways, to find a neurologist with the Autonomic Disorders subspecialty (you need, at minimum, both a neurologist [most important] and a cardiologist/electrophysiologist who deals with autonomic disorders regularly) go to this website: https://www.ucns.org/Online/Online/Diplomate_Directory.aspx

Under "Please choose a subspecialty" (drop-down menu), select "Autonomic Disorders" and press the blue button "Find"

If you want to get better and stay healthy: do not go to Reddit/Social Media/etc. It is a really toxic place with respect to this illness. People on there are extremely dramatic. Also, avoid the organization Dysautonomia International. The people who run it (primarily women--and I am female myself) are drama queens. They also post inaccurate information on their official Facebook feeds all the time. Also, don't read stuff posted to random blogs. Content that is not moderated about dysautonomia tends to really get wild and can be quite harmful. Trust me on this. If you want to get better with this illness, you have to be careful about what you expose yourself to on the internet.

I hope this helps and I hope to hear from you.

I too read Why We Sleep, and found it quite interesting. Then I found out that there is some controversy regarding its claims.

Previous discussions on HN:

* https://news.ycombinator.com/item?id=21546850

* https://news.ycombinator.com/item?id=26684519

TBH I haven't made up my mind about the book yet, just thought you should know about these things too.

I found these critiques soon after reading the book. However, there are many other researches cited in the book and I would take them on face value for the not very scientific but somewhat valid (for me) reason of personal experience.

I got into the habit of all night slogs right since high school and it persisted till recently due to the pressure of sticking to deadlines. So I could relate well to several experiments cited in the book.

Considering we do not know nearly as much as we ought to know yet, I would rather err on the side of caution and give far more importance to sleep than I have previously.

The article specifically mentions a previous study that showed that gains from short rests were greater than those from sleep:

> In a previous study, led by former NIH postdoctoral fellow Marlene Bönstrup, M.D., Dr. Cohen’s team showed that most of these gains happened during short rests, and not when the subjects were typing. Moreover, the gains were greater than those made after a night’s sleep and were correlated with a decrease in the size of brain waves, called beta rhythms.

Anyone who has learned an alternative keyboard layout will tell you this is true. When you get a new keyboard (say, with linear layout), you will struggle like crazy during the day doing typing tests like https://monkeytype.com/ back to back like 20 times, and your speed won't improve –it will deteriorate.

But give it a good night's sleep or at least a few hours not doing the exercises, you will come back at higher WPM speeds. And at some point you wake up and it just "clicks" in your brain, congrats you just re-wired your brain.

You know how people, especially software developers, complain about being interrupted by others at work? I, as a dev, have recently realized that this can sometimes be a good thing. It gives the brain a few seconds to unconsciously process whatever we are working on. I have found a few times recently to return to my work with a slightly freshened mind, sometimes having a better solution than the one I was working on.

Now, this said, there is a balance. We don't want to be interrupted very often but sometimes taking a break, even for a few seconds can be a good thing.

Being interrupted with a question that forces you to flush all your brain caches and load up a different context is not the same as a nap or walk-induced interruption ;)
I'd rather take break intentionally than having to context-switch due to somebody else
Being interrupted by my own rising task-fatigue, when I am at the natural end of a flow-cycle is very different from being interrupted in the middle of flow.

Other people’s interruptions destroy my productivity. The balance there is zero outside interruptions.

When I finish a flow cycle, my attention naturally goes to something else, a perfect time for a break.

I noticed this while learning to drive. When I practiced driving continually, I was learning little. But when I gave few days breaks in the middle, I found myself much more comfortable.

Even with exercise, I've observed that intermittent rest has helped.

I’ve definitely noticed this teaching the guitar. If students first have a really good run at learning something, particularly something mechanical when you’re working on a piece/exercise, if you stop at some point and chat about/distract them with something briefly, subsequently when they start playing it again it’s sometimes as if ‘magically’ what they were working on previous to the break has been ‘digested’ on a more sub-conscious level or something, and afterwards they play it much better… I wondered if it was because initially the conscious mind was over-thinking and grappling with mapping a task from individual pieces (bit like: https://en.m.wikipedia.org/wiki/Blind_men_and_an_elephant ) …or even self-sabotaging, and the break helped somehow to reach a more ‘flow-y’, confident state where the brain has a starting point to work from - some sort of pre-prepared mental ‘map’…
When I get 'lazy' taking a long timeout and doing something as mental as washing dishes always result in me getting back to the PC in a flow state.