EDIT: Reworked and moved to Main following Gunnar_Zarncke's advice.

Related to: Book Review: How Learning Works, Build Small Skills in the Right Order, What are useful skills to learn at university?

This article is organized into three sections focusing on attention, processing and recall respectively. The advice in each section is roughly organised in order of usefulness, although your mileage may vary. It's best to view this as a menu of study techniques rather than an in depth guide.

Just follow at the links provided if you wish to learn more about any point. Links with a lettered superscript[a] generally link to a part of a YouTube video while those with a numbered superscript[1] link to an article. Links without any superscript generally link to another LessWrong page.

Paying Attention

Attention is very important for learning. Where you spend it directly determines which areas of your brain you'll develop while studying and learning new skills.
  1. Split your study up into 25 minute chunks, separated by five minute breaks[a]
    Also known as the Pomodoro Technique[b]. This one is simple to implement but tremendously effective. It will protect you from attention burnout, increase your useful study-time, and help prevent distractions from becoming procrastination by setting up a Schelling fence around your breaks.
  2. Focus on one task at a time[1]
    Multitasking is one of the worst things you can do while studying, it can reduce your productivity by up to 40% and divides your attention up unnecessarily which will impair your ability absorb new information. If social media and the internet is of a particular distraction to you, tools such as Stay Focused can help you stay on track.
  3. Set up your study environment[a]
    Exploit situational psychology by making your environment more conducive to study; identify cues that cause you to procrastinate, remove them if possible, and set up cues for studying. Mentioned in the video, a 'study lamp' can make an effective cue providing it is only ever used for studying. Additionally joining a study group can be an effective way to do this (a good example being the LessWrong Study Hall).
  4. Choose the right music[2]
    There are a few rules of thumb to follow here. Avoid listening to music while trying to absorb new information, though if your aural environment is particularly distracting then music without lyrics or white noise can be useful. Don't use unfamiliar music or music with lyrics in as this will unnecessarily tax your ability to focus. Music can increase productivity for mundane or well-practiced tasks involving low mental effort.

Learning Material

Before going any further I'd advise you to watch this video[c]. It's an excellent explanation of why just going over material isn't enough to actually learn it and additionally dispels a few myths about the important factors in learning.
  1. Understand the principles behind 'deep processing'[c]
    The key thing to understand here is that the more you relate a new concept to ones previously learned, the more likely you are to remember it. This is far more effective than learning by rote, not only does it improve recall but it also improves your ability to apply the material. A study strategy that forces you to process things deeply is called to as an orienting task[c].
  2. Develop your metacognition[c]
    Metacognition refers to your beliefs about how well you know the material you're studying. Overconfidence here is negatively correlated with academic success (see the video) and can prevent you from updating on new material[d]. One of the reasons for this negative correlation is that overconfident learners spend less time on material than they should. Being sure to test yourself on your knowledge regularly can go a long way to combating this. 
  3. Understand the difference between recognition and recollection[a]
    Related to the previous point, a sense of recognition is one of the biggest causes of overconfidence when reviewing material. A good solution is to test yourself on your ability to recall material before you review it. Not only will doing so help you avoid mistaking recognition for recollection, but knowing what you don't know will help target your revision. 
  4. Troubleshoot your understanding[e]
    In most subjects, concepts have a chain of dependencies with advanced concepts depending on the more fundamental ones (in mathematics this chain is particularly long). If you're having trouble understanding a new concept it is very unlikely that you're inherently bad at understanding that concept, rather there's a flaw in your understanding of the more fundamental concepts that lead up to it. Target your understanding of those and understanding the concept in question will become much easier.

Holding onto Information

Once you've processed the material effectively you need to be able to recall it efficiently. While deep processing helps you get information into long term memory, getting it to stay there is a different matter entirely. Memory follows what's known as the forgetting curve[3]. Forgetting has not so much to do with losing the information, but rather having trouble retrieving it – and as far as learning goes you haven't really learned something until you can effectively retrieve the information.
  1. Test yourself on material[4]
    Practicing retrieval has a dramatic effect on your ability to recall information. Key to this method is ensuring your cues are appropriate to the way you're going to be test, so past paper questions tend to be best. When using flashcards it is important to make sure that the cues require you to not only recall the information, but process it on a deep level too. 
  2. Make use of spaced repetition[4]
    Spaced repetition is testing yourself on material over incrementally larger periods of time (an hour, a day, a week, a month and so on). The idea is to test yourself on information just as you're about to forget it and as it turns out, it is far more efficient than just blindly testing yourself on material over and over. Keeping track of when to review information can be a pain, fortunately there's plenty of spaced repetition software out there to do that for you (I personally find Mnemosyne is simple to implement and use).
  3. Get some sleep[a]
    Sleep is absolutely crucial for retention. If you must cram, make sure you do it the night before the exam, if you do things the other way round your memory will be considerably worse off for it. In general make sure you get a good nights sleep every day that you've studied. If you're having trouble sleeping due to spending a lot of time at a computer f.lux might be helpful to you.

Video Sources:

New to LessWrong?

New Comment
16 comments, sorted by Click to highlight new comments since: Today at 3:35 AM


You consistently use this where you probably want 'absorb'; they are not the same. Admittedly the usage is metaphorical, but learning something is closer to "take in or soak up by chemical or physical action, typically gradually" than it is to "adhesion of atoms, ions, or molecules from a gas, liquid, or dissolved solid to a surface". Presumably you don't want the information to just form a thin film on top of the learner.

Presumably you don't want the information to just form a thin film on top of the learner.

It seems you just invented a perfect term to describe (one of) the revealed preferences of the education system.

Just want to say I'm glad you're back and don't let the bastards/SJWs grind you down.

Fixed it. I don't think I've ever consciously registered that adsorb != absorb, so thanks for that.

Avoid listening to music while trying to adsorb new information

I recently spent under $35 to get industrial earmuffs and earplugs, for a combined total of 64 db of noise reduction. Single most cost-effective investment I've made in my learning (not counting the $0 ones).

I recently spent $300 on noise-cancelling earbuds. I live in the middle of San Francisco, so it's pretty noisy. I'd tried earplugs and found them uncomfortable and the noise reduction unimpressive. The earbuds have been great for productivity in general (both at work and studying at home). I highly recommend them if you're in a noisy area and can afford it.

it kills your productivity by up to 40%

I would replace "kills" with "reduces." Generally, when you kill something, it is dead, not mostly dead.

Good point, thanks, fixed it.

Move to Main. One learns by rote, not wrote.

This has accumulate a lot of only positive up-votes in a short time and no comments. From my experience this post is very close to a Main post as it is clearly disseminating generally useful and well-backed information.

I propose the following changes which I believe are required for a main post:

  • Look up one or two sequences or other posts for which this could be a follow-up.

  • There are quite a few YouTube links (which makes this very easily exercisable advice) intermingled with pointers to articles and more abstract references. I think it would make a more well-research impression if you could make the sources of your advice more clear. Maybe add a section at the bottom explicitly naming the sources (title and author).

  • Remove the disclaimer (or reformulate it to just clearly name the audience).

Thanks for the pointers; I'll make the changes you've proposed and move it to main at some point over the next day.

Look up one or two sequences or other posts for which this could be a follow-up.

I'm having trouble finding an appropriate post, did you have a particular one in mind?

The sequences indeed do not have much on study and learning (EY is autodidact). At least http://lesswrong.com/lw/3nn/scientific_selfhelp_the_state_of_our_knowledge/ has the following section

Study methods

Organize for clarity the information you want to learn, for example in an outline (Einstein & McDaniel 2004; Tigner 1999; McDaniel et al. 1996). Cramming doesn't work (Wong 2006). Set up a schedule for studying (Allgood et al. 2000). Test yourself on the material (Karpicke & Roediger 2003; Roediger & Karpicke 2006a; Roediger & Karpicke 2006b; Agarwal et al. 2008; Butler & Roediger 2008), and do so repeatedly, with 24 hours or more between study sessions (Rohrer & Taylor 2006; Seabrook et al 2005; Cepeda et al. 2006; Rohrer et al. 2005; Karpicke & Roediger 2007). Basically: use Anki.

To retain studied information more effectively, try acrostics (Hermann et al. 2002), the link method (Iaccino 1996; Worthen 1997); and the method of loci (Massen & Vaterrodt-Plunnecke 2006; Moe & De Beni 2004; Moe & De Beni 2005).

And then I found the following posts outside teh sequences which are at least somewhat relevant:

As study and learning is underrepresented in the Sequences, maybe a sequence for that could be started. Alas I don't know the (social) protocol for doing so.


Get some sleep[a]

Damn, damn, damn, damn, damn. It's not just that I've always been something of a night-owl while all of society around me functions on a "get up with the sun" morning schedule. It's that the night is often the single uninterrupted block of many hours I have for studying or coding.

Also, would someone happen to have tips for dealing with grad-school-level time pressures while maintaining a healthy sleep schedule?

TLDR: I managed to fix my terrible sleep pattern by creating the right habits.

I've been there, up until a month ago actually.

I've tried a whole slew of things to fix my sleeping pattern over the past couple of years. F.lux, conservative use of melatonin, and cutting down on caffeine all helped but none of them really fixed the problem.

What I found was that I'd often stay up late in order to get more done, and it would feel like I was getting more done (where in actual fact I was just gaining more hours now in exchange for losing more hours in the future). Alongside this my pattern was so hectic that any attempt to sleep at a "normal" time was thwarted by a lack of tiredness, I could use melatonin to 'reset' this, but it'd rarely stay that way.

The first thing that helped was sitting down and working out hour by hour how much time I actually have in a week; this prevented me from thinking I could gain more time by staying up later. The second thing was forming good habits around my sleep. Habit's typically follow a trigger-routine-reward pattern and require fairly quick feedback. As a result building a habit where the routine is sleeping for eight hours is quite hard.

Instead I appended two patterns either side of the time I wished to sleep, the first with the goal of making it easier for me to sleep, and the second with the goal of making it easier for me to get up.

The pre-sleep pattern followed:

Cue: 'Hey it's 10:30pm'

Routine:Turning off technology->Reading->Meditation

Reward: Mug of hot-chocolate

While the post-sleep pattern followed:

Cue: Alarm goes off,

Routine: Get out of bed.

Reward: Breakfast.

Since doing this I've been awake at 8 am every morning with little trouble, and the existence of those habits has made easy to add other habits into my routine. Breakfast, for example, is now a cue to go out running on days when I don't have lectures (this is very surprising for me, I've received several comments along the lines of "Who are you and what have you done with the real you" since I began doing this).

I hope you find this useful.

Just occurred to me that I could put TAPs I want to install into Anki! Like the cue could be the thing on the front of the card, and the routine+reward can be the text on the back.


Also, would someone happen to have tips for dealing with grad-school-level time pressures while maintaining a healthy sleep schedule?

I happen to have a) grad-school level time pressures, and b) a healthy sleep schedule. I'd never have done it own my own, it just worked out because I had a bunch of kids. But the kids are actually incidental. What kids do is make it so you get up at the same time every day, because once they're up, they find some way to destroy your house or kill themselves in 20 min. So you have to get up.

What I learned from all this, what I wished I'd known all along, was that the amount of sleep you get is, within reason, not that important. What matters is getting up and going to bed at exactly the same time every day, even on weekends.