How is Ultralearning Different?

Recently, I tried to introduce the idea of ultralearning: deep, aggressive self-education. Since I’ve been discussing it more frequently, I thought it would be useful to clarify more precisely what I mean by it.

To the casual reader this may seem pedantic, but I’m hoping for the more frequent reader of this blog this will help articulate, in my mind, what are the key distinctions between a whole host of related concepts I tend to throw about.

Some Things Ultralearning Isn’t

Part of the difficulty with this new concept is that I’ve spent a lot of time talking about related concepts: efficient studying, rapid learning, regular self-education, active learning, etc. While there’s certainly overlap with these concepts, I wouldn’t say they are interchangeable.

1. Ultralearning Isn’t Just Learning Faster

Rapid learning, or learning something over an usually short period of time (or, alternatively over a relatively few number of total studying hours), is something I’ve written about previously. Trying to pass linear algebra exam in ten days of learning or learning to draw portraits proficiently in a hundred hours are possible examples.

Ultralearning differs from this concept in two ways. First, I would say a key characteristic of ultralearning is that it must be highly self-directed. It’s possible to rapidly learn while following the careful guidance of a teacher or coach, but I wouldn’t call this an ultralearning effort. Second, an ultralearning project could aim for incredible depth and breadth, even if it was pursued over a longer period of time.

2. Ultralearning Isn’t Just Self-Education

Similarly, I wouldn’t say all self-education is ultralearning. In order to qualify, the approach used must be particularly aggressive, intensive and deliberate. Learning physics by grinding through hard problems in practice sessions is an ultralearning approach. Casually breezing through some books on physics probably isn’t.

I would actually argue most self-education isn’t ultralearning. This is simply because, as a recreational activity, most people prefer to learn casually, without a high degree of mental effort.

Side Note: By this measure, my current cognitive science learning project wouldn’t yet be ultralearning, despite my earlier classification. I’m hoping this will adjust as I’m able to more rigorously practice it, but I’m still exploring the best way to do that at the moment.

3. Ultralearning Isn’t Just Learning Effectively

Learning something in an effective way doesn’t mean you’re using ultralearning. I want to maintain that distinction because I feel the contours of what we know about effective learning (deliberate practice, active recall, spacing effect, etc.) are broad enough that they could plausibly host different competing strategies or styles which could be more effective for different people.

For instance, someone slowly learning programming by working their way through a guided programming curriculum and practice projects can certainly learn effectively, even though I wouldn’t describe their pursuit as ultralearning.

What Is Ultralearning?

In short, for some learning effort to fit the concept of ultralearning, I would argue it has to have the following characteristics:

  1. Highly self-directed. It is possible to use ultralearning within a formal curriculum. But what is distinctive about ultralearning projects is that the student is ultimately taking a very active role in choosing what to learn, what not to learn and how to do it.
  2. Aggressive learning tactics. Ultralearning is typified by tactics that dive straight into the hardest and most difficult part. Naturally, these tactics are mentally demanding and often uncomfortable, so many people who are choosing how to learn prefer a more casual approach.

In short, ultralearning is an uncommon strategy because it is hard. It is hard because it requires a lot more planning and self-motivation than less self-directed approaches. It it also hard because the learning approach itself is more mentally demanding than other alternatives.

Why Ultralearn?

Given ultralearning is difficult, why bother doing it? Why not use a different strategy that relies more on guided instruction or less strenuous learning tactics?

The short answer is that because I believe, if you can succeed in getting past the difficulties of ultralearning, you can do some amazing things with it. I’ve spent the last few years exploring some of those things. I don’t think ultralearning is the only factor that made my projects possible, but I do believe it was a necessary one.

The same difficulties of ultralearning are also what makes it powerful:

  1. Self-directedness allows you to learn what you want, when you want, how you want to learn it. Yes, following coaches and teachers can make some things simpler. But often they come with extra restrictions and baggage. If I hadn’t self-educated my computer science background, I highly doubt I would have bothered to go back to school for four years to do it.
  2. Aggressive learning tactics let you learn better and faster. The Year Without English is the perfect example of this. Yes, only speaking the target language from Day 1 is really hard. But, it can cut years off of your learning approach than using the more typical casual book study of languages does.

Who Should Ultralearn?

Ultralearning is not for everyone.

Frankly, most people don’t actually care about learning enough to bother with it. This is true of all self-education, so it’s doubly true for ultralearning.

I can’t tell you the number of conversations I’ve had with people who are planning to travel to a new country, want to learn the language and ask my advice. When I tell them to not speak English, they nod their head, but I can tell they won’t do it. They just don’t care enough. That’s totally fine, but it’s important to admit that.

Some people have told me they can’t ultralearn because they have kids, or a job, or can’t travel (in the case of languages), or are older. This confuses the idea of ultralearning with the specific approach I used.

The heart of ultralearning is a MacGyver-esque ability to work with whatever resources you have available. It doesn’t matter if you only have an hour per week, no access to resources or time. What matters is the desire to construct a self-education program and the desire to pursue it aggressively, even if that means taking mentally uncomfortable steps.

For most of these people, I suspect are like my friends who travel. They want a convenient excuse for not doing it, but perhaps the true one is better: ultralearning is difficult and whatever they want to learn isn’t a priority enough to merit the effort it would require.

However, the reason I write about ultralearning isn’t to persuade everyone to do it. Instead it’s to show the possibilities for the few people who are interested in taking up the challenge. For the people who want to know the thrill of arriving in a country with little linguistic ability and staying committed to learning through immersion. For the people who want to push themselves harder than they ever thought possible on a skill they’ve been stuck with for decades. For those who want to amass entire fields of knowledge in a timescale that allows them to act on it.

I hope to continue writing about ultralearning and featuring more of my own and others’ projects that fit this paradigm of self-education. Not to convince everyone that I’m right, but to encourage a few people to learn more than they think they are capable of.

  • Tom

    Interesting read but I still don’t see how ultralearning differs from learning. Replace the word ultralearning in the article for learning and it seems pretty much the same thing to me. Except when you say: ultralearning isn’t for everyone. This is funny because I think it depends on the subject. Not all subjects can interest everybody. Some people will be more interested than others depending on what they are learning.

  • Shawn

    From what I’ve read, what makes ultralearning different is the method employed. Is The Year Without English the only method to learn? If it isn’t, what are some of the principles of this method that can be analogously applied to another method, for say, someone who can only study the foreign language one hour each day?

    From a marketer’s perspective, most self-help learnings focused on one of these three, or a combination of them: Effectiveness, Adherence, Efficiency. There’s usually a trade off between one of these.

    Effectiveness: pretty self-explanatory.
    Adherence: How likely are you stick to the program you prescribe yourself? Eating cabbages for one week might help you lose 10pound, but how long would the average person adhere to such a program?
    Efficiency: How long does it takes to master that skill? It matters little if you have the best method but take 10 years to learn a skill. In other words, the ROI on time and effort spent.

    The Year Without English might be very effective and efficiency, but because it’s hard as you said, the adherence level is low.

    On the other hand, learning languages in classroom might be fun (high adherence), and it might be somewhat effective, but it takes maybe 4 times longer to master the language. Low efficiency.

    I think what’s different about ultralearning is that it prioritizes Effectiveness and Efficiency over Adherence.

  • Mutasem

    I don’t have any experience with ultralearning and I am only starting to practice self-directed learning. But it seems to me that the high intensity (in time and difficulty) is what gives ultralearning its momentum. Thoughts?

  • See the “What Is Ultralearning?” section.

    Not all learning is highly self-directed. It’s common to learn in a class, where the curriculum is set by someone else.

    Not all learning targets the most difficult part of a subject. It’s easier to repeat the easy parts of a class (e.g., solve easy problems) than to focus on the difficult parts.

    That’s why “ultralearning is not for everyone.” Not because of the subject, but because of the learning methods.

  • Matt Serna

    I’m a huge fan of the ultralearning philosophy, and hope you keep writing more about it. I’ve noticed the correlation between difficulty and efficacy first hand while learning Brazilian Portuguese, learning more in a few days in Brazil spent without speaking English than in over a month of home study beforehand.

    There’s no doubt in my mind that, at least for busy people, this approach makes the most sense. Life is short, and free time is so scarce, so I’d much rather take the “hard route” to learning if it’s going to help me be more efficient with my time.

  • Nice article……..I really appreciated your thinking over ultra learning but don’t you think it is a very difficult process as i am one of the type who believe in last night studies.

  • Bjarke Tan

    Do you think ultralearning could work for personal devolpment? For example overcoming fears faster or improve your social skills faster That would be amazing (it’s still amazing on its own of course)

  • Caro

    I wonder if, for an artist, putting on hold all her usual art tools and art languages of creation and expression for the singular purpose to learn a completely new art language, might qualify.
    For instance, let’s picture a painter wanting to learn music (from absolute beginner level: can’t read, nor hum music); or a musician wanting to learn painting (from absolute beginner level : can’t draw a straight line).
    Then for one year, the artist will have to express herself/himself in the new language. Supposedly the artist can’t be and act without making art.
    Would that qualify as a basis for “Ultra learning”?
    If yes, then. …That would be very interesting to know.
    Thank you very much, dearest Scott, for another supremely enlightening article. And thank you everyone for reading my muse speaking. ..

  • Scott Young

    The goal isn’t what makes it ultralearning, it’s the method.

    As an example: You could learn very casually art through some kind of spontaneous, self-directed practice. That would be self-education, but not ultralearning since it lacks the particular aggressiveness or intensity of approach.

    Alternatively, you could learn intensely, but in a formal art program where you don’t have to think much about how to structure your learning but simply keeping up with assignments. That wouldn’t be ultralearning because there’s little control over the curriculum design.

    This isn’t to say either of those approaches are necessarily wrong. It’s just to limit my definition so when I talk about ultralearning I’m meaning something quite specific instead of just lumping together any learning goal into one category.

  • Scott Young

    I don’t think the concept makes sense if you extend it beyond learning. So if you’re overcoming fears by learning, maybe? But I don’t know what that would look like.

  • Scott Young

    Yes, it is very difficult. But it works. That’s my point.

  • Scott Young

    High-intensity is sort of a hallmark of ultralearning, but I hesitate to make it the defining element (after all, if a lower-intensity strategy actually worked better, I think it would qualify). However, I do feel there’s a tendency to pursue moderate-to-high intensity learning activities as being somewhat more time efficient than lower-intensity ones and that plays out over many very different learning tasks.

  • Scott Young

    So the question of adherence is interesting. I think you could split it into two components: willingness to start and willingness to persevere. That is, there are some goals that sound really hard and people are reluctant to try them, but maybe aren’t as difficult, whereas there are some goals that sound easy but turn out to be very difficult to stick with.

    My feeling is that the adherence difficulty of the no-English rule is actually only moderately difficult if you’re able to travel to another country. However the reluctance to try the method is through the roof.

    I think immersive language learning is a type of ultralearning, but I wouldn’t say it’s the only way to learn a language (or the only approach that would qualify as ultralearning). If you did one hour a day, but it was high-intensity immersive practice instead of low-intensity flashcards/games, I would say that counts.

  • Andrew

    Hi Scott,

    I am currently trying out Ultra Learning on some of the MIT courses such as 18.06 (linear algebra) and 6.041 (probabilistic systems), and I think some people may have the following question as well:

    During phase 2 of ultralearning a subject — how much of the problem sets did you do to improve on weak areas? I noticed that MIT’s midterm questions are slightly easier than problem set questions, but explores the concepts less deeply. However, doing all questions would take more than 3-5 days to complete. So I guess more generally: how did you sample problems to address what you are weak on without over-doing it?


  • Vivian

    I think you can. Make a detailed plan of all the things you avoid and your plan to overcome them. Use flooding instead of gradual exposure. If you have done the difficult thing, the rest becomes easier. if you’re afraid of socialising, force yourself to go to 4 parties and talk to at least 10 people at each party, for example.

    From :

    Graded exposure vs flooding

    Most exposure therapists use a graded approach in which mildly feared stimuli are targeted first, followed by more strongly feared stimuli. This approach involves constructing an exposure hierarchy in which feared stimuli are ranked according to their anticipated fear reaction (Table 1). Traditionally, higher-level exposures are not attempted until the patient’s fear subsides for the lower-level exposure. By contrast, some therapists have used flooding, in which the most difficult stimuli are addressed from the beginning of treatment (an older variant, implosive therapy, is not discussed in this article). In clinical practice, these approaches appear equally effective; however, most patients and clinicians choose a graded approach because of the personal comfort level.

  • Scott Young

    I’m not saying there’s nothing analogous to ultralearning in other domains, simply that the concept loses usefulness once it comes to stand in for very general ideas about self-improvement.

  • Fascinating. Having never heard the term before, it makes perfect sense.

    I wonder if a person who self-learns and studies for career advancement in a career they hate, would this be considered Ultra Learning?

  • Bjarke Tan

    Hi 🙂
    If i use the feynman technique, overlearning, active recall and distributed pratice and trying to learn deeply the way cal newport describes in hes book called deep work on 4 books i have chosen to learn my self and specefic hours i plan to learn 6 days a week would it be ultralearning or just effectiv learning? If it’s just effective learning do you have any suggestions on how i can make it more like ultralearning?(my goal is to be able to use and understand the information over a long period of time :)) Kind regards 🙂