Scott H. Young's Blog, page 3

22 Thoughts on Using AI to Learn Better

I’ve been thinking about (and tinkering with) AI a lot lately. Specifically, I’m focused on how the new tools can be used to learn better.



Ultralearning was premised, in part, on the idea that new technologies can help autodidacts accelerate their learning. Open courseware enabled, for the first time, someone like myself to study from actual MIT classes. Machine translation, Skype tutoring and spaced repetition software transformed the possibilities for learning languages. Even learning to draw by hand can be accelerated if you are able to superimpose your reference images to easily diagnose mistakes.

But those technological leaps pale in comparison to the advances in AI we’ve experienced in the last couple of years. The possibilities for using technology to assist with principles I identified in my 2019 book are vast—meta-learning, on-demand explanations, limitless practice problems, corrective feedback and more.

Given the size of the topic, I’ve been a little reluctant to write about it, as I feel it deserves much more in-depth effort than a single blog post could ever provide. However, as that will take me some time to arrive at, I figured I’d share some tentative thoughts:

Necessary and Unnecessary Difficulty

1. AI can make learning easier. This is a double-edged sword. While it can reduce some of the unnecessary pain of learning by making explanations, examples and guidance more available, it also has the obvious potential to undercut the necessary effort required for learning.

2. Necessary effort includes things like attending to facts, forming a mental model of the situation, practice, retrieval, and noticing gaps in your own abilities and understanding.

3. Unnecessary effort includes things like difficulties due to poor explanations, insufficient practice materials, materials that are too hard/easy, trial-and-error problem solving, lack of feedback, etc.

4. This dual nature means that, like most other tech changes, AI will have ambiguous effects on learning, broadly speaking. Some people will use it to learn more effectively; others will simply use it to cheat on their homework.

5. Furthermore, what constitutes “necessary” effort is both theoretically contentious and poorly understood by learners. Therefore, the risk that even well-meaning students undercut their learning by skipping over the “necessary” parts is non-trivial.

The Value of Learning Skills

6. At the same time AI makes learning easier, it also changes the value of what is learned. Many skills that can be performed well by AI are going to rapidly lose their economic value. Basic illustration, translation, and coding skills are going to be devalued as a near-costless competitor emerges.

7. But this does not mean learning, itself, is a waste of time. In the near-term, many AI-adjacent skills may even increase in value as “centaurs,” or human/AI collaborative pairs, are more productive than either on their own. 1

8. Nor is it the case that, just because an LLM can do a skill, learning it as a human being is a waste of time. To think efficiently, much knowledge and skill needs to be in your brain, not on silicon. Thus, just as teaching children to add numbers has remained a core aspect of mathematics education despite decades of cheap calculators, so too will it remain important to learn a lot of skills that AI can already perform at superhuman levels.

9. This means using AI to accelerate your learning in the skills it is already proficient in is the most obvious use case for AI-assisted learning. Despite the precarious position of professional translation, for instance, this is one of the best times to study a new language.

Verifying AI Output

10. Much is made of hallucinations, but nearly all sources have a degree of fallibility. I remember the moral panic about Wikipedia when it first arrived, but today, Wikipedia is more accurate than pretty much every non-academic source (and even then you’d be alarmed at how often citations are misleading).

11. The easiest way to sidestep hallucinations wholesale is to use AI in a verifiable way. Verifiable prompts produce answers that may have been difficult to generate, but are easy to check. Since P!=NP2 this is enormously useful. For instance, asking AI to give the primary sources used means you can easily check whether the research says what the AI claims.

12. Analogies, metaphors, visual mnemonics and other “help me remember this” types of prompting are also an example of verifiable AI output—here hallucinations are irrelevant because the answer isn’t factual; AI generating dozens of examples of creative “sounds like” imagery for a keyword mnemonic isn’t an issue of accuracy beyond whether the words actually do sound alike. 3

AI May Not Be a Good Teacher, but It Might Be a Great Tutor

13. Currently, a major bottleneck in AI usefulness is the clarity and precision of the prompt it is given. Responses can be vastly more helpful if you’re really clear on what you want. Many disappointments of AI use come from expecting a good one-shot response, though it may take a few iterations on your prompt before you can articulate exactly what you need.

14. This prompt-dependency means that AI can be a great tutor, even though it is a mediocre teacher. For most people, a traditional course/textbook/curriculum will be better than “going blind” into AI and expecting it to teach you in an effective way. But, if you already know your sticking point, then AI is great for filling in the gaps.

15. For example, a key step in the Feynman Technique is to map your understanding of a topic so you can identify a precise question. In the original formulation, that then turned to close reading to try to resolve your confusion, but now it is the perfect time to ask an AI.4

How I Am Using AI for Learning: AI’s Strengths and Weaknesses (So Far…)

16. I now do a lot of my reading “AI assisted,” meaning that when I read a textbook or paper, I query a reasoning model like o3 to ask follow-up questions like if there are updates based on more recent research, the empirical status of claims the author is making and do robustness checks to see if an author’s viewpoint is mainstream or unorthodox.

17. But I haven’t given up on real books yet. Deep Research and other tools can do a decent literature review if you have a niche question, but it’s undoubtedly worse than a book or review paper written by an actual expert.

18. In addition to knowledge and explanations, AI can also help with cognitive skills. It can be hard to find good practice problems and questions, but AI can generate them.

19. I’ve used AI a lot to help practice Macedonian grammar, something that’s difficult to do with flashcards (because you’re learning a rule, not a fact), and which doesn’t have the ample resources that major world languages offer.

20. Still, fine-tuning to get useful output can also be hard. Repetitive practice quickly devolves into loops of highly similar questions/answers upon repeat prompting. Getting an API key and prompting the AI through software is probably more reliable than using the chat interface.

21. Coding agents are getting good enough now that bespoke learning software for specific tasks is approaching the threshold of feasibility from a cost-benefit standpoint. I recently made a script to help my Chinese listening skills by downloading a YouTube video, making a transcript, summarizing it in English and providing a table of key words to help with watching. Something like this would have been incredibly helpful in my intensive language learning periods, but coding it myself from scratch would have taken so long that it would have eaten up most of the benefits.

My Final Thoughts

22. Ultimately, in spite of the new wonders (and worrisome dangers) of AI, it’s important to underline that learning itself hasn’t changed. You still have the same brain, and the rules for how learning works still apply.

In all this, I’ve entirely side-stepped the question of whether AI is “good” as a whole. As someone whose livelihood depends on many of the skills that AI now performs, I’m as uncertain as anyone about the answer to that question.

Still, I take the invention of LLMs and their likely continued progress to be brute facts of reality. As the world changes, the most valuable skill will be learning to adapt.

The post 22 Thoughts on Using AI to Learn Better appeared first on Scott H Young.

Reflection – Day One

I’ve just begun a new month in my year-long Foundations project. This month is focused on reflection, where I’m restarting a daily journaling habit and digging into some of the research on thinking and feeling better.

For those interested, you can check out my previous seven months here:

Fitness: Start, End, Books.Productivity: Start, End, Books.Money: Start, End, Books.Food: Start, End, Books.Reading: Start, End and Books.Outreach: Start, End, Books.Sleep: Start, End, Books.Why Journal?

The keystone habit for this month is to write in a journal daily. To prevent this from becoming an  onerous commitment (especially in light of the other eleven habits I’m working to maintain), I’m not holding myself to any minimum amount. Writing a single word is fine on days when I’m super busy. Of course, I expect I’ll do more than that most days.

The motivations for journaling are manifold:

Writing makes you smarter. Working memory, the ability of the mind to hold simultaneous ideas, is strongly linked to intelligence—and is famously limited. Writing liberates working memory by allowing ephemeral thoughts and ideas to have a permanent holding place. While the idea that writing makes you smarter may sound hyperbolic, consider multiplying three-digit numbers in your head versus on paper to see that it is clearly true.Writing allows you to capture thoughts. A major tenet of Cognitive Behavioral Therapy (the current gold-standard therapy for depression, anxiety and more) is that we engage in automatic thoughts that produce feelings and actions, often resulting in a vicious cycle where negative thoughts create negative feelings and actions, which then reinforce negative beliefs. Usually that loop occurs without much self-reflection or awareness. But if you write thoughts down, you now have an external object you can interrogate, asking yourself, in essence, “Wait, do I really think that?”Writing helps you solve problems. The reason the math problem is so much easier on paper than in your head is that writing allows you to offload components of a complex puzzle so you can organize a solution. But this is true of non-math problems too! Often, our gut provides an intuitive answer, but if you want a reflective answer, writing can facilitate much better plans.

Of course, these are all benefits of fairly unstructured journaling—where you simply write down your train of thoughts and see where it goes. Journaling can be made more specific, ranging from a daily record of events to meditations on specific themes like gratitude or empathy.

My Experience with Journaling

Journaling is a practice I’ve done on and off for my entire adult life. In some ways, this blog is an extension of that practice. My inner life has been profoundly shaped by both my personal and public writing.

I’ve tended to rely more on journaling when my problems in life have been more difficult to solve and when I’ve had fewer people to discuss them with. (Conversations can also serve a lot of the same functions as solitary reflection, albeit with different strengths and weaknesses.)

However, like many good habits, it’s easy to let something that works slip because you get busy or simply forget. I’m hoping bringing back journaling, especially in a low-key way that doesn’t mandate a heavy time commitment, will help me deal with some of the real problems in my life as well as the more imagined problems where my own reaction to the situation is truly at issue.

I’ve written before that I’m a somewhat anxious person by temperament—I tend to plan and worry by nature. However, many of these idle worries are completely useless, neither producing action nor alerting me to an unseen danger. I don’t expect journaling regularly will eliminate these emotions, which presumably come from a non-conscious part of my mind, but I do think it will help me catch myself when I get absorbed in some current worry and break out of the cycle of unhelpful thinking.

I also want to use journaling to recapture some of the original motivational benefits it brought me in my younger days. Writing about my goals and vision for life helped me reinforce a plan of action. Too often, I get into a reactive mode where my daily decisions seem to be made out of necessity, rather than being open to all the possibilities.

Reflection, Beyond Writing

Of course, journaling is just one way to adopt a reflective practice. There are others:

Mindfulness meditationConversations with close friendsCreative self-expressionTherapy

I’m choosing journaling for both myself and the people following along in the course because I think that, of the techniques for reflection, it requires the least training to do it well. Meditation, in contrast, is probably best learned through a retreat. Not everyone can really tap into their self-awareness with art without a lot of training. Similarly, conversations can be delicate if you don’t have the right person to chat with, and therapy is not accessible to many.

But I’m hoping that my reading and research for this month will extend beyond this particular technique. I want to use this month as an opportunity to re-read some classic works in the psychology of thinking, cognitive therapy, mindfulness and more.

As always, if you want to join along with me for the month, the keystone habit is simple: Get yourself a paper journal (or open a new document on your computer/phone) and write in it every day. Even just one word counts, so don’t feel bad if you don’t have time to get to the bottom of all your life issues at the end of a tiring day.


Toward the end of the month, I’ll share some notes from the books I read and reflect about my own reflections!

A Little Update

Although this month’s focus is reflection, I thought I’d share a little update to one of the previous foundations I have been working on: fitness.

Last Sunday, I finished my first-ever marathon!1

This represents a major improvement from last year, when I would have struggled to finish a ten kilometer run, never mind a bit over forty-two. Running a marathon had always been on my bucket list, but it wasn’t until I started exercising daily that the idea became more than a distant daydream.

The post Reflection – Day One appeared first on Scott H Young.

Sleep – Month-End Update

As I write this, I’m reaching the end of my month-long effort to sleep better. This is part of my year-long project to improve my foundations in life, following previous months dedicated to fitness, productivity, money, food, reading and outreach.

Overall, I’m pleased with how the month went. According to my Fitbit, the average time I’ve spent actually asleep this month was seven hours and fourteen minutes, compared to my average for the previous six months of six hours and forty-eight minutes. This is an improvement of twenty-six minutes per night.

Twenty-six minutes may not seem like much but, subjectively, the difference has been dramatic. I’ve felt a lot better most days, which has made it easier to concentrate at work—and hopefully made me a bit more pleasant to be around.

Sleeping Better: What Worked for Me

I started this month fairly pessimistic about improving my sleep. As I mentioned in my introduction, I have two small children, so mid-night interruptions in sleep and early awakenings can make sleeping a regular eight hours difficult. I had largely resigned myself to accepting I wouldn’t sleep well until our kids are a bit older.

While sleep this month was a bit easier than it was the previous month, the frequency of interruptions didn’t actually change all that much. As I review my sleep journal, I was woken up at least once before 5:30 a.m. on nearly 50% of the nights this month. (Although, fortunately, our daughter managed to get back to her typical 5:30-6:00 a.m. wake-up time, after last month’s bout of 4 a.m. wake-ups.)

Instead, the improvement in my sleep was really simple: I became deliberate about going to bed earlier. Instead of lingering in bed until 10:30-11:00 p.m., I set a curfew for myself not to watch television or use any devices after 9 p.m. This led to me going to sleep about half an hour earlier, which in turn meant roughly half an hour more sleep each night, interruptions notwithstanding.

“Go to bed earlier” is such brain-dead simple advice that I think the fact that I didn’t earnestly try it until now requires some self-reflection.

I think my failure to apply this advice came from a few places:

Drinking too much coffee. I had always told myself that caffeine didn’t impact my sleep, in part because when I did try to sleep I fell asleep quickly. But it may be that my afternoon coffees were bumping back my perceived sleepiness by half an hour or more.Prioritizing entertainment over sleep. Like many, I was tempted to watch one more episode, check email one last time or otherwise stay up a little later than I should. Especially after a busy day, it was often hard to cut short my entertainment time to get to sleep. Having the curfew limit has been enormously helpful in removing the temptation to procrastinate on going to sleep.I had a bad attitude. In retrospect, it was easy to point to my kids’ sleep habits as the culprit in my own sleep difficulties. And while these may have been the proximate cause of some poor nights of sleep, my fixation on this prevented me from recognizing how much of my sleep was under my control.The Limits of Sleeping Better

While my improved sleep has been a welcome change, doing the research for this month also helped me shift my attitudes toward sleep.

One idea that I found particularly helpful was Jade Wu’s comment that tiredness is not the same as sleepiness. Sleepiness is a lack of sleep. Tiredness is a lack of energy. The two are not interchangeable.

Sleepiness means you need sleep—and you will sleep if given the opportunity. Tiredness, in contrast, can come from stress, boredom, mental overload and more—it means you could use a break, but if you don’t fall asleep right away when given the opportunity, you’re probably not sleepy.

Reading Wu’s book, Hello Sleep, also helped me recognize that I had some unhelpful beliefs and attitudes about sleep that were exacerbating my problems. In particular, I had the notion that any time I wasn’t brimming with energy, the issue was due to not sleeping well enough in the night.

Now, I realize this was probably not the case. While there were definitely some days in my past where I could have dropped into a deep nap, if given the chance, there were probably also a lot of days where my energy levels would have perked up almost instantly if I were to switch to hanging out with a friend, watching a fun television show or doing an activity I really enjoy.

Of course, if you’re extremely tired (but aren’t technically sleepy) such a distinction may seem like semantics. But I found it helpful, because it suggested that sometimes I needed to look elsewhere from just my sleep itself to fix moments of low energy. Sometimes what I needed was to get outside, chat with a friend or do something fun.

Similarly, I used to get a bit of frustration surrounding my sleep quality. My deep sleep scores provided by my Fitbit were always under my desired benchmark. But the accuracy of those metrics is fairly low, so I’m learning to trust that my body gets the rest it needs. If I needed more deep sleep, I’d get it, so worrying about sleep quality is a waste of time.

Plans for Future Sleep

Above all, my biggest takeaway from the month was that my sleep will be good if I make it a priority, and that sleeping well is under my control, even if I don’t always get to choose the ideal conditions to make it happen.

I don’t plan to be obsessive about forcing an early bed time on all occasions, especially if that means missing out on opportunities to socialize with people who don’t go to bed at 9:30 p.m. But I do think my decision to cut out screen time after 9 p.m. forced me to realize how much I had been trading off sleep for entertainment of dubious value.

I also plan on sticking with my one-coffee-per-day rule, keeping it decaf after that. While the allure of a mid-afternoon pick-me-up is appealing, I think I’m better off without it.

Finally, I’m going to be better at distinguishing genuine sleepiness from the desire to take a break in my day-to-day life.

Updates on Other Foundations

Some quick updates on the previous foundations:

Fitness — Continuing to go well. In seven months, I’ve only missed two days. (Although I’ve allowed myself some easy options, like stretching, on days my body needs to recover.)Productivity — The full-capture system is still going well. There’s always more to do than time to do it in, but I’m pretty happy with how my current system keeps the stuff that really matters from slipping through the cracks.Money — Just finished a big annual review and budget for the year.Food — I’ve been pretty good about sticking to the dietary changes I started previously. My weight has been holding steady at around 15 lbs. lighter than my weight at the beginning of the year.Reading — The research for this project has kept me busy, so I expect to keep on track to finish 8 to 10 books each month.Outreach — Fairly successful, although there are still improvements to be made here. I think being more rigorous about tracking might help.

That’s it for this month. Next month, I’m focusing on reflection, which is a month dedicated to thinking about thinking.

The post Sleep – Month-End Update appeared first on Scott H Young.

Eight Books for Getting a Better Night’s Rest

I’m wrapping up the seventh month of my year-long Foundations project. This month’s focus is sleep. I’ll share some personal reflections in the next post. In this post, I’ll share the books I read and key takeaways on sleep.

For those interested, you can also read my updates and book lists from previous months here:

Fitness: Start, End, Books.Productivity: Start, End, Books.Money: Start, End, Books.Food: Start, End, Books.Reading: Start, End and Books.Outreach: Start, End, Books.The 3-Minute Summary of What I Learned This Month

Sleep matters. Modest sleep loss impairs cognition, mood and immune system. Total deprivation is lethal.

Despite that, we devalue sleep culturally. We valorize those who push through sleep deprivation—even though many cognitive tests show that twenty-four hours without rest impairs us as much as being legally drunk.

Sleep is much more than simple unconsciousness. Our brain is highly active when we sleep. Spindle events transfer memories from the hippocampus to the cortex. Glial cells shrink allowing cerebrospinal fluid to rinse the brain and remove waste products. Dreams occur throughout sleep with the most vivid and bizarre occurring during REM sleep toward the end of the night, although we remember few of them because circuits for memory are actively suppressed.

Our need to sleep is driven by two processes. The first, Process S, accumulates sleep pressure throughout the day. It is probably regulated through adenosine. The other, Process C, cycles between day and night. Process C is signaled with melatonin and requires daily doses of bright light during the day and darkness at night to calibrate.

Ideally, these two processes synchronize to make you sleepy at night and awake during the day. But caffeine (which temporarily blocks adenosine receptors) and indoor lighting can mess with these basic rhythms, making it harder to fall sleep.

Insomnia is not sleep deprivation. If you are sleepy enough, you will sleep. The most successful therapy for insomnia is actually to limit how much time you spend in bed—forcing you to accumulate enough sleep drive to make it possible to fall asleep quickly. Worrying about sleep is one of the best ways to ensure you struggle to sleep well.

Scientifically, we still don’t know why we sleep, though there are many possible theories. The lack of consensus may be because sleep actually performs many different physiological functions that have all co-located to our period of daily dormancy.

Notes on the Books I Read About Sleep


Side note: I didn’t include Matthew Walker’s popular book, Why We Sleep. See this footnote for an explanation.1

1. Hello Sleep by Jade Wu

Insomnia is when you desperately want to sleep but can’t fall asleep or stay asleep.

The most successful treatment for insomnia is somewhat paradoxical. Instead of trying harder to sleep more (which causes you to stress out about sleeping, making it harder to fall asleep), you do the opposite: limit how much time you spend in bed.

This forces you to build up sleep drive, making it easier to fall asleep. It also helps you condition the association between your bed and sleeping, instead of your bed and television, social media or, worst of all, *trying* to fall asleep.

One of Wu’s most surprising arguments was that those who suffer from insomnia aren’t, in general, sleep deprived. They may have poor quality or consistency of sleep. They may feel tired and awful. But they usually aren’t sleep deprived—otherwise they would naturally fall asleep.

It seems like part of insomnia is itself sustained by the anxiety you have over your own sleeping patterns. Trying too hard to sleep well can get in the way of sleeping well!

2. Wild Nights by Benjamin Reiss

What’s the “proper” way to sleep? If you’re like me, I’m guessing you imagine something like this: you turn off the lights, go into a room alone or with your significant other, then you fall asleep for eight hours straight, rising to an alarm clock. This is followed by an uninterrupted sixteen hours of activity before you sleep again.

Reiss points out that this modern ideal of sleeping is actually incredibly weird when compared to how most human beings have traditionally slept.

Far more common was the pattern of having two sleeps: going to sleep for a few hours, waking up for some quiet mid-night activity, then going back for a second sleep. Naps, too, were remarkably common cross-culturally. And historically, sleep was almost never a private affair—whole families, including guests, typically slept together in the same bed.

Our modern angst about sleep can be traced back to the invention of indoor lighting, which severed our sleep-wake cycle from the sun.

In his book, Reiss explores more of our sleep-related weirdness, from the parent-sleep-industrial complex, to the deeply intertwined history of sleep and racism, and even the thesis that Walden was precipitated by Henry David Thoreau’s difficulty getting a good night’s sleep.

3. The Neuroscience of Sleep and Dreams by Patrick McNamara

Why do we sleep? Science has a lot of theories, but is far from a consensus over sleep’s primary purpose.

This itself ought to be an amazing fact. Compare sleep to nutrition. We’re far from having achieved perfect knowledge of nutritional science, but there’s no controversy amongst experts over why we eat.

Sleep is evolutionarily expensive. It requires an animal to forego feeding, exploring or mating for hours at a time—all while putting itself in a position that makes it vulnerable to predators. Despite that, virtually all animals sleep. Dolphins sleep one half of their brain at a time so they won’t drown, and bears must rouse themselves from their hibernation in order to allow their brains to sleep.

 Perhaps the lack of knowledge here is disappointing. But in our modern age where everything seems to have long been discovered, theorized and explained, I find it exciting that such a basic question about human existence is still shrouded in mystery.

4. Sleep: A Very Short Introduction by Steven Lockley and Russell Foster

I found this to be a good general summary of sleep science, from Process S and Process C to the four stages of sleeping and what we know about their functions.

One of the most interesting tidbits in this book concerned the best practice for overcoming jet lag. The key is to carefully time bright light exposure to nudge your internal clock in the desired direction. In general, evening light delays your clock and daytime light advances it.

For instance, a traveller leaving New York at 7 a.m. and landing in London at 7 p.m. (local time) after a 7-hour flight is five hours “behind” the English time zone, i.e., his body still thinks it is 2 p.m. Light exposure during the latter part of the flight or when he lands in London, however, will work to delay his internal clock, which is opposite of the adaptation needed. Therefore, minimizing light exposure in the latter hours of the flight and upon landing will hasten his adaptation to the new time zone.

Tragically, the necessity of light exposure to entrain this daily rhythm means that most completely blind people suffer from a desynchrony of their internal clock, flipping completely out of phase every month or so, as their suprachiasmatic nucleus does not get the needed pulse of light to keep it aligned with the Earth’s rotation.

5. Dreaming: A Very Short Introduction by J. Allan Hobson

Dreaming is mysterious. It is also poorly understood.

Some of this is probably due to the difficulty of the subject. Dreams are difficult to remember, control and interpret. Sleeping in a lab distorts the kinds of dreams we have, and MRI machines are noisy, uncomfortable and expensive, making the bread-and-butter neuroscientific technique used for studying countless other mental phenomena difficult to apply.

But some of our misunderstanding also seems to be a direct result of the pernicious influence of Sigmund Freud and his followers’ ideas about dreams. For decades they had a stranglehold on theories of dreaming, despite many of Freud’s tenets being directly contradicted by data.

Unbelievably, rapid eye movement (REM) sleep wasn’t discovered until the 1950s. Anyone who has watched someone as they were sleeping could have made such a discovery.

Hobson persuasively argues that the strangeness of dream content has distracted us from dream form. We focus on the unusual things that happen in dreams, searching for explanations to their meaning, rather than documenting the way dream consciousness systematically differs from waking consciousness (highly visual, poor self-awareness, etc.). Hobson then links these formal features of dreams to the brain systems that are selectively activated or downregulated during dreaming, arguing that the actual content in dreams is largely meaningless.

6. The Unappreciated Power of Naps with Jade Wu

In this Audible-only lecture series, Wu (also the author of Hello Sleep) shares research surrounding the power of naps.

Citing a range of studies, Wu documents that naps can boost our mental and physical performance, suggesting that napping, long taken as part of cultural tradition in places such as China and Spain, has been underrated in our modern expectation of sixteen-hour wakefulness.

Still, napping can be a double-edged sword. Sleep too long or too deeply and you can find yourself too groggy afterward to function effectively. Naps can also eat away at your sleep drive, making it hard to fall asleep at night.

To nap effectively, therefore, Wu argues that we should follow the rule “quick and dirty, under thirty”: keep naps less than half an hour to prevent entering into slow wave sleep. Wu also notes that naps work best when done consistently, on the same schedule, like regular sleep.

7. The Secret World of Sleep by Penelope Lewis

While we still may not know the full purpose of sleep, it is fairly clear that, in complex organisms, learning is one of the primary beneficiaries of a good night’s rest.

Lewis, a neuroscientist, reviews the basics of how learning works in the brain and then shares the science of how sleep works to consolidate, enhance, prune and selectively delete memories. She shares research showing that sleep helps with reinforcing important information, strengthening skills, abstracting away details and forming creative associations.

She also pokes holes in Hobson’s theory that the contents of dreams don’t matter, suggesting a few possible roles for dreams ranging from dialing down the emotional content of our memories to managing our social lives.

8. Caffeinated by Murray Carpenter

The flip side of our cultural denigration of getting sufficient sleep is the elevation of the chemical that makes us think we don’t need it. Caffeine has long been our society’s drug of choice, with coffee, tea, and sodas enabling us to push through sleep deprivation.

As a drug, caffeine is difficult to judge. On one hand, it has remarkable benefits for cognition and physical performance, minimal side-effects and a relatively high toxic dose.

On the other hand, caffeine is a drug that has a physiological effect on the human body. Furthermore, it is reinforcing, which means we are inclined to keep consuming more of it.

Carpenter makes a provocative analogy between the liberal addition of caffeine to sugar-containing beverages and the addition of nicotine to cigarettes. The former are implicated as a major contributor to the obesity epidemic, and it is no coincidence that soft drink companies find it useful to add this mildly addictive substance to their products.

At the same time, caffeine is barely regulated, working its way into not just the traditional coffees and colas, but juice, supplements and even chewing gum. We’re engaging in a worldwide experiment with upping our caffeine intake, and I think Carpenter is right to be at least slightly alarmed at the possible downstream consequences.

_ _ _

That’s it for books this month. Next week, I’ll post some reflections on my personal efforts to sleep better this month!

The post Eight Books for Getting a Better Night’s Rest appeared first on Scott H Young.

Rapid Learner is Now Open for a New Session

About once per year, I offer a new public session of my course, Rapid Learner. This is a six-week program that is designed to make you a better student, professional and lifelong learner. If you’ve found my essays on the science of learning helpful, or found my learning projects (MIT challenge, The Year Without English, Foundations) interesting, this course shows you how to do it.

Click here to sign up for Rapid Learner.

This is the new edition of the course, which includes 20+ newly-recorded lessons, deep dives, walk-throughs and more. If you’re looking to get better at learning difficult things, this course is the place to start.

This course is designed for students who want to do better in school with less stress and studying, professionals who need to master hard skills to get ahead and people who just like learning and want to do it better. The course is divided into six weeks where you’ll learn:

How to set up a learning project you’ll actually finish.How to manage your time and energy to learn, on top of your work.The ingredient common to all successful learning efforts.The process to deeply understand any idea.How to remember anything.How to master and maintain everything you learn.

If you have any questions about the course at all, please email me directly and I’ll do my best to answer them!

Registration will only remain open until midnight on Friday, April 18th, 2025, (Pacific time).

The post Rapid Learner is Now Open for a New Session appeared first on Scott H Young.

Lesson 4: Happiness is the expansion of possibility

In the prior three lessons, I’ve shared how to become a project finisher (not just a starter), how to design effective learning projects, and finally how to create flow so learning becomes a joy, not a grind.

Today, I want to step back and consider the question of why to learn in the first place. Learning is not just a means to a better job, fitter body or more interesting hobby (although it can do all those things) but one of the ends of life itself.

What Makes You Happy?

The exact shape of a life well-lived has been debated for millennia. While we may never have a “final” answer, many different philosophies put an expansion of consciousness at the apex of any hierarchy of human needs.

Abraham Maslow called it self-actualization. Aristotle called it eudaimonia. Socrates thought only the reflective life was worth living. And throughout the ages, many other sages have placed inner cultivation above material rewards and social status.

Life, at its best, isn’t about mere survival and status, but self-expansion. When we see our possibilities in life expanding, we are happy with our lives. When we see them static or contracting, life feels dull—even if we have wealth or social standing.

Learning is the process of self-expansion. It is the expanding of the mind to encompass new possibilities, and with new possibilities come vibrancy and enthusiasm.

If Learning Makes Us Happy, Why is Studying Hard?

Humans are intrinsically curious. We enjoy learning. We seek out experiences of mastery, growth and new ideas.

Unfortunately, I think the process of learning has often been dominated by institutions that thwart our natural curiosity. Schools foster competition among students. Credentials wall off opportunities to a narrow class. Sports pick winners and discourage losers.

This socialization means that, for many, we begin to associate “learning” with pain and frustration. Our innate curiosity remains, but we push this urge to grow and expand into other areas: video games, social media, television shows, “safe” areas where we can feel some facsimile of our desire for self-expansion without the threat of discouragement or failure.

Entertainment is fine. But just as watching Friends is a poor substitute for having friends, limiting all your experiences of mastery to video games, or all of your thirst for knowledge to social media gossip is, fundamentally, a limit on how your self can expand.

Rekindle Your Desire to Learn

My life’s work tries to help people rediscover their original curiosity. To get excited about learning, in the knowledge that they actually can finish the projects they daydream about. You actually can learn French, programming, or ballroom dancing. You actually can change your career, improve your health, transform your relationship, or understand your own psyche.

A major obstacle to rediscovering curiosity is simply lacking the right tools: your projects never seem to get off the ground, you try to get better but constantly get stuck, you’re beset by doubts about your own talent, aptitudes or even motivations.

I’d like to be your guide in changing your relationship with learning. I don’t just want to show you a new technique for memorizing facts, or a strategy for practicing skills (although there are plenty of those). Instead, I want to show you a different way to live: less focused on achieving a specific goal, and more on cultivating a continuous process of becoming a better you.

On Monday, I’m going to be opening my popular, six-week course, Rapid Learner, for a new session. I hope you’ll join me in building the tools for a more curious life, together.

The post Lesson 4: Happiness is the expansion of possibility appeared first on Scott H Young.

Lesson 3: The surprising secret to flow

This is the third lesson in a four-part series on how to design and finish successful learning projects. I’ve written it in anticipation of my six-week course Rapid Learner reopening on Monday. In the first lesson, I shared how to go from being a starter to a finisher. In the second one, I shared the keys to effective meta-learning, or learning how to learn a particular skill.

Today, I want to talk about actually doing the work. What separates a delightful and fun process of learning from an endless grind? What, in other words, separates those who excel at doing the work of learning from those merely daydreaming about mastering it?

Flow: Between Boredom and Frustration

The late psychologist Mihaly Csikszentmihalyi pioneered the concept of flow to define the state of pleasurable absorption in a mentally challenging and meaningful task. 

Learning demands concentration and has the potential to be quite meaningful, both of which are needed to produce the flow state.

But more often, learning produces the opposite: boredom and frustration. Boredom fuels our desire to escape—think of constantly glancing at the schoolroom clock as you wait for the bell to signal a lecture is finally over. Frustration occurs when there is no progress, despite effort; it happens when we’re working hard but feel completely stuck.

Flow matters because we’ll only do the work of learning if the joys of learning consistently exceed the pain.

A learning project may have high extrinsic rewards (better job, health, or relationship), or even delayed intrinsic rewards (satisfaction, pride, a new hobby). But if its moment-to-moment experience is highly negative, it will be difficult to sustain in the face of alluring, low-cost alternatives like games or social media.

In contrast, if the experience of learning is pleasurable (or at least not too unpleasant), you’ll be able to sustain it much longer. This is the secret behind tools like Duolingo, which is a much more enjoyable tool to learn languages than most alternatives.1

How to Flow

How can you create and sustain a flow state in learning? 

The answer is simple: flow occurs when tasks are not too hard (frustration) and not too easy (boredom). Add to this clear and visible markers of progress, and even superficially mundane activities can become addictively compelling.

This is what game designers do when they want to create compelling games: set the difficulty just right and create constant progress markers to sustain engagement.

Games can be optimized entirely for flow. But learning projects, at least effective ones, must also be designed to acquire the knowledge you actually need. This other purpose often puts constraints on how you maximize flow in your learning.

However, assuming you have a more compelling reason to learn than simply filling time, we don’t need to make learning as compelling as a video game to get it done. All we need to do is make sure it is compelling enough so we don’t have difficulty doing the work consistently.

In general, there are three tools you can use to improve the flow in your learning project:

Flow Enhancer #1: Make boring parts more engaging

Boredom is one failure mode of flow. If you’ve ever gotten stuck when reading a dull textbook and found yourself drifting off as you re-read the same passage, you’ve seen how boredom can derail learning.

Finding a more expressive lecturer or better author to read is one solution. But even if you can’t change your materials, you can make processing them more engaging by changing the technique you use while reading them.

For instance, by keeping a stack of index cards and writing a few bullet points summarizing each page after you read it, you take a completely passive activity and turn it into a more active one. This assists with learning, but also makes the activity itself less boring.

Flow Enhancer #2: Make the frustrating parts easier

Frustration is the other failure mode of flow. It occurs when you’re working on something that is beyond your current skill level, so you’re experiencing frequent negative feedback signals from the environment.


While some frustration is almost inevitable in a learning project, it isn’t a necessary emotion for learning. The more we can reduce frustration in our efforts, the more enjoyable we can make learning—without sacrificing effectiveness.

For instance, if you’re working through a set of problems and solutions, you can adjust the difficulty level dynamically to keep it challenging, but not frustrating. If you can’t solve the problems alone, try solving them with a hint. If you can’t do that, try omitting one of the steps of the solution and see if you can fill it in. If you can’t do that, try explaining the solution to yourself as you read through it. By dynamically adjusting difficulty downward to where it is challenging yet doable, you can increase flow, decrease frustration and make it easier to stick to the work.

Flow Enhancer #3: Make progress more visible

Beyond choosing the right difficulty, it helps to break your learning into small units you can make steady progress on. Then make that progress conspicuous in your learning efforts.

Flashcard softwares do this well by giving you a constantly increasing counter showing “cards learned” so you can watch your progress inch forward. But you can do this for almost any variable that is associated with progress:

Number of problems solved.Pages read.Sketches made.Exercises completed.Rehearsals performed.

Choose a variable that always improves (setbacks are ~3x as costly to motivation as forward progress is helpful), is fine-grained enough that you see it go up every time you study, and reliably matches the learning activity you’re actually trying to accomplish. 

Then, make sure this variable is highly visible in your actual learning efforts. If you don’t “feel” its presence when you’re studying, it won’t work to motivate you.

_ _ _

In my full course, Rapid Learner, I’ll go through much more than flow in the equation of how you can get your learning done—from productivity systems, energy management and the nitty-gritty of studying. On Monday, I’ll be opening the course for a new session. In the next lesson, the last in this preview series, I’ll talk about imagining a life devoted to learning new things, and why it’s the key, not just to success, but to happiness itself.

The post Lesson 3: The surprising secret to flow appeared first on Scott H Young.

Lesson 2: The three questions for effective meta-learning

In anticipation of a new session of my popular course, Rapid Learner, I’m sharing a four-part lesson series on how to create successful learning projects. In the first lesson, I distinguished the key differences between project finishers and project starters.

Today, I’d like to discuss how to make the projects you undertake effective. How can you ensure that the projects you work on actually help you get better at the things you’re trying to get better at?

To succeed in our efforts at improvement, we need two types of knowledge:

Knowledge of the domain. In Spanish, this includes the words “agua” and the difference between “estar” and “ser.” In programming, the difference between a variable and a constant. In calculus, how to take a derivative.Knowledge of how the domain is organized. Learning Spanish requires learning vocabulary, grammar and pronunciation. Learning programming requires remembering syntax and having a mental model of what each line of code is doing. Learning calculus requires memorizing transformation rules, as well as developing a spatial intuition about rates of change.

Meta-learning is the second kind of knowledge. It is your “map” of a domain that you use to plan what you intend to learn. And, just as in real navigation, when you have a good map, you’re less likely to wander aimlessly on your way to your destination.

To create this map successfully, there are three major questions to keep in mind:

Question #1: How do other people draw the map?

Learning anything on your own has a bootstrapping problem. Learning and meta-learning are intertwined, so the less we know about a subject, the less we know about how that subject is organized. 

This leads to our first principle: before embarking on any new learning project, try to figure out how other people organize efforts to learn it. What do those people see as the major divisions in skill, knowledge and practice?

Is the skill taught in school? If so, how do people organize curricula around it? Is it taught in books, courses or apprenticeship programs? How do those typically proceed?


Fortunately, research here has gotten much easier. Google and, more recently, LLM tools like ChatGPT can offer starting points for understanding novel topics. You can now get a glance at how experts break down a topic with just a few clicks.

Question #2: What facts, concepts and procedures must be acquired?

Once you have a rough map, it’s time to fill in the details. All learning is composed of a few basic “atoms” of knowledge in three main types:

Facts—things which need to be memorized. Words, terms, formulas, dates, jargon.Concepts—things which need to be understood. Principles, ideas, narratives, explanations.Procedures—things which need to be performed. Methods, steps, actions, techniques.

This applies across fields. Learning Spanish breaks down into facts (vocabulary), concepts (such as the difference between estar/ser), and procedures (like grammar and pronunciation). Learning programming breaks into facts (like syntax), concepts (design patterns), and procedures (like how to write a for-loop). Learning calculus has facts (e.g., cosine is the derivative of sine), concepts (such as what a limit is), and procedures (like the chain rule).

This framework transforms an otherwise amorphous process of improvement into, essentially, a to-do list. To learn a skill you simply need to perform the techniques that are useful for learning each type of knowledge, such as memorization for facts, explanation for concepts, and practice for procedures.

Question #3: What fine-tuning and feedback are needed?

Drawing the map and identifying the “atoms” of the skill you want to learn is powerful, but it is rarely sufficient. Maps never fully reflect the territory—what looks like a straight road may unexpectedly be a dead-end. What appears to be a simple concept or procedure may disguise enormous tacit knowledge and intuition.

Simply checking off items on the to-do list can backfire if the way you are learning those “atoms” doesn’t match with how they need to combine in real skills and situations. As such, the process of successful learning requires constantly rewriting the map and reshuffling the to-do list. You learn a bit, see what’s working and what isn’t, and adjust your approach accordingly.

This means good learning projects require two activities in parallel:

Drills to work through the “atoms” of learning you’ve identified. This means reading books, memorizing terms, practicing procedures and understanding the essential concepts.Direct practice both to test how closely your learning efforts match reality, and to integrate and combine the “atoms” in ways that are useful. This includes solving problems, writing code, or using the language to communicate. Direct practice means applying the skill you are learning to real life.Bringing it all together

The end result is a learning project that is both practical and flexible.

Practical, because it is grounded in expert understanding of how the skill or subject is organized, and because it has been broken down from vague notions of “improvement” into specific facts, concepts and procedures.

Flexible, because by combining drills to master the atoms with doing direct practice to test out your map, you ensure you can make any needed changes to your project so those atoms combine successfully into real skills.

There is, of course, more to this process than I can explain in one lesson. This was a major motivation behind developing my full, six-week course Rapid Learner. Teaching the art and science of meta-learning, this course will guide you through how to design projects to master any skill or subject you care about. I’ll send out registration information on Monday. In the next lesson, I’d like to turn from meta-learning to productivity. After all, a plan is only paper. Effective learning means execution—how can you go from someone with great plans, to someone who consistently gets the work done?

The post Lesson 2: The three questions for effective meta-learning appeared first on Scott H Young.

Lesson 1: How to start projects you’ll actually finish

If I was to describe one practice that, when sustained over time, could produce the greatest benefits for your life, it would be this: regularly starting—and completing—meaningful projects.

Projects are bigger than individual tasks, but still aim to complete some concrete set of actions. A project can be all sorts of things. It can be a goal to start exercising, a plan to learn French, moving to a new city, or getting a new job.


Most of us have many projects, including the active strivings of our daily life, as well as daydreams about hobbies we’d like to learn, sports we’d like to try or career moves we might want to make.

But even though we all have projects, there is an enormous difference in our propensity to complete them.

Some people are finishers. Their life is a steady conveyor belt of realized goals and projects.

Other people are starters. They have a lot of enthusiasm for new efforts, but somehow fail to reach a conclusion with most of the efforts they begin.

If you feel like you’re more of a starter than a finisher, that’s not a reason to be alarmed. Indeed, I have a lot of sympathy for you; I used to be a frequent starter and infrequent finisher myself.


And starters should be proud. Even though you may not complete all your projects, there is at least a strong desire to do things and enthusiasm for life. It’s far better to be constantly seized by ideas for your life than to have no aspirations at all.

However, in both my personal experience and work with students over the last two decades, I know that becoming a consistent finisher is not always easy, even when you know it can produce dramatic benefits over time.

So today, let me share with you a three-step process you can use to become a more consistent finisher:

Be specificBe brief Only one project at a time

1. Be specific

Projects that are more specific are also more successful. The raw materials that spark our motivations are often sprawling and vague. It takes work to shape them into a format we can make progress on.

Consider the following projects, ordered from most general to most specific:

I want to be healthier.I want to get in shape.I want to start exercising more.I want to start running.I want to run every day.I wake up early and run every morning.

Each rung you ascend on the ladder of specificity makes your project more actionable, and more likely that you’ll successfully complete it before moving on.

2. Be brief

Motivation fades with time. The race you began with a sprint often ends with a slog. There’s a simple fix for this: choose shorter periods of time to focus.

For open-ended projects where improvement is continuous (e.g., fitness, language learning, research on a topic), limit your first several projects to one-month sprints. Remember: you can always stack projects working on the same goal back-to-back, but working in short bursts keeps you focused.

If your project is defined by all-or-nothing achievements (e.g., launching a business, writing a book, graduating from college), then I recommend either finding short-term milestones to focus your effort (e.g., getting one client, writing a book proposal, passing your next exam) OR being pickier when taking on longer projects. A four-year project should inspire much more thought before embarking than a four-week effort.

3. Only one project at a time

Once you pick a project, commit to finish it or to deliberately put it aside before picking a new one. Resist the urge to start any new projects until you’ve exited out of your last one.

While there’s always a little slipperiness in what, exactly, constitutes a project, the intention is what matters most. Habits, routine work and the “whirlwind” of your daily life will continue, but steel yourself against committing to new projects until your previous one officially concludes.1

Becoming a project finisher, not just a starter, isn’t something that happens all at once. Instead, it’s a meta-habit, a kind of practice you establish and reinforce through repeated experience. It takes some time and practice to learn, but by choosing the right sorts of projects—and completing them consistently—you can take your life in any direction you wish.

_ _ _

Next week, I’m going to be holding a new session of my course, Rapid Learner. This course teaches the strategy I’ve used for my learning projects, such as the MIT Challenge, Year Without English and, most recently, Foundations. In this six-week course, I’ll teach you not only how to design successful projects, but to optimize your learning and results. But first, I’d like to share how you can go from finishing projects to making those projects effective—choosing activities that are likely to accelerate your learning.

The post Lesson 1: How to start projects you’ll actually finish appeared first on Scott H Young.

Sleep: Day One

Today is the first day of the seventh month in my year-long Foundations project. This month’s focus is sleep. 

In case you’re interested, here are my notes from the previous six months:

Fitness: Start, End, Books.Productivity: Start, End, Books.Money: Start, End, Books.Food: Start, End, Books.Reading: Start, End and Books.Outreach: Start, End, Books.

Sleep is essential for health, productivity and well-being. Yet often we don’t give it the attention it deserves. While those who shirk on fitness or healthy eating are often shamed in our culture, lack of sleep is valorized—getting too little shut-eye is a badge of honor demonstrating our work ethic and stoicism.

Recently, the popular discussion around sleep has begun to assign it the importance it deserves. Even so, culture changes slowly. When I was discussing this list of foundations with a friend, one of them gave me a quizzical look when I suggested an entire month focusing on sleeping better. Exercise, sure. But sleep?

Despite recent attention, sleep remains underrated as a driver of good health, positive mood, productive energy and all-round well-being.

My Sleep (or Lack Thereof)

Sleep is one of the hardest foundations for me personally. Much of this is owing to being the father of two small children who, lovable though they are, are not always conducive to getting eight uninterrupted hours every night.

My typical sleep routine isn’t bad. I usually go to bed between 9:30 and 10:00 pm and wake up between 5:30 and 6:00 am. On days without interruptions, this is probably close to the amount of sleep my body actually needs. In the rare cases where I do sleep past 6:00 am, I usually wake up spontaneously between 6:30 and 7:00 am.

 The challenge I have is not with my typical sleep routine, but how many nights are atypical. While both of my kids (not quite 5 and 2) sleep through the night, at least one of them wakes up before 5:30 about 50% of days.

Sometimes it’s just a quick interruption that doesn’t impact my sleep much. In other cases, an interruption becomes an early wake-up. My two-year old, for instance, was waking around 4:00 or 4:30 am nearly every day for most of the previous month. It seems like we’ve gotten her back on a normal schedule, but travel, colds or other minor changes in routines can set us back into an unfortunate sleeping rhythm.

I also tend to function poorly when low on sleep. Some people seem to need more sleep to function at their best, and others can get by with 4-6 hours without it seeming to bother them. I’m definitely in the former group, and before having kids I almost always got eight hours of sleep each night.

Still, my sleep situation could definitely be worse. I have no problems of insomnia—it rarely takes me more than 5-10 minutes to fall asleep when I lay down, and I rarely wake up in the night for more than a few minutes. I’m also an excellent napper—I can even fall asleep and wake up in about ten minutes if I need it. I also don’t, as far as I know, have any health problems associated with sleeping that can be so frustrating for many.

My Plan for Improving My Sleep

Leading up to this month, I noticed I had a bit of pessimism about whether or not I would be able to do anything to improve my sleep. My wife and I are well-versed in the behavioral strategies to get kids to sleep better (consistent bedtime routines, wake-up lights, self-soothing, etc.), and I already follow a lot of the advice to improve my sleep (consistent bedtime routines, going to bed early, avoiding caffeine late in the day).

 Some problems in life aren’t easily fixable. Fortunately, sleeping problems due to kids are typically temporary, so some of this may just be the sort of things I have to wait out for a year or two.

However, as I started a little advance reading for this month and did some brainstorming, I realized there are actually a lot of things I could do better. On top of that, even if I don’t have solutions now, spending a month rigorously documenting my sleep might point to solutions I hadn’t considered before.

So here are a few things I’m going to try to improve my sleep quality this month:

Start taking brief naps again. I used to be a near-daily napper. Twenty to thirty minutes is usually plenty, and it can make a big difference. I mostly stopped this habit once I started working in an office—there was never a comfortable place to stretch out, even if I knew it would be better for my productivity to nap rather than to keep working. I think if I bring a yoga mat and a small pillow I might be able to get a good setup to make 20-minute naps a regular part of my day again.Be more consistent about sleeping early, especially after a night of poor sleep. If I don’t sleep well, my hardest times are usually around 4:00 to 6:00 pm when I feel like a zombie. By then whatever caffeine was in my system has worn off, and it is still hours until I would normally sleep. However, by 9pm, I am typically fairly alert, even if I only got a few hours of sleep the night before. I may not have total control over my sleep, but I can at least inch my bedtime a bit earlier when I’m in a sleep deficit.Cut down to one cup of coffee per day. I drink a lot of coffee. Some days I even have three or four cups of coffee throughout the day. Since I tend to fall asleep quickly, and the research-backed benefits of caffeine consumption likely outweigh the costs, cutting back wasn’t a priority for me initially. However, for this month I’m going to stick to only my morning coffee as an experiment.Avoid television or exercise after 8:30 pm. While I was a consistent 6:00 am jogger when I began this year-long project, the combination of a few months of kids disrupting sleep and my wife breaking her foot pushed running out of this slot and forced me to work out later. However, I find exercise after 8:00 pm makes it much harder for me to fall asleep, so I’m going to do my best to keep it earlier. I don’t watch a lot of television before sleep, but I’ll keep notes on this too in case this is happening more than I realize.Keep a sleep journal. I’m going to write some brief notes every day this month of when I went to sleep, any interruptions, and my overall mood. This, plus the sleep tracking on my Fitbit, should help me make some empirical observations about my sleep rather than just relying on my memory.

Overall, as I consider these changes, I’m cautiously optimistic. Hopefully I can improve the baseline quality (and quantity) of my sleep at night and, barring that, at least get some restorative naps throughout the day to improve my functioning when my sleep has been poor. As always, I’ll keep you updated with how it goes at the end of the month!

The post Sleep: Day One appeared first on Scott H Young.