Noi Sek ♂

The Pacemaker's Approach to Mastery

Learning to Learn efficiently

14th of Feb, 2017

How many skills will you master in your lifetime?

Setting the Stage

I came upon the thought of ‘Mastery’ the other day while musing with a close friend of mine. A loose thread that I was happy to pull, and pull, and pull until I had unraveled the entire picture and was left to ponder the mess I had made for myself. It stuck with me in the peculiar way that errant, persistent thoughts do, until I finally decided to give it the lead and see where it took me. Following that thread lead me down a rabbit hole of research papers, unfounded conjecture, and undalterated hearsay. But, it also produced a number of very interesting insights worth sharing. In trying to determine what exactly it is behind both learning and mastering skills effectively, I decided to write this article as an attempted guide to finding that thing you love, and rapidly becoming an expert in it. No promises will be made that this article isn't more of the aforementioned conjecture and hearsay than research, but I do like to think that my journey bore at least a few pertinent fruits.

Learning and improving is often thought of as an iterative process. One of equivalent exchange. You put effort in, and fair as fair does, out comes skill as your just reward. But that's not quite how it works-- and I'd like to prove it. When it comes to learning I've always thought that people tend to confuse the intent of their actions with the efficacy. In those moments where you make the decision to say to yourself "I would like to learn this Thing", or "I would like to increase my mastery over Thing"; are you really putting your best foot forward or are you just practicing for the sake of meeting a personal quota? Is the value of your input on par with the level of your expected output? Are the sum of your efforts on par with the level of improvement that you would like to see?

Those appear to be very basic questions, especially to you as the reader, but very few people actually consider asking themselves these things. I would like to make the case that the value of practice is not in the repetition (muscle memory centric endeavours aside) nearly as much as it is in forcing yourself to adapt to new scenarios. In examining the Franco Corelli's, the John Carmack's, and the Kim Jung Gi's of the world, how many of them are a product of homogeny rather than constant, self-directed change?

Very few, if any is the answer. So how did they do it?

Traditional Mastery

Well, often the path to mastery is the long and boring one. Sometimes you'll hear a story that involves the use of the word 'prodigy' at least once, but most will tell you that to truly master something, it takes Ten Thousand (10,000) hours of practice, practice, and practice.

I think that's a romantic notion. Throwing a lifetime's worth of effort into your passions and emerging at the peak of your years as a bonafide Michaelangelo-esque (Or Bob Ross-ish, at least) master of your craft. A fine and dandy proposition for those with an excess of time and a spritely disposition. But let's put that into perspective.

Ten Thousand hours.

That's 1,250 eight-hour days-- or 3.42 years at 100% consistency with no weekends.

18.6% percent of your waking hours over the course of 10 years, or 44.3% of your available free time when working a 9AM-5PM job with a 30 minute commute over the same period of time (completely excluding the need to do anything else)

Alternatively: just under 14 years of 3 hour sessions, five days a week, every week of every month of every year.

An astounding amount of time, to say the least. Probably more time than most can even properly conceptualize in terms of the daily sacrifices a regimen that strict requires.

A case could be made that anyone who cannot make the time to master [skill] does not value it enough, but I would like to think of it more as placing the value of one's time above market price. It follows that when confronted with a price-tag that unfavorable, one might seek more efficient uses elsewhere. Ten thousands hours is an absolutely phenomenal amount of time for anyone, no matter how empassioned.

Naturally, in the words of Python's Raymond Hettinger, I thought to myself "There must be a better way!"

A Streamlined Approach

I've long held a theory that the journey to the colloquial 10,000th hour time to mastery (TTM) could be smoothed significantly with some slight modifications to the learning process. Do not misunderstand me here, a dozen dozen papers have convinced me that the average person (intelligence or innate talent notwithstanding) learning the average task will still require ~10,000 hours to reach mastery, I just also believe that one could approach something very close to it with ~40% of that time given the proper approach-- and I'll outline it here for you.

Before we continue, an excerpt from The role of deliberate practice in the acquisition of expert performance [4] to help explain:

Howard's (2011) case study of the three Polgár sisters provides further support for our conclusion. Beginning at a young age, the sisters received several hours of chess instruction every day from chess grandmasters and their father, a chess teacher and author of several chess books. Using practice estimates obtained from biographical and autobiographical accounts, Howard found that the sisters differed both in the highest rating they achieved and in the amount of practice they accumulated to reach that rating. For example, one sister's peak rating was 2735 in an estimated 59,904 h of practice, whereas another sister's was 2577–more than a standard deviation lower–in an estimated 79,248 h of practice. Howard also found that the two sisters who became grandmasters had accumulated a great deal more practice by the time they reached their peak rating than had the eight grandmasters in his sample who reached top-ten in the world (M = 14,020.5 h, SD = 7373.96 h).

This snippet concerns an experiment wherein researchers measured the relative skill of three sisters over time as they were trained to play chess. Constraining the subjects allowed the researchers to control for variance in environmental conditions, heritage, and external factors with a reasonable measure of confidence.

The author of the paper then concludes from the above that practice is an unreliable measure of relative skill, and that time to mastery is more reliably predicted by personality (level of emotionality) and memory (ability to hold and recall complex concepts)-- but I choose to draw a slightly different conclusion. I agree with his statement on working memory being a strongly correlating factor for TTM, but as we can't control for that, we will ignore it for now. The key here is the subject's personality. My working theory is that personality is not only a red herring for the 'level of emotionality' (tolerance for training), but also for how individuals actually approach the process of practice.

To expand, what I'm suggesting is that some individuals are better at extracting more value out of the same amount of practice time. Not that they are always inherently better at a skill, but that they are un-consciously better at training themselves. Despite applying the same deliberate level of practice, the methods that more 'masterful' individuals use are far more efficient when it comes to actual, measurable progression. Chess lends itself very well to rote and memorization, which I believe implies that the real difference between subjects was one of adaptation. I believe that the best of the three was most able to identify her weaknesses and build upon them to solidify her skill very rapidly, and that the worst among them relied on the muscle memory of repetition in order to reach the same level.

In other words: The quantity of practice is irrelevant. The most important factor is quality. Sound familiar?

Again, it all sounds very basic, but now we have at least a meagre amount of scientific backing. Let's talk about how we actually improve our training methods.

An effective training / practice / learning regimen must:

Maintain a tight feedback loop.

Maintaining a tight feedback loop is chief among all possible things that you can immediately improve. A very, very large portion of attaining skill at a rapid rate is the ability to identify your weaknesses and actively combat them. Whether that means recording videos of yourself, making audio tapings, or acquiring an independent third party reviewer, you MUST constantly assess and iterate. Practice will build muscle memory (mental and physical), but without the controlling factor of goal oriented revision, it is entirely possible to become rooted in a path where you are only reinforcing bad habits.

Conversely, never allow yourself to become too comfortable within a task. Choose complex challenges, always focus on the domains that you enjoy the least, and avoid the domains that you feel most comfortable in. It is good to make sure that you're enjoying yourself, but make sure that you are not doing it at the expense of your development. If you must spend 20 minutes in your comfort zone for every hour that you spend out of it, so be it, but do not allow yourself to become preoccupied with a region of your domain just because it is the part that comes most naturally to you.

For the artist, this means drawing highly technical pieces instead of purely creative ones. For the software engineer, this means learning different programming styles and languages rather than sticking with what is familiar. For the chess player, this means establishing worst case scenario victories from the perspective of your opponent, rather than running through permutations of opening sequences. These sorts of activities are all tightly coupled, whether you realize it or not. It may not make sense to you that a football player practices Yoga, but does the increase in flexibility and tendon strength make them better at their jobs? Absolutely. In a similar fashion, everything that you learn is complementary irregardless of whether or not you feel that you need to be proficient in it.

Be daring, record yourself, evaluate, and iterate.

Be sustained for no longer than 90 minutes

There are two parts to making sure that your practice sessions are no longer than 90 minutes. First among them is the principle of intensity. You shouldn't feel as if you need to break a personal record every time you sit down, but the intensity of whatever it is you're doing should preclude the ability to do it for longer than 90 minutes. If you're able to sustain the same level of effort at the end of 90 minutes as when you started...it's time to move on to something more challenging! Whether that benchmark is met with mental, physical, or emotional exhaustion is dependent on what it is you're trying to achieve.

Second of the two parts is the concept of 'chunking'. Breaking the subdomains of your given domain into even more manageable portions that allow you to objectively measure and push your progression. You already do this at a subconscious level when you choose to learn a task, but take it a step further and try to moderate your practice sessions into self-contained, digestable chunks. Not only will you be able to self-correct at a finer level, you'll also recall the information you've ingested much more accurately. A thought exercise-- were you tasked with teaching someone to retain a list of, say, one hundred words in a specified order, how would you go about doing it? Which would be most effective; to run drills for 24 hours in order to memorize the list in one day, to spend three days asking them to memorize 33 words a day, or to spend two weeks and allow the person to memorize 7 words at a time? If your goal is total, 100% recall, whose memory would you trust the most in six months?

Keep your sessions short and meaningful. Intense, self-contained, and focused.

Be repeated no more than 3-4 times a week

Making sure that an activity is not / can not be excessively repeated applies more obviously to physical endeavours than to mental ones, but for most the necessity of recovery time is indiscriminate. In much the same way that you must allow the body to heal after particularly strenuous exercise, the brain must also be given time to process the things put before it. In fact, you probably already know the name of the process involved! The REM phase of sleep is the mental analogue to giving your body a well earned rest, and your body seeks it just in order to process the events of the last 16 hours. It follows that, much as your muscles take an extended period of rest to regenerate stronger than they were before, allowing your brain enough time to forge and re-forge neural connections is an important part of the learning process.

What does that mean? Try not to practice on sequential days, and give yourself some time to recover and reflect. You'll be better off for it. This point ties loosely into the last point of 'Never continue beyond the point of failure'.

Footnote: I have a sub-theory here based off of a soon-to-be-sourced study that it may also be at least as effective to pair this strategy with a 2 week on / off training cycle (allowing you to target more than one skill at once).

Never continue beyond the point of failure

Your brain is a machine! A hot, wet, stinking mechanical mind with mechanical thoughts, and convoluted wetware machinations! Much like any other machine, your brain seeks to be efficient. In that efficiency, it follows the path of least resistance in order to build connections between related concepts, actions, thoughts, ideas, memories, scents, sights, and smells. This means that every time you put pencil to paper, your brain gets a little better at turning idea into thought, thought into motion, and motion into the complex characters you call letters. This also means, however, that every time you fail... you get a little better at it. Be careful in reinforcing the same bad habits that you seek to eliminate, and stop immediately the moment that your form or technique is compromised.

When your fingers are too tired to continue playing properly: Stop!

When your brain is too taxed to recall accurately anymore: Stop!

When your body is too fatigued to stay awake: Stop!

Do not allow your brain to build connections in failure. When the river has carved itself deeply into the mountainside, one does not simply ask the river to head in another direction. "Powering through" will only lead to long-term damage in the long run. Avoid it all costs.

To recap, an effective training / practice / learning regimen must

  • Maintain a tight feedback loop.
  • Be sustained for no longer than 90 minutes
  • Be repeated no more than 3-4 times a week
  • Never continue beyond the point of failure

Even with this approach, there will still be times when you truly do hit a wall-- as far as I've seen in my own personal experiences, mastery over time is defined much more by periods of flat plataeus than by sharp peaks.

A Revised Formula for Time to Mastery

So, in bringing this back around to the original thought process that triggered all of this, I did a bit of napkin math and came up with the following formula:

# This really represents time to 90th percentile skill, rather than 99th
hours_to_mastery = 3500

# For practical reasons, assume that you start at 30 and cease learning at 60.
current_age = 30
years_of_practice = (60 - current_age)

# An average of 30 minutes a day, at 70 minute sessions 3 days a week
weekly_practice = 3.5 # Hours
yearly_practice = (weekly_practice * 4) * 12

# Let's pad in some room for error: missed days etc. We're not perfect.
efficiency = 0.8

years_to_mastery = ( hours_to_mastery / ((weekly_practice * 4 * 12) * efficiency) )
potentially_masterable_skills = ( years_of_practice / years_to_mastery )

The results of this can be seen at the top of the article.

Notes

Brave is the soul that can tolerate this level of commercial-grade drivel for as long as you have. Congratulations, and my thanks for following along thus far. I hope that this article has been of some interest to you, but the journey is not yet complete! I am considering this a living document at the moment as I continue to build more evidence behind my (sometimes arbitrary) axioms.

There are also a few sticking points that bear mentioning:

  1. As I'm spared the burden of peer review and the need to actually validate much of the above, this is all very, very tenuous and theoretical. There are a few unaddressed holes, flaws, assumptions, and necessary explanations that I have entirely skipped for the purpose of brevity-- I must encourage that you view some of the references yourself to draw your own conclusions!

  2. All of the above assumes that you are of average innate ability to acquire a skill before attempting to do so. Much of the research that I have read implies that not only is innate ability a major factor in TTM, but that it accounts for up to 70% (!) of TTM variance. This leaves only the last 30% to optimize. Fear not, however, as research also suggests that humans tend to be drawn naturally to the things they are innately suited to excelling in. If you're passioniate about something, go for it!

  3. If it is not already obvious, the 'Revised Formula for Time to Mastery' is actually a formula for time to 'Almost Mastery' or "Indistinguishable to a Moderate Layman Mastery".

I will leave you with the following excerpt [1].

Although there are surely notable counter-examples in science, one would be well-advised to employ common sense (i.e., critical thinking) when considering various claims of universal statements in the domain of expert and elite performance.

Further Reading

  1. How Learning Works: Seven Research-Based Principles for Smart Teaching

  2. Intelligence (Research Journal)

  3. LessWrong on Learning

  4. The Spacing Effect

  5. Although I can't yet recommend it, I'm looking forward to reading Intelligence: Its Structure, Growth and Action, Volume 35. Haven't yet read it as it's quite expensive!

  6. There are an astonishingly high number of papers on neural networks and machine learning that echo traditional insights from human learning almost 1:1, start here and spend some time researching them if you are interested.

Primary References


Back Home
Discourse is fun. Tweet @noisek or ...
Leave A Comment!