How to learn anything faster
The secret to accelerated learning? It's all about iterations, not hours
I've long been fascinated with people who are able to learn extremely quickly. Whether it was Scott Young's MIT Challenge (finishing the entire MIT computer science undergraduate degree in a year), language polyglotism, or Max Deutsch's one-month quest to train to face chess grandmaster Magnus Carlsen, I have always been fascinated with these "ultralearners."1
As I dug in deeper, the unifying factor was that these ultralearners also did a much better job in their practice. These learners were able to reduce a complex skill into discrete components and use deliberate practice to drill each distinct skill. They ascertained the minimum set of skills necessary for a task (what Tim Ferriss would call the 20% of skills required for 80% of the performance, deriving from the 80/20 rule), and then they would then fanatically drill them over and over.2
Crucially, they all showed that learning is a function of the number of iterations you complete, rather than the number of hours you spend. While time spent is certainly correlated with iterations, it's the iterations that are actually meaningful to create improvement.3 I'm reminded of the painful quote on experience:
Twenty years of experience is often one year of experience repeated twenty times.
Ultralearners in all fields have identified the essential components of a skill and then accumulate large number of repetitions over time to acquire expertise. They are the people who are able to accumulate twenty years of practice in one year.
Regardless, I'm always trying to keep my iteration speed as quick as possible (something that I often struggle with), because it allows me to maximize learning.
And in the spirit of pushing my own learning, some open questions:
What qualifies as an "iteration" of learning? Certainly it's context dependent, but are there any universal formulations? My working model is that one iteration is defined as one pass through the observe, orient, decide, act (OODA loop).
How do you best approach learning in wicked environments? Deliberate practice is very suited for kind learning environments where there are repeatable patterns; quick, accurate, and rapid feedback; defined rules and boundaries. Wicked learning environments are where the exact same action may return different results or there are nearly infinite dimensions in the problem (e.g. investing). Therefore, deliberate practice is extremely difficult in wicked learning environments.
Scott Young’s book Ultralearning is an excellent resource on this topic.
It also seems that these ultralearners (the good ones, at least) have a high bar for what qualifies as "practice." As the old adage goes: Practice begins when you get it right.
Later, I would stumble upon research like Anders Ericsson’s Peak and Daniel Coyle’s The Little Book of Talent that would confirm these ideas. As Coyle writes: "Deep practice is not measured in minutes or hours, but in the number of high-quality reaches and repetitions you make—basically, how many new connections you form in your brain. Instead of counting minutes or hours, count reaches and reps."