00:00:00
What is metalarning and how do I learn how to learn better? >> DSS deconstruction, selection, sequencing,
00:00:07
stakes, deconstruction, which is taking a fairly ambiguous goal like learn to swim
00:00:15
or learn Japanese. None of those are actually very descriptive. Deconstructing any one of those is
00:00:22
breaking it down into constituent parts. You can do that very effectively with the help of an expert. For instance, I
00:00:28
mean, if you want to find a silver medalist from the Olympics two Olympics ago, you can probably get on a Zoom call
00:00:36
with them for $100 an hour, maybe $50 an hour. Like, you do have access to world-class
00:00:44
talent. They would help you figure out, all right, there all these different possible components. The next part,
00:00:48
which is selection, Prao's law. So, you're picking the 20% that will give
00:00:52
you 80% of what you want. Let's just use language learning in that case. You can
00:00:57
very easily find word frequency lists for any given language. Hundreds of thousands of words you could learn. But
00:01:03
with the most frequently used 1500, you can get to reasonable conversational fluency in almost any language in 8 to
00:01:11
12 weeks without question if you approach it methodically. But you need the right material first. The next S is
00:01:19
sequencing. Putting it in the right order. This is the magic sauce that gets lost a lot which is what is a logical
00:01:26
sequence for learning any given skill. What do you practice first? So in the case of swimming for instance forget
00:01:33
about breathing like you need to figure out like getting comfortable putting your head under water etc etc. Last S
00:01:38
stands for stakes which means incentives. So how do you ensure that you will do actually what it is you say
00:01:45
you're committing to doing? If more information were the answer we'd all be
00:01:48
billionaires with six pack abs. So information is clearly not sufficient. It's necessary but not sufficient.
00:01:55
Incentives drive behavior change, right? You could give uh 500 bucks to a friend
00:02:01
or 100 bucks, whatever. Does that the amount doesn't really matter. And if you
00:02:04
don't do what you say you're going to do, they donate it to like your most
00:02:07
hated political candidate in your name. Right? That's another one that I've seen
00:02:11
work really well. If people expect some kind of like linear incremental progress, it just ain't going to happen.
00:02:16
And so most people quit before they hit any real inflection points. That's it.
00:02:20
That DSS, deconstruction, selection, sequencing, stakes. And if you just check those
00:02:27
boxes moving that order, uh your ability to learn will hockey stick in a really meaningful way.