Cedric Chin, (Chin 2023)

Summary

Thoughts

Notes

#+begin_quote Let’s say that you want to get good. Let’s say that you want to get good fast. And suppose you’re aiming to be top of your field. How do you go about doing it?

The typical answer you’ll get is probably something along the lines of ‘go do deliberate practice’ [Deliberate practice], or ‘learn to interleave learning’ or ‘go do spaced repetition’. These answers draw from a rich body of research on practice methods, expertise, and expertise acceleration.

That’s the good news. The bad news is that pop versions of these ideas often don’t work that well. You need to pay careful attention to what the research actually says — and what contexts they come from — before you can apply it. Or you need to try these ideas out in practice, and report back on what works.

There are, broadly speaking, three approaches to accelerating expertise that Commoncog is interested in. The names we use for these three categories are somewhat unique to this site; we’ve chosen them to delineate several categories of expertise acquisition that — in reality — have some overlap.

  1. The Deliberate Practice tradition. This body of work was started by K. Anders Ericsson in the 90s, and broadly discusses practice approaches that accelerate skill acquisition. Ericsson’s work is primarily concerned with something called ‘Deliberate Practice’ (DP). Deliberate Practice is a technical term, it does not mean ‘practicing deliberately’. It is a specific type of practice with a specific set of properties. DP is still considered the gold standard today — it is the best, most studied form of practice that produces the best results in certain fields. (Notice the implication: there are lousier forms of practice that are not as good as DP, and there are fields where the best performers do not become good through DP). The main problem with DP is that coming up with DP exercises is not free — someone has to spend time developing effective DP exercises for your skill domain.
  2. The Naturalistic Decision Making approach. What happens when you don’t want to spend time coming up with effective DP exercises? The answer is: you cheat. You find a bunch of real world experts, extract the tacit expertise from their heads, and then build training methods around what they’ve already figured out in their field of work. This body of work is best typified by the work of the Naturalistic Decision Making community (or NDM) — a sub-branch of psychology that has developed methods for studying expertise in real world environments. This topic covers expertise extraction methods, theories of expertise in real world environments, and training methods that leverage extracted tacit mental models of expertise.
  3. The Trial and Error approach. The above two research approaches beg one other question: how did these experts become experts in the first place? DP experts become experts because prior generations of practitioners took the time to develop DP exercises that may accelerate skill development. How did those first experts become good? Similarly, NDM researchers seek out existing experts in real world domains — often with no history of DP-style exercises — and then explicate their tacit expertise. How do those experts acquire their expertise in the first place? The answer here is, broadly: they learn through trial and error. But then what separates the masters from the merely good practitioners? This third topic of expertise is perhaps the smallest and the least developed, but no less interesting because many of us operate in domains where we have to get good through time consuming trial and error.

Deliberate Practice

The following are highlights on Commoncog’s coverage of DP:

Naturalistic Decision Making

There are more articles on tacit expertise in Commoncog’s archives. Here are some highlights:

Learning from Trial and Error

How do you learn better from trial and error?