Ericsson (and Pool)’s Peak is a salute to the merits of deliberate practice, a framework for learning skills in the most effective manner possible. It is rooted in several principles: (1) building up to overall goals via a sequence of small, specific objectives, (2) receiving clear, frequent, immediate feedback (ideally involving some kind of expert supervision), (3) adapting the curriculum or training sessions s.t. you are constantly at the edge of your comfort zone, and (4) engaging in regular, intense/focused study, with your mind fully invested in the task at hand and always working on making it past obstacles. The end result of deliberate practice is the hardwiring of increasingly-refined, skill-specific “mental representations” in your brain. Mental representations are essentially (conscious or unconscious) constructs in our minds which we use to process things – everything, actually. A transparent example would be the mnemonic structures used by digit memorizers (e.g. chunks of 3 or 4 digits seen as cross-country running times, stored in a hierarchical tree structure).
The authors argue that deliberate practice can be used by all people of all ages to attain potentially limitless levels of competence in their disciplines of choice. Furthermore, Ericsson and Pool assert that this is more or less the route that all experts take in order to reach absurd degrees of aptitude, demonstrating that it’s ultimately the plasticity of our brains, not “inborn talent”, that allows us to excel. They concede that there are some cases in which genes or early development turn out to be important, e.g. that high IQ helps chess players learn faster early on, but maintain that in the long run it’s smart practice that gets high-performing people places. Always.
I had heard most of the advice in this book before reading it, but it was presented so effectively, in such a cohesive package, that I couldn’t help but feel awed and invigorated reading it. (To be fair, this is maybe the first self-improvement book I’ve read.) Also, a lot of the concepts in this book relate very closely to modern machine learning concepts, which I found interesting. Like deliberate practice, ML/deep learning is all about rewiring pathways to learn better internal representations, driven by some sort of supervision signal. At one point in the book, they even refer to the relevant physical structure as a network of neurons. They also talked about continuing to do things that work, and trying to adjust things that don’t work – like RL – and going from simulation to reality – like sim2real – and constructing training sets of labeled mammogram examples – like any old supervised learning paradigm.
One note is that the book seemed to be all about framing success stories in a deliberate-practice sense. I don’t know if the idea in itself is incredibly revolutionary (although it is definitely nice to have all of these solid principles laid out for people to process), nor if there are success stories that actually didn’t come from deliberate practice, nor if there are failure stories that actually did use deliberate practice in manners that seemed like they should have worked. Of course you can rationalize and explain away anything, but essentially it’s possible that deliberate practice is not the end-all silver bullet that the authors want to push for. Or maybe the technique is so general that it would be hard for it not to fit all these success stories. I don’t know. I believe heavily in the power of belief, so I choose to trust that the premise works, that anyone can do anything (our brains are set up for it), and that these principles, applied correctly, are a promising path to self-actualization.
There were a lot of nice quotes and a lot of insight and many, many more examples in the book than I feel like recording in this reflection, so if you’re interested I recommend reading it yourself. For example, contrary to the immediate notion of the 10,000 hour rule, doing the same thing over and over isn’t going to make you better at it. You need to be pushing your boundaries (working on whatever you need to improve on in the moment), i.e. not repeating the same thing across 10,000 hours. Tennis players will plateau if all they do is play the game a lot. To continually improve, they need to keep identifying particular points of the game to work on and grinding those. Doctors also traditionally don’t get better with experience because they just do the same things over and over again and don’t push their comfort zones.
Other quick notes: (1) training isn’t supposed to be fun, so seeking performance means a lot of no-joy periods. Practice sessions should typically (definitely at first) be limited to about one hour in length, since we usually can’t focus intensely for much longer than this. You cannot just go through the motions – your mind must be completely focused on the task at hand or most of the benefits of practice are going to be lost. Experts need a lot of sleep because they spend so much time focusing intensely! (2) In deliberate practice, skills are the end, not knowledge. We get knowledge as part of the process of developing the skills. (3) There are a lot of good teaching tips in this book (e.g. Wieman approach). (4) Ben Franklin is a legend.