Where I got this from

This post is primarily a way for me to summarize and consolidate what I've learned from Barbara Oakley's "Learning How to Learn". I also incorporate advice I've gotten from professors, from personal reading, and experience.

what to do and why

learn in building blocks

Everything we know is built out of mental chunks. These chunks are bits of mental data bound together by shared use, context, or meaning - think microskills or phonemes to words. Once disparate pieces of knowledge have conglomerated into a chunk it is treated as a single unit by the brain and it takes little effort to recall and use it. Chunks take so little effort to use that they can become part of the base knowledge that the brain links together into another chunk.

When you learn how to skateboard, you don't just "learn how to skateboard." First, you learn how to position your feet, you learn how to squat, you learn how to shift your feet; then you learn how to use all those skills to push forward and to brake. Then, when learning to skate switch you don't start with learning how to swap feet while moving. You start with learning how to balance while leading with your non-dominant leg.

Mix up your learning

When reading documentation for a software product I often find myself frustrated by the fact that I can't understand how the part I'm reading about fits into the whole, or form even a vague systematic view of the knowledge base. By jumping around the target knowledge area, we can interleave many sub areas of knowledge and begin to see how they fit together before any one part is fully congealed, allowing us to form a clearer more systemic understanding of the whole. Think of it like loading an interlaced image vs an uninterlaced image. The interlaced one loads a lower resolution of the image faster, and then refines it as more data comes in, as a result we can view and understand the image well before it's done loading. An un-interlaced image that loads line by line may load each segment in perfect detail, but it takes much longer for that detail to become legible, since we have less of the image to work with.

Connect your knowledge

Knowing things is a snowball process. The more you know, the easier it becomes to learn something new. By looking for way's we can analogize or relate what we're learning to what we're already familiar with we can help ourselves organize our knowledge for retrieval and build that snowball. Metaphors, mnemonics, and visualizations are all useful tools here. We can also link knowledge by context. Learn about the historical context of what we're learning about and what it was a response to.

Process over product

We've all heard some aphorism about loving the journey not the destination, the doing more than the achieving, so on and so forth. These aphorisms are often presented as some sort of moral truism to follow. Such a presentation annoys me and I'm not in the habit of following advice that annoys me. Here's the amoral case: THE BRAIN IS LAZY. Thinking is calorically intense and our body wants to save energy on every occasion. A process can become a low energy routine, an outcome cannot.

We talked earlier about how chunks convert the hard work of recalling lots of information into the low effort work of recalling one unit of information. Routines are another example of this same chunking. We can chunk a process, making it easy. We can't chunk a product, in fact by focussing on the product we're making it even harder for ourselves because we constantly need to evaluate how close we are to our goal and what needs to be done. By focussing on the doing we can learn and chunk out a routine that lets us offload action to our subconscious.

In other words: if you want to learn to play guitar, focus on playing notes and chords and stringing them together, don't focus on nailing a perfect rendition of "through the fire and flames".

have a quitting time

A common piece of productivity advice is to just work on something for a few minutes and then the momentum will carry you through, this is often paired with mentioning the Pomodoro method and how 25 minutes of work followed by a 5 minute break will make you somehow crush your deadlines. If you're like me, this is not true. In fact it is actively detrimental to building your capacity.

Any act of self deception is going to hurt us, even if we do it to protect ourselves, even if it does protect us. If you tell yourself, as I once tried to, that "oh I'm only going to work for 5 minutes and then stop" while secretly hoping and planning on momentum kicking in and carrying through, some part of you will know you are lying to it, and it will stop trusting you. Some part of you will feel betrayed when you keep working past the five minutes. Some part of you may say 5 minutes isnt enough of a break, it will say "I can't even get to the break room and back in 5 minutes" and will grow resentful that it works with no reward.

I spent years ignoring these parts and as a result they stopped listening to me. I had lost my ability to trust myself and my own psyche was in active rebellion against me. Loss of self trust makes us anxious, it makes us second guess and agonize over every decision, it shatters self confidence, it makes accomplishing even basic tasks hard because we no longer truly believe in our ability to achieve any of the requisite steps.

Once we are in a place of low to no self trust, if we want to be able to feel functional again, we have to pick a time when the work ends and we have to stick to it. Over and over again. If we say I will work on this for two minutes and then stop. We stop after 2 minutes, even if we think we got nothing done. we stop at 2 minutes and we do this over, and over, and over, until all those worn out distrusting parts can trust us again, and we have to keep that trust.

Any hardship is easier to endure if we know exactly when it will end.

The real, not torture Pomodoro method

The power of the Pomodoro method is that it gives us a quitting time. We can tell ourselves that, "after two pomodoro cycles I will not have to do this again today" and mean it. If after those cycles we're not done, do two more tomorrow. The other power is we get to reward ourselves for our work. That doesn't mean a 5 minute break. A lot of the time what we're trying to do is hard and draining work, a short break seems profoundly unsatisfying. If our reward feels unsatisfying, then it's not going to generate the dopamine that actually rewards our brain. So ignore the break and find an actual reward, maybe go get a coffee, or play a level of a video game or something.

The loosest version of this system is pick a time period that feels light and manageable, work for that long, recuperate, and reward yourself for a job well done. Repeat that until your quitting time for the day, then QUIT.

Consult other people whenever you can

There's lots of institutional knowledge out there that very sadly (and sometimes for malicious reasons) cannot be acquired from a book. A lot of that knowledge is very useful and will save us a lot of trouble. There is no way to learn it without consulting with other, more experienced, people, people who remember learning what we're now trying to learn. If we try to learn everything ourselves from scratch, we will trip and fall many times before we learn to walk, get lost in the woods before we stumble onto the path and so on. We are all greatly advantaged by having someone to consult with even for just a few moments that can point out ruts we might fall in, guide our technique, and provide a roadmap of what we need to learn.

If we try to learn everything on our own, we might by chance discover a few things in the process of fumbling in the dark that noone has seen before, but, by and large we will find ourselves laying the foundations to a cathedral already built elsewhere. Whenever possible, go to the built cathedral and climb the stairs.

Experiential learning can only be learned by experience

In the world of sports coaching there is a paradigm referred to as Environmental Dynamics or the Constraint Lead Approach. The problem with conventional coaching that CLA aims to solve is one, that any conceivable drill regimen can't capture the infinite possibility space of an actual game, two, controlling the body's movement with the speed and precision necessary for sports performance is incredibly difficult to capture in language.

What a constraints lead approach proposes is: that we have an automatic understanding of how to move our body and build fine motor skills through subconscious neurological adaptations, therefore we train a player not by telling them how to play better, or by telling them what better play looks like, we train them by altering the constraints of the game and its environment to encourage developing the skills they need to improve. For example, if we have a baseball player that is hitting the ball too low, avoid instructing them in the minutiae of how to tense their shoulders or where on the ball they should aim to make contact. Put a wall in front of them and tell them to hit the ball over it. The player now has a way to instantly and intuitively evaluate their performance and can automatically explore their movement space to find an efficient solution. Another example might be to eliminate part of an athlete's movement. If a swimmer's kicks seem weak we might give them a flotation device to hold with their hands and tell them to only move by kicking until their speed improves.

Note that this approach only works in a situation where the fundamentals of our domain come to us automatically. We don't learn to code by typing into a text document until we chance into writing valid Python. But, once we are familiar enough with Python's syntax, keywords, and standard library that we can begin to easily and automatically explore solutions to problems without having to consult a syntax guide, we can improve our code style and efficiency through similar constraint lead exercises. For example, we might practice eliminating nesting by telling ourselves we aren't allowed to use an if-statement in our solution, or we might force ourselves to work purely within an object oriented or functional paradigm for a while.

practice retrieval

What we call skill acquisition is at its core, automatic memory retrieval. When the neural pathway for a certain memory is so burned in that activating and using that information happens unconsciously. Immediately after learning something, try to remember it. Maybe we read a book chapter, before going to take notes we should take thirty seconds to recall as much of what we read as possible. This applies to meetings, coaching, and anything else.

Take days off

Information enters long term memory from short term memory via a process called consolidation. Consolidating a memory once isn't enough to wear in the neural pathways and retrieval will still be difficult and inefficient. If we want to build up automatic recall we have to practice retrieving the memory from long term storage and then reconsolidating it. What this means is we have to let information leave short term memory, and then recall it back and let it leave short term memory over and over again to build up the pathways for that information. Typically, this retrieval practice is called spaced repetition, this often looks like using flash cards once a week or once a month, or drilling our sport of choice on a regular basis. If you're like me, doing something at the same time once a week or once a month outside of work can be intimidating, difficult, or soul crushingly boring. Instead I would recommend re-framing from spaced repetition to taking breaks.

We can spend a few hours practicing a skill one day, then take a few days off, and come back to it. By thinking about it as a break we prevent ourselves from getting locked into a set schedule, side-stepping some of the intimidating overhead and fear of stagnation, while, still remembering to come back to our skill every now and then.

Keep it short

We can only maintain focus for a few hours at a time at most. Many professional athlete's have short, focussed training sessions of around 40 minutes focussing on one skill. They might do more than one session in a day, but one session at a time.

Practice deliberately

Deliberate practice is a bit of a buzzword. What it means is to avoid simply showing up and unconsciously do the actions of your target skill. Instead we want to be attentive, mindful, and focussed on developing specific chunks of our target knowledge. If I'm practicing archery I want to show up and work on having consistent shoulder tension for twenty minutes, then take a break and evaluate my form by shooting at the target and paying attention to where my inconsistencies are.

The other part of deliberate practice is that we must run towards our weaknesses. To build a full skill set we can't just do the parts we're good at, we have to locate our points of failure and strengthen them.

pick the most minimal example you can, then do it

When reading a math paper, it's good practice to find the simplest example of the principle being described and work through it to see how the machinery works. It's also a good idea pretty much everywhere else.

Take notes, after you've digested the information

Making our own reference material feels great and allows us to test how well we know our stuff. But, If we do it too early we might feel like we're learning without actually committing anything to memory. We need to let the knowledge percolate, organize and be recalled once or twice before we start writing notes, otherwise we're simply copying information down without thought. The other danger of making reference notes too early is we don't have a good understanding of the material yet and can't organize our thoughts in a way that will be useful to our future selves.

This doesn't mean we can't take notes during the learning process. We just need to be sure we've considered and practiced retrieving the material first. And, unless our learning notes are exceptionally well organized, we should probably come back to do a second pass to produce our actual reference material.

Notes and tips

• Write your to-do list the night before, that way you have time to brace and prepare in our sleep
• make notes indexable and useable. Use headings, subheadings, accent colors, and even tables of contents to make it easy to find and retrieve the information you need when you need it.
• Leave lots of empty space in your notes so that you can annotate and correct anything you right down.
• use a mind palace to store ordered sequences
• use odd images full of vivid sensory details as mnemonic devices to remember and cluster concepts.
• If reading a document, know that the author is trying to guide us through a specific thought process, if we pay attention to the conventions they use to present something to us, and where else we've seen them used, it can help us find new associations that help us better understand the subject.
• Learn with other people, what clicks with one person won't click with another, by discussing, even amongst newcomers everyone comes away with a more well rounded understanding.
• Sometimes after making progress for a while, we will feel like we plateued or got worse, this is actually a normal phase of learning and can occur when our brain is going through the process of reanalyzing and restructuring our knowledge.
• If you can turn learning into a game, do it. Play not only encourages exploration, but lowers our mental barriers to acquiring new information and rules.
• You are under no obligation to read the entirety of a textbook or article. Everything on earth is just reference material. Skip to the part that interests you right now, digest it, then leave the book behind until you need it again.

proposed procedure

1. try it out: its difficult to know what people are talking about if you have no experience. Don't try to do it well, just try to get a feel for it.
2. research: remember to skip around the knowledge area to build a holistic picture and look for ways to relate the knowledge back to things we already know.
3. take a break
4. set goals and drills: once we know a little bit more about the domain we can start setting up small and achievabel practice goals and drills. For example, learning to make websites:
   1. make a single static page with no styling
   2. make a single static page with styling
   3. add javascript interactions to some of the elements
   4. make a multipage website
5. deliberate practice
   1. do project or drill
   2. evaluate
   3. take a break
   4. research
   5. repeat

example: taking up running

I bounced off running a couple times, but then I started playing tag with a friend and got competitive about it. I took up running to improve my stamina and better my odds. Day 1 I picked a trail near my home and ran it without any prior running research. I realized that my stamina was worse than I thought, but also found myself frustrated with how heavy my foot-falls felt and how much my body shook with each step.

The next day I watched some videos and read some articles about running focussed on fixing my heavy footfalls and jostling gait. I learned that the primary variable affecting the smoothness of my run is cadence (how many steps per minute) and that how heavy a footfall feels is primarily a factor of how far ahead of the center of gravity the foot contacts the ground. With these in mind on my next few runs I evaluated how my feet and body felt while trying out different cadences and postures until I found a way to run comfortably and stay acceptably light on my feet.

I then realized I still wasn't satisfied with my endurance, after each run I was exhausted and I didn't feel closer to being able to meet my distance goals even if each run was no longer horribly uncomfortable. I did a little more research and learned that the exhaustion I felt was likely a result of poor pacing. I started setting challenges for my runs like "run as slow as possible without walking" or "run twenty minutes without walking or stopping, slow down the run instead". My goal shifted from improving my time to reducing exhaustion. Now, I'm working on alignment and technique, my goal has shifted to making my weight distribution neutral to forward heavy and I will walk long distances adjusting my torso and feet, looking for a stance that feels like it draws me forward.