Shared Physics
by Roman Kudryashov
Shared Physics

How Learning Works: Components, Systems, and Loops

Published on 8 min read

While listening to Josh Kaufman’s Farnham Street interview the other day, it dawned on me that there are a few interconnected and important concepts related to learning that come together in an interesting and foundational way. Not only that, but I had never thought about them too deeply and at times used them interchangeably. It's like Charlie Munger once said about mental models: “All this stuff is really quite obvious and yet most people don’t really know it in a way where they can use it.”

This realization was humbling, but also helpful: there exists a system for turbo-charging your ability to get better! And more importantly, it doesn’t rely on anything patentable or trademarkable in order to get its point across!

Without further ado, here are the components, definitions, and how they come together:

The Components and Loops of Learning

1. Knowledge is just information.
It’s the bits — the 1s and 0s collected over time that add up to knowledge. Knowledge on its own is useless. To quote Derek Sivers, "If more information was the answer, then we'd all be billionaires with perfect abs."

2. Experience is lived and firsthand knowledge.
Any knowledge that is not firsthand is necessarily incomplete. There's no book or description that can be substituted for your personal psychological and physiological reaction to something in the world.

Living in the world creates an accumulation of first-hand knowledge. The act of doing something in the world creates experiences as well. Experiences can both create and apply knowledge, but that is different from a skill because…

3. A Skill  is the ability to do something in a replicable way.
Doing something once does not mean you are skilled. When you are able to do something again and again, in different contexts, then you are skilled.

So far, pretty straight forward. But I reckon there’s a lot of concepts around the “getting better” part of the the knowledge-experience-skills framework that most people don’t think too deeply about, especially if they haven’t been involved in sports in a semi-professional manner. I know I would often stop with "I learn stuff, practice it, and become more skilled. What more is there to it?"

So to jump right in:

4. Performance is an individual instance of your application of skills.
For example: when you go to the theater, you’re watching a bunch of skilled folks apply their skills publicly for you. The show you see is a performance. It might be good, it might be bad. Someone might have an off day or really nail it. If performances are uneven, you can make an assumption the people are unskilled (because they can’t do it in a replicable way). It might go without saying, but Training is repeated performances intended to improve skills.

But how do you learn and improve? What are better ways to perform or train? This is often augmented by Teachers, and different types of teachers play different roles in the learning process. Once you leave school, there are two types of teaching most people primarily rely on: Coaching and Mentoring.

5. Coaching is external guidance and feedback on your performance.
Coaching helps improve skills and performance through calibration facilitated by an honest third party. A coach is someone who can watch you when you can’t watch yourself. In sports, the concept of coaching is so ingrained as to be inseparable from the act of sporting itself (there might be an argument that sporting without a coach is the heart of amateurism). But it isn’t just for the realm of sports: people have vocal coaches, business coaches. Many skills-teachers are coaches in some degree, as are some therapists. That said, in the business world, I think coaching is extremely underutilized.

6. Mentoring is a subset of coaching primarily focused on the creation of knowledge.
Mentoring is used in business and academia more often than coaching, and so I think it makes sense to distinguish between the two. Whereas coaching is focused on your performance, mentoring is focused on your knowledge. When one mentors, they give advice, guidance, and perspective. People who are good mentees are those who are able to take that knowledge and apply it through training to get better.

It's also helpful to distinguish between a mentor and a book – while many will refer to a library of recorded knowledge as "the world's greatest teacher", mentoring is different in that it is optimized for you and your specific situation. While both create more knowledge, mentoring does so in a way directly applicable to your circumstances.

7. Both coaching and mentoring rely on Feedback, which is information presented back to you from the world, after you have acted in some way.
It is usually about your performance or training, but can be broader in the case of mentoring. Feedback can come from folks not involved directly in your learning process as well. We tend to call this unsolicited feedback and it has negative connotations because it is either not presented in a structured way for you to ingest, or because it comes at a time when you are not ready or open to it.

But in most instances, feedback is a good thing! Sometimes it needs to be taken with a grain of salt, sometimes it needs to be filtered, and sometimes it needs to slap you in the face, but feedback is good because it is how the outside world is reflected back to you! And yes, feedback can be physical or literal — punching something the wrong way or connecting wires the right way will give you literal feedback in the form of pain or a light bulb coming on.

8.Logging is the act of recording feedback to measure changes in performance over time.
I used to wonder why people recorded their weight lifting reps in a notebook when I went to the gym. Ok — I didn’t wonder too hard: it creates a log that you can use to measure your progress. I never recorded anything for myself because I didn’t need to — staying in shape through a repetition of some exercises is extremely different from a program of self improvement.

It was actually a design professor who moonlit as a triathlete that highlighted the uses of logging for me: he used it not just for measuring improvement over time, but also as an early warning system when he was pushing himself too hard and his body was reacting in subtle but measurable in deviation-from-the-baseline ways.

Logging is extremely important because it can also be used to gain distance from a performance or a training in order for you to evaluate it in a less subjective way when a coach or a mentor is not available. Logging also creates value for non-physical improvement: many people keep decision journals to reflect on their own decisions after the fact to calibrate their heuristics (mental models and shortcuts). Daily journaling and even project management (whether through tools like Jira/Asana or bullet journaling… or keeping post its with your daily to-dos like I happen to do) can  generate valuable metadata that can inform your learning cycle.

However, none of this can happen without:

9.Reflection, which is the intentional and internal process of turning feedback into knowledge.
This is the work of learning: processing external information and turning it into knowledge that informs the practice and development of your skills. As you reflect, you re-calibrate any new knowledge against old knowledge and figure out how this should inform your future actions.

When all of these components are combined, this becomes a...

10.System, which is how different elements come together and interact with each other.
Systems can be finite (think of a Rube-Goldberg machine that’s run once) or infinite, which is explored in fun detail through the great "Finite and Infinite Games" by James Carse. As he describes it,

A finite game is played for the purpose of winning, an infinite game for the purpose of continuing the play. Finite games are those instrumental activities - from sports to politics to wars - in which the participants obey rules, recognize boundaries and announce winners and losers. The infinite game - there is only one - includes any authentic interaction, from touching to culture, that changes rules, plays with boundaries and exists solely for the purpose of continuing the game.

Learning is an infinite system in the sense that it continues to change and evolve over time. More importantly, learning is a specific recurrent system:

11. Learning is an infinite and self-reinforcing system called a loop.
Self-reinforcing loops turn into flywheels in the Jim Collins’ “Good to Great” sense (or see his "Turning the Flywheel" monogram) of moving faster and faster through the cycle and requiring less effort each run-through. Positive (or additive) loops become immensely powerful agents of growth, while negative (or subtractive) loops can grow out of control and doom an effort — corporate, interpersonal, or emotional. Loops can be broken or interrupted but because they tend to reinforce themselves, the longer a loop continues, the harder it is to stop it.

One of the more famous loops nowadays is the OODA loop from military strategist and pilot John Boyd, who argued that all interactions are governed by competing loops of Observing-Orienting-Deciding-Acting, and that victory comes to those who are able to get “inside” someone else’s loop and perform it faster than them - essentially, being able to adapt to changing circumstances and also change those circumstances ahead of someone else’s decisions.

For fun (and because it often gets really simplified), here is Boyd's loop in its detail:

There are many other such loops as well. These loops can be intentionally additive – as in the case of intentional learning – or you can think about these loops with the intent to disrupt them.  Vipassana meditation is one such loop, recognizing that too often we are reflexive to the world rather than intentional, and emphasizing the careful observation of the self in order to add intentionality between what we observe and how we act. Similarly, paramedics use a "Access-Adjust-Act" loop when providing help in the field. Even Agile software development is a loop of this sort, with its emphasis on Plan-Do-Reflect cycles.

At the heart of each of these loops are the steps of:

  • Figuring out what’s going on by gathering information
  • Processing/filtering it
  • Using that processed information to inform your actions
  • Acting on that
  • … and starting over to figure out what the new reality is.

The aforementioned components of the learning process are one such loop than can be systematically used for self improvement, and they highlight some specific differentiations of what observation, assessment, and action are if your intention is to learn better.

Recognizing and clearly understanding them as components, what their purposes are, and how the come together is what enables us to go from having a haphazard learning process to being an active participant in taking advantage of the self-reinforcing properties.

So if I were to go back and draw the lines between all of these components, it would look like this:

  • Knowing
    Knowledge, experience, skills
  • Doing
    Training, performance
  • Assessing
    Coaching, mentoring, feedback, logging, reflection

You can't really skip any of these components without harming the learning process in some way, but you can "weigh" some more than others. For example, a good coach might also be a mentor, or you can use direct performance as a form of training – in the 1970s, the US Air Force learned from "Exercise Red Flag" trainings that real air combat was the best training for air combat.

The key insights are not the components themselves, but understanding what part each component plays and how it relates to the other components. Looking back on this – and looking back on my own learning process (reflection) – I realize where I have had gaps and where I can be more intentional. Everyone knows to train and to read to get more information, but now I understand where a coach versus a mentor might help, and what specifically I would want to get from them, depending on what I am trying to learn. While I've always been reflective, I've never done the meta work of logging in order to reflect more deeply of progress or warning signals over time.

Again, I wish this was something I had “learned” earlier. While I understood each individual component on their own, I never realized how they come together in a self-reinforcing loop, or how to maximize the value of each individual component through the learning journey. The crazy thing about school and education is that it often creates knowledge (and sometimes skills), but lost in the curriculum and individual subjects is the important meta skills of actually "learning".

But hopefully you can learn that now.

After publishing, reader David Buresund sent in this fantastic conceptual map of how the pieces come together. Thank you!

Thanks for reading

Was this useful? Interesting? Have something to add? Let me know.

Seriously, I love getting email and hearing from readers. Shoot me a note at and I promise I'll respond.

You can also sign up for email or RSS updates when there's a new post.