how to rapidly assimilate new information

post by dhruvmethi · 2024-10-24T02:18:00.648Z · LW · GW · 3 comments

Contents

  map out your current understanding
  targeted research
  apply the research
None
3 comments

Crossposted from here, where you can read more of my writing.

The quantity or extent of your knowledge can be crudely described by the formula below:

Quantity of experiences × Amount extracted from each experience

This means that you have two main levers to improve your knowledge base (i.e. learn):

I’m going to spend the rest of this article diving deeper into the second point above because I truly believe it’s one of the most essential skills anyone can attain. It’s a force multiplier; while we don’t have complete control over the quantity of experiences we have and the opportunities available to us, we have a dramatic amount of control over how we engage with and extract meaning from the things that we do experience. It then becomes a positive feedback loop because as you improve that skill, you hop on a fast track towards mastery over certain domains that then inevitably increase the opportunities available to you, putting you in more interesting and dynamic situations that push you to the boundaries of your existing knowledge and give you more opportunities for further learning.

I use a three-step process to maximize my learnings from each experience:

  1. Map out your current understanding — this is a crucial step that people often skip. Without making an honest attempt to explain the status quo, you won’t have a good grasp on what you know and what you don’t. This self-awareness is necessary for improvement in any domain (hence why the first step of any coaching program is always an initial skills assessment).
  2. Perform targeted research — once you’ve identified the gaps in your knowledge, you can perform effective targeted research and screen the information you come across based on whether it’s improving your understanding of your experience or not.
  3. Apply the research — the shorter the gap between when you learn something and when you apply it, the easier it is to retain. This isn’t a groundbreaking statement by any means. Still, it has a meaningful impact on the things that you should focus your learning towards: it’s much more effective to focus your efforts on problems that you are currently experiencing as opposed to preparing for problems that you may face in the future.

map out your current understanding

The goal of this step is to come up with a list of interesting, targeted questions that you can then figure out how to answer.

To get there, you have to first write a set of statements that encapsulate your current knowledge. Each of these statements should generally follow the following pattern:

Each action should describe something that happens, and each outcome should describe the observed result. Here are a few examples:

I’ll use the example of a car since it’s a good way to understand the concept. You can map out a rudimentary understanding of how a car works by using the following table:

(Of course, all of this is only true if you operate an automatic car. Other sorcery happens if you operate a manual one)

Most of the time, this is enough knowledge to operate a car effectively. Clearly, however, this leaves a lot out of the picture! A lot is happening under the hood, literally. And this fact is painfully obvious if you’ve ever encountered a situation in which your car stops operating as it should and you have no idea what to do.

This brings me to a more accurate formula describing each statement:

I read this blog titled “Reality has a surprising amount of detail” a while ago, and it lives rent-free inside my head almost daily. It’s so true; there is so much hidden complexity behind everything we see and experience.

In the case of the car, it is not obvious at all how pushing down a pedal propels a car forward! There’s so, so much that happens and has to go right for that action to result in the outcome that we see and expect. There is typically a really large gap between the actions we take and the outcomes we experience, and learning things means bridging the gap between action and outcome so that our “mental table” of actions and outcomes gets updated with many more actions and intermediate outcomes. This is how we demystify things.

Once we have created a mental table, we can then create really good, targeted questions that’ll help us uncover the magic that we don’t yet understand. In the case of the car, the questions can look like this:

This is a much better starting point for research than simply googling “How does a car work?” And naturally, as a result, it has a much better chance of actually translating into something meaningful for you.

targeted research

This step is fairly straightforward since we’ve done a lot of the heavy lifting by crafting targeted questions that are highly coupled with the bounds of our current knowledge. If you want to move extremely quickly (like I always do), the key thing here is to quickly iterate and update your mental model.

Let’s say we’re starting to perform research on the connection between a gas pedal and the car moving forward. A quick Google search will yield the following statement:

As the gas pedal is pressed, it turns a pivot that in turn pulls the throttle wire.

This is useful information! We should stop here and update our actions/outcomes table to this:

 

This is a meaningful step forward, and we’ve made progress in demystifying what happens. This new action/outcome entry in the table yields its own set of questions (what the fuck is a throttle wire?) that will be much more confusing but quite interesting to answer since we’re now playing with concepts, things, and ideas that are beyond our current frame of understanding.

Eventually, if you persist through the confusion and keep moving through the iterative process, you’ll end up with more than just a clusterfuck of browser tabs, but a better understanding of how everything works together.

apply the research

This is a critical step for the work you have done up to this point to yield any fruit. Without the opportunity to apply the knowledge you have gained, it will atrophy and become insubstantial faster than you expect.

The application of your research must involve creating or doing something; basically, any activity that activates the senses. I don’t exactly know why this is. Perhaps the usage of our senses activates different memory storage mechanisms that work really well when paired with the increased knowledge that we have put in our brains. Either way, it’s undeniable that a basketball player cannot get better by only watching videos on shooting, that an engineer cannot get better only by watching lectures, and that a leader cannot get better only by reading books.

This means that the most effective learning loops start when you’re faced with a concrete problem that you need to solve (i.e. your car breaks down, you need to figure out how to get your teammates to work harder, etc.) The problem serves as something that can anchor the focus of your learning loop to keep it constrained and effective, and it allows you to instantly apply things that you’re learning.

You can only deeply understand how important this is when you look at a few examples of this done wrong.

One example is how schools teach history. It’s completely backward! Using the principles outlined here, the teaching should begin with a set of phenomena that students can relate to right now as an anchor for their learning loop. If you’re teaching U.S. History, you should begin by having students outline current phenomena that they’re experiencing, and then create statements that articulate why students think these things are happening. Once the gaps in their understanding are evident, history then becomes the explanatory tool that helps them deepen their understanding of the present and be more conscious citizens. Instead, we currently start with something archaic and then move forward, burying the lede and only allowing students to understand the relevance of everything that they’ve learned at the end when their curriculum has caught up to the present day, by which point most of the things they’ve already learned have been forgotten.

I also get this wrong all the time. I frequently get caught up in rabbit holes reading about the problems other people are facing and how they’re solving them in incredible ways, and I get distracted and try to mimic them, thinking that by following a path they’ve shown to be successful, I’ll be able to attain similar rewards. For example, I constantly see so many entrepreneurs attaining dramatic wealth and status, and I delude myself into thinking that this is something that I really want even though I have no concrete problem that becoming an entrepreneur would solve. This is a red herring! The only way I’ll be able to create something meaningful with my life is to stay ruthlessly disciplined in solving the plenty of problems I am experiencing in my current life and attaining mastery over those. That means saying no to:

No matter how much I crave knowledge in all of these domains, I will invest the time to learn these when problems pop up in my life that require me to attain this knowledge to solve them. In the meantime, I’m going to stay locked in on solving the following problems:

Trust me, I’ve got my hands full, and after many years of chasing other people’s dreams and ambitions, I’m finally finding contentment with focusing on the problems that are squarely in my domain, having faith that staying focused and obtaining mastery over these domains will bring me everything that I want in a rapid and fulfilling way.

Coming back to the original point: the application of research and learnings is the fastest way to attain mastery. If you’re not completely focused on attaining mastery over something, you likely need to rethink your priorities.

While the process itself is relatively straightforward, it’s the principles beneath it that give it tremendous power and demand it to be taken with a level of seriousness.

It can also be applied to the past! You can unpack previous experiences you’ve had in teams, relationships, or whatever, and perform the same iterative process: map out your understanding of what happened and then perform targeted research to better understand what took place to prepare you for similar situations in the future. You won’t have the opportunity to immediately apply your learnings, but your past sensory knowledge of what occurred is a sufficient anchor for the process to still be meaningful and give you insight into what could’ve happened differently for the outcomes to change. I think this is largely what happens in therapy, and it is really effective at reducing the pain you experience from the past.

Either way, people have achieved tremendous things when they’ve focused on problems they’re experiencing in their current life and applying this learning loop to them. There was a story recently about someone who figured out how to treat their own breast cancer by injecting the cancer cells with viruses. That’s incredible! There are so many stories of things like this, and I particularly like this blog (Slime Mold Time Mold) which consistently puts out the most interesting research studies that challenge our A → B understanding of many things that we take for granted. I also love this story of a guy who put his toaster in the dishwasher to challenge our conventional understanding of how electricity works:

I put a toaster in the dishwasher.  I know; some of you have just decided that I am a total moron, and won’t read further.  That’s OK.  I learned two very important lessons from this little experiment:  (1) It is very difficult to discern the difference between Conventional Wisdom and Conventional Ignorance;  (2) When Conventional Ignorance is challenged, things can get nasty.

(Spoiler alert: the toaster continued to work.)

Everyone has the power to improve their lives and make tremendous contributions to the world. All it takes is discovering the frontiers of your knowledge and working intently and diligently to expand it.

3 comments

Comments sorted by top scores.

comment by StartAtTheEnd · 2024-10-24T22:50:02.136Z · LW(p) · GW(p)

I will not argue that any of what you said is wrong, because I don't believe it is, but I've personally found that learning too much too fast makes me sick. "Rapid" learning may be fine, but anything faster could have serious trade-offs. Consuming and digesting food os similar enough to consuming and digesting knowledge that many intuitions carry over (like the neusea for overeating, or getting tired of eating the same thing for too long, etc). 

When cramming for exams, I would sometimes go through 4000 pages in about two weeks, and it would result in a sort of confusion and nausea, and I'd have lots of loose ends and scattered thoughts floating around. Now, I didn't always do the exercises like I should, and my learning was more theoretical than practical, so it may just be that I didn't finish anything before moving on to the next part, leaving my knowledge unsolidified. So "sort of understanding" is definitely not a good stage to stop at, that's my mistake, and most people here probably know better than to do that.

However, I've heard of people trying really hard to study or remember something for hours a day, and then forgetting other important events going back like two weeks. Like memories of last weekend just disappearing and such. I'm not sure if older knowledge is at risk (if you can accidentally erase important things if you're too aggressive in your learning).

Maybe some people on here have stories to share? Not that it's likely. You need to be really low in conscientiousness to be as unstructured as I am, and to have a messy desk, a messy house, messy notes, and be obsessed enough with something that you forget to eat, or forget if it's currently morning or evening. And people like 'us' don't fit in here since we avoid "tedius" things, leading to messy and informal writing, and leading us to avoid knowledge which doesn't interests us but which is relevant if one wants to write an article on a subject. Perhaps med students have enough of a workload to understand the consequences of excessive learning, but I don't know how many of them use this site.

Apologies if your areas of interest doesn't extend to what I'm discussing here.

Replies from: dhruvmethi
↑ comment by dhruvmethi · 2024-10-24T23:11:16.186Z · LW(p) · GW(p)

I think that's completely valid, and I've often experienced that as well. I think, though, if you're properly taking the time to apply what you've learned and build sensory experiences based on the things you're learning, you'll have an artificial cap on the pace at which you can consume knowledge and be forced to learn at a speed that allows you to digest things fully and have things properly integrate with your previous base of knowledge instead of replacing things that were in your head before.

Not to say I am or anyone is good at applying everything they learned, and not to say that everything you come across should be properly assimilated, because most of it isn't really useful at helping you address the problems that you're facing. But I think if you take seriously the notion that you have to apply things to truly assimilate them, I think you'll find a healthier balance.

Replies from: StartAtTheEnd
↑ comment by StartAtTheEnd · 2024-10-25T00:01:44.892Z · LW(p) · GW(p)

Most of my learning took place in my head, causing it to be isolated from other senses, so that's likely one of the reasons. In some of the examples I know of people forgetting other things, they did things like learning 2000 digits of pi in 3 days, which is exactly something which doesn't really connect to anything else. So you're likely correct (at least, I don't know enough instances of forgetting to make any counter-arguments) 

most of it isn't really useful at helping you address the problems that you're facing

This is a rather commonly known technique, but you can work backwards from the problems, learning everything related to them. Rather than learning a lot and hoping that you can solve whatever problems might appear.

What I personally did, which might have been unhealthy, was wanting to fully understand what I was working with in general. So I'd always throw myself at material 5 years of studies above what I currently understood. When introduced to the Bayes chain rule, I started looking into the nature of chain rules, wanting to know how many existed across mathematics and if they were connected with one another. Doing things like this isn't always a waste of time, though, sometimes you really can skip ahead. If you Google summaries of about 100 different books written by people who are experts in their fields or highly intelligent in general, you will gain a lot of insights into things.