<ul><li>Learn-as-needed approach. You have a problem that needs solving, so you go and learn the things that will solve your problem.&nbsp;</li><li>Learn-everything-you-can approach. Pick things which seem interesting and learn as much as you can. Don’t worry about whether you will find somewhere to use it later.</li></ul>

There are 2 broad approaches to self-education 📖 

Scott H Young

Learning

Self Improvement

Skills and knowledge transfer far less than we would expect them to. So “broad-based” education is mostly a myth. What we learn is usually specific and often stuck to the contexts where it was learned.

Many times it’s not obvious that knowledge can solve a problem until after you’ve acquired it. Unlike physical tools, mental ones are effectively invisible until after you’ve already learned them.&nbsp;

<div>
There's two, broad approaches to self-education. 
The first is a learn-as-needed approach. You have a problem that needs solving, so you go and learn the things that will solve your problem. 
The second is a learn-everything-you-can approach. Pick things which seem interesting and learn as much as you can. Don't worry about whether you will find somewhere to use it later.
Rather than argue for one case or the other, I'd like to present what I think are the best arguments for each approach, and leave it up to you to decide where you want to go between these two extremes.
<h2>Why Learn-as-Needed Makes Sense</h2>
<figure><img src="https://www.scotthyoung.com/blog/wp-content/uploads/2018/12/learn-as-needed.png"></figure>
A common finding in cognitive psychology is that skills and knowledge transfer far less than we would naively expect them to.
Consider a few (scary) results from research:

<ul>
	<li>Students who took high-school psychology do no better in a university class than those who don't.</li>
	<li>Students taught how to calculate the area of a rectangle, nonetheless struggle to calculate the square footage of a room for the problem of buying carpet.</li>
	<li>Despite years of education in reading, writing and arithmetic, the average American cannot answer many seemingly basic questions, such as how to "calculate the total cost of ordering office supplies, using a page from an office supplies catalog."</li>
</ul>

This suggests that "broad-based" education is mostly a myth. What we learn is usually specific and often stuck to the contexts where it was learned. Given the extreme volume of knowledge in the world, this favors an approach which is ruthlessly practical. Learning things in the vain hope that they will someday be useful is often going to lead to wasted time.
A second motivation for learn-as-needed is that it allows for better learning. Since the need is known before the learning, the actual lessons are properly stored and remembered. Needing to learn a programming trick to solve a problem you actually face will be remembered more deeply than a textbook example you don't care about.
Learning-as-needed, also needn't be an anti-intellectual or anti-educational approach. Knowledge solves life's problems, so if you look closely there's almost always problems in your life you could solve by learning more about it. As long as you don't shy away from learning new things, a person with this attitude could nonetheless spend a lifetime learning and solving their actual issues at hand.
<h2>Why Learn-Everything-You-Can Makes Sense</h2>
<figure><img src="https://www.scotthyoung.com/blog/wp-content/uploads/2018/12/learn-everything-you-can.png"></figure>
Transfer may be a stubborn problem, but those who have deeper experience and more exposure tend to apply what they learn more broadly. 
My own anecdotal experience suggests that those who have a deeper foundation of factual knowledge are better at creating analogies, pictures and extending the knowledge to new situations.
The most powerful argument for this approach, is that many times it's not obvious that knowledge can solve a problem until after you've acquired it. Unlike physical tools, mental ones are effectively invisible until after you've already learned them. Without learning things that seem unnecessary at the beginning, you can't discover how they're applied.
Learn-everything-you-can may also end up encouraging a higher volume of learning overall than learn-as-needed. This is because often there is an easier (short-term) way to solve a problem that doesn't involve learning something new. You can hire someone, ask for help or use an off-the-shelf solution. 
Learning has a lot of fixed costs, but the additional cost of applying previously learned skills is much lower. Therefore a learn-everything-you-can attitude may encourage higher investment than one which is overly practical.
<h2>Which Should You Apply?</h2>
Obviously the two modes aren't mutually exclusive. You may learn some things out of interest and others out of a ruthless practical necessity. I think this question boils down more to-which end should you push yourself towards?
My feeling is that the actual research on transfer and education casts a harsher light on many learning activities which are completely divorced from practical applications. Ethics classes may help you understand philosophy, but they don't make you a better person. Critical thinking classes may teach you what modus ponens means, but not how to avoid being duped.
This suggests that leaning too hard on an academic approach which is removed from real situations, may not yield the instrumental benefits in being able to think better and do more in the world.
However, at the same time, those who completely avoid learning theories often end up unable to see how new knowledge could solve their problems. Learning enough about different domains to be able to recognize the mental tools they have and where they work, may therefore be more efficient than mastering each tool in the toolkit. 
</div>

There's two, broad approaches to self-education. The first is a learn-as-needed approach. You have a problem that needs solving, so you go and learn the things that will solve your problem. The second is a learn-everything-you-can approach. Pick things which seem interesting and learn as much as you can.

Bite‑sized knowledge
to upgrade
your career

Should You Learn Things You Don't Plan on Using?

There are 2 broad approaches to self-education 📖

The Value of Learning "Useless" Things | Scott H Young

5 strategies for remembering everything you learn

The Science Behind Building General Skills | Scott H Young

Bite‑sized knowledgeto upgradeyour career

Should You Learn Things You Don't Plan on Using?

There are 2 broad approaches to self-education 📖

The Value of Learning "Useless" Things | Scott H Young

5 strategies for remembering everything you learn

The Science Behind Building General Skills | Scott H Young

Bite‑sized knowledge
to upgrade
your career