Cognitive load theory and its applications for learning
scotthyoung.com
scotthyoung.com
Of course we have limits to how much we can do at once, but learning isn't like filling a bucket. Learning takes active work, effort. An analogy would be like saying 'weight lifting is easier with tiny weights' or 'don't lift too much weight.' Yes, but that's obvious, useless, and forgets the presumed goal of weight lifting, which requires putting in some effort. Similarly, you wouldn't force kids to listen to lectures about baseball or other sports and memorizing all the rules before letting them play.
Some of the other things mentioned on the poster article are false or only apply to rote, trivial learning. Remember most of this type of traditional psych research “includes participants who have no specific interest in learning the domain involved and who are also given a very short study time.”
Worked examples, like other passive learning situations, can cause an 'illusion of knowledge' - feeling like you know and understand, but not really.
Look at pre-worked answers, build rote knowledge, etc, before trying to solve problems? We learn more doing the exact opposite - see research on 'productive failure'. We learn best in context, when we have a need to learn.
Our intuitions about learning, teaching, etc, are often the exact opposite of reality. Here are just a couple of examples:
Khan Academy and the effectiveness of science videos https://youtu.be/eVtCO84MDj8
Measuring actual learning versus feeling of learning in response to being actively engaged in the classroom https://www.pnas.org/content/116/39/19251?cid=nwsltrtn&sourc...
As a former educator, my experience is in opposition to this. Students who looked at worked examples and then did practice problems improved their learning rapidly but often felt like they weren't learning that much. Students who worked really hard to "figure it out" and failed often felt really intense feelings of gratification and deep understanding but, in fact, often had a worse understanding of material. I learned (as one of the studies you links indicates) that feelings of deep understanding or learning aren't very correlated with educational outcomes except insofar as they keep you motivated.
For any flaws CLT has I've never seen an alternative theory of learning that seems provide guidance that is even half as effective, both professionally and personally (I've A|B tested both myself and students a fair bit).
Also, I think it's wrong to position CLT as opposed to active learning techniques. As long as students don't have to "discover" any essential knowledge, everything is kosher according to CLT. So e.g. peer instruction is fine (e.g. the Mazur paper under the other blog post "Evidence for Various Research-based Instructional Strategies: Countering Critiques").
Of course, since you mostly test them how good they are at repeating given solutions you wont see this difference in classes. But this difference is very apparent to how they perform after they graduate, there is a reason many people with stellar grades are horrible in practice and vice versa. If you never practiced working under conditions where you don't know exactly how to solve things then you are basically worthless at solving anything people care about.
Edit: I'd say that it is great to learn your tools that way, you don't need to understand them perfectly. But your core competence is worth investing the effort to practice solving novel problems. A physicist can learn math by just repeating solved examples, but a mathematician probably should practice solving things on his own. And solving things yourself is a lot easier in undergrad than waiting for your PHD studies etc, it is the same kind of practice just way easier since the material is easier.
Very few teachers do this in math/science/technology classes. Teachers in these subjects understand that it's the design/problem solving process that is being taught, not the particular details which are only retained on an as-needed basis.
The reality is that formalized education is rote by design. The reason many on HN were frustrated by formal education is because they are not the average student. When you are two standard deviations from the mean in ability (in either direction), school will feel like a very painful grind. Most on HN could do the entire K-12 curriculum in a fraction of the time that it takes the average student. However, the education system is designed for the average student, who by definition are the majority. Students who learn very rapidly tend to get tired of the "training wheels" which are employed in formal education. However, only once you have seen the full spectrum of student ability from the other side as an educator do you understand why they are necessary.
The real sad part of all this is that we lie to students and tell them that computer science, which is by far the most abstract and cognitively demanding subject taught in school/college, can be learned and even mastered by all. This demeans the field of computing, and often can make students feel inadequate, as they are told that everyone finds computer science very easy to understand. When faced with a challenging concept, they then immediately feel that they are stupid for not understanding it immediately.
Cognitive load theory helps with difficult subjects like CS, because it prepares the learner for intense effort. When I teach CS, the first sentence I utter to my students is: "This is going to be the hardest subject you've ever taken. Prepare yourself, and don't give up."
What I would say is that the thing that makes a lesson good for helping students handle unfamiliar problems well directly hamper the more concrete skill acquisition within the lesson and so you don't want to mix the two uncritically. Student outcomes are much better when teachers are deliberate about which lessons are about handling unfamiliar concepts and which are about more concrete skills like adding fractions.
Surely these two claims are mutually exclusive?
He follows that phrase with examples of more constrained theories.
I’m not sure why it’s so different than tutorials, where the opposite happens. Everyone just looks at the answers and gets nowhere. Maybe it’s something about understanding or caring. There’s a huge difference in learning when you get information that you want to know, that you need to know, versus you hope you’ll have a use for it some time later.
3x
---- = 12
2
3x = 24
x = 8
This contrasts with directing the learner to solve that initial equation, or to write a particular program. The utility (per this article and quickly tracking down some more material on it) is more for the first time learners, and as they learn more they'll end up getting more value from working problems out themselves (versus studying more worked examples).
Similarly seeing something written out is not the same as writing it (in terms of the processing going on in our brain)
My typical problem when confronted with an entirely new class of tasks or an entirely new programming language is "where the hell do I even start".