How do children learn?

Post Reply New Topic

Most of them learn randomly. First a child experiences a situation: I touched the stove, and I got hurt. Very soon she learns a rule to prevent such situations: Don’t touch stoves. Then she experiences similar situations and begins to improve her rule: Don’t touch things that make fire or turn red. This new rule works for more than just situations involving stoves. It helps her in dealing with far more situations than her first rule did. So with rules, situations are easier to understand which means that with rules, situations are more easily controlled, even if one has never experienced a specific situation before.

Then she learns a logic: Beware of electric and gas lines and machines because our flesh is conductive and not flame-retardant. Notice that a logic works for more than one rule; some logics apply to only a few rules while others apply to billions or more. So with logic, rules are easier to understand which means that situations are even more controllable, rules are more easily understood, the task of determining which rules to apply in certain situations is made much simpler, and finally rules are more effortlessly applied in those situations.

But this process of learning is far too chaotic. There is far too much entropy, i.e. the amount of chaos, in this method of learning. More chaos means more possibilities. Consider language. The more possibilities that a statement could be interpreted to, the more ambiguous that statement is. More ambiguity equates to more error in understanding, which slows the learning process. So how do we make this less random? How do we reduce the entropy of the educative process?

Let’s revisit the process of learning. First a newborn learns situations like: When I touch the stove, I get hurt. Imagine these as points in the empty space of a newborn’s mind (see Figure 1). Then they learn more situations and they begin to learn some rules like: Don’t touch hot things. These are vectors in the space (see Figure 2). A vector is a geometric entity that has both length and direction; think of it as an arrow. Note that when a rule changes from Don’t touch the stove, to Don’t touch things that make fire or turn red, this change is represented as the changing of a vector; it could be changing in length and/or direction. Note that the more similar situations you learn, the more likely you are to realize that you should make a new rule, i.e. the more points you’ve learned that lie along a straight path in your knowledge network, the more likely you are to realize that you should consider adding a vector along that path. If your mind makes this realization, then a new vector is installed along that path. Hence you’ve learned a new rule by projecting and more importantly, you’ll be able to tackle new similar situations that you’ve never experienced nor heard of or even imagined previously.

Then the newborn learns a logic like: Beware of electric and gas lines and machines because our flesh is conductive and not flame-retardant. This is represented by the local superstructure of vectors (see Figure 3). Note that the more similar rules you learn, the more likely you are to realize that you should make a new logic, i.e. the more vectors you’ve learned that are connected with each other, the more likely you are to realize that you should make a superstructure of those vectors. If your mind makes this realization, then a new superstructure of logic is installed on top of those vectors. Hence you’ve learned a new logic by projecting and more importantly, you’ll be able to tackle new similar situations and rules that you’ve never experienced nor heard of or even imagined previously.

With logic, rules and situations are less necessary to be learned because they can now be projected instantaneously, i.e. on the fly. What does it mean to be able to project rules and situations? Well most of this article is my mind's projections. I did not learn these things from a teacher, nor by reading. Instead, I learned them by projecting. The more logic one learns, the more accurately she will be able to project rules and situations, i.e. learn rules and situations without the help of teachers or even reading.

Full article with charts:
https://docs.google.com/document/d/1ztt ... t?hl=en_US

I thought they learned through curiosity, as if the child asked themselves "What would happen if I touched the stove?"

I think as a kid nearly everything I learned was through reading, mostly on my own and not because they made us at school. But not all kids are natural-born readers and the though of a kid not knowing how to read by the time they're seven because one of those extreme-hippie-parents thinks they need to learn everything on their own no matter how long it takes makes my skin crawl. It's dangerous to not how to read! If your doctor gives you medicine how will you no how much to take and not over or under-dose without reading the label???
And then there's math. I have always been bad at math, always hated it. As a kid I was afraid I would not be able to handle adult responsibilities like buying my own groceries because of it. I didn't have to like math but I still needed to learn it and without the extra help I got I might not have.

There's more than one way children can learn. The example in the first paragraph seems to include two of them.

Learning not to touch a stove by getting burned is an example of "Reinforcement". Generalizing this pattern of knowledge to anything that produces a flame is an example of "Accommodation". (These processes represent two important theories of learning, respectively, "behaviorism" and "constructivism".)

However, there are many other ways children can arrive at such conclusions. They might learn these lessons socially, for instance, by seeing someone else touch a stove and get burned, and observing how people react to other flames.

Besides answering questions they ask, it may be helpful to expose them to more social experiences, and to model the behaviors and attitudes you want to see them develop. Just don't go burning your hand on the stove

By sitting at a desk reading books and being lectured to for 7 hours a day. Duh!

By Hebbian processes of activation in NMDA receptors, in the hippocampus and other regions. That's the neuroscientific answer.

I've always thought it best to treat the learning process as a black box. I myself am very interested in how the learning process takes place, if only so that I can shorten the whole process to achieve my goal of evil genius.

All I can observe are:
1. Trial and error.
2. Internalization, then feedback to (1) again for verification. Repeat.
3. Lots of sleep.
4. Red Bull.

I seriously fear for my liver.

This looks to be a very elaborate paper. The only thing that disturbes me is that, for all of its math and logic, there are only 3 resources visible from which information was drawn.

There is a chineese proverb that sums up what is being said and what is currently empirically validated by educators.

"Tell me and I'll forget; show me and I may remember; involve me and I'll understand.”


I could go a lot more into this but, it's very late. I'll make a more elaborate post tomorrow XD. Good read though.