[hariharan_uno]

Not related to the app, but I wish people stop measuring education with time. In my opinion, kids must spend quite a bit of time understanding basic stuff. Information overdose reduces the quality of understanding. So, the metric shouldn't be "18 months of maths in 6 weeks". A child who learnt the same stuff in 18 months is able to comprehend advanced stuff in the later stages than a kid who learnt it in 6 weeks. Again, this is not a criticism about the app. We just need to give time for kids to learn stuff.

[robert_tweed]

There's no evidence that is the case though. This app is highly interactive and provides immediate feedback, so the child knows whether they are doing things right or wrong.

The fact that it's more like a game means it's much more geared to the way that children learn than just working through a bunch of problems from a book or on a blackboard. It also allows the kids to progress at their own rate: spending more time on bits they don't understand until they do understand them. Class learning tends to be subject to the lowest common denominator, or conversely some kids at the lower end of the scale get left behind.

There's significant evidence that learning by interacting is the best way to promote retention. It's probably also fair to assume that the app itself does not simply promote rote learning of facts either, but rather understanding how the concepts work and repeatedly applying them to unfamiliar problems.

That seems to me to be far more optimal use of the time and creates a solid foundation for going on to understand more advanced concepts later.

Alternatively, why not simply spend 18 months working through even more problems on the app to get better and better? I suspect that the law of diminishing returns kicks in pretty quickly and you don't get much benefit simply through longer exposure at the same level. Instead, you can just introduce more advanced concepts earlier and keep on learning. That doesn't necessarily result in information overload, so long as you are continually building on top of a foundation of throughly understanding the simpler concepts and continuing to use them.

For example, once you understand addition you don't really ever stop using it and forget how to do it. Rather, you go on to learn about multiplication, exponentiation, etc. and hopefully, do so realising that they are all just fancy kinds of addition. I think it's much better if teachers can focus on ensuring that their pupils have that kind of understanding and are capable of applying it, than for example making sure that they can recite all the multiplication tables.

Of course, how you should learn maths differs a bit depending on whether your ultimate goal is to be a greengrocer or a computer scientist. Hopefully we're mostly focusing on the latter now.

[e12e]

Indeed. If we had unlimited resources, the best way to teach would probably one teacher per one (or two-three) child(ren). When you're working closely with one student, you can focus on just what that student needs to learn/needs help with. That probably doesn't scale up to the wide level of schooling that a modern industrialized world needs though (Maybe we should leverage that wasted extra efficiency we've picked up to recruit everyone in the work force to work 4 days at their job, and 1 day as a teacher?).

It sounds like this kind of computer-assisted learning might be able to increase teaching/learning efficiency similarly to what can be achieved with good one-on-one tutoring.