[EliRivers]

"As I've got older I felt I learned stuff much more quickly than when I started college."

Drifting into a personal monologue, I know I am a lot better at learning now than I was 15 years ago. Learning advanced mathematics now is much easier than it was then (measured in the rather subjective unit of "amount learned / time taken"). Languages also are much easier. The only difference is the amount of time I put in now compared to the amount of time I put in then. If I was in full-time education now (instead of a half hour to an hour a few times a week) I would be yomping through textbooks and courses.

I think some of it is that I just know so much more now, and I've got so much more experience of joining knowledge up and making use of the combined result. I've also got so much more confidence in my ability to learn, and I know that if I'm struggling, grinding through does get results. I don't get demoralised, I don't wonder if I'm ever going to be able to understand it; I just do it.

[jnbiche]

At around 40, I'm now learning things like math quicker, but then I forget them a year later, whereas the things I learned in college I still remember now.

Is this normal, I wonder?

You might say, well, if you're forgetting them a year later then you didn't really learn them, but I disagree. I feel more complete mastery of a subject matter more now than I did in my teens. No, I forget things I learn much more quickly now, and the only way to not forget them is to practice them at least each week, and that rapidly becomes a time sink.

[markism]

This might be as simple as the fact that you're spacing your learning out less. In college you learned things over a span of 15 weeks, giving you many opportunities for spaced repetition, which is one of the most effective methods to make material stick. Now I'm assuming you learn material in much less time than a semester, so there's less repetition and therefore less long term retention.

https://en.wikipedia.org/wiki/Spaced_repetition

[jnbiche]

Thanks for this suggestion -- this may well be the cause. Perhaps I'll start trying spaced repetition learning.

[trbvm2]

This is an expected result of a storage system that has finite capacity being tasked with storing a lifetime of useful information.

Research on this topic points to the notion that the brain is pretty actively trying to ignore things and toss out information that is no longer useful. It seems that the heuristic the brain uses to determine what should be kept and what should be discarded is related to how well some new information fits with information already in memory and how often an area of memory is revisited.

If you were tasked with fitting a lifetime of useful information on a 100 petabyte hard drive, how would you go about it? When you inevitably run out of free space, how would you continue to store new information?

http://brainworldmagazine.com/learning-memory-how-do-we-reme... http://neuroscience.uth.tmc.edu/s4/chapter07.html

[jlgray]

I agree completely about learning math faster, or maybe easier. My thinking has been that now we have a much better scaffold to construct knowledge on. Calculus, linear algebra, proofs, thinking in arbitrary numbers of dimensions, etc. are all much more ingrained than they were in undergrad, and as a result you don't have to spend as much effort on the details.

I wish I could say I didn't forget things when I was was in university, but I still remember the frustration of coming back from summer vacation and having forgotten half of what we learned last semester.

[lordnacho]

Same here. Machine learning wasn't in vogue when I graduated, but now I'm actually able to help my younger brother through the last courses in college. Quite surprising, because I'd have sworn when I was that age it would take weeks to understand something like Random Forest.

I reckon part of the reason is that when there's no pressure on you, you can explore around a question (how do you classify this particular thing...) rather than try to steer right towards the answer as fast as possible. Which isn't actually as fast as when you read around.