>>10217821There's also an argument for the brain re-using components. In arithmetic, hand centers light up, and people who learn binary on their fingers deal better with binary in itself, and performance on tasks worsens if a manual task is simultaneously trying to use the hands while someone is (unintentionally) use that region for math.
I knew a guy who became blind from diabetes type 1 in his 20s, and before and after blindness he could do assembly programming with little to no errors. A well rounded person who just had an unfortunate life from diabetes.
I personally believe general problem solving involves an annealing process: Many regions activate, and then as the brain progresses through information, fewer regions become active, until you achieve the correct answer or the most preferable (easiest to think) method. In this style, strategies may vary on occasion because some are just mentally easier to handle at the moment, or resources available restrict methods. Transient errors are more likely because you're working on a probability your current idea will work.
The other method would be "perfect rote". The problem has a clearly defined chain of parts, and only by remembering each step accurately can you proceed to solve it. This involves a concreteness to methods, so more restricted, longer lasting connections are needed.
Note these two methods would be organically at odds with eachother: you can't have a perfect method that links all cells, and you can't have an annealing process that only involves a single chain. Most strategies would involve pieces of both, and problems would be more suited to one form or another. This also presents a challenge to an "expert" because if you have one strategy method, it's more or less preferable, but if you conversely have weak binding, your memory and consistency will suffer over time. In my opinion this is the origin between conflict between general learning and rote learning.
Cont.