Earlier quoted context omitted.
> particularly when humanities topics are far more amenable to self-study than STEM topics. Really? I feel exactly the opposite, especially for CS. How do you become a good writer without feedback from a good writer (or, at least, critical reader)?
CS is an outlier. I imagine mathematics could be a good outlier as well. Its probably easier to find someone to help criticize your writing as opposed to getting your hands on expensive scientific / engineering equipment (both because of cost and regulations) however.
In particular, the popular and effective Moore method of teaching math relies heavily on interaction with other students.
https://en.wikipedia.org/wiki/Moore_method
>[T]he content of the course is usually presented in whole or in part by the students themselves. Instead of using a textbook, the students are given a list of definitions and theorems which they are to prove and present in class, leading them through the subject material.