I have a mind-blowing insight for you. African tribes have figured this out. The way they teach history is by re-enacting events from myth and history. Those "silly" and noisy dances make a lot of sense.
As Confucius said,
I hear and I forget.
I see and I remember.
I do and I understand.
Participation in a tribal ceremony has the potential to teach history best - with highest chance of remembering.
I think there is a lot more "dark memory" that can be triggered by conversation or the right context.
I also think we translate a lot of the information we read so it integrates with our own understanding or context. For example, I think I better understand why the US responded to COVID like it did after reading US political history from 1970-2020.
Basically, I think humans are very good at detecting, remembering, and applying patterns for our own purposes, and books introduce us to patterns.
I actually agree with that. I think it depends on the nature of the material, how much else you already know about the subject, your intention in reading the book, etc. I was being a little overly general in phrasing it in terms of only walking away with "a vague memory".
The whole point of TFA is precisely this, combined with questions about how books might be better structured to facilitate a more active role.
I'm not saying we shouldn't, just pointing out that it's not practical to expect that after one causal read through of a book. At least books as they are today. Which leads to...
As I see it, the author's question is: can we create a new medium which affords such active engagement with the works we read more effectively?
That's fair. That, to me, turns the question into "how can we make books that work better?" instead of "why don't books work?" And I think that's a valid and useful question.
To the author's point that books aren't necessarily built around a theory of cognition / learning, and could be improved by incorporating that, I'd say "sure". I'm not at all opposed to the idea of making books better. I was really responding more to the earlier part of the essay, which seemed focused on hammering on the "books don't work" idea.
So the OP apparently thinks a novel construct can do what Yeshivas and Madresehs and Monastaries have been doing the past couple of thousand years.
Now if the OP can come up with an object that makes the user rock back and forth while in a trancelike state, then he probably deserves the academic fame that he clearly desires.
As someone who has deep experience of learning from novels and communicating what they have to offer, the alternate protocol you outline for learning from books applies also to novels.
Under the auspices of literature there is much enduring value that readers can glean by reading through once, rereading passages of interest, and synthesizing an interpretation or response.
If you want to learn from a book, you have to play a more active role. Personally I find that the process of learning from a book involves A. read it once, straight through like a novel - maybe underlining key passages in pencil, then B. reading again and taking detailed notes as you read, and possibly looking up citations, or related references (the Wikipedia page for a particular term or phrase, for example), then C. re-reading your notes a couple of times as well as reading and re-reading supplemental material that you found, and finally D. writing your own synthesis of the material.
Yes, that's a lot of work and yes it takes a long time and no, not every book justifies doing all that. Sometimes it's OK to just read the book like a novel, and walk away with just a vague memory of the highest level themes and maybe detailed recall or one or two specific points that resonated with you. It all depends on why you're reading the book and what you hope to get out of it.
Edit: and if it's a book with exercises, then add "do the exercises" as another step. That said, I see reading / working through a textbook to really "learn a subject" as being a slightly different thing from reading the typical pop-sci book or business book or whatever, that doesn't generally have exercises. "Learning a subject" to me usually involves using multiple books, videos, articles, maybe writing some code, etc.