When you ask people why they read non-fiction, they are likely to tell you that their primary motivation is to learn. But are people’s behaviors really consistent with this being their motivation? Almost all of the information that we read is already forgotten days after having read it, and most of what remains is forgotten after months. Even those facts or ideas that are particularly interesting or useful to us are forgotten the significant majority of the time, or at least, only partially remembered. If we are lucky enough to be reminded of the same material at some point, we may recall having learned it, or be able to learn it more quickly than we would have if we had no seen it before, but little of such knowledge is actively available in our brains should we ever want to apply it.
If our goal is actually to learn information for the long-term, then it is necessary that we review what we learn. Highlighting and note taking likely help memories form, but become far more powerful techniques when you take the time to actually review the notes. It takes us about 3 to 12 hours to read a non-fiction book (depending on length, difficulty, our level of focus, and our reading speed). An especially compelling book may contain 20 ideas that are new to us and that we find interesting or useful, spread across, say, roughly 7 hours of reading. If taking notes on the most interesting ideas were to take 3 minutes per idea, that would add an extra hour to the reading process. If we then spent 2 hours, cumulatively, over the next 6 months reviewing our notes, the total time to consume the book would now be 10 hours rather than the usual 7. But instead of remembering only a small fraction of those 20 ideas in the long-term (at a level that makes them actually applicable) we could now remember far more.
So, to give an example, if taking notes and then reviewing them causes us to recall 8 of the book’s 20 new/interesting/useful ideas long-term rather than just 4, then we’ve just retained a whole extra compelling book worth of ideas for a mere 3 extra hours of time spent (rather than the usual 7). That would make this reading/noting taking/reviewing process far more efficient for learning than just reading non-fiction book after non-fiction book in the way that it is typically done. In other words, why spend an extra 7 hours reading yet another book that will be almost entirely forgotten, rather than just spending 3 hours to retain more than double of what you have already read? In my note taking experience, increases in efficiency of this magnitude are very realistic to achieve.
You can do much better than this, however. Flash cards (which lead you to try to recall something, and then check the answer) provide a significantly more efficient way to memorize than passively reading notes. And smart flash card software, like Anki or SuperMemo, can accelerate things even more. They use “spaced repetition” algorithms which automatically decide when to test you on different material based on when you’ve been tested on it in the past, as well as how difficult you found the material to remember in your past attempts. So you are only infrequently quizzed on knowledge that you are already likely to know, whereas knowledge you are likely about to forget is frequently refreshed. This means that you rapidly assimilate whatever is in your flash card database. With note taking plus spaced repetition software, we should be able to make our acquisition of knowledge many times more efficient in terms of interesting and useful ideas retained per hour of time invested.
So why do so few people use these techniques? It’s likely explained by a variety of factors. People don’t just read non-fiction books to learn, but also they read them for fun, out of boredom, so that they can say they have read them, and because they feel like reading non-fiction is good for them. Note taking is also not particularly enjoyable, so may detract somewhat from the appeal of reading. But even those most motivated to learn rarely use these methods, probably in many cases because they just don’t think about it. In school people often do highlight and sometimes even review their notes, but outside of the exam context these good habits disappear. And most people have never even heard of spaced repetition software.
So if you truly are motivated to learn, download a free copy of Anki, highlight (only the most important, interesting, surprising, and useful stuff) as you read, and when done, quickly type up that content as Anki flashcards. Then, spend a little time each week reviewing cards in Anki. I predict you’ll be very impressed at how efficiently you can assimilate knowledge!
Regardless of how much smarter reading anything makes us, fiction is more likely to make us dumber.
http://lesswrong.com/lw/k9/the_logical_fallacy_of_generalization_from/
Non-fiction is also likely to make us dumber, making learning a poor reason to read the way people do, which is as you say.
http://www.youtube.com/watch?v=RoEEDKwzNBw
http://lesswrong.com/lw/14q/why_youre_stuck_in_a_narrative/