Amen. You're on the right track with taking a problem-based approach rather than focusing on "studying" content. I would bet the farm (if I had one) that scores for every student on SDN would increase if they BEGAN their studies by taking and carefully analyzing either one AAMC exam, or at least a good portion of questions from the AAMC Section Bank.
IMO, you need to learn first what the MCAT will look like/require of you. Then you need to learn by
testing yourself with accurate MCAT passages/questions. The"Content First"/"AAMC Last" approach is leading too many students into surprising test day experiences--the real exam ends up being far more experimental, acronym-laden, and reasoning-based than they expected. They've memorized some crap...but the exam didn't ask them to regurgitate that crap....
Here's how I approach this process with my students, which I learned myself from my mentors at Altius:
STEP 1: Learn what real AAMC MCAT questions look like. You need to attempt/analyze at least 100+ to really get it.
STEP 2: Regularly attempt passage-based problems on MCAT science topics--THEN learn/review those basic science concepts in the CONTEXT of how they were tested by those questions.
STEP 3: REPEAT until you rarely encounter a problem or passage that leaves you feeling uncomfortable or unprepared.
Most students on here spend MONTHS doing content review and never see a real AAMC question until a month, or even a week or two, before the real deal...YIKES!!
With my students, I do problem-based or nothing. If you ask me a question during tutoring I ask you a question. I'll reason through a topic with you, but never give you the answer--because I know that you need to practice ARRIVING at an answer given a question. Most students study the other way around..."Tell me the answers...then ask me questions." The classes I teach are non-lecture format where we START the class with realistic MCAT questions on the assigned topic (i.e., enzyme binding)...then...only A-F-T-E-R everyone has seen and attempted a few enzyme-based MCAT passages, T-H-E-N we discuss the principles of enzyme binding.
Both attention and retention skyrocket using that approach compared to content review first, then test.
There's actually science behind this stuff...its not just my opinion. If you're a serious consumer of knowledge, check out a couple research papers below. In the literature this idea is called "The Testing Effect," "Test-Enhanced Learning," or "Retrieval Practice"...
My all-time favorite statement from one of these researchers:
"Subjects predicted they would recall more after repeated studying, even though testing enhanced long-term retention more than re-studying." In other words, the students participating in the study THOUGHT that additional studying would increase their recall, but it was actually testing before or during the study period itself that worked best for them. That's the exact misunderstanding in which most students here on SDN persist...and why the "status quo" will probably continue to be "Content Review First, THEN Practice Tests" for the foreseeable future.
SAGE Journals: Your gateway to world-class journal research
http://psycnet.apa.org/doiLanding?doi=10.1037/a0019902
https://www.gwern.net/docs/spacedrepetition/1991-mcdaniel.pdf
http://psych.wustl.edu/memory/Roddy article PDF's/Roediger & Butler (2011)_TCS.pdf
Examining the testing effect with open‐ and closed‐book tests