You point out an important choice facing medical students and how they approach their preclinical education.
What is the purpose of learning knowledge that is deemed "interesting" by PhD researchers or by subspecialist clinicians but is otherwise irrelevant to the current clinical practice of medicine or is not something that you can expect every good doctor in any field to know?
Is it that there is a small chance (and it's small, because it's low-yield) that this knowledge-for-knowledge's-sake may BECOME relevant later and that learning it NOW will be EASIER than learning it LATER?
I don't think that's worthwhile, in terms of efficiency. Rather than learning how BID functions in apoptosis now in case there is a cancer drug 20 years from now that targets it, I would rather learn it in 20 years, if and only if there is that drug, when I have significantly more perspective on how important knowing this stuff is to my area of clinical expertise. You COULD learn it once now, at considerable difficulty because it lacks certain relevance and meaning in your life, and then you let that memory decay exponentially without any added meaning or repetition for the next 20 years. It doesn't come up on boards (because it's low-yield), it doesn't come up on the wards (because the drug doesn't exist yet), it doesn't come up in your research (because you choose to specialize in an area that, statistically, is highly unlikely to deal with this one tidbit of minutia). Then 20 years from now, the strength of that memory trace is <1% of its initial strength. Now you read about this new miracle drug, you have to almost essentially re-learn the entire concept to get that memory from 1% to 100%. I would rather learn it (go from 0% to 100%) in 20 years because it will have an added layer of MEANING and relevance. When items of knowledge are elaborated with meaning, they stick in your memory better, i.e. they are easier to learn. Further, I'll know that the effort I'm putting into it is going to be worth it because it will be relevant, rather than just knowledge for knowledge's sake.
Preclinical medical education is not like a liberal arts education. They're not trying to get us to open our minds and think about things in new ways across different disciplines and learn knowledge for knowledge's sake. Preclinical medical education is professional school. It's about dumping a ton of knowledge in a single domain on us, forcing us to create declarative memories for everything, hoping some of it sticks and that we can apply it in the clinical context.
You might object, "Remember you supposed to be a clinical scientist." meaning we are supposed to be advancing the field and that we are a step above the mid-levels, nurses, PAs, technicians. I get that. I don't think teaching us a broad variety of esoteric KNOWLEDGE is the way to cultivate the SKILLS it takes to be a clinical scientist who can advance the field. We can only be experts in a small slice of medicine and forge ahead there, and our success in that endeavor will not be determined by how much of it we learned in the first two years of medical school. My ideal curriculum would pare down the content of first two years into the most clinically relevant things in order to create time for research. A few progressive-thinking schools are already doing this. Duke has had this idea for decades with its 1 year preclinical model. Vanderbilt is following in its steps. Many other schools have the 1.5 year, like Penn and Emory.