SAT考得高，MCAT就没问题？
https://forums.studentdoctor.net/threads/is-mcat-prep-about-hours-put-in-or-does-it-have-to-do-with-innate-ability.1431109/#post-22397831
I did take the old exam. All you needed back in my day was ~6 weeks to do either a review course or read a set of books. There were only a handful of discrete knowledge based questions, and maybe 1 page worth of equations to know. Maybe it changed significantly with the update.
Although, you'll notice there are many people out there who can briefly review and score in the top few percent, and many people out there who study up multiple times and consistently get non-competitive scores. I guess you could make up a logic there, like that a majority of college premeds don't know how to study, but then you'd still have to explain the narrow retest interval and why so many struggle to improve by studying more. As comparison for a knowledge based exam, the USMLE has massive intervals (16 point interval to contain 65% confidence) and doing more practice questions is the best predictor of scoring higher.
Compare that to MCAT. Brief search just turned up this small study - no significant MCAT correlation to GPA, major, etc but very significant correlations to SAT. Their value for SAT total to MCAT total was 0.45, which is quite good in the field of psychometrics; for comparison SAT verbal to MCAT verbal was 0.60 and that's about the highest you can find in studies like this.
It's not an IQ test in that it doesn't only test for g, but it's much more a passage based aptitude test than it is a content knowledge check.
Edit:
To throw a couple more sources in there:
1993 academic medicine, Montague and Frei - SAT was highest predictor of MCAT explaining 41% of variance (for comparison, all other model components including GPA collectively explained only 21% of variance)
1987 journal ntl med assc, Carmichael et al - SAT highest predictor explaining 57% of variance
So this has been a durable finding for decades!
