Read the intro section, the graphs and their captions, and then try to figure out everything in the paper from that. Look at the results section as needed. Then read the discussion section to see if the conclusions you made from the data are correct.
Vlad makes an interesting point I never thought about before, however I'm not sure how much the average abstract will help you since usually they're summaries of what the article is about; basically just "this problem exists so we did this experiment to find out more about it and found out this". Rarely do they ever go into any real depth, much less enough depth to jump straight to figures and know what's going on, unless maybe you're knowledgeable about the subject/field. As an undergrad reading a professional journal you most likely aren't. Even if you are, that's likely because the article matches up with your major or area of research, in which case it's not going to help you much for the MCAT since the MCAT is going to throw random stuff at you which, thanks to Murphy's Law, are almost certain to only be your weakest topics.
