Remember that the MCAT is a kind of performance. Practice is key. It exercises your knowledge-base and makes the ideas accessible. Practice teaches you how the exam is itself a kind of choreographed exercise where timing is critical. BerkReviewTeach's advice is very good to focus in your practice to learn to recognize the particular kinds of questions that have a way of turning into time sinks and to have a strategy for those based around saving time. Practice teaches you catch yourself before losing ten minutes on one question. Don't do it. One question isn't worth risking a crisis in the last two passages.
In my experience the questions that cost a lot of time are generally going to be the quantitative problems and the deeper conceptual ones. In quantitative problems often difficulty with tiime can occur where you take a wrong turn in set-up. You might have started using a Newton's laws formula when work & energy would have given the answer much more quickly. Questions that also cause problems with time are the most sophisticated conceptually oriented ones such as those oriented around judgment regarding research conclusions. It's hard to give up or reassess a question you have already spent time on, so there can be a sunk cost fallacy that compounds the problem.
With regard to the conceptually subtle questions, in these you can find yourself caught between two answers. You are going back and forth between the passage and the answers. In this scenario there is a tendency, especially if you are having a good exam, to turn the perfect into the enemy of the good. Accept missing a question or two even if you are aiming for Harvard. Don't worry about being perfect. Very good is enough. If it helps remember that you're probably going to get the question right anyway. That's just probability. If you are stuck between two answers, your chance of getting the question right is greater than 50% because instinct is on your side. Trust your instinct and move on. Step back and instead of thinking about which answer choice is the most appealing, try asking yourself which is least worst. Make your choice. It can help acceptance to make a notation so that you know you can go back at the end of the section if you have time to spare. Move on.
For the overall management of pace I suggest checking in every half hour, not more than that, to see if you are on track for 20 questions / half hour. Of course this isn't in stone because things can vary depending on questions per passage and how things fall, but it's a good rule of thumb to catch yourself if you if you are having problems. Check in at roughly the half hour mark and make sure you are at around 20 questions. At the one hour mark, see that you are around 40. If you are on pace, you should finish with at least a few minutes to spare. If you have run into a tough passage or two and find yourself falling behind, then you need to bring out plan B mode to catch up. In plan B mode, you don't have the luxury to look back at a passage so much. Every passage has a question or two that is going to be difficult. Just two or three times in plan B mode, you will need to identify that question right away and stop yourself from dwelling on it because you are using the most difficult questions to harvest time back. Make a good guess quickly on three difficult questions where you have narrowed down to two answers . You probably got two of them right anyway. It's a small sacrifice to make sure you have time for the last two passages where there are ten easy questions in jeapardy. Try not to fall behind, but if you do, go to plan B mode for a few passages, and at the next check-in you will be back on pace.