I'm looking for research labs on campus to work for this summer and (hopefully) for the coming semesters as well. Last summer I worked for a research fellow who I did a lot of grunt work for and the topic of research was something that I lost interest in so I didn't gain anything from it. Now I'm looking for labs earlier than I did last year and for a topic that I'm truly interested in. Should I look for a big or small lab? How much of a factor should I place if the PI is well established in his/her field?
I actually have experience in both a large lab of about 20 people between grads, UGs, and post docs (with a well known PI, professor emeritus) and with a small lab with 3 or so grads, 1 post doc and a PI that was an assistant professor. For the purposes of medical school admissions I don't think that the stature of the PI matters much.
One advantage of a large lab is that they will be publishing papers much more frequently so you
might have a better chance at getting your name on a paper in a large lab. But then again, you will not likely get to know the PI very well because he/she will be far to busy to spend much time with you.
On the other hand, when I worked in a small lab we did not publish many papers (I did not get published). However, I spent many many hours working directly with my PI and he was able to really be a personal mentor for me for the few years that I was in his lab. Not only did this make for a great experience for me as an undergrad, but I think it made for a MUCH better LOR.
In the end I would argue that a small lab is better for pre-med research experience. Even if you do publish, I would argue that the only thing that really matters is what you can show you actually accomplished/learned from your experience. You might have your name on a paper, but if you can't prove that you actually did something meaningful then it's not that impressive. In my opinion, the best way to prove to adcoms that you did something of value and grew as a scientist as a result of your work, is to have a strong recommendation from your PI, which as I said is much more likely to happen in a small lab.