What PhysioProffe said about not letting students be corresponding or communicating authors, as well as Drug Monkey’s post about authorship a few weeks back made me think that some of these differences are driven by the style of research amongst the sub-disciplines. It’s not just a clinical science/basic science distinction (and I’m not even going to get into the atrocities that are clinical phd’s). It’s more correlated with the culture of research. In disciplines that have lots of field work (such as ecology, paleontology, even some kinds of anthropology or astronomy), or fields where there can be a large theoretical component (some parts of Biomed Engineering or population genetics or math or some parts of physics), students do their own projects and produce single authored papers. These projects don’t need a large lab with lots of supplies to make the research feasible, and there are still lots of little funds to buy plane tickets to exotic places. The expectations for graduation (at least in the biological ones above) are 3-5 single or first authored papers. Because those folks tend to have to sing for their supper (i.e., TA), even now, they can come out of grad school and walk successfully into a faculty job.
I draw both kinds of trainees into my lab (as post-docs or PhD students). Although there is large within group variation, it is clear that some portion can be ascribed to this background, especially concerning expectations with respect to group interaction and mentor involvement. Because the selection gradient may be steeper for the independent folks, they tend to survive better when thrown into the big bad world of faculty responsibilities. That is, the one who survive to get a degree have had to learn how to run a project on their own. They’ve already had to be successful at getting funded. That doesn’t mean that there aren’t other things that need to be learned, not the least of which is how to play with others.