Simon Bridge said:
It is totally up to the supervisor and the school.
This is completely true. I worked with two different professors as an undergraduate. One told me he wanted code to do "X", gave me a recent paper that he had published, and told me to code it using FORTRAN - in such a way that we could efficiently implement it using parallel processing. I would check in with him periodically with my progress. He didn't really know how to do what he wanted me to do, so I couldn't get much help from him.
The other professor started out by training me in the different experimental procedures performed in his lab and gave me a few blackboard lectures over different methods of calculating the quantities of interest from the data collected. As time went on, I was given more freedom to write my own code, find my own software, perform my own experiments, and finally, write up my results for a presentation.
Comparing the two methods, I flourished much more under the second professor's guidance. It's not that I am not a good independent worker
now, but, when you are just starting to get involved in real research, a little guidance at the beginning goes a long way. When you speak with professors about the prospect of working with them, try to get a feel as to what responsibilities you will have, and how much help will be available if you need it. Different people work better under different systems. Of course, you will eventually have to be a good independent worker if you want to get a Ph.D. But a little help along the way never hurt anyone.