Dr. Liebmann: Thanks for your interest. Here are some comments to the points you have raised:
I. One important distinction when considering this point is whether excititoxic or 6-OHDA lesions of the PFC have been made. With both types of lesions, effects of destruction of specific subareas (as in our study) have not been studied with respect to subcortical effects. Lesions of the whole mPFC have produced somewhat inconsistent results. 6-OHDA lesions tend to increase subcortical dopaminergic activity, while excitotoxic lesions tend to have opposite effects. But again, these findings have not been reported consistently.
II. The major distinction that I can think of right now is the distribution of dopamine input to the mPFC which is densest in infralimbic and prelimbic areas but muss less dense in cingulate areas. So it is quite likely that the different subareas are also distinct in physiological terms, as you have suggested. And actually, the lack of effect of cingulate lesions on morphine and cocaine CPP might be due to the fact that not much dopamine is present in that area (if one assumes that prefrontal dopamine is somehow important for the CPP-inducing effects of these drugs).
III. I´m afraid I cannot give you an informed answer to this question but I would suppose that blood-brain barrier effects to not play an important role.
IV. I agree that self-administration would certainly be more ´to the point´ than CPP because it is the closest we can get to human drug taking behaviour. I do not quite agree, however, that self-administration is easier and better established than CPP. CPP experiments are in fact quite easy to run, with no surgical preparation of the animals and with minimal training and technical requirements. Self-administration experiments are VERY time-consuming and very difficult in practical terms (i.e. there are many things that can go wrong during the course of the experiment). Therefore I would say that CPP experiments are definitely ´easier´ to run than self-admin. experiments. I also don´t think that CPP is not well established. It is used in many many labs and almost 100 papers are published each year on this topic. Thus, although there are still some theoretical and practical problems with certain CPP procedures, I would say that it is one of the best (at at least, one of the most widely) established paradigms in behavioural pharmacology.
I hope these comments answer some of your questions.