dritalin - I wrote one of the chapters you refer to in the "Educing Information" report. Others have offered the wisdom of their experience in their here (and there are some very experienced HUMINTers among them). In the HUMINT world that experience is remarkably valuable.

Having said that, anecdotes are dominating the public debate about intelligence interviewing. Some people are saying "I know a case where a guy was subjected to X and then he talked." Does that mean that X works?

My I re-frame your question a bit? Instead of trying to decide "what works better with Afghan EPWs", it might be useful to think about how you would develop a plan to determine what kind of approaches might be more or less successful with a particular detainee, from whom you are seeking particular kinds of information, in a particular context. Whether a strategy "works" is not simply a matter of whether a detainee "talks" or provides information. The objective is accurate, useful information in a strategically-relevant timeframe. It is certainly possible to create conditions in the interrogation that interfere with the accuracy of information.

Regarding the use of coercive techniques, the FM sets the ground rules for now. Those techniques do not appear to have a promising future in US interrogation policy. I did a recent book chapter with a couple of colleagues on recent trends in US interrogation policy and practice which I have on my SelectedWorks, if you're interested: http://works.bepress.com/randy_borum/ (Jedburgh - I hope providing this link does not violate the forum's ROE. If so, let me know and I will remove it. I was just trying to be responsive to the RFI - Thank you, sir)