Par (sorry, can't get to my umlaut symbol) Your questions cluster around, "How does the person select the problem context?" As you say, if the issue is dancing better, then it is quite obvious that selecting a dancing context is appropriate. Let's look at some of the alternative problems you present. 1. You mention a person who wants to be happier in their life. I would recommend that the person select the context in which they are least happy -the strategy being that if this person succeeds here, the rest will tumble. 2. heroine addiction - the point in the context where the decision to take the drug occurs 3. housecleaning - simply the house itself will serve 4. having cancer - typically, a person with cancer experiences some of the contexts in their lives as more impacted by the cancer than other - those will serve well. 5. the incredibly fat issue - as you correctly stated, if the issue is eating (which is very frequently involved) then the fork and knife will work. If the issue is metabolic, then a combination of the involuntary signals with framing to ensure that the unconscious understands and accepts the objective of the exercise followed by a new code game has proven effective. General comment: the art of selecting context is serious task. In general, when we select a context, it must be concrete enough to ensure that the visual and auditory stimuli that define the selected (and hallucinated) context overlap strongly enough with what we will actually experience when we re-enter that set of contexts. The differences created through deployment of the high-performance state will generalize to all members of the set of contexts represented by the selected context. I have also noted that the selected contexts are never actually context historically available - they tend rather to be an amalgam of several distinct contexts - this works very well as it ensures greater generalization. However, please note that in my experience the selected context should be concrete as the point of this context selection is to serve as a container for the changes and to identify along which specific sensory lines the changes will generalize themselves. It seems very difficult to change anything abstract - abstact differences will emerge but through the manipulations of concrete representations. All the best, John