Hello Patrick. I've been reading through all the recent threads here.  I have some concerns about your approach to "modelling"...Based on your description of your work and these threads I presently have the perception that you are considering your approach to modelling to be "equal but different" to the model presented in WITW.  I have not observed you do your work or seen the results, but based on your writing here I have to wonder about how similar in purpose, method, or results your model is to that presented in WITW. Meta-programs as far as I can tell focus on content...More specifically they are attention filters imposed by the observer.  In other words if I approach an exemplar to model and elicit a set of meta-programs am I left with a clue about how they do what they do?  Do I have any clue about the sequence of states they occupy in achieving the outcome I'm modelling?  Modelling using meta-programs seems to pre-suppose a number of things which I suspect are generally invalid: 1) State is irrelevant 2) Sequencing is irrelevant 3) Non-verbal elements of the skill are irrelevant 4) Experts in a field are capable of objective analysis and insight as to their own behavior. I took your iWam survey tonight.  It began by imposing a context called work and then asking a detailed series of questions to elicit meta-programs.  One problem with such an approach is that there is nothing present to actually elicit my ways of thinking about a large set of contexts that you lump together as "work".  For virtually every question in iWam I sequenced through a series of 10-12 "contexts" that occur in my "work" on a daily or weekly basis.  Presented with iWam in the context of "work" there is simply no way to answer the questions consistently without thinking of specific contexts such as, "writing a detailed technical specification", "presenting a proposal to executive management", "conducting a team meeting", "brainstorming solutions to a crisis situation", "defining a plan to mitigate risk on a crucial project", "working with individuals in other companies to define industry-wide standards", etc.  The answers are very very different based on context.  iWam captures none of the differences between contexts, let alone the crucial skills of being able to elegantly move between these contexts. Assuming someone wanted to model what I do and found 2 other people with similar capabilities and you administered iWam my belief is that your results would be fairly random unless you specifically define a context.  If you were able to narrow down a specific context that was common to all 3 exemplars you may succeed at coming up with a list of characteristics that describe how we behave, but I am not convinced that you would have any beginnings of a clue as to how to replicate these behaviors or build a transferrable model.  Sure you'd be able to talk about it, but how specifically would you put this into action? The main aspect of NLP that attracted me to study NLP in the first place was the very idea that people in NLP could do modelling.  To my great disappointment over the last 11 years is that virtually all written descriptions I've seen of "NLP Modelling" projects is that they give you a list of characteristics held by the exemplar, but they very rarely provide anything useful for people in actually performing the skill that was "modelled".  It is virtually impossible to distinguish most of these "NLP models" from the 500 non-NLP books at your local book store with titles like "20 Things Winners Do". As far as I'm aware modelling has always been a form of "accelerated learning".  To go into the context with a set of distinctions like meta-programs that already impose specific filters on attention assumes you already know what you should and shouldn't pay attention to.  Now if you knew that I would have to think you either already have the skill or a model of the skill and there would be no bloody point in doing modelling at all.  As far as I can tell WITW proposes a model that basically says I don't have a clue about what's important (how could I?), I'm going to put aside everything else I know and just mimic until I can do the same things and produce the same results as the exemplar.  Once this is stable in the neurology, and only then (if ever) will I even begin to try to apply any kinds of f2 filters to this in order to build a transferrable model. From my perspective your approach to "modelling" is a way to stay stuck by never learning anything new and the latter is a way to constantly learn new things and surprise and delight yourself.  These are not equivalent ways of approaching the task of learning.  They strike me as fundamentally different. That said I don't dispute that you can get results with your content modelling approach or that being able to perform statistical analysis of the results of your profile is good for marketing and convincing scientists.  I am however at a loss at present to see how either of those issues are relevant to the pursuit of building models of excellence. I am presently unable to imagine how if Richard and John had had iWam back in the early 1970's that it would have been much help in eliciting useful patterns from Milton H. Erickson...or you can pick your top 3 favorite therapists from the early 70's.  Knowing that Milton was results oriented, and more into options than procedures might not cut it at replicating his skills.  Maybe that's just my personal limitation at this point in my development. Am I completely mis-understanding what you are trying to accomplish in your modelling work? Take good care, Loren