личные инструменты

Моделирование по видео

Re:Modeling without the immediate presence of the person to be modeled by John Grinder on Wed Apr 21, 2004 5:00 am

A careful reading of the model of NLP modeling presented in Whispering will unequivocally indicate that the absence of the model (live and available in the context where the patterning to be explicated is available) and the absence of any explicit representation of the context of observation and especially the responses of the person or group in that parallel context that NLP modeling is simply not applicable.

As I have stated a number of times on this website, there are a vast number of activities that are covered by the term modeling - all of them perfectly legitimate attempts to learn about patterning in the world. What distinguishes NLP modeling from all of these other forms are

1. the unconscious assimilation (all f2 filters suspended - the know nothing state)

2. the modeler's disciplined imitative practice (still in the know nothing state) of the patterning assimilated in the original context in a parallel context until the criterion has been achieved = the modeler can elicit in the parallel context the same set of responses that the model elicited in his/her/their original context in roughly the same time frames and with roughly the same quality.

OK, given all that, let's look carefully at your question. I will assume unless you signal otherwise that that the first of the distinguishing characteristics of NLP modeling - the suspension of f2 filters is adequate comprehensible from the descriptions in Whispering. This is not to imply that it is easy to do but it is simple enough to appreciate - the specific strategies for achieving such know nothing states is an independent question.

So, we focus on the second of the characteristics that distinguishes NLP modeling from all other forms of modeling. Let's say that we have a preformer, well known for her ability to X. This set of behaviors (X) occurs in same set of relatively well-defined contexts (C) and in those contexts when the model is doing X, she is (typically) consistently using the responses of the person or group is an integral part of how she does X. Let's say that that paragon, this avatar, this formidible representative of excellence of doing X in C has been captured on videos (thousands of hours of such behavior). We watch and listen to the videos and yes indeed, we intuitively (the only basis for making the call - intuition - as we as yet have no explicit model to use as an evaluation metric) judge the tapes to contain the patterning we wish to model. It is likely that there is adequate representation of the model in these thousands of hours of videos (assuming our intuitions are accurate) and thus, we could learn to imitate the patterning (X) presented by the model.

What's missing? - the entire phase III of modeling as presented in Whispering

In particular, to the extent that the model is offering patterning (X) that is interactive - that is, the model takes into account on an ongoing basis the non-verbal and verbal responses of the person or group as an integral part of determining what to do next enroute to the objective or successful conclusion of the interaction - to that extent, we have no way of appreciating (and therefore modeling) the interaction effect. However the situation is even more difficult, the videos will be focused on the model. And as we have already mentioned, thus removes an essential element in the patterning X. What is the group doing in response to the model that influences how the model selects (albeit, unconsciously) the next step in the process. Further, there is no representation on the video of the actual context - a prerequisite to deciding how we are to create a parallel context in which with disciplined practice, we will attempt to satisfy the explicit criterion described above. We are fundamentally in the position of attempting to pattern the game of tennis while laboring under the restrictioin that we may observe only half of the tennis court.

In summary, then, the absence of the responses to the model and the absence any explicit representation of the context precludes an application of NLP modeling.

I wish to point out that there is a largely unexplored and important implication of the statement above. Namely, that there is a special set of modeling contexts in which it is possible to apply NLP modeling in which there is no live model. This highly circumscribed set has three defining characteristics

1. there is adequate representation of the direct preformance of the model

2. there is no interaction effect present in the context in which the patterning (X) occurs

3. the responses of the person, group… can be somehow explicated

Did you get it?

The most obvious subset of this highly circumscribed set is certain types of athletic performance. Tennis is a highly interactive athletic activity = take, for contrast, the performance of the legendary Carl Lewis - a famous American sprinter. Say that our modeling task is to create an explicit model of Lewis' extraordinary sprinting ability (= X). Say we have adequate representation on video of his sprinting - note please that there are serious issues here for example, do we need videos of warm-ups, how close a set of video shots of his breathing, facial expressions, eye movement patterning… when does the performance begin? We will pass these difficulties without comment for now. However, it is important to note that the devil is in the term adequate when considering this characteristic.

Is there an interaction effect in sprinting - my experience says yes, from sprinting experiences in high school competitiion - it made a significant difference whether the competition was ahead or behind me in the sprint. So, perhaps shooting competition (an asynchronous activity - which thereby reduces the interaction effect) would be a clearer example. Let's pass this difficulty by acting as if it is for the argument of little significance.

What is the set of responses that the person, group… is making (outside of the interaction effects themselves) that determines whether in the parallel practice context the modeler has achieved criterion has been achieved. Here we get a break - the criterion is the time in which the sprint is completed - in this particular case, we have as close to an objective criterion as humanly possible.

So, have a careful look at the difficulties still inherent in the application of NLP modeling to any model who is not present (whether dead or simply unavailable) and note that there are limited set of such possibilities where the criteria for NLP modeling may be satisfiable.

At any rate, my simple request is to make the distinction proposed by Carmen and me in Whispering - model any way you wish however please limit describing what you do as NLP modeling just in those activities that meet the specifications of NLP modeling as outlined here and as practiced by Grinder and Bandler in their creation of the field of NLP.