Part Thirteen: Part A Re-Modelling Perceptual Positioning and Processing

This article first appeared in Rapport 56, Summer 2002

“Everybody has got to be somewhere.”

Spike Milligan


The human mind is a fascinating subject to investigate. As a modeller it is the most complex and mysterious of subjects to model. NLP has made a large contribution to our understanding of the structure and function of mind:  our subjective processing. One of the major aims of NLP is to expand and improve our subjective processing skills.

Increasing our understanding of our perceptual processing can add to our effectiveness and potentially enable us to understand others better and to assist them to improve their processing.

In this article I will investigate perceptual processing and evaluate the Perceptual Position model, outlining some new models and exercises to develop your perceptual processing.

Early experience with perceptual processing

My early experience with meditation and hypnosis had made it clear to me how fascinating the mind is. As a young psychology student I experimented with a variety of different models. The Gestalt therapy of Fritz Perls was very interesting and I gained a lot through practicing the extensive range of ‘awareness’ exercises (see “Awareness” by John O. Stevens (Steve Andreas as he is now called) for an excellent and still very relevant and useful book also the first half of Gestalt Therapy by Perls, Hefferline and Goodman).

I remember one Gestalt exercise from over twenty years ago, in a workshop in London where I also first heard about Bandler and Grinder.  In this particular exercise we were told to look around the room and open to any object that we felt a connection with. We were then instructed to become that object and to describe how it felt being the object. For all of us, life as the object was one of “being used, taken for granted, ignored unwanted, not appreciated enough”, and such like moans and complaints. Certainly there was an absence of celebration of life, quality, design, happy functioning and so on. It was after all a context for therapy and change rather than a celebration of life!

Two major issues are highlighted in this example. The first is how our own personal issues easily and powerfully surfaced when we lived as the object. The second was how powerful the context was in directing us to select “therapeutic” content and not some other, more positive, content. This demonstrates the powerful influence context has on any perceptual processing.

As I followed up recommendations to read the work of Bandler and Grinder, I managed to get all their early books via a relative in America and gained further models and tools for improving perceptual processing. In “Changing with Families” Bandler, Grinder and Satir had combined their talents and outlined some very useful patterns. One of Satir’s models was the relationship between Self, Other and Context. Satir had identified five ‘types’ of behavioural stances that people adopt under stress. Bandler and Grinder were able to extend these models to include language and sensory processing. I particularly liked the integrated nature of all the models and this inspired in me the value of quality modelling.

I would summarise the three levels of their modelling as follows:

Processing Area Self Other Context
Satir Categoryter Placater Blamer Computer
Bandler and Grinder Feel Visual External Auditory Digital

Satir’s other categories of Distracter and Leveller correspond to all three being used. The Leveller is a coordination of all three; the Distracter is a non-coordinated sequencing through them.

A pattern that worked well with this model was Bandler and Grinder’s language pattern of shifting reference. I found this to be a useful tool for expanding awareness and understanding in any situation. Shifting from self to other or other to self has many uses, for example:

  1.  Subject: “ If I said that to him he would get really angry”.
    Therapist: “If he said it to you would you get really angry?”
  2. Subject: “He doesn’t know how to express his feelings.”
    Therapist: “If he was you what would he be expressing about his feelings now?”

I had ample opportunity to practise the family therapy patterns and to use these patterns in my work as a groupworker with adolescents and families. One of the perceptual skills that is very important when working in groups is tracking the group as a whole, while maintaining an awareness of each individual’s point of view. This takes some practice and is very tiring at first but is crucial for intensive work. The self, multiple other and context worked very well as a large map for integrating the specific sensory acuity and language patterns.

All of these models and skills increased my appreciation of the importance of improving my own perceptual processing and that of my clients.


The Perceptual Position (PP) Model

One model that is used to expand what we are attending to in any communication is the Perceptual Positions (PP) model. The model was developed in the mid 80’s and comprises of three perceptual positions.

The first position is fully associated in your own physical space, operating from your own perspective.
The second position is to be associated as another person, experiencing the world from their perspective. Dilts and DeLozier give examples of “walk a mile in their shoes,” “sit on the other side of the desk.” Encyclopaedia p.941
The third position is an observer position beyond yourself and the other/s in the communication.

Since its creation there has been one main addition to the PP model. Robert McDonald added a 4th  Position, identifying with the whole communication system and the basis for a “We” or “Group” perspective.

The Benefits of PP model

The model is easy to learn and to teach. This appeals to students and trainers alike. As Dilts states the PP model has been become very popular in NLP.

“Since their development, Perceptual Positions have been incorporated as a major part of many NLP techniques. The ability to take multiple perspectives is an essential skill for leadership, teaching, therapy and wisdom. Reimprinting, Meta Mirror, Meta Map, Aligning Perceptual Positions (from Connirae Andreas ‘Core’ Transformation work), and the various NLP techniques used for Conflict Integration, Mediation and Negotiation, all use Perceptual Position as a primary modality of producing change and achieving desired outcomes.” Dilts and DeLozier, (2000), p943

Another benefit identified by Dilts and DeLozier is:

“One of the most useful aspects of DeLozier and Grinder’s reformulations was that they provided an operational process by which people could enter and experience each position, that could be connected with specific language patterns, physiology and internal representations (the three primary operators of NLP). Like all other NLP distinctions, perceptual positions are characterized by specific physical, cognitive, and linguistic patterns.” Dilts and DeLozier, (2000), p940

Evaluating the basis for the PP model

When I first came across the PP model I wondered “why?” To me it was a retrograde step as it was less precise than the existing Satir model of Self, Other, and Context that I had been using. The simplicity of the PP model means that it is easy to learn and is therefore very popular for beginners. The popularity that this simple model has developed within the NLP community at more advanced levels is however very concerning. Rather than being the beginning of constant improvement it sadly seems to be an end of development in this area for many people.
The important NLP questions are “Is the model useful?” If yes, then the next questions are “Can it be more useful?” From there we can go on to explore “What does it cover? “What does it not cover?  “Can we make improvements within it?” “Can we add to it?” Given the obvious importance of perceptual processing, any improvements could be highly rewarding for personal and professional development.

I will begin by evaluating the basis for the PP model and then the model itself.

“The NLP notion of perceptual position was originally formulated by John Grinder and Judith Delozier (1987) as operational extensions of the earlier NLP concepts of ‘referential index.’ ‘meta position’ and Gregory Bateson’s concept of ‘double’ and ‘triple description.’ Dilts and Delozier (2000), p. 938

I will take these three elements that they used and evaluate their use of them as the basis for the PP model. I will then evaluate the model itself.

From the summary above although some of the early elements of NLP are included the most obvious one, Satir’s work, is not included. In terms of the main three referred to:

1.Referential Index
This is a shift in attention from self to other or other to self. It is not a real shift only a shift in subjective experience, a shift in reference. It is stating the obvious, but in my experience of how the PP model is used it is not obvious to some in practice, when you move to 2nd position you are not really shifting position. What you are doing is rearranging yourself to process from a difference point of view within your current information base. You cannot become the other person; you can approximate their posture and perceptual processing. The approximation will never be perfect, as you cannot have all the detailed information. This is an interpersonal example of the well known “butterfly effect” that limits the accuracy of weather forecasting: an unknown detail, like the flapping of a butterflies wing, will increasingly undermine any predictive model through time. The more detail you have as the basis for prediction the more accurate your predictions can be but they will be increasingly wrong as time increases.

This is why 2nd positioning someone you know well with information about what they are thinking and trying to do will be more detailed and accurate than a stranger chosen without prior knowledge. This is easily tested.

2. Meta Position
A ‘Meta position’ is not a function of distance but a qualitative change in processing; holistic and evaluative. The inaccurate use of ‘meta’ in NLP for any disassociated or distanced position has irritated me for many years. Observing something does not mean that you are ‘meta’ to it. It may be easier to be ‘meta’ from a distance but distance does not guarantee it. As you are reading this you are attending to it, making sense of it. That is not a ‘meta’ position. You can continue reading and think about what I am trying to do by evaluating and where this might lead; is it relevant and useful? You are now at a ‘meta’ position in relation to this article. Both are useful. The important point is that distance, or position is not the key operator, they do not create the necessary processing required for a ‘meta’ process. I will be presenting some of my re-modelling of ‘meta’ in a future article.

3. Bateson’s Double or Triple Description.
“Even when the participants within the communication loop do not agree, their relationship is enhanced and the possibility of future cooperation is created when they are able to shift perceptual positions in relationship to the interaction. This shifting of perceptual position is referred to as “triple description” because there are, minimally, three different perceptual positions occurring within a communication loop at any time: those of me/myself (first position), the other individual (second position), and the witnessing of the interaction between these two (third position). Dilts and Delozier (2000), p

This is based on the ‘Multiple Description’ work of Bateson. Bateson was quite clear that there was more to “double description” than simply having two descriptions:

“But the two-eyed way of seeing is itself an act of comparison.” (Mind and Nature, p.87)

Bateson emphasised multiple description in relation to singular and many uncoordinated descriptions. The difference is encased in the old sayings that “someone could have ten years experience of ten times one year’s experience” and “It’s not just the experience but what you do with the experience that is important”.

The assumption of a more complex process taking place when all that is guaranteed is a an increase in the number of single descriptions is an example of incomplete modelling. It is an inaccurate application to state that three perceptual positions create a triple description. They are merely three single descriptions (which could create overload and confusion). To become a triple description some additional processing is necessary.

Exercise: A simple exercise will demonstrate the possibilities.

Take two toilet roll holders (or any tubes) and bring them up to your eyes, one for each eye, like disconnected binoculars

  1. Move independently at the same time
  2. Move together, coordinated
  3. Notice the difference.

The first will be two separate sights that will be 2-D and disorientating, the second will be stable but the basis for “binocular vision”.

Dilts and DeLozier seem unaware of this difference as do Ginder and DeLozier (1987, see pages, 65, 180, 322, 358) who make the mistake of describing situations where there is more than one description as being “double” or “triple” descriptions. This mistake in ‘epistemology’ is somewhat ironic on the part of Grinder and DeLozier as the problems of unsound epistemology are one of the main themes of their book.

Impoverished, Additive and Multiple Description

As a modeller I have been interested to note the use and misuse of Bateson’s distinction. I have identified three possibilities:

  1. Impoverished description
  2. Additive Description
  3. Multiple Description

Improving our processing will move us from impoverished through additive to multiple descriptions, only if we build a structure that can function to coordinate the new descriptions.

This is an example of incomplete or incompetent modelling. They do not present the other possibilities or how the multiple description works. The reader is left with a piece of dogma. This is not only poor modelling: it is misleading, non-developmental and unscientific. In practical NLP terms it is also the basis for poor or inconsistent results when used with techniques because following the directives of the techniques will result in only the additive not in the necessary and useful multiple description. The student has to add other elements themselves for it to work. If they don’t it will be an addition of content, which can be useful but not the quality shift expected. If they do add other things to get the quality shift they will do so from a wide range of possibilities.

There is a qualitative difference between ‘Impoverished’, ‘Additive’ and ‘Multiple’ description. I knew this experientially as a groupworker twenty years ago. Consistantly my new colleagues and myself as a beginner suffered from ‘Impoverished descriptions. We were unaware of what was happening. Adding more information was very useful but was not, on its own, a ‘multiple’ description. Adding more descriptions gives an increase in information but no qualitative difference until the information is interconnected, compared, contrasted and evaluated. My perspective would then shift from “additive” to “multiple”. There were significant other skills involved, skills that are not identified or included in the PP model.

The 4th position continues the aim of extending our perceptual information about the whole communication system. Similar claims of “multiple” description are made for the fourth position and for all three positions collectively. The claim is that they are a “multiple description, a triple description. A position, single or compound is insufficient to create this. A “Multiple perspective” requires more than simple thinking; it requires holistic or at least systemic thinking. This is not included in their Perceptual Positions model and was one of the areas where I added new DBM models to extend our understanding.

Why was this particular model developed and what further improvements can we make?

Dilts and DeLozier go on to outline :

“The basis for the various perceptual positions comes from the fact that relational experiences always involve more than one individual in the communication loop. The ability to understand the communication loop, and the ebb and flow of events that occur within the loop, is a powerful tool, enabling people to both improve communication and produce ecological outcomes.” Dilts and DeLozier (2000), p. 340

This is so important that is warrants further development. Any weakness or limitations, and any improvements will influence this range of NLP techniques, models and patterns

In terms of the PP Model there is scope for further modelling and improvements.

  • The PP model is not explicit about what to do when in the position.
  • The difference between fantasy and reality can be blurred.
  • There are other useful possibilities not included in the basic model.
  • It is not clear how the different models work together.

We will now outline some additions to the PP model and some new models that together with the PP model form an integrate model for Perceptual Processing


Bandler, Richard & Grinder, John, The Structure of Magic vol.1, Science and Behaviour Books, Inc 1975
Bandler, Richard, Grinder, John, and Satir, Virginia, Changing with Families, Science and Behaviour Books, 1976
Bateson, Gregory, Mind and Nature, Dutton, 1979
Bateson, Gregory, Angels Fear, Rider 1988
Bateson, Gregory and Ruesch, Jurgen, Communication, The Social Matrix Of Psychiatry, Norton & Co. 1951
Bateson, Gregory, Sacred Unity, Further Steps To An Ecology Of Mind, Harper Collins 1991
Dilts, Robert and DeLozier, Judith, Encyclopaedia of NLP,
Grinder, John, and Bandler, Richard, The Structure of Magic vol.I1, Science and Behaviour Books, Inc 1976
Grinder, John, DeLozier, Judith, Turtles All the Way Down, Grinder, DeLozier and Associates 1987
McWhirter, Re-Modelling NLP, Part 1 – Models and Modelling, Rapport (1998)
McWhirter, Re-Modelling NLP, Part 2 – Re-Modelling Language, Rapport (1999)
McWhirter, Re-Modelling NLP, Part 3 – Feeling, Conflict and Integration, Rapport (1999)
McWhirter, Re-Modelling NLP, Part 4 – Basic Structures and Processes, Rapport (1999)
McWhirter, Re-Modelling NLP, Part 5 – Planning, Problem-Solving, Outcomes and Achieving, Rapport (1999)
McWhirter, Re-Modelling NLP, Part 6 – Understanding Change, Rapport (2000)
McWhirter, Re-Modelling NLP, Part 7 – Facilitating Change, Rapport (2000)
McWhirter, Re-Modelling NLP, Part 8 – Performing Change, Rapport (2000)
McWhirter, Re-Modelling NLP, Part 9 – Organising Change, Rapport (2001)
McWhirter, Re-Modelling NLP, Part 10 – Unconscious Processes and Hypnosis, Rapport (2001)
McWhirter, Re-Modelling NLP, Part 11 – Metaphors, Rapport(2001)
McWhirter, Re-Modelling NLP, Part 12A – Hypnotic Inductions and Hypnotherapy, Rapport (2001)
McWhirter, Re-Modelling NLP, Part 12B – Hypnotic Inductions and Hypnotherapy, Rapport (2002)
Perls, F., Hefferline, Ralph, F., and Goodman, Paul, Gestalt Therapy, Penguin Books 1973 (1st published 1951)
Stevens, John, O., Awareness, Real Peoples Press, 1971


DBM is a registered trademark of Sensory Systems Training.

John can be contacted at:
Sensory Systems Training
162 Queens Drive
Queens Park
Glasgow G42 8QN
Phone: 0141 424 4177