NLP got its start through the process of modeling.
Modeling is how the Milton Model and Meta Model were created.
First, Richard Bandler had modeled Fritz Perls and Virginia Satir on his own and began forming study groups on-campus at the University of California.
John Grinder was an assistant professor at the time, and eventually, the 2 of them ran into each other.
Both of them were fascinated by Richard’s ability and wanted to formalize exactly what he was doing.
Grinder made an offer to Bandler that went something like “If you teach me what you are doing, then I will tell you what you are doing.”
From there, the rest is history.
The idea behind modeling is simple: You find someone who has above-average skill in a particular domain, you figure out how they do it, and you install that skill either in yourself or other people.
Modeling achieves 2 particular goals: getting a particular result and learning explicitly how to do it.
Not only does the modeler benefit from learning a new skill, but the person being modeled will gain a deeper understanding of how they do what they do.
Basic Principles of Modeling
NLP Modeling Applications
According to the dictionary, a model is a “system or thing used as an example to follow or imitate.”
This comes close to describing the NLP modeling process.
In NLP Modeling, we are creating a thing to be “followed and imitated” to achieve a specific result.
A common application of NLP modeling includes repeating or refining a performance by specifying the steps followed by elite performers or during optimal examples of the activity.
For example, let’s take the game of chess.
As of 2022, the top person in chess is a guy named Magnus Carlsen, considered to be one of the greatest chess players of all time.
If you wanted to model certain chess-related skills, then Magnus would be an excellent candidate.
Modeling is flexible in the sense that you don’t have to apply the skill you learned in the same context that you originally found it.
You can formalize a process through modeling and apply it to a completely different context.
The possibilities are nearly endless.
Levels of Modeling
When it comes to modeling effective behaviors, there are multiple levels we may need to look at.
The levels of modeling mirror the Neurological Levels, created by NLP Master Trainer Robert Dilts so if you’re already familiar with that, then you’re halfway towards understanding the levels of modeling.
If not, that’s ok. We’ll go over each one.
Each level contains 2 sub-levels called a macro-level and a micro-level.
Let’s dive in.
Environment
The first level is known as the environment. At this level, we are considering when and where the person is performing the activity. The environment determines the external opportunities and the constraints that the person must react to.
The macro-environment includes things like the general habit and social context.
The micro-environment involves specific locations like an office, classroom, auditorium, etc.
In addition to the environment itself, we also look at the influence the person is having on their environment.
Behavior
Behavior is specifically what the person is doing and how they’re reacting in their environment.
Macro-behaviors include things like general patterns, communications, and styles.
Micro-behaviors include things like tasks, specific work routines, and work habits.
Capabilities
Capabilities are how a person generates behaviors in a particular context. This can be in the form of a mental map, plan, or strategy.
Macro-capabilities include general strategies and skills like learning, memory, motivation, decision-making, and creativity.
Micro-capabilities include things like visualization, internal self-dialogue, or how a person specifically uses their sense or cognitive capabilities during a particular behavior or task.
Beliefs and Values
Beliefs and values tell us why a person does a particular thing in a particular time and place.
They help to provide reinforcement that either supports or inhibits capabilities.
On a macro-level, beliefs and values relate to the type of meaning, cause-effect relations, and boundaries people place on events or perceive in the surrounding world.
On a micro-level, a person’s beliefs and values can relate to processes on other levels. A person may have beliefs about their environment, behavior, capabilities, etc.
A person can also have beliefs about their beliefs called a Meta-State, but that’s a topic we’ll save for another blog post.
Identity
We all have an identity in relation to our beliefs, capabilities, and actions, within a given environment. In other words, we need to consider the who behind the what, when, where, why, and how.
In order to establish a particular role, one must have a mission within the larger system that he or she is operating.
Spiritual
In the context of NLP modeling, spiritual refers to how a person relates to the people around him such as family, colleagues, social class, culture, and so on. This is in regards to the larger system that one is a part of.
Vision
The vision is based on the spiritual level and helps to define the purpose of an individual’s behaviors, abilities, beliefs, values, and identity within the larger system that he or she is a part of.
This level tends to be what ties everything together and gives all the other levels their ultimate meaning.
NLP Master Trainer Robert Dilts suggests we visualize the relationship of each level as a network of generative systems that focus or converge on the identity of the individual as the core of the modeling process.
Modeling Capabilities
Most NLP modeling processes focus on the capabilities level, also known as the “how-to”. Capabilities are what connect beliefs and values to specific behaviors.
Without the “how”, knowing what one is supposed to do and why to do it is ineffective.
Even though a strong emphasis is placed on modeling capabilities, that doesn’t mean it’s the only level that is looked at. Creating an effective model is often a gestalt of values, beliefs, and specific behaviors that produce the desired capacity.
A huge benefit of modeling capabilities is that they tend to be nonlinear in application.
You can model a particular skill and apply it to a variety of different situations, tasks, or contexts.
We can understand how skills are organized using the TOTE model. The TOTE model provides a feedback loop between goals, the choices and means to accomplish those goals, and the evidence used to assess progress towards those goals.
According to the TOTE model, all mental and behavioral programs revolve around having a fixed goal and a variable means to attain that goal. As we think, we set goals in our mind, either consciously or unconsciously, and develop a TEST for when that goal has been achieved.
If the goal hasn’t been achieved, then we OPERATE to change something or do something to get closer to the goal.
When our TEST criteria have been satisfied, we then EXIT onto the next step.
In order to model a particular skill or performance, we must identify each of the key elements of the TOTE related to that skill or performance:
The performer’s goals.
The evidence and evidence procedures used by the performer to determine progress towards that goal.
The set of choices used by the performer to get to the goal and the specific behaviors used to implement those choices.
The way the performer responds if the goal is not initially achieved.
When it comes to modeling the TOTE elements of an exceptional performer, there are different levels we can consider such as the ones previously mentioned. We can consider the performer’s goals in terms of behavior, capability, beliefs and values, etc.
Different levels of goals require different levels of evidence and operations and have different levels of problems.
Some skills and capabilities are made up of other skills and capabilities. In that case, each sub-skill would have to be modeled separately. If you wanted to become a great chess player, you would have to master several skills such as visualization, calculation, evaluating, planning, etc.
We can apply the modeling process to different levels of complexity with respect to different skills and capabilities.
Here are some examples:
Simple behavioral skills: Specific, concrete, observable actions that take place within a short period of time, anywhere from a few seconds to a few minutes. Simple skills include things like getting into a particular state, shooting a basket, brushing your teeth, etc.
Simple cognitive skills: Specific, easily identifiable, and testable mental processes which occur within a short period of time. Examples include remembering names, spelling, and acquiring simple vocabulary.
Simple linguistic skills: recognition and use of key words, phrases, and questions.
Complex behavioral skills: A combination of simple behavioral actions done in a particular way. Examples include performing surgery, making a presentation, acting a part in a film, etc.
Complex linguistic skills: The skilled use of language in a highly dynamic, often spontaneous, situation. Examples include persuasion, negotiation, verbal reframing, storytelling, etc.
Complex Cognitive skills - Those which require a synthesis or sequence of other simple thinking skills. Examples include solving a math problem, designing a house, planning a wedding, etc.
Before doing any sort of modeling, one must consider the level, or chunk size, they will be focusing on. Distinctions and procedures for modeling one level of skill may be ineffective at modeling a different level.
Modeling Methodology
Phases of the Modeling Process
Before we dive into the specifics of modeling, it may be useful if we got a high-level overview of how modeling works.
Modeling can be broken up into 3 distinct phases.
There’s also some preparation we need to do beforehand to ensure we get started off on the right foot.
Let’s go over each one.
Preparation
Select a person who has the capability you wish to model.
You must then determine the following:
The context in which you will do the modeling.
Where and when you will have access to the person being modeled.
What relationship do you want with the person being modeled.
What state you will be in while doing the modeling.
Establishing the appropriate conditions, anchors, and lifelines, that will allow you to fully commit to the project.
Phase 1: Unconscious Uptake
In this phase, you are to engage the person being modeled in an example of the desired performance or capability within the appropriate context.
You start modeling them by going into “second position” (more on that later) in order to build intuitions about the skills the person is demonstrating.
You’re not looking for any specific patterns in this phase.
Simply take on the physiology and posture of the model and attempt to identify yourself with him or her internally.
Don’t try to consciously understand what they’re doing.
Not yet.
A useful state to adopt in this phase is the state of “not knowing”. This is where you drop all previous mental maps and assumptions with reference to one’s ongoing experience.
Once you have a good set of intuitions from being in “second position” with the person you’re modeling, you want to arrange for a context where you can use the skill you’ve been modeling.
Try out the skill within that context “as if” you’re the person you’ve been modeling.
Then, try to achieve the same result by just being “yourself”.
This will give you a “double description” of the particular skill you’re modeling.
When the result you get is roughly the same as those the person being modeled gets, then the first phase will be complete.
Phase 2: The Subtraction Process
In this phase, the goal is to sort out what is essential in the model’s behavior versus what is unnecessary and be explicit about the strategies and behaviors you modeled. You can use your “first position” (aka first-person) behaviors as a reference since you’re able to get similar results to the person being modeled.
Your main task is to clarify and define the specific cognitive and behavioral steps that are required to produce the desired results within the chosen context. You also need to systematically leave out pieces of any of the behaviors or strategies you’ve identified in order to see what makes a difference.
Anything that you leave out that doesn’t make any difference to the responses you get is not essential to the model.
When you leave out something that does make a difference to the results you get, then you’ve found a crucial part of the model.
Once you’ve completed this stage, you will have a “triple description” of the model: you will have a minimum model of how you replicate the model’s capabilities for yourself i.e. “first position”, you will have “second position” intuitions of the model’s capabilities that you developed from placing yourself in his or her shoes, and you will have a “third position” perspective from which you can notice the difference between the way you replicate the model’s capabilities and how that person manifested the capability in his or her original way.
Phase 3: Design
In the final phase of modeling, you take everything you’ve learned from the previous phases and design a context and procedure which enables others to learn the skills you have modeled, and thus be able to get the results that the person who served as the model has been able to achieve.
To produce the design, you must integrate all 3 perceptual positions, first, second, and third.
Rather than mimic or imitate the specific steps followed by the person being modeled, it is generally more effective to create the appropriate reference experience for the learners that will help them discover and develop the particular “circuits” they will need to perform the skill effectively.
Different people have different levels of conscious and unconscious competencies as their starting states. Some may be able to combine multiple steps of a procedure together into one step, while others may need a particular step broken down into sub-skills.
Steps of the Modeling Process
Another way to look at modeling is through an 8-step process. The first 6 steps cover both the first and second phases of the modeling process we mentioned previously. The last 2 steps deal primarily with the design phase of the modeling process.
Here they are:
Step 1. Engage the skill to be modeled within the appropriate context.
Step 2. Gather information from multiple perspectives.
Step 3. Filter for relevant features and patterns.
Step 4. Organize the patterns into a coherent structure/model.
Step 5. Test the usefulness and effectiveness of the model by trying them out in specific situations and contexts to make sure you achieve the desired result.
Step 6. Reduce the model to its simplest form that will produce the desired result.
Step 7. Identify the best procedures to transfer, or install, the specific skills identified in the modeling process.
Step 8. Determine the most appropriate instruments to measure the results of the model, and find the limits or edge of the model’s validity.
NLP Modeling Perceptual Positions
There are 4 primary positions from which information is gathered:
First position: “My Perspective” - Trying something out for ourselves and exploring the way we do it.
Second Position: “Your Perspective” - Standing in the shoes of the person to be modeled and attempting to think and act like the other person as much as possible.
Third Position: “Meta Position” - We observe the person being modeled as they interact with other people (ourselves included) as an uninvolved witness. We suspend personal judgments and notice only what our senses perceive.
Fourth Position - “Group Referencing” - Perceiving a situation from the perspective of the whole system, or the “relational field” involved in the situation. It’s an intuitive synthesis of all prior perspectives in order to get a sense of the entire gestalt.
At the very least, you need the first three positions for effective behavioral modeling.
Perceiving a situation from multiple perspectives allows one to gain broader insight and understanding with respect to that event or behavior.
Implicit vs. Explicit Modeling
Behaviors can be modeled either implicitly or explicitly.
Implicit modeling involves moving to “second position” with the modeling subject. You’re stepping into that person’s shoes and adopting their perspective as if it’s your own. From here, you can start building personal intuitions about that individual’s subject experience.
This is a more unconscious form of modeling and is used to examine behavior as a whole.
Implicit modeling tends to be analog in nature and is comparable to a child’s perspective.
It is an inductive process where we take in and perceive patterns of the world around us.
Explicit modeling involves moving to “third position” to describe the explicit structure of the modeling subject’s experience so that it can be transferred to others. This is considered a more conscious form of modeling and one of its main goals is to break down the behavior into its component parts.
We need both explicit and implicit modeling in order to model effectively. Without the implicit phase, there is no effective intuition base from which to build an explicit model. Without the explicit phase, the information that has been modeled cannot be turned into tools and techniques to be transferred to others.
Implicit modeling is more than enough if you want to help a person develop personal, unconscious competence with the desired behavior. If you want to create a technique, procedure, or skill set, that can be taught or transferred to others, then explicit modeling is required.
NLP itself was born from both implicit and explicit modeling. It started off with Richard Bandler implicitly modeling Fritz Perls and Virginia Satir through the use of videotapes and direct experience. Even though he was good at what he did, he couldn’t figure out exactly how he was doing it.
John Grinder was responsible for the explicit modeling. As I mentioned at the beginning of this post, Grinder made an offer to Bandler that enriched both of them in the process. Bandler showed Grinder what he was doing, and Grinder explained to Bandler what Bandler was doing.
Elicitation
Elicitation is how we gather information and identify relevant features and patterns related to the TOTEs of the person being modeled.
In addition to questionnaires and interviews, it’s useful to take a more active role in gathering information like role-plays, simulations, and “real life” observation of the expert in the appropriate context.
The most common NLP elicitation methods either has the person recall and relive a specific experience or carry out a task that presupposes or triggers a particular capability, strategy, resource, or problem.
Let’s say you have a person who’s a great chess player. You can ask them to recall a time when they were playing a chess game or you can have them play chess against a computer on the spot.
One advantage of the first method is that it allows the modeling subject to gain some distance from the experience in order to reflect on its structure. If there are aspects of the experience that are unpleasant, it will be easier for them to dissociate.
The disadvantage of the first method is that it may give mixed or contaminated signals because the person has to continually go through the process of recall. They may move in and out of the state being accessed which may create confusion for the viewer.
There’s also a lot of filtering (i.e. deleting, distorting, and generalizing) that occurs as they are determining what to access and present to others.
The second method’s main advantage is that it provides immediate “higher quality” information about the experiences or states to be explored and utilized. There’s less conscious filtering and more spontaneous, unconscious cues available.
The main disadvantage of the second method is that the modeling subject tends to get caught up in the content of the experience and is unable to reflect on the process and develop metacognition. There’s also the risk that the modeling subject becomes overly self-conscious about what they’re doing, which may lead to discomfort and dissociation from the experience.
Finding Relevant Patterns
A particular skill or capability, regardless of its level of sophistication or complexity, is made up of several dimensions relating to the functions defined by the TOTE. Successful performance of any type requires the ability to conceptualize, analyze, observe, follow procedures, interact with others, and manage relationships, to some degree.
Conceptualization, Analysis, and Observation are necessary for creating effective TESTs. They relate to establishing goals and determining evidence for success.
Following procedures, interacting with others, and managing relationships, are aspects of the “Operations” necessary to effectively reach the goal and satisfy the evidence needed for the “TEST” phase of a particular TOTE.
Here’s a breakdown of each dimension:
Conceptualization - The ability to conceptualize the whole and relate or fit something into the larger framework.
What is the purpose of the skill or ability?
When would you use it? Under what circumstances?
How does it fit with other competencies?
Analysis - The ability to break something down to its component parts; to categorize its elements.
What distinctions are most relevant to successfully performing this skill?
What do those distinctions indicate?
Observation - The ability to gather information in real-time.
What is most relevant to observe to successfully perform this skill?
What, specifically, do you need to be able to observe?
What cues or patterns are most important?
Following procedures - The ability to recall and enact a sequential set of steps that leads to a certain outcome.
What is the key sequence of actions necessary to successfully perform this skill? At what level or chunk size are they?
When and where is it important to follow the sequence of steps precisely? When and where is it important to be flexible?
To what degree do particular steps rely on particular observations?
Interacting with others - The ability to systematically elicit and react to the ongoing behavioral response of others.
Which possible reactions (on your part) go with which actions initiated by others? What is the intended result of those reactions?
What particular actions (on your part) are intended to elicit particular reactions from others? What is the desired result of those reactions?
What cues let you know when it is time to act, react or change actions?
Managing relationships - The ability to recognize and select appropriate behavior in relation to roles, norms, contexts, etc.
Under which conditions is it important to vary the pattern of interaction?
How does the internal state, both within yourself and others, influence and alter what you do or how you do it?
What desired state, both within yourself and others, is the intended result of the procedure or interaction?
Getting the answers to these questions will help you determine what needs to be taught or provided in order to transfer the capability to others and what will be the evidence that the capability has transferred over.
Finding relevant patterns involve 2 key processes: feature detection and pattern recognition.
Let’s cover each one.
Feature Detection
Features are specific qualities or characteristics that we decide to filter for as we’re modeling. This includes characteristics like rep systems, linguistic patterns, eye movements, and so on.
The features we look for determine the kind of patterns we find.
Here are the most common features used in the NLP modeling process:
Physiology - Postural patterns, gestures, eye movements, and nonverbal patterns like voice tone and tempo.
Cognitive Strategies - Noticing any particular rep systems, submodality patterns, and habitual cognitive sequences.
Meta-Program Patterns - General organizational patterns like time perception, relationship to significant other, orientation towards, goals, etc.
Meta-Patterns - Patterns in the way that the modeling subject communicates or relates to others involved in the situation.
We can classify distinctions in terms of their level of capability that is the focus of a particular modeling project:
Simple Behavioral - Specific physical cues and actions.
Simple cognitive - Rep systems and modalities.
Simple Linguistic - Meta Model patterns and predicates
Complex Behavioral - Perceptual Positions.
Complex Cognitive - Meta Program Patterns and Logical Levels.
Complex Linguistic - Sleight of Mouth Patterns
Pattern Recognition
Pattern recognition is the process or procedures used to identify the particular features or distinctions that are the most important for achieving a goal or result.
The most fundamental way to accomplish this is to find a group of people that demonstrate the desired capability or are able to achieve the desired result and find the similarities and differences between them with respect to the features and characteristics you want to explore.
Your main goal is to determine the specific features that are common to the processes used by all of the individuals being modeled.
In NLP, there’s a method called contrastive analysis, which involves observing which features are always present when a particular result is achieved and which are always absent when a result is not achieved.
For example, a person may notice during their creative state that they have constructed visual images and positive internal self-talk. When they are unable to reach that state, there are no such images and the presence of a critical internal voice.
Contrastive analysis is not a foolproof method. For instance, one can use contrastive analysis to show that the main works of genius were done by white Caucasian males of European descent. However, this does not mean that women and non-European people are incapable of producing works of genius.
Defining A Modeling Project
Here’s a typical sequence of steps for defining a modeling project:
Conduct a needs analysis to determine the specific issues, contexts and skills to be addressed.
Select the individuals to be modeled.
Set up and carry modeling scenarios and procedures in order to engage the capabilities and performance to be examined and gather the necessary information.
Identify the relevant patterns in the behaviors, strategies, beliefs, etc. of the individuals that have been modeled.
Organize the patterns that have been discovered into a descriptive and prescriptive structure.
Experimentally test and refine the model by trying it out in relevant context(s) to see if it achieves the desired result.
Design effective installation/intervention procedures and tools in order to transfer and apply the key elements of the model to others.
Measure the results obtained by applying the model.
Here’s a link to a worksheet to help you define your modeling project.
Throughout your modeling project, it is important to define your evidence and evidence procedure through each phase, including preparation.
Here are some useful questions to ask yourself at each stage:
Preparation
What criteria will you use (have you used) to identify the individual(s) you will be modeling?
How do (will) you know that the person(s) to be modeled have the desired skill?
How do (will) others know that the person(s) to be modeled have the desired skill?
Phases 1 and 2 (Gathering info via implicit and explicit modeling)
What criteria will you use to know that you have effectively modeled the capability?
How does the person being modeled know that he or she has the desired skill or has achieved the desired result?
How do (will) you know that you have learned the skill possessed by the model(s)?
How will others know that you have learned the skill possessed by the model(s)?
Phase 3 (Design)
What criteria will you use to know you have effectively transferred the ability?
How will you know that others have learned the skill possessed by the model?
How will others know that they have learned the skill possessed by the model?
Answering these questions will help you determine and design the most effective modeling processes.
Modeling Strategies
Modeling strategies are subprocesses within the overall NLP Modeling methodology.
Modeling strategies tend to be used during explicit modeling or when creating tools and procedures to transfer capabilities to others.
Modeling strategies utilize both inductive and deductive processes.
Inductive processes involve perceiving the world around us.
Deductive processes are those through which we describe and act based on our perceptions.
The difference between inductive and deductive processes is the same as the difference between understanding a language versus speaking it.
Micro Modeling Strategies
Micro Modeling Strategies involve modeling pieces of a specific skill such as a simple behavioral, simple cognitive, or a simple linguistic ability.
Here’s a simple 3-step process for micro modeling:
Identify the skill you want to find out about.
Have the person to be modeled demonstrate an example of that skill in a specific context.
Elicit the person’s TOTE for applying that skill in that context.
TOTE Questions
What is the context in which you commonly use the skill to be modeled?
What are the goals and objectives that guide your actions as you apply the skill in this context?
What do you use as evidence to know that you are accomplishing your goals?
What do you do to get to those goals? - what are some specific steps and activities that you use to achieve your goals in this context?
When you experience unexpected problems and difficulties in achieving your goals in this context, what specific steps or activities do you take to correct them?
Macro Modeling Strategies
Macro Modeling Strategies involve identifying component skills of a more complex or involved ability such as a complex behavioral, complex cognitive, or complex linguistic ability.
Here’s a 3-step process for Macro Modeling:
Engage the person in a context that requires the ability.
Identify specific behavioral examples and demonstrations of the ability to be modeled.
Starting with the behavior, elicit various levels of the process (how, why, who) that support the behavior.
Multi-level Modeling Questions
What is the context or environment you are exploring? When and where does the capability or activity occur?
What are the specific behaviors with the capability that you're exploring? What aspects of the behavior are particularly significant in order to achieve the desired result?
What internal thoughts and capabilities are associated with that behavior? How do you think when you are acting in that way?
What values are expressed by your behavior and capabilities?
What beliefs provide motivation for your thoughts and activity?
Who are you if you engage in those particular beliefs, values, capabilities, and behaviors? What is your identity?
What is your mission? Who else are you serving with this activity?
What is your vision of the larger system in which you are pursuing that mission?
Applied Modeling Strategies
An effective applied modeling strategy is made up of 3 parts:
Identifying key capabilities possessed by the individuals who are able to achieve a certain outcome or result.
Specifying particular individuals who may benefit from being able to learn those capabilities and achieve those results.
Defining which of those capabilities are most needed by individuals who require the skill or desire to achieve the result.
One common approach to applied modeling is to identify a need or problem and then find someone who possesses the capabilities and resources necessary to effectively deal with the need or problem.
Another approach involves identifying the capabilities possessed by individuals who are able to achieve a particular outcome, and then identifying a group of individuals who could benefit the most from those capabilities.
Applied modeling can help with putting the information we gathered from other modeling strategies into practice. We can structure this information gathering using a “Present State-Desired State” format known as the SCORE Model.
The SCORE Model allows us to define the essential features of a particular problem space.
Here’s a breakdown letter-by-letter:
S - Define the Symptoms associated with the present or problem state.
C - What are the Causes of those symptoms?
O - What is the desired Outcome that would replace those symptoms?
R - What are the Resources necessary to transform the symptoms and their causes?
E - What are the long-term Effects of achieving the outcome?
Generally speaking, the symptoms and causes are embodied by people who need or desire the capability to be modeled.
The outcome, effects, and resources are embodied by the individuals to be modeled.
Here’s a basic applied modeling strategy:
Identify the full SCORE defining the “problem space” to be addressed by the modeling project.
Elicit the following:
A multi-level description of the “problem space” of the individual who need the resource being modeled.
A multi-level description of the capabilities of the individuals who possess the resources necessary to reach the desired state.
Transfer the relevant levels of resources possessed by the successful individuals to the individuals needing those capabilities.
Applied Modeling Questions
Symptom - What are the specific, measurable, or observable symptoms to be addressed by the modeling project?
Causes - What are the causes of those symptoms?
Outcomes - What is the outcome or desired state to be reached, that the individual(s) to be modeled is able to demonstrate consistently?
What resources does the person being modeled have that allow him or her to:
Consistently reach the desired outcome.
Deal effectively with the symptoms.
Address and transform the causes of the symptom.
Move in the direction of longer term positive effects
To Be Elicited From Individuals That Need The Resource
Are there any contextual or environmental constraints that the individual who needs the capability must contend with?
What specific behaviors are they currently engaged in? What are their problem behaviors?
What specific cognitive capabilities are they lacking or have that causes them trouble?
What beliefs do they have that either limit or disempower them?
What values or hierarchy of values, are the individuals who need the capability, operating from?
How do they perceive themselves?
Do they have any sense of mission or vision with which to organize their identity?
To Be Elicited From Individuals Who Have The Resource
Are there any contextual or environmental opportunities that the person has?
What observable behaviors demonstrated by the person being modeled, are different from those that need the capability?
What specific mental capabilities or cognitive strategies are employed by the person being modeled?
What beliefs do the person being modeled have that allow them to cope more effectively?
What values, or hierarchy of values, does the person being modeled operate from?
How does the person being modeled perceive themselves?
What type of vision and mission does the person being modeled use to organize their activity?
Applied Modeling and the “Back-Propagation” Process
The final stages of Applied Modeling share many similarities with back-propagation in neural network technology.
Neural networks are computer structures based on the way in which the brain functions. They are typically made of a number of interconnected elements that are used to create a type of model of some pattern or phenomenon. The model is formed as a function of “weights” or the strength of connections between the elements in the network. The “inner” model determines the output of the network.
A common strategy used by neural networks is called backpropagation.
To give you an example, let’s say a neural network is designed to recognize animals.
For argument’s sake, the image of a penguin is fed into the network.
The image becomes coded as a pattern of reactions in the network based on the current “weightings” of the interconnections.
As a product of this pattern, the network outputs a result, let’s say a lion.
If there’s a discrepancy between the expected result and the actual result, the weightings are adjusted accordingly.
The image is fed into the network again and the process repeats itself.
After a number of cycles, the output of the network will begin to match the desired result more often. In other words, the computer has created a useful model, one that achieves the desired result given a particular input.
A similar method is used in NLP to refine a model after it has been established.
In the case of NLP, the human being would be the “neural network” and the different elements includes things like rep systems, beliefs, values, language patterns, etc.
Focusing on one particular distinction is like the process of giving weight to an element in a computer network. Where we place our attention creates an “attractor” that stimulates “self-organizing” behavior in a person.
For instance, noticing a person’s facial expressions rather than the types of clothes they’re wearing will alter the way you respond to that person.
Here’s how to apply the backpropagation approach to applied modeling:
Try out the steps, strategies, and distinctions defined by the model, within the appropriate context.
Notice the results you achieve and compare them to the desired result.
Adjust the steps and distinctions proposed by the model in order to make a “better approximation”.
Try out the new adjustments to the model, and continue to repeat the process until you (or the person for whom the model is intended) can achieve the required “threshold” of the desired result.
Code Congruence
Gregory Bateson, a highly influential figure across many fields like social science, anthropology, linguistics, and others, once said that “If you want to think about something, it is best to think about that thing in the same way in which that thing thunk.”
Bateson's understanding of code congruency asserts that the most effective and ecological models are those in which the relationships in the model match the relationships between the system of elements of the phenomenon which we are modeling.
An example of code congruency is the path of the planets. When people believed that Earth was the center of the solar system, they had to come up with complicated math equations to account for the movements of the other planets.
When things changed and the Sun was placed at the center, the math became a lot simpler and more elegant.
Models which are not code congruent may be useful in some cases, but they have their limits.
Code Congruence in Behavioral Modeling
In the early days of NLP, Richard Bandler and John Grinder decided to conduct a Modeling Seminar where they would model the work of Virginia Satir.
They decided to break it up over 2 days. For the first day, Virginia would work with a family in the morning, demonstrating her approach to Family Therapy. Bandler and Grinder would reflect on her work and describe some of the key behavioral and linguistic patterns she had applied during the session.
The next morning, she would work with another family, leaving the last afternoon for a final reflection and closing remarks.
The first morning went as expected.
Later that afternoon, Bandler and Grinder described how Virginia "anchored" various family members using nonverbal cues, how she led various individuals into certain states, how she created and triggered various responses in the family members, and so on.
The next morning was a complete disaster. Virginia was unhappy with her work and so were the family members and everyone else involved.
According to Bateson's code congruence, Bandler and Grinder described Satir's actions in mechanical, cause and effect terms, which is probably not the way that Virginia herself thought about what she did, consciously or unconsciously.
Her poor performance on the second day was not brought on by the fact that she was conscious of the process, but rather, because the code used to model her process was not congruent with the structure of the actual process.
Code congruency does not have to do with the accuracy of the content of the code or model. The important thing is that the relationships between the elements and events in the model is congruent with the relationships between the elements and events making up the system we're modeling.