NLP began with a question: what do brilliant therapists do differently? The method of finding out was modelling: studying an expert closely and trying to copy what they do. It is the founding idea of NLP, and still its most sensible one.
The NLP version
NLP modelling usually involves:
- Choose an expert with a clear, repeatable skill.
- Observe their behaviour. What do they do and say, in what order?
- Explore their thinking. What do they notice, believe and picture?
- Build a model. A simple description of the pattern.
- Test it. Can you, or someone else, now do it?
- Refine and teach it.
Bandler and Grinder used this to study Fritz Perls, Virginia Satir and Milton Erickson (NLP part 1). Many NLP techniques were produced this way.
Why it’s harder than it looks
Experts can’t fully explain themselves. Much expertise is tacit. Research on self-report found that people are often unaware of what really drives their behaviour and offer plausible-sounding explanations instead. (Nisbett and Wilson’s 1977 paper is the classic; not rechecked here.)
Copying behaviour isn’t the same as having skill. Years of practice, knowledge and judgement sit under what you can see.
Models lose detail. Some Ericksonians say the Milton Model simplifies how Erickson worked (NLP part 3), and the research on NLP’s patterns has been poor (NLP part 2). The model was only as good as the observation, and nobody checked it systematically.
Survivorship and halo. We tend to pick experts because they’re famous, not because we’ve confirmed that their methods caused their results.
What expertise research adds
Psychologists have studied how experts get good. A central finding, from work by Anders Ericsson and colleagues, is that top performance usually comes from deliberate practice: focused, effortful work on specific weaknesses, with feedback, over years. (From general knowledge; specific figures not checked here.) A model of what experts do can show you where to aim, but the skill still has to be built.
Related methods, such as cognitive task analysis, are used in some professions to map how experts make decisions. They use structured interviews and observation, with more care than a typical NLP model.
Modelling well, yourself
- Pick a specific skill, not a whole person. “How she opens a difficult conversation,” not “be like her.”
- Watch more than once, ideally several examples.
- Ask them, but also watch what they do. The two can differ.
- Write the steps in your own words.
- Practise one step at a time, with feedback from a person or recording.
- Compare your attempts with the expert’s, and adjust.
- Keep what suits you. You don’t need to copy their style.
In hypnosis
Hypnotists have always learned this way: by watching skilled practitioners, then practising with feedback. Good training combines observation, supervised practice and theory (see hypnosis series, part 39).
The bottom line
Learning from the best is a good strategy. Modelling works as a method when it’s careful, repeated and tested, and when it’s followed by real practice.
Next in the series: Neuro-Linguistic Psychotherapy: The Professional Route for NLP in Therapy
This article is for general information and is not medical advice.