The Swish, States, and Neuroscience: Continuing the Dialogue with Steve Andreas

Neuroscience, Nuclear Physics, French Cooking and the Swish

This post is in response to a continuing dialogue between Steve Andreas and myself, revolving around the swish (but occasionally veering off onto other topics). Steve and I hope this dialogue can act as an example of deep yet respectful debate of important areas of NLP.  Of course I am also deeply enjoying this dialogue and hope Steve and the rest of the NLP community who are reading this discussion are as well.

As usual, I’d like to take one aspect of Steve’s prior post and explore it a little more deeply. The aspect I have chosen is Steve’s wonderful line: “I have great respect for neuroscience, and also for nuclear physics. However, I don’t think either one has much to tell us (so far) about learning and change”.

That’s rather like saying I respect French cooking and nuclear physics, but neither one will help me entertain my friends from Paris! In this blog post I’ll focus on how neuroscience may indeed be able to assist with one of the key points of disagreement between Steve and myself: the difference between the ‘slingshot’ swish (aka Steve’s “butt joint”)  and the ‘standard’ swish (aka Steve’s “lap joint”).

Before we dive into the client’s brain and swish their neurons, I’ll preface by saying the principal use of neuroscience in change-work (IMO) is that it provides a complete and compelling set of metaphors to create change-potential in clients. That’s why I wrote a book on Neuroscience for coaches (‘Keeping the Brain in Mind’) with my co-author Melissa Tiers. Explaining how the processes you (as coach) will use with your clients actually rewire their brain, you set them up for success. For example I coach a number of hard-nosed New York business men and women on a long term basis. They may initially be reluctant to try any technique they consider “woo-woo”, but as soon as I explain how their ‘working memory’ operates, they are keen to get started as those “woo-woo” techniques have now become scientific. Science is the new religion.

Hebb’s Law and Spike Timing Dependent Plasticity

OK, let’s dive into that brain. Inside your client’s brain (let’s say they came for smoking cessation), you see a tangle of long cells, with sparks of electricity moving between them. These are your client’s neurons. When one neuron (the pre-synaptic neuron) fires, the neighboring neuron (the post-synaptic neuron) can either fire in response, or fail to fire in response. Whether the neighbor fires depends on a number of complex factors, one of which is how receptive the second neuron is to signal from the first neuron, i.e. whether or not the two neurons consider themselves to be ‘friends’ or ‘enemies’.

How do the neurons decide whether to be friends or enemies? Pretty much the same way you do! If one neuron fires, and its neighbor fires in response (“Hi neighbor, how are you today?”; “I’m just great, how are you?”) then they become friends; and the second neuron becomes more likely to fire when the first neuron fires. But if the first neuron’s greeting is ignored by being met with a chilly ‘failure to fire’ response, then the neurons become 'enemies' and second neuron actually becomes less likely to fire when the first one fires. 

This process is called ‘Spike Timing Dependent Plasticity’ (STDP). This is the basis of the famous Hebb’s Law which is somewhat inaccurately stated as, “neurons that fire together, wire together”. As we have seen, it would be more accurate, but less poetic, to say “neurons that fire sequentially, wire together”.

Self-Directed Neuro-Plasticity

Now let’s focus on two specific neurons, the pre-synaptic “smoking neuron”, and right next to it, the post-synaptic “ideal-future-self” neuron. I do realize these are not actually individual neurons, rather they would be neural networks, but let’s stick with the simple case first! What do you see when the “smoking” neuron fires a spark of electricity to its “ideal-future-self” neighbor across the synapse? How does the “ideal future-self” neuron respond? Unfortunately, it doesn’t respond at all, because you haven’t yet run the swish pattern with your client, and the neurons are currently enemies! After all, if these neurons were wired together, the client wouldn’t be a smoker because every time she saw a cigarette she would imagine how she wanted to be a non-smoker.

We want to wire these two neurons together so that whenever your client sees their cigarettes, they imagine their ideal-future-self. Hebb’s Law (or rather STDP) says we first get the smoking neuron to fire (by asking our client to imagine seeing her pack or cigarettes), then immediately afterward get the ideal-future-self neuron to fire (by swishing into the picture of their ideal-future-self). Repeat this a few times, and voila the neurons become friends, and the ideal-future-self neuron will fire each and every time the smoking neuron fires. As Steve rightly says we are linking the two pictures (or in this description, linking the two neurons) and this can be done very quickly. Indeed it must be done quickly if STDP is to work. This is why we speed the swish up until it is done very quickly. Incidentally, this is also why the cue image (say the pack of cigarettes) always precedes the outcome image, because we have to fire the pre-synaptic neuron first for STDP.

So the swish helps the client to literally rewire their own brain, what Dr. Jeffrey Schwartz calls self-directed neuro plasticity (see Dr. Schwartz’ book ‘The Mind and the Brain’).

Now Steve draws an erroneous conclusion. He says that because we are swapping images (which is fast), or wiring neurons (which is even faster), therefore (and I quote) “a simultaneous transition will be much more dependable than the sequential slingshot swish”. Clearly the one statement doesn’t at all follow from the other, any more than saying “because the pre-synaptic neuron fires first and only afterward does the post-synaptic neuron fire, then the sequential transition is more dependable”. In NLP jargon this is a ‘cause-effect statement’ from the Milton Model, but it’s not good logic.

Sequential or Simultaneous, That Is the Question

So what’s an NLPer to do to decide between the two swishes? Fortunately we have a wealth of data on ‘sequential’ versus ‘simultaneous’ neural firing from the field of classical conditioning. Now the white-coated ‘behaviorists’ who get excited by ringing bells and salivating dogs, have their own jargon, which I will use here so readers can take a look at the underlying research if they wish. They call sequential firing “trace conditioning”, and call the type of ‘simultaneous’ firing Steve is referring to “delay conditioning”. To make this clearer, if Pavlov rang his bell and then fed his dog that would be trace conditioning, but if he rang the bell and then fed the dog while the bell was still ringing, it would be delay conditioning.

So what does the world of neuroscience have to say about trace versus delay conditioning? It turns out they say some pretty interesting things, and at least one VERY interesting thing.

So first and foremost, both trace and delay conditioning are effective. Outside of some specific contexts (that I will get to in a moment) they seem to be pretty equal. One context in which they are not equal, where delay conditioning (Steve’s lap joint) is better, is if there is a significant delay between the stimulus and response. For example if Pavlov rang the bell then waited for five minutes before feeding his dog, ‘delay conditioning’ would likely be more effective because the bell would still be ringing when the dog was fed. Obviously there is no significant delay between trigger picture and outcome picture in the swish, so this is not an issue (unless you are a fan of the slooowww-swiiisshh pattern).

Another context where ‘delay conditioning’ is also more effective is where there is significant damage to the subject’s pre-frontal cortex (PFC) or hippocampus. It appears that trace conditioning uses a different neural pathway, one that travels through the hippocampus and PFC, while delay conditioning requires only a cerebellum and brain stem. Therefore if you have a client who has had a full frontal lobotomy, or has serious brain damage to the PFC or hippocampus as a result of say a stroke, or if you’re a frog-whisperer and your client is an amphibian, you should definitely stick to the classical (lap joint) swish. Otherwise feel free to use either, confident that both are effective. Or use the one that butters your parsnips, or better yet the one that butters your client’s parsnips.

Actually this comes to an important principle of NLP, namely that, as NLPers, we do not use exclusive ‘or’s’. Meaning as the classical and slingshot swish actually use different neural pathways, you could try using both (either sequentially or simultaneously – just kidding!).

Neural Networks and Hebbian Engrams

Now let’s expand the discussion from individual neurons to the more accurate neural networks. Because neighboring neurons either become ‘friends’ (more-and-more likely to fire) or ‘enemies’ (less-and-less likely to fire), your brain becomes tribal, forming cliques of neurons that all tend to fire together and also stop other cliques from firing. This process is described by the famous psychologist Gordon Allport as follows:

If the inputs to a system cause the same pattern of activity to occur repeatedly, the set of active elements constituting that pattern will become increasingly strongly inter-associated. That is, each element will tend to turn on every other element and (with negative weights) to turn off the elements that do not form part of the pattern. To put it another way, the pattern as a whole will become 'auto-associated'. We may call a learned (auto-associated) pattern an engram. 

If that’s a bit of a mouth-full, think of the Jets and the Sharks from West Side Story, you’re either a Jet, or you’re a Shark (but not both), and when the Jets visit Doc’s Drug Store, the Sharks generally stay away, and vice versa. Revisiting our client’s neurons, the smoking neuron (let’s call it Tony) has a whole gang of other neurons, let’s call them the Jets, associated with it. These Jets will include memories of smoking, contexts where the client smokes, beliefs about smoking, the feelings associated with smoking, and so on. And the ideal-future-self neuron (which we will call Chino) has its own gang (the Sharks) of neurons relating to how that ideal-future self will look, feel, behave (OK Steve no behaviors) and so on.

When we use the swish to wire Tony and Chino together, we end up wiring all the ideal-future-self resources onto all the smoking contexts and triggers - we may have to swish several contexts to fully generalize of course. At the end of the swish, the Jets and the Sharks become friends, reconciled through Smoking-Tony’s demise, and the Maria neuron sings “One Hand-One Heart” as the credits roll.

Choosing the ‘Appropriate Resource’

I will comment briefly on Steve’s points about resource selection. I completely agree that the resource should be “appropriate” for the problem. However that doesn’t get us very far: what is appropriate and what is not? There is something in NLP called an ‘ecology check’, which seeks to ensure the change being made by the client will be ecological for all aspects of their life. For example if a client wants a high energy confidence state to do something silly, the coach has to deal with the lack of ecology as a first step.

In HNLP (Humanistic Neuro Linguistic Psychology) we say, “the conscious mind has no business choosing the resource-state”. This includes the hypnotist/coach’s conscious mind. HNLP presupposes it is up to the client’s unconscious to choose the best resource. This might be a high energy resource, or a lower energy ‘end-state’ type of resource. If a high energy resource state is chosen then would then tend to moderate into a lower energy end-state over time (as in the discussion of Bella – see previous post). The principal tool for achieving this is pacing and leading.

Steve seems to imply that high energy resource states are by their nature bad (“a thousand red flags flutter in my mind”). High energy states can indeed be risky, that’s why we still use the Greek word hubris for those who tempt the gods. However, I would hazard to say that for every person who has made a bad decision due to over-confidence (the number of Las Vegas gamblers shows there are many), there is a person who has failed to take action toward their goals and dreams due to low energy lack of confidence. Some who only rarely access high energy states often “live lives of quiet desperation, and go to the grave with their song still in them”. It cuts both ways, and many people admire James Dean over the man who retires from his desk job and spends the rest of his days playing bingo until expiring unremarked at the age of 90.

A great example of a higher energy state is Dr. Richard Bandler’s client session called ‘Shyness’. Dr. Bandler does a classical new strategy installation. If you have seen the video of the session, you will recall Dr. Bandler installs an initial step in the strategy where he asks the client to make a sort of primal yell. BTW this takes place inside the client’s head, not out-loud when introducing himself to a young lady! This could be interpreted as a high energy sense of confidence (and why not). There is a similar step Tony Robbins uses in his ‘stuttering’ demonstration (that Steve refers to in an earlier post), where Robbins leads the client to make a ‘warrior’ yell; again this generates a high-energy state in the client. What you may not realize when seeing these is that this high energy yell also takes back control of the voice from the word and language based left-brain, to the more emotional right brain (there’s always more going on!). In any case, Bandler does a follow up with his shyness client, who sits back in his chair relaxed (wearing badly fitting shorts if memory serves), talking as if meeting girls was never a problem for him. The high energy state transforms into an end state.

Amygdala Hijack

Finally, another piece of neuroscience. When talking of phobias, anger, fear and embarrassment, and similar states, these are moderated  by the so-called ‘amygdala hijack’. The optic nerve runs from the eyes to the visual cortex at the back of the brain, where it's processed into a meaningful representation of the world, which is transmitted forward to the PFC at the front of the brain for rationale decision making. This route is too slow to prevent me from being eaten by a saber-tooth tiger (“Oh is that a saber tooth tiger over there? Hmm that’s interesting, perhaps I should run…”; CHOMP!!!). Therefore, to save me from danger  my brain takes a feed from the optic nerve and runs it straight into the amygdala, before the information even reaches the visual cortex. The amygdala is responsible for fight-or-flight responses, and if my amygdala has any doubt about what it’s seeing, it sends a jolt of adrenaline into my body that I feel as a phobic response, or physical fear, or embarrassment (social fear), or anger. This is known as the amygdala hijack because it hijacks the rational mind, and is why I can feel afraid when I see a snake-like twig on the ground at night, and why these emotions can instantaneously interrupt everyday emotions.

The phobia cure, and similar techniques, target ‘amygdala hijack’. We calm the amygdala (“imagine you’re sitting in a comfortable theater, about to watch an old grainy black and white movie”), while gradually introducing the stimulus. This ultimately results in a rational ‘so what’ response, as the information follows the longer pathway through the visual cortex and PFC (for a visual phobia).   Notice that this depotentiation of the amygdala hijack follows the meta pattern: there is the phobic response; gradual exposure to the trigger in a dissociated manner; a calm response (the resource); and finally the ability to associate into the context (have the bee land on you) and retain the resource-calmness, i.e. the ‘collapse’. I totally agree with Steve that client’s don’t generally go from a phobic response to ‘confidence’, but again this doesn’t mean high energy confidence is never appropriate.

Conclusion

I am greatly enjoying this ongoing discussion with Steve, and I hope it provides some interesting jumping off points for readers own thoughts, discussions and experiments. Due to modern imaging techniques such as fMRI, neuroscience is making leaps and bounds in our understanding of how the brain works. I would strongly encourage anyone interested in NLP, how the brain and mind work, or the human condition in general to become familiar with the principles of neuroscience. These will aid you in choosing a solid direction for your own experimentation with clients.

The tai chi master was delighted after three years of training that his student was able to replicate the tai chi  forms he had been taught. He told the student to go away and practice by himself for a further three years and then return. After three years the student returned, crestfallen and confessed that even while practicing diligently, he had lost 30% of what his teacher had taught. The teacher was disappointed and sent the student away to practice for another three years. Three years later the student returned and said he had lost 60% of his learnings. Once more the teacher sent him away to practice for three more years. On his return the student told the teacher he had now lost 100% of his learnings, at which point the teacher congratulated him as he had made the art his own.

-          - Ancient Chinese story