LeDoux's main interest is the study from a neurological point of view of fear. One of the interesting comments he makes is that we shouldn't necessarily assume that there is one single emotional system in the brain. (The ``limbic system,'' whose existence is currently in considerable doubt; p. 101.) It is quite possible that different emotions are handled by very different systems (p. 103).

I was especially interested in what LeDoux has to say because of the way it relates to phobias and its relevance to NLP (NeuroLinguistic Programming), and I have some comments to make. However regular readers of the newsgroup will be aware that I am no psychologist, and even less am I any sort of expert on neurology. Any expert-sounding knowledge below is taken directly from LeDoux's book.

The NLP phobia cure does not have to do with any particular hypotheses of what phobias are or how they arise, but in some of the NLP trainings I went through, it was stated that phobias are simply examples of classical conditioning (an ``anchored response,'' in NLP terminology). Richard Bandler has often claimed to be able to easily create phobias in subjects, and part of the NLP lore is that in the early days, Bandler and other NLPers would sometimes give phobias to unsuspecting subjects in shopping malls. Presumably Bandler was not concerned with the ethical problem involved since he was supremely confident of his own ability to cure phobias. In any case, in more recent days, to the best of my knowledge, he restricts himself to giving phobias to volunteers at his seminars.

I don't know whether the responses Bandler created were true phobias or just a conditioned response that seemed something like a phobia, and I don't know what his technique is. (It was never something that seemed to me of any great clinical value, although perhaps one could give someone who wanted to stop smoking a phobia of cigarettes.) Presumably, Bandler's approach was something like that used by John Watson in the case of Little Albert. (``John Watson claimed to have conditioned an animal phobia in an eleven-month old boy, Little Albert, by making a loud clanging sound while the boy was happily playing with a rat.'' I doubt that Watson had any greater informed consent from Little Albert than Bandler had from the Santa Cruz shoppers.)

The explanation of a phobia as simply a conditioned response never quite made sense to me. In the first place, this explanation doesn't account for the extreme nature of phobic fear, especially in relationship to the often mild nature of the stimulus for the phobia. (Of course the overt stimulus for the phobia is merely the CS, and one may hypothesize that the original US was quite terrifying. Still, this idea always seemed doubtful to me.) Although people say things like, ``I have a mild phobia of ...'' or ``I have a fairly strong phobia of ...'' and ordinary fear can, at the extreme, reach the point of absolute terror, it has never seemed to me that there's a continuum of which mild apprehension forms one end and phobic fear that other. As I've mentioned, I'm no expert on psychology, but it has always seemed to me that there's a difference of kind as well as degree between ordinary fear and phobic fear.

Also, people with phobias very seldom have any conscious memory of some traumatic experience that gave rise to the phobia. This is a point that LeDoux brings up himself. He says that people don't remember the experience that created a phobia because extreme fear interferes with memory formation and retention (pp. 239-244). Maybe so, but friends and relatives of the subjects don't remember the experience originally causing the fright either, and yet one would expect such experiences, presumably extremely intense, to be quite notable. Furthermore, people who have been through quite terrifying experiences don't usually develop phobias as a result. A woman who has been raped, for instance, often finds the locale of the event extremely unpleasant afterwards, and may feel extremely uneasy when subsequently going to that neighborhood, but as far as I know there is rarely an actual phobia induced.

In any case, the information in LeDoux's book suggests to me a hypothesis that would make phobic fear differ in kind as well as degree from ordinary fear.

Fear, according to LeDoux, seems to be generated in the portion of the brain called the amygdala. When fear is a response to an external stimulus, the sensory information from the stimulus gets communicated to the amygdala; only at that point does fear arise.

Now the interesting thing here is that there are two different pathways to the amygdala (pp. 163-165). The first, the thalmic pathway, is crude but fast. One is walking through the woods, hears a strange noise, and a signal is sent from the thalamus to the amygdala, which causes one to suddenly become very alert and, in many cases, to freeze. The thalamus is not able to make very fine sensory discriminations, but it's able to alert the amygdala that something has been sensed which one might possibly want to be afraid of.

But the thalamus will also send a signal to the neocortex, which is where high level processing of sensory information is done. The neocortex will then send a second signal to the amygdala, saying, in one case, ``It's okay, that was only a twig snapping,'' or in the other, ``That sounded a whole lot like a rattle snake; better take some defensive measures.'' (There is also an involvement of the hippocampus in responses where the cue is contextual, but not in responses to a specific sensory stimulus; pp. 167-168).

Now my hypothesis (and to some extent LeDoux suggests this rather tentatively himself; pp. 254-255) is that however it may be that phobias get created, the structure of a phobia is that this thalamus-neocortex-amygdala connection is not made. This definitely agrees with the memory of the experiences I used to have when I had a fear of heights. There was an immediate, extremely strong fear, and it was extremely difficult and sometimes impossible to get myself to do a neocortical check: ``Is this situation really unsafe?'' (Normally the neocortical check occurs in a fraction of a second; p. 163).

Now so far, my explanation seems highly speculative. But I think that we see more evidence for it if we look at the process by which phobias are extinguished.

There are two common behavioral cures for phobias: implosion (or flooding, I don't quite know the distinction between these terms) and progressive desensitization. My own thoughts and what's in LeDoux's book relate only to desensitization.

The process is a well-known one in behavioral psychology and is called extinction. Seligman, for one, studied a situation in which rats had been conditioned to associate the sound of a buzzer with an electric shock, and therefore to jump over a barrier in their cage when the buzzer sounded. ``Especially important to Seligman was the fact that avoidance conditioning vanishes quickly if the animal is prevented from making the avoidance response and alternative solutions for escape or avoidance are not provided.... [If the barrier is replaced by a wall] preventing the avoidance response, the rat soon learns that the buzzer is no longer followed by a shock and begins to ignore the buzzer.''

In the case of progressive desensitization, the subject is first taught to systematically relax, and to maintain the state of relaxation when the phobic stimulus is presented. Relaxation and fear are not compatible states, so, paraphrasing the quote above, we can say that, ``The phobic response is prevented when the phobic stimulus is presented, so the subject soon learns to maintain a neutral feeling in the presence of the phobic stimulus.'' It makes sense in terms of Seligman's experiments, and furthermore it's just good common sense.

What happens in terms of the brain might seem quite apparent. Since the link from the thalamus to the amygdala corresponding to the phobic is now no longer useful, it quickly atrophies, and the subject no longer has a phobia. (Or, more likely, what happens is that the amygdala ``forgets'' that this particular stimulus from the thalamus corresponds to fear.)

If this is one's concept of extinction, then one can see why psychologists would consider it ``absurd'' that there could be a form of desensitization which can work in five or ten minutes, as NLP practitioners claim.

(But here, I have to say, there's a big difference between psychologists and researchers in the natural sciences. Namely, what does one do when confronted with the possibility of a phenomenon that may be incompatible with one's theory? In psychology, or at least certain areas, the response seems to be that since the phenomenon is clearly impossible, there's no reason to waste any time investigating it. In the natural sciences, the attitude would be that it's of urgent importance to investigate this phenomenon. It provides a way of putting theory to an even more demanding test, and may suggest modifications in the theory.)

But in fact, as LeDoux explains, the explanation proposed above is not what happens in cases of extinction. The response by the amygdala to a particular kind of information relayed from the thalamus, once learned, does not decay and apparently survives for the life of the subject (pp. 150-252). What happens in the case of extinction is apparently that the neocortex learns to send, via the medial prefrontal cortex, a signal to the amygdala saying essentially, ``Chill out, everything's okay.'' (``When Maria Morgan made lesions of [the medial prefrontal corex], rats continued to act fearfully in the presence of a conditioned fear stimulus long after rats without lesions had stopped acting afraid,'' p. 248).

Now if what happens in extinction is that the brain is learning to send an ``all clear'' signal from the visual cortex (since most phobias are in response to visual stimuli) to the amygdala, it becomes at least a little more plausible that this could be accomplished by a five or ten-minute visualization.

In the NLP phobia cure, instead of trying to teach the subject relaxation, one asks him to go through a rather complicated visualization whose purpose is to have the subject dissociate, that is to direct their attention away from their own body. One asks the subject to imagine being in a movie theatre, and then float up out of their body (I like to ask them to float up into the projection booth, so they can be in control of the imagined experience) and watch themselves watching a poor quality black and white film.

It's always seemed to me that there are two possible explanations for why this is useful. The first is that in having the subject imagine being out of their body, one is keeping them in a state where they will be unable to feel fear. (If a subject said, ``I felt a little afraid,'' I would ask, ``Did you keep yourself floating in the theatre or did you let yourself 'fall into' the movie?'')

The second possible explanation, though, is that it's just a process of distraction. One is simply giving the subject a complicated task to do, one which is likely to hold the subject's attention because it seems superficially related to the phobia. In this way, it would be like asking the subject to sing a complicated song while confronting the phobic stimulus. (In the Bay Area, it is well known that many people find driving across the Bay Bridge a frightful experience, and some people cope by singing at the top of their lungs for the length of the bridge.)

If it is true, however, that what one is really doing is teaching the brain to send an ``all clear'' signal from the visual neocortex to the amygdala, then it is conceivable that the really important aspect of the process is the fact that it is a visualization. One is simply teaching the subject to direct attention to the visual cortex when the phobic stimulus is experienced. Perhaps some other complicated visualization would serve the same purpose.

Bandler has always maintained that in the phobia cure it's absolutely essential that the movie the subject watches not be focused on the phobic stimulus, but instead show the subject confronting the stimulus. Whether this is in fact true or not is something I don't know. I have certainly never experimented with it. In terms of NLP theory (which unfortunately has never really been written down systematically anywhere), this makes a certain sense. In terms of the functioning of the brain, it's a possible topic for investigation.

In NLP, at least, asking subjects to visualize themselves is an important component of many techniques. It seems as if visualizing oneself activates some important process in the brain.

The desensitization process described above used to be the totality of the NLP phobia cure, which in those days was called Visual-Kinesthetic Dissociation. But about fifteen or twenty years ago, a second step was added. In this step, the subject ``steps into'' the movie at the end (after the phobic encounter is over), now seeing everything in full three-dimensional color, through his own eyes as a participant in the experience, and then runs the whole experience backwards to the beginning like a VCR backspacing very very fast. This second step is repeated five or ten times.

Aside from the fact that this second step is also a complicated visualization (much more difficult for most subjects than the first), I don't know how to explain the fact that it apparently increases the power of the technique.

Finally, NLPers have always claimed that the phobia cure is equally effective as a way of neutralizing traumatic memories. To me, what's in LeDoux's book certainly seems to confirm that the same process would work for neutralizing phobias and neutralizing emotional memory. The main difference is that in the case of phobias, the stimulus is external and in the case of memories it's internally generated.

--
Trying to understand learning by studying schooling is rather like trying to understand sexuality by studying bordellos.       -- Mary Catherine Bateson, Peripheral Visions