Confessions of a Working Psychotherapist

In my previous posts, I explored the apparent split between thoughts and feelings.  It seems that we can have feelings without conscious thought, even without the “automatic thoughts” that Beck claimed to precede emotions.  Our ability to change feelings through conscious thought is quite limited.

I’ve suggested that two different types of processes are in operation.  One is logical thought based on a type of verbal processing.  The other is a feeling process based in pattern recognition.  The latter may be based upon neural networks that are similar to those used for pattern recognition elsewhere in the brain.  In contrast to verbal thought, the pattern recognition process doesn’t provide us with a direct method to examine why we feel a certain way in reaction to a situation  That information is likely to be contained in a tangle of excitatory and inhibitory connections between neurons.  When we try to explain our feelings, it is more of an examination of the outputs of that process than a direct, conscious knowledge of its process.  Those outputs probably include Beck’s automatic thoughts.

How we feel about things sometimes has little connection to what we think of them — and this accounts for a large portion of the misery in this world.  Over the past few weeks, I’ve been thinking about this.  In particular, I’ve been concerned with developing testable hypotheses and the clinical tools.

I’ve decided to leave the question of experimental proof alone.  My reading in contemporary neuroscience has convinced me that within twenty years, we will probably have a very accurate idea of how emotions and thinking interact.  This will happen mainly through examinations of brain activity using functional MRI, PET scans, and devices that measure electromagnetic activity in small groups of brain cells directly.  The mechanism of interaction between thought and feeling has become the realm of technology-driven science.

Even though this model may not be right, it may be good enough for right now, especially if it accounts for clinical phenomena and leads us to new treatment methods.  If feelings are produced through a neural network pattern recognition process, we can predict that many of the same phenomena observed in neural networks for pattern recognition would also be seen in emotional responses.  These would include: (1) the strengthening of responses through repeated triggering with positive feedback; (2) adaptation to a constant stimulus without feedback; and (3) mutual inhibition of competing networks.

We actually see examples of (1) and (2) frequently in clinical practice.  OCD gives us a good example of feelings being strengthened by feedback.  Performing a compulsive act temporarily reduces anxiety, which increases the strength of the compulsion.  OCD is effectively treated using exposure with response prevention, which gives us a good example of adaptation of feelings in the absence of feedback.

So let’s examine the last of the three: competition/mutual inhibition.  This needs a bit more explanation than the other two.  We can start with a visual system example that you may have seen.  In a number of classic visual gestalt illusions, enough information is present to construct two very different pictures from the same image, eg two faces versus a vase, or a death’s head versus a lady at her dressing table.  Here are a few examples:

 

One way of seeing the image initially dominates our view, until, through some indescribable process, we suddenly ‘get’ the other picture.  We have difficulty seeing both simultaneously, though we know both are there.  It’s as if one interpretation of the image competes with and inhibits the other interpretation.

Something similar happens when therapists reframe patients’ descriptions of situations.  The patient comes in with a description of an event that includes a great deal of negative explanation, interpretation and/or prediction.  By changing the way the elements of the situation are used to construct its meaning, the therapist provides a benign or positive alternative interpretation.  Hopefully, this transforms the emotional response to the situation into something less troubling.

Unfortunately, it can be a daunting process. Therapists often say that patients resist these reframes, meaning that they argue about them.  Often, despite accepting the reframe as valid from a rational point of view, patients insist that their original interpretation was correct from a feeling point of view.  They may also feel that the therapist is invalidating their feelings and world view.  (Which is true: their feelings are the problem.)

A competition/inhibition framework is useful in understanding the situation.  Reframing works at the level of our conscious constructs of situations and events, our verbalized logical responses.  But our feelings operate by way of a different process.  When a reframe works, patients feelings shift to the new construction.  When it doesn’t work, it’s like trying to see the alternative image in an illusion.  The more you try, the more frustrating it becomes.

Beck’s cognitive restructuring has similar problems.  Patients use rational responses to counter negative automatic thoughts, which Beck believes are the source of troubling emotions.  Yet patient often discover that they can believe in their rational responses while still feeling that their painful automatic thoughts are also true.

What we need is a method to work on feelings more directly.  I believe we can do this using the competition / inhibition effect.  If we want to change patients’ painful automatic reactions to situations, the best way may be to create competing reactions that generate positive feelings.  We can hope that those positive feelings will inhibit the painful feelings that caused them to seek therapy.

Over the past three months, I have been working on a method for treating patients with core beliefs/schemas that include those Young & Klosko labelled as Unrelenting Standards, Punitiveness, Defectiveness, Failure, and Negativity.  These patients, however cooperative, are usually treatment resistant, and they have been through prior therapies with limited results.

My method of treatment has been to identify or construct verbal statements that sum up these patients’ underlying negative feelings about themselves.  These are usually harshly judgmental and global in scope, eg “I’m a complete f___ up.  Anything I touch, I ruin.”  Such statements account for patients’ feelings in a wide variety of situations.  They correspond to Beck’s core beliefs or Young & Klosko’s schemas.  My argument, however, is that these statements are verbalizations that represent a pattern recognition process that is non-verbal.  They are reports about the results of that process, rather than the process itself.  Confusion occurs, in part, because talking about feelings often triggers production of the feelings by way of the same non-verbal process that is triggered by events, memories, and images.  If you’re a veteran with PTSD, talking about the traumatic situation triggers the same process as seeing a movie, remembering the battle, etc.

Rather than refuting these verbalized core beliefs/schemas, I work with patients to find statements that elicit positive feelings, that are arguably true, and which compete with their core beliefs/schemas at the feeling level.

These statements cannot be simple opposites of the statements describing core beliefs, since patients can usually see the falseness of positive global statements.  They are often rejected as ridiculous.  If they generate positive feelings, they are still dismissed as oddities.  If you tried the exercise in part one of this series, that was probably your reaction.  [*** insert link ***] A patient can accept: “I f___ up everything I touch or do,” but they reject as patently false: “I’m always right, and I never make mistakes.”  Likewise, statements that are accurate but too limited in scope fail to trigger positive feelings, “I don’t make too many mistakes.”  Statements which don’t trigger positive feelings are just worthless, however true they may be, eg: “Everybody makes mistakes.”

In contrast, consider:  ”I work hard to check my work, and I hardly ever make mistakes.”  It should elicit positive feelings, while also passing the test for truth.  Patients can even accept it as true in the face of  counter-examples.

Creating these statements has to be a collaborative, creative process, in which patient and therapist work together to find statements which are objectively true, which seem true to the patient, which generate strong positive feelings, and which are sufficiently related to the core belief to compete.  I ask patients to say these out loud and to report on the feelings elicited.

I am currently developing techniques to use these statements effectively and efficiently.  Patients report positive results when they use them in a manner similar to a three-column Daily Record of Dysfunctional Thoughts.  They also object to the effort required.  Homework compliance is no better than in traditional Cognitive Therapy.  In the coming weeks, I plan on developing a protocol for linking visual images to positive, competing statements, and the use of both the statements and their linked images in a form of therapeutic meditation.

Please feel free to share your feedback.

So what’s going on here?  Let’s start with what we know.

• We routinely experience feelings, both positive and negative, which have no basis in rational thought.
• We can alter our feelings with statements that have little or no basis in rational thought.
• We respond emotionally, sometimes against our better judgment, to a wide variety of stimuli, eg: music in a minor key, the sight of a child crying, manipulative television advertisements, and political propaganda.
• We can provide verbal explanations for many of our feelings, but these are frequently incomplete or unlikely.  They seem to be after the fact explanations of what must be , rather than a real time explanation of what is.
• We have our own introspective experience of the split between thought and feeling, which seems to correspond to widely reported descriptions of others.

I think the most parsimonious explanation is that we have two mental processes running simultaneously.  One process is the conscious, linguistically based, manipulation of symbols: logical thought.  The other may best be described as a pattern recognition process that is outside of conscious awareness or control, that provides no explanation of its inner workings, and that provides only a summary output of its reactions to the world.  It generates feelings.

There’s an old, mildly humorous, adage: “If it walks like a duck, and if it talks like a duck, it’s probably a duck.”  The humor comes from the difference between how we think about things versus how we recognize objects in the world.  Standing at the edge of a pond, we don’t normally go through a checklist of visual and auditory features of the wildlife around us.  Our perceptions are automatically put through a pattern recognition process which abstracts information from the environment.  Normally we don’t ‘see’ feathered, billed, waddling, web-footed, quacking things.  We ‘see’ ducks.  As long as the objects in our visual field are recognizable, this process is immediate and itself imperceptible.

An interesting aspect of this process is that it discards much of our raw perceptions.  Unless we force ourselves to take note of the properties of objects, we usually discard that information, keeping only the awareness of the abstracted information: the ducks.  This only breaks down when we see something we don’t recognize, let’s say a platypus.  Is it a duck?  We then use a conscious, logical process to review its characteristics.  This happens a few times until we learn to automatically recognize a platypus.  Once we’ve built up an internal platypus detector, the component information is routinely discarded.

Something similar to duck/platypus recognition happens frequently in psychotherapy.  An inherently ambiguous situation triggers an interpretation, leading to positive or negative feelings.  The process is automatic and nearly immediate.  Unless we consciously review the elements of the memory for alternative explanations, those elements are discarded from our thinking.  Our first, automatic, interpretation appears to be an objective truth about whatever occurred.  It’s difficult to give up that interpretation, and many people react defensively, insisting that their interpretation is the only sensible one.  It is only after considerable self-examination that we can accept the possibility of other ‘truths’.

Imagine a wife who complains that her husband yells at her all the time, causing emotional distress.  Her interpretation is that he is bossy, and that his yelling is either angry or bullying.  Hypothetically, she may have a history of having been yelled at by other people: a yelling father, boss, or ex-husband.  These experiences may have resulted in her development of a ‘yelling detector’ that is triggered by her experience.  It is only after examining the situation and determining that the yelling occurred in a noisy environment and that her husband suffers from hearing loss, that she can learn to interpret the situation as something else.  Until this reconsideration of evidence and reinterpretation occurs, her life filled with angry and bullying yelling is her truth.

Much of the work of the brain occurs via neural networks that categorize and abstract data.  Our knowledge of the workings of the complex and hyper-complex cells of the visual cortex gives the best example.  Through a succession of data abstraction networks we recognize brightness differences, lines of connected differences, and shapes of connected lines.  This is done through groups of nerve cells organized in layers, each with weighted excitatory and inhibitory connections to the nerve cells in the next layer.

These types of neural networks has been studied extensively.  They are the basis of much of our current computer technology for voice recognition and character recognition.  One interesting feature of pattern detection neural networks, is that they do not use templates for comparison.  Everything is accomplished through connections that either stimulate or inhibit other cells.  We don’t compare the perceived image of a duck to an idealized image we have stored in our heads.  We either see a duck or not.   The internal process cannot be reported in any sensible way.

If in fact our feelings come from a more elaborate pattern recognition system based on the same rules of organization, this would account for many of the phenomena that we’ve noted.  The process would not be part of an unconscious, so much as not having the stuff necessary to be part of conscious thought.  It is different and apart, rather than repressed.  We can examine the inputs to the network and its outputs, but the connections and their positive or negative weights cannot be symbolized or explored through introspection.

Much of what we’ve learned to feel about our world has probably developed through the creation of pattern recognition neural networks.  If so, this learning has occurred not in terms of a series of rules applied to situations, but through a process of weighted positive and negative connections between nerve cells.

When our feelings about events in our lives change, it is uncertain whether we have changed the pattern of connections and connection weights in existing pattern recognition networks or whether we just create new networks and inhibit those which existed earlier.  If we have more than one network activated by the same inputs, with differing output results, the result would be … mixed feelings.

This discussion of feeling-generating pattern-recognition neural networks is completely speculative.  It accounts for many phenomena, it is based on known patterns of operation of the brain, and it is relatively parsimonious compared to other explanations.  However, in order to be properly hypothetical it has to generate hypotheses that can be tested.    I’ll discuss that and the implications for clinical practice in the next section.

Copyright © Etan Ben-Ami, LCSW 2011  All Rights Reserved

In Part I, I demonstrated the common phenomenon of feeling or believing in expectations, demands, or standards which we know to be unreasonable from an objective, logical perspective.  These feelings and beliefs are usually harsh and absolute, and the more we think about them, the less justified and more irrational they seem.  At the same time, they hold great emotional power either to vex us, or more rarely, to fill us with pride and good feeling (hopefully as demonstrated).

I’ve spent years thinking about this phenomenon and examining it in a clinical context.

In clinical practice, it usually comes up when a patient says something like this: “I’m not an idiot, and I’m not crazy.  I know it’s wrong to [fill in the  blank], but I can’t help feeling that way.”  The subject can be anything from expectations of perfection, sensitivity about appearance, anger at an old insult or injury, the loss of love, difficulty accepting aging — nearly anything that you can imagine produces strong, unpleasant emotions.

What gives?  Why this split?

How can our thinking be so different from our feelings?

Nearly every school of psychotherapy has confronted this problem on some level.  Psychoanalytic thinking has offered several explanations, even in Freud’s lifetime, and more since.  I can’t do justice to them all, since I’m not a student of analysis, and this is just a blog entry.  Most seem to involve conflicts between conscious and unconscious elements of personality and thought, usually based in Freud’s structural model of id, ego, and superego.  Most analysts would probably identify the problem of anxiety and guilt over the inability to achieve unrealistic expectations as a problem of a punitive superego.  A further explanation would refer to distorted internalized images of parents and their demands.  (Admittedly, this is an over-simplification.)

I’ve never found the psychoanalytic explanation satisfying.  Better minds than my own have argued for and against it, so I’ll just raise a few problems pertinent to this issue.  It seems based on an undifferentiated brain that hosts a mind with invisible virtual structures.  Yet its function changes in response to psychoactive drugs, disease, and physical damage.  Why are some pieces conscious and some unconscious, and can we really make the unconscious conscious?  Is the awareness gained from a completed analysis real or just a believable explanation?  Why can I accept an intellectual argument against the feelings/beliefs generated by my punitive superego, yet continue to feel the full weight of guilt or anxiety.  Ouch!

Aaron Beck, the creator of Cognitive Therapy, dealt with the issue very differently.  He avoided the explanations of the structural model, instead hypothesizing a structure of thought and belief.  We all have core beliefs, mostly created in childhood and with understandable distortions.  These are simple, general, and absolute.  When applied to common situations in our lives they generate intermediate level beliefs.  These in turn generate automatic thoughts, frequently judgmental or punitive, which in turn generate our emotions.  Feelings don’t exist as an independent entity.  They’re a combination of automatic thought and elicited emotion, with thought always coming first.  Changing beliefs is a matter of confronting automatic thoughts, intermediate, and core beliefs with rationality.

I’ve wanted to believe in the cognitive model.  I’ve even based treatment on it.  I’m not so sure anymore.  I know that I do not experience my feelings as automatic thoughts or my emotions as preceded by automatic thoughts.  When I observe myself through close and deliberate introspection, it seems like automatic thoughts are produced by feelings, and that the emotions come first.  I know that my most troubling negative feelings are completely immune to my attempts to modify them via my rational process: Beck’s cognitive restructuring.  God knows, I’ve tried to apply to myself what I’ve taught to others.  It doesn’t work nearly as well as I’d like.

Jeffrey Young and Janet Klosko’s Schema Therapy offers an extension of Beck’s model of core beliefs.  Like Beck, they believe that as children we develop many mistaken beliefs about ourselves and our world.  Those ideas that don’t have much emotional force are easily corrected: eg, few grade school children have difficulty accepting a model of the solar system.

Erroneous ideas that have emotional weight are treated differently.  We combine the memory of our confirming experiences, our emotional reactions to those memories, and the resulting beliefs, predictions, and habitual responses to situations that trigger them into something termed a schema.  We usually refuse to consider contrary evidence or alternative explanations.  They become encysted in consciousness.  Also, they usually function outside of our awareness until triggered by events.  You can think of them as functions and subroutines in a computer program.  They execute and then they’re dormant again, making them unavailable for debugging except via their output while active.

You can also think of them in terms of the process an experienced driver goes through in making a left turn in traffic.  A new driver considers every factor carefully and consciously.  An experienced driver performs the turn almost automatically (including swearing at traffic) and may have difficulty explaining just what he or she did.

I like the schema model, and I use it in my daily work.  Still, it doesn’t explain the phenomena adequately.  It doesn’t tell me why my thoughts and feelings don’t match up.  It doesn’t tell me why I can’t just re-educate my patients’ more noxious schemas and tell them to cut it out.  It feels as if we have two brains.

Part III will have my model.

Copyright © Etan Ben-Ami, LCSW 2011  All Rights Reserved

Do you ever find yourself thinking one thing and feeling something completely different is true?  Are the feelings usually troubling?  Could you turn them into statements, perhaps centered around the words “should” and “must”?  Perhaps they are harsh statements about your inability to live up to those “shoulds” and “musts”?   Here are some examples: “I must not make mistakes.  I should be kind to everyone.  I don’t deserve to be loved.”  Does any of this sound (agonizingly) familiar?

Well congratulations, you’re human!  To the best of my knowledge, no dog, cat, cow, goat or other non-hominid has ever complained about this kind of thing.  The jury is out on signing chimps & gorillas and various large marine mammals who appear to talk to each other.  I believe it is a universal human phenomenon based in our development of language and logic.

But before we get into that, let’s try a small demonstration of this split.  Take a piece of paper and write down a few of your stronger, more disturbing, more generally demanding “musts” and “shoulds”.   Some people will have many of these, and find them more powerful and painful than others.   Even if you don’t have many, you probably have some.

Try to avoid reasonable, nuanced, statements delimited by practical considerations, eg: “I should pay my taxes by April 15 or the IRS will fine me.”  Better would be: “I must not be late.”  It’s more general, absolute, and probably more emotionally powerful, if that’s your issue.  Try for a list of statements that are somewhat painful to read.

Now take your list and re-write every demand as a sort of positive statement about yourself, however untrue.  ”I must not make mistakes,” becomes: “I never make mistakes.”  ”I should never do anything that would make people angry,” becomes: “No matter what I do, people are never angry at me.”  You may find these statements boastful or ridiculous.  That’s okay.  They’re clearly impossible and irrational.  Your conscious, objective, logical, adult mind should look at them that way.

We’re almost done.  Just read all your positive statements out loud, if you’re in private, or silently, but with feeling if you’re in a public place.  Read them as if you sincerely believe them to be absolute truths about you.

Are we having fun, yet?  Doesn’t it feel good to say those things?  The fact that they’re ridiculous doesn’t matter.  They make you feel good, at least temporarily.  In general, the more painful the original demand statement, the better you feel when you read its impossible, boastful opposite.  But these things are ridiculous, crazy, impossible.  Pretty cool effect, huh?

You can shred the list now.

(Coming soon, Part II: Explanations.)

Copyright © Etan Ben-Ami, LCSW 2011  All Rights Reserved

Here is another great little article on cognitive errors that we see everyday.  If you feel you have all of them, well … at one time or another, we’ve all made these mistakes.  The benefit of understanding these cognitive errors is that we can begin to catch ourselves and correct them to some degree.  Ultimately, we have to accept that these errors are nearly impossible to completely avoid.