Professional Boundaries, Scientific Models, and Hemispheric Specialization

by Siegfried Othmer | December 15th, 2005

Sometimes an actual case history does more to establish a principle than mere enunciation. My mother-in-law is 94 years old, and is doing neurofeedback every day to maintain her level of function. Family members are pitching in as they can to keep her in good spirits. Among the family members is a son-in-law who is also a Harvard-trained psychiatrist. Recently it transpired in casual conversation that he had caused some medication to be prescribed for one reason or another. Since we are doing neurofeedback and have to judge any changes that occur with respect to neurofeedback strategies, it would have been nice to know that a medication was being prescribed.

An old-line doc, on the other hand, feels that this kind of information should not be shared with anyone. Besides, what can neurofeedback possibly matter to anything? We have no standing to know about the prescription, and no reason to know it. This, incidentally, is from a person who has known about neurofeedback since we first undertook the training with our son Brian in 1985. He was a witness to Brian’s progress. He was also aware that my father-in-law undertook neurofeedback for years in the late eighties for his dementia and Parkinson’s. Every time the man got away from the neurofeedback instrument for a few weeks, his wife would notice his decline in mental function. That function would then be nicely recovered once he was back on home on the instrument.

This doc was originally taught in the tradition of psychoanalysis and then made the transition to biological psychiatry without a hiccup. Back in the sixties, he could dress someone down in full paragraphs peppered with Freudian epithets without pausing for breath. By now, the vocabulary has changed. Throughout, one thing has remained unaltered: It is difficult for him to remain civil when the discussion is about neurofeedback. One imagines a process similar to what might have gone on with Russell Barkley. Hearing about neurofeedback early on, it was easy to dismiss. And now it is difficult to shift ground. There is also the wrinkle that it is difficult to be a hero in one’s own family.

So what is going to happen when it dawns on psychiatrists such as my brother-in-law that neurofeedback is actually much more comprehensive, and much more nuanced, than psychopharmacology alone? Without a doubt a mentality such as his would wish to draw the technique behind the curtain of medical authority and to become the official administrator of the neurofeedback “medicine.” After all, it will have to be “balanced off” against the medication, so it should all be done by one authority.

But one senses the world changing underfoot with respect to the acquiescence of the public to the traditional medical model. Indeed the elderly are still faithfully buying in, taking all their medications as directed, and often slipping nonetheless into dysfunction. Such functional deterioration can always be dismissed as the inevitable consequence of mere aging. One is not likely to be seen failing at anti-aging medicine since at some point failure is expected. A slightly younger generation, on the other hand, is taking command of their own lives through education in self-care through dietary supplements, exercise, and other means. Parents of difficult children are gradually becoming acquainted with a more natural medicine, one that is concerned with proper functioning of the body than with targeting some “disease vector” or with re-shaping mental functioning with the usual drugs.

The major problems of disregulation are coming to be seen as multi-faceted, requiring remedies from many different medical and other competences. Does anyone remember the fuss about “pygnogenol” being a remedy for ADHD? In the paradigm of the day, it had to be a categorical remedy or none at all. Its impact had to be “statistically significant” across a representative population of ADHD kids or it was not to be taken seriously. Pygnogenol is an antioxidant. Does efficacy for some ADHD kids not simply hint at the possibility that antioxidants should be considered as one of many remedies that might be invoked for better physiological functioning of a compromised system?

Or in the case of the autistic spectrum, one can look at the case of “secretin.” For a while, it was ballyhooed as a potential remedy for autism, and then it was just as assuredly whittled down to size by the usual bulldozer of statistical argument, just as in the case of pygnogenol. But some kids clearly benefited significantly from the remedy. Should that not also fit some kind of model? Instead a kind of statistical fundamentalism erased the data. Secretin did not meet threshold.

It is quite inevitable that the success in understanding the autism spectrum as a multi-faceted disorder will lead to a more general reframing of chronic conditions of disregulation in similar terms. Anti-aging medicine is another case in point–it does not have a narrow target. Chronic pain management is another. ADHD, our bread-and-butter preoccupation with neurofeedback, will follow autism into being modeled as a much more complex disorder than it has been heretofore. Just to see how far along these lines we have already come, consider a vignette from the past:

At a time when my own encounter with the neurofeedback virus was still in its virulently infectious phase (meaning that nearly everyone I met ended up hearing about it), I talked to my brother-in-law about some of our early successes with Tourette Syndrome. He dismissed the discussion with a wave of his hand: “Tourette’s is not a problem. You just give Haldol.” It is now some twenty years later, and even he would agree that matters are much more complicated than that. Tourette Syndrome is now also considered a spectrum disorder. David Comings has been promoting that view for nearly that same length of time, but his early views as an outlier have since become mainstream. At the outset of his Tourette Syndrome research, even he was looking for the “Tourette’s gene” (singular). Through his research and that of others, that has become quite a palette of genes. Matters are much more complex than he allowed for, even given the spectrum concept that he was promoting.

Consider the “natural medicine” and “natural pharmacy” approach that has emerged for TS, as discussed in the new volume, Tics and Tourette’s, by Sheila Rogers. Medication management here plays only a very minor role. Instead a variety of tic triggers are identified that point the way in various directions for remedy. Many of the issues are in fact common to those we engage with autism. In future, the same will be true of ADHD and of childhood Bipolar Disorder. Ultimately, these conditions may even come to be seen as substantially iatrogenic. Certainly in the flurry of cases that suddenly made their appearance with bipolar symptoms, the strong suspicion arises that much of this incidence was the direct result of pharmacologically induced disregulation. These children could not tolerate stimulants and anti-depressants, and their nervous systems were further disregulated thereby. But even when it comes to ADHD, one can argue that in the larger sense the exclusive pharmacological focus of the medical model displaced the more wholesome remedies that should have been looked to all along. We have had iatrogenic ADHD by virtue of medical neglect of the spectrum aspect of ADHD.

A Systems Perspective

It is only a matter of time until a systems perspective will take hold, and eventually even find formal expression. Such a systems perspective will be the natural framework for an integrating, boundary-breaking technique like neurofeedback. The natural course of events will be for neurofeedback to be accepted as part and parcel of a congenial theoretical framework, and that of necessity implies a systems orientation.

Such a systems perspective will eclipse the current formulation of mental dysfunction in terms of distinct disorders with discrete boundaries. What will ultimately sustain the old model beyond its dotage is the fact that health care delivery is organized at all levels around that model. The system will continue to act as if the model were valid, just as it continues to spill words about the “chemical deficiency” model of psychopharmacology, despite the fact that it is not graced with any formal support.

All of the above has implications for how we move forward, and about where we should direct our efforts. Our field has very much bifurcated around this very issue. An analogy that comes up for me in this regard is the very different orientation to the game of football by the offense and the defense. The offense is about imposing a particular order on things and implementing a particular strategy, whereas the defense is oriented toward tackling whatever complexity it confronts. For the more model-driven perspective, the orientation is a strategic one. In order to persuade the scientific community the research design must be a standard one if the results are to be welcomed. For the defensive team, the response is always a tactical one: how do I manage the particular conundrum that a particular client represents. To this question all past clinical experience is relevant. Research designed out of this perspective would be organized very differently.

For the offense, neurofeedback must be shaped as much as possible into a mold that fits within the standard view. One delights in the fact that evidence for neurofeedback efficacy can finally be seen in functional MRI plots, for example. That represents tangible evidence of change from an independent source. In practice, the QEEG tealeaves are inspected for evidence of the disorder. Any deviations that are found are ipso facto declared to be a feature of the identified disorder, and become the target of remediation. Clinical evidence is sorted with respect to certain canonical disorders. The diagnosis matters because only group data counts, and only statistical analysis of such data can survive into publication.

The defense team in neurofeedback, on the other hand, realizes that whether a particular diagnostic threshold is crossed has not the slightest import for what is to be done clinically. The diagnostic threshold is a nullity in the clinical world of neurofeedback. Disregulation is on a continuum. There are no thresholds. There is no stasis. Today’s measurement will not be tomorrow’s. An arbitrary threshold crossing on one day may not be confirmed on the next. No matter. Disregulation is the target quite irrespective of the diagnostic presentation. The research design has to be in terms of characterization along many dimensions.

In the spectrum model, no single remedy must bear the burden of efficacy by itself. It is fully to be expected that every remedy brought to bear may contribute only incrementally to overall remediation for the vast majority in a clinical cohort, with perhaps only a fraction of that cohort exhibiting dramatic change in response to one intervention or another. That fraction does not have to be a statistically significant sample, and the benefit does not even have to be visible across the board. As an example, Linus Pauling entered upon the study of Vitamin C entirely on the basis of his review of published studies that officially had shown negative outcome. That of course was in the standard paradigm of statistical appraisal of whole-group effects. He simply looked beyond the statistics and found things that should not have been ignored.

On another occasion, psychologists reviewed a British study of the effect of vitamin supplementation on measured IQ. The formal results of the study were negative, but the psychologists found, on more fine-grained review, that some of the children in the group had made startling changes in IQ that could not readily be explained away. The study had been done on upper-class British children who would not be expected to be nutritionally deficient as a group. On the other hand, it is not unlikely that mental performance might have been constrained by a vitamin deficiency in some small subset of these children for one reason or another. The results, as they turned out, where just what one might have expected.

It is not unlikely that measured IQ is multiply-constrained, and that there is a wide variety of physiological pathways in which test performance might suffer in an individual case. A spectrum model must apply, and it follows that the usual statistical assumptions of group homogeneity are not applicable and should not be held up as the prevailing standard.

Once the intrinsic complexity of our favorite disorders of disregulation is recognized, the implications for research design follow immediately. Advantages then accrue to the “defensive” rather than the “offensive” orientation toward the clinical data and to research. Some arbitrary ordering principle is not imposed upon the research paradigm or upon the research subjects; rather, any order becomes a matter of a posteriori discovery and of subsequent confirmation out of the data themselves. Research hypotheses must be grand and encompassing; data acquisition must be multi-pronged and multi-dimensional; data analyses must be multi-factorial. Research becomes more observational than prescriptive.

This is not far removed from the world of the clinician, and this is where the neurofeedback clinician has been all along. The handholds and verities of research in the classical mode are not available to the clinician. The default remedy is to accrue information one client at a time, and to abstract from these clinical narratives certain patterns of responding that show some consistency over time and different subjects. Now it goes without saying that this process can also go badly wrong, as was beginning to be the case with the technique of facilitated communication. Formal research is designed precisely to rise above our tendency to alight upon ephemeral correlations and imbue them with significance. Even a large group of professionals can be persuaded of its own efficacy, and it’s off into the ozone. So there do need to be some tethers back to objective reality. These can come later in the scheme of things. And in the case of neurofeedback they are plentifully available in all of the CPT and IQ and MMPI data that have been accumulated over the years.

We assert, then, that the “defensive” or observational research posture is a viable point of departure, as well as a fertile source of research hypotheses that can subsequently be subjected to greater scrutiny. With what has already occurred in this field to date, we would assert that the “existence theorem” of neurofeedback has already been adequately proved. That is to say, neurofeedback efficacy cannot in all cases be down-sized to a mere placebo effect or other non-specific effect of the training. More affirmatively still, disorders of disregulation respond broadly to the neurofeedback challenge. The literature in support of this proposition is now persuasive. Now it will be argued that the rest of the world has not yet been persuaded, and it would be my response to say that we do not have to hold up the next step in the research paradigm until that milestone is met. Ignorance, like poverty, will always be with us.

Without any apologies we can now move forward into investigating the more interesting subsidiary questions that can illuminate for us the particulars of the disregulation model. At the same time, and in recognition of the centrality of the spectrum concept, we should not even implicitly yield the power of decision to the standard “null-hypothesis-testing.” The evidence derived from such testing is fine so far as it goes. That is to say, if the data are sufficiently robust to contradict the null hypothesis, then the results that satisfy the rest of the world will also suffice for us. On the other hand, if the null hypothesis is not contradicted, that is not necessarily the end of the story. The spectrum concept mandates that data be evaluated with a finer screen. Prospectively, we should not submit casually to the “null-hypothesis-testing” mandate. I for one will not take the bit.

Professional Implications of the Spectrum Model

On one of the e-lists a discussion has been ongoing with respect to the obligations of neurofeedback practitioners when working with classic diagnostic categories. The essential point is that unlicensed neurofeedback clinicians have no business claiming publicly to be working with and even “treating” specific diagnoses. That is the purview of the licensed professions who have been professionally qualified with respect to all the clinical issues surrounding diagnosis and treatment.

A remedy suggests itself immediately from the perspective of the disregulation model and the spectrum concept. The classic model does not serve us at all in any respect whatsoever, except of course for the fact that this is the model that our potential customers carry around in their heads. Somehow we have to reach people where they are. We do that be reframing all of the issues with a view toward the disregulation model and the spectrum concept. By the same token, it does not serve us at all to speak in the language of treatment and cure. It also does not properly represent what we accomplish. The appropriate model is that we aspire to achieve functional improvement in the context of whatever diagnosed condition, genetic endowment, nutritional status, or psychological stressors may prevail.

In other words, we move affirmatively to remove ourselves from the line of fire that is likely to be directed our way by mainstream vigilantes. One of the advantages accruing to us when we confront the standard model is that it presents to us a rather fixed target. It is therefore easy for us to orient ourselves with respect to such a target. When people confront us with the language of “chemical deficiency,” we segue to the language of disregulation. When people ask, does neurofeedback cure a particular condition, we segue to the broader domain of the dimensions of function and dysfunction. Any particular disorder can be discussed in terms of the functional domains in which deficits are typically observed for that disorder. It is these domains on which we focus, and for which we furnish evidence of efficacy. We divide and conquer.

We act in a manner that is consistent with our reality, which is that the classical disorders “do not carve nature at its joints.” We mention them only as a point of departure, and only to draw the connection with our own preferred model. If this is done consistently, then our words will align with our actions, our policies, and our particular remedy, and we will also avoid presenting a target of convenience to those for whom neurofeedback poses a threat. The operative phrase is that we “render unto Caesar what is Caesar’s,” and we abandon any reference or orientation toward the canonical disorders.

The systems perspective imposes yet one other important constraint on the way that we work, which is that we cannot work alone. We must ally ourselves with a variety of other professions that address different aspects of the “systems failure.” Now in line with the increasingly transient quality of personal and professional relationships, it is unlikely that many of such competences will end up co-resident at the neurofeedback office. What becomes established instead is a virtual network of professional connections, one in which the individual links may each have a certain transient quality. The existence of such a network renders it superfluous to have any other specialty–psychotherapy or otherwise–co-resident with the neurofeedback practice. In the systems perspective, no single specialty commands a preferential place in the systems approach. It is the entire network that confers stability and equilibrium to the enterprise, not whatever professional happens to be co-located with the neurofeedback practitioner, if any.

This brings us to the final linchpin of the systems concept, which is that it could benefit from a “systems integrator.” What has always been missing in our various systems of organizing health care is the over-arching perspective that makes sense collectively out of everything that is being recommended to a particular person. By default, the parent or patient typically occupies that role, often without explicit awareness of the centrality of that role. From the vantage point that we have accumulated through our neurofeedback work, it makes sense that the most comprehensively integrative perspective goes along with the most integrative of approaches, namely neurofeedback.

So perhaps the neurofeedback clinician comes closest to providing the needed integrative perspective. It is not where most of us are at the moment, but we could slip into that role nicely. On the other hand, medical practitioners tend to rule themselves out as system integrators by virtue of the compartmentalized thinking inculcated by the medical model. Medical practitioners have shown themselves unwilling, by and large, to embrace those who are not within the medical community. The spectrum concept represents almost an opposite pole that needs to be out there as an alternative focal point for the organization of clinical and personal data.

There is actually a kind of unfortunate precedent for this, in that the gatekeepers at insurance companies fill this role by default. They are also typically not medically licensed. Their own institutional bias is obvious, so they cannot fill that role as we intend for it to function. But it is interesting that the medical world has already moved to where critical decisions with respect to care are being made by a non-medical person at some remove from the issues.

Two Perspectives; Two Futures

Those who find themselves in the leadership of the neurofeedback organizations feel constrained by various pressures to establish neurofeedback according to the newly emplaced canons of “evidence-based medicine.” Somehow we are to thread the needle on large-scale controlled studies with homogeneous sample populations with random assignment for each distinct disorder, all to be done without major funding and without institutional sponsorship. Somehow neurofeedback is to be tested in a manner that strips away any of the contributions that a psychologist or other trainer might actually bring to the proceedings–motivation, engagement, support, understanding, interaction, therapeutic alliance, responsiveness to events, adaptation, redirection, the celebration of success, and finally closure to a successful “hero’s journey.” Neurofeedback must be tested naked, like throwing infant monkeys in with Harlow’s wire mothers. (“They give milk, don’t they?”)

In deference to such standards, there must also be denunciations of anything else that falls short of the gold standard, lest the impression be given that one might be swayed by anything less than the best of evidence. Fealty to such standards also mandates a kind of denial of everything that already exists, or alternatively the insinuation that what exists is somehow counterfeit, or at any rate is making claims dangerously outrunning the facts.

In this model, it is urgent that we reify the objects of training by direct observation–lumps on an EEG; coherence anomalies, etc. Otherwise, what would be doing? Along with that, we seek “tangible” evidence of having actually accomplished something. So this crew looks to fMRI data for confirmation! Everything short of such tangible evidence remains a little suspect. Somehow the autistic child making eye contact, the attachment-disordered child learning to engage, is not enough. Curiously, all of this wholesale denial of “psychology” is being done in the interest of psychology. Now without the formalism of psychology, attachment disorder would never have been codified in the first place. So why cannot the resolution of attachment disorder be judged also entirely within the realm of psychology? They did not discover attachment disorder in an MRI scan, after all.

The roadmap for this worldview is to rely as much as possible on objectively measurable data, principally the EEG, to guide the neurofeedback. It should be observed, before we go too far down that road, that almost anything that can be determined in a quantitative EEG will also be doable in near real time in the future, and thus embedded automatically into the feedback paradigm over the timescale of a session. The clinician will have less and less to do with it, and will not even have to see the data. All will be managed by complex algorithms embedded in software. Then the gold standard of totally objective targeting will have been reached, and the hand of man will no longer be needed in the process. Hurrah for psychology!

In the meantime, the official research agenda mandates fixed protocols for evaluation. The “one-size-fits-all” approach may be disparaged in other contexts, but in fact it constitutes the researchers’ ideal. The approach that comes closest to this at the moment is Joel Lubar’s and Vince Monastra’s (not Val Brown’s!), and this is surely being done in the interest of clean research designs. The use of a single protocol in turn means that the population has to be selected to fit that protocol. So in the end a kind of self-consistency prevails: the population sample has been matched to the technique. But reality has surely been turned on its head. It’s not that other folks are not also trainable; it’s merely that they may need protocol flexibility.

Then there is the other reality, one that grasps the inherent complexity of any human interaction as impactful as the delivery of neurofeedback in aid of a profoundly disregulated nervous system. The conversation that takes place in a psychodynamic encounter–with the trainee and with the family system–now acquires another dimension: an interaction with the nervous system that organizes these psychological states and family systems interactions. The training becomes a learning experience for both clinician and client, one in which the client grows in awareness of his physiology, of his capacity for mastery, and of larger dimensions of the self. For the clinician, it is a matter of “the art of the possible.” How much challenge can the given nervous system take? What is the right pacing of the training? When should new challenges be introduced?

One observation is central to all of this: In the dynamic approach to training, it never gets any easier. Competence increases over time, but the challenge presented to the clinician rises comparably. Whatever domains of competence the clinician possesses, they each present means of assessing progress, and they confront the clinician with choices. The larger the area of competence, the greater the challenge to manage well. Whatever competence one brings to the table, nature matches it on the other side.

The opposite is the case for the first approach. Once a technique is validly established, it becomes teachable as a procedure and matters plateau. All the inherent complexity has been written out of the script by the “science.” If it is not in the research paper, then it does not matter. Let’s hear it once again for psychology!

The thought that this complex, dynamic interaction could become subservient to the reductionist scientific agenda should offend any clinical psychologist deeply. The scientific enterprise can study neurofeedback all it wants with reductionist models and on its own terms. When it imposes that paradigm on the clinical practice of neurofeedback, on the other hand, it constitutes overreach. The competent neurofeedback clinician is right to regard the scientific study of neurofeedback by such methodology as an entertaining sidelight and a charming irrelevance. None should be intimidated by it.

Summary and Conclusion

In the two approaches to neurofeedback just discussed, we see reflected once again the dichotomy of the two hemispheres. The world of science has had its extraordinary success with a consciously reductionist perspective in which experiments were designed to limit the variables in play. This is the world of physics and chemistry, where the outcome is something that can approach finished status, like the law of gravity. The ideal is science cast as in a static mold, and then institutional mechanisms can established themselves around those same static molds.

Even with our brief venture into the 21st century, that already has a certain 20th century datedness about it. We can already sense that the left-hemisphere propensity toward dividing the problem and conquering the pieces does not do justice to the whole. We can already intimate that the left-hemisphere preoccupation with controlling variables cannot be pushed forward into domains where the key variables cannot be controlled.

In one of the last entries of Brian’s diary in 1991, he wrote about wanting to study the “science of complexity.” He certainly grasped the key scientific challenge of the day. In some natural events, complexity is of the essence and it will not serve to extract the pieces. Meteorology and climate change are cases in point. International finance comes to mind. The operation of the human nervous system is another case in point, and perhaps the best. In the face of intrinsic, perhaps irreducible complexity, the emphasis must shift to observation along many dimensions, and the science must shift from the “offensive” posture of planning our scientific campaigns to the “defensive” posture of opportunistically registering what phenomena nature makes available to us to observe.

We are in the Lewis and Clark phase of the exploration of brain function. The clinician is of necessity having to come to terms with the complex. It’s not complete science, to be sure. But a scientific discipline does apply to the skills of observation, and this does furnish the essential observational data that allows theoretical, model-based science to function. The skilled clinician by nature brings a right-hemisphere, integrative perspective to his task. Complexity is the clinical reality. The traditional scientist is like the altricial birds that have to be nurtured to maturity for some time in the nest. The clinician is more like the precocial birds, like the ducklings that can be tossed in the water right after clambering out of the shell, and they can already swim.

In our human development, we are both altricial and precocial. We are altricial with respect to motor function and cognitive function, and we are precocial with respect to emotional attachment. We come into the world as prepared for attachment as ducklings come for swimming. But societally we acknowledge the hierarchy of the acquired skill. We naturally defer to the primacy of left-brain intelligence, to the intellectual gifts that are not so evenly and universally distributed. Only now are cognitive scientists coming to acknowledge that right-hemisphere skills also represent higher function. It is the right hemisphere that is endowed by nature to manage the complexity of our world in real time.

Psychodynamic treatment and neurofeedback training are always conducted as a real-time process. Science almost never is. We are talking about almost incommensurate competences. We therefore have in neurofeedback the laughable situation that altricial birds, while still being nurtured in the nest, are complaining to the ducklings in the water that they cannot swim–or, since everyone can see that they are in swimming, that they are not using the right strokes. In fact, it is the clinician that is pointing the way toward the science of complexity by stirring up the raw material of science with which we must engage.

Leave a Reply