On the Use of EEG Operant Conditioning as a Treatment for Affective Disorders, including Reactive Attachment Disorder and Borderline Personality Disorder

Siegfried Othmer, Ph.D.
Chief Scientist, EEG Institute
February, 2002

We may be on the threshold of a significant breakthrough in our understanding of brain function, and this new understanding may have significant implications for the remediation of some of the most intractable psychopathologies. In this brief treatment, only the most cursory perspective can be given on these emerging developments.

The prevailing paradigm for the understanding of mental disorders centers on the understanding of neuromodulator systems, in that psychopharmacological agents are the usual remedies. Over the last few years, increasing reliance has been placed on the use of agents referred to as anti-convulsants by neurologists, and as mood stabilizers by psychiatrists. These have been applied to the most intractable of psychiatric disorders. These are characterized typically by sudden, uncontrolled changes in the state of the brain. Among the psychopathologies, this class of conditions is least well managed medically.

In order to understand those conditions that are secondary to the existence of instabilities in brain function, we must understand the mechanisms by which the brain organizes its own states. This must of necessity require a model that treats the brain in the time domain, and in terms of its bioelectrical interactions. Such a model is rather tangential to neuromodulator-based brain models. An entirely new formalism is required. The cognitive neurosciences have attempted to address this problem at a number of levels.

At the most fundamental level, that of the action potential, it has been largely a dead-end road to try to understand how information is encoded in the brain (the neural code problem). Information is generally carried by the brain as a group property, or an ensemble property, and hence we must instrument ourselves to observe entire ensembles. This is in practice challenging. Nevertheless, this approach has led to key insights, in particular the principle that collectivities of neuronal firings distinguish themselves in the timing domain by means of unique timing signatures. That is, timing defines belonging. This organizing principle is known as "time binding."

The existence of time binding as a fundamental aspect of the neural code imposes a more global organizational burden upon the brain. That is to say, the orderly processing of information up the chain requires that timing coordination be organized along the entire pathway, or the signal is "gated" into oblivion. It is perhaps best to illustrate this principle with an example. It is now known that visual processing occurs in the human brain at a 40-Hz basic rhythm. The brain "snap-shots" the world periodically and moves the signal along subsequently in packets of 25 msec duration. The sequential passage of the signal through the brain is, however, largely driven by neuronal delay transit delay times, which are a given. The brain can only act negatively upon this signal by failing to pass it on at the next synaptic gate. Otherwise it continues to reflect its original timing signature. Thus the original timing established by packet formation in striate cortex imposes a temporal ordering on all down-stream processes that rely in any way on visual-channel signals. Presumably this carries through all the way to the motor act.

It is clear then that the brain must organize itself globally in the time domain for proper function. The means for doing this have been gradually uncovered by means of in-vitro experiments on certain brain networks, in particular the thalamocortical networks. It is also necessary to observe the brain in terms of its collective activities, and the EEG is suitable to this enterprise. In fact, it has been shown that thalamocortical networks are responsible for the obvious EEG rhythms that have been studied since the 1920s. Whereas matters are not yet well understood, it is argued that there are cortical-cortical, thalamo-cortical, cortico-thalamic, and brainstem influences on the key periodicities that we observe in the EEG.

If we are so crucially dependent for good function on the organization of brain timing, it would not be surprising to find that certain psychopathologies are traceable to problems or breakdowns in brain timing. Once this model is adopted, even hypothetically or conditionally, it becomes apparent that many psychopathologies can be fruitfully regarded from this perspective. This insight was presented at the Neurosciences meeting in 1999 in Miami by Rodolfo Llinas, who initially proposed that "thalamocortical dysrhythmias" could be the basis for understanding conditions as disparate as major depression, tinnitus, Parkinson's, and neurogenic pain.

Llinas had only sparse evidence for his rather bold proposal. Four cases, one per condition, had been studied via MEG (Magnetoencephalography---which differs from EEG largely in that it can study the brain at depth). A similar pattern was observed for all, namely that there was an inappropriate level of "coupling" (i.e., correlation) between the high frequencies and the low in these conditions. It can be argued that good brain function requires a coordination across the frequency domain between the higher frequencies (such as those involved in visual processing) and the lower frequencies which establish the overall pacing of brain activity. The higher frequencies are committed to organizing localized function, and the lower frequencies to the more distributed, integrative functions. A breakdown in this relationship allows for the appearance of phantom sensory experience (tinnitus, phantom pain, the positive symptoms of schizophrenia, for example), and the initiation of inappropriate movements (motor and vocal tics of Tourette Syndrome, Parkinson's, sub-vocalization in schizophrenia, etc.).

Llinas' presentation resonated so strongly with the prepared audience that his paper was accepted for immediate publication (sans peer review!) in the Proceedings of the National Academy of Sciences. This is almost unprecedented. His paper was shortly followed up by a note by David McCormick in which he allowed himself the speculation (there were no data) that thalamocortical dysrhythmias could be the "Rosetta Stone" of a large variety of psychiatric disorders.

One of the unfortunate consequences of the necessary compartmentalization of science is that cognitive neuroscientists are not ordinarily exposed to the literature on mental disorders and their treatment. Hence it is not surprising that a parallel development could take place in the clinical world that supports the model of thalamocortical dysrhythmias, but is probably entirely unknown to these researchers. In the late sixties, the first experiments were performed in which brain function and behavior were systematically altered in cats by operant conditioning of EEG spindle-burst phenomena (the sensorimotor rhythm). Fortuitously, it was observed that such operant conditioning significantly altered the threshold for chemically induced seizures. Subsequently, it was shown that seizure incidence could be systematically reduced in medically refractory cases of epilepsy.

Unfortunately, no brain model existed at the time in terms of which these results would make sense. Hence, these scientifically "premature" findings were neglected by neurology in its rather single-minded focus on pharmacological and surgical remedies. Clinically, however, the work with EEG reinforcement evolved to the point where a wide variety of psychopathologies are now known to yield to operant conditioning of EEG frequency-based parameters. And just as sensorimotor rhythm training can be seen as a kind of bio-electrical anti-convulsant, developments over time have paralleled those in psychiatry almost precisely, leading to the use of similar techniques as "mood stabilizers" for psychiatric conditions.

Martin Teicher showed back in 1984 that there was a high correlation of symptoms of temporal lobe epilepsy with abuse histories, with the highest correlation for those with a history of both physical and sexual abuse. We would like to interpret these results as indicating that psychological trauma is sufficient by itself to produce the functional instabilities to which the EEG gives witness, even absent any physical insult. These findings further support the commonality of mechanisms that may underlie the seizure susceptibility of interest to neurology and the mood instability of interest to psychiatry. By whatever pathway one gets there, the brain will have become unstable.

The good news, however, is that the brain is sufficiently plastic in the general case that more stable function may be achievable through operant conditioning techniques based on EEG phenomenology. This rather astounding claim, made entirely on the basis of opportunistic empirical findings in clinical practice, must of course have sounder grounding in controlled research. In particular, the clinical methodology that has evolved over the last several years (all based on the early work with seizure control), has been shown efficacious in applications to Borderline Personality Disorder, Reactive Attachment Disorder, Episodic Explosive Disorder, Post-Traumatic Stress Disorder, Sociopathy and Conduct Disorder, and Dissociative Identity Disorder.

These preliminary findings lend substance to the proposition that even major, treatment-resistant psychopathologies are susceptible of remediation by direct appeal to the self-regulatory mechanisms of the brain, as manifested in the EEG. This means in particular that models based on "critical periods" for emotional development need to be revisited. It appears that the requisite neural circuitry for the restoration of emotional regulation is in place, and merely requires training to improve functional organization and re-integration.

EEG neurofeedback, then, is emerging as a versatile technique to appeal directly to specific brain networks involved in emotional stabilization and regulation. Even though we are dealing with largely sub-cortical regulatory activity, there is a cortical manifestation that can serve as the operant. Reinforcement parameters are altered throughout the training routine in order to optimize the response. Such a response can be obvious even within a particular session, or become apparent between sessions. The dependence of outcome on the specifics of reinforcement implies a kind of built-in control for every trial. Optimization of the training means that we are dealing throughout the process with a kind of A-B design, since one is always confronted with choices in terms of clinical protocols. This built-in control, and the rapidity of brain response, means that the clinical findings to date are robust, and highly deserving of validation through more refined controlled studies in a university setting.

The relative simplicity of this technique bodes well for accessibility to the mental health professional. The non-invasive character of the technique means that research validation could lead to rapid insertion into the clinical environment via a broad range of mental health professionals, and on a large scale, in order to address a wide range of the most serious psychopathologies that are currently intractable and therefore under-served.

Proc Natl Acad Sci U S A 1999 Dec 21;96(26):15222-7

Thalamocortical dysrhythmia: A neurological and neuropsychiatric
syndrome characterized by magnetoencephalography.

Llinas RR, Ribary U, Jeanmonod D, Kronberg E, Mitra PP

Department of Physiology and Neuroscience, New York University
School of Medicine, 550 First Avenue, New York, NY
10016, USA.

Spontaneous magnetoencephalographic activity was recorded in awake, healthy human controls and in patients suffering from neurogenic pain, tinnitus, Parkinson's disease, or depression. Compared with controls, patients showed increased low-frequency theta rhythmicity, in conjunction with a widespread and marked increase of coherence among high- and low-frequency oscillations. These data indicate the presence of a
thalamocortical dysrhythmia, which we propose is responsible for all the above mentioned conditions. This coherent theta activity, the result of a resonant interaction between thalamus and cortex, is due to the generation of low-threshold calcium spike
bursts by thalamic cells. The presence of these bursts is directly related to thalamic cell hyperpolarization, brought about by either excess inhibition or disfacilitation. The emergence of positive clinical symptoms is viewed as resulting from ectopic gamma-band activation, which we refer to as the "edge effect." This effect is observable as increased coherence between low- and high-frequency oscillations, probably resulting from inhibitory asymmetry between high- and low-frequency thalamocortical modules at
the cortical level.

Nature Medicine 5: 1349-1351 (1999)

Are thalamocortical rhythms the rosetta stone of a subset of
neurological disorders?

McCormick, David A.

Recent evidence indicates that "dysrythmias" cause alterations in the normal function of the thalamocortical loop and lead to various types of neurological disorders. Will decoding this rhythm help us to understand the basis of movement disorders, chronic pain, and even neuropsychological disorders?