Available via license: CC BY
Content may be subject to copyright.
Article Not peer-reviewed version
Endogenous Neuromodulation at
Infra-Low Frequency: Method and
Theory
Siegfried Othmer * and Susan FitzGerald Othmer
Posted Date: 17 October 2023
doi: 10.20944/preprints202310.1085.v1
Keywords: endogenous neuromodulation; neurofeedback; Infra-Low frequency; slow cortical potential;
intrinsic connectivity networks; infra-slow fluctuations; EEG biofeedback; developmental trauma
Preprints.org is a free multidiscipline platform providing preprint service that
is dedicated to making early versions of research outputs permanently
available and citable. Preprints posted at Preprints.org appear in Web of
Science, Crossref, Google Scholar, Scilit, Europe PMC.
Copyright: This is an open access article distributed under the Creative Commons
Attribution License which permits unrestricted use, distribution, and reproduction in any
medium, provided the original work is properly cited.
Article
Endogenous Neuromodulation at Infra-Low
Frequency: Method and Theory
Siegfried Othmer and Susan F. Othmer
The EEG Institute, Los Angeles, CA, USA
* Correspondence: Siegfried Othmer siegfried@eeginstitute.com
Abstract： Clinical work conducted over the last seventeen years at the EEG Institute in Los Angeles and by
other neurofeedback providers around the world has demonstrated the utility of extending frequency-based
neurofeedback deep into the infra-low frequency (ILF) region, using the method of endogenous
neuromodulation described herein. The method is characterized by the absence of any overt reinforcements,
which makes it possible to extend the clinical reach to extremely low frequencies. As the training frequency is
lowered, the signal becomes more difficult to discriminate, and ultimately it can only be discerned by the brain
itself, in the process of endogenous neuromodulation. The method emulates how the brain does skill learning
in general: It must observe itself performing the skill, with feedback on its performance. While the immediate
target of ILF neurofeedback is enhanced self-regulatory competence--with symptomatic relief and functional
recovery the secondary consequences, progressive lowering of the target frequencies has led to improved
outcomes in application to challenging dysfunctions such as episodic suicidality, migraine, seizures, and
bipolar mood swings. The work has also yielded insights into how the frequency domain is organized. The
training proceeds best at frequencies that are specific to each individual, and these are referred to as optimal
response frequencies (ORFs). These frequencies differ for various placements but stand in two fixed
relationships to one another, one that holds over the EEG spectral range, and another that holds over the entire
ILF range. Training in the ILF region engages the dynamics of the glial-neuronal networks, which govern tonic,
resting state regulation. The collective clinical experience with ILF neuromodulation within a large practitioner
network supports the case for making protocol-based, individualized ‘homeodynamic’ regulation a therapeutic
priority, particularly for our most impacted clinical populations: addiction, trauma formations, traumatic brain
injury, and the dementias. The case is made for further outcome studies and foundational research.
Keywords: endogenous neuromodulation; neurofeedback; Infra-Low frequency; slow cortical
potential; intrinsic connectivity networks; infra-slow fluctuations; EEG biofeedback; developmental
trauma
Introduction
The frequency-based techniques of neuromodulation, in all their variety, have benefited from
considerable innovation over the past few years (Evans and Turner, 2017). Nearly all manipulate one
or another parameter governing the EEG spectrum with either a reinforcement or stimulation-based
technique. There are also methods that modulate the Slow Cortical Potential, both challenge- and
stimulation-based, to be discussed further below. We concern ourselves here with the alternative of
endogenous neuromodulation, in which the brain engages directly with the spectrally limited EEG
or Slow Cortical Potential. More specifically, features are extracted that reflect the mechanisms by
which the frequency domain is organized. This more naturalistic way of working with low-frequency
signals extends neuromodulation techniques beyond the reach of prescriptive methods such as
operant conditioning. It appears to be more effective than prescriptive methods in some generality,
in that it merely supports and augments how the brain learns and maintains self-regulation natively.
The method offers unique advantages when it is extended deep into the infra-low frequency region,
the focus of this paper.
In brief, the method involves having a client watch the time course of the narrow-band filtered
Slow Cortical Potential as it unfolds over the course of a 20-to-50-minute session. The signal is derived
Disclaimer/Publisher’s Note: The statements, opinions, and data contained in all publications are solely those of the individual author(s) and
contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting
from any ideas, methods, instructions, or products referred to in the content.
Preprints (www.preprints.org) | NOT PEER-REVIEWED | Posted: 17 October 2023 doi:10.20944/preprints202310.1085.v1
© 2023 by the author(s). Distributed under a Creative Commons CC BY license.
2
from a differential montage at selected cortical sites. Since the signal is slowly varying and therefore
of limited interest to the trainee, it is typically imbedded in a video game or movie. For example, the
signal may be used to modulate the size of the screen display; or it may govern the speed of a rocket.
Auditory and tactile feedback complement the visual representation for a more immersive
experience. While the trainee is engaged with the movie or the game, the brain tracks the signal of
interest. As soon as the brain detects the relevance of the signal to its task of state management, it
incorporates the signal into its regulatory schema as an additional control loop. Since there are no
cues or stimuli to provoke or direct the brain’s response, the challenge must be entirely self-generated
and self-referential. Thus, the response is based on the brain’s ‘detection and correction’ activities,
operating on the dynamics of the signal being tracked, rather than on the more conventional
algorithmic inputs external to the brain.
To function in real life, the brain must operate on a prediction model. This applies not only to
action-perception cycles but to state management as well (Sterling, 2012). Endogenous
neuromodulation rests on this operational framework. A plausible model of the process may be
delineated as follows: At the outset, the brain detects a correlation between a feature in the sensory
signal stream and its internal state, progressively confirming the provisional hypothesis of correlation
by virtue of its persistence. The signal is projected forward in time by way of predictive coding, wity
successful prediction confirming the correlation. The brain will then act to bring closure between its
prediction and the unfolding reality. This constitutes the self-generated challenge. It can be
understood as a continuous process of Bayesian inference (Friston, 2010). The operation of all aspects
of this continuous process are concurrent rather than sequential, which has been described in terms
of circular causality (Freeman, 2006).
Thus, by tracking the surface potential, we have unblinded the brain to a core self-regulatory
activity, the dynamic management of tonic cortical activation. We are therefore justified in using
language that ‘anthropomorphizes’ the brain, so to speak, because clearly the brain is the agent in
this process. Volitional involvement is not required. The method can be used even with infants and
with people in diminished states of consciousness.
In the clinical setting, the process becomes an iterative encounter between first- and third-person
perspectives, one that allows the clinician to discern the state dependence of dysregulation status.
This facilitates the migration to the optimal training parameters in terms of placement and target
frequency (Bagdasaryan and Le Van Quyen, 2013). Over the course of sessions, this becomes a
scaffolding process in which foundational issues in the hierarchy of regulation are targeted first,
setting the stage for addressing subsidiary objectives.
The Historical Context
Before the method is described further, the context for this novel developmental thrust toward
low-frequency training is presented. The confluence of three major developments in the 1960’s to
1980’s laid the foundation for the development of ILF neurofeedback. There was the concurrent
development 1) of classical biofeedback, with its emphasis on training the regulation of the
autonomic nervous system with the aid of measures of peripheral physiology, and 2) of EEG
biofeedback (the common terminology at the time) that was mainly promoting brain stability (with
a focus on epilepsy) and hyperactivity, with a focus on ADHD. Both objectives were pursued by way
of calming motoric excitability. The third major development was the pursuit of Slow Cortical
Potential training in Europe during the eighties, again with the objective of calming cortical
excitability.
Therapeutic EEG biofeedback was discovered quite accidentally in the late sixties in animal
work. Under umbrella of sleep research, cats were operantly conditioned to enhance the incidence of
spindle-bursts at the frequency of the sleep spindle, 12-15 Hz. This bursting activity was labeled the
sensorimotor rhythm (SMR) (Wyrwicka and Sterman, 1968). Down-training of the bursting activity
was demonstrated as well. To prove that operant conditioning had occurred, two groups of cats were
trained successively in both directions, in a crossover design. Motoric excitability diminished with
the SMR reinforcement training, and sleep parameters were altered, as indexed by sleep spindle
Preprints (www.preprints.org) | NOT PEER-REVIEWED | Posted: 17 October 2023 doi:10.20944/preprints202310.1085.v1
3
density (Sterman et al.,1970). Months later, these same cats were committed to a test of the toxic effects
of low levels of monomethylhydrazine, a rocket fuel (Sterman, 1976). This substance depletes GABA
stores, which renders cats highly susceptible to seizure. Surprisingly, half of the group demonstrated
heightened resistance to seizure onset, and upon inspection it was found that these were the cats that
had finished with SMR up-training in the prior experiment. The learned behavior change had not
suffered the extinction that commonly attends operant conditioning designs after reinforcement
ceases. Here we had an establishing study that quite by happenstance was both blinded and
controlled, as well as suitably powered. The researchers had been unaware of the ground-breaking
import of the serial experiments that they had just consummated. No respectable scientist would
have posited what was found here as a hypothesis a priori.
Human subjects research followed, as interest in the method flourished for some years (Lubar
et al, 1981; Lantz and Sterman, 1988). A summary of those studies was published in 2000 (Sterman,
2000). As all subjects in those studies had been treated to medical standards and were medically
stable, the further reduction of seizure incidence by some 60% (mean value) renders the method
comparable to what can be achieved with a second or third anti-convulsant. Often, medication could
be titrated down or even withdrawn entirely. And not infrequently (~5%) the trainees became entirely
seizure-free. Nevertheless, Federal funding of this research terminated in 1985, and the method has
been kept alive subsequently mainly in clinical work and via self-funded studies. It had been a case
of ‘prematurity in science’—a discovery made before its time (Hook, 2002).
The third tributary to the confluence was the work originating in Europe to train the Slow
Cortical Potential (SCP) toward diminished cortical excitability by way of a repetitive transient
challenge to the trainee to shift the SCP toward increased positivity (Rockstroh et al, 1989; Birbaumer
et al, 1990; Birbaumer, 1999). In this learning task, clients are challenged to alter the ambient level of
the SCP by several microvolts within nominally eight seconds (Strehl, 2009). For the first time, tonic
state regulation—the focus of nearly all of biofeedback—was being trained directly with a
neurofeedback technique. This broadened the clinical agenda considerably.
An extensive research history exists for SCP-based neurofeedback, and the clinical footprint
largely overlaps with what can be accomplished with frequency-based training in the low beta band.
The core claim of moderating cortical excitability has implications for epilepsy (Rockstroh et al, 1993;
Kotchoubey et al, 2001), migraines (Siniatchkin et al, 2000), and even schizophrenia (Schneider et al,
1992; Gruzelier et al, 1999). The predominant application has been to the ADHD spectrum, just as in
the case of SMR and low beta-band training (Strehl et al, 2006).
In all these applications, the objective is diminished excitability as a learned response, with
consequences for tonic state regulation. This objective is explicit in the last of the studies cited above,
by Gruzelier et al. Here a shift in the hemispheric asymmetry of negativity (i.e., excitability) is directly
targeted by way of the differential signal at {C3 – C4}. The training is challenge-based, within the
customary eight-second window. Thus, all the above techniques for training the SCP rely on phasic
responding to train tonic activation. Nevertheless, this study can be seen in retrospect as another
technological precursor to the technique of endogenous neuromodulation under discussion here.
The entire field of biofeedback collectively made the case for the clinical utility of enhancing
autonomic regulation, and of restoring sympathetic/parasympathetic balance. SCP training then
enlarged the agenda for tonic state regulation to cover a broader range of state variables, as well as
to enhance cerebral stability. And EEG frequency-based training enhanced our capacity to focus on
specific functional domains. The stage was thus set for the exploration of the role of infra-low
frequencies in state regulation. The transition into ILF NF was more a matter of continuity rather than
of discontinuity, however, so our treatment of the topic necessarily takes us back to the origins of
therapeutic neurofeedback.
It is helpful, in that regard, to survey much of the spectral range of clinical interest. The results
are shown in Figure 1, which presents the frequency-based analysis of a three-hour 19-channel record
while the participant is passively watching a movie. We observe a power-law relationship, and in
this case a single exponent holds throughout the range of 0.1mHz to 10 Hz. A power-law relationship
characterizes dynamically self-organizing systems (Plenz and Niebur, 2014). The observed power-
Preprints (www.preprints.org) | NOT PEER-REVIEWED | Posted: 17 October 2023 doi:10.20944/preprints202310.1085.v1
4
law exponent of ~1.5 is the expected value for self-organizing systems operating at the edge of
criticality, conditions under which responsiveness is maximized within the constraints of system
stability (Beggs JM and Plenz D, 2003; Kitzbichler MD et al, 2009). In the present instance, this single
exponent applies equally to the EEG and SCP regimes. Implicit here is a frequency hierarchy of
regulation in which each frequency is responsible for its corresponding niche in the frequency
domain, as well as setting the tone, so to speak, for activities organized at higher frequencies. The
phase of the low-frequency signal modulates the amplitude of higher bands (He et al, 2010; Buzsáki
and Watson, 2013). The relationship holds in the infra-low frequency region (Monto et al, 2008). This
constitutes a potent impetus to train tonic state regulation by operating in the infra-low frequency
region.
Figure 1. Shown is the spectral power (i.e., [voltage]2) over six orders of magnitude in frequency for
a 19-site montage. Individual channels are shown in red, and the average value is shown in purple.
An eye-ball fit is shown in blue. A power-law relationship is indicated over the entire range. The
power-law coefficient is 1.55. The deviation at higher EEG frequencies may well be explained by scalp
muscle activity, or by excitement attributable to the movie being watched by the volunteer over the
course of three hours. (Data courtesy of Bee Medic, Switzerland.).
Early Developments in Therapeutic Neurofeedback
Neurofeedback did not evolve in the US in a climate favorable to innovation. After all, it emerged
as a clinical discipline well before the concept of brain plasticity was generally accepted, so it was
widely dismissed as an expensive placebo (Thibault et al, 2018). In response, clinicians hewed closely
to the protocol that had been validated in research by Sterman and Lubar: the operant conditioning
of the sensorimotor rhythm (12-15 Hz) and low-beta band (15-18 Hz). Training was typically done on
the left sensorimotor strip, at {C3 – T3}, i.e., in bipolar montage. Sterman had carved one road into
the jungle, so to speak, and everyone was on it (Sterman and Friar, 1972). In the 1985-to-1988-time
frame, our group developed the NeuroCybernetics system as a computerization of Sterman’s
laboratory instrument, and it was oriented entirely to this protocol.
For application to ADHD, Lubar moved the training of the SMR band to Cz (in referential
montage), following Tansey, where the theta-band excess, the key signature of dysregulation in
ADHD, was observed to be at its extremum along the central strip (Tansey, 1985). An inhibit strategy
of cueing the brain with respect to episodes of dysregulation was installed by Sterman to assure the
integrity of the discrete rewards, and this evolved into a shaping strategy (systematic down-training
of excess theta-band activity) at the hands of Joel Lubar (1997). We adopted both changes.
Preprints (www.preprints.org) | NOT PEER-REVIEWED | Posted: 17 October 2023 doi:10.20944/preprints202310.1085.v1
5
Protocol innovation was first potentiated with the availability of digital EEG analysis to the
private practitioner in the early nineties. This permitted targeting dysregulation wherever it was
observed in cortex, and that also served more generally to make the climate within the field more
open to innovation. The resulting proliferation of methods is encompassed by the term ‘qEEG-based
neurofeedback.’ The downside was the elevation of the localization hypothesis of neuropsychology
(just as we were entering the realm of modern network theory!), and with it a shift toward single-site
training with referential placement. This development also shifted the perspective to a dysfunction-
focus, contrary to all the traditions in biofeedback, which focused on functional enhancement. The
quantification of EEG behavior brought in train aspirations toward a manualizable, ‘scientific
neurofeedback’, with the implicit mandate that matters had to be done in this way.
As the mechanisms-based training of Sterman and Lubar was function-focused, it was protocol-
driven, with little or no individualization. Innovation therefore needed to be hypothesis-driven,
based on functional neuroanatomy. As theory was in short supply, innovation transpired much more
slowly. In our early work with ADHD, we moved from Cz to C4 with SMR-band training in 1992,
and reliably established that the two hemispheres needed to be trained differently. For many years,
the standard protocol became the combination of C3-beta training (15-18 Hz) with C4-SMR (12-15
Hz) for thousands of clinicians (Kaiser & Othmer, 2000). The right-hemisphere placement had a
calming influence, while the left-hemisphere training was more activating. We now had a predictable
handle on the regulation of central arousal: lower training frequencies took us to lower arousal levels.
The next critical step was a return to bipolar montage for all our protocols. This allowed us to migrate
off the sensorimotor strip while keeping one sensor on the familiar sensorimotor or temporal site (C3
or T3). More fundamentally, the training variable once again became the relationship between two
sites: we were training connectivity relationships.
The “C3-beta/C4-SMR” protocol found extensive representation in the literature through the
work of John Gruzelier and colleagues for over a decade (Egner and Gruzelier, 2001; Gruzelier and
Egner, 2005; Gruzelier et al, 2014). The protocol was also the impetus for a large-scale controlled study
of neurofeedback-aided addiction treatment (Scott et al, 2005). Also notable was the controlled study
by Thomas Fuchs in which the protocol was compared against stimulant medication (Fuchs et al,
2003). The attempt to discriminate between SMR and beta training was published in two papers
(Kaiser and Othmer, 1997; Barnea et al, 2004). Our multi-site retrospective appraisal of the protocol
in application to ADHD was published in 2000 (Kaiser and Othmer, 2000).
It was mainly families whose children did not do well with medication that were attracted to the
training. It was possible to resolve the sleep problems as well as the tics and bruxism that sometimes
cropped up with stimulants. Commonly we were engaged with the comorbid disruptive behavior
disorders that don’t resolve with stimulants. As these findings were considered anecdotal, they were
taken up in a book chapter (Othmer et al, 1999a). The broader reach of the protocol, to the anxiety-
depression spectrum, sleep disorders, pain syndromes, brain injury, and the autistic spectrum was
likewise consigned to a book chapter (Othmer et al, 1999b).
The singular experience at our clinic of a client becoming manic with SMR-training made it clear
that even standard training frequencies entailed significant risks, and that in consequence training
frequencies needed to be individualized. The resulting shift in clinical focus to the regulation of
central arousal led to the observation that a more optimal response could be obtained over a wider
range of training frequencies. This led to the critical observation that potentiated protocol innovation
in this approach going forward, the finding that highly dysregulated clients trained best at very
specific frequencies. The search for an optimal response frequency (ORF) in each client became the
driver that gradually extended the training across the entire EEG spectrum, but mainly moved us
progressively lower in the EEG band. The most severely dysregulated clients typically ended up
training at the lowest frequency we had available in the software, thus gradually pushing the
boundaries ever lower. As practitioners learned how to work with these new lower frequencies and
their clientele experienced better outcomes, a corresponding shift occurred in the clientele, in that
practitioner started to attract ever more complex presentations.
Working Models
Preprints (www.preprints.org) | NOT PEER-REVIEWED | Posted: 17 October 2023 doi:10.20944/preprints202310.1085.v1
6
The journey to lower frequency has been underway for a quarter century, with seventeen of
those years spent exploring the ILF regime. The centrality of early attachment deficits and
developmental trauma in our most challenging clients became apparent, and this has relevance to
chronic medical disease as well. Our approach has been organized around several models.
The Trauma Model
“PTSD [Post-Traumatic Stress Disorder] and Traumatic Brain
Injury are the leading cause of sustained physical,
neurological, cognitive, and behavioral deficits in military
personnel and the civilian population.”
Mingxiong Huang, PhD (Presentation at Combat
Operational Stress Conference, 2010)
The connection between psychophysiological state and chronic medical disease was first
brought to scientific notice with the “Adverse Childhood Experiences (ACE) Study” in the late
nineties (Felitti et al, 1998). Here the focus was on overt traumatic experiences or contexts of living:
psychological, physical, or sexual abuse, etc., during formative years. Evaluation of some 9500 adult
questionnaires yielded the finding that “Persons who had experienced four or more categories of
childhood exposure had 4- to 12-fold increased health risks for alcoholism, drug abuse, depression,
and suicide attempt.” With respect to chronic medical disease, with four or more ACEs the risk of
ischemic heart disease, cancer, lung disease, liver disease, and diabetes was found to be elevated by
a factor of two.
These findings, although troubling in themselves, under-estimate the medical risk because the
most severely impacted individuals most likely had already attritioned out of the pool and were no
longer available for the interview. A later study found that those with six or more ACE’s died 20
years earlier than those without (Brown et al, 2009). A prospective study could resolve the issue. As
it happens, the Grant study of 256 Harvard students of the late forties and early fifties had already
filled that niche. When the 160 remaining participants were re-evaluated in the 1990s, it was found
that if the person had grown up in a positive emotional environment, the incidence of the usual
diseases of aging was 25%. If they had grown up in an adverse emotional climate, the incidence was
89%. The risk multiplier was an astounding 3.6. In other words, matters did not have to rise to the
level of discrete, countable adverse events to create medical risk. The Grant study results informed
our thinking from that time forward.
Results were similar for a Scandinavian study that evaluated the somatic health impact of
psychological stress. In tracking the increased incidence of disability-related pensions over a five-
year period starting at age 65, a dramatic correlation with early psychological stressors was noted. 3-
7 stress factors sufficed to yield a doubling of the incidence of disability-related pension over the
period. 9-12 stress factors yielded a risk multiplier of four. Overall, “over a quarter of… disability
pensions granted for somatic diagnoses could be attributed to psychological distress.” Even more
concerning, “…even mild psychological distress was associated with later onset of long-term
disability” (Rai et al, 2011).
The implication is that our general health status (mental and physical well-being) is closely
correlated with early emotional upbringing. Harry Harlow, the psychologist who conducted the
study in which rhesus monkeys were taken from their mothers to be nourished by wire surrogates,
encapsulated his results cogently: “Learning to love, like learning to walk and talk, can’t be put off
too long without crippling effects.” The infant learns to love by being loved—unconditionally. When
that does not occur, or it occurs inconsistently and unreliably, the infant’s orientation to safety and
threat is altered. This threat-consciousness is physiologically encoded, with impacts on arousal
regulation and the autonomic nervous system (Porges, 2011; van der Kolk, 1994). Tonic regulation of
state is fundamentally altered, in consequence of a shift toward over-arousal and even hyper-arousal.
Any tendency toward hyper-excitability—genetically rooted or trauma-induced—is exacerbated.
Preprints (www.preprints.org) | NOT PEER-REVIEWED | Posted: 17 October 2023 doi:10.20944/preprints202310.1085.v1
7
Altered tonic regulation of state thus predisposes to the emergence or compounding of cerebral
dysregulation.
The Dysregulation Cascade
Consider the finding that even a single change in domicile in a 14-year-old teenager doubles the
cumulative risk (to middle age) of attempted suicide, substance abuse, and violent offending. What
appears to be a minor traumatic episode may in fact be quite significant in some children’s lives. The
only reasonable explanation is that this apparently ‘minor’ trauma has outsize consequences for a
subset of the teenage population that is already at risk by virtue of their prior history (Webb et al,
2016).
Traumatic experiences therefore cannot be appraised in isolation. They are connected by what
we have termed the dysregulation cascade, in which the thread of continuity is provided by way of
neural network—and visceral/autonomic—encoding of the dysfunction. Vulnerability compounds
by virtue of repeat exposure before full recovery has been effectuated. A representative model for the
dysregulation cascade is the boxing match. Boxing is in essence an assault upon the functional
organization of the brain, one that allows little time for recovery between rounds. But substantial
recovery typically ensues between bouts, or else the boxer has a short career.
Post-traumatic stress disorder (PTSD) is also subject to the dysregulation cascade. Most cases of
Vietnam era PTSD resolved without resort to professional care. (PTSD was not even formally
recognized until the eighties.) Those cases that did not resolve were largely refractory to conventional
treatment. These cases were likely rooted in early trauma. This pattern became very clear with the
application of ILF neurofeedback to combat-related PTSD during the recent wars. The population
divided into ‘fast responders’ who recovered quite quickly in the training (twenty sessions or less), a
larger category that recovered in forty sessions, and a substantial cohort who failed to recover fully,
or required many more sessions, or both. Members of the latter group were much more likely to
report early life crises in their history. They are recognized in the European ICD formalism as cases
of Complex PTSD (WHO, 2022).
This discussion must also include the topic of minor traumatic brain injury (mTBI) and its
chronic sequelae, an issue that has been as much neglected in medicine as emotional trauma. There
is nothing minor about the chronicity of mTBI (Herrera-Escobar and Schneider, 2022). The label was
introduced in wartime for any head injury that did not involve skull fracture or a penetrating wound.
One major thrust in the early days of EEG neurofeedback was working with the persistent deficits of
mTBI, which were commonplace in the population. The deficits are largely remediable with
neurofeedback, even years after injury. The capacity for recovery is not lost. Left unremediated,
however, mTBI is yet another contributor to the dysregulation cascade. Significantly, it is being
recognized that a succession of minor brain insults (such as headers in soccer) can eventuate in
diagnosable mTBI if the incidence is sustained at a sufficiently high level (Lipton et al, 2013). Head
hits are a predominant risk factor as well for Chronic Toxic Encephalopathy (Daneshvar et al, 2023).
In every brain at risk, there exists an ongoing contention between such episodic functional
decrements and the available resources of recovery.
As in the case of minor brain insults, physical and emotional, there are also cumulative effects
from the burdens of low socio-economic status (SES). A recent study quantified the impact. Low SES
was shown to involve a hazard ratio of 2 for cerebral hemorrhage, and of about 1.7 for obesity, self-
harm, poisoning (i.e., substance abuse), and psychotic disorders. The authors then showed that
starting with the psychiatric conditions of self-harm, substance abuse and psychotic disorder a
cascade of some sixteen medical conditions of increasing mortality risk could be identified: liver and
renal disease, ischemic heart disease, cerebral infarction, chronic obstructive bronchitis, lung cancer,
and dementia (Kivimäki et al, 2020). A thread of continuity exists that begins with deficiencies in
early attachment or developmental trauma, compounds with adverse life events, and eventuates in
chronic medical disease and foreshortened life span. The increased mortality occasioned by poverty
has been tracked prospective study that is now in its fifth decade. Recently published results revealed
Preprints (www.preprints.org) | NOT PEER-REVIEWED | Posted: 17 October 2023 doi:10.20944/preprints202310.1085.v1
8
an 80% elevation of mortality risk by age 53 among those raised in poverty and in circumstances of
either crowded housing or of parental separation (Yu et al, 2022).
The Causal Chain
That brings us, then, to the question of mechanisms. This topic has been extensively explored by
Martin Teicher and colleagues at Harvard. The term ‘maltreatment trauma,’ which includes both
overt mistreatment and abject neglect, physical as well as sexual and emotional abuse, is applied to
this population. The impact is so substantial that in the characterization of mental disorders, those
with a maltreatment history may well constitute a distinct ecophenotype (Teicher and Samson, 2013).
Evidence for altered brain development already exists (Teicher et al, 2016). Of particular interest
to us is evidence for altered functional connectivity, as illustrated in Figure 2. Here principal
connectivity relationships are discerned with respect to key hubs of interest. In unexposed brains one
sees a healthy network under the aegis of the left anterior cingulate, directing a confident interaction
with the outside world. By contrast, in the exposed brain the well-elaborated network informs the
right anterior insula, which has primary responsibility for securing personal safety, in collaboration
with the right precuneus. At the same time, the left pre-frontal circuitry is relatively impoverished in
comparison with unexposed individuals. The trauma history has fundamentally altered both how
the brain engages with the outside world and how it manages its internal regulatory regime to assure
its own safety (Teicher et al, 2014).
Figure 2. Network differences are identified for maltreated versus unexposed young adults. The
circled dots indicate key regions of interest: the left anterior cingulate cortex, the right anterior insula,
and the right precuneus. Connectivity analysis yielded the primary linked nodes that were found to
be common within each cohort. These are nodes with direct connections to the region of interest
Jointly these linkages determine the relative importance of the three regions of interest for the two
cohorts. (Source: Teicher et al, 2016; Original source: Teicher et al, 2014.).
Preprints (www.preprints.org) | NOT PEER-REVIEWED | Posted: 17 October 2023 doi:10.20944/preprints202310.1085.v1
9
The differences shown in Figure 2 lie in the functional domain, and are therefore accessible to a
training strategy, at least in principle. However, in both cases we are seeing patterns of connectivity
that are congruent with the person’s life experience. What is the impetus for change? We know for
example from cases of chronic anxiety that, despite the associated discomforts, the state of anxiety
may come to confer a certain sense of security to the patient and therefore may not be so readily
relinquished. The same may hold true for over-arousal and hyper-vigilance in a person with combat-
related PTSD. These may well continue to be experienced as the life preservers they once were. An
effective therapeutic strategy would need to offer the trainee the visceral experience of calmness in a
safe context to effect a global physiological ‘reset.’ The approach of endogenous neuromodulation,
which goes to the foundations of the regulatory regime that took shape originally in an environment
of threat, appears to accomplish this, based on tracked symptom changes over the course of ILF
treatment (Othmer et al, 2011; Kirk and Dahl, 2022; Spreyermann, 2022)
Based on the above, we can reasonably argue that mental dysfunctions lie in the causal chain of
chronic medical disease, and that early childhood trauma accounts for the relatively intractable
portion of the healthcare burden. One might invoke Pareto’s Law on this point, conjecturing that
some 75% of medical costs may be attributable to nominally 25% of the population, and that this
cohort consists largely of those who endured early childhood trauma or physical brain trauma. Quite
literally, there is no health without mental health. It follows that a comprehensive strategy to address
the risk of chronic medical illness in a prevention model involves utilizing effective methods of
resolving these early emotional or physical brain traumas.
The Regulatory Hierarchy as Therapeutic Hierarchy
Working at extremely low frequencies enables us to engage the foundations of the regulatory
hierarchy in three aspects: 1) the developmental hierarchy (arousal and affect regulation); 2) the
functional hierarchy (from the more general to the more specific; from the more distributed to the
more localized; from the more contextual to the more specifically task-oriented); and 3) the hierarchy
in the frequency domain. In the latter, the lower frequencies establish the conditions for the dynamics
unfolding at higher frequencies. This hierarchy extends into the gamma range of frequencies, and as
low as we can reliably measure into the low-frequency domain. The implication of our clinical success
is that the intrinsic connectivity networks are sufficiently plastic so that re-normalization of function
can be mediated by way of ILF Neurofeedback, despite their being rooted in early development. The
above insights have been consolidated through clinical work over the past sixteen years. The relevant
citations are covered below in the section on Results.
The Method
The training strategy involves sequentially addressing the core regulatory domains of 1) cerebral
stability, 2) arousal regulation, 3) autonomic regulation, 4) affect regulation, and 5) executive
function. A complementary target is the left parietal heteromodal cortex. We review each in turn.
Cerebral Stability
The fundamental burden of any self-organizing control system is the maintenance of
unconditional stability, sufficient to sustain basic functionality. Brain instabilities violating this
criterion include seizures, migraines, panic attacks, asthmatic episodes, syncope, vertigo, sleep apnea,
narcoleptic events, and suicidal episodes. The ability to address this disparate range of instabilities
with ILF NF supports the view that it constitutes a comprehensive approach to the improvement of
cerebral self-regulatory competence. Since the presence of such instabilities interferes with the
subtlety that characterizes good regulation, addressing them must constitute the primary objective
of a training strategy toward functional normalization.
The management of seizure susceptibility with neurofeedback has more solid literature support
than prevails for any other condition (Sterman and Egner, 2006; Egner and Sterman, 2006). Excellent
results have also been reported for the remediation of migraine risk in a clinical cohort that had been
Preprints (www.preprints.org) | NOT PEER-REVIEWED | Posted: 17 October 2023 doi:10.20944/preprints202310.1085.v1
10
treated to medical standards (Walker, 2011). With a quantitative EEG (qEEG )-based protocol, more
than 50% of trainees ended up migraine-free for the one-year follow-up, versus none among those
who relied on medication alone. Another 40% improved by more than 50% in incidence, versus 9%
for the medication-only group. Only 2% failed to improve incidence significantly.
These excellent results notwithstanding, a significant step forward was taken with the
realization that the classic brain instabilities tend to respond best to inter-hemispheric placements, as
shown in Figure 3a. This discovery was made initially in working with migraines. With lateralized
placement, a lateralized migraine may well just escape to the other hemisphere; targeting it there just
brings it back. With inter-hemispheric placement, no such escape is available, so the migraine
vanishes. Typically, this occurs within session, and often within just a few minutes. In yet other cases,
the migraine is set on a path to resolution that takes a little longer. With extended training with the
same protocol, migraine incidence typically subsides substantially, paralleling Walker’s findings. An
outcome study on migraine headache using this method has just been published (Legarda et al, 2022)
Figure 3. Principal protocols explored over the course of development. 2a) Inter-hemispheric
placements at homotopic sites; 2b) Lateralized placements with T3 or T4 in common; 2c) Principal
protocols employed in ILF NF.
Reliance on inter-hemispheric placement goes back to the early seventies, when Douglas Quirk
used SMR reinforcement in bipolar montage at {C3 – C4} with violent offenders at the Ontario
Correctional Institute in Canada. In combination with galvanic skin response biofeedback, he was
able to reduce recidivism in selected violent offenders by a factor of more than three (from 65% to
20%, as documented in three-year follow-up) with 30 sessions of training. His work was the first
application of inter-hemispheric training to a cerebral instability, as well as the first large-scale clinical
utilization of NF (Quirk, 1995). Quirk claimed to have worked successfully with more than 2700
inmates.
Once the impact of inter-hemispheric placements was recognized for migraines, it did not take
long to establish that the same held true for brain instabilities in considerable generality. A key
cerebral vulnerability appears to lie in the coordination between the two hemispheres. This
coordination is managed at such a subtle level that a prescriptive training strategy is likely
unavailing. Here the approach of endogenous neuromodulation is obligatory—both for its capacity
for refined individualization and for its preferential access to the deep infra-low frequency domain.
To re-establish control, the brain merely needs information about its own functioning; it is not in need
of instruction.
Consider, in this regard, the attempt by Gruzelier et al to shift hemispheric negativity in
schizophrenia that was referred to previously (Gruzelier et al, 1999). If this procedure were adopted
for bipolar disorder, the brain would be driven into mania in one case, and into depression in the
other. The balance point would remain elusive to any prescriptive type of training. With endogenous
neuromodulation, the balance point can be located by fine adjustment of the training frequency.
Preprints (www.preprints.org) | NOT PEER-REVIEWED | Posted: 17 October 2023 doi:10.20944/preprints202310.1085.v1
11
Training too high takes the brain in a manic direction; training too low takes it toward the depressive
state. This can take place in a matter of minutes in rapid responders. Training at the ORF serves to
reinforce the network status that maintains stability against excursions into either mania or
depression. In practice, the search of the optimal response frequency (ORF) might take a few sessions.
The process is illustrated in Figure 4. Positive attributes can be combined to yield an Index of
Functionality; adverse observables can be combined to yield an Index of Dysregulation. An optimum
in both parameters is achievable at the optimal response frequency, and the therapeutic journey
typically begins with the search for the ORF, particularly in highly symptomatic, and thus highly
reactive/responsive clients.
Figure 4. Both positive and negative features contribute to identifying the Optimal Response
Frequency in a trainee. These can be grouped into an Index of Dysregulation and an Index of
Functionality. Typically, the process is less formalized: a fluid and dynamic interactive process that
moves the client over the behavior surface, gradually converging on the ORF.
When the brain is exposed to information relative to its state, just acting upon that information
accelerates its journey through state space. The clinician adjusts the available parameters—placement
and target frequency—to move the process toward its most propitious outcome. In a highly
symptomatic individual, the journey to identify the ORF may well encounter adverse attractors in
the attractor landscape (Freeman, 2006), which would call for rapid accommodation on the part of
the therapist. There are compensations: the highly symptomatic individual is likely to be very
sensitive to parameter change, thus facilitating the optimization procedure by way of good reporting.
Conversely, when the trainee is less parametrically sensitive, then training at the ORF is also less
critical.
In the search for a client’s optimal training parameters throughout the training process, we are
compelled to operate clinically with what is termed “ipsative trend analysis,” i.e., the discernment of
change induced via the training process in our clinical observables (Ulrich, 2020). In this process, the
client’s report on subjective experience of change is heavily relied upon—if the client is capable of
reporting, and particularly if the client is symptomatic. Symptom changes within session are the
primary drivers of the optimization procedure. In the absence of symptoms that can be tracked within
session, the process relies on state shifts with respect to issues on which the trainee can report:
alertness, calmness, emotional ambience, etc. The objective is to reach the state at which the client
feels maximally alert, calm, and euthymic. That specifies the conditions under which improved self-
regulatory competence is achieved most efficiently in terms of both target frequency and placement.
The conventional measures used in biofeedback can also be helpful here: galvanic skin response,
finger temperature, heart rate and its variability, as well as scalp muscle tension. Over the course of
Preprints (www.preprints.org) | NOT PEER-REVIEWED | Posted: 17 October 2023 doi:10.20944/preprints202310.1085.v1
12
sessions, additional information is recruited regarding the quality of sleep, and status with respect to
drives such a hunger, pain, and cravings, depending on the client’s issues.
The above approach was adopted in the late nineties, when we were still working exclusively in
the EEG spectral range. Here the ORF may have to be fine-tuned at the 1 – 3 % level. It may migrate
modestly—and slowly—over the course of sessions, or when other protocols are introduced, but we
regard it as a basically stable characteristic of a particular nervous system, at least over the time scales
relevant in therapy.
Given the centrality of brain stability in neuroregulation, the hypothesis that all our clinical
objectives might be met with inter-hemispheric training at various homotopic sites was evaluated in
the early 2000s, concurrent with the migration down to the lowest EEG frequencies (Othmer and
Othmer, 2005). Three placements turned out to be of greatest utility: {T3 - T4}, {P3 - P4,} and {Fp1 -
Fp2}. Frontal placement {F3 - F4} played a role mainly in the anxiety-depression spectrum.
Continuous performance test data, which we had been acquiring systematically on our clients since
1990, continued to yield satisfactory results (Putman et al, 2005). Nevertheless, over the course of two
years the addition of lateralized placements was indicated, as shown in Figure 3b. All our lateralized
montages had either T3 or T4 in common. Very quickly, {T4-P4} and {T4-Fp2} came to dominate in
the practice, indicating a right-side priority in the training—a complete reversal from the early days
of the field, when the focus of cognitive neuroscience was on left-hemisphere function almost
exclusively. At that time, right hemisphere function was still terra incognita.
What has survived the test of time in terms of principal training sites is shown in Figure 3c.
Observe that with the addition of lateralized protocols the inter-hemispheric training defaulted
largely to {T3 - T4}. This schema turns out to be independent of whether one is training in the EEG
band or at infra-low frequencies. The same networks are being targeted in essentially the same
manner; we are merely appealing to different regions of the frequency hierarchy. However, the
frequency range selected for training turns out to be determinative for outcomes.
Arousal Regulation
Beyond the primary challenge of assuring cerebral stability, the secondary burden of a self-
regulatory control system is to maintain the appropriate setpoint of activation for optimal
functionality. The foundational variable in this category is central arousal. This concept traces back
to the origins of the neurosciences. As far back as 1895, Freud and Breuer asserted that “a certain
measure of arousal exists in the conductive pathways of the resting, waking, engagement-capable
brain.” Since that time, the term arousal has typically referred to phasic arousal, the response to a
challenge, whereas our present usage concerns tonic arousal level, in line with Freud and Breuer.
Arousal is treated as a global state variable, one that characterizes brain state as a unitary entity. The
term activation then remains available to refer to more localized excitation, either within regions or
within functional domains. A contemporary perspective on arousal is given by Ross and Van
Bockstaele (2021). The intimate connection of arousal level with infra-slow fluctuations is reinforced
in a recent report by Sihn and Kim (2022).
The whole-brain character of arousal level could serve only a heuristic function for Freud and
Breuer, and the same holds true now. We don’t have a good measure. Nevertheless, within the frame
of the regulatory hierarchy in which we operate, the training of arousal regulation is the second-
highest priority after cerebral stability. One placement has dominated in this task throughout much
of our history: {T4 - P4}. A profound and global sense of calm—possibly even unique in the trainee’s
life experience—can descend upon a person even within a single session with this protocol. In most
cases, of course, we labor for such gains over several sessions. Quite commonly, both cerebral stability
and arousal regulation must be addressed within the very first session.
Autonomic Regulation
The next priority in the regulation of core states is the autonomic nervous system (ANS). Many
instabilities involve autonomic dysregulation as the primary issue: asthma, panic, narcolepsy, sleep
apnea, and Postural Orthostatic Tachycardia Syndrome (POTS). Much autonomic dysregulation is
Preprints (www.preprints.org) | NOT PEER-REVIEWED | Posted: 17 October 2023 doi:10.20944/preprints202310.1085.v1
13
involved in migraine, particularly cluster headaches. But we have already noted that all these
instabilities respond to essentially the same protocol, {T3 - T4}, so this does not further complicate the
agenda. As for the critical issue of the tonic relationship of the sympathetic and parasympathetic arms
of the ANS, {T4 - P4} training plays the dual role of calming over-arousal and of promoting
parasympathetic activation. Thus, with the two protocols discussed, which jointly serve as our
standard starting protocols, the issue of autonomic regulation is well covered. We have here an
approach that can be broadly helpful with issues of dysautonomia. (It remains true, however, that
the traditional biofeedback techniques retain their utility in the refinement of regulatory competence.)
Affect Regulation
The next protocol priority is affect regulation, for which we draw on right-prefrontal training at
{T4 - Fp2}. This is never a starting point, even if affect regulation is the primary clinical objective. The
clinician is obligated to ascertain how a client responds to the two starting protocols before venturing
to train the affective domain. It is difficult to generalize about this placement because ‘affect
regulation’ covers a lot of ground. The new placement is integrated into the training protocol, in
which every placement that continues to be useful is retained.
Executive Function
Executive function is targeted with {T3 - Fp1}, and as with affect regulation, a foundation must
first be laid with the starting protocols. Training is never begun with this placement, and in many
cases optimal training parameters are difficult to establish in the ILF. Practitioners may then resort to
legacy protocols in the EEG range.
Left Parietal Heteromodal Cortex
Left parietal training at {T3 - P3} can be helpful in the processing of detail. It is therefore relevant
to the resolution of learning disabilities (dyslexia, dyscalculia, and dysgraphia). It is usually the last
to be introduced into the training schema, as the groundwork needs to be laid for best outcomes.
Collectively, these five constitute the primary protocols utilized in the ILF regime. They are illustrated
in Figure 3c. They respect the principal cortical divisions, hemispheric and front-back.
Synchrony Training in the ILF Regime
Paradoxically, the midline has been left out of the picture. There is one additional ILF protocol
that must be mentioned. We often find it beneficial—and sometimes essential—to also train the
coordination of the frontal and parietal hubs of the Default Mode Network with an ILF Synchrony
protocol that promotes comodulation of the key frequencies in the 0.01 to 0.1 Hz spectral range. This
calls for midline placement (referential) at frontal and parietal sites, in sum-of-channels
configuration, {AFz} + {Pz}, to promote the common-mode component. The target frequency can also
be critical in this protocol. Most clients never move beyond these primary placements in their ILF
training. A fuller complement of protocols is covered in the Protocol Guide, which is currently in its
Seventh Edition (Othmer, 2019).
The Frequency Domain
Extension of Training into the Infra-Low Frequency Domain
The evolution of frequency-based training through its critical transition into the infra-low
frequency domain has had little representation in the literature to date. A case study on anxiety and
fear in a cancer patient demonstrates that training could be done productively even in the delta and
theta bands via endogenous neuromodulation (Benioudakis et al, 2016). A study on Complex
Regional Pain Syndrome captured the state of the art just prior to entry into the ILF regime (Jensen
MP et at, 2007). The larger scope of NF in application to chronic pain is covered in a book chapter
(Othmer and Othmer, 2006).
Preprints (www.preprints.org) | NOT PEER-REVIEWED | Posted: 17 October 2023 doi:10.20944/preprints202310.1085.v1
14
With availability of suitable software at our clinic in 2006, the training frequency could be
extended to a bandpass of 0.1 Hz, down from its prior limit of 1.5 Hz (0-3Hz bandpass) of the
NeuroCybernetics system. Within the first six months, some two-thirds of our clients ended up
training at that lowest frequency, a compelling argument to extend the frequency range further. That,
however, required new instrumentation design (Cygnet, Bee Group, Switzerland,). 10 mHz in corner
frequency was reached in 2008 with the new instrument. However, the pattern of cases piling up at
the lowest frequency just repeated. Within just six months, the training band was extended down to
1mHz.
Every step down in frequency presented a somewhat greater challenge to signal processing, and
a certain novelty to the clinician. Over time, however, the pattern repeated, with some two-thirds
training at the lowest frequency, so the range was extended to 0.1mHz in 2010. This once again
pushed the limits of the available architecture, which called for yet another upgrade of the signal
acquisition system, both hardware and software. This became available in 2013. By 2015, a repeat of
the above pattern forced another extension of the range to 10 µHz (microHz). By 2017 we were
working down to 1 µHz, and in 2019 the range was extended to 0.1 µHz. At the beginning of 2022,
training in the 0.01 µHz range became available—and almost immediately became indispensable.
With every downward step, our clinical reach extended to some people we had not been able to
help adequately before. Despite the increasing technical challenges as we went lower, the essential
experience of the training—in the perspective of the trainee—was very similar across the entire range.
In particular, the response to alteration of the training frequency was just as prompt at low
frequencies as at higher ones. The brain is of course keying on the dynamics prevailing at the target
frequency; it is not aware of the target frequency itself, which disappears entirely from view. Even
signal extraction deep in the ILF spectrum must reflect real-time dynamics or it would be irrelevant
to our project. The observable dynamics at issue lie in the spectral regime covered in fMRI
measurements, 0.1-200 mHz, and that holds for all target frequencies that lie below 0.1mHz.
Optimal Response Frequency: A Resonance Phenomenon
With each highly responsive client, an optimal response frequency can be found at which the
training is clearly ‘better’ than at neighboring frequencies, both above and below. We are still far from
an understanding the neurophysiological basis of what we are dealing with here, but the clinical
realities are difficult to dispute. This behavior exhibits the characteristic of a resonance phenomenon,
as illustrated in Figure 5.
To explain our observations, resonance is best viewed in the complex plane (Feynman, 1972).
The real axis yields the magnitude of the response, which peaks sharply at the resonance frequency.
It is described by the parameter Q, the ratio of center frequency to width of the curve. Q falls in the
range of 10-20, which serves to confirm the resonance model. The imaginary axis yields the phase
response. The resulting system response differs above and below the ORF in a phase-sensitive
manner. While the phase response is complex and may be difficult for the client to describe, it is
generally perceived as adverse in both cases (Othmer, 2009). That further supports the resonance
model.
Figure 5. The physiological response of a dysregulated system near the optimal response frequency
(ORF) is illustrated. a) The magnitude of the response, the responsivity, manifests as a narrow peak
at the ORF with respect to neighboring frequencies, the core characteristic of resonance. b) The phase
Preprints (www.preprints.org) | NOT PEER-REVIEWED | Posted: 17 October 2023 doi:10.20944/preprints202310.1085.v1
15
response of a resonant system is illustrated for reference (A). The physiological response exhibits a
phase-sensitive aspect that differs above and below the resonance frequency, and is restricted to the
immediate neighborhood of the ORF, where phase depends strongly on frequency, as shown in (B).
c) The net physiological response as training frequency is migrated over the narrow range is
illustrated. The optimum response is bracketed above and below by adverse tendencies. Shown are
three levels of responsivity (positive) and reactivity (negative), which correlate with dysregulation
status.
The strong frequency dependence becomes observable by virtue of the dysregulation status of
the trainee. The variables at issue here are multiple, consisting of all that are being tracked through
the training as indices to the physiological state of each individual. At the resonance frequency, the
functional status is optimized, the dysregulation status is minimized (given the prevailing
constraints), and the responsivity, the apparent rate of progress in training, appears to be optimized.
Clients have sometimes referred to this as the “sweet spot.” The more severe the dysregulation status,
the greater is the necessity to train at the ORF. Fortunately, this covaries with the ability to
discriminate it, as indicated in Figure 5c.
For example, it was observed early on that a migraine aura that emerged in a particular client
with training at 3.9 Hz could be disrupted by moving to 4.0 Hz. However, with return to 3.9 Hz the
aura would be rekindled. Movement back to 4.0 would cause the aura to subside once again. This
phenomenon was reproducible, all in the course of a single session. Such a sharp frequency
dependence is characteristic of the resonance response, in contrast to the gentler dependence on
frequency observed elsewhere across the spectrum. This kind of evidence established the case for the
resonance model for the ORF originally in the EEG spectral range some 25 years ago.
The stability of the ORF over the course of sessions—it tends to migrate only incrementally and
slowly with a given protocol—argues for the hypothesis that the brain organizes the frequency
domain around these key frequencies, which are dispersed throughout the ILF and EEG ranges.
Indeed, such a frequency hierarchy has already been conjectured (Gollo et al, 2015). Whereas this is
a very congenial hypothesis, one is nevertheless struck by just how deep into the ILF these
characteristic frequencies go. But then we also know that bipolar and depressive cycles can extend
over many months in a quasi-periodic manner, and that could be reflective of one of these deeper
rhythms.
The Frequency Rules
The frequency rules that apply to ORFs in lateralized placements are shown in Figure 6 (Othmer
et al, 2013; Othmer and Othmer, 2016; Othmer and Othmer, 2017). They divide into a high-
frequency and a low-frequency domain and are observed to be consistent within their respective
regions. Both a linear and a logarithmic plot illustrate the frequency relationships, which are
arithmetic in the EEG range and geometric in the ILF range. The left hemisphere optimizes at 2 Hz
higher than the right and optimizes at a factor of two higher in the ILF range. The Delta band is the
transition region in the two domains, as the two criteria coincide at a right-hemisphere frequency of
2 Hz and a left-hemisphere frequency of 4 Hz (i.e., 2 x 2 = 2 + 2).
Different rules apply to inter-hemispheric placements. This is easily said, but it was a long
process to tease them out. That was not the case for the lateralized placements, where the legacy
protocols of “C3-beta and C4-SMR” differed by 3 Hz, and it was merely a matter of systematic trial
to establish that the optimal difference was really 2 Hz rather than 3. For the inter-hemispheric
placements, it was noted that the cortical resting rhythms differed by nominally 4 Hz between the
posterior sensory regime (i.e., 10 Hz) and the central somatomotor regime (~14 Hz). Thus, it was
found that posterior placements optimize at 4 Hz lower than central (and a factor of four lower in the
ILF), except for posterior temporal sites (T5/T6), which train identically to T3/T4. Frontal placements,
in turn, optimize at 2 Hz lower than central placements (a factor of two lower in the ILF). In actual
practice, with most clients optimizing at the lowest available frequency in the ILF so much of the
Preprints (www.preprints.org) | NOT PEER-REVIEWED | Posted: 17 October 2023 doi:10.20944/preprints202310.1085.v1
16
time, and with T3-T4 having such a broad clinical footprint, the opportunity to probe other inter-
hemispheric placements at yet lower frequencies has been limited until recently.
Figure 6. The frequency rules that apply to ORFs in lateralized placements are illustrated. They divide
between the EEG range and the ILF but remain consistent within each. The transition region is the
upper delta band. The relationship is arithmetic in the EEG range, and geometric in the ILF—as
indeed must be the case.
It is appropriate at this point for the authors to acknowledge that some of what has been
presented no doubt strains credulity on first exposure. We have had the benefit of years to do reality-
testing, progressing incrementally by way of Bayesian inference, cumulatively
confirming/disconfirming provisional hypotheses. The entire construct rests substantially on ‘soft
data,’ namely the subjective reporting of highly symptomatic clients with respect to categories that
are not readily quantifiable—and inherently variable. Particular observations are often non-
repeatable, as the state vector is itinerant. And yet what has emerged is the hardest of testable
hypotheses: the frequency rules. They are constantly under test within the global practitioner
network. The existence of the frequency rules validates the ORF principle, and the existence of the
ORF in turn validates the approach for elucidating it. There is no alternative to the within-subject
design for proving out these hypotheses. Every clinical case is its own A/B design, with an ongoing
protocol contingency at every juncture.
Mechanisms Implications
The most striking feature of this entire development is the long-term trend toward ever lower
target frequencies. This pattern is driven by the issue of arousal regulation. Those whose training
begins with the {T4 - P4} placement exclusively tend to train at the lowest frequencies. By contrast,
those whose training begins with the inter-hemispheric protocol exclusively have their target
frequencies much more broadly distributed. Over the course of training, most clients (>90%) are likely
to experience both starting protocols. And it is found that in such cases the inter-hemispheric
placement is more tightly constrained in terms of ORF, as is to be expected for cerebral instabilities.
In generality, all right-lateralized placements train at the same ORF as inter-hemispheric placement
at {T3 - T4}. This fact alone serves to establish the primacy of {T3 - T4} among alternative inter-
hemispheric placements.
The implication of the above divergence in spectral distributions is that our two principal failure
modes arise from different sources. Arousal dysregulation tends to be the result of environmental
insult or persistent duress, whereas brain instabilities tend to have a genetic foundation, one that
promotes the vulnerability to hyper-excitability. The training of both failure modes is favored at the
ORF, as the frequency domain organization imposes its own constraints.
Preprints (www.preprints.org) | NOT PEER-REVIEWED | Posted: 17 October 2023 doi:10.20944/preprints202310.1085.v1
17
Another striking consistency is the right-side dominance that has emerged with training in the
ILF regime, with the combination of T4-P4 and T4-Fp2. This has all been empirically driven, with the
brain effectively in a controlling role with respect to protocol priorities. However, it is consistent with
the focus on early developmental priorities. Right-side function is the first to mature in infancy and
early childhood (Chiron, 1997; Gao, 2014). More broadly, we observe that the two principal failure
modes that have been identified are not just relevant to those afflicted with early trauma, or are
otherwise highly deficited, but to our clinical population in general. Our two starting protocols are
essentially universally applicable, which implies that we are all subject to the identified key
vulnerabilities. Right-hemisphere function is therefore our primary clinical target across almost the
entire range of mental-health related dysfunctions that we encounter, and across our entire clinical
population, with only rare exceptions, if any.
The Neurofeedback Therapist’s ‘Systems Perspective’ and its Foundations
The ‘systems perspective’, from the standpoint of a neurofeedback therapist, is shown in Figure
7. We assert the primacy of the regulatory arc that begins with interoception, informs autonomic
regulation, and affect regulation, and ultimately governs the tonic ambient of central arousal. All
these core functions reside in a state of intimate co-regulation. They are under the primary
management of the right hemisphere, and this coordination is organized in the ILF regime. Personal
history stamps the proceedings via hippocampal, afferent vagal and cranial nerve pathways. The
history of traditional biofeedback has long testified to the intimate association of affect regulation
with the autonomic nervous system. The latter is trained in order to tame the former, as for example
in the management of anxiety.
The foundation for the above conception was laid years ago in the animal work of Nina
Aladjalova, whose book became available in the English language in 1964 (Aladjalova, 1964).
Aladjalova studied the infra-slow rhythmic potential oscillations (ISPOs) at great length. “A single
stimulation of the reticular formation immediately elicits an arousal reaction in the EEG of the cortex,
but has no effect on infraslow activity,” she writes. “This reaction is apparently regulated by the rapid
regulatory system. Stimulation of the ventromedial part of the hypothalamus…intensifies infraslow
cortical activity within 30-40 minutes. This reaction is presumably regulated by the slow regulatory
system.”
“…infraslow activity is intensified by certain actions after a long latency period, 30-100 and 120-
200 minutes later. We conjectured that this phenomenon reflects the activity of the slow control
system of the brain…not only to automatically adjust the system to keeping internal environment
constant but actively to establish a new level of activity” (Aladjalova, 1964)
Preprints (www.preprints.org) | NOT PEER-REVIEWED | Posted: 17 October 2023 doi:10.20944/preprints202310.1085.v1
18
Figure 7. The systems perspective of the ILF neurofeedback practitioner is illustrated. The therapeutic
priority lies with the regulatory arc of interoception, autonomic regulation, affect regulation, and tonic
central arousal. .
Even in an optimal functioning context, this remains the priority in training. We have been
engaged with the slow control system that Aladjalova first identified, and it has been more a matter
of homeodynamics than of homeostasis. This system is centrally regulated by the hypothalamus,
which governs our internal milieu—autonomic function, sleep-wake cycle, ultradian rhythms, etc.
The brainstem is of course central to arousal regulation via the ascending reticular activating system,
and as such constitutes the top of the cerebral regulatory hierarchy. Phasic arousal is subject to input
from the thalamocortical networks. The tonic level of arousal is in turn informed by a second
regulatory hierarchy, under hypothalamic control.
It is appropriate at this point to make the connection, as best we can, between the principal
protocols and the targeted core state variables. These protocols were already well-established before
our acquaintance with intrinsic connectivity networks (ICNs). Our working model in the late nineties
was that we were engaging the multi-modal association areas, the highest level of integration of our
sensorium. These are the last areas to mature in neural development, being associated with the
highest level of plasticity. According to a more recent study, these regions also exhibit the highest
connectivity gradient (Margulies et al, 2014). That is the equivalent of saying that these sites are the
most highly integrative in character.
In the perspective of the ICN model, we are training the sites where the Default Mode Network
is accessible to us at the cortical surface at lateralized sites. The two criteria are convergent. Further,
the multi-modal association areas serve as input to the salience network, so we are training the nexus
of the Sensorium, the Default Mode and the Salience network. Our early work in the ILF regime
was influenced by Buckner et al. (2008), which examined the connectivity relationships among the
hubs of the DMN. We’ve created our own graphical representation of the data presented there, and
this is shown in Figure 8. In addition to the general argument that it is most efficient to train the
relationships between the hubs, there is the subsidiary argument that one would like to train those
linkages that exhibit the highest connectivity. Those network linkages that the brain keeps under the
tightest control make for the most discriminating sources of information back to the brain. These
considerations further underpin the primacy of T3, T4, P3, and P4 in our protocols. Buckner’s data
also supports the case for right-side priority in the training. Observe that the connectivities are
generally larger on the right side, and that T4 is more intimately inter-connected than T3. This biases
us toward right hemisphere training.
Preprints (www.preprints.org) | NOT PEER-REVIEWED | Posted: 17 October 2023 doi:10.20944/preprints202310.1085.v1
19
Figure 8. Connectivity relationships are indicated for the hubs of the Default Mode Network that are
directly relevant to neurofeedback protocol decision-making (according to Buckner et al, 2008). A
right-hemisphere bias toward greater connectivities is observed. LTC: Lateral Temporal Cortex (T3
and T4); PCC: Posterior Cingulate Cortex; vMPFC: ventromedial pre-frontal cortex; dMPFC:
dorsomedial pre-frontal cortex; IPL: Inferior Parietal Lobule (P3 and P4). .
In order to illuminate the relationship between the two hemispheres, we draw on a seminal
paper that yields information flow among the principal hubs of the Default Mode Network. These
hubs were originally identified in the study of microstates by the Lehman group in Switzerland
(Lehmann et al, 1987). Each of the microstates is identified with one of four hubs of the DMN, three
in the posterior region and one with an anterior locus. The two principal hubs, anterior and posterior,
lie along the midline, and two posterior hubs are lateralized.
Information flow among these hubs has been determined by means of a measure of directed
coherence in the alpha and low beta bands, leading to the finding that information flow was dominant
from the left hemisphere to the right, as well as from the left to the midline hub, relative to the flows
the other way (Lehmann et al, 2014). The imbalance can be substantial. The clear implication is that
with respect to the regulatory role of the lower EEG bands, the left hemisphere is in a commanding
position with respect to the right hemisphere.
A division of responsibilities is indicated. We may infer from the above that a reciprocal
relationship exists in which the right hemisphere bears the primary burden of organizing our resting
states, and in that capacity also governs the left hemisphere. The left hemisphere, in turn, supervises
our engagement with the outside world, and in that role also governs right hemisphere function. The
ILF regime plays a primary role in organizing the resting state configuration, whereas the EEG regime
handles the complexity, the coordination, the immediacy, and the temporal precision required for
our interface with the outside world. The upper delta band falls in the middle ground, with right
hemisphere primacy extending up to 2 Hz, and left hemisphere primacy extending down to 4 Hz,
according to the frequency rules.
Vinod Menon proposed the triple network model of psychopathology in 2010 (preprint),
postulating that “aberrant organization and functioning of the Central Executive Network, the
Salience Network, and the Default Mode Network are prominent features of several major psychiatric
and neurological conditions (Menon, 2011). With our limited focus on remediation—as opposed to
phenomenology—it now appears that psychopathology is much more explicitly rooted in the Default
Mode Network and the Salience Network, and particularly in their tonic regulation within the ILF
regime. The Central Executive is essentially missing from the conversation. A right-hemisphere bias
has also been identified (Schore, 1997). This, then, defines our agenda with respect to the primary
regulatory arc in Figure 7.
Foundational Research
Basic research into ILF neurofeedback via endogenous neuromodulation has taken place to date
solely within Russia, where high-level interest in biofeedback has existed for decades. One of the
startling clinical observations has been the common experience of robust first-session effects in the
ILF training, although single-session effects on excitability have previously been reported in EEG-
band training as well (Ros et al, 2010). It was therefore of critical interest to establish whether such
first-session effects could be corroborated in physiological measurements. This challenge was
undertaken by Dr. Olga Dobrushina of the International Institute of Psychosomatic Health in
Moscow, with the collaboration of the Treatment and Rehabilitation Center of the Russian Federation,
also located in Moscow.
The specific objective was to identify the networks engaged in the process of covert
neurofeedback by distinguishing them from what transpires in sham neurofeedback. 52 healthy
volunteers were recruited to a single session of ILF NF under uniform conditions, and resting state
fMRI data were acquired immediately prior to and again following the session (Dobrushina et al,
2020). It is important to note that this study was in no way intended to serve as an efficacy study.
Preprints (www.preprints.org) | NOT PEER-REVIEWED | Posted: 17 October 2023 doi:10.20944/preprints202310.1085.v1
20
First of all, efficacy was no longer in question, and secondly, one would not let the question of efficacy
turn on a single session of training. The experiment was process-focused more than outcome-focused:
What changes are induced by virtue of the act of training on (possibly transient) network
organization?
Significant changes were observed in both groups, and there were systematic findings among
the members of each group, despite their heterogeneity. In the veridical training group, “increased
connectivity was observed through a network consisting of the right and left inferior lateral occipital
cortex, right dorsolateral prefrontal cortex and striatum nuclei.” The sham training group also
demonstrated systematic change; it differed significantly both from the pre-training condition as well
as from the veridical training cohort.
As a corollary finding, this major study contributes to the cumulative body of evidence that
sham neurofeedback is not to be considered a neutral control condition in neurofeedback. Under the
given blinded conditions, every participant entered the study with the assumption of undergoing an
active process. In the search for persistent correlations that will inevitably be mobilized during the
session, the brains in the veridical group experience closure and get to settle down to an actual
feedback process, whereas the brains in the sham group remain in perpetual search status, one that
is never graced with lasting success. This difference is sufficient to yield divergent results in imaging,
which testifies both to the salience of the feedback on ILF activity and to the consequential nature of
the sham training challenge.
At the Institute of the Human Brain in St. Petersburg, interest in neurofeedback has existed since
the nineties. Several studies have been done in recent years to characterize ILF NF. The first of these
involved the training of three cases of depression with 20 sessions of ILF NF. Based on three different
rating scales, an average improvement of nominally 80% in symptom severity was found, and this
finding held up upon one-year follow-up. Substantial changes were also observed in EEG spectral
parameters in each of the participants (Grin-Yatsenko et al, 2018).
The second initiative was a controlled study in which ILF NF was compared to Heart Rate
Variability (HRV) training as an active control condition (Grin-Yatsenko et al., 2020). The study was
performed on a non-clinical population, so the main objective was to track changes in the spectral
characteristics in the ILF regime. After completion of 20 training sessions, all nine trainees in the NF
group indicated general improvement in their perceived health status—decreased reactivity to
stressors; improved body and spatial awareness; and improved stability of mood. They also noticed
improvements in energy level and in cognitive performance. In the HRV group, six of eight reported
heightened stress tolerance and a greater ability to relax, while the remaining two could not identify
any notable change in their state.
The average changes observed in spectral power in the 0.01 to 0.5 Hz band for the two groups
were substantial for all standard 19 sites. All 9 trainees in the ILF NF group demonstrated significant
increase in spectral power. The outcomes for the HRV group were more variable, but on average
showed a decline in spectral power at all sites. The decline did not reach statistical significance.
Additionally, rhythmic oscillations in the 0.06 to 0.12 Hz band became more prominent in 8 of 9 NF
trainees. Power in that band increased in only 3 of 8 among the HRV controls, with either no change
or a decline in the others. Extended spectral analysis of the data has just been published, revealing
a systematic increase in a 0.02 Hz region along with the increased rhythmicity in the 0.08 Hz region
(Grin-Yatsenko et al., 2023).
Limited though they are, these data are suggestive of the proposition that ILF NF is effecting
readily observable—indeed macroscopic—changes in the frequency-based organization within the
upper ILF range, the range of interest in fMRI research. Even a modest recruitment of neuronal
populations from distributed activity into correlated activity within transient rhythmic oscillations—
infra-slow fluctuations—suffices to account for the substantial increase in spectral power. Tendencies
toward the expected power-law distribution are also discernible.
The theoretical aspects of neurofeedback have also attracted academic attention (Sitaram et al,
2017). Among the several models for neurofeedback, covert, closed-loop neurofeedback as well as
the skill learning model are discussed. The category of covert, closed-loop neurofeedback
Preprints (www.preprints.org) | NOT PEER-REVIEWED | Posted: 17 October 2023 doi:10.20944/preprints202310.1085.v1
21
encompasses endogenous neuromodulation. The feedback is continuous, and the training process
lies beneath awareness. The method at issue here can also be modeled as the enhancement of the
brain’s core skill of self-regulation. In fact, the method emulates how the brain does skill learning in
generality: It must observe itself performing the skill. There must be feedback on its performance.
Results
ILF neurofeedback has seen little representation in the clinical literature to date. There were three
early publications in the neurology literature. One paper focused on side effects of anti-epileptic
drugs (Legarda et al, 2011). Another focused on the method, with PTSD as a case example (Othmer
et al, 2011). A broader perspective on the clinical reach of ILF neurofeedback was offered in 2013
(Othmer et al. 2013). Here the use of the term endogenous neuromodulation to refer to the absence
of overt reinforcers was introduced—likely for the first time. A systematic mixed studies review has
just been published (Bazzana et al, 2022). It reports on 18 studies of ILF NF and the alternative of
reward-based training on Infra-Slow Fluctuations.
With respect to the issue of emotional trauma specifically, early results for combat-related PTSD
were reported by Othmer (2009). More recent publications of note include dramatic recoveries from
combat-related PTSD and TBI observed in a small pilot study (Carlson and Ross, 2021) and rapid
recovery from a case of complex PTSD (Gerge, 2021). Just published are four additional reports. One
is a controlled study of 36 patients suffering from an eating disorder and comorbid PTSD (Winkeler
et al, 2022). A second documents the impact of ILF NF on children under the care of the State, most
of whom have a trauma history (Fleischman, 2022). A third paper reflects on a multi-year history of
working with PTSD using ILF NF in combination with conventional therapies (Spreyermann, 2022).
A fourth reports on extended ILF NF training of a veteran with PTSD (Kirk and Dahl, 2022). The
broader clinical perspective on ILF NF—including PTSD—has been covered in the book titled
Restoring the Brain, which is presently in its second edition (Kirk, 2015, 2020).
Since ILF neurofeedback matured entirely in clinical practice, no formal efficacy studies have
been published to date. It’s a little late now, and furthermore, conventional group studies are
problematic for various reasons. The ‘special competence’ that may be claimed for endogenous
neuromodulation in the ILF regime is the exploitation of latent neuronal/glial plasticity in highly
impaired systems, particularly those in which the deficits are rooted in early development. Such
special competence is difficult to quantify, given the range of organicity prevailing in such highly
impaired neural systems, the individualization of training protocols, and the variance contributed by
clinician proficiency. Within the domain of clinical practice, ILF neurofeedback should not be
researched as a procedure at all, but rather as a therapy—by way of outcome studies. Foundational
mechanisms research is different matter.
The immediate target of ILF neurofeedback is enhanced self-regulatory competence (with
symptomatic relief and functional recovery the secondary consequence). We draw upon the
Continuous Performance Test, a pressured choice-reaction time test, to demonstrate that self-
regulatory competence can be systematically enhanced. This test is routinely given to clients seeking
ILF neurofeedback. It has been in continuous use in our practice since 1990. Whereas the original
purpose was to assess ADHD children, the test is now used to appraise performance relative to our
two core concerns, arousal regulation and cerebral stability. One may regard omission errors, for
example, as indexing subtle discontinuities of state (“attentional lapses”), one end of the instability
continuum. Commission errors, on the other hand, likely have an arousal level dependence.
For the above purposes we rely on the QIKtest (Bee Medic, Switzerland), which slightly adapted
the design of the Test of Variables of Attention (T.O.V.A. ®) (Leark et al., 2007) to a handheld device
that gave us 0.1 msec timing accuracy. Central features of the TOVA include an invariant inter-
stimulus interval of 2 sec to minimize novelty, and a split between a low-demand and high-demand
phase, in which the ratios of target to non-target are 2:7 and 7:2, respectively. The QIKtest
arranged for a return to a low-demand phase for a more complete picture. It also facilitated data
collection on a central server, which made it possible to derive population-based norms. These are
particularly necessary for the discrete errors, where the distributions are distinctly non-Gaussian.
Preprints (www.preprints.org) | NOT PEER-REVIEWED | Posted: 17 October 2023 doi:10.20944/preprints202310.1085.v1
22
Non-parametric norms were first established in 2013 based on some 50,000 records, which yielded
more than 500 samples for each of forty age and gender bins. These norms have subsequently been
refined.
Cumulative clinical results for the practitioner network at large are evaluated periodically, and
the results for errors of commission are shown in Figure 9 for the period 2018-2022. Plotted is the
incidence of cases for a given number of errors per test. In the highly deficited region, incidence of
cases is reduced by a factor of 2.5 after nominally twenty sessions of training. The plot demonstrates
that anyone sufficiently functional to take the test is quite likely also capable of improving their scores
with training. For errors of omission, the corresponding improvement factor was 1.9, and for reaction
time outliers, it was 1.4. Surveying the reaction time distribution with the advantage of large numbers
had made it clear that reaction time outliers had to be treated as a distinct entity, extending beyond
the exponential tail of the expected ex-Gaussian distribution (Leth-Steensen et al, 2000).
Figure 9. Pre-post distributions of commission errors are shown here for a non-selective sample of
15,260 cases (data from 2018-2022). Incidence is plotted for the entire range of number of errors per
test, out to the pathological extreme of 200 errors. Because of scatter, the data have been smoothed to
reveal the underlying trends, and the superimposed curves were by visual inspection. With nominally
twenty sessions of training, a reduction in incidence by a factor of 2.5 is indicated for the highly
deficited region.
The data are plotted as cumulative distributions in Figure 10. The change in median score can
be expressed in clinically meaningful terms by way of the norming data. The pre-training median
value of 10.4 commission errors translates to a normative score for 13.7-year-olds, and 4.8 errors to a
normative score for 20-year-olds. Such data are even more meaningful when related to age-
segregated cohorts. A convenient division is into a 6-9-year (inclusive), and a 10-19-year cohort.
Preprints (www.preprints.org) | NOT PEER-REVIEWED | Posted: 17 October 2023 doi:10.20944/preprints202310.1085.v1
23
Figure 10. Cumulative plot of commission error incidence from Figure 9. The median score before
training is 10.4 errors, versus 4.8 errors after nominally 20 sessions. Based on our norming data, this
translates to an improvement in effective mental maturity from 13.7 to 20 years.
With respect to impulsivity, the median mental age for the younger cohort is less than six years
of age at intake. As norms don’t extend to younger ages, a value of 6 years is assigned. After
nominally twenty sessions of training, the median mental age is 10 years. For the older cohort (10-
19), the median age is 9 years at intake, and after 20 sessions we obtain a score of 17 years. Observe
that after twenty sessions the younger cohort outperforms the older cohort at intake. Results for all
four primary measures of the CPT are shown in Table 1. Impulsivity is the quickest to respond, with
the median score rising to above normative values. This is followed by inattention (as indexed by
omission errors), for which the median reaches normative values. Mean reaction time and variability
in reaction time take longer to plateau. Those whose level of function is not normalized in twenty
sessions are well advised to continue the training.
Surprisingly, it was observed some years ago at the EEG Institute that outcomes for omission
and commission errors were not predictable on the basis of intake data. Shown in Figure 11 are the
results obtained on a subset of 87 trainees whose starting standard score was 85 or less (<16th
percentile) out of a total cohort of 350 successive data sets. Data are plotted rank-ordered by starting
value. When non-responders are excluded from the plot (i.e., those showing less than 5 pt.
improvement in standard score), it is found that the likelihood of reaching a score of 100 (i.e., zero
omission errors) is independent of the starting value, as shown in split-half comparison—at about
50%. It appears that initial test results tell us very little about the potential for functional recovery.
Only the training itself can disclose what is possible. Results for impulsivity show a similar tendency,
along with fewer non-responders (5% of those scoring < 85). These data show that remediation of the
cardinal symptoms of ADHD are achievable with a primary reliance on the two principal targets—
cerebral stability and arousal regulation—that typically dominate in the first 20 sessions of training.
Table 1. Changes in median mental age with nominally twenty sessions of ILF NF are shown for our
two age ranges. Data refer to the entire practitioner network that contributes to the database. .
Preprints (www.preprints.org) | NOT PEER-REVIEWED | Posted: 17 October 2023 doi:10.20944/preprints202310.1085.v1
24
Age Range
Category
Pre-
Post
Delta
6-9 years, inclusive
Impulse control
6
10
4.0
Inattention
6.2
8.5
2.3
Mean reaction time
7.3
8.0
0.7
Variability in reaction time
6.6
7.6
1.0
10-19 years, inclusive
Impulse control
9.0
17.0
8.0
Inattention
10.3
13.8
3.5
Mean reaction time
12.1
12.6
0.5
Variability in reaction time
10.4
11.8
1.4
Figure 11. Pre- and post-data are shown for omission errors for the 87 clients scoring 85 or less in
standard score at intake, out of a sample of 350 successive cases seen at the EEG Institute in Los
Angeles. The likelihood of scoring zero omission errors (i.e., standard score of 100) after nominally 20
sessions of training is independent of starting value. Non-responders were removed for clarity, as
they tend to cluster at the bottom. They numbered 27 in the cohort of 114 scoring < 85 (i.e., 24%,
amounting to 8% of the total sample of 350).
Conclusion
A method is described for the exploitation of endogenous neuromodulation to target the
foundations of our regulatory and developmental hierarchies, with a particular view to the
remediation of developmental deficits as well as the consequences of early emotional and physical
trauma. The objective is to forestall any subsequent dysregulation cascade early in the state of
formation, and thus lay the basis for constituting a resilient self. More generally, it is proposed that
the foundation of a better-regulated physiology reduces the severity and improves the manageability
of mental dysfunctions in considerable generality. For the population at large, this work should be
understood in the frame of optimal functioning.
These objectives drive a focus on the infra-low frequency domain, the foundation of the
frequency hierarchy, with a particular emphasis on right-hemisphere function and on inter-
hemispheric coordination. The training is ideally conducted in observance of the principle of optimal
response frequency, and in cognizance of established frequency rules that relate different protocols.
The remedy provided by way of endogenous neuromodulation is available at any age.
The substantial functional improvements that are available even for highly deficited individuals,
with modest investments of professional time, present us with the ethical mandate to make such
training available to children in need early in their development, as well as to adults enduring the
handicaps bequeathed to them from their early upbringing.
Preprints (www.preprints.org) | NOT PEER-REVIEWED | Posted: 17 October 2023 doi:10.20944/preprints202310.1085.v1
25
Author Contributions： Conceptualization, Susan Othmer; Investigation, Susan Othmer; Methodology, Susan
Othmer; Writing – original draft, Siegfried Othmer; Writing – review & editing, Siegfried Othmer.
Funding： Funding for the preparation and the publication of the manuscript is provided by the Anthony and
Jeanne Pritzker Family Foundation under a grant of funds to the Brian Othmer Foundation.
Institutional Review Board Statement：All of the authors’ findings, as reported here, were acquired in a clinical
setting over the course of many years, and none involved formal research studies that called for the involvement
of an IRB.
Informed Consent Statement： Informed consent was obtained on all our clients at intake, giving permission
for the utilization of their clinical data in an anonymized fashion.
Data Availability Statement： All the data presented here remain accessible on the original server.
Acknowledgments ： Since this manuscript recapitulates a 38-year development, it is appropriate to
acknowledge the engineers responsible for our three generations of neurofeedback instrumentation: Edward
Dillingham (NeuroCybernetics), Howard Lightstone (EEGer), and Bernhard Wandernoth (Cygnet). We
acknowledge Marco Versace for his development of the CPT analysis program, EEG Expert, and thank John
Putman for his analyses of CPT data. We also gratefully acknowledge funding support from the Anthony and
Jeanne Pritzker Family Foundation for this publication project. I thank Martha Herbert for her critical review of
the manuscript.
Conflict of Interest： The authors declare that the research was conducted in the absence of any commercial or
financial relationships that could be construed as a potential conflict of interest.
References
Aladjalova, N.A. (1964). Slow Electrical Processes in the Brain. Elsevier
Bagdasaryan, J. and Le Van Quyen, M. (2013). Experiencing your brain: neurofeedback as a new bridge between
neuroscience and phenomenology. Frontiers in Human Neuroscience, 7 (680), 1-10 doi:
10.3389/fnhum.2013.00680
Barnea, A., Rassis, A., Raz, A., Othmer, S., Zaidel, E. (2004). Effects of Neurofeedback on Hemispheric Attention
Networks, Brain and Cognition, (November 2004)
Bazzana F, Finzi S, Di Fini G, Veglia F (2022). Infra-Low Frequency Neurofeedback: a systematic mixed studies
review. Front. Hum. Neurosci. 16:920659. doi: 10.3389/fnhum.2022.920659
Beggs, J.M. and Plenz, D. (2003) Neuronal avalanches in neocortical circuits. J Neurosci, 23(35):11167-77.
Benioudakis, E.S., Kountzaki, S., Batzou, K., Markogiannaki, K., Seliniotaki, T., Darakis, E., Saridaki, M., Vergoti,
A., and Nestoros, J.N. (2016). Can Neurofeedback Decrease Anxiety and Fear in Cancer Patients, Postepy
Psychiatrii i Neurologii 25(1), 59-65 http://dx.doi.org/10.1016/j.pin.2015.12.001
Birbaumer, N., Elbert, T., Canavan, A.G.M., Rockstroh, B. (1990) Slow Potentials of the Cerebral Cortex and
Behavior, Physiological Reviews, 70(1), 1-41.
Birbaumer, N. (1999). Slow cortical potentials: Plasticity, operant control, and behavioral effects. The
Neuroscientist, 5, 74–78.
Brown, B.B. (1974). New Mind, New Body: Bio-Feedback: New Directions for the Mind, Harper and Row, New
York
Brown, D.W., Anda, R.F., Tiemeier, H., Felitti, V.J., & Edwards, V.J., Croft, J.B. Giles, W.H. (2009). Adverse
Childhood Experiences and the Risk of Premature Mortality, Am. J. Preventive Med. 37(5), 389-396
DOI: 10.1016/j.amepre.2009.06.021
Buckner, R.L., Andrews-Hanna, J.R., Schacter, D.L. (2008). The brain's default network: anatomy, function, and
relevance to disease. Ann N Y Acad Sci. 1124(1), 1-38
Buzsáki, G. and Watson, B.O. (2012). Brain rhythms and neural syntax: implications for efficient coding of
cognitive content and neuropsychiatric disease. Dialogues in Clinical Neuroscience, 14(4): 345-367
Carlson, J., and Ross, G. W. (2021). Neurofeedback impact on chronic headache, sleep, and attention disorders
experienced by veterans with mild traumatic brain injury: a pilot study. Biofeedback 49, 2–9. doi:
10.5298/1081-5937-49.01.01
Chiron, C., Jambaque, I., Nabbout, R., Lounes, R., Syrota, A., & Duklac, O. (1997). The right brain hemisphere is
dominant in human infants. Brain, 120, 1057-1065
Daneshvar, D.H., Nair, E.S., Baucom, Z.H. et al. (2023). Leveraging football accelerometer data to quantify
associations between repetitive head impacts and chronic traumatic encephalopathy in males. Nat Commun
14, 3470 (2023). doi.org/10.1038/s41467-023-39183-0
Dobrushina, O. R., Vlasova, R. M., Rumshiskaya, A. D., Litvinova, L. D., Mershina, E. A., Sinitsyn, V. E., et al.
(2020b). Modulation of Intrinsic Brain Connectivity by Implicit Electroencephalographic Neurofeedback.
Frontiers in Human Neuroscience 14. doi:10.3389/fnhum.2020.00192
Preprints (www.preprints.org) | NOT PEER-REVIEWED | Posted: 17 October 2023 doi:10.20944/preprints202310.1085.v1
26
Egner, T. and Gruzelier, J.H. (2001). Learned self-regulation of EEG frequency components affects attention and
event-related brain potentials in humans. Neuroreport 12(18): 4155-4159
Egner, T., Sterman, M.B. (2006). Neurofeedback treatment of epilepsy: from basic rationale to practical
application, Expert Rev. Neurotherapeutics 6(2), 247-257 doi:10.1586/14737175.6.2.247
Evans J.R. and Turner R., editors (2017). Rhythmic Stimulation Procedures in Neuromodulation, Elsevier,
London
Felitti VJ, Anda RJ, Nordenberg D, Williamson DF, Spitz AM, Edwards V, Marks JS (1998) Relationship of
childhood abuse and household dysfunction to many of the leading causes of death in adults. The adverse
childhood experiences (ACE) study. Am J Prevent Med 14:245-58.
Feynman, R.P. (1972). The Feynman Lectures on Physics, Vol. 1, Addison Wesley.
Internet-accessible: https://www.feynmanlectures.caltech.edu/I_23.html (accessed 12/27/22)
Fisher, S.F. (2014). Neurofeedback in the Treatment of Developmental Trauma: Calming the Fear-Driven Brain,
WW Norton, New York
Fleischman M (2022). Documenting the Impact of ILF Neurofeedback on Underserved Populations with
Complex Clinical Presentations. Front. Hum. Neurosci. 16:921491. doi: 10.3389/fnhum.2022.921491
Freeman, W.J. (2006). Definitions of state variables and state space for brain-computer interfaces, Cognitive
Neurodynamics 1(1): 3-14 http://dx.doi.org/10.1007/s11571-006-9001-x
Friston, K.J. (2010). The free-energy principle: a unified brain theory? Nat. Rev. Neurosci., 11, pp. 127-138
doi:10.1038/nrn2787
Fuchs, T., Birbaumer, N., Lutzenberger, W., Gruzelier, J.H. and Kaiser, J. (2003) Neurofeedback treatment for
attention-deficit/hyperactivity disorder in children: a comparison with methylphenidate. Appl.
Pychophysiol. Biofeedback, 28: 1–12.
Gao, W., Alcauter, S., Smith, J.K., Gilmore, J.H., Lin, W. (2014) Development of human brain cortical network
architecture during infancy. Brain Structure and Function, pp. 1-14 DOI: 10.1007/s00429-014-0710-3
Gerge, A. (2020). A multifaceted case-vignette integrating neurofeedback and EMDR in the treatment of complex
PTSD. European Journal of Trauma & Dissociation, 4(3), 100157. https://doi.org/10.1016/j.ejtd.2020.100157
Gevirtz, R. (2013). The promise of heart rate variability biofeedback: evidence-based applications. Biofeedback
41, 110–120. doi: 10.5298/1081-5937-41.3.01
Gollo, L.L., Zalesky, A., Hutchinson, R.M., van den Heuvel, M., Breakspear, M. (2015). Dwelling quietly in the
rich club: brain network determinants of slow cortical fluctuations. Phil. Trans. R. Soc. B 370:20140165
http://dx.doi.org/10.1098/rstb.2014.0165
Green, E., and Green, A. (1977) Beyond Biofeedback, Delacorte Press/Seymour Lawrence
Grin-Yatsenko, V., Othmer, S., Ponomarev, V.A., Evdokimov, S., Konoplev, Y., and Kropotov, J. D. (2018). Infra-
Low Frequency Neurofeedback in Depression: Three case studies. NeuroRegulation 5, 30–42.
doi:10.15540/nr.5.1.30
Grin-Yatsenko, V.A., Kara, O., Evdolimov, S.A., Gregory, M., Othmer, S., Kropotov, J.D. (2020), Infra-Low
Frequency Neurofeedback Modulates Infra-Slow Oscillations of Brain Potentials: A Controlled Study, J
Biomed Eng Res 4(104): 1-10
Grin-Yatsenko, V.A., Ponomarev, V.A., Kropotov, J. D. (2023). Changes of the Infra-Slow EEG Fluctuations of
the Brain Potentials under Influence of Infra-Low Frequency Neurofeedback, Journal of Evolutionary
Biochemistry and Physiology, 59(3), pp. 831–840. DOI:
10.1134/S002209302303016X
Gruzelier J, Hardman E, Wild J, et al (1999). Learned control of interhemispheric slow potential negativity in
schizophrenia. International Journal of Psychophysiology 1999;34:341-8
Gruzelier, J.H., Egner, T. (2005). Critical validation studies of neurofeedback, Child Adolesc Psychiatric Clin N
Am 14, 83-104 doi:10.1016/j.chc.2004.07.002
Gruzelier, J.H., Foks, M., Steffert, T., Chen, M.J.-L, Ros. T (2014). Beneficial outcome from EEG-neurofeedback
on creative music performance, attention and well-being in school children, Biological Psychology, 95: 86-
95, https://doi.org/10.1016/j.biopsycho.2013.04.005
He, B.J., Zempel, J.M., Snyder, A.Z., and Raichle, M.E. (2010). The temporal structures and functional significance
of scale-free brain activity, Neuron, 66(3): 353–369. doi:10.1016/j.neuron.2010.04.020.
Herrera-Escobar, J.P. and Schneider, J.C., (2022). From Survival to Survivorship—Framing Traumatic Injury as
a Chronic Condition, N Enlg J Med 387: 581-583
Hook, E.B., editor (2002) Prematurity in Scientific Discovery: On Resistance and Neglect, U. California Press
Jensen, M. P., Grierson, C., Tracy-Smith, V., Bacigalupi, S. C., & Othmer, S. (2007). Neurofeedback treatment for
pain associated with complex regional pain syndrome type I. Journal of Neurotherapy, 11(1), 45-53.
https://doi. org/10.1300/J184v11n01_04
Kaiser, D.A. and Othmer, S. (1997) Efficacy of SMR-Beta Neurofeedback on Attentional Processes, Biofeedback
and Self-Regulation (4):299-312
Kaiser, D.A., Othmer, S. (2000). Effect of Neurofeedback on Variables of Attention in a Large Multi-Center Trial,
Journal of Neurotherapy, 4(1), (2000), pp.5-15
Preprints (www.preprints.org) | NOT PEER-REVIEWED | Posted: 17 October 2023 doi:10.20944/preprints202310.1085.v1
27
Kirk HW (2015, 2020). Restoring the Brain, Neurofeedback as an Integrative Approach to Health. Taylor and
Francis, Boca Raton FL
Kirk HW and Dahl MG (2022) Infra Low Frequency Neurofeedback Training for Trauma Recovery: A Case
Report. Front. Hum. Neurosci. 16:905823. doi: 10.3389/fnhum.2022.905823
Kitzbichler, M.G., Smith, M.L., Christiansen, S.R., et al (2009) Broadband Criticality of Human Brain Network
Synchronization. PLOS Comput Biol, 5(3):e1000314. doi:10.1371/journal.pcbi.1000314
Kivimäki M, Batty GD, Pentti J, Shipley MJ, Sipilä PN, Nybert SJ, Suominen SB, Oksanen T, Stenholm S,
Virtanen M, Marmot MG, Singh-Manoux A., Brunner EJ , Lindbohm JV, Ferrie JE, Vahtera J (2020).
Association between socioeconomic status and the development of mental and physical health conditions
in adulthood: a multi-cohort study. The Lancet Public Health 5(3) E140-E149
https://doi.org/10.1016/S2468-2667(19)30248-8
Kotchoubey B, Strehl U, Uhlmann C, et al (2001). Modification of slow cortical potentials in patients with
refractory epilepsy: a controlled outcome study. Epilepsia 42:406-16.
Lantz D and Sterman MB (1988). Neuropsychological assessment of subjects with uncontrolled epilepsy: effects
of EEG biofeedback training. Epilepsia 29(2), 163-171.
Leark, R. A., Dupuy, M.S., Greenberg, L.M., Corman, C.L. and Kindschi, C.L. (2007). TOVA: Test of Variables of
Attention, Professional Guide. Los Alamitos, CA: Universal Attention Disorders Inc.
Lehmann, D., Ozaki, H., & Pal, I. (1987). EEG alpha map series: brain micro-states by space-oriented adaptive
segmentation. Electroencephalography and Clinical Neurophysiology, 67(3), 271–288
Lehmann, D., Pascual-Marqui, R.D., Milz, P., Kochi, K., Faber, P., Yoshimura, M., Kinoshita, T. (2014). The resting
microstate networks (RMN): cortical distributions, dynamics, and frequency specific information flow.
Available from: http://arxiv.org/abs/1411.1949.
Legarda, S. B., McMahon, D., Othmer, S., & Othmer, S. (2011). Clinical neurofeedback: Case studies, proposed
mechanism, and implications for pediatric neurology practice. Journal of Child Neurology, 26(8), 1045-
1051. http://dx.doi.org/10.1177/0883073811405052
Leth-Steensen, C., Elbaz, Z.K., and Douglas, V.I. (2000). Mean response times, variability, and skew in the
responding of ADHD children: a response time distributional approach, Acta Psychologica, 104, 167-190
Lipton, M. L., Kim, N., Zimmerman, M. E., Kim, M., Stewart, W. F., Branch, C. A., & Lipton, R. B. (2013). Soccer
heading is associated with white matter microstructural and cognitive abnormalities. Radiology, 268(3),
850–857. https://doi.org/10.1148/radiol.13130545
Lubar J. F., Shabsin, H. S., Natelson, S. E., Holder, G. S., Woodson, S. F., Pamplin, W. E., & Krulikowski, D. I.
(1981). EEG operant conditioning in intractable epileptics. Archives of Neurology, 38, 700-704.
Lubar, J. F. (1997). Neocortical Dynamics: Implications for Understanding the Role of Neurofeedback and
Related Techniques for the Enhancement of Attention. Applied Psychophysiology and Biofeedback, 22(2),
111-126
Margulies, D.S., Ghosh, S.S., Goulas, A., Falkiewicz, M., Huntenburg, J.M., Langs, G., Bezgin, G., Eickhoff,
S.B., Castellanos, F.X., Petrides, M., Jefferies, E., & Smallwood, J. (2016). Situating the default-mode
network along a principal gradient of macroscale cortical organization. Proceedings of the National
Academy of Sciences, 113 (44) 12574-12579 DOI:10.1073/pnas.1608282113
Menon, V. (2011). Large-scale brain network and psychopathology: a unifying triple network model. Trends
Cogn Sci, 10:483-506
Monto, S., Palva, S., Voipio, J., Palva J.M. (2008). Very Slow EEG Fluctuations Predict the Dynamics of Stimulus
Detection and Oscillation Amplitudes in Humans, J. Neuroscience, 28(33): 8268-8272
Othmer, S., Othmer, S.F., and Kaiser, D.A. (1999a). EEG Biofeedback: Training for AD/HD and Related
Disruptive Behavior Disorders, In Understanding, Diagnosing, and Treating AD/HD in Children and
Adolescents, An Integrative Approach, James A. Incorvaia, Bonnie S. Mark-Goldstein, and Donald
Tessmer, editors, Aronson Press, Northvale, NJ, pp.235-296
Othmer, S., Othmer, S.F., and Kaiser, D.A. (1999b). EEG Biofeedback: An Emerging Model for Its Global Efficacy,
In Introduction to Quantitative EEG and Neurofeedback, James R. Evans and Andrew Abarbanel, editors,
Academic Press, San Diego, pp. 243-310
Othmer, S., Othmer, S. (2006). Efficacy of neurofeedback for pain management. In: Boswell M, Cole BE, eds.
Weiner’s Pain Management: A Practical Guide for Clinicians. 7th ed. CRC Press; 719-739
Othmer, S.F. and Othmer, S. (2007) Interhemispheric EEG Training: Clinical Experience and Conceptual Models,
Chapter 5 in Handbook of Neurofeedback, Dynamics and Clinical Applications, James R. Evans, Ed., The
Haworth Medical Press, New York, p.109-136
Othmer, S. (2009). Neuromodulation Technologies: An Attempt at Classification, Chapter 1 in Introduction to
QEEG and Neurofeedback: Advanced Theory and Applications (Second Edition), Thomas Budzynski, T.,
Evans, J.R., and Andrew Abarbanel, A., Eds, Elsevier, pp. 3-26
Othmer, S., & Othmer, S. F. (2009). Post-Traumatic Stress Disorder – The Neurofeedback Remedy. Biofeedback,
37(1), 24–31. http://dx.doi.org/10.5298/1081-5937-37.1.24
Preprints (www.preprints.org) | NOT PEER-REVIEWED | Posted: 17 October 2023 doi:10.20944/preprints202310.1085.v1
28
Othmer, S., Othmer, S., Legarda, S., (2011). Clinical Neurofeedback: Training Brain Behavior, Treatment
Strategies - Pediatric Neurology and Psychiatry, 2(1):67-73
Othmer, S., Othmer S.F., Kaiser, D.A., and Putman, J. (2013). Endogenous Neuromodulation at Infra-Low
Frequencies, Seminars in Pediatric Neurology, 20(4): 246-260 (2013) doi:10.1016/j.spen.2013.10.006
Othmer, S. and Othmer, S.F. (2016). Infra-Low-Frequency Neurofeedback for Optimum Performance.
Biofeedback, 44, 81-89. DOI: 10.5298/1081-5937-44.2.07
Othmer, S. and Othmer, S.F (2017). Development of the Othmer Method of Neurofeedback (PDF). Available
from:
https://www.researchgate.net/publication/317868978_Development_of_the_Othmer_Method_of_Neurofe
edback [accessed Dec 26 2022].
Othmer, S.F. (2019). Protocol Guide for Neurofeedback Clinicians, 7th Edition (2019), EEG Info, Los Angeles
Plenz D, and Niebur E (2014). Criticality in Neural Systems, Wiley-VCH, Weinheim, Germany
Porges, S.W. (2011). The Polyvagal Theory; Neurophysiological Foundations of Emotions, Attachment,
Communication, and Self-Regulation, WW Norton, New York
Putman, J.A., Othmer, S.F., Othmer, S., and Pollock, V.E. (2005), TOVA Results Following Inter-Hemispheric
Bipolar Training, Journal of Neurotherapy, 9(1), 27-36 (2005)
Quirk, D. A. (1995). Composite biofeedback conditioning and dangerous offenders: III. Journal of Neurotherapy.
1 (2), 44-54.
Rai, D., Kosidou, K., Lundberg, M., et al (2011). J. Epidemiological Health, doi:10.1136/jech.2010.119644
Rockstroh, B., Elbert, T., Birbaumer, N., Wolf, P., Duchting-Roth, A., Reker, M., et al. (1993). Cortical self-
regulation in patients with epilepsies. Epilepsy Research, 14, 63–72.
Rockstroh B, Elbert T, Canavan AG, Lutzenberger W, Birbaumer N. 1989. Slow Cortical Potentials and Behavior,
Second ed. Baltimore: Urban & Schwarzenberg.
Ros, T., Munneke, M.A.M., Ruge, D., Gruzelier, J.H., Rothwell, J.C. (2010) Endogenous control of waking brain
rhythms induces neuroplasticity in humans, European Journal of Neuroscience, 31 (4), 770-778.
DOI:10.1111/j.1460-9568.2010.07100.x
Ros, T., Moseley, M.J., Bloom, P.A., Benjamin, L., Parkinson, L.A., Gruzelier, J. H. (2009). Optimizing surgical
skills with EEG neurofeedback, BMC Neuroscience, 10:87
Ross, J.A. and Van Bockstaele, E.J. (2021) The Locus Coeruleus- Norepinephrine System in Stress and Arousal:
Unraveling Historical, Current, and Future Perspectives. Front. Psychiatry 11:601519. doi:
10.3389/fpsyt.2020.601519
Schore, A.N. (1997). Early organization of the nonlinear right brain and development of a predisposition to
psychiatric disorders. Development and Psychopathology, 9 (1997), 595–631
Schneider, F., Rockstroh, B., Heimann, H., et al (1992): Self-regulation of slow cortical potentials in psychiatric
patients: Schizophrenia. Biofeedback & Self-Regulation 17:277-92.
Scott, W.C., Kaiser, D.A., Othmer, S., and Sideroff, S.I. (2005). Effects of an EEG Biofeedback Protocol on a Mixed
Substance Abusing Population, American Journal of Drug and Alcohol Abuse, 31(3), 455-469
doi:10.1081/ADA200056807
Sihn, D., and Kim S-P. (2022). Brain Infraslow Activity Correlates with Arousal Levels, Front. Neurosci.,
16:765585. doi: 10.3389/fnins.2022.765585
Siniatchkin, M., Hierundar, A., Kropp, P., Kuhnert, R., Gerber,W. D., & Stephani, U. (2000). Self-regulation of
slow cortical potentials in children with migraine: An exploratory study. Applied Psychophysiology &
Biofeedback, 25, 13–32.
Sitaram, R., Ros, T., Stoeckel, L., Haller, S., Scharnowski, F., Lewis-Peacock, J., et al. (2017). Closed-loop brain
training: the science of neurofeedback. Nature Reviews Neuroscience 18, 86– 100. doi:10.1038/nrn.2016.164
Spreyermann, R. (2022). Infra Low Frequency Neurofeedback for PTSD: A Therapist's Perspective . Front. Hum.
Neurosci. 16:893830. doi: 10.3389/fnhum.2022.893830
Sterling, P. (2012). Allostasis: A model of predictive regulation. Physiology and Behavior, 106, 5-15
doi:10.1016/j.physbeh.2011.06.004
Sterman, M.B., Howe, R.C., and MacDonald, L.R. (1970). Facilitation of spindle-burst sleep by conditioning of
electroencephalographic activity while awake. Science, 167: 1146-1148
Sterman, M.B. (1976). Effects of Brain Surgery and EEG Operant Conditioning on Seizure Latency following
Monomethylhydrazine Intoxication of the Cat. Experimental Neurology, 50, 757-765
Sterman ,M.B. (2000). Basic concepts and clinical findings in the treatment of seizure disorders with operant
conditioning. Clin. Electroenceph. 31(1), 45-55
Sterman, M.B., Egner, T. (2006). Foundation and Practice of Neurofeedback for the Treatment of Epilepsy,
Applied Psychophysiology and Biofeedback, 31(1), 21-35. 10.1007/s10484-006-9002-x
Sterman, M.B. and Friar, L. (1972). Suppression of seizures in an epileptic following sensorimotor EEG feedback
training. Electroencephalogr. Clin Neurophysiol. 33(1), 89-95. doi.org/10.1016/0013-4694(72)90028-4
Strehl, U., Leins, U., Goth, G., Klinger, Ch., Hinterberger, Th., & Birbaumer, N. (2006). Self-regulation of slow
cortical potentials—A new treatment for children with ADHD. Pediatrics, 118, 1530–1540.
Preprints (www.preprints.org) | NOT PEER-REVIEWED | Posted: 17 October 2023 doi:10.20944/preprints202310.1085.v1
29
Strehl, U. (2009) Slow Cortical Potentials Neurofeedback, Journal of Neurotherapy: 13(2): 117-126 DOI:
10.1080/10874200902885936
Tansey, M. A. (1984). EEG sensorimotor rhythm biofeedback training: Some effects on the neurological
precursors of learning disabilities. International J of Psychophys 3, 85-99
Teicher, M. H. & Samson, J. A. (2013) Childhood maltreatment and psychopathology: a case for ecophenotypic
variants as clinically and neurobiologically distinct subtypes. Am. J. Psychiatry 170, 1114–1133 (2013).
Teicher, M. H., Anderson, C. M., Ohashi, K. & Polcari, A. (2014). Childhood maltreatment: altered network
centrality of cingulate, precuneus, temporal pole and insula. Biol. Psychiatry 76, 297–305
Teicher, M.H. Samson, J.A., Anderson, Carl M., Ohashi, K. (2016). The effects of childhood maltreatment on brain
structure, function, and connectivity. Nature Reviews/Neuroscience 17, 652-666
Thibault, R. T., Lifshitz, M., & Raz, A. (2018). The climate of neurofeedback: scientific rigour and the perils of
ideology. Brain, 141(2), e11–e11. https://doi.org/10.1093/brain/awx330
Ulrich, G. (2020). Ipsative Trend Assessment (ITA), An EEG-based Approach for Semi-quantitative Evaluation
of Disease Courses, unpublished monograph
Van der Kolk, B. (1994) The Body Keeps the Score: Memory and the Evolving Psychobiology of Posttraumatic
Stress, Harvard Review of Psychiatry, 1:5, 253-265, DOI: 10.3109/10673229409017088
Walker, J.E. (2011). QEEG-Guided Neurofeedback for Recurrent Migraine Headaches, Clin EEG and
Neuroscience, 42(1), 59-61
Webb, R.T., Pedersen, C.B., Mok, P.L.H. (2016) Adverse Outcomes to Early Middle Age Linked with Childhood
Residential Mobility. Am. J. Prev. Med., 51(3), 291-300 doi: 10.1016/j.amepre.2016.04.011
Winkeler A, Markus Winkeler M, Hartmut Imgart H (2022). Infra-Low Frequency Neurofeedback in the
treatment of patients with chronic eating disorder and comorbid post-traumatic stress disorder . Front.
Hum. Neurosci. 16:890682. doi: 10.3389/fnhum.2022.890682
World Health Organization (2022). ICD 11 for Mortality and Morbidity Statistics: 6B41 Complex posttraumatic stress
disorder. Available online at: http://id.who.int/icd/entity/585833559 (accessed August, 2022).
Wyrwicka, W., and Sterman, M.B. (1968). Instrumental conditioning of sensorimotor EEG spindles in the waking
cat. Physiol. Behav. 3: 703-707
Yu, J, Patel, R.A., Haynie, DL, Vidal-Ribas, P., Govender, T., Sundaram, R., and Gilman, S.E. (2022), Adverse
Childhood Experiences and premature mortality through mid-adulthood: A five-decade prospective
study, Lancet, published online: https://www.thelancet.com/action/showPdf?pii=S2667-
193X%2822%2900166-1 https://doi.org/10.1016/j. lana.2022.100349
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those
of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s)
disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or
products referred to in the content.
Preprints (www.preprints.org) | NOT PEER-REVIEWED | Posted: 17 October 2023 doi:10.20944/preprints202310.1085.v1
