ArticlePDF Available

Infra-Low-Frequency Neurofeedback for Optimum Performance

  • The EEG Institute, a dba of EEG Info

Abstract and Figures

A method of nonprescriptive neurofeedback is described that is based on the brain interacting with its own tonic slow cortical potential. In the absence of any explicit guidance by the clinician, the training depends entirely on the brain's response to the unfolding signal. When this training is performed under optimal conditions in terms of placement and target frequency, there is a bias toward optimal functioning. The brain uses the information for its own benefit. The outcomes of the training are either comparable to or exceed expectations based on conventional electroencelphalogram band-based neurofeedback. Results are shown for a cognitive skills test for an unselected clinical population.
Content may be subject to copyright.
Infra-Low Frequency Neurofeedback for
Optimum Performance
By Siegfried Othmer, Ph.D., and Sue Othmer,
The EEG Institute, Los Angeles, CA, USA
This is the text with color figures for an article published in Biofeedback, 44(2), pp. 81-89 (2016)
DOI: 10.5298/1081-5937-44.2.07 An error that appeared in the published version has been corrected here.
A method of non-prescriptive neurofeedback is described that is based on the brain
interacting with its own tonic slow cortical potential. In the absence of any explicit
guidance by the clinician, the training depends entirely on the brain’s response to the
unfolding signal. When this training is performed under optimal conditions in terms of
placement and target frequency, there is a bias toward optimal functioning. The brain
utilizes the information for its own benefit. The outcomes of the training are either
comparable to or exceed expectations based on conventional EEG band-based
neurofeedback. Results are shown for a cognitive skills test for an unselected clinical
Key words: neurofeedback, slow cortical potential, infra-low frequency training, optimum
The common objective of neurofeedback is the enhancement of cerebral function and
thus of organismic functional competence. This objective is not inherently deficit-
focused. On the contrary, the method depends entirely on the enhancement of function
that already exists. It is therefore a more organic perspective to regard neurofeedback
generally as a method of achieving optimal functioning. One may even take that view if
the starting point is a state of substantial dysfunction. Such dysfunctions may introduce
some constraints, but the objective remains the same. Dysfunction subsides by virtue of
improved function. What makes this view most appealing is that in the case of brain
training with neurofeedback one is not confronted with a headroom limit. One can
always do better in some respect or other. This is in contrast to much of traditional
biofeedback, where the objective is to maintain good regulation of certain physiological
variables, and once those objectives are achieved there is nothing more to aspire to.
The fact that most neurofeedback is currently being conducted in a clinical setting
tends to shift the perspective to the remediation of deficits. Thus even the language
used to communicate with clients tends to adopt that perspective. This is unsurprising
for a number of reasons. The entire healthcare system is deficit-focused, and
consequently clients tend to adopt that perspective as well. A somatic complaint or
functional deficit will naturally focus the attention narrowly. Moreover,
characterization methods used in neurofeedback tend to focus on the discernment of
deficits, both to make the case for neurofeedback and to guide the training. For
example, statistical parameter mapping is commonly used to determine deviations from
normative behavior. Such deviations are typically linked to the diagnosis and then
become targets for a ‘normalization paradigm’ in neurofeedback.
The notion that neurofeedback clinicians are actually prescribing a remedy for
particular complaints is in need of a sober, cold-eyed critique. If truth be told,
neurofeedback clinicians have lots of evidence that the brain may not respond as
directed in training situations. Or the predicted outcome may not be achieved with the
recommended protocol. Or the EEG does not change in the direction implied by the
protocol. When the brain is subjected to the close-order drill that is operant
conditioning, the response has greater variation than our models imply.
It is therefore more appropriate to regard the rewards and inhibits of a traditional
neurofeedback protocol as a provocation or a challenge rather than as an explicit
instruction. This is an easy case to make for the inhibit aspect of a training protocol.
After all, the brain is merely being alerted to its transient indiscretions, and it is left to its
own devices for a response. But even in the case of a targeted reward, matters are often
not what they seem. Response formation depends upon the brain assigning meaning to
the episodic reward, whereupon the brain is in a position to exercise all of its degrees of
freedom in response. The clinician is not in good control of that process.
It has been our own experience over the years that the brain responded far too quickly to
the training in many cases to be reasonably attributed to an operant conditioning
response. In addition to the expected slow and gradual learning curves, we were
observing state shifts and symptom relief that were surprisingly rapid and unexpected.
When we first let this be known, it led to a lot of initial skepticism about our clinical
findings. Our observations were not the problem, however. It was the model in terms of
which such results were inconceivable. The brain was deriving more information from
the signal than we thought we were providing. While we as scientist-practitioners were
focusing on the operant conditioning aspects of the design, the brain was appraising the
signal more comprehensively. Over time, we came to abandon the operant conditioning
aspect of the training entirely, relying instead solely on the brain’s observation of its own
output, as reflected in the EEG. This came about quite naturally rather than by virtue of
an explicit decision, as described in the following section. The signal was now
continuous, allowing the brain to experience it rather than merely to observe it. The
feedback became more organic, more captivating and thus more effective, quicker to
reach the goal. Once we gave up trying to dictate to the brain in an attempt to prescribe
outcomes, the brain placed itself totally in charge of the process.
Once the neurofeedback process is left to the discretion of the brain in its execution, it
becomes apparent that the brain utilizes all the information available to it in the cause of
better regulation. The training process that has emerged is entirely analogous to the one
by which the brain acquired its self-regulatory skills in the first place, during early
development. We have simply given the brain the benefit of additional information, so it
is the beneficiary of much more direct feedback on its own regulatory activity than is
otherwise available. That in turn allows the brain’s natural proclivity toward self-
optimization to be executed in a larger workspace and with greater refinement. This
process is most appropriately seen in the frame of training toward optimum functioning,
a matter of skill learning. Within the biofeedback field, this has traditionally been referred
to as training toward mastery.
The emergence of this new kind of brain training is described briefly herein, and results
are presented that make the case for the optimum functioning model. Since the results
are achieved in the absence of any bias imposed on the feedback signal, the case is
made that the brain utilizes the feedback signal in its own best interests, that is to say
with a bias toward optimal performance.
Mechanisms-based training
The original sensorimotor rhythm (SMR)/beta training approaches by Sterman and
Lubar were aiming at achieving better regulatory function in first order. Sterman’s SMR
training was intended to achieve better regulation of motoric excitability. In its
application to seizure control, the protocol was therefore invariant, independent of the
locus of any seizure focus (Sterman, 2000). The inhibits on the theta-band and high-beta
band activity were intended to shore up the integrity of the rewards in the SMR-band.
They were not thought to play any intrinsic training role in their own right when they
were first instituted.
In Sterman’s original training of cats, there was no implication that a functional deficit
was being targeted (Sterman et al., 1970). They were normal cats (or at least they were
before they had electrodes implanted in their brains). The quality of their sleep improved
by virtue of the training. The fact that the training effected control of chemically induced
seizures was incidental to the original objective of the training, which was to investigate
the effects on sleep (Sterman, 1976). In retrospect, this can be seen entirely in an
optimum functioning frame.
The addition of beta1 training by Lubar was aiming at improved cognitive function in
Attention Deficit Hyperactivity Disorder (ADHD) (Lubar & Lubar, 1984). His training
was the first to assign an EEG shaping role to the theta-band inhibition. However, as
already indicated, the inhibit function was passive in that it placed no imperative demand
upon the brain, but rather merely informed it of its status of ranging outside of the bounds
of good performance. The categorical remedy was the restoration of better performance,
but the remedy was not narrowly specified.
The inhibit-based training was later augmented with quantitative electroencephalogram
(QEEG)-based targeting, which shaped thinking more toward a deficit focus. However,
matters had not been fundamentally altered. The availability of the QEEG simply
enlarged the parameter space in which an appropriate inhibit strategy was to be
determined. The response of the brain was still discretionary; also, the reward strategy
largely remained unaffected. And even when QEEG measurements were used to inform
a strategy of low-level stimulation, whether optical, auditory, or electromagnetic, these
stimulations likewise served mainly in the role of a provocation or a challenge, rather
than as a specification of an imperative target for the training.
The development path of the reward scheme in the Sterman/Lubar paradigm took a very
different form at our hands. The availability of video feedback facilitated the presentation
of the full dynamics of the training band along with the threshold crossings. In addition to
this being more engaging to the trainee, it turned out to be more informative for the brain
as well. Trainees responded more quickly and more profoundly, particularly with the use
of bipolar montage as opposed to referential placement. It was observed that some
individuals were exquisitely sensitive to the particulars of the reward frequency, and that
discovery introduced the concept of the Optimum Reward Frequency, or ORF, which has
guided our work ever since. With each trainee, the effort was undertaken to optimize the
reward, or target frequency.
Such sensitivity to the particulars of the reward seems hard to believe when one inspects
the signal at issue. An example of three training bands, separated by 0.5 Hz, is shown in
Figure 1, along with the spectrals for the same bands. It is difficult to tell the difference
between the signals in the time domain, and the differences likewise seem quite modest
in the frequency domain. Nevertheless, sensitive responders can systematically
distinguish between the different bands in training, and they can often do so very
quickly, on the timescale of one to six minutes. Some individuals are sensitive to
differences even smaller than 0.5 Hz. The ORFs determined for each individual are
quasi-stable, changing only slowly with training if at all.
Figure 1. Three filtered traces derived from the same source, with center frequencies spaced 0.5 Hz apart, at 9.5 Hz,
10 Hz, and 10.5 Hz, respectively. The broadband EEG is shown in the top trace, and the bottom trace shows the
rectified and smoothed feedback signal for one of the three traces. The compressed spectral array is shown on the
right for each of the filtered traces.
The sensitivity to target frequency was a particular issue with those afflicted with
instability of brain function such as seizures, migraines, panic attacks, asthmatic
episodes, and Bipolar Disorder. With individualization of the training frequency the
clinical reach was extended to a much broader range of conditions, and it extended to
more complex clinical manifestations. The frequency range eventually covered the
entire conventional EEG range of 0.5 to 40 Hz. The distribution of target frequencies
was strongly skewed toward the low end of the frequency range, and the most common
target frequency was the lowest available in the software at hand: 0.1 Hz. This led to the
exploration of the tonic Slow Cortical Potential as a training vehicle in 2006.
Infra-Low Frequency (ILF) Training
In the range of 0.1 Hz and below, the)training is done on the basis of simple waveform-
following, in which the brain merely witnesses the time course of the tonic Slow
Cortical Potential (SCP), which directly reflects cortical excitability (Elbert, 1993). The
target frequency is too low for conventional amplitude-based training with
thresholding, so perforce operant conditioning had to be abandoned formally as the
operative model. It had already become inoperative by virtue of the rapid response we
had been observing, but with the abandonment of thresholding no vestige of the operant
conditioning model remained. Placement was always bipolar. The dynamics of the
signal then reflect the fluctuating differential cortical activation between the two active
scalp sites, and the brain is observed to engage quite effectively with that information.
The gradual migration of the training to ever lower frequencies has resulted in a process
that is entirely brain-based. The slowly meandering, relatively featureless signal holds
no inherent interest except for the brain that produced it in the first place. And the
process only gets underway once the brain recognizes its own agency with respect to
the fluctuating signal. The key to that recognition must be the ongoing dynamics in the
The approach differs from the well-known SCP training that operates on transient
behavior in that it removes the dependence on cognitive engagement with the task
(Birbaumer, 1999). In the new approach, there is no overt challenge. There is not even a
requirement that the trainee be apprised of the actual signal, or even to be aware of the
training procedure at all. Nevertheless, the trainee typically responds fairly promptly to
the signal with a shift in arousal level, alertness, and vigilance, and such within-session
response permits the optimization of the target frequency. The objective is to determine
the frequency at which the trainee is optimally calm, alert, and euthymic. The training
proceeds under conditions of the best-regulated state accessible to that nervous system at
that moment. In the absence of felt or reported within-session response, the training is
optimized on the basis of session-to-session changes.
The frequency-specificity of the conventional training carries over into the infra-low
frequency region. For that reason, we refer to this approach as infra-low frequency (ILF)
training. The new approach completely took over our practice in 2006, and appeared to
yield better and more rapid results for all clinical conditions typically seen in a
neurofeedback practice. Over time, the clinical reach was broadened to cover more
challenging clients by extending the frequency range of training to below 0.01Hz. Even
at such low frequencies, the dynamical aspects of the signal are sufficient to engage the
brain, and to do so quite promptly.
Electrode placements carried over from the higher-frequency region. These have in fact
remained fairly invariant since the late nineties, but within that basic framework some
tactical shifts have taken place. Brain instabilities were uniformly addressed with the
inter-hemispheric placement T3-T41 (Othmer, 2015). This has been found to be more
effective clinically than the C3-C4 placement that was more commonplace early on in the
field (Quirk, 1995). It has therefore become the default starting placement for the brain
Lateralized placements include principally T4-P4, T3-Fp1, and T4-Fp2. Whereas there
was an early pre-occupation within the field with left-side placements, the extension of
the work to the low frequencies has been accompanied by a shift toward right-side
training as a priority. This is readily explained on the basis that core regulatory function
is organized at low frequencies, and it involves right hemisphere priority. We have
progressively moved toward those issues that are primary in our developmental
hierarchy, issues that may not be as effectively addressed with higher-frequency training.
These consist of arousal regulation, affect regulation, autonomic regulation, and
interoception. These core issues are always involved in neurofeedback to some degree,
but perhaps not as effectively and efficiently as with infra-low frequency training
targeting the right hemisphere.
In practice, only two protocols are candidates at the outset of training, depending on
whether arousal regulation or brain stability dominates in the clinical presentation. These
are T4-P4 and T3-T4, as shown in Figure 2. In some instances, both are called for from
the outset. Then, depending on the need, others of the four standard placements are
added to the protocol. Further downstream, yet other protocols may be added for more
specific purposes.
Figure 2. Starting placements for Infra-Low Frequency neurofeedback. These are either used exclusively in the first
session, or they are used sequentially in the event that both are required.
This approach reflects a very clear ‘hierarchy of needs’ that becomes apparent as the
various protocols are evaluated for inclusion. A kind of scaffolding applies to this
process as the early protocols lay the foundation for the protocols to follow. Some of the
protocols that eventually become necessary might not have been tolerated at the outset.
One has the clear sense of recapitulating the person’s original developmental hierarchy
and of facilitating a kind of re-ordering and of functional re-normalization.
The hemispheric division of labor plays a determinative role in the clinical decision-
making process. In line with the well-known approach/withdrawal dichotomy, the right
hemisphere takes primary responsibility for issues of core state regulation and of the
vegetative domain. Concomitantly, cortical resting states are more broadly and
intimately connected on the right side (Buckner, et al. 2008). Being primarily responsible
for interoception, the right hemisphere sees to one’s sense of safety (Sridharan, et al.
2008). The left hemisphere exercises a primary responsibility with respect to executive
function and engagement with the outside world. The quality of the latter is conditional
on the functionality of the former. Hence the key to good left-hemisphere function
actually lies with the right hemisphere.
Assessing progress in training is firstly a matter of tracking symptom severity, and
secondarily one of assessing the quality of regulation broadly: the quality of sleep; of
emotional regulation; of alertness and vigilance; etc. Despite the very limited targeting,
the breadth of impact makes it apparent that the whole brain is affected in this process,
including in particular left-hemisphere function, even before it has been explicitly
targeted. Left-hemisphere functionality can be readily assessed with an instrument such
as a Continuous Performance Test (CPT).
The CPT is a pressured choice reaction time test that allows one to characterize a variety
of functions (Othmer, 2014). We utilize the QIKtest (, which was
designed to emulate the TOVA ® (Test of Variables of Attention) (Leark, 1996). It
allows a comparison of stimulus-sparse and stimulus-intensive challenges and thus
explores the arousal-level dependence of functionality. By tracking the variation in
performance over the session it also yields information on the stability of brain function
and on the capacity to maintain vigilance under the challenge of tedium. Results were
analyzed using EEG Expert ( Initially the QIKtest relied on TOVA
norms. However, as the data accumulated on a central server from a large practitioner
network, it became apparent that the distributions of the discrete errors were distinctly
non-Gaussian. Both omission and commission errors exhibited ‘long-tail’ (power-law)
behavior. This finding invalidated the use of Gaussian statistics in the determination of
normative performance. Hence the QIKtest analysis came to rely on population-based
norms that were established on a database of over 50,000 records. The use of such norms
is contingent only on the availability of a representative and statistically robust sample.
Non-parametric statistics were used throughout. The resulting percentile scores were then
converted back into equivalent standard scores by means of the conversion that applies to
a Gaussian distribution, for ease of inspection and to facilitate comparisons in terms of
familiar categories (e.g., standard scores).
We utilize the QIKtest as an evaluation tool and progress measure with everyone who is
capable of taking it, and for that reason it presents an unparalleled opportunity for the
evaluation of neurofeedback with a single instrument across the entire range of
functional and dysfunctional populations.
Clinical Results
Impulsivity, as indexed by errors of commission, presents an almost ideal measure to
track because the deficit lies in the functional domain and should in principle be subject
to normalization. The brain has to be functional in order to be impulsive. On the other
hand, the ‘normal’ range of performance leaves room for improvement. This is in
contrast to the case of omission errors, where organicity plays a much larger role.
Whereas non-responders may make up only about five percent of a clinical population
with respect to impulsivity, they may constitute as many as 25% of the population with
respect to errors of omission. Additionally, the objective of zero omission errors is
commonly met by many, even in a clinical population, and is then not available for
further improvement. So the inattention measure loses discrimination. For both of the
above reasons, the impulsivity measure is preferred over the inattention measure to
appraise neurofeedback in an optimum functioning paradigm.
Results of the QIKtest for the impulsivity measure are shown in Figure 3 for 5,746
clients who received nominally twenty sessions of infra-low frequency neurofeedback
training. The pre-training distribution is shown in green; the post-training results are
shown in red. The normative distribution is shown in black. The data have been
smoothed by means of near-neighbor averaging for greater clarity.
Impulsivity Score
Figure 3. The distribution of impulsivity score for a non-selected clinical population of 5,746 is shown both before and
after twenty sessions of ILF neurofeedback, in green and red, respectively. The norm is shown in black. The dotted
curve shows the difference pre-post. The actual distributions are not Gaussian-distributed, but have been converted to
Gaussian equivalent for ease of inspection. See text for discussion.
The result of the training experience was to move the distribution above norms
systematically. The deficited portion of the distribution became significantly depleted.
The effect size is approximately 0.75 even for this population, for which there has been
no prior selection of a deficited pool. The greatest improvement from pre-to-post is found
at one standard deviation above norms. The probability of scoring two standard
deviations above the mean doubled with the training, despite the fact that the score was
already above norms at the outset.
The same data set can also be used to evaluate what happens with a population in
deficit with respect to impulsivity by arbitrarily limiting the sample to those who scored
poorly at the outset. A cutoff of one standard deviation below norms means that we are
looking at the bottom 16% of the entire distribution in terms of impulsivity. A degree of
homogeneity of the sample was also sought by limiting the age range to 10-19. The
resulting sample size of 578 represents the subset of the population of Figure 3 that met
both criteria.
Figure 4. The distributions for impulsivity score before and after training are shown for a subset of the
population in Figure 2. The pool is selected in terms of age (10-19 years) and is restricted to those who initially
scored more than one standard deviation below norms. See text for discussion.
The results are shown in Figure 4 in the form of cumulative distributions. The post-
training data reveal that 30% of the trainees ended up scoring above norms. The median
score has improved by one standard deviation; the effect size is approximately unity for
this deficited population. If the standard score of 85 is taken as the threshold for normal
functionality, the training has moved two-thirds of all trainees into the functional range
within twenty sessions. The pool of individuals who function in deep deficit is even more
strongly depleted, with the cohort scoring below 70 (at the second percentile level)
reduced by a factor of four.
As stated earlier, the training needs to be conducted at the optimal response frequency, the
ORF. A consistent finding with all placements and with all individuals is that the left
hemisphere optimizes at twice the frequency of the right for all training in the ILF range
(Othmer, 2013). This contrasts with the earlier finding of a difference of two Hertz
between the left and the right ORFs in the conventional EEG range of frequencies. The
crossover between these two regimes is at two Hertz on the right. These relationships are
illustrated in Figure 5. These relationships have by now been confirmed by thousands of
practitioners over a period of many years—fifteen in the case of the EEG frequency range;
nine in the case of the ILF range. Exceptions to these frequency relationships have been
reported by clinicians, but they tend to be quite rare.
Fig. 5: Hemispheric differences in the optimum target
frequency over the entire EEG spectrum
Figure 5. A fixed relationship prevails between the optimum response frequencies in the left and right hemispheres. In
the EEG range above two Hz, the left hemisphere training optimizes at two Hertz higher than the right. Below two Hz,
the relationship is harmonic. The left hemisphere training optimizes at twice the frequency of the right.
The data presented encompass the entirety of such data that was available for ILF
training. The sample population was unselected, consisting of the complete set of pre-
post data available in our database over the timeframe of 2006 to 2014. Consequently,
the data covered the time period over which the clinical method was developed from
its initial beginnings in 2006. In fact, the method is still in a state of ongoing
refinement. Moreover, the data were contributed by hundreds of clinicians of varying
levels of experience and of clinical acumen. The clients were in many cases those for
whom a bit of impulsivity would be seen as the least of their problems. The clinical
focus was not on the matter of resolving impulsivity, by and large. For all the above
reasons, the data of Figures 3 and 4 can be considered a valid reflection of real-world
experience with ILF training in actual clinical settings with typical clients.
The significant import is first of all that the findings based on the above sample are
statistically robust. Secondly, the data cement the case that neurofeedback in general,
and ILF training in particular, is a method of training for optimal functioning rather
than the mere remediation of deficits. This follows from the fact that the training
population is moved to better than normative performance, and that the relative
improvement with respect to prior performance increases monotonically with score.
A third observation is that many of the trainees were still on their first one or two training
protocols by the twenty-session milestone, most likely targeting the right hemisphere,
and yet benefit was observed for what is seen as a performance issue that strongly
implicates left-hemisphere function. This makes the case for the whole-brain training
effects even of lateralized placements. The possibility remains that further gains might be
in prospect as left-hemisphere training is incorporated. It is also known that the benefits
of training are not fully exploited by twenty sessions, particularly for those who remain
in deficit at twenty sessions.
Significantly, the above results cannot be explained in terms of the placebo. First of all,
there is no placebo model for the population shift above norms as seen in Figure 3.
Secondly, it must be recognized that the neurofeedback training was effectively covert.
Trainees were unaware of what signal they were training on, and many were unaware
that they were training their brains at all, at least until they experienced the effects on
their physiological state. Yet others were resolutely skeptical until they had to come to
terms with their own responsiveness to the signal. Hence there was no cognitive or
volitional aspect of the training process that could have mobilized the placebo response.
There was no signal with which they could usefully engage even if that had been their
intention. Third, the desired training effects were only available at the optimum response
frequency, the ORF. That kind of specificity rules out the placebo model. Fourth, since
the ORF is not known at the outset, the effects of the initial training may not accord with
the desired objectives of the training, and may even be contrary to the expectations of the
client. This constitutes an argument against the placebo-as-wish-fulfillment being
responsible for the effects.
It is true that the protocol also incorporates an inhibit component that is not too
dissimilar from other inhibit schemes that are commonly used in the field. Although the
inhibit-based feedback is not obtrusive, it is not covert. Whereas a trainee is unlikely to
be distracted or engaged by it, it is a readily discernible signal. Since the inhibit scheme
is not substantially distinguishable from what has been standardly available in the field,
it also cannot account for clinical effects that are clearly stronger than what was
observed before, when similar inhibit schemes were in effect.
Moreover, attention of the trainee is not usually called to the presence of the inhibits,
since these do not constitute an action item for the trainee. Questions about these subtle
intrusions into the feedback signal are not usually raised until after the training effects
become unambiguously apparent, at which time the client’s curiosity is aroused on the
question of how that might have come about. After all, the client had not been “doing
anything.” By this point, the question of a placebo effect has already been resolved with
respect to that particular individual. Nevertheless, it may be argued that the inhibits are
sufficient to mobilize a placebo response on the part of the trainee. To that proposition,
the response is as follows:
The most compelling argument against the placebo model for ILF neurofeedback is the
frequency relationship illustrated in Figure 5. The determination of the ORF is in each
case based entirely on the report of the trainee on his or her own subjective experience of
the training. The trainee is of course blind to the curve, and yet the client reports
invariably conform to the relationships of Figure 5. This demonstrates that the clinical
experience of the training is governed predominantly by what happens with the ILF
component of the training rather than with the inhibits. If neurofeedback were a placebo,
or even if the results were attributable mainly to the inhibit aspect of the training, then the
relationship expressed in Figure 5 could not have been discovered. Instead, we have the
relationship confirmed with every individual who experience both left and right-
hemisphere training. Every such confirmation is an argument against the placebo model
for that individual.
The above relationship points to an underlying ordering principle in the frequency domain
that relates right-hemisphere to left-hemisphere function. Observing that inter-hemispheric
training optimizes at the same frequency as right-lateralized placements allows one to
propose that the right hemisphere plays a dominant role in core state regulation. For
optimum response frequencies that fall below two Hz, the left hemisphere organizes itself
at the second harmonic with respect to the right hemisphere frequency. For optimum
response frequencies greater than four Hz, the left hemisphere coordinates with respect to
the right at a frequency two Hz higher than the right. At the present time, there is no
independent objective evidence to support the existence of the posited frequency
relationships. This presents an intriguing hypothesis to be pursued.
Dominance is imputed to the left hemisphere for executive function, motor planning, and
the integration of sensory input with motor output. The left hemisphere can be thought of
as principally responsible for the text of our lives, whereas the right hemisphere has the
burden of setting the context. The imputation of left hemisphere priority in the EEG range
has recently received support from the analysis of resting microstates (Pascual-Marqui et
al, 2014). Of four basic microstates, one has a left posterior hub and one has a right
posterior hub. The two remaining hubs have a front-back orientation. The posterior hub on
the midline is common to all four microstates, and there is only one frontal hub.
Information flow was found to dominate strongly from left to right, and from left to
middle, over the reverse flow. Results were obtained for the alpha and low-beta frequency
On the above assumption, it can be argued that ILF training is concerned with those
regulatory functions that are the primary burden of the right hemisphere. Since these are
also the functions organized in early childhood, there is complete congruence between
the hierarchy of regulation, the developmental hierarchy, and the hierarchy of
organization in the frequency domain. In a comparison with our earlier higher-frequency
trainings, in our judgment there appears to be a clear advantage in terms of both training
efficiency and outcomes if neurofeedback is begun with respect for the regulatory
hierarchy, even though the data presented here do not speak to that issue.
Just as the results rule out an explanation in terms of a placebo response, the method
likewise cannot be explained in terms of an operant conditioning model. The brain is
merely engaged with the unfolding dynamics of the differential cortical activation, as
reflected in the tonic SCP. This is closely analogous to what happens in ordinary
skill learning. The brain gets feedback on the performance of the skill, but the
feedback is a mere correlate of the actual activity. In the present instance the self-
regulatory skill of the brain is at issue, to which the EEG---or the SCP in this case---
is the operative correlate.
The mechanism appears to be the re-normalization or re-ordering of the functional
coupling of our core connectivity networks, by virtue of ‘activity-mediated plasticity.’
Altered functional connectivity has been postulated to be a key failure mechanism in
psychopathology (Menon, 2011). The brain’s observation of its own state in EEG
feedback propels it into novel state configurations, and every such state opens up the
‘near neighborhood’ of possible states. Every such state is susceptible to reinforcement
and consolidation.
Summary and Conclusion
Infra-Low Frequency training is an emerging approach to neurofeedback that is
intrinsically function-oriented, as opposed to targeting dysfunction. The clinical results
cannot be explained on the basis of a placebo model; hence the results stand on their
own, even absent validation via a placebo-controlled design. The method cannot be
described in terms of the standard operant conditioning model; nor does the method rely
on conscious mediation. Instead the results are explained in terms of conventional skill
In its essence, the training must be understood in the optimum functioning frame, as
these results are achieved without explicit guidance or micro-management by the
clinician. The clinician’s role is one of discerning which ‘window into brain function’ is
most salient for the brain’s burden of enhancing its own functional competence.
By working at extremely low frequencies with a right-hemisphere bias, the method
addresses the foundations of our regulatory hierarchy, thus opening the door to the
therapeutic relief of conditions traceable to early childhood developmental deficits or
misdirections. As such, ILF training has allowed us to address challenging clinical
presentations that did not yield to our earlier, higher-frequency protocols. In this work,
there is a complete continuum between working with a clinical population and one
concerned entirely with optimal functioning. The methods and means are the same.
Consequently, ILF training has added a new dimension to the various neurotherapy
approaches that have been maturing at their respective pace over the past several
Birbaumer, N. (1999). Slow cortical potentials: Plasticity, operant control and behavior
effects. Neuroscientist, 5(2), 74-78.
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, 1-38.
EEG Expert, Hamburg, Germany, available at
Elbert, T. (1993). Slow cortical potentials reflect the regulation of cortical excitability. In
C. W. McCallum (Ed.), Slow potential changes in the human brain (pp. 235-251). New
York, NY: Plenum.
Leark, R., Dupuy, T. R., Greenberg, L., Kindschi, C., & Corman, C. L., (1996). Test of
variables of attention: Professional manual (version 7). Los Alamitos, CA: Universal
Attention Disorders.
Lubar, J. O., & Lubar, J. F. (1984). Electroencephalographic biofeedback of SMR and
beta for treatment of attention deficit disorders in a clinical setting. Biofeedback and
Self-Regulation, 9, 1-23.
Menon, V. (2011). Large-scale brain networks and psychopathology: A unifying triple
network model. Trends in Cognitive Sciences, 15(10), 483-506.
Othmer, S., Othmer, S.F., Kaiser, D.A., Putman, J. (2013). Endogenous
Neuromodulation at Infra-Low Frequencies. Seminars in Pediatric Neurology, 20(4),
Othmer, S. (2014). The Role of the Continuous Performance Test.
Othmer, S.F. (2015). Protocol guide for neurofeedback clinicians (5th Edition).
Los Angeles, CA: EEG Info Publications.
Pascual-Marqui, R. D., Lehmann, D., Faber, P., Milz, P., Kochi, K., Yoshimura, M.,
Nishida, K., Isotani, T., Kinoshita, T. (2014). The resting microstate networks (RMN):
cortical distributions, dynamics, and frequency specific information flow. Cornell
University Library. Available at
QIKtest. Bee Medic GmbH, Kirchberg, Switzerland. Available at
Quirk, D. A. (1995). Composite biofeedback conditioning and dangerous offenders: III.
Journal of Neurotherapy 1(2), 44-54.
Sridharan, D., Levitin, D. J., & Menon, V. (2008). A critical role for the right fronto-
insular cortex in switching between central-executive and default-mode networks.
Proceedings of the National Academy of Sciences, USA, 105(34), 12569-12574. doi:
Sterman, M.B. (2000). Basic concepts and clinical findings in the treatment of seizure
disorders with EEG operant conditioning. Clinical Electroencephalography, 31(1), 45-
Sterman, M.B., Howe, R.D., & Macdonald, L.R. (1970). Facilitation of spindle-burst
sleep by conditioning of electroencephalographic activity while awake. Science, 167,
Sterman, M.B. (1976). Effects of brain surgery and EEG operant conditioning on seizure
latency following monomethylhydrazine intoxication in the cat. Exp. Neurol., 50, 757-
... By allowing the brain to obtain more salient information about its own processes, the brains capacity to selfregulate in a proper manner improves accordingly. It is assumed that operant conditioning underlies the effects of neurofeedback, through adaptation of self-regulatory processes (Othmer et al., 2013;Othmer & Othmer, 2016, 2017. However, the mechanisms which underpins the effect of neurofeedback has been a topic of debate (Ioannides, 2018) The human brain is both highly unitary and integrated. ...
... It is plausible to assume that disruptions within the regulatory mechanisms and the RSN consequently affect an individual during active states. If chronic pain is caused by maladaptive plasticity within RSN and disruptions in ISO, neurofeedback which aims to renormalize the abnormal resting state rhythms might decrease symptom severity (Othmer et al. 2013(Othmer et al. , 2016(Othmer et al. , 2017. Researchers have used sensory motor rhythm neurofeedback with positive outcomes on pain and fatigue in FM patients (Kayiran, et al., 2010). ...
... Specific electrode placement is individualised by the clinician based on the symptoms and clinical presentation. It is recommended that one should initiate treatment with a bipolar montage with a T3-T4 placement, as this display both strong effect and broader clinical efficacy (Othmer & Othmer, 2016, 2017Othmer et al., 2013). ...
Full-text available
Background: One of the main motives for why individuals seek medical attention is pain. Fibromyalgia (FM) is a condition characterized by chronic pain, fatigue, and cognitive complaints, which severely disrupts an individual’s quality of life. Medical providers and researchers have not been able to find a There is no cohesive theory of why some individuals have fibromyalgia. Consequently, there is a lack of adequate diagnostic tools, unsatisfactory treatment, and uncertainty amongst patients. Previous studies have found fibromyalgia patients to display significant alterations in central mechanisms, functional connectivity in the resting-state networks and cortical areas identified as the Dynamic Pain Connectome (DPC). Aims: This study consists of two parts. It aims to (1) identify whether individuals suffering with fibromyalgia significantly differ in the temporal dynamics of the brain, and if this is related to cortical areas involved in the DPC. The second part wishes to (2) investigate the clinical benefits of infra-low frequency neurofeedback treatment (ILF-NFT) on fibromyalgia symptoms. Method: FM patients received ILF-NFT, which included pre- and post-treatment clinical measures with a 19-channel EEG recording and self-reports of symptom severity. Power spectra analysis was conducted to look for deviations in the theta, alpha and beta frequency, derived from frontal, central, temporal, and parietal electrodes. Results: A Wilcoxon Signed-Rank Test found significant decreases in symptoms following ILF-NFT, indicating that the treatment targets cortical activity associated with pain, fatigue, and cognitive complaints. Several of the participants had deviations which were source localized in key DPC-nodes. The limitations of this study are further discussed.
... The data of Smith and colleagues, employing an adaptation of the method used here, supports the hypothesis that ILF training preferentially influences autonom-ic nervous system regulation and thus improves the emotional equilibrium of patients, which in turn positively influences attention and working memory [39]. Further evidence with respect to vigilance and attention was recently documented in a large-scale compilation of pre-and post-training continuous performance test data on a clinical population [42]. Clinically significant improvement in performance was consistently observed among those in deficit, largely irrespective of the clinical condition being targeted in the training. ...
Full-text available
A formal comparison of Infra-Low Frequency Neurofeedback with an active control condition, Heart Rate Variability training, is undertaken in the present research. 17 participants 21-50 years of age with no history of neurological or psychiatric diseases conditions, but reporting about some physiological or psychological complaints were involved in the study. Participant progress was monitored by means of Visual Go/NoGo test performance and spectral power of slow EEG oscillations during the test before and after twenty sessions of training. Outcomes favored Infra-Low Frequency Neurofeedback training over Heart Rate Variability training with respect to health status and Visual Go/NoGo test results. Significant elevation in amplitudes in the Infra-Low Frequency range was observed only for the Neurofeedback cohort.
A good number of veterans while serving in recent combat zones experienced blast injuries resulting in traumatic brain injuries (TBIs), 80% of which were mild (m) with 25%–50% having prolonged postconcussive symptoms (PCSs). Neurofeedback (NFB) has demonstrated a decent degree of efficacy with mTBI PCSs in civilian and veteran populations. Using infra-low frequency NFB, the authors conducted a pilot study to determine the feasibility and initial efficacy with veterans. Because these results were promising, funding for a full clinical trial was subsequently applied for and acquired.
Full-text available
Electroencephalographic neurofeedback (EEG NF) can improve quality of life (QoL) and reduce distress by modifying the amplitude of selected brain frequencies. This study aims to investigate the effects of NF therapy on QoL and self-efficacy in cancer patients and to explore age-related reactions. In a waitlist control paradigm, psychometric data (EORTC QLQ-C30, General Self-Efficacy Scale) of 20 patients were collected at three different time points, each five weeks apart. An outpatient 10-session NF intervention (mobile) was conducted between the second and third measurement point. QoL and self-efficacy changed significantly over time (QoL: F(2,36) = 5.294, p < .05, η² = .227; Self-efficacy: F(2,26) = 8.178, p < .05, η² = .386). While QoL increased in younger patients, older patients initially showed a decrease in QoL, which then increased during intervention. Younger patients did not differ from older patients in QoL in both waitlist control (T0-T1) and intervention phase (T1–T2). QoL in older patients significantly differed between waitlist control and intervention phase (Z = − 2.023, p < .05, d = 1.085). Self-efficacy increased in both age categories. Younger and older patients did not differ in self-efficacy in waitlist control, but in intervention phase (F(1,16) = 7.014, p < .05, η² = .319). The current findings suggest that NF therapy is a promising treatment modality for improving QoL in cancer patients. Our study reveals NF being a tool to influence self-efficacy, which should receive more appreciation in clinical care. However, the effect of NF in different age groups as well as the influence on further cancer-related symptoms should be investigated in future research.
Full-text available
This paper will review what is conventionally known of sleep homeostasis and focus on insomnia as a primary manifestation of brain dysregulation, whether as a solitary symptom or as part of a larger syndrome such as post-traumatic stress disorder, PTSD. It will discuss in brief behavioral/mindfulness treatments that have been used to treat neurologic diseases, as this is germane to the phenomenology of neurofeedback (NF). It will explore how neurofeedback may work at the subconscious level and cover the current clinical experience of the effectiveness of this technique in the treatment of insomnia. It will conclude with a case presentation.
Full-text available
A 38-year-old army officer started therapy in 2020 with a four-year history of auditory hallucinations and delusions of reference, persecution and grandeur, symptoms that were resistant to traditional antipsychotic medications. He follows an integrative psychotherapy program that aims to reduce his anxiety, continues his antipsychotic medications, and has Infra-Low Frequency Neurofeedback. After his initial assessment he had a 40 min session of Infra-Low Frequency Neurofeedback before any other kind of intervention. Before and immediately after the session he completed the SCL-90 scale and the Visual Analog Scale covering 20 aspects of his psychological and physical state as well as his schizophrenic symptoms. This first Neurofeedback session had dramatic effects on his psychotic symptoms, levels of anxiety and psychosomatic condition, before his first psychotherapy session and/or any changes in his antipsychotic medication. The above results have great importance due to the severity and chronicity of schizophrenia. Informed consent was obtained from the participant for the publication of this case report (including all data and images).
Full-text available
Autism spectrum disorder (ASD) is a neural and mental developmental disorder that impacts brain connectivity and information processing. Although application of the infra-low frequency (ILF) neurofeedback procedure has been shown to lead to significant changes in functional connectivity in multiple areas and neuronal networks of the brain, rather limited data are available in the literature for the efficacy of this technique in a therapeutic context to treat ASD. Here we present the case study of a 5-year-old boy with ASD, who received a treatment of 26 sessions of ILF neurofeedback over a 6-month period. A systematic and quantitative tracking of core ASD symptoms in several categories was used to document behavioral changes over time. The ILF neurofeedback intervention decreased the average symptom severity of every category to a remarkable degree, with the strongest effect (80 and 77% mean severity reduction) for physical and sleep symptoms and the lowest influence on behavioral symptoms (15% mean severity reduction). This case study is representative of clinical experience, and thus shows that ILF neurofeedback is a practical and effective therapeutic instrument to treat ASD in children.
Full-text available
Social anxiety disorder has been widely recognised as one of the most commonly diagnosed mental disorders. Individuals with social anxiety disorder experience difficulties during social interactions that are essential in the regular functioning of daily routines; perpetually motivating research into the aetiology, maintenance and treatment methods. Traditionally, social and clinical neuroscience studies incorporated protocols testing one participant at a time. However, it has been recently suggested that such protocols are unable to directly assess social interaction performance, which can be revealed by testing multiple individuals simultaneously. The principle of two-person neuroscience highlights the interpersonal aspect of social interactions that observes behaviour and brain activity from both (or all) constituents of the interaction, rather than analysing on an individual level or an individual observation of a social situation. Therefore, two-person neuroscience could be a promising direction for assessment and intervention of the social anxiety disorder. In this paper, we propose a novel paradigm which integrates two-person neuroscience in a neurofeedback protocol. Neurofeedback and interbrain synchrony, a branch of two-person neuroscience, are discussed in their own capacities for their relationship with social anxiety disorder and relevance to the paradigm. The newly proposed paradigm sets out to assess the social interaction performance using interbrain synchrony between interacting individuals, and to employ a multi-user neurofeedback protocol for intervention of the social anxiety.
Full-text available
There are several different methods of neurofeedback, most of which presume an operant conditioning model whereby the subject learns to control their brain activity in particular regions of the brain and/or at particular brainwave frequencies based on reinforcement. One method, however, called infra-low frequency [ILF] neurofeedback cannot be explained through this paradigm, yet it has profound effects on brain function. Like a conductor of a symphony, recent evidence demonstrates that the primary ILF (typically between 0.01–0.1 Hz), which correlates with the fluctuation of oxygenated and deoxygenated blood in the brain, regulates all of the classic brainwave bands (i.e. alpha, theta, delta, beta, gamma). The success of ILF neurofeedback suggests that all forms of neurofeedback may work through a similar mechanism that does not fit the operant conditioning paradigm. This chapter focuses on the possible mechanisms of action for ILF neurofeedback, which may be generalized, based on current evidence.
The aim of this pilot study was to assess whether neurofeedback (NFB) can be useful in the treatment of impulsive behavior in long-term abstinent cocaine and heroin addicts. A single-blind sham-controlled NFB protocol was carried out to assess the effects of NFB on impulsivity in 20 (10 + 10) cocaine and heroin long-term abstinent addicts ( Diagnostic and Statistical Manual of Mental Disorders [4th ed., text rev.; DSM-IV-TR]). Psychotic and neurologic diseases were excluded. Participants underwent 40 NFB sessions based on the very slow cortical potential range. Inhibitory deficits were specifically addressed through right and left prefrontal training. Clinical improvement was measured with Likert-type scales, the Hamilton Depression Rating Scale, and the State–Trait Anxiety Inventory, and impulsivity was assessed using the Barratt Impulsiveness Scale and the Continuous Performance Test. Although the results are preliminary due to the small sample size, the NFB-treated group showed a significant clinical improvement, including symptoms of anxiety and depression, with two differentiated time periods. No significant clinical improvement was found in the control group. A significant decrease in the post- versus pre-treatment measures of global impulsivity, nonplanning impulsivity, and error commission measures was found in the NFB-treated group; effect size ( d Korr ) in the pre–post control design was moderate. No significant change was found in the control group. Despite the limitations of this study, the results suggest that NFB is better than placebo in improving impulsivity and clinical symptoms of anxiety and depression in long-term abstinent cocaine- and heroin-dependent individuals.
Full-text available
A brain microstate is characterized by a unique, fixed spatial distribution of electrically active neurons with time varying amplitude. It is hypothesized that a microstate implements a functional/physiological state of the brain during which specific neural computations are performed. Based on this hypothesis, brain electrical activity is modeled as a time sequence of non-overlapping microstates with variable, finite durations (Lehmann and Skrandies 1980, 1984; Lehmann et al 1987). In this study, EEG recordings from 109 participants during eyes closed resting condition are modeled with four microstates. In a first part, a new confirmatory statistics method is introduced for the determination of the cortical distributions of electric neuronal activity that generate each microstate. All microstates have common posterior cingulate generators, while three microstates additionally include activity in the left occipital/parietal, right occipital/parietal, and anterior cingulate cortices. This appears to be a fragmented version of the metabolically (PET/fMRI) computed default mode network (DMN), supporting the notion that these four regions activate sequentially at high time resolution, and that slow metabolic imaging corresponds to a low-pass filtered version. In the second part of this study, the microstate amplitude time series are used as the basis for estimating the strength, directionality, and spectral characteristics (i.e., which oscillations are preferentially transmitted) of the connections that are mediated by the microstate transitions. The results show that the posterior cingulate is an important hub, sending alpha and beta oscillatory information to all other microstate generator regions. Interestingly, beyond alpha, beta oscillations are essential in the maintenance of the brain during resting state.
Full-text available
Neuromodulation in the bioelectrical domain is an attractive option for the remediation of functionally based deficits. Most of the interest to date has focused on exogenous methods, such as repetitive transcranial magnetic stimulation, transient direct current stimulation, vagus nerve stimulation, and deep brain stimulation. Much less attention has been given to endogenous methods of exploiting latent brain plasticity. These have reached a level of sophistication and maturity that invites attention. Over the last 7 years, the domain of infralow frequencies has been exploited productively for the enhancement of neuroregulation. The principal mechanism is putatively the renormalization of functional connectivity of our resting-state networks. The endogeneous techniques are particularly attractive for the pediatric population, where they can be utilized before dysfunctional patterns of brain behavior become consolidated and further elaborated into clinical syndromes.
Full-text available
A model is described which postulates ways in which regulatory circuits within the brain might generate electrical activity underlying spontaneous EEG fluctuations and event-related slow potentials (ERP) of the brain. We suggest that slow potentials represent a measure of the excitability of cortical neuronal networks and that this excitability must be regulated within distinct limits. The regulation has reflexive characteristics but must contain anticipatory elements as well. Five main postulates are basic to the model: (1) The electrical sources of event-related slow potentials and other large amplitude EEG activity reside in the dendritic trees of cortical pyramidal neurons. (2) Surface negative potentials represent a measure for the excitability in the underlying neural tissue, surface positivity signifies widespread absence of facilitation. (3) Slow potentials indicate availability and spatial allocation of resources for information processing performed by the underlying neural tissue. Direction of attention and memory search will generate negative potentials in those neural assemblies which process the respective concepts. Memory storage and updating of information require that large proportions of the neural network be shut off and therefore they are accompanied by widespread positive waves of high amplitude. The electrical source of these waves can be traced to neural populations not involved in the storage process. (4) Conscious processes arise above a certain amount of cortical excitability; they appear when threshold levels of negative DC shifts are exceeded. (5) The regulation of cortical excitability and thus of slow potentials is achieved in part via a feedback loop which runs through the basal ganglia, other subcortical structures and the thalamus, returning to the cortex. This feedback loop generates permanent fluctuations resulting in EEG waves. The loop has necessarily non-linear characteristics and therefore the EEG may best be analysed with methods from non-linear systems theory. This paper describes evidence for these postulates and suggests ways of further testing the model.
Slow cortical potentials are negative or positive polarizations of the electroencephalo gram (EEG) or magnetic field changes in the magnetoencephalogram (MEG) that last from 300 ms to several seconds. They originate in depolarizations of the apical dendritic tree in the upper cortical layers that are caused by synchronous firing, mainly from thal amocortical afferents. Functionally they constitute a threshold regulation mechanism for local excitatory mobilization (negative slow potentials) or inhibition (positive slow poten tials) of cortical networks. Humans can learn to voluntarily regulate these potentials after operant training using immediate feedback and positive reinforcement for self-generated slow potentials shifts. After learned self-regulation of negative slow cortical potentials, motor and cognitive performance of various tasks improves, depending upon the specific cortical location of the learned response. Learned reduction of cortical negativity in creases seizure threshold and improves drug-resistant epilepsies. The learned self-reg ulation of slow cortical potentials is based on a redistribution of attentional resources and depends cntically on a prefrontal and thalamic attention system. Finally, a thought trans lation device uses the slow potential self-regulation skill in totally paralyzed locked-in patients for communication with a language-shaping computer system. NEURO SCIENTIST 5:74-78, 1999
Six children were provided with long-term biofeedback and academic treatment for attention deficit disorders. Their symptoms were primarily specific learning disabilities, and, in some cases, there were varying degrees of hyperkinesis. The training consisted of two sessions per week for 10 to 27 months, with a gradual phase-out. Feedback was provided for either increasing 12-to 15-Hz SMR or 16- to 20-Hz beta activity. Inhibit circuits were employed for blocking the SMR or beta when either gross movement, excessive EMG, or theta (4–8 Hz) activity was present. Treatment also consisted of combining the biofeedback with academic training, including reading, arithmetic, and spatial tasks to improve their attention. All children increased SMR or beta and decreased slow EEG and EMG activity. Changes could be seen in their power spectra after training in terms of increased beta and decreased slow activity. All six children demonstrated considerable improvement in their schoolwork in terms of grades or achievement test scores. None of the children are currently on any medications for hyperkinetic behavior. The results indicate that EEG biofeedback training, if applied comprehensively, can be highly effective in helping to remediate children who are experiencing attention deficit disorders.
The science of large-scale brain networks offers a powerful paradigm for investigating cognitive and affective dysfunction in psychiatric and neurological disorders. This review examines recent conceptual and methodological developments which are contributing to a paradigm shift in the study of psychopathology. I summarize methods for characterizing aberrant brain networks and demonstrate how network analysis provides novel insights into dysfunctional brain architecture. Deficits in access, engagement and disengagement of large-scale neurocognitive networks are shown to play a prominent role in several disorders including schizophrenia, depression, anxiety, dementia and autism. Synthesizing recent research, I propose a triple network model of aberrant saliency mapping and cognitive dysfunction in psychopathology, emphasizing the surprising parallels that are beginning to emerge across psychiatric and neurological disorders.
Cognitively demanding tasks that evoke activation in the brain's central-executive network (CEN) have been consistently shown to evoke decreased activation (deactivation) in the default-mode network (DMN). The neural mechanisms underlying this switch between activation and deactivation of large-scale brain networks remain completely unknown. Here, we use functional magnetic resonance imaging (fMRI) to investigate the mechanisms underlying switching of brain networks in three different experiments. We first examined this switching process in an auditory event segmentation task. We observed significant activation of the CEN and deactivation of the DMN, along with activation of a third network comprising the right fronto-insular cortex (rFIC) and anterior cingulate cortex (ACC), when participants perceived salient auditory event boundaries. Using chronometric techniques and Granger causality analysis, we show that the rFIC-ACC network, and the rFIC, in particular, plays a critical and causal role in switching between the CEN and the DMN. We replicated this causal connectivity pattern in two additional experiments: (i) a visual attention “oddball” task and (ii) a task-free resting state. These results indicate that the rFIC is likely to play a major role in switching between distinct brain networks across task paradigms and stimulus modalities. Our findings have important implications for a unified view of network mechanisms underlying both exogenous and endogenous cognitive control. • brain networks • cognitive control • insula • attention • prefrontal cortex
Monomethylhydrazine (MMH) is a highly convulsive methyl derivative of hydrazine. The generalized tonic-clonic seizures elicited by this compound are unusual because of a characteristic latent period between administration and seizures. Evidence indicates that the duration of this latent period can reflect seizure susceptibility in relation to this drug. In the present study this concept was utilized to evaluate the influence of chronic electrode implantation and subsequent EEG operant conditioning on seizure susceptibility in the cat. Thirty cats were studied in three groups of 10 each. One group consisted of unoperated animals, another of operated animals with a diversity of electrode placements and experimental treatments, and the third of operated animals provided with 3 months of sensorimotor EEG operant conditioning. The EEG pattern rewarded was rhythmic, 12 to 16 cps activity, termed the sensorimotor rhythm. Seizure response was measured as the latency, in minutes postinjection, to the onset of generalized tonic-clonic seizures following intraperitoneal administration of 10 mg/kg MMH. Operated animals with either no EEG conditioning or noncontingent conditioning showed significantly shorter and more stable seizure latencies than either the unoperated group or the operated group with sensorimotor-rhythm conditioning. These data indicate that the surgical procedures used increased seizure susceptibility in this paradigm, and that sensorimotor-rhythm operant conditioning countered this effect. Furthermore, the marked variability in seizure latencies noted among unoperated and trained animals suggested individual differences in seizure susceptibility and in response to operant conditioning, respectively.