In modern neuroscience centers, the EEG is still the tool of choice in evaluation of convulsive epilepsy, as well as some other non-convulsive forms, such as staring episodes seen in “absence epilepsy” typically as a 3/second spike and wave dominant anteriorly, or temporal lobe epilepsy, which is seen as a “notched” slow wave discharge fronto-temporally.

The EEG can now be processed through algorithms, such as spike dipole localization software, to identify the “seizure focus” cortically, or spectral averaging to look for changes in the underlying EEG rhythmicity due to the disorder.

One of the difficulty with the two later categories is that they are not always identified as forms of epilepsy, and thus can be mis-diagnosed based on behavior alone as some other disorders, including ADD/ADHD in absence epilepsy “spells” where the attentional process is disturbed by the discharge taking segments of time out of the cognitive streaming of perception, or from discharge in sensory areas.  These segments being removed do not have any conscious awareness of the event for the person experiencing the blips missing from their cognitive process, and they will have trouble tracking on-going events, like driving or listening to a speech or lecture.  Imagine missing a few here and there, to tens of seconds from your awareness, and see if you don’t have “attentional deficits”.

The other major areas of misdiagnosis are of a “schizophrenic” or “psychotic” nature.  This occurs when the discharges are frontal or temporal and disturbing local cortical function, and may be seen as a range of presentations from hallucinations or emotional outbursts of rage, or even “fits of laughter” in “Gelastic seizures”.   Temporal Lobe Epilepsy (TLE) is a particularly difficult one to properly diagnose in the absence of the EEG.

The importance of these missed-diagnoses can be quite severe, with the use of medications to treat the symptoms often being contra-indicated by the epilepsy.  One example of this is TLE that is assumed to be psychosis, since antipsychotic medications lower the seizure threshold, and make the person worse, which can then be responded to with more antipsychotics, spiraling the person into a progressively worsened condition.  The use of stimulants in epilepsy is a controversial area, as the effect of stimulants for inattention in known and treated epileptics may be acceptable, though throwing a stimulant at an undiagnosed epileptic can have severe negative consequences.

The real issue is that IF YOU DO NOT LOOK, YOU WILL NOT SEE… and in epilepsy, looking requires the EEG, as the gold standard.

In surgical approaches, the EEG is used to identify whether there are multiple foci, which generally will preclude a good outcome (you remove the brain tissue and the seizures do not change).

I will post some images of the WIDE variety of morphologic presentation that epilepsy can take, so that some understanding of the task of the Electroencephalographer and Epileptologist can be better appreciated by those who think it is straight-forward.

Thanks for your attention to these obscure issues.

Jay

These results are from 300 seconds of linked ear EEG data, note the dominant slower alpha peak frontally…. And the raw coherence values of that linked ear data. The raw EEG file sample is also included, so you can see the waveforms these values are being drawn from.

Linked Ears

Raw Coherence Values of Linked Ear Data

Raw EEG

The same exact 300 seconds of EEG data, reprocessed now with the weighted average montage.  Note the difference in spectra, and waveform!!!  The temporal slower alpha is now seen as the source of that slower alpha content.

The alpha hypercoherence in the EEG is easily seen in this data, but not in the linked ears.

This shows that you need to find the EEG montage that shows the actual EEG data for your case first, and THEN calculate coherence.

Weighted Average Spectra

Weighted Average Waveform

Weighted Average Raw

The images below show the Spectral plot, coherence plot and raw EEGs. Just like the other montages did. The Cz coherences are so inflated with field effects they are at 0.8 across the full spectrum at some sites, obviously artifactually high.

Spectral Plot

Coherence Plot

Raw EEG

Technical Issues in qEEG Power point Presentation

Continue Reading to see parts 2 and 3.

Part 2 Discusses Z-Scores and Standard Deviations

Following the careful recording of the EEG, the quantitative analysis is begun with the sampling of the data to be used in the analysis by the Fourier transform. The Fourier analysis assumes there are no transients (epileptic discharges, episodic voltage changes etc.) or state changes (light sleep, drug effect, mental task, etc.), so these must be avoided when selecting data for analysis in qEEG for eyes closed resting database comparison. There are some eyes open and task databases available more recently (Hudspeth, Sterman, Duffy etc.)

Transients are an event with a rapid onset and ending, with an increase in amplitude of greater than 50% over the ongoing activity. Epileptiform activity is one common example of this phenomenon. The only time transient discharges may validly be included is when dipole localization or “mapping” of the sources of this activity is the intent, and in this circumstance only significant discharges should be sampled, with the ongoing background treated as the state change and eliminated.

Less rigorous analysis selection standards exist for data not intended for database comparison, such as reading or other task related data. Usually these task data will only be used for gross comparison to the more carefully collected steady state data, looking for gross changes in brain function. When there is an intent to compare to an eyes closed normative database, transients will increased the variability of the dataset, but will be averaged out in the mapping unless persistent or very prominent. These transients will simultaneously alter the dataset’s standard deviation from the norm.

State changes include typically sleep stages collected for an eyes closed awake database comparison. Stage 1 sleep is a subtle drowsy state, where people recorded will usually deny being drowsy when alerted. The alpha is beginning to wax and wane, with subtle increases in theta and occasionally slow rolling eye movements and a decreased EMG tonus. This is just prior to stage 2, where an object being held will slip from grasp, alerting the client, with most realizing the drowsing is present in stage 2. Many people doing a repetitive task or automatic pilot type task will be in stage 1 without being aware of time passing. This is a dangerous situation if response to change is needed, as there are delays in reaction time associated with this state.

The problem with stage 1 (drowsing) being added to a dataset is that it is a mixing of states, violating the FFT assumptions and making database comparison validity more than merely suspect. The inclusion invalidates comparison.

The task of artifacting is to sample enough data to provide reliable maps while maintaining the validity of the sampling, not taking state changes or transients. The amount of data acquired should provide highly repeatable or reliable mapping. The time required to achieve these results is different for each frequency.

Beta becomes reliable in the first 30 to 45 seconds, with alpha following at 60-90 seconds. The intermittent nature of theta makes it the least easily established reliability, with 120-180 seconds required. Delta is reliable at about 120 seconds. Our lab tries to get 120 seconds of data when possible.

Reliability may be established another way, with split-half replication. This actually looks at the lability within the sampled data by looking for the invariant similarity of the repeatable parts of the two data sets, with the more variable parts looked at with less confidence.

The time span or length of the epochs selected determines the sensitivity to the slower frequencies. The Fourier transform has to have a completed waveform in the data epoch sampling for the frequency to be detected and quantified. A 1/2 Hz low frequency sensitivity is thus achieved only when 2 second epoch lengths or longer are sampled. There is a downside to too long an epoch. It increases the likelihood of including state changes and transients, though when clean state stable data exists in long episodes, it should be sampled.

Some equipment has preset epoch lengths and interactions between the sampling rates and the epoch length, which is problematic in sampling data flexibly. The epochs in some equipment are not able to be adjusted in time to ’slide’ past artifacts, making the data artifacting such that it is difficult to sample the clean data in a record with intermittents like eye movement or other transients. Careful selection of the equipment should precede entry into this field.

The artifacting is mostly concerned with eliminating the more common artifacts of eye movement and EMG, as well as movement or electrode artifacts. This cleaning of the data is part of the art of doing good qEEGs, though the science of adequate sampling and the assumptions of the Fourier must be kept in mind.

New artifacts introduced by the digital processing.

The digital recording and processing of the raw analog waveform of the EEG should be understood in technical detail to properly interpret the resultant maps and numerical tables of findings.

The EEG is digitally converted from the analog data by an analog-to-digital or A-D converter and a resultant digital dataset is derived. The digitizing sampling rate and the bit length of the computer data will determine the resolution of the resultant image and tabular datasets. The faster the sampling rate, the faster the frequency that can be resolved, with a minimal sampling frequency defined by the Nyquist principle as 2 times the frequency being resolved.

Proper reproduction of the EEG for visual perspective requires a more conservative sampling rate than the 2:1 Nyquist ratio. This greater than 2:1 ratio must be set by the individual’s preferences. Few would choose less than 128 samples per second, most would prefer 500/second to 1000/second. (The manufacturer’s with set buffer sizes for the epochs, like the Lexicor, will however need to look at the impact in loss of lower frequency sensitivity. For this situation 128 to 256 is the highest reasonable choice.)

The channel sampling should be simultaneous, to avoid remontaging error or slew. If not simultaneous, a faster sampling rates will reduce this error (as will burst mode sampling) reducing the phase or time base error to a minimum.

The bit length of the computer “word” processed by the CPU effects the amplitude resolution in qEEG, irrespective the sampling rate. The longer the bit length, the better the resolution ( a bit length of 12 is acceptable, but 16 is preferred. Older units will have 8 bit processing and are not fully adequate without further scaling adjustments).

The epochs selected during artifacting will all have an abrupt start and stop, without a zero voltage point at each end of all the epochs. This sudden voltage is seen by the FFT (fast Fourier transform) of the computer analysis as a square wave at the start and stop of each epoch. The resultant output of the Fourier is that all frequencies were present at that point to “reconstruct” the square wave. This is termed “leakage artifact”, or “Gibb’s artifact”.

The result of the Gibb’s artifact is that if a 10 Hz waveform was put into the FFT, a spectral plot of the output would have a rise of the baseline where all frequencies were used to reconstruct these abrupt starts and stops of the epochs. There would be a spectral peaking at 10 Hz, with a tapered response and a broadened base to the frequency plot.

To correct for the leakage or Gibb’s artifact, a “windowing” filter is used. The result of this windowing is the return to the baseline of the generally elevated spectral plot mentioned previously. There is a residual broadening of the idealized spectral peak at 10 Hz. This residual artifact of the broadening of the spectrum following the windowing is “smearing” artifact.

The windowing used in qEEG is usually a Hanning window. This filter slowly ramps up the start and ramps down at the end, to avoid the apparent square wave the FFT sees. Other windowing techniques are triangular, Blackman, Hamming, Meyer’s , with the lack of windowing occasionally referred to as a “rectangular window”. A full discussion of these details, contrasting the different styles of windowing is outside the scope of this chapter.

“Aliasing” is an artifact caused by a frequency source near the sampling rate (and above the nyquist sampling rate) so that a beat frequency is created as an alias of the source frequency/sampling rate interaction. Aliasing filters are used to control for this artifact in all modern devices.

Database issues

The datasets derived from the artifacted EEG are the starting point for the comparison of these data points to the databases used in qEEG analysis.

The databases are the means and standard deviations used to establish the significance probabilities for the observed measurements. These will be represented as Z-Scores, roughly these may be seen as the standard
deviations from the normative database of the data points in the dataset. They are calculated as the patient mean minus the database mean, divided by the standard deviation of the population.

Z-scores are reported in tabular form and may be mapped. Significance probability mapping ( a term derived by Frank Duffy, M.D. of Harvard’s Children’s Hospital) allows the interpreting individual to view the spatial distribution and extent of deviation in an easily discernible display. 1.96 Z-score deviation is equal to 2 standard deviations, and 3.08 Z-score deviation is three standard deviations.

Normative databases are constructed with highly screened normal individuals with an age range establishing the limits of the database. The database is constructed controlling for socio-economic and other demographic influences. Importantly, the databases must be established with different norms for male and female, to account for the significantly different neurophysiologic structures and rates of development of the male and female brains.

The database used should be selected to match the end use and population to be seen. The age range may be critical for those seeing children or the elderly. For others, the presence of multivariate stepwise discriminants used in determining the likelihood of membership in one of two or more clinical groupings will be critical in the selection.

For others, the presence of an eyes open database for use in clinical eyes open work, or the task related data will be critical.

The database from E.Roy John, Ph.D. of NYU’s Brain Reasearch Laboratory contains more than univariate measures, with multivariate parametric statistical evaluations of normal and many clinical subgroups. This has stepwise discriminant analysis associated with it. Discriminants for head trauma are also available from Thatcher. The Duffy database is used, though is not is commercial distribution now that the Nicholet BEAM instrumentation does not carry it. Sterman has an age limited performance based database available, with new databases such as Hudspeth’s coming into availability recently as well.

In using the discriminants, if available in a database, care should be exercised to assure the applicability of the discriminant to the client being evaluated. The client must fit the conditions that were set for the construction of the discriminant.

If a discriminant is set up to decide the likelihood of being a member of group A or group B, a member of another group, C, will be classified as A or B, not properly identified as another type, type C. This weakness of discriminants must be controlled in the selection and use of the discriminant, not after it has been performed.

Displays

The displays in the qEEG report will be presented as a progressive analysis of the data. The displays may be sorted into those that are closer to and farther from the raw EEG, being farther from the EEG as more statistical manipulations are performed. The farther from the raw data one goes, the easier it is to mistakenly interpret artifacts as real or misinterpret relational data.

To avoid these easy and eventually certain mistakes, the full visual interpretation of the EEG must precede any review of the analyzed data. Only then should this be followed by a review of the raw amplitude mapping, in as detailed a frequency display as possible. This may be followed by broad band analysis. The amplitude mapping should be followed by the power and then the relative power analysis. Only after this step-wise evaluation this should absolute or relative power Z score or other database comparisons be done.

Following the spectral evaluation, the statistically extracted measures of symmetry, coherence and phase are evaluated, without being as likely to make a mistake.

The presence of artifact should be expected in clinical work. In clinical work the luxury of prolonged recordings and rejection from a study due to artifacts is not present as it is in academic situations. The progressive step by step evaluation will control for these situations as best they may be.

To err may be human, but it should be controlled and accounted for with methodologic routine to the extent humanly possible.

The ascending reticular activating system stimulates the diffuse thalamic projection system and sets the general arousal level of the brain. With an increase in the CNS arousal level, there is an increase in the mean frequency of alpha and a decreased slowing. With decreases in arousal there is a slowing of the alpha, as well as eventually an increase in diffusely distributed slowing ( a mixture of diffuse lower voltage delta and theta, usually with a weak vertex prominence in linked ear montages).

When this diffuse slowing with slower alpha is seen, a vertex or central sensory-motor strip beta training will slightly speed up the alpha and decrease the slowing seen. A frontal beta minima seen in the data may respond to a more anterior placement for the beta training.

This increased beta training should correspond to a brain stem shift in RAS activation with increased norepinephrine level’s stimulating effect and results in increasing vigilance.

When this slowing and alpha pattern is seen, but with alpha intruding frontally (occasionally with less of the slowing) , the protocol should include some parietal “high alpha”, defined as 11-16 Hz in the classical EEG literature, but usually in NT, this is from 10 or 11 to 14 Hz.

This parietal high alpha training shifts the alpha mean frequency higher, decreasing the diffuse projection system’s frontal alpha and increasing the posteriorly distributed specific projection system alpha (though at a slightly faster frequency distribution). Often it is the slower alpha frequencies that are intruding frontally, and they are reduced with this shift.

Focal slowing

Focal slowing that is not an artifact (such as a pulse, electrode, electrodermal, eye movement or other artifactual source of slowing) should be evaluated by an electroencephalographer or neurologist.

The focal slowing may be from a tumor, ischemia, stroke, trauma, inflamation or other medical condition. The etiology should be identified prior to any intervention or consideration of NT.

Should the etiology be known, generally the reduction of the slowing and enhancement of faster activity improves the brain function following NT. Slowing is reported in specific learning disabilities and sensory processing problems (Chabot et al., SSNR Aspen,1997)

Faster alpha variants, not low voltage

The alpha frequencies may be faster than usual, sometimes corresponding with anxiousness or hypervigilance. These situations can present with complaints about attentional problems, with the hypervigilance acting as a source of increased distraction, but with the process differing from more usual ADD/ADHD presentations.

The faster alpha often has increased EMG associated with it. Patients with these findings seem to respond with paradoxical increased anxiety if EMG relaxation is tried without first addressing the hypervigilance of the faster alpha. The ‘letting down of the guard’ is anxiety producing.

Training the slower end of the normal alpha band parietally seems to have a strong positive effect on these individuals, with SMR training used by some therapists. If the EMG remains, the subsequent relaxation therapy seems to work without the increased anxiety following the NF intervention.

Frontal lobe disturbances

The frontal lobes inhibit distractions and inappropriate impulsively motivated behavior, control affective mood states and attentional states. The frontal lobes also set the motor strips tone via inhibitory control loops involving subcortical structures. The frontal lobe has general regulatory control over the entire rest of the brain.

In attentional and affective disorders as well as motor dyscontrol such as hyperkinetic disturbances the locus of the dysfunction is commonly the frontal lobe.

Work recently done on ADD/ADHD and affective disorders shows a variety of frontal disturbances seen with the qEEG. These varieties include slowing in theta, alpha intrusion and even excess beta (Chabot et al, 1998 SSNR, Austin). Non- systematically replicated research showed the qEEG to predict the response of the patient to medications. The theta types responded to stimulants, with the frontal alpha, especially when the mean frequency of alpha was slowed, responding to amphetamines. Other researchers show some of the alpha types to respond to SSRI type antidepressants (such as the OCD responsive type in work by L. Prichep and the depressives with frontal alpha in work by S. Suffin and Emory).

In NT, the qEEG may be used to adjust the intervention, with frontal theta responding to beta protocols with suppression of the excessive theta. The alpha frontal types respond less well to frontal alpha downtraining than to posterior high alpha training with concurrent frontal beta training. The frontal beta type seeming to respond to normal frequency alpha training posteriorly and frontal beta suppression if the beta is still excessive (with any slowing noted suppressed as well). The frontal beta excess type is often a difficult patient in my experience.

Frontal beta minima can be seen in frontal disturbances and seem to respond well the NF intervention. Areas commonly seen are F7 in attentional problems, F8 in impulsivity and F3 or Fp1 in depression. The locus or even presence of a minima is difficult to predict behaviorally, as other disturbances of function may yield the same behaviors.

Right frontal training and frontal symmetry

The commonly held belief that in NF the right frontal lobe should be avoided needs to be explored and understood before discarding it.

Frontal alpha and beta interhemispheric ratios seem to correspond well with the perceptual style of the subject. Right hemispheric dominant (more beta and less alpha than the left) subjects have a “glass is half empty” perception and a lower mood state, or more depression.

NF with the frontal lobes needs to keep the dominance on the left, or to establish such a dominance to avoid deteriorating mood states and perceptual styles in the client.

This has led some to avoid the right frontal training, or frontal training in general. This would be a drastically limiting elimination of potentially efficacious NF intervention on brain function, given the importance in brain pathophysiology and functional regulation of the frontal lobes.

If care is taken to measure and assure the desired lateral symmetry, it is my experience that the frontal lobe training on either side may be done without significant difficulty. Without this knowledge and care applied to protocol considerations, it could be a large source of client dissatisfaction.

Spindling excessive beta

This pattern has been reported to be associated with ‘cortical irritability’, viral or toxic encephalopathies and in epilepsy. It has a classically defined higher voltage beta occasionally even exceeding 20 microvolts. This abnormal beta is seen in waxing and waning spindles over the effected cortex. This pattern is seen in less than 10% of the ADD/ADHD and affective disordered population, but when seen, it is an important finding.

I have seen this excessive spindling beta in areas associated with pre-epileptic auras. In one case it was seen occipitally during visual auras and in another fronto-temporally with auras effecting the client more subtly as a smell or even a remembrance.

This pattern responds very badly to any beta training, exacerbating the symptom complex. Beta training is strongly contraindicated.

Beta suppression directly in the area of concern has shown good clinical response. The band of frequencies to be suppressed should be selected based on individual profiles, not by standard bands. I have seen broader bands like14-22 Hz or bands as narrow as 14-16 Hz in excess, with higher 20-30 Hz beta occasionaly involved as well, with many variations.

The customizing of these interventions would be very difficult, if not impossible without the qEEG to provide location, distribution and frequency range information to the NF practitioner.

Areas without significant activity
seen in any EEG band

The EEG can be seen with areas of decreased amplitude, not just a beta minima, an area of general amplitude minima. This phenomenon is seen well in mapping and has been reported in areas of cortical dysfunction.

The EEG requires the generation of alternating currents for any voltages to be measured in the EEG. This requires the activity of neurons in an area with the increased blood flow and glucose metabolism associated with these cellular processes. The flowing of blood in the brain is regulated by the concentration of bicarbonate ion measured as PCO2, a metabolic by-product of the burning of glucose and the generation of energy within the Krebbs cycle. The ADP/ATP cycle within the mitochondria is the site of this electro-chemical interplay.

In careful work done by Pribram, the time frame for these events has been studied, with the slower DC activity preceding the cellular action potential. This DC system has been used since the 1970’s in Europe in NF, showing that the shift to electro-positivity can even stop an epileptiform discharge from occurring (N. Birbaumer).

This phenomenon also can be seen in qEEG as an area of decreased voltages in all bands, progressing from beta through alpha to the slower frequencies. The area is effectively shut down and is not functioning. When the brain’s electropositivity increases enough, the AC activity seen in the EEG is inhibited.

These areas have been observed frontally in attentional and affective disorders (Gunkelman, SSNR1997), and are reported in observations of individuals who have been brain washed and have given the locus of control over to others (personal communication with Brownback, Mason and associates).

Training these frequency “dead areas” is something newer in the field, but it seems to respond to beta training, suppressing any slower frequencies that may be still present. Beta is correlated highly with PET measurements of metabolic activity (I.A. Cook, A. Leuchter et al, 1998 UCLA)

Generally low magnitudes

The occurance or a low voltage EEG is considered a normal variant when it is a low voltage fast EEG. When the low voltages appear slow however, it is a diffuse and non-specific abnormality. The difference between the two patterns is somewhat more qualitative than quantitative. The morphologic presentation differs more significantly than the magnitude differences in the quantitative analysis.

The magnitudes are all low, and in relative terms the power will look slowed in both cases, though the faster morphologic pattern is a normal variant. When a low voltage slow pattern is seen and is confirmed not to be drowsing, it should be evaluated for metabolic, toxic or other diffuffuse encephalopathies such as degenerative or post hypoxic etiologies.

The low voltage slow type is reported in dementias as an early EEG change. This seems to respond to high alpha training from 10 or 11 to 14 Hz. This is the same EEG effect as nootropic medications (smart drugs) will provide.

The low voltage fast type usually corresponds with anxious, nervous and hypervigilant individuals. Though not pathognomonic, it is commonly seen in alcoholism and alcohol free members of families of alcoholics with a strong family history.

Interestingly, in research on this pattern, it is shown to respond to alcohol by suddenly having alpha that is well formed. The alpha will slow and rhythmic slower activity will increase if higher doses are given. The state is reported in euphoric terms by the research subject. This euphoria is also reported associated with alpha induced by opiates.

The low voltage fast pattern responds well to alpha training with a normal alpha distribution of 8-12 or 9-11 Hz. The learning curve for this is well established, having a fifth order curve fit. The phases of the experience are well defined by the curve.

There is an initial increase in alpha due to the habituation to the clinical setting, with subsequent decreases in alpha during the active attempts at controling the alpha. These phases are followed by the release of the active attempts and a return to the habituated level. This is followed by passive volitional attempts and the eventual acquisition of voluntary control seen as the exponential increase at the end of training.

The subsequent alpha/theta training is commonly used by neurotherapists in these cases when there is a concurrent addiction or intense life stress or trauma by history.

Temporal lobe alpha

When alpha is seen in the temporal lobe, it can be from a variety of causes, indicating that a more complex neurofeedback protocol response may be needed.

The alpha from old head trauma is usually a faster alpha variant, adjusted for the individuals alpha ‘tuning’. This is seen over a year or 2 from the time of the trauma, after the acute healing and swelling have long dissipated. It replaces the slowing which may be seen initially.

Temporal alpha may also be an effect seen in response to a decrease in ipsilateral frontal lobe activity. The decrease in uncinate fasciculus or inferior longitudinal fasciculus stimulation from the frontal lobe allows the temporal lobe to be idle. This usually will be seen with one of the frontal lobe patterns discussed earlier.

The temporal idling may be cleared up with the direct frontal work discussed earlier, but may require lower band beta training directly on the temporal site.

Strong alpha at T5 or T6 can contaminate the ear references, yielding a false image of frontal alpha in the qEEG. The ears, having alpha present and the frontal lobes without alpha are compared in the differential amplifier. The amplifier will show alpha in the frontal channel falsely, which has to be controlled for by using a variety of montages in reviewing the data. The frontal alpha will not be seen with sequential or non-contaminated reference montages, such as Cz or in more sophisticated equipment the ‘common average’ or even the ‘Hjorth’ montage may be used to give a more pure look at the data.

The temporal lobes seem sensitive to excessively fast beta training, with 14 Hz training, 12-15 Hz, 14-16 Hz or other lower band beta used more commonly than a higher frequency intervention due to this sensitivity. In my experience, these bands seem to be best adjusted based on clinical response, not to any obvious spectral loss in the CSA or amplitude mapping.

The post traumatic faster alpha seems to respond to coherence training, seemingly reconnecting the functional relationships. This requires the use of qEEG coherence measurements, as without this, the training sites and bands would not be evaluated or selectable based on any objective criteria.

A note on coherence and phase

The cortex is full of neural connections, all electrically active, though not all seen with the EEG. The EEG is measuring summations of radially generated action potentials from pyramidal cells, not the laterally oriented cortical-cortical tracts.

The raw EEG is blind to nearly 2/3 of the electrical activity of the cortex. Much of which is seen by the Magnetoencephalography (MEG), a measure of the “magnetic” activity of the brain (actually lateral current flow of intracellular activity). The MEG is however blind to the extracellular potentials arranged radially which comprise the EEG.

The longitudinal myelinated fiber tracts are the high speed web of the cortex, not the organized subcortical radiations of the thalamo-cortical systems or other subcortical-cortical projection pathways. This network of local arcuate, longitudinal, fronto-temporal (uncinate) and interhemispheric collosal tracts are invisible to the raw EEG.

The activity of this invisible network, however, can be inferred from the coherence and phase relationships between areas “hooked up” by the network. This spatial distribution, “connectivity” reflection of the subsurface activity is obviously quite complex. The covariance of power at two sites (coherence) or their covariance in time (phase) give measure within the EEG to activities within this invisible network.

The view of the raw coherence and phase data yields little to the non-expert. When compared to an age and sex matched normative database, displayed as a Z-score, the magnitude and direction of variance, including the significance of the variance become much more evident.

There are two types of phase relationships seen in qEEG, conducted phase differences and propogated phase differences. The propogated phase is seen where a focal phase reversal indicates the source of an EEG phenomenon. The conductive phase is simply the time delay due to propogation along neural pathways.

The phase measurement reflects the correlation of covariance in time of activity at two sites. This temporal relationship can be slowed by damage to tracts due to demyelinating, structural or toxic/metabolic influences. The phase relationship may be faster with increased nerve conduction velocity, or by volume conduction through fluids and as a field effect.

The simultaneous projection to various cortical locations of synchronized volleys from thalamo-cortical radiations seen during thalamic activity paced by the ventral-medial or reticular nucleus is a “highly connected” state, with high phase synchrony. This unique synchronized state is predictive of meditative expertise in Zen meditators (Gevins et al.). It involves progressively generalizing increased phase synchrony in alpha, followed by increases in theta synchrony (possibly representing slowed alpha).

Similar findings have been seen in Yogic meditators, though with differences in habituation to repetitive sensory stimulation. The Yogic meditators habituated faster than normal and the Zen meditators failed to habituate. Interestingly, this reflects the philosophical views of the two meditative techniques. The Zen meditation aspires to novelty of experience while the Yogic traditions emphasize the imaginary nature of the external reality.

The phase relationship is measured as difference in the degree of arc of tangents of two waveforms, measured at a point in time. This is reported as degrees of phase shift from 1-180 degrees, or time synchrony of the waveforms, reported in milliseconds.

Often confused with phase is the EEG measure of coherence. The confusion comes from the use of the term “coherence” as an adjective descriptor for phase. This misuse of a technical term as a descriptor in the same field of interest is problematic. The term “phase coherence” should be eliminated and replaced by “phase synchrony” or, most properly, by specifying a phase relationship from 0-180 degrees of phase shift or by giving the Z-score deviation from normal. Please retain the use of coherence for it’s technically proper role within this field.

Coherence is the cross correlation of the power (or amplitude) of activity at two sites. Sites that covary highly are presumably processing related cortical/subcortical volleys or have a high “connectivity”. This is reported as a value from 1 to 0 (on some older equipment, as a value from 100% to 0% coherence).

The cortical-cortical long fiber connections compete with the shorter connections. As Bob Thatcher says, “The close siblings can speak to each other, but not to their distant cousin at the same time.”. This is called the two compartmental model of coherence.

There is a third compartment I believe, the subcortical-cortical compartment, which would explain the observation of high coherence locally having decreased connections locally. The coherence is from the subcortical source being connected to both sites, not the cortical-cortical compartment’s connectivity.

Coherence is most easily viewed as morphological similarity, with correction for absolute magnitude and is irrespective of the time synchronization. When two waveforms are shifted in time until maximal coherence values are attained, the time base shift is the phase delay of the waveforms.

Structural damage produces reliable, consistent patterns of change in phase and coherence patterning in the qEEG, though functional influences account for variable findings. This variance makes interpretation of the data subject to having to rely only on strong patterns of variance, not isolated findings.

The proper conservative interpretation of these data must also be viewed in the context of the entire constellation of findings from the rest of the patients EEG, qEEG and clinical presentation.

When training changes in coherence or phase using neurofeedback, over-training (shifting beyond a normal relationship range) needs to be avoided. This requires merely setting a proper normalized goal or value as a training target.

There is currently what I believe to be an inappropriate practice in coherence NT. This is the use if a single channel EEG machine used sequentially (old term bipolar) with the assumption that this will feed back coherence. This assumption suggests that increased highly coherent activity cancels, thus decreasing the amplitude of the channel. Thus training amplitude could train coherence.

This is a grossly false model of coherence feedback. This model assumes the phase of the activity to be synchronous. It also assumes that the synchronous activity has amplitude equivalence. In reality the coherence calculation equally weights amplitudes, correcting for asymmetries, the amplifier does not. The calculation of coherence is not time locked, measuring covariance independent of the time locked nature of the amplifiers response.

I believe we are not yet in the presence of enough information about the clinical impacts and needed protocol controls to make full use of the clinical application of these types of training. The clinical application of these techniques is highly fruitful ground, needing systematic clinically valid research and protocol development.

More fully conclusive research is needed based on the ongoing application of this training before rules for intervention are presented.

I have nothing against the careful clinical use of coherence or phase training. When the more “traditional” neurofeedback applications fail, these phase and coherence based interventions should be empirically tried. With more trials using the proper measures, design and controls, advances will be made.

The departing caveat

Now that I have completed this codification of some of the interventions seen currently in qEEG based NT, I can say without question that this will be a source of future embarrassment. To capture a snapshot during a field’s rapid development stage is a guaranteed red face in the future. If you don’t believe this, look at your own childhood photographs!

So, please be gentle with your ridicule, and don’t mistake this art to be a science… not just yet.