Addendum to:

 The Neurofeedback Book

An Introduction to Basic Concepts in Applied Psychophysiology



Lessons from Clinical Experience

Using NFB + BFB



Anterior Cingulate –Thalamus – Beta Spindling – SMR—Slow Wave Excess – RSA

A Few of the Keys to Understanding Successful Intervention


Michael Thompson M.D. & Lynda Thompson Ph.D.





Based on work done with more than 2000 clients over the past 12 years, we wish to share our best guesses about why we have successful intervention using neurofeedback (NFB) and biofeedback (BFB). This is constantly being refined and expanded as we learn more.


Goal: Provide a brief response to the question:

ŇWhat have been the common factors in clients helped using NFB + BFB?Ó


I.            Areas of Work

ADD Centre/Biofeedback Institute of Toronto does assessment and intervention to:



A.     Optimize Performance with executives, students and athletes.

This has become an important area of the ADD CentreŐs work.  The goal is to:

ADD Performance

ADD Confidence

ADD Efficiency

ADD Productivity


In sports, work and academics, an alert, focused, relaxed approach and a rapid recovery from stress characterizes the winners. Neurofeedback when combined with Biofeedback and Metacognitive Strategies offers high functioning clients a uniquely effective and intellectually satisfying approach to meeting these objectives.




B.        Ameliorate Symptoms of Common Disorders

Clients age 6 - 65 who have Disorders including:

               i.     ADHD

              ii.     Learning Disabilities &/or Memory Difficulties

            iii.     Autistic Spectrum Disorders, especially AspergerŐs Syndrome

            iv.     Affective Disorders (OCD, Anxiety, Panic, Labile emotions and Depression).

Plus a smaller number with:

              v.     Movement Disorders (TouretteŐs, Bruxism, Dystonia with ParkinsonŐs)

            vi.     Seizures disorders

           vii.     Head injuries

         viii.     Headache, Chronic Pain (relatively infrequent at our Centre)


Co-morbidity is common; e.g., ADD + LD; ADD + AspergerŐs in children; ADD + Anxiety or ADD + Depression in adults




II.          Symptoms which are usually in common to all presentations


Deficiencies in attention and concentration:

á      may be combined with anxiety and tension

á      may also be accompanied by movements that are not consciously directed or wanted (hyperactivity, tics, bruxism, seizures)





III.       Common EEG and Stress Assessment findings

A. EEG Findings


   i.       ADHD

Excess slow wave activity; that is, sustained, high amplitude theta (3-7 Hz) and/or excessive low frequency alpha (8-10 Hz) in central and/or frontal regions, combined with a dip in SMR (13-15 Hz) (Remember that short bursts of synchronous 6 Hz  theta [probably hippocampal in origin] is good and is associated with encoding and recall of memories.)

And /Or

High amplitude, high frequency Beta (usually above 20 Hz)

Occasionally Beta Spindling, usually between 19 & 37 Hz but occasionally at a lower frequency between 14 and 18 Hz.


LORETA may show increased slow wave activity in prefrontal regions and the anterior cingulate or, alternatively, beta spindling may be seen to originate in the anterior cingulate, Brodmann Area 24 (usually accompanied by symptoms of affective dysregulation, anxiety, panic and / or anger)

Note:  The LORETA findings must be considered preliminary as these observations are from a small number of cases.


ii.              LD & Memory Problem

Dyslexia: decreased activity (high amplitude slow wave (3-10 Hz) over WernickeŐs area. LORETA may show it to originate in the Insula (Left)

And / or

Low amplitude beta (14-18 Hz)


Non-verbal Learning Disorder: (NVLD) decreased activity (high amplitude slow wave (3-10 Hz) in the parietal region (usually PZ, P4, T6 areas). LORETA may show some of this to originate in the Cuneus.


Memory Difficulties:  May be partially due to ADHD (working memory weak in those with ADHD) and show the same EEG patterns. In addition, usually find with LORETA that there is slowing in parahippocampal area– this affects short-term memory - and/or posterior cingulate (Brodmann 31) which may relate to long-term memory and the synthesis and integration of sensory information.


iii.            AspergerŐs Syndrome:

a.     With Sensory Aprosodia T6 may show increased slow wave activity (theta and/or lo-alpha) compared to T5. LORETA findings may include  Anterior Cingulate –Brodmann area 24 - activity outside data base range. This is often  slowing and spindling.

b.     Comodulation and Coherence findings may show disconnect between the right parietal region, P4, and T6 and the left frontal lobe.

c.     With Motor Aprosodia may see slow wave activity at F4 and/or beta spindling. (Beta spindling may be seen at Fz and appear, using LORETA, to stem from the anterior cingulate)

 Autistic Spectrum Disorders

Usually observe high amplitude slow wave activity at Cz and Pz. There may also be slowing throughout the parietal region. LORETA may indicate the origin of much of this slowing to be in the area of the cuneus. The cuneus is the centre of the sensory association cortex: visual, auditory, kinaesthetic. It is a key area for the integration, synthesis, organization and manipulation of sensory input.


iv.            Affective Disorders (OCD, Anxiety, Panic, Labile emotions and Depression) and Slow Recovery from Stress:

Depression Usually observe less activation in the left frontal lobe as compared to the right. This is consistent with Richard DavidsonŐs research on emotions.

Panic Disorder May observe high activity (increased beta activity) in the right frontal lobe compared to the left.

Anxiety, Panic, Labile Emotions: These symptoms usually correspond to a dip in SMR amplitude between 13-15 Hz and this may correspond to the symptom of ruminations and be accompanied by very high amplitude, high frequency beta (often spindling beta) between 23 and 35 Hz.

Anxiety and emotional intensity may also show a rise in beta activity 19-22 Hz. This may correspond with low amplitude high alpha (11-12 Hz).


Those with the above conditions, and also people with slow recovery from stress, show psychophysiological changes. See ŇStress AssessmentÓ below.


v.              Movement Disorders (TouretteŐs, Bruxism, Dystonia with ParkinsonŐs)

The key finding can be a very marked dip in SMR between 13-15 Hz


vi.            Seizures:  The key finding is the characteristic spike and wave pattern. One may also see bursts of very high amplitude synchronous theta.


vii.           Head Injury: A right-left ÔdisconnectŐ is often observed. There may be high slow wave (initially delta, later low frequency theta) at coup and contra-coup sites. Usually the EEG is low amplitude. Alpha is lower amplitude in occipital-parietal region (amplitude closer to that of frontal and central region). The eyes closed peak alpha frequency drops (slower).


viii.         Headache, Chronic Pain: (Infrequent at our Centre) Abnormal psychophysiological variables, low amplitude high alpha (11-12 Hz) and SMR, high amplitude high frequency beta. May also see high amplitude theta.


Stress Assessment

Clients seeking self-regulation skills for stress management usually show: low peripheral skin temperature, muscle tension, no synchrony between heart rate variability and respiration, respiration shallow, frequent and irregular. All of these variables worsen and electrodermal response (EDR) rises with stress. (Variables may differ in chronic stress – flat EDR, high skin temperature.) In these adults full recovery to baseline does not occur in the  2 minute recovery period after each test stressor.


Example case showing EEG, brain map, LORETA – see appendix, Case #1


IV.        Major goal of training in all of these clients:

The conscious/unconscious ability to sustain a mental state of: Relaxed, Calm, Focused Concentration with the ability to switch from broad open-awareness to specific focus combined with using appropriate problem solving strategies without interference from unnecessary circular thinking or actions.




V.          Training Program

NONE of the following concepts  should ever be used as a ÔprotocolŐ. We never use that term. How to assess the client and decide on the appropriate training for each individual is the subject matter of our Professional Training Workshops.


In our experience, the major components of the training programs, although individualized according to EEG and Stress Assessment findings, seem, on review, to always include:


a.     Children (Note: exact frequency ranges and site(s) individualized)

Decrease Dominant slow wave between 3 and 10 Hz

Increase SMR (usually 12-15 or 13-15 Hz)

Increase Beta (usually 15-18 Hz) while carrying out academic tasks

The site is usually Cz for some of the sessions

C4 for some sessions when impulsivity plus hyperactivity are a major problem

WernickeŐs area when dyslexia is a problem (Site close to centre of ÔXŐ formed with lines drawn between  C3-T5 & P3-T3)

Other areas as dictated by QEEG findings


metacognitive strategies



b.    Adults  (Note: exact frequency ranges and site(s) individualized)

Decrease Dominant slow wave between 3 and 10 Hz

Increase SMR (usually 13-15 Hz)

Increase Beta 15-18 Hz while carrying out academic tasks


Decrease beta spindling when present. QEEG determines frequency (usually above 20 Hz) and site. (Only occasionally is this a factor with children)

Increase high alpha 11-12 Hz if low


Encourage diaphragmatic breathing at 6 breaths per minute (BrPM) with synchrony with high amplitude heart rate variability. This must be accompanied by a decrease in muscle tension, normal skin temperature, and normal fluctuations of electrodermal response (EDR).



Metacognitive strategies




Specific Training Programs:

               i.         ADHD

As above but usually train up SMR at both Cz and C2 or C4 when impulsivity and hyperactivity are present.


              ii.         Learning Disabilities &/or Memory Difficulties

As above but train over the site responsible for that area of learning if EEG shows decreased activity in that area: e.g., dyslexia – training to improve reading comprehension would be over the left temporal-parietal junction –WernickeŐs area and the Insula.


            iii.         Autistic Spectrum Disorders, especially AspergerŐs Syndrome

As above for the ADHD symptom component but add FCz training to decrease dominant slow wave and beta spindles (when present) and train up SMR. Training up the SMR rhythm is thought to help stabilize the cortex, particularly in the area of the anterior cingulate (BrodmannŐs area 24). Also train down the dominant slow wave and activate T6 area to help with the sensory aprosodia symptoms. Train at F4 to normalize the EEG if abnormal - for motor aprosodia symptoms.


            iv.         Affective Disorders (OCD, Anxiety, Panic, Labile emotions and Depression):

á      For all these disorders, normalize the EEG abnormalities stemming from anterior cingulate dysfunction (including beta spindling).  Also normalize maladaptive psychophysiological reactions to stress using biofeedback with particular emphasis on cardiac variability.


á      OCD (Obsessive Compulsive Disorder): Pay particular attention to anterior cingulate with training at FCz. Train down dominant slow wave and the high frequency beta.


á      Anxiety and Panic: Pay particular attention to right frontal abnormalities (train down over-activation: spindling beta, high frequency beta). Train up SMR (C4) and 11-12 Hz alpha (if low) and pair this with improving psychophysiological variables using biofeedback.

á      Depression: Pay particular attention to increasing left frontal activation relative to right frontal. This may be assisted by:          (1.) two channel training with a common reference to Cz (or linked ears to decrease left frontal low frequency alpha compared to right); (2.) decreasing left frontal dominant slow wave while increasing low beta 14-17 Hz; (3.) Increasing a competing high frequency alpha (11-12 Hz) generator by increasing this hi-alpha at P4 or P2.


              v.         Movement Disorders (TouretteŐs, Bruxism, Dystonia with ParkinsonŐs)

Increase SMR (C3, Cz, C4). ÔPairŐ this SMR training with biofeedback to decrease sympathetic drive by training diaphragmatic breathing (6BrPM), decreased EMG, increased peripheral skin temperature and normalization of EDR. (Combination may affect muscle spindles.)

            vi.         Seizures disorders

Increase SMR (C3, Cz, C4). Decrease dominant slow activity near seizure origin.


           vii.         Head injuries

á      Decrease dominant low frequency slow wave near coup and contra-coup sites.

á      Normalize (usually increase) coherence and/or comodulation which may, for example, be showing a ÔdisconnectŐ between hemispheres.  May also show anterior/posterior disconnects.

á      Increase posterior alpha – may involve Ňbrain-brighteningÓ using Pz site for training up 11-12 Hz alpha.

         viii.         Headache, Chronic Pain (Infrequent at our Centre)

á      Biofeedback to normalize psychophysiological variables while raising SMR (We have hypothesized that this combination may affect muscle spindles and be a factor in helping fibromyalgia.)

á      Increase high alpha and decrease high beta when present.

á      Decrease the dominant high amplitude theta when present (See Stu DonaldsonŐs work on fibromyalgia).



VI.       Results of Training:

Overview: At the ADD Centre, all clients attend initial and progress testing interviews in which both subjective and objective variables are evaluated and compared. Their symptom history is reviewed and questionnaires are answered. Pre and post training objective testing includes the QEEG (single, two channel and 19 channel), two continuous performance tests (TOVA (Test Of Variables of Attention), IVA (Intermediate Visual Auditory test for attention variables). In addition, a stress test for psychophysiological variables is recorded for adults, and psychometric testing is done for children and most adults.


a.  Individual Pre –Post training Evaluation

All clients demonstrate significant improvements in one or more key areas if they have a supportive family and if they attend a sufficient number of sessions. This varies with presenting symptoms – 40 sessions for most adults, 40-60 for children but more if comorbidity with LD, AspergerŐs, autism, seizures.


b.  Reviews of pre – post testing carried out at the institute:

  1. ADHD: average IQ gain > 11 pts, symptoms normalized, significant academic gains
  2. Academics / Learning:  > 1.5 year gain in 5 months (40 sessions) in reading comprehension
  3. Autistic Spectrum, mostly AspergerŐs (n= 126) – significant improvements on all measures of social functioning
  4. Normalization of EEG patterns


c.  Long  Term Follow-up Review in progress:

Random selection of clients who completed their training (long term, 2- 10 year follow-up).


Brief Summary of Results of ADHD, AspergerŐs, and Dyslexia Reviews - See Appendix #2


VII.      When will Clients NOT do as well as expected?

Overall one can reasonably expect that clients will significantly improve and meet the majority of their objectives provided that the following criteria are met:  Clients must be chosen who exhibit one or more of the problems outlined above. These clients should complete an adequate number of sessions. The client who has a good family (or community support) for doing the program will do better.

It is unwise to expect that clients will meet their objectives if:

1.     They do not do a sufficient number of sessions. (This is usually about 40 sessions for ADD without hyperactivity) and more than 60 for more difficult conditions such as Autistic Spectrum disorders, severe learning disabilities, and so on).

2.     There are major family problems – separation and divorce proceedings are occurring, one parent does not support the child coming to the program, and so on.

3.     Borderline personality disorder in the client or parent of a client.

Motivation of child, adolescent, is important but that is usually our job with the parents to find ways to help the child become motivated.



VIII.   Conclusion

The anterior cingulate is the ÔhubŐ of affective control and of decision-making. It is also, through its connections to the extrapyramidal system and the basal ganglia, involved in movement disorders. It is therefore central to the dysfunction found in the disorders listed above.  The thalamus is the origin of the rhythmic waves (alpha, theta, SMR) that are one key to NFB training paradigms. Beta spindling signals cortical instability. The stability of the cortex may be improved with SMR training. ŇPairingÓ the NFB training with reduction in sympathetic drive and normalizing the limbic-cortical-hypothalamic-adrenal axis using BFB (especially respiration and heart rate variability training) accelerates normalization. The combination of NFB, BFB and strategies improves client self-regulation and therefore performance.


Triple A



The key to understanding the effectiveness of NFB combined with BFB and metacognitive strategies in many of the above areas is linking our understanding of the Amygdala, Hypothalamus, Pituitary, Adrenal (AHPA) axis with the key control centres that link it to conscious cognitive response systems (thalamic-cortical circuits and the anterior cingulate connections to both cortical and sub-cortical structures.

The anterior cingulate, affect division: Brodmann areas 24, 25, 33, is part of the rostral limbic system, a matrix of structures engaged in similar functions. This matrix includes: amygdala, periaqueductal grey, autonomic brain stem motor nuclei, ventral striatum, orbitofrontal and anterior insular cortices. This system assesses the motivational content of, and assigns emotional valence to, internal and external stimuli. It regulates context-dependent behaviours. It regulates emotional responses and behaviour through the autonomic nervous system (ANS), endocrine functions and conditioned emotional learning.


The AHPA axis (see Thompson pp106-108) controls the automatic and unconscious aspects of the stress response. This response system is controlled by interaction between brain stem nuclei (including autonomic nervous system nuclei and the locus coeruleus) and the amygdala-hypothalamus-pituitary-adrenal axis (AHPA) with the frontal lobes and hippocampus having a major influence through connections to the amygdala and hypothalamus.



Remember, Neurofeedback Training is a Learning Procedure, it Takes Time

With adults or parents, the trainer can use analogies that remind them that it takes time and, with enough training, things should finally fall into place. Learning a new skill takes time and practice. Second, be patient, in the early stages it may seem that very little is changing. Some parents enjoy sayings such as: there are orchids that can take 9 years to bloom; or, ninety percent of the growth of the Chinese Bamboo tree is in the fifth year!  Third, donŐt give up.  If you stop pumping the old hand water pump when you need water you must re-prime it and start all over again.



APPENDIX #1  CASE EXAMPLE:     Female, Age 42 (Two Doctorates) Combined ADHD, AspergerŐs, Panic, Learning Disability


EEG findings included: with eyes open on and off task, 4-8/16-20 ratio >3, dip in EEG amplitude at 14 Hz. and a marked rise in amplitude at 21 and 30 Hz (spindling beta) observed at Fz, Cz and F4.


With eyes closed the Laplacian montage, relative power, brain map corresponded to beta spindling seen at 23 Hz in the raw EEG


The LORETA analysis in the eyes closed condition showed 23 Hz beta activity at 1.6 SD from the Neuroguide (NG) data base mean originating in the anterior cingulate, Brodmann area 24.

EXAMPLE #2:  Male, age 8, diagnosed elsewhere as Autistic. Not conversing.

Constantly looking around and moving. Extremely difficult to control his behaviour which occurred in frequent bursts of movement. At these times he did not appear to understand either verbal or non-verbal communication.


EEG findings: Many abnormalities were observed in the raw EEG, brain maps and LORETA. These included: Spike and wave activity at T3. With LORETA this appeared to originate in the left Insula (>6 SD from the data base mean at 3 Hz). This site may account for the language comprehension difficulties. Later in the record bursts of 3 per second spike and wave activity were observed at T6. This site may explain the sensory aprosodia symptoms.














Unfortunately the site of the seizure activity appeared in the above to be the insula.














 (A question arises, could the anterior commissure be part of a linkage between these seizure sites?)                                                


In addition to the foregoing, the anterior cingulate, Brodmann area 24, also demonstrated EEG abnormalities. In the LORETA shown below, 12 Hz is >3SD above the NG data base mean.


Bursts of 30 Hz beta spindling were also observed in the raw EEG in the left frontal region and this deviation from the neuroguide data base mean is seen in the LORETA image below.


All results are significant P = > .01  (only exception in IVA noted below)

Data collected after 40 sessions of training. Training proceeded for a further 20 sessions for many of the children due to serious initial presentation.


Wechsler IQ Scores

i. ADD  n = 103








Full (+ 11)




Verbal (+ 11)



Perf. (+ 12)




ii. AspergerŐs   n = 36









Full Scale IQ




Verbal IQ




Performance IQ






i. ADD  n = 139




Attent.(+ 16)



Impuls.(+ 18)



React.T.(+ 5)



Variab.(+ 22)




ii. AspergerŐs  n = 77;        Initially more anxious and careful than ADHD clients.













reaction time









AspergerŐs Clients  n = 63









Auditory Response Control




Visual Response Control not significant




Auditory Attention




Visual Attention









Wd.R.(+ 7)



Spell.(+ 7)



Arith.(+ 8)



ii. AspergerŐs   n= 55



















i.  ADD Clients: 110 of 111 showed significant improvement in theta / beta ratio


ii.   AspergerŐs Clients  (4-8 Hz / 16-20 Hz)       n = 84











Pw ratio (4-8)2 / (13-21 Hz)2





Uv ratio  4-8 / 16-21





Uv ratio  3-7/12-15






NOTE:  With some Child clients we could not obtain valid EEG in first session.

Post-training testing:  Not done on all adult clients because they often stop training because they have achieved their objectives and/or they are moving.


AspergerŐs Clients













Australian AspergerŐs Scale












DSM IV checklist






ConnersŐ Global Index






Further decreases were seen when >40 sessions done. (with >60 sessions, virtually all AspergerŐs questionnaires were in the normal range.)


Selected References:

      Devinsky, Orin., Morrell, Martha,  Vogt, Brent, (1995) Contributions of Anterior Cingulate Cortex to behaviour. Brain, 118, 279-306


     Monastra, V. J., Monastra, D., & George, S. (2002). The effects of stimulant therapy, EEG biofeedback and parenting on primary symptoms of ADHD. Applied Psychophysiology and Biofeedback, 27(4), 272-250.


      Sterman,  M. B. (2000). Basic concepts and clinical findings in the treatment of seizure disorders with EEG operant conditioning. Clinical Electroencephalography, 31(1), 45-55.


Thompson L. & Thompson M. (1998). Neurofeedback Combined with Training in Metacognitive Strategies: Effectiveness in Students with ADD, Applied Psychophysiology and Biofeedback, 23(4),  243-263.


Thompson, M. & Thompson, L., The Neurofeedback Book: An Introduction to Basic

Concepts in Applied Psychophysiology, Wheat Ridge, CO: Association for Applied Psychophysiology and Biofeedback.  (Available through: