Author: K. Thornton, PhD -- Source: Article Source

Abstract
Understanding of the functioning of human brain has been a significant goal of modern neuroscience. Concomitant with that goal is the desire to improve the brain’s native abilities. Previous research in the Neurotherapy (EEG biofeedback) field has amply demonstrated the ability of classical operant conditioning methodology to accomplish this task in reference to a set of broad cognitive abilities as measured by IQ tests.

It is well known, however, that different individuals possess abilities in differing degrees, which must be reflected in the underlying physical process. The quantitative EEG (qEEG) provides relevant (to human cognition) measures of the brain’s electrophysiological functioning, which can be addressed by operant conditioning methodology. To understand how a specific ability works, the problem of auditory memory functioning was evaluated by development of a qEEG activation database.

The information obtained in that research was employed in Neurotherapy treatment protocols of a group of brain injured and learning disability subjects, and resulted in respective gains of 119% (brain injury) and 400% (learning disabled) in auditory memory functioning.

Numerous methods have been employed in the cognitive rehabilitation of the brain injured as well as psychoeducational interventions with the learning disabled and special education population. These approaches have not fared as well as the Neurotherapy approach with respect to their ability to increase native abilities in these groups.

What is the qEEG?

The quantitative EEG (qEEG) is a digitized recording of EEG analog data. Interpretations of the traditional EEG were hampered by the limitations of problems in human judgment reliability. With the advent of the qEEG came several advantages in understanding the EEG signal: 1-the data could be analyzed mathematically in reference to a normative database and 2-different classes of variables could be assigned to the data obtained.

The raw signal is generally divided up according to the following divisions and nomenclature: Delta (0-4 waves per second or Hertz), Theta (4-8 Hertz), Alpha (8-13 Hertz) and Beta (frequencies greater than 13 Hertz). The nature of a signal at a particular location involves nomenclature such as relative power, peak amplitude, peak frequency, and magnitude.

The second class of variables addresses the similarity of the waveforms (via algorithms) between two locations and employs such terms as coherence and phase. Connection activity (coherence and phase values) reflect the activity of the long, myelinated neurons (in the white matter of the brain) connecting the different locations.

The reliability of the qEEG has historically been questioned, but consistent empirical support has indicated highly respectable reliability values, often in the 90 range (Gasser et al., 1985; Salinsky et al., 1991; Corsi-Cabrera et al., 1997); Arruda et al., 1996; Lund et al., 1995; Burgess & Gruzelier, 1993; Harmony et al., 1993; Pollock et al., 1991; Hamilton-Bruce et al., 1991).

During the past 20 years, the field has been rapidly developing its empirical support and clinical knowledge base.

Relationship to Cognition

Much of the original work on the relationship between the qEEG signal and cognition examined the eyes closed data and then correlated those values with well known cognitive measures, such as the IQ test. Giannitrapani (1985) was able to relate performance on the WISC Arithmetic and Comprehension scores to the qEEG beta values. Thatcher & Walker (1985) related increased WISC IQ scores to short inter-hemispheric connections, especially in the posterior regions.

Relationship to Clinical conditions
Learning Problems

Eyes Closed comparisons
The research studies conducted have characterized clinical groups (ADD, AD/HD, learning disabled) by specific qEEG patterns and have generally found elevated levels of theta and delta activity and decreased beta activity levels. The additional problems noted in the literature are problems in connection activity (coherence and phase). As the task changes across the research studies, distinct patterns of deficits manifest themselves, thus providing a clue on the brain’s differential response pattern.

Fein et al., (1986) (N=113) were able to consistently show decreased beta power (19-24 Hz) in dyslexics (eyes closed condition) but no differences in the other bands. Harmony et al. (1990) (N=81) were able to identify children with poor educational evaluations on the basis of high levels of absolute power of delta (left frontal and temporal areas), increased theta values (with subjects matched for SES), and increased alpha at occipital areas (for the subjects with good evaluations). Byring et al. (1991) (N=44) were able to differentiate between spelling disabled children and normals on the visual basis of excess slow activity (especially in temporal regions) and quantitative basis in terms of low alpha and beta powers, and high complexity (spread of frequencies) in the parietal-occipital regions in the spelling disabled children.

Ackerman et al. (1995) (N=119) were able to show that dysphonetic readers had significantly higher values in the theta and delta bands. Both phonetic and dysphonetic poor readers had lower beta activity than Attention Deficit Disorder subjects with adequate reading skills.

Connection Issues (coherence and phase relationships)

Several authors have indicated a connection problem in these children. Marosi et al. (1992) (N=152) were able to demonstrate different patterns of coherence maturation in normal and learning disabled children under the eyes closed condition. Chabot et al. (1996) (N=407) were able to distinguish between normals and ADD/ADHD and Learning Disabilities on the basis of QEEG variables under the eyes closed condition (coherence, relative power, asymmetry and location issues). Evans & Park (1996) were able to identify significant deviations from a normative database in a group of 8 dyslexic children and 2 adults. These were evident in the left posterior region (in particular the P3 position) in terms of reduced coherences and usually involved the theta bandwidth.

Under Activation Conditions
The activation approach has generally yielded a richer set of information on the brain’s response pattern. The research cited focus the deficits in the left temporal and posterior regions.

Mann et al. (1992) (N=25) were able to demonstrate increased amplitude theta activity (4-7.5 Hz) and decreased beta1 (12.75-21 Hz) (compared to normals) in ADD subjects when subjects were reading or drawing. The increased theta activity was more prominent in frontal regions, while decreased beta was significantly lower in temporal regions. For children with dyseidetic disorders (difficulty with visual spatial processing for whole word recognition) there was increased left temporal theta in the t3-p3 region. Lubar et al. (1985) had obtained similar results.

Sklar et al. (1973) (N=25) demonstrated that dyslexic children had more theta activity (3-7 Hz) in the parietal region (rest condition) as well as more activity in the 16-32 Hz range than normals, who had more 9-14 Hz activity (alpha). Under the reading task condition, however, the normals increased activity in the 16-32 Hz range, while the dyslexics decreased activity in this Hertz range. Within the same hemisphere, the coherences were higher for the dyslexics but lower between homologous connections (similar positions) across the hemispheres than normals (reflecting possible problems in the corpus callosum).

Duffy et al. (1980) (N=18) found differences between normals and dyslexics in terms of the bifrontal areas as well as the expected left temporal and left posterior quadrant. The activation tests produced more prominent group differences.

Dyslexics were noted to have increased alpha during activation conditions (relative to controls). Dykman et al. (1982) (N=10) employed a complex visual search task and recorded over the central and parietal sites. They were able to differentiate between the groups (hyperactive, learning disabled, mixed and normal children) on the basis of two frequencies – 16-20 Hz and 7-10 Hz.

Duffy et al. (1988) were able to demonstrate increased alpha in dyslexics under rest and activation conditions as compared to normals in the left posterior, left anterolateral frontal, left midtemporal and bilateral medial frontal areas. Gutierrez & Corsi-Cabrera (1988) (N=8) monitered EEG activity during spatial, verbal and one demanding mixed task to determine possible hemisphere and performance effects. They found no significant differences between performance levels but increased beta power in the left posterior region across all tasks, as well as decreased alpha relative power and increased theta relative power.

Galin et al. (1992) (N=113) concluded in an activation procedure for dyslexic and normal readers that the theta activity in the temporal lobes was the main discriminating variable between the two groups. Valentino et al. (1993) conducted an auditory continuous performance task (N=27) and compared performance levels with EEG variables. They found an increase in beta power (especially in fronto-temporal and left temporal sites), decreases in alpha and posterior theta, and increased anterior theta and delta. The lower performing group had decreased left temporal beta power, while the good performers had more anterior beta and less posterior alpha and theta.

Klimesch et al. (1993) were able to show that during memory retrieval, the alpha frequency (frequency range not stated) of good performers was 1.25 Hz higher than bad performers. Also, during retrieval, alpha desynchronization was more pronounced for bad performers than good performers. Special cognitive tasks such as reading, classification and recognition, as well as attentional demands tended to reduce the power within the alpha band. They also noted that mental tasks and task difficulty, in particular, lead to an increase in alpha frequency only for the difficult but not for the easy tasks. In addition, alpha frequency increase selectively in the hemisphere, which is dominant for a particular task.

In conclusion, the pattern of qEEG abnormalities in the learning disabled child is one of 1-elevated levels of delta, theta and sometimes alpha as well as decreased levels of beta and 2-decreased connection (phase and coherence) patterns with a focus in the left temporal and left posterior regions.

Traumatic Brain Injury

Previous qEEG research – Eyes Closed Discrimination

Original research on the TBI patient addressed the issue of discriminating between a normal and a subject with a TBI (Thatcher et al. 1991). The Thatcher et al. (1989) studies very effectively addressed this problem and obtained discriminate values in the .90 ranges across three independent samples. The predominant finding was decreased posterior alpha and increased posterior beta activity, frontal connection issues and some long cortico-cortico connection patterns. Extending the research into neuropsychological and MRI variables, Thatcher et al. (1998) were able to demonstrate a relationship between increased theta amplitudes and increased white matter T2 MRI relaxation times (indicator of dysfunction) in a sample of mild traumatic brain injury (MTBI) subjects. Decreased alpha and beta amplitudes were associated with lengthened gray matter T2 MRI relaxation times.

These measures were correlated with neuropsychological measures, which indicated decreased cognitive function. The subjects were 10 days to 11 years post injury. This study integrated MRI, qEEG (eyes closed) and neuropsychological measures in a sample of MTBI subjects.

Randolph & Miller (1988) found variability of the EEG a critical component in discriminating head injured patients from normals. Prichep & John (1992) were also able to discriminate head injured from normals on the basis of the qEEG signal. Further studies by Hooshmand et al. (1989) found mostly mild, nonspecific generalized slowing in the TBI patient. Tabano et al. (1988) found an increase in the mean power of the lower alpha range (8-10 Hz) and reduction in fast alpha (10.5-13.5 Hz) with an accompanying shift of the mean alpha frequency to lower values. They also reported a reduction in fast beta (20.5- 36 Hz) activity.
All the studies have focused on frequency ranges below the 32-Hertz range and have not examined the issues of location or task. Collectively, the studies have indicated elevated posterior beta reactions (and decreased alpha), frontal connection issues, decreased alpha and beta amplitudes, increased variability, and nonspecific generalized slowing. Some of the studies appear to have conflicting results (i.e. increased posterior beta, reduction in fast beta) possibly due to different frequency ranges studied.

Activation Conditions
In two studies (Thornton, 1999a, 1999b), addressed eyes closed and three activation conditions (auditory and visual attention, listening to paragraphs). The research employed frequencies higher (32–64 Hz) than normally employed in discriminant analysis and demonstrated encouraging results in distinguishing between normals and MTBI subjects.

The eyes closed comparison resulted in an 82% correct classification (N=81 of which 29 were Traumatic Brain Injury (TBI)) and the activation analysis yielded hit rates of 95% (auditory attention task – N=67), 91% (visual attention task – N=69) and 88% (listening to paragraphs – N=86). The listening to paragraphs task required the least number of variables to discriminate. The variables, which were most often involved in successful discrimination, were high frequency (32-64 Hertz) variables emanating from the frontal lobes. The amount of time elapsed since the accident had no effect on improving the variables affected by the head injury (thus time does not heal). Normals and TBI patients were combined into one group and a correlational analysis was conducted to determine what correlates with successful recall in this combined group. The results indicated a clear pattern of lowered frontal lobe coherence and phase beta2 involvement in the TBI group (Figure #1), which negatively correlated with performance in this combined group.

In a normal adult group, auditory memory performance is positively correlated with coherence alpha projections from predominantly left hemisphere locations (T3, F7 and others) (Thornton, 2000c). Within the head injured group, the positive correlates of successful recall are the right temporal location (T4 – Phase Beta1) and left frontal (F7-Phase Beta2) (Thornton, 2002b).

Traumatic Brain Injury Interventions
Scope of the Problem

Every 21 seconds, one person in the US sustains a Traumatic Brain Injury (TBI). An estimated 5.3 million Americans - a little more than 2% of the US population - currently live with disabilities resulting from traumatic brain injury. Every year 1.5 million Americans sustain a traumatic brain injury.

Every year more than 80,000 people experience the beginning of long-term disability as a result of a TBI. More than 50,000 people die every year as a result of a traumatic brain injury. The causes of TBI include airway obstruction, near-drowning, throat swelling, choking, strangulation, crush injuries to the chest, electrical shock or lightening strike, trauma to the head and/or neck, traumatic brain injury with or without skull fracture, blood loss from open wounds, artery impingement from forceful impact, shock, vascular disruption, heart attack, stroke, arteriovenous malformation (AVM), aneurysm, intracranial surgery, infectious disease, intracranial tumors, metabolic disorders, meningitis, certain venereal diseases, AIDS, insect-carried diseases, brain tumors, hypo/hyperglycemia, hepatic encephalopathy, uremic encephalopathy, seizure disorders and toxic exposure- poisonous chemicals and gases, such as carbon monoxide poisoning.

The groups, which experience TBIs most frequently, are the elderly, adolescents and young adults. The cost of a TBI in the U.S. is 48.3 billion dollars a year. The hospitalization costs are 31.7 billion a year and the fatality cost is 16.6 billion a year. Car accidents (44%) and falls (26%) are the leading cause of TBI. (Brain Injury Association of America)

Previous Research on Interventions with TBI
Cognitive Rehabilitation

The literature on memory improvement in brain-injured patients has reported different intervention approaches (repetitive practice, strategies, visualization, PQRST, etc.) and has found generally minimal to mixed results. The broad issues in the area concern approaches, effectiveness, generalization (to other tasks and everyday life), long term maintenance of improvement and the subject’s self report of changes.

One of the initial reviews in this area (Benedict, 1989) concluded, “findings regarding the effectiveness of memory remediation interventions have been inconsistent”, adding that methodological inadequacies have hindered the identification of specific treatment effects. The literature is clear that simple repetitive practice is of minimal or no aid in improving memory for recall exercises (Glisky & Schacter, 1986). Specific techniques such as visualization, method of loci and cognitive strategies have shown different degrees of effectiveness. However, researchers in the field generally agree that these approaches face the problem that the subject does not continue the use of the strategy (Freeman et al., 1992). Significant improvements from repetitive recall drills have not been found (McKinlay, 1992). Internal memory aids such as imagery instructions are employed less than external memory aids, but patients generally employ neither.

Individuals with mild brain injury have shown increases of 22% on short term recall (logical memory) and 86% delayed recall with treatment addressing both psychosocial issues and cognitive approaches (Ryan & Ruff, 1988). The researchers did not find any differences in improvement between the type of treatment provided.

Employing the PQRST (Preview, Question, Read, State, Test) (Benedict & Wechsler, 1992) in a single case design with individuals who were brain injured resulted in only marginal results in one subject, while another study (Grafman, 1984) obtained improvement in 40% of brain injured subjecsts on 50% of the memory tests employed. A memory notebook treatment (Schmitter-Edgecombe et al., 1995) program obtained no improvement in logical memory scores.

Learning Disabilities Interventions
Scope of the Problem

The federal government has spent 350 billion dollars over a twenty-year period on special education programs (U.S. Dept of Education), which were designed to improve the cognitive functioning of the special education child. There were an estimated 6.5 million children who were expected to require special education in 2002 (Office of Special Education).

Approximately 63% of these children have specific learning disabilities or speech/language problems (US government) with no concomitant physical disability. The societal implications of improving the cognitive functioning of the special education student are significant as the percentage of inmates requiring special education in adult correctional facilities is 28-43% (vs. 5% in normal population) (Winter, 1997). It has been reported that 82% of prison inmates in the U.S. are dropouts. In New York City, the average cost of incarcerating a youth is $55,300 a year. (Winter, 1997).

Despite the enormity of the social and educational problem, the interventions currently employed have largely been unsuccessful in obtaining significant and meaningful results. Lyon & Moats (1988) concluded that: “It is difficult, if not impossible, to find any evidence beyond testimonials and anecdotal reports that support the assumptions, treatment methods, and stated outcomes associated with medical and psycho educational models, there is overwhelming empirical and clinical data indicating that medical and psycho educational models, as they are presently conceived and used, are inadequate for determining what and how to teach learning disabled students.” (p.225)

Eleven years later the conclusion remains the same as Birsh (1999) concluded, "despite the widespread inclusion of multisensory techniques in remedial programs for dyslexic students and a strong belief among practitioners using these techniques that they work, there was little empirical evidence to support the techniques' theoretical premises." (Pg 7).

Psychoeducational
The special education and learning disability intervention research arena is comprised predominantly of psychoeducational type interventions that focus on associational networks (Orton-Gillingham, Lindamood Bell), improving phonic pronunciation ability (phonic intervention programs), individual tutoring, Fast Forward computerized interventions and resource room interventions. Almost all of the programs have focused on reading and verbal ability issues as measured by standardized instruments. Although important, it is a curious phenomena that the intervention focus and educational field has failed to address issues of math, memory and other cognitive skills with as much effort and resources, as these cognitive skills are also important to functioning in society.
Orton-Gillingham

Many schools currently employ the Orton-Gillingham Multisensory Methods. This method states that dyslexia is caused by neurophysiological-based disabilities that may be helped by multisensory teaching techniques that provide linkages between the visual, auditory and kinesthetic senses. It “concentrates on fusing smaller units (letters, sounds, and syllables) into more complex wholes (words)”. Oakland and Black (1998) stated that reviews of the treatment literature on dyslexia “reveals a limited number of scientifically sound and clinically relevant reports of significant treatment effects.” (p. 336)

The research they conducted with the Orton-Gillingham method indicated no difference between teacher interventions (350 sessions) or 350 Orton-Gillingham videotape sessions presentation on any measure. The main effect was for age. As you get older you get better, regardless of the treatment method. The research also obtained “modest” improvements on reading comprehension, word recognition, and polysyllabic words when comparing the control to the experimental group. There was no effect on Spelling or monosyllabic decoding skills.

Lindamood-Bell
The intervention model follows the five components of reading – phonemic awareness, phonics, fluency, vocabulary and comprehension specified in the No Child Left Behind Act. The program is conceptualized as a sensory-cognitive approach, which involves imagery as well as other exercises in the interventions.

Results of the Lindamood-Bell treatment approach have been published on the web at http://www.lindamoodbell.com. The treatment reported involves 103 to 148 hours of intervention over a period of 6 weeks. Improvements were reported (2002) for a series of cognitive abilities (Decoding - 7 to 16 percentile rank increases, Comprehension - 11 to 18 percentile rank increases) and by prior diagnosis (dyslexia - 0 to 26 percentile rank increases, ADD - 0 to 24 percentile rank increases, ADHD - 0 to 22 percentile rank increases). The increased scores were on standard measures of verbal abilities.

Tutoring
One adequately designed study to date documents improved (to appropriate grade level) reading as a result of 75 hours of tutoring intervention. (Vellutino et al., 1996) The gains were maintained after the interventions ended.

Phonics
Foorman et al. (1997, 1998) studied children at risk for reading failure and had the children engage in phonological approaches for a full school year, which resulted in improvements in phonological analysis skills. However, additional studies by Foorman et al. showed that the child’s verbal intelligence predicted end of year performance more than the interventions, suggesting the program was not effective. Foorman et al. (1997a) involved students eligible for Title 1 in a multi level approach and were able to demonstrate gains in reading levels. Difficulties in maintenance of gains (Olson et al., 1997) and problems of generalization to reading ability and other skills have been noted by some authors.

Fast-Forward
Fast forward is a computer based reading intervention program consisting of seven adaptive exercises. These exercises are designed to improve auditory and language processing by using nonlinguistic and acoustically modified linguistic speed (rapid frequency transitions in speech are slowed and amplified). Typically the treatment involves 100 sessions occurring over a 4-6 week period and has obtained gains of 1-2 grade level improvement in reading skills.

A national field study of the method (Scientific Learning website) indicated an improvement of 15% in auditory discrimination (N=130) (Goldman Fristoe), over 1 SD on following directions (Token Test) (N=329), 14% improvement in receptive language skills and 10% improvement on expressive language skills on the Clinical Evaluation of Language Fundamentals (CELF-3) (N=77). An average of 10% improvement was achieved in overall language skills (N=77) (Test of Language Development, Primary (TOLD-P:2)


Temple (n=12) found improvements in reading on word identification, word attack, passage comprehension, oral language and rapid naming, but did not show gains on the letter rhyme task. The fMRI changes that were noted were in the left temporo-parietal cortex and inferior frontal gyrus. The improved brain response correlated with only the phonic word training exercise of the Fast Forward program and oral language ability. Interestingly, there was no correlation between increased activity on the fMRI and the improved reading scores. The control group did not demonstrate changes on the fMRI or reading scores. (Temple et al., 2003)

Resource Room

Resource room interventions have been employed by the school systems to address learning problems. One study has addressed the issue of the comparative effectiveness of resource room interventions versus EEG biofeedback (Orlando, Rivera 2003). The author obtained superior results of EEG biofeedback to resource room (control group) across a variety of reading measures (basic, comprehension, composite) and IQ. The resource room control scored significantly lower in basic reading, reading comprehension and reading composite, and lowered full-scale IQ scores. Therefore resource room allocation lowers a child's abilities in reading and general intelligence. The author noted that re-evaluation data often indicates that cognitive functioning frequently decreases in the re-evaluation process, especially at the six-year period from the initial evaluation.

What is Neurotherapy?

Neurotherapy (or EEG biofeedback) is the operant conditioning of the electropysiological signals of the brain, which are recorded at the scalp and produced by a thin layer of cortical gray matter (3 mm) located just below the skull.
Electrodes are placed on the scalp of a subject and the electrical information is sent to a recording unit. The unit employs a software interface to graphically present the status of selected qEEG variables to the subject in a visual modality. Reward and inhibit parameters are software selected. The software then provides the reward (airplane rising, higher tones, or other complex visual imagery) to the brain with the aim of changing the qEEG signal in a desired direction. As the brain is an adaptive organ, it attempts to satisfy the demands made upon it by the software and changes its activity to meet these requests. The exact mechanism is unknown.

Scientific Research Status

Almost all of the original research was directed towards changing intelligence scores (as measured by IQ tests) and attentional abilities (as measured by the TOVA and similar instruments). Four independent researchers have been able to demonstrate significant increases on IQ tests averaging 15 points (one standard deviation) as a result of EEG biofeedback (Linden et al., (1984) (N=18); Tansey, (1991) (N=24); Othmer & Othmer, (1992) (N=15); and Thompson & Thompson (1998) (N=98)). A consistent finding generated across many research studies has been significant improvements in attentional abilities. Othmer & Kaiser (2000) were able to demonstrate the effect in the largest study to date (N=1089). The importance of IQ changes resides in the empirical fact that the best predictor of future success in life is the IQ score (Brody, 1991). The groups usually addressed with these interventions have been the learning disabled and the LD, ADD and AD/HD groups.
All of the interventions employed protocols directed at the C3-C4 sensorimotor strip and attempted to increase beta microvolts (from 13 to 22 Hertz) and decrease theta microvolts. The number of sessions averaged between 20-40 for the IQ studies and about 40 for the attentional studies.

Developmental of activation database with higher frequencies for specific cognitive functions

The author was involved in the development of a qEEG activation database with a sample size of 90 (normal subjects with no history of learning problems, ADD, neurological problems or any other significant medical problems which could affect CNS functioning). Subjects were engaged in 22 generally standard neuropsychological tasks. These tasks included visual and auditory attention, auditory (word lists and paragraphs) and reading memory (input, immediate and delayed recall), spelling, pronunciation of nonsense words, mathematics (multiplication tables, internal spatial addition of double digit numbers), memory for the names of faces, memory for location of objects, a “to-do” list and autobiographical memory. Additional emotional tasks included the experiencing of love, happiness and sadness. The data was analyzed for the relationship between success at a task (recall score, % correct, etc.) and the qEEG variables to determine the relationship between the qEEG variables and effective cognitive functioning.

Value of database for differential diagnosis of specific cognitive problems
The database offers 3 unique advantages over traditional eyes closed databases. First, the frequency range was extended to 64 Hertz (in contrast to the normal 32 Hertz range available in the eyes closed databases) to allow for a fuller understanding of the nature and relative importance of the higher frequency. Second, the analysis done under activation conditions reveals results, which are conceptually closer to the object of concern, cognitive function, than an eyes closed comparison. Third, the advantage of knowing the relationships between the qEEG variables and cognitive performance allows the clinician to see exactly what is deficient in the electrophysiological functioning when compared to the activation database.

Summary and Conclusions

The employment of an activation qEEG allows for very precise intervention protocols to improve cognitive functioning. The problems that becomes evident in this type of evaluation often will indicate more problems than can be addressed with 40 sessions, although the cognitive ability may still improve dramatically. Unanswered questions in this area concern the generalizability of the gains, both in terms of cognition and qEEG variables.

There have been difficulties addressing some of the brain conditions encountered where the approach has not resulted in such dramatic results. These cases, in particular, have involved toxic encephalopathy and chemotherapy to the brain. There are some other cases where the subject’s qEEG values had improved but the cognitive performance had not improved sufficiently to justify further interventions. For example, one subject increased their memory score from 1 to 2 after 40 sessions and the treatment was discontinued, even though the qEEG values had improved. In such cases, it is presumed that there is an underlying subcortical problem that is causing the cognitive problems. It is difficult, at this point in time, to predict the conditions under which the brain will not respond.

Overall, the data presented above reflect improvement in about 80-85% of the subjects addressed who completed the full 40-session program. Of particular interest to note is the responsiveness of the learning disabled subjects, who surpassed all other groups in terms of the magnitudes of the cognitive performance gains. This data provides an encouraging alternative intervention to a rather large population of students in our school systems (5-15%).

The ability of this activation database Neurotherapy approach to improve a cognitive ability (in this case memory) in some 40 sessions or 20 hours by such a large degree while other programs are requiring anywhere from 70 to 350 hours of interventions to obtain relatively small percentage improvements on the tested skill area strongly argues for a paradigm shift in how we conduct special education and cognitive rehabilitation in the United States.