1996-1997

PROGRAM EVALUATION
EEG NEUROFEEDBACK
NEW VISIONS SCHOOL – 1996-97

The EEG neurofeedback department of A Chance to Grow/New Visions School was established in 1991 to provide neurofeedback training for students in schools where the agency was located, for persons referred by professionals in the community, and subsequently, for students of New Visions School when it opened in 1993. New Visions School was the first public school in the U.S. to utilize neurofeedback training for its students. This report details the results of neurofeedback training for New Visions School students during the 1996-97 school year. Results are determined using the Test Of Variables of Attention (TOVA), The Conners’ Continuous Performance Test (CPT), client self-reports, teacher evaluations, and parent communications

NEUROFEEDBACK PROGRAM
The EEG neurofeedback department of A Chance to Grow, Inc. (ACTG) was initially established in 1991 to provide neurofeedback training for students in the Minneapolis Public Schools where the agency was located. The program also served agency staff, their families and a few clients from the community.

The EEG neurofeedback program is designed to help students correct problems in brain function. This is done through a ‘feedback’ process which shows moment by moment alterations in brain activity. The student observes these alterations as changes on a computer screen and hears them as changes in audio tones as well. S/he then learns, through practice, to move brain activity in the desired direction. Once accomplished, this learning appears to be maintained by the student and the reported effects seem to permeate many areas of neurological, psychological, and physiological functioning.

Training Program
Students generally receive two 1/2 hour sessions of neurofeedback training per week. Occasionally there are compelling reasons to increase the frequency of sessions, so some students with severe impairments may come to the clinic daily or twice daily for a period of time. Training is accomplished using protocols established by the foremost clinicians and researchers in the field of EEG neurofeedback, modified to reflect our experience working within this unique environment. Individual training procedures are developed for each student to address presenting concerns. Training variables included sensor placement, frequency selection, type of visual and auditory feedback, and of course, frequency of sessions. These variables are monitored and adjusted as training proceeds.

Generally, students are taught to enhance or increase higher frequency (12 to 20 hz) brainwave activity associated with increased attention and focus. They are concurrently taught to inhibit or reduce lower frequency (4 to 7 hz) brainwave activity associated with decreased attention and focus. Students are also taught to inhibit EMG (muscle activity) to encourage relaxation. The specific range of frequencies encouraged is determined by a variety of screening techniques. Alterations in training frequencies are made based upon the ongoing evaluation processes.

Personnel
Staff members have backgrounds in education, psychology, and EEG technology. Staff are certified in EEG neurofeedback or are pursuing such certification. Staff members attend regular continuing education programs devoted to EEG neurofeedback, neurophysiology, neuropsychology, assessment, general and educational psychology, and education.

PROGRAM EVALUATION

Test Of Variables of Attention (TOVA) and Conners’ Continuous Performance Test (CPT)
The TOVA and the Conners’ CPT are continuous performance tests which evaluate four basic measures of arousal and alertness. These functions are attention, impulsivity, response time and variability. Attention refers to the ability to remain on task during the entire test. Impulsivity represents the tendency to respond to the incorrect stimulus (the non-target symbol), and/or the inability to inhibit such a response to the incorrect stimulus. Response time measures how long the student takes to process and respond to the signal. And finally, variability measures the student’s ability to maintain a reasonably steady focus (as determined by fluctuations in response time) during the entire test. The two tests measure and report this information in different ways, but the concepts are similar. The tests have been compared to other methods of screening and evaluating children and adults for attentional problems and have been found to be essentially consistent and valid. Normative data was gathered from ‘typical’ individuals matched for age and gender.

Testing Results
Sixty-five NVS students received neurofeedback services during the 1996-97 school year. The TOVA or Conners’ CPT was administered to nearly all students prior to training. Four students did not possess the minimum skill level needed to take either test and therefore were not tested.

Thirty-three students (51%) were given repeat tests after a minimum of 20 neurofeedback sessions. Sixteen students began neurofeedback training late in the school year and did not receive the minimum 20 sessions required for re-testing. Five students either left the school or left the program for various reasons. Seven students were not re-tested at the end of the year due to absence, schedule conflicts, or other difficulties. Of the 33 students who received pre and post- testing, 18 received the TOVA and 15 received the Conners’ CPT.

TOVA Results
Initial TOVA results indicated that all 18 students receiving the TOVA were below the typical range in at least one scale on the pre-test. Three were below typical in only 1 scale, five were below typical range in only two scales, eight were below in three scales, and two were below the typical range in all four scales. Eighty-three percent of these students improved at least 0.5 standard deviations (SD) in one or more scales upon re-test. Generally 0.5 standard deviations from the mean is considered a significant change. The TOVA Interpretation Manual (‘1991 by Lawrence M. Greenberg) states under:

Significance
‘In general a change of more than 0.5 standard deviations (in either direction) is significant.’

Forty-four percent of the students improved in attention, thirty-nine percent showed decreased impulsivity, thirty-nine percent had improved response times, and sixty-one percent had better variability scores. These improvements indicate that most of those students who were tested experienced significantly less problems in factors which may interfere with learning. These factors include such things as attention problems, impulsive behavior, slow information processing, and lack of consistency in skills, abilities, and achievement.

Attention
It is clear that the ability to sit still and pay attention for age appropriate periods of time is an essential skill for any student. Simply being able to attend to information being presented, in whatever form, is required before the student can make use of that information to pass a test or write a report or respond in class. Being able to spend time studying also depends to a large degree on the student’s ability to keep his or her attention on the task long enough to learn the information or complete the assignment. This is why measuring attention can provide useful information for predicting a student’s academic performance.

Impulsivity
The ability to resist impulsive actions may be even more important than attention as a general life skill. It can translate into the ability to resist the temptation to respond to teasing or provocation from another student, to resist peer pressure to use drugs or engage in dangerous or self-defeating behaviors. It can represent the ability to think through an action rather than responding with the first impulsive response. Impulsive students often respond with words and actions which are inappropriate to the task or situation because they lack the ability to ‘stop and think’. This is generally not a choice for most impulsive students. They require specific brainwave training to correct this tendency, not punishment or lectures.

Response Time
Response time improvements are actually thought to be improvements in the processing ability of the brain. Generally a faster response time, in the presence of stable attention and reduced impulsivity, indicates improvement in the student’s ability to receive, process, and respond to information. If a student is a ‘slow processor’, s/he is either not processing the data as fast as his or her peers or is responding cautiously because of a desire to make fewer mistakes.

Variability
Finally, the ability to maintain steady attention for longer and longer periods of time is important for progressing through the ever increasing demands of school, and eventually work. According to the TOVA Interpretation Manual:

‘Variability is the single most important variable in correlating TOVA response and attention deficit disorders. It is clearly the most powerful variable and, in older subjects, may be the only variable that is deviant from the norm. We have begun to view variability as the hallmark of attention deficit disorders. It’s as though the average response time and error rates are not as meaningful as the extreme swings in performance. We know clinically that an individual with attention deficit disorder can, at times, attend to tasks and respond appropriately. However, this level of performance cannot be sustained, and their performance often deteriorates within any given period of time.’

(emphasis added.)

Eighty-three percent of students tested improved 0.5 standard deviations from their original test scores on at least one scale, seventy-two percent improved 1 standard deviation, fifty percent improved 1.5 standard deviations, and thirty-three percent improved 2 standard deviations from their baseline scores prior to neurofeedback training. A gain of one standard deviation is highly significant and indicates a change in that student which is unlikely to be attributed to random factors.

This chart shows the percentages of students improving by 0.5, 1, 1.5, and 2 standard deviations (SD) on at least one TOVA scale.


This chart shows the percentage of students improving 0.5 standard deviations in each scale


These charts show pre and post-test TOVA for 18 students
(higher scores are better)

More than half of the students tested showed improvements in three TOVA scales, indicating a global change in functioning in a majority of these students. Most students improved in the variability scale which can be a crucial indicator of the ability to maintain attention in the classroom. . Improvements in attention were also quite common, suggesting that these students experienced not only a more steady attention (variability), their overall ability to maintain attention also improved. Many of those same students also showed decreased levels of impulsivity, suggesting that they will be able to resist impulses which often result in behavior and disciplinary problems. Students with initial scores in the ‘normal’ range for a particular scale showed less profound changes in post test results. For example, students who were not impulsive to begin with did not show much change in the impulsivity scale. Students who were extremely impulsive on the pre-test, generally improved markedly on this scale on the post test.

Case Histories
The following TOVA test scores and brief histories of individual students illustrate how the training translates into specific behavior changes and improvement in skills. These examples are typical of the progress experienced by many of our students as they participated in EEG neurofeedback training.


TOVA Results Based On Standard Scores (higher scores are better)

This 12 year old girl was initially seen for severe problems with attention and also some impulsivity. She was resistant, appeared angry, had problems with low self esteem, and did not want to do neurofeedback training. During her initial TOVA, she needed to be reminded 4 times to respond only to the target image (evidence of impulsivity). Following 54 sessions of neurofeedback training, she had improved significantly. Her attention scale improved by 72 points, impulsivity scale improved by 34 points and her variability scale improved by 18 points. The response time scale showed a 9 point decrease and probably reflected a more careful approach to the test. Her reading scores improved and teachers and other staff members remarked on her improved attitude and self-image.

TOVA Results Based On Standard Scores (higher scores are better)

This 12 year old boy was referred for hyperactivity, impulsivity, behavior concerns, and learning delays. Parents indicated that he was anxious and depressed, always on the go, frustrated, and gave up easily. The classroom teacher reported verbal and physical aggression, poor peer relationships, and off task behavior in the classroom. Following 60 sessions of neurofeedback, TOVA results showed marked improvement. He was less impulsive and had better attention. Behavior and attitude improved, though he remained a difficult student. His impulsivity scores improved, but remained in the problematic range and he may need more sessions, and/or other interventions to completely resolve this issue.

TOVA Results Based on Standard Scores (higher scores are better)

This 11 year old girl, diagnosed as ADD and on Ritalin, was referred to neurofeedback for attention and hyperactivity problems. She was highly variable in her responses to the initial TOVA and was moving almost constantly. Parents reported many symptoms of hyperactivity. She received neurofeedback sessions daily for some time initially and has shown significant improvement in her attention and variability scores though impulsivity remains a problem. It should be noted that the initial TOVA was administered following this student’s regular dosage of Ritalin, while the final TOVA was administered after the Ritalin had been discontinued.

Conners’ Continuous Performance Test (CPT) Results
Fifteen students received the Conners’ CPT before and after training. Their results indicate improvement in many of the most important scales of this test. The following graph shows the percentage of students improving 0.5 standard deviations in each of the 10 scales for which T-scores are available. Nearly half (47%) of the students improved on 5 of the 10 scales, indicating global improvement. Seventy- three percent of the students improved 0.5 SD on scale 3, which is one measure of variability.

Percentage of Students Improving 0.5 SD by Scale

The Conners’ Continuous Performance Test User’s Manual (Conners, 1995) states:

‘The reaction time variability is often a better measure than speed of how consistently the respondent can focus his or her attention, and is much less affected by response sets or expectations than speed. Variability, and not just the average level of speed, is a useful measure of attention.’

Other test scales which measure variability indicated that, of this group, 53% improved on scale 4, 60% improved on scale six, 47% improved on scale eight, 53% improved on both scales nine and ten.

The following chart shows the percentage of students improving on each of the variability scales and a composite percentage for all five variability scales.

The following graphs and individual summaries show changes based on data from the Conners’ CPT. The data is in the form of changes in T-scores from pre-test to post-test. Typically, improvement in T-scores is shown by decreases in the score, represented as a negative number.

These T-scores have been converted to positive numbers to make the graphs easier to understand and so improvement is shown as change in a positive direction. Changes of 5 points in either direction is equivalent to 0.5 standard deviations. Changes of 10 points equals 1 standard deviation of change. Improvement of 5 points is significant and improvement of 10 points or more is considered highly significant.

Conners’ CPT T-Scores (converted) for Student # 26

This 13 year old boy was referred for attention problems, sleep disorder, low self-esteem, and temper tantrums. Parents endorsed 15 of 19 symptoms of ADD on a brief rating scale prior to training. Following neurofeedback training, he scored 1 or more standard deviations of improvement in 9 of 10 scales on the CPT. His teachers reported that he was much more focused and attentive in class, sleep was no longer a problem and self-esteem had improved.

Conners’ CPT T-Scores (converted) for Student # 19
Chart shows the amount of change in T-scores from pre-test to post-test
(positive changes show improvement)

This 12 year old boy was referred for attention problems and hyperactivity. He had only 18 sessions of neurofeedback training but showed substantial improvement in his CPT scores. He was new to the school this year and was seen by his teachers as initially quite anxious. He appeared to have problems with self-esteem and also was affected by a long-standing family illness. He was able to make excellent gains in reading (1.8 years gain this year) and appeared to get his anxiety and hyperactivity under greater control.

Conners’ CPT T-Scores (converted) for Student # 30
Chart shows the amount of change in T-scores from pre-test to post-test
(positive changes show improvement)


This 11 year old boy was referred for attention problems. He was exceptionally tense and anxious and appeared quite socially inhibited. He was also significantly below age level academically. He received 26 sessions of neurofeedback training with excellent results. His initial CPT results indicated markedly atypical scores both in attention and in variability. Post-testing results showed statistically significant improvement in 8 of 10 scales. He became much more relaxed and confident and made a 1.5 year gain in his reading scores. Teachers reported he appeared more confident and had developed several friendships in the classroom.

Summary
These results represent some of the gains made by students served during the 1996-97 school year at New Visions School. We noticed a significant difference in scores with increased numbers of sessions. Students receiving only 20 to 30 sessions often did not show as much improvement as those students receiving 40 sessions or more. Many researchers recommend 40 to 60 sessions of neurofeedback training to assure that training goals are met and that improvements are maintained following the cessation of training. The population served by this program often experiences more severe impairments than the typical ADD/ADHD participants included in most research studies. Some of our students have received more than 200 sessions over the course of several years as part of a comprehensive program developed by A Chance To Grow to help them maintain a higher level of functioning than would normally be expected of them.

Forty-two students were served during the 1994-95 school year. Fifty-three students were served during the 1995-96 school year and sixty-five students received EEG neurofeedback services during the 1996-97 school year. Projections for future years include a further increase in number of students served and an increase in total sessions for those students.

Promising gains by students and modest increases in the numbers of students receiving neurofeedback services only serve to highlight the need for greater capacity and newer technology to enable the neurofeedback program to provide this service to more students with greater effectiveness. Only about a third of New Visions School students were able to make use of neurofeedback training this past year, even though staff had identified many more who could have benefited from this intervention. Increased efficiency will enable the program to see a few more students during the next school year, however more equipment, a larger facility, and more staff will be required to truly serve the needs of all of our students.

Future Programs/Research Initiatives
A Chance To Grow and the EEG neurofeedback department will be conducting a State of Minnesota funded demonstration project during the 1997-98 school year. We will be bringing EEG neurofeedback training into regular public elementary schools to offer this service to approximately 40 students. We will conduct extensive pre and post-testing and will disseminate the information we gather to school districts and organizations throughout the state as well as attempting to publish the results in professional peer reviewed journals.

This initiative will hopefully be followed by the integration of this technology into many more schools in Minnesota and throughout the country.

APPENDIX

Equipment
Neurofeedback training was conducted using two Neurocybernetics EEG biofeedback systems (16340 Roscoe Blvd, Suite 120 * Van Nuys, California 91406-1201 818-891-6789 ) and two PHL EEG biofeedback systems (Biocom Technologies, P.O. Box 2729, Poulsbo, WA 98370, [206]-853-5090 ). The Neurocybernetics systems use a two computer configuration with an isolated EEG signal amplifier. The feedback display is in the form of a ‘Mazes’ game similar to pacman and ‘Islands’ or ‘Highways’ games which represent the signals as three boxes which change size as EEG (brainwave activity) and EMG (muscle tension) signals change in amplitude. This information is displayed against an interesting and changing visual background. The PHL systems use a single computer configuration, also with an isolated EEG signal amplifier manufactured by J&J Engineering. Physiological information is presented in a wide variety of graph, picture, and game displays. In addition to EEG, the Neurodata systems display other physiological data such as EMG, EDR (sweat gland response) and thermal (peripheral vasodilatation). This information is quite useful in evaluating students who are anxious. These modalities also help such students learn anxiety response control.

TOVA
The TOVA is a continuous performance test which evaluates four basic measures of arousal and alertness. These functions are attention, impulsivity, response time and variability. Attention refers to the ability to remain on task during the entire twenty-two minute test. The impulsivity scale represents the tendency to respond to the incorrect signal, and/or the inability to inhibit such a response to the incorrect signal. Response time measures how long the student takes to process and respond to the signal. And finally, variability measures the student’s ability to maintain a reasonably steady focus (as determined by fluctuations in response time) during the entire test. The test has been compared to other methods of screening and evaluating children and adults for attentional problems and has been found to be essentially consistent and valid. Normative data was gathered from ‘typical’ individuals matched for age and gender.

Conners’ Continuous Performance Test (CPT)
The CPT is similar to the TOVA in design and purpose. It does however introduce several other variables which have proven interesting when evaluating students in the EEG neurofeedback program. The TOVA presents the images at a fixed interval while the CPT presents the images at intervals from 1 to 4 seconds. This change in inter-stimulus interval (isi) elicits differing responses in students depending upon the specific difficulties they experience in the learning environment.

Students with classic attention problems may be able to respond correctly to stimuli presented moderately fast (2 second intervals) but may have problems processing the stimuli accurately when they are presented quickly (1 second intervals). Some clients may also experience some ‘drift’ or fall off in their ability to stay with the task when the stimuli occurs less frequently (4 second intervals). The results of this variability in the frequency of stimulation are reported as changes in response speed and consistency when the interval changes. The CPT also uses letters as its stimuli and this introduces the factor of discriminating between stimuli which appear similar (the x and y for example). This variable is reported as perceptual sensitivity.

Like the TOVA, the CPT also reports specific response time (RT) data and then evaluates variations in this data over the course of the test to determine several different measures of variability. One variability measure is the hit response time standard error (SE) taken from all the individual responses. Another is the variability of standard errors which is the standard deviation of the 18 standard error values calculated from each sub-block of responses within the test. Hit standard block change measures indicate whether the individual’s reaction times became more or less consistent as the test progressed. Hit response time inter- stimulus interval change registers a slowing or speeding of reaction time as the time between targets increased. Finally, the hit standard error inter-stimulus interval change records the level of increased or decreased consistency in response time as the time between targets increased.

Another measure of alertness in the CPT is that of attentiveness, which is measured by how well the individual discriminated between targets and non-targets (stimuli). Further measures of variability, attention, and impulsivity involve analysis of the students response to the test in general. Was the student cautious and responding slowly to avoid mistakes, or did the student favor quicker response times at the expense of accuracy? Also did the student’s responses slow down, speed up, or remain the same from the beginning of the test to the end? Did the student slow down his or her responses as the interval between letters increased or decreased and did he or she have more errors of omission or commission compared to the response time and variability scales.

The complexity and level of detail in the Conners’ CPT appear to make it an effective tool for evaluating students both prior to and following EEG neurofeedback training. Our experience with the Conners’ CPT indicates that it may generate more of a negative practice effect than the TOVA. This means that student scores may decline during subsequent test administrations due to the inherently boring nature of the test. This is a common factor for all continuous performance tests. Some of them, however seem to be more sensitive to this effect than others. The results of the CPT seemed to reflect the actual improvement in student’s performance on other measures, so we are hopeful that it will continue to provide us with useful data. This was our first year using the Conners’ CPT and we will continue to assess its effectiveness as an evaluation tool.


New Visions School
Michael Joyce, MA NLP

3820 Emerson Ave.
Minneapolis, MN 55412
(612) 521-2266 F-9647


Email: mjoyce@mail.actg.org

More information about New Visions School

Comments are closed.