Applied Psychophysiology and Biofeedback, Vol. 27, No. 4, December 2002 ( C° 2002)

The Effects of Stimulant Therapy, EEG Biofeedback,

and Parenting Style on the Primary Symptoms

of Attention-Deficit/Hyperactivity Disorder1

 

Vincent J. Monastra,2;3 Donna M. Monastra,2 and Susan George2

One hundred children, ages 6–19, who were diagnosed with attention-deficit/hyperactivity

disorder (ADHD), either inattentive or combined types, participated in a study examining

the effects of Ritalin, EEG biofeedback, and parenting style on the primary symptoms of

ADHD. All of the patients participated in a 1-year, multimodal, outpatient program that

included Ritalin, parent counseling, and academic support at school (either a 504 Plan or

an IEP). Fifty-one of the participants also received EEG biofeedback therapy. Posttreatment

assessments were conducted both with and without stimulant therapy. Significant improvement

was noted on the Test of Variables of Attention (TOVA; L. M. Greenberg, 1996) and

the Attention Deficit Disorders Evaluation Scale (ADDES; S. B. McCarney, 1995) when

participants were tested while using Ritalin. However, only those who had received EEG

biofeedback sustained these gains when tested without Ritalin. The results of a Quantitative

Electroencephalographic Scanning Process (QEEG-Scan; V. J. Monastra et al., 1999)

revealed significant reduction in cortical slowing only in patients who had received EEG

biofeedback. Behavioral measures indicated that parenting style exerted a significant moderating

effect on the expression of behavioral symptoms at home but not at school.

KEY WORDS: EEG biofeedback; ADHD; Ritalin; parenting style; outcome studies.

Attention Deficit-Hyperactivity Disorder (ADHD) is a psychiatric disorder, characterized

by the primary symptoms of inattention and/or impulsivity and hyperactivity, that

is evident in approximately 3–5% of school-aged children (American Psychiatric Association,

1994). Although currently defined in terms of behavioral symptoms, there is evidence

that the core symptoms of ADHD can be associated with metabolic (Zametkin et al., 1990;

Zametkin & Rapoport, 1987), circulatory (Amen, Paldi, & Thisted, 1993), neuroanatomical

(Casey et al., 1997; Hynd et al., 1993), and electrophysiological abnormalities (Chabot,

Merkin, Wood, Davenport , & Serfontein, 1996; Chabot & Serfontein, 1996; Mann, Lubar,

1Preliminary reports of this research were presented at the Eleventh Annual Conference of CH.A.D.D. held in

Washington , DC , in October, 1999, and at the Annual Convention of the American Psychological Association,

which convened in Washington , DC , in August, 2000.

2FPI Attention Disorders Clinic, Endicott , New York .

3Address all correspondence to Vincent J. Monastra, PhD, FPI Attention Disorders Clinic, 2102 E. Main Street,

Endicott , New York 13760 ; e-mail: poppidoc@aol.com.

231

1090-0586/02/1200-0231/0 C° 2002 Plenum Publishing Corporation

232 Monastra, Monastra, and George

Zimmerman, Milller,&Muenchen, 1992; Monastra et al., 1999; Monastra, Lubar,& Linden ,

2001). In addition to the primary characteristics of ADHD, there are multiple secondary

symptoms that are frequently noted, including learning disorders, anxiety, depression and

other mood disorders, tic disorders, and conduct disorders (Spencer, Biederman, &Wilens,

1999). Estimates of the incidence of these secondary or comorbid symptoms range from 50

to 90% (Barkley, 1998; Spencer et al., 1999).

Because of the significant impairment of academic, social, family, and vocational functioning

that is caused by ADHD and these comorbid conditions (summarized by Barkley,

1996; Hinshaw, 1992), considerable scientific effort has been directed at developing effective

pharmacological and psychological treatments. As reviewed by Spencer et al. (1996)

and Swanson et al. (1993), the vast majority of these studies have indicated that pharmacological

treatments can exert a positive effect on the core symptoms of inattention,

impulsivity, and hyperactivity. However, as noted by Barkley (1998), approximately 35–

45% of patients diagnosed with an “Inattentive” Type of ADHD and 10–30% of those with

a “Combined” Type of ADHD fail to respond to medications. In addition, systematic review

of the effects of pharmacological treatment on cognition, academic achievement, and

social skills (Bennett, Brown, Craver, & Anderson, 1999; Brown & Sawyer, 1988; National

Institute of Health , 1998) fails to support a conceptualization that the wide range of clinical

problems presented by ADHD patients can be effectively treated by medication alone.

Based on a clinical perspective that many of the functional impairments associated with

ADHD are not responsive to medication treatments, researchers have examined the role of

various behavioral therapies in the development of academic, social, and attentional abilities.

Such studies have examined the effects of “reinforced” instruction in psychosocial skills

in school and “camp” settings (e.g., MTA Cooperative Group, 1999; Pelham, Wheeler, &

Chronis, 1998), at home via parent training (Anastopoulos, Shelton ,DuPaul,&Guevremont,

1993), or through a combination of these approaches (MTA Cooperative Group, 1999). The

outcome of these studies suggests that although pharmacological treatments for ADHD

are effective in treating core ADHD symptoms, a combination of such treatments with

social skills and parent training yielded additional improvements in secondary areas of

psychosocial functioning (e.g. learning, behavioral, emotional, social, and family problems).

However, there is no evidence that these clinical improvements continue in the absence of

sustained, long-term treatment with stimulant medication.

Because of concerns about the risks of long-term treatment with stimulants (Breggin,

1998; Jensen et al., 1999) examination of the effects of “nonpharmacological” treatments for

ADHD has been encouraged (Breggin, 1998). Among these treatments, EEG biofeedback,

a type of behavioral therapy developed to target the core ADHD symptoms of inattention,

impulsivity, and hyperactivity, has “generated considerable interest” (National Institute of

Health, 1998).

The initial description of the use of EEG biofeedback in the treatment of ADHD

was reported in a pair of case studies (Lubar & Shouse, 1976; Shouse & Lubar, 1979). In

their first study, Lubar and Shouse (1976) presented the application of operant conditioning

techniques to reinforce specific types of electrophysiological activity for the purpose of

treating the core symptoms of ADHD. Similar to other operant conditioning paradigms, this

treatment involved providing patients with visual and auditory “feedback” for certain “neuronal

behaviors.” Based on earlier studies by Sterman and his colleagues (summarized in

Sterman, 1996), Lubar and Shouse (1976) hypothesized that reinforcing increased production

of electrophysiological activity within either the 12–15 Hz (SMR) or 16–20 Hz (beta)

EEG Biofeedback Treatment for ADHD 233

ranges, while attempting to decrease “slower” cortical activity (4–8 Hz; theta), would result

in reduction of impulsivity/hyperactivity and improvement of attention when recordings

were obtained over either the sensorimotor region or the central frontal region. Their initial

findings were consistent with this hypothsis as reduced hyperactive behavior and improved

attention were reported in these early case studies.

Despite the positive clinical outcome of Lubar and Shouse’s application of operant

conditioning principles to treat ADHD by reinforcing electrophysiological activity within

specific frequency bands, there have been few published reports of controlled group studies

examining the efficacy of EEG biofeedback. As reviewed by Nash (2000), the majority of

reports assess efficacy via analysis of multiple case studies, in which patient performance

on certain measures (e.g., intelligence, academic skills, behavioral rating scales, continuous

performance tests) is compared pre- and posttreatment with EEG biofeedback. Although

these published case studies (e.g., Alhambra , Fowler,& Alhambra , 1995; Lubar, Swartwood,

Swartwood, & O’Donnell, 1995; Lubar, Swartwood, Swartwood, & Timmermann, 1996;

Thompson&Thompson, 1998) have yielded generally positive results, additional controlled

clinical research was needed in order to address issues of treatment efficacy.

To date, only two controlled group studies have been published. The first ( Linden ,

Habib, & Radojevic, 1996) utilized a randomized design and compared the effects of

40 sessions of EEG biofeedback (theta suppression/beta enhancement) with a “waiting

list” control. A total of 18 patients (aged 5–15) participated in the study. Treatment sessions

were conducted over a 6-month period. Medication therapy was not provided for members

of either group. Results indicated improvement on a measure of intelligence, and reduced

ADHD symptoms on a behavior rating scale in the biofeedback group.

The second “controlled” studywas conducted by Rossiter and LaVaque (1995). In their

design, 46 participants (aged 8–21) were given the opportunity to select participation in an

EEG biofeedback group or a stimulant therapy group (titrated Ritalin). Twenty sessions

of biofeedback were provided over a 3-month period. Pre- and posttreatment assessment

for both groups consisted of behavioral rating scales and the Test of Variables of Attention

(TOVA). Both groups showed significant improvement on dependent measures. There was

no significant differences between the Ritalin and the biofeedback groups.

Although these two prior investigations of EEG biofeedback utilized controlled group

designs and reported positive response on multiple dependent measures, examination of

the methodology revealed several limitations, including small sample size and absence of

follow-up data. In addition, although published EEG biofeedback protocols were utilized,

examination of electrophysiological variables in response to treatment was not conducted

and the designs did not provide a basis for comparing the unique contributions of EEG

biofeedback and other “active” (e.g., Ritalin) or “placebo” treatments. Finally, although an

attempt was made to limit other types of psychological interventions, it was evident that at

least informal “parent counseling” and other nonspecific forms of counseling were provided

to some of the participants.

The purpose of the present study was to examine the effects of EEG biofeedback and

Ritalin on the primary symptoms of ADHD, as well as, on neuropsychological and electrophysiological

measures, while controlling for other commonly provided types of clinical

interventions (stimulant therapy, parent counseling, school consultation). Because previous

controlled studies of EEG biofeedback had not provided extensive follow-up data, examination

of treatment effects was conducted 1 year after initial evaluation. Because Ritalin

has been shown to yield only short-term clinical effects (see review by Barkley, 1998) and

234 Monastra, Monastra, and George

there is case study evidence that suggests the effects of EEG biofeedback may be more

enduring (Lubar, 1995; Tansey, 1993), the use of a dismantling design was considered appropriate

for examining transitory versus sustained clinical effects. As a result, examination

of patients, both while being treated with Ritalin and following a medication “wash out”

period, seemed required. In addition, because there is evidence that systematic use of reinforcement

principles by parents/caretakers can contribute to improved social functioning

of children diagnosed with ADHD (Pisterman, McGrath, Firestone, & Goodman, 1988;

Pollard, Ward, & Barkley, 1983), evaluation of the effects of parenting style was considered

necessary in order to clarify the effects of EEG biofeedback and Ritalin on behavioral

characteristics of ADHD.

Our hypotheses were as follows. First, given prior reports suggesting the efficacy of

both stimulant therapy and EEG biofeedback, we predicted that participants being treated

with Ritalin alone or in combination with EEG biofeedback would show improvements on

behavioral and neuropsychological tests of attention and impulse control during posttreatment

evaluations conducted while using medication. Second, given the absence of long-term

clinical effect of Ritalin, as well as, the lack of measurable change on QEEG indicators

of cortical arousal over frontal and central cortical regions following administration of

methylphenidate (Lubar et al., 1996), and the case reports of reduced cortical slowing following

EEG biofeedback (Thompson&Thompson, 1998), we anticipated that only patients

who received EEG biofeedback would demonstrate improvement on QEEG measures. Furthermore,

we hypothesized that only patients who had received EEG biofeedback as part

of treatment would show sustained improvement on behavioral, neuropsychological, and

QEEG measures when tested after a 1-week medication “wash-out” period. Finally, because

systematic use of reinforcement strategies by parents/caretakers has been associated with

improved social functioning in patients diagnosed with ADHD, we predicted that parenting

style would emerge as a moderating variable on behavioral measures, regardless of the

inclusion of EEG biofeedback.

METHODS

Participants

One hundred children, ages 6–19 (83 males; 17 females), and their parents participated

in this study. Based on parental preference, patients participated in either a Comprehensive

Clinical Care (CCC) program, which included medication management, parent counseling,

and school consultation, or a CCC plus EEG Biofeedback program (CCCCB). All were

diagnosed withADHD(24:ADHD,inattentive; 76:ADHD,combined) by a licensed clinical

psychologist, based on DSM-IV criteria. None had a history of prior treatment for ADHD.

As reflected in Table I, the composition of the two groups was comparable with respect

to participant age, gender, diagnosis, intelligence, parental education, marital status, and

median family income.

All participants were screened by The Family Psychology Institute, a private outpatient

psychological clinic located in a region of Upstate New York with a population of

approximately 500,000 within a 50 mile radius of the clinic. Physicians, schools, and mental

health professionals located near the Institute referred the individuals who participated

in the study. Individuals who had previously been diagnosed and treated for ADHD or other

EEG Biofeedback Treatment for ADHD 235

Table I. Characteristics of Clinic Samples

CCC CCCCB

Gender

Male/female (N) 40=9 43=8

Age (years)

Mean 10.0 10.0

SD 3.7 3.1

Diagnosis

ADHD, Inattentive (N) 14 10

ADHD, Combined (N) 35 41

Intelligence quotient

Mean 105.9 105.2

SD 8.6 11.2

Parents

Highest grade

Mean 16.2 15.5

SD 2.5 2.3

Median income $50,000–60,000 $50,000–60,000

Marital status

Married (N) 49 50

Separated/divorced (N) 0 1

Note. CCC D comprehensive clinical care group; CCCCB D comprehensive

clinical care plus biofeedback group.

psychiatric or medical disorders that could affect attentional functions were excluded from

this study.

Procedure

Pretreatment Screening

Following physician evaluation of each participant for medical conditions (other than

ADHD), which could cause symptoms of inattention and hyperactivity (e.g., anemia, hypoglycemia,

thyroid disorders), parents were interviewed by a licensed clinical psychologist

using Barkley and Murphy’s structured format (Barkley & Murphy, 1998). This parental

interview format provides extensive information regarding medical, developmental, academic,

and social history, and serves as a foundation for comparing patient clinical history

with DSM-IV criteria for ADHD and other psychiatric disorders. In addition, this interview

provided a format for examining the behavioral management methods used by parents, and

the degree of parental consistency. In order to be accepted for participation in the study,

participants were required to meet all DSM-IV criteria for ADHD.

Subsequently, the parents completed the Home Version of the Attention Deficit Disorders

Evaluation Scale (ADDES; McCarney, 1995). The ADDES is a behavioral rating scale

that provides an indication of the frequency of ADHD symptoms and a basis for comparison

with “nonimpaired” age peers. A School Version of the ADDES was completed by the

child’s teachers. In order to be included in this study, each participant needed to be rated as

displaying a significantly greater frequency of ADHD symptoms than same age peers on

both the Home and School Version of the ADDES. Specifically, a standard score below 7

was required on the Inattentive and/or Hyperactive/Impulsive subscales of both the Home

and School Versions of the ADDES.

236 Monastra, Monastra, and George

Children meeting the behavioral criteria for inclusion in the study were then evaluated

with the Test of Variables of Attention (TOVA; Greenberg, 1996). Because of the potential

rater bias associated with behavioral rating scales like the ADDES, the use of a computeradministered/

scored test of attentional abilitieswas considered desirable in order to obtain an

“objective” measure of attention and capacity for impulse control. Continuous Performance

Tests, like the TOVA, provide an assessment of an individual’s performance on a task that

requires tracking of visual stimuli with differential response/nonresponse to target and

nontarget stimuli. In this study, errors of inattention (i.e., failure to respond to a target

stimulus) and impulsivity (i.e., response to a nontarget stimuli), as well as, response rate

and the consistency of response rate (variability) were obtained.

The TOVA was selected because it has been utilized as one of the outcome measures in

previous biofeedback studies (e.g., Lubar et al., 1995; Rossiter&LaVaque, 1995; Thompson

&Thompson, 1998) and because it has been shown to demonstrate adequate criterion related

validity when compared with physician diagnosis ofADHD(Monastra et al., 2001). In order

to be included as a participant in the study, a standard score below80was required on at least

one of the TOVA subscales (i.e., Omissions, Commissions, Response Rate, or Variability).

Finally, a Quantitative Electroencephalographic (QEEG) Scanning Process (Monastra

et al., 1999) was conducted using the Autogenics A-620 Electroencephalograph (Wood

Dale , IL ) with associated Assessment Software (Wood Dale , IL ). Because prior QEEG

research (Chabot et al., 1996; Chabot & Serfontein, 1996; Mann et al., 1992; Monastra

et al., 1999, 2001) indicated that patients with ADHD typically exhibit excessive “slow

wave” activity (4–8 Hz), relative to “fast” EEG activity (13–21 Hz) over central-midline

and frontal locations, Monastra et al.’s QEEG assessment was conducted in order to insure

that only ADHD patients showing this type of QEEG profile were included in the study

(Monastra et al., 1999).

Monastra et al.’s protocol involves a comparison of electrophysiological power recorded

at 4–8 Hz (“theta”) and 13–21 Hz (“beta”) (Monastra et al., 1999). QEEG recordings

are obtained from the vertex (Cz) with ear references. In the Monastra et al. (1999) process,

a ratio comparing the power recorded within the “theta” and “beta” frequency bands is

calculated based on QEEG data obtained during four, 90-s tasks (Baseline, Silent Reading,

Listening, and Drawing). The overall average of these “power ratios” is then determined

in order to obtain the electrophysiologically-based Attention Index. Participants needed to

exhibit an Attention Index that was at least 1.5 SD greater than age peers based on the

database provided by Monastra et al. (1999) in order to be included in this study.

All evaluations were completed between 9:00 a.m. and 3:00 p.m. None of the children

were tested within 48 hr of using any type of medication. Participants were invited

to participate in this study provided that the results of clinical interview, behavioral

rating scales (both Home and School), the TOVA, and the QEEG Scan were all positive

for ADHD. Subtype differentiation was made on the basis of interview and rating

scales.

Treatment Phase

Following the pretreatment screening, those participants who met inclusion criteria

were interviewed with their parents. A review of the treatments that would be provided

to participants in this study was given at that time. These interventions included stimulant

EEG Biofeedback Treatment for ADHD 237

therapy, parent counseling (individual and group), school consultation to establish/monitor

a program of academic support, and EEG biofeedback. All participants received stimulant

therapy, parent counseling, and school consultation. In addition, EEG biofeedback was

offered and included in the treatment program of 51 of the participants. A brief description

of each treatment follows.

Stimulant Therapy. As noted previously, all participants in the study were treated with

Ritalin. Dosage was titrated as follows. Initially, all participants were prescribed a 5 mg

dose, t.i.d. for 1 week. After a week, parents and teachers completed the Side Effects Rating

Scale (Barkley & Murphy, 1998) and the child was tested with the TOVA. If the standard

scores for all TOVA subtests were within 1.0 SD of age peers and IQ, and parent and teacher

ratings for adverse side effects indicated that the medication was well-tolerated, no change

in dose was made. If at least one of the TOVA subtest scores remained within the “clinical

range” (i.e.,>1:5 SD below age peers and IQ) dose was increased by 2.5 mg per dose and

the child was retested after 1 week. This process continued until all TOVA subtests were

within the nonclinical range. The average daily dose of Ritalin (following titration) was

25 mg, t.i.d. ( 10–10–5 ) for the CCC group (range: 15–45 mg/day). This was the same as

the average dosage administered to members of the CCCCB group (range: 15–45 mg/day).

Parent Counseling. Using the model presented by Anastopoulos, Smith, and Wien

(1998) as a foundation, all parents participated in a 10 session parenting class, followed

by individual consultation on an “as needed” basis. The program described by Anastopoulos

et al. (1998) consists of a series of educational “steps” designed to increase parental

understanding of the causes of ADHD, as well as, the role of positive parental attention

and systematic use of reinforcement strategies in reducing the functional impairments

associated with ADHD. Our parenting class also included presentations on “Problem

solving with preteens/teens” (Robin, 1998), “Nutrition,” and “The educational rights of

children with ADHD.” At the conclusion of our parenting classes, each parent had developed

and was attempting to implement a program of systematic reinforcement to address

either primary or secondary ADHD symptoms. The mean number of clinical contact

hours for parenting classes/individual consultations was 27 for the CCC group; 25 for the

CCCCB group.

School Consultation. At the conclusion of the pretreatment screening, those participants,

who met inclusion criteria, were referred by their parents to the Committee for

Special Educational Services in their home school district. Federal regulations under the

Individuals with Disabilities Act (IDEA) and the Rehabilitation Act of 1973 specify that

individuals diagnosed with ADHD are to be evaluated by their school district in order

to determine the presence/degree of learning and functional disabilities. This evaluation

is to be completed within 45 days of receipt of a letter requesting such an evaluation by

a parent (or other caregiver). Based on this school evaluation, a program of academic

remediation and/or accommodation is to be developed and monitored on an ongoing

basis.

In accordance with these laws, each of our participants was evaluated by their school

districts, and either an individualized educational program (IEP) or a plan of academic

support/accommodation (“504 Plan”) was developed, implemented, and revised with our

assistance. The mean number of on-site school consultations was three for both the CCC

and the CCC+B groups (range: 1–7). Weekly “progress” reports, listing any incomplete

assignments, upcoming projects and tests, and any behavioral incidents, were also reviewed

with the parents in order to insure parental reinforcement of “on-task” behavior at school.

238 Monastra, Monastra, and George

EEG Biofeedback. For those patients whose parents selected EEG biofeedback, individual,

weekly “attention training” sessions, lasting 30–40 min, were also provided using

the Lubar Protocol (Lubar et al., 1995). In addition to the visual and auditory feedback

that was provided by the Autogenics A-620 Neurofeedback System (Wood Dale, IL) each

time that the child produced 0.5 s of improved arousal over the frontal cortex, participants

were also reinforced for their efforts using a “point system.” When the patient accumulated

a total of 20 points (representing improved EEG performance on 20 “training” tasks)

they could exchange these points for a cash “reward” of $15. Participants typically accumulated

a sufficient number of points to earn such a “reward” every three to four sessions.

EEG biofeedback sessions were conducted until the patient exhibited a degree of

cortical slowing on the QEEG scan that was within 1.0 SD of age peers, based on the

Monastra et al. (1999) database, and were able to maintain this level of arousal for 40 min

in each of three consecutive treatment sessions. All of the participants in the CCC+B group

achieved this criterion. The average number of sessions needed to reach this goal was

43 (range: 34–50).

Posttreatment Assessment

One year after the intake evaluation, each patient was reevaluated using the ADDES

(Home and School), the TOVA, and the QEEG Scan. The first posttreatment assessmentwas

conducted while the patient was being treated with Ritalin. A second posttreatment assessment

was conducted after a 1-week medication “washout” period. During the “washout”

period, no stimulant medications were administered.

In addition to evaluating patient progress, parenting style was evaluated at the conclusion

of treatment based on interview. Parenting style was rated as “systematic” if parents

reported use of time out, removal of privileges, and use of earned privileges “most of the

time.” Parenting style was rated as “nonsystematic” if the parents failed to report use of

a combination of “reward” and “response cost” techniques, “most of the time,” or if they

reported use of physical punishment, acquiescence to child, or avoidance of the child, “most

of the time.”

Statistical analysis consisted of ANOVA to examine the main and interactional effects

of Ritalin, EEG biofeedback, and parenting style on behavioral, neuropsychological,

and QEEG measures. Post hoc analysis of significant main and interactional effects was

conducted using Tukey’s Honest Significant Difference Test. All statistical analyses were

conducted using the Statistica Software Program (StatSoft, 1995). An alpha level of at least

.05 was used for all statistical tests.

RESULTS

Pretreatment Assessment

Prior to conducting a statistical analysis of the main and interactional treatment effects

of EEG biofeedback and parenting style, analysis of pretreatment scores on behavioral,

neuropsychological, and electrophysiological measures was conducted in order to insure

that theCCCandCCCCBgroups were comparable in terms of initial severity of impairment.

The mean pretreatment scores on the ADDES, TOVA, and QEEG Scan are provided in

EEG Biofeedback Treatment for ADHD 239

Table II. Pretreatment Assessment: Without Ritalin

CCC CCCCB

Dependent measure Mean SD Mean SD F(1; 98) p

Attention Deficit Disorders Evaluation Scales: Standard scores

ADDES: Home

Inattentive 3.92 2.02 4.22 2.23 0.48 .50

Hyperactive 6.02 3.53 5.09 3.48 1.72 .20

ADDES: School

Inattentive 4.61 1.22 4.69 1.12 0.11 .70

Hyperactive 5.88 3.79 5.14 3.03 1.17 .30

Test of Variables of Attention (TOVA): Standard scores

Inattention 77.00 27.08 69.57 27.46 1.86 .18

Impulsivity 74.96 25.96 68.98 24.89 2.07 .23

Response time 85.35 20.39 87.94 18.85 0.44 .51

Variability 64.57 17.33 62.45 18.71 0.34 .56

Quantitative EEG scanning process

QEEG: Attention Indexa 5.85 2.30 5.77 1.80 0.04 .85

aAttention Index D mean theta/beta power ratio, averaged for four tasks.

Table II. The results of this initial analysis revealed no significant group differences on any

of the behavioral, neuropsychological, or QEEG measures.

Posttreatment Assessment

Behavioral Measures

One of the primary goals of this study was to examine whether EEG biofeedback

exerted any effect on behavioral, neuropsychological, or electrophysiological measures

beyond that associated with stimulant therapy. As a result, parent and teacher observations

on the ADDES were obtained on two occasions, 1 year following initial assessment. These

observations were first recorded while the participants were still being treated with Ritalin.

A second posttreatment assessment was conducted after a 1-week period in which no

stimulant therapy was provided. A summary of the mean standard scores derived from the

ADDES during these two evaluations is provided in Table III. Standard scores below 7 on

the ADDES are considered to be within the impaired range.

Initial inspection of the mean standard scores for inattentive and hyperactive/impulsive

behaviors (presented in Table III), reveals a pattern of continued impairment in the CCC

group, both at home and school. When tested 1 year after beginning stimulant therapy

(Ritalin), no indication of sustained improvement was suggested by group data, regardless

of the inclusion of medication or consideration of parenting style. All of the group means

remained under 7, an indication of the need for continued intervention.

In contrast, mean standard scores for the group that had received EEG biofeedback

(CCCCB) suggested sustained improvement, regardless of the use of Ritalin, when tested

1 year after the initial evaluation. These treatment gains were reported by parents and

teachers. Inspection of the data contained in Table III revealed group means above the

clinical cut-off score of 7 regardless of the use of Ritalin. In addition, the moderating

influence of parenting style was also suggested in Table III, as those patients whose parents

systematically employed reinforcement strategies demonstrated improved attention and

240 Monastra, Monastra, and George

Table III. Posttreatment Assessment: Behavioral Measure

Attention Deficit Disorders Evaluation Scales: Standard scoresa

Home School

Inattentive Hyperactive Inattentive Hyperactive

Mean SD Mean SD Mean SD Mean SD

Comprehensive clinical care group

With Ritalin 4.63 0.95 6.06 3.14 4.96 0.82 5.96 3.44

Without Ritalin 3.10 0.91 4.51 3.79 3.29 1.06 4.53 3.76

Ritalin and systematic parenting 4.67 0.99 5.91 3.27 4.97 0.92 5.76 3.56

Systematic parenting without Ritalin 3.12 0.96 4.45 3.89 3.24 1.20 4.27 3.87

Ritalin without systematic parenting 4.56 0.89 6.38 2.92 4.94 0.57 6.38 3.24

No Ritalin and nonsystematic parenting 3.06 0.85 4.63 3.70 3.38 0.72 5.06 3.60

Comprehensive clinical care plus

biofeedback group

With Ritalin 8.59 1.86 8.65 2.16 9.35 0.72 9.63 1.09

Without Ritalin 8.16 2.10 8.37 2.35 9.53 0.61 9.69 0.84

Ritalin and systematic parenting 9.22 1.36 9.49 1.56 9.38 0.72 9.73 1.10

Systematic parenting without Ritalin 9.19 1.05 9.51 1.39 9.68 0.53 9.84 0.80

Ritalin without systematic parenting 6.93 2.02 6.43 1.99 9.29 0.73 9.36 1.08

No Ritalin and nonsystematic parenting 5.43 1.70 5.36 1.55 9.14 0.66 9.29 0.83

aStandard scores below 7 are considered indicative of impaired functioning.

reduced hyperactivity and impulsivity at home. Illustrations of the moderating effects of

parenting style are presented in Figs. 1 and 2.

Statistical analysis of the posttreatment data, obtained while participants were being

treated with Ritalin, revealed a significant main effect associated with EEG biofeedback.

Analysis of variance results indicated that the group whose treatment included EEG

biofeedback (CCCCB Group) showed significant greater attention, F(1; 98) D 177:62;

p < :001, and less hyperactive/impulsive behaviors, F(1; 98) D 23:18 ; p < :001, at home

compared to participants whose treatment did not include EEG biofeedback (CCC group).

A similar degree of sustained improvement was also evident in the ratings of teachers,

who rated the children in the CCCCB group as more attentive, F(1; 98) D 821:30; p <

:001, and less hyperactive/impulsive, F(1; 98) D 52:49; p < :001, than those in the CCC

group.

Further analysis of the main effect of EEG biofeedback was conducted on data obtained

after a 1-week medication “washout.” This data is also included in Table III. ANOVA

results revealed that sustained improvement was reported by parents and teachers only in

the CCCCB group, who continued to display significantly fewer inattentive behaviors at

home, F(1; 98) D 239:54; p < :001, and at school, F(1; 98) D 1313:13; p < :001, than

the members of the CCC group. Similarly, the CCCCB group exhibited signficantly fewer

hyperactive and impulsive behaviors than the members of the CCC group both at home,

F(1; 98) D 37:81; p < :001, and at school, F(1; 98) D 91:02; p < :001.

Another primary goal of this study was to examine the effect of parenting style on the

manifestation of inattentive and hyperactive/impulsive behaviors. Particular interest was

placed on determining whether participants whose parents/guardians were using a “systematic”

type of parenting (i.e., consistent use of rewards and response cost strategies) would

exhibit fewer behavioral symptoms than participants whose parents were “nonsystematic”

in their parenting style.

EEG Biofeedback Treatment for ADHD 241

Fig. 1. Plot of the mean standard scores for the Inattentive Scale of the ADDES, 1 year after beginning treatment.

Scores of 6 or less indicate continued impairment. Ratings were obtained from parents while their child was

being treated with medication (IN MED) and after a 1-week medication “wash-out” (IN NOMED). Graph

depics results for two-way interaction, Parenting style (SYS: Systematic vs. NON-SYS: Nonsystematic) £ EEG

biofeedback (Yes vs. No.)

In order to clarify the contributions of EEG biofeedback and parenting style on behavioral

measures obtained from parents and teachers, interactional effects were analyzed. The

results of a statistical analysis of parental ratings indicated a significant interaction between

parenting style and EEG biofeedback. These interactional effects were evident when participants

were rated while being treated with Ritalin [Inattention: F(1; 96) D 14:73; p < :001;

Hyperactive/Impulsive: F(1; 96) D 10:23 ; p < :001], and following a 1-week medicationwashout

[Inattentive: F(1; 96) D 15:45 ; p < :001; Hyperactive/Impulsive: F(1; 96) D 5:72; p < :02].

As clarified by post hoc analyses (Tukey HSD), no significant effect of parenting

style was noted in the CCC group, regardless of the use of medication. However, in the

CCC+B group, participants whose parents consistently used effective reinforcement strategies

showed significant reduction in symptoms. This pattern was noted at 1-year follow-up

when patients were tested while using Ritalin (p < :001), as well as, after a 1-week medication

washout (p < :001).

No evidence of an interactional effect was evident in statistical analyses of teacher

ratings. This finding was noted both when participants were being treated with Ritalin

[Inattentive: F(1; 96) D 0:03; p D :86; Hyperactive/Impulsive: F(1; 96) D 0:79; p D :38],

as well as, after a 1-week medication washout [Inattentive: F(1; 96) D 0:33; p D :56;

Hyperactive/Impulsive: F(1; 96) D 1:77; p D :19]. Post hoc analyses (Tukey HSD) revealed

that significant improvements in attention and behavioral control at school was

noted in the CCCCB group, regardless of parental style or use of medication (p < :001).

In contrast, no statistically significant improvements on behavioral measures were noted in

the CCC group, whose treatment did not include EEG biofeedback.

242 Monastra, Monastra, and George

Fig. 2. Plot of the mean standard scores for the Hyperactive/Impulsive Scale of the ADDES, 1 year after

beginning treatment. Scores of 6 or less indicate continued impairment. Ratings were obtained from parents

while their child was being treatment with medication (HY MED) and after a 1-week medication “wash-out”

(HY NOMED). Graph depicts results for the two-way interaction, Parenting style (SYS: Systematic vs. NONSYS:

Nonsystematic) £ EEG biofeedback (Yes vs. No).

Neuropsychological Measure

In order to assess whether EEG biofeedback contributed to sustained improvement on

a computerized test of attention and impulse control, participants in this study were retested

with the TOVA, 1 year after their initial evaluation. During this posttreatment period, the

TOVA was administered on two occasions, once while being treated with Ritalin and again

after a 1-week medication washout period. As described previously, errors of inattention

(i.e., failure to respond to a target stimulus) and impulsivity (i.e., response to a nontarget

stimuli), as well as, response rate and the consistency of response rate (variability) were

obtained in order to assess the sustained effects of EEG biofeedback. Standard scores below

80 on any of the TOVA subscales are considered to be significantly less than anticipated in

individuals with average intelligence (such as our sample).

Table IV presents the standard scores and results of ANOVAs for both the CCC and

CCCCB groups when participants were tested 1 year after their initial assessment. The

upper half of this table reflects the positive effects of stimulant therapy in the CCC and the

CCCCB groups. The mean scores for both groups were well within the unimpaired range.

Comparison between both groups showed there was no significant difference between the

performance of the two groups on any of the four TOVA subscales.

Following a 1-week discontinuation of Ritalin, participants from both groups were

reevaluated with the TOVA. The results of this subsequent assessment are summarized in

the lower half of Table IV. Significant differences in performance were noted between the

CCC and CCCCB groups on all TOVA subscales. The CCC group exhibited an anticipated

EEG Biofeedback Treatment for ADHD 243

Table IV. Posttreatment Assessment: Neuropsychological Measure

Test of Variables of Attention (TOVA)

Comprehensive clinical

Comprehensive clinical care care C biofeedback

Dependent measure Mean SD Mean SD F(1; 98) p

With Ritalin

Inattention 102.24 5.89 101.45 7.21 0.36 .55

Impulsivity 103.96 7.60 101.10 10.78 2.34 .13

Response time 100.65 9.16 102.20 10.67 0.59 .44

Variability 98.98 10.53 100.10 9.56 0.31 .58

Without Ritalin

Inattention 76.24 22.71 98.92 7.65 45.35 <.001

Impulsivity 79.82 23.48 95.16 14.67 15.48 <.001

Response time 88.24 17.05 97.02 8.19 10.89 <.001

Variability 64.04 12.44 94.39 9.49 189.11 <.001

return to baseline level of performance once Ritalin was discontinued. However, the group

that had received EEG biofeedback (CCCCB) sustained a level of performance that was

well within the unimpaired range. The TOVA scores in the CCCCB group revealed a level

of performance that was significantly higher than the CCC group and the CCCCB baseline

measures.

QEEG Scanning Process

In order to evaluate whether the effects of EEG biofeedback were simply a placebo

effect, inclusion of a “biological” measure was considered essential. Because QEEG studies

published by three research teams (Chabot&Serfontein, 1996; Mann et al., 1992; Monastra

et al., 1999) had revealed significant differences in the degree of cortical “arousal” measured

electrophysiologically over central and frontal locations, the use of QEEG assessment was

considered appropriate as a “biological” measure of attention. Monastra et al.’s QEEG

Scanning Process (Monastra et al., 1999) was selected for this study due to the availability

of a published normative database, as well as, published reliability and cross validation