EEG Differences in ADHD-Combined Type
During Baseline and Cognitive Tasks
Jeffery N. Swartwood, PhD*, Michie O. Swartwood, PhD*, Joel F. Lubar, PhD, and
DeAnna L. Timmermann, PhD
This study examines the relation between neurologic,
behavioral, and performance indicators of attention-
deficit hyperactivity disorder. Twenty-three males age
nine to 11 years with attention-deficit hyperactivity
disorder, including symptoms of hyperactivity, and 23
matched controls served as participants. Differences
between groups were investigated using referential
19-channel quantitative electroencephalogram, behav-
ioral rating scale data, and continuous performance
test data. Results from the behavioral data were con-
sistent with previous research. Behavioral ratings for
participants with attention-deficit hyperactivity disor-
der were significantly more negative than controls.
Control participants performed significantly better on
the continuous performance test, with fewer errors,
faster reaction times, and less variability in reaction
time. Electroencephalogram results indicated differ-
ences between participants with attention-deficit hy-
peractivity disorder and control participants primarily
in the alpha bandpass, with evidence of increased alpha
in posterior regions during baseline for the group with
attention-deficit hyperactivity disorder. Additionally,
participants with attention-deficit hyperactivity disor-
der manifested decreased alpha in left frontal regions
when reading. The results are discussed in terms of
possible differences in electroencephalographic data as
a function of degree of hyperactivity, as well as the
impact of task specificity on the electroencephalogram.
2003 by Elsevier Inc. All rights reserved.
Swartwood JN, Swartwood MO, Lubar JF, Timmermann
DL. EEG differences in ADHD-combined type during
baseline and cognitive tasks. Pediatr Neurol 2003;28:
199-204.
From the *Department of Psychology, State University of New York
at Cortland, Cortland, New York and theDepartment of Psychology,
University of Tennessee at Knoxville, Knoxville, Tennessee.
2003 by Elsevier Inc. All rights reserved.
doi:10.1016/S0887-8994(02)00514-3 0887-8994/03/$see front matter
Introduction
Among the psychiatric diagnoses of childhood, atten-
tion-deficit hyperactivity disorder (ADHD) is one of the
most debated. Establishing a distinct syndrome has been
difficult because of such issues as a lack of differentiation
from other behavioral and learning disorders, low levels of
agreement across different measures, and a lack of a single
etiology or a consistent response to treatment [1,2]. Part of
the problem may be that the current descriptors of the
disorder are too broad, relying heavily on reports of
behavioral symptoms by parents or teachers. The devel-
opment of more objective measures would enhance the
validity and reliability of an ADHD diagnosis. Although
ADHD is currently a behaviorally driven diagnosis, re-
search is beginning to provide evidence for the use of
neurometric tools to facilitate the diagnosis.
One such neurometric that has been actively investi-
gated is quantitative electroencephalogram (EEG). Con-
sensus, however, regarding EEG differences in ADHD
across different studies has not been reached. Several
studies have demonstrated that individuals with ADHD
exhibit increases in slow-wave activity, particularly in the
frontal regions of the brain, which is sometimes accom-
panied by a decrease in fast wave activity in the posterior
region [3,4]. Other studies have failed to localize the
excess in slow-wave activity to the frontal regions [5,6]
but have found a diffuse increase in slow-wave activity
from recording sites across the entire brain. In addition to
these findings, Clarke et al. [6,7] found an increase in
fast-wave activity in the frontal region of the brain for a
subset of the participants with ADHD. Clarke [5] noted
that a greater proportion of children with ADHD-com-
bined type had this increase in fast-wave activity com-
pared with participants with ADHD-inattentive type.
It is possible that some of the variability in findings
across studies may be due to differences in the participant
Communications should be addressed to:
Dr. Swartwood; SUNY Cortland, Psychology Department,
Cortland, NY 13045.
Received April 2, 2002; accepted August 26, 2002.
Swartwood et al: EEG Differences in ADHD 199
population that are revealed by EEG data but not neces-
sarily discriminated by using standard diagnostic criteria.
Other sources of interstudy variability may be due to
procedural differences during the process of EEG collec-
tion. Some researchers record EEG data from the partici-
pants during an eyes closed baseline condition [5,6],
whereas others have examined participants during en-
gaged tasks (reading, listening, and drawing) in addition to
baseline conditions [4,8]. General conclusions regarding
research in this area depends on these participant and
procedural differences. For example, although the Mann et
al. [4] study investigated EEG changes during both base-
line and cognitive tasks in children with ADHD-inatten-
tive type, whether or not the same results would be
observed in those children diagnosed as ADHD-combined
type is not known.
The present study compares the EEGs of children with
ADHD-combined type on a variety of baseline and cog-
nitive tasks with the EEGs of typically developing chil-
dren. Given the results of research with nonhyperactive
children with ADHD during cognitive tasks [4,7], it is
expected that children with ADHD-combined type will
exhibit increased theta activity, possibly localized in
frontal areas. Additionally, research from Clarke [5,6]
indicates that participants with ADHD-combined type
may exhibit increases in beta. This study seeks to integrate
and expand previous EEG diagnostic research and to
further develop an electrophysiologic profile in children
with ADHD-combined type.
Method
Participants
Participants for this study were 23 male children diagnosed with
attention-deficit hyperactivity disorder (ADHD-combined type) and 23
male control children. The age range of the participants was from 9 to 11
years. Participation was voluntary, and participants were recruited
through either the local school system, Children and Adults with
Attention Deficit/Hyperactivity Disorder (C.H.A.D.D.) group, or a neu-
rology clinic. Participants were not known to the investigators before
their participation in the study. No monetary payment was used to recruit
the participants; however, parents did receive results from the neurophys-
iological and behavioral test batteries.
Participants were included in the ADHD group if they met the
following criteria: (1) they had been diagnosed by a physician as having
ADHD according to the Diagnostic and Statistical Manual and had
complaints of hyperactivity, inattention, and impulsivity; (2) they were
currently being treated with methylphenidate; (3) they had not been
diagnosed as having a learning disability; (4) they were rated by a parent
as having symptoms of ADHD (greater than one standard deviation from
the mean) on the McCarney Attention Deficit Disorder Evaluation Scale
(ADDES); and (5) they scored in the average range on the Wide Range
Achievement Test (WRAT). All participants in the ADHD group had
medication administration suspended for at least 48 hours before testing.
Participants in the control group were age matched to participants in
the ADHD group. Control participants were not rated on any subscale of
the ADDES as having symptoms consistent with a diagnosis of ADHD.
Any individual who volunteered for the study as a control participant
who was rated as one standard deviation or more from the mean on the
200 PEDIATRIC NEUROLOGY Vol. 28 No. 3
ADDES rating scale norms was not included in the study. Additionally,
subjects in the control group were not diagnosed with other learning or
behavior disorders.
Materials
Participants were tested using several neurological and psychological
test materials. The McCarney Attention Deficit Disorder Evaluation
Scale (ADDES) [9] was used to collect behavior ratings from parents
regarding their child. The Wide Range Achievement Test-Revised
(WRAT-R) [10] was used to screen academic performance and to
determine appropriate materials for the EEG conditions. The Test of
Variables of Attention (TOVA) [11], a normed continuous performance
test, was used as a measure to assess performance differences between
groups. During the TOVA, visual stimuli were briefly presented every 2
seconds for 22.5 minutes. Participants followed standardized instructions
to watch the screen and click a button whenever a colored square
appeared at the top portion of an outer square (target stimulus). If the
square appeared at the bottom portion of the outer square (nontarget
stimulus), participants were told to refrain from clicking. Scores derived
from the TOVA provided measures of errors of omission, errors of
commission, mean correct response time, and response variability.
Referential electroencephalographic data were recorded through the use
of the Lexicor NRS-24 (Lexicor Health Systems: Boulder, CO) equip-
ment and software.
Procedure
The process for administering neurologic and physiologic tests was the
same for both the ADHD and control groups. Testing took approximately
3 hours on a single day. Parents completed the McCarney Attention
Deficit Disorder Evaluation Scale (ADDES) while the children were
tested. Data were collected in the same order for all participants. The
WRAT-R was administered first, followed by the TOVA, and finally the
referential 19-channel EEG was recorded. Individual scores on the
WRAT-R were used to determine appropriate materials for the child
during the reading and story conditions.
The quantitative referential 19-channel EEG data were collected using
the International 10-20 system for electrode placement, including the
central locations. Electrode placements were made using an electrode cap
(Electro Cap Co.: Eaton, OH) and Electrogel conductive gel (Weaver and
Co.: Aurora, CO). An electrode was placed at Oz using a discrete
electrode to provide an additional recording site. Linked ear electrodes
were used as the reference. A cotton ball was placed behind the ear
electrodes to isolate the referent from potential sources of contamination.
Data were collected under six conditions: eyes open baseline, reading,
drawing, the Ravens Progressive Matrices [12], listening to a story, and
the coding task from the WISC-R [13]. The EEG was recorded for 2
minutes for each condition.
Data management
All EEG data were visually inspected and artifact rejected before any
statistical analyses were performed. Segments of EEG that involved
obvious ocular or excessive muscle activity were deleted. Each partici-
pants total amount of artifact free EEG used for analyses varied, but at
least 1 minute of data were analyzed for each condition. Artifact rejection
was completed by two extensively trained researchers with an inter-rater
reliability coefficient of 0.92. Bandpasses were defined as follows: delta
1 (D1) 0.75-1.75 Hz, delta 2 (D2) 2-3.75 Hz, theta (TH) 4-7.75
Hz, alpha (AL) 8-12.75, beta 1 (B1) 13-21.75, and beta 2 (B2)
22-31 Hz.
Results Table 1. Significant t tests on log transformed data

Psychometric measures ADHD Control t-Test

Frequency/Condition Location Mean Mean Value
There were no significant differences in age or grade
Delta 1/Story
between the ADHD group and the normal control Fp1 0.35 0.48
group. Results from the ADDES, which were used to 2.78
establish the groups of children with and without ADHD, Fp2 0.38 0.45 2.15*
F8 0.38 0.44 2.2*
indicated significantly lower scores (poorer perfor- Fz 0.43 0.50 2.44*
mance) for the ADHD group on all subscales: inattention Alpha/Baseline
[t(44) 9.53, P 0.0005], impulsivity [t(44) 7.50, P O2 0.46 0.60 2.02*
T5 0.56 0.70 2.16*
0.0005], and hyperactivity [t(44) 7.97, P 0.0005]. T6 0.52 0.69 2.44*
Results from the TOVA continuous performance test Alpha/Reading
revealed that the ADHD group had significantly poorer F3 0.90 0.83 2.07*
performance on scores of omissions [t(44) 2.40, P F7 0.98 0.92 2.16*
Alpha/Coding
0.021)], reaction time [t(44) 4.49, P 0.0005], and T5 0.78 0.87
variability [t(44) 3.48, P 0.001]. Academic perfor- 2.84
mance as measured by the WRAT-R indicated signifi-
Note: Percentage EEG activity at a particular location increases as the
cantly lower scores for the ADHD group on spelling mean of the log transformed data becomes less negative.
[t(44) 5.72, P 0.0005], reading [t(44) 4.26, P * P 0.05
0.0005], and mathematics [t(44) 3.09, P 0.003]
P 0.01
sub-tests; however, standard scores for the children with
ADHD were all average and ranged from 90-105. Stan-
Correlations among EEG, ADDES, and TOVA
dard scores for the normal control group ranged from
107-119, which placed them between one and two stan- To assess possible relationships among EEG, ADDES,
dard deviations from the mean. and TOVA for the participants with ADHD, a Pearson
Product Moment correlation coefficient matrix was com-
puted. Only the EEG measures that significantly differed
EEG measures according to the initial multiphase analysis were correlated
with the three subscales of the ADDES and the four
Data analyses were conducted using a multiphase ap- measures generated from the TOVA. The significant
proach [4,14] that limits what portion of the data is results from this correlational matrix are presented in
analyzed, thereby reducing experiment-wise error. This Table 2.
approach uses a combination of data reduction techniques
and traditional parametric statistics. First, the total number Discussion
of data points to be compared was reduced using a Psychometric measures
principle component analysis (PCA) on the 19 recording
locations. This analysis reduced the data to a smaller set The TOVA results, excluding the commissions score,
of linear combinations that accounted for the majority of revealed significant differences between the children with
variance in the original data. Then, individual t tests on and without ADHD. For number of omissions, reaction
the actual data from all 19 recording sites were performed time, and reaction time variability measurements, the
for those factors from a given condition demonstrating a ADHD group had poorer performance compared with the
significant difference between groups. Raw data were log normal control group. These findings, in conjunction with
transformed to address the naturally skewed distribution the significant differences on the ADDES that were
initially used to establish group membership, provide
of the EEG.
support that the two groups of participants in this study
The significant findings from the PCA are summar-
demonstrated distinct patterns of behavior.
ized in Table 1 and graphically illustrated in Figure 1.
Analyses were based on percent power. Figure 1 reveals
increased right frontal delta 1 activity in the ADHD EEG measures
group during the story condition, increased alpha The EEG data revealed significant differences between
activity in the ADHD group in the posterior regions children with and without ADHD; these differences were
during the baseline condition, increased alpha activity task and bandpass specific. Differences were seen in the
in the normal control group during the reading condi- alpha band pass during the eyes open baseline condition.
tion in the left frontal region, and increased alpha Significantly more alpha was identified in the participants
activity in the ADHD group during the coding task at with ADHD at T5, T6, and O2. These findings may
T5 only. indicate that the participants with ADHD failed to sup-

Swartwood et al: EEG Differences in ADHD 201

 Figure 1. Significant differences in EEG
data by cognitive condition.

Table 2. Significant correlations between ADDES, TOVA, and press alpha activity during an eyes open baseline relative
EEG for the ADHD group
to normal controls. This finding is consistent with previous
Correlation Probability research [3] using baseline measures in individuals with
Test Subscale/EEG data Coefficient Level ADHD and symptoms of impulsivity/hyperactivity. In
normal individuals, alpha activity in the posterior regions
ADHD Group
TOVA reaction time/F7-TH(drawing) 0.4687 0.024 of the brain attenuates when the eyes open [15]. Reviews
TOVA reaction time/FZ-AL(ravens) 0.5346 0.009 of alpha-related activity have generally supported the
TOVA reaction time/T5-TH(story) 0.5203 0.011 relation of this desynchronization in the posterior regions
TOVA variability/FZ-AL (ravens) 0.5066 0.014
ADDES impulsivity/T5-B1(coding) 0.4359 0.038
of the brain to engagement of the occipital lobes in the
processing of incoming visual stimuli [15]. The failure of
Note. EEG data is in the following format: location-frequency participants with ADHD to suppress alpha during a base-
(condition). Bandpasses were defined as follows: delta 1 (D1)
line condition supports the behavioral research findings
0.751.75 Hz, delta 2 (D2) 23.75 Hz, theta (TH) 4 7.75 Hz,
alpha (AL) 8 12.75, beta 1 (B1) 1321.75, and beta 2 (B2) that children with ADHD are not able to attend to and
2231 Hz. process visual stimuli as efficiently as children without
ADHD. It should also be noted that research has demon-
Abbreviations:
strated that alpha activity is impacted by anxiety [16] and
ADDES Attention Deficit Disorders Evaluation Scale increased during certain meditative states [17]; both of
TOVA Test of Variables of Attention which involve attentional processes.

202 PEDIATRIC NEUROLOGY Vol. 28 No. 3

 Alpha production was also found to differ during the
coding condition at T5; participants with ADHD produced
significantly more alpha at this location. Since the ADHD
group also had significantly more alpha-wave production at
T5 during the baseline condition, the higher alpha seen
during the coding condition could represent a persistence of
the abnormally high alpha waves observed during the base-
line. Alpha persistence in the left parietal region may be
indicative of parietal underactivation during the coding task.
The parietal lobes are involved in writing [18], language [19],
and drawing [20], all of which are factors that could be
expected to be active during the coding condition.
The finding of increased left frontal alpha activity in the
normal control group during the reading condition is
difficult to interpret. Researchers typically interpret the
presence of alpha activity in a particular region to signify
inactivation of that region [21]. It is interesting to note,
however, that increases in frontal alpha were also found
during a reading condition by Swartwood [14] in children
with ADHD while they were on methylphenidate. Perhaps
methylphenidate normalized the participants frontal alpha
similar to the effect seen in the control participants in the
present study. Additional research is necessary to further
elucidate the importance of frontal alpha during reading.
Results during the story condition indicate significant
differences in the delta 1 bandpass at FP1, FP2, FZ, and
F8. For each of these locations, the mean percent power
for the ADHD group was significantly higher than the
normal control group. These differences cluster around
frontal regions (FP1, FP2, FZ) with an asymmetrical
location on the right (F8). It is possible that control
participants followed the instructions to visually fixate on
a point better than the participants with ADHD, since this
is a common problem for children with ADHD [22].
Alternatively, this finding may have its origin in a possible
delay in the functional maturity of the brain in children
with ADHD, as suggested by Matsuura et al. [23].
Correlations among EEG, TOVA, and ADDES
Correlations were conducted to assess possible relations
among the EEG, TOVA, and the ADDES for the partici-
pants with ADHD. The standard score for reaction time on
the TOVA was negatively related to theta activity for left
frontal and temporal regions during story and drawing
tasks (Table 2). That is, higher percent power in theta was
associated with lower standard scores in reaction time on
the TOVA. This is consistent with suggestions by Mann
[4] that attention decreases as theta power increases. This
decrease in attention probably made it more difficult for
the ADHD group to respond promptly to the target stimuli,
resulting in the lower reaction time scores. The other
significant correlations for the ADHD group were between
a single test and a single EEG variable. Without any
cohesive pattern, a reliable interpretation of the data is not
feasible.
General conclusions
Results from the present study indicate few EEG dif-
ferences between male children with ADHD-combined
type and matched control participants. The significant
differences that were indicated by this study depended on
the task presented to the participants. Mann et al. [4] found
consistent differences between ADD participants without
hyperactivity and control participants on baseline and
cognitively engaged tasks. The majority of these differ-
ences were associated with increased theta in the central
regions and decreased beta in posterior regions. The
current findings differ in that there were not only fewer
significant differences overall, but those differences that
were found appeared to center in the alpha frequency. The
most likely explanation for this discrepancy lies in the
participant selection process. In the Mann et al. study the
participants were specifically chosen because they were
free from complaints of hyperactivity. The present study
specifically chose participants that experienced hyperac-
tivity and impulsivity; therefore, the different results
obtained in the two studies may be due to inherent
neurologic differences between children with and without
hyperactivity.
It may also be the case that the cognitive tasks chosen in
this particular study did not maximize potential differ-
ences between participants with ADHD-combined type
and those children not diagnosed as ADHD. Differences
between children with and without ADHD are often
elusive, and they are often observable only when elicited
by environmental demands or consequences [1]. Future
research exploring differences between children with and
without ADHD during tasks specifically designed to tap
executive functions, such as the ability to learn from
consequences, to self-monitor behavior, and to inhibit
unsuccessful behaviors, may yield more consistent find-
ings, as these behaviors represent fundamental problems
for individuals with this disorder. Furthermore, studies
could make additional within participant comparisons,
examining how a participants baseline EEG differs from
their task-related EEG and how regional patterns vary
according to task (i.e., left/right or frontal/occipital).
As a final note, future studies need to highlight meth-
odologic differences when comparing experimental results
across studies. Differences in participant population, EEG
quantification, and statistical analyses are likely to impact
results and may be a factor in the apparent conflicting
results in ADHD research regarding EEG and cognitive
activity. A meta-analysis of the existing research in this
area would help to clarify our present understanding of the
disorder and provide insight into directions for future
research.
References
[1] Barkley RA. Attention deficit hyperactivity disorder, A hand-
book for diagnosis and treatment. New York: Guilford Press, 1998:3-55.
Swartwood et al: EEG Differences in ADHD 203
 [2] Rutter M. Cognitive deficits in the pathogenesis of autism.
J Child Psychol Psychiatry 1983;24:513-31.
[3] Lazzaro I, Gordon E, Whitmont S, et al. Quantified EEG activity
in adolescent attention deficit hyperactivity disorder. Clin Electroen-
cephalogr 1998;29:37-42.
[4] Mann CA, Lubar JF, Zimmerman AW, Miller CA, Muenchen
RA. Quantitative analysis of EEG in boys with attention deficit-
hyperactivity disorder: Controlled study with clinical implications. Pe-
diatr Neurol 1992;8:30-6.
[5] Clark A, Barry R, McCarthy R, Selikowitz M. Excess beta
activity in children with attention-deficit/hyperactivity disorder: An
atypical electrophysiological group. Psychiatry Res 2001(b);103:205-18.
[6] Clarke A, Barry R, McCarthy R, Selikowitz M. EEG analysis in
attention-deficit/hyperactivity disorder: A comparative study of two
subtypes. Psychiatry Res 1998;81:19-29.
[7] Clarke A, Barry R, McCarthy R, Selikowitz M. EEG differences
in two subtypes of attention-deficit/hyperactivity disorder. Psychophys-
iology 2001(a);38:212-21.
[8] Janzen T, Graap K, Stephanson S, Marshall W, Fitzsimmons G.
Differences in baseline EEG measures for ADD and normally achieving
preadolescent males. Biofeedback Self Regul 1995;20:65-82.
[9] McCarney SB. The attention deficit disorders Evaluation Scale
Home Version Technical Manual. Columbia, MO: Hawthorne Educa-
tional Services, Inc., 1989:1-5.
[10] Jastak JF, Jastak SR. Wide range achievement test-revised
(WRAT-R). Wilmington: JASTAK Assessment Systems, 1984:2-5.
[11] Greenberg L. Test of variables of attention. Los Alamitos, CA:
Universal Attention Disorders, 1998:1-3.
[12] Raven JC. Guide to the Standard Progressive Matrices. London:
H. K. Lewis, 1960:1-4.
204 PEDIATRIC NEUROLOGY Vol. 28 No. 3
[13] Wechsler D. Wechsler intelligence scale for children-Revised
(WISC-R). New York: Psychological Corporation, 1974:1-10.
[14] Swartwood M, Swartwood J, Lubar J, Timmermann D,
Zimmerman A, Muenchen R. Methylphenidate effects on EEG, be-
havior, and performance in boys with ADHD. Pediatr Neurol 1998;18:
244-50.
[15] Williamson S, Kaufman L, Lu Z, Wang J, Karron D. Study of
human occipital alpha rhythm: The alphon hypothesis and alpha suppres-
sion. Int J Psychophysiol 1997;26:63-76.
[16] Hardt J, Kamiya J. Anxiety change through electroencephalo-
graphic alpha feedback seen only in high anxiety subjects. Science
1978;201(4350):79-81.
[17] Travis F. Autonomic and EEG patterns distinguish transcend-
ing from other experiences during Transcendental Mediation practice. Int
J Psychophysiol 2001;42:1-9.
[18] Kimura D. Acquisition of a motor skill after left hemisphere
damage. Brain 1977;100:527-42.
[19] Simon O, Mangin J, Cohen L, Le Bihan D, Dehaene S.
Topographical layout of hand, eye calculation, and language related areas
in the human parietal lobe. Neuron 2002;33:475-87.
[20] Warrington EK, James M, Kinsbourne M. Drawing disability
in relation to laterality of cerebral lesion. Brain 1966;89:53-82.
[21] Henriques JB, Davidson RJ. Left frontal hypoactivation in
depression. J Abnorm Psychol 1991;100:535-45.
[22] DuPaul GJ, Rapport MD. Does methylphenidate normalize the
classroom performance of children with attention deficit disorder? J Am
Acad Child Adolesc Psychiatry 1993;32:190-8.
[23] Matsuura M, Okubo Y, Toru M, et al. A cross-national EEG
study of children with emotional and behavioral problems: A WHO
collaborative study in the western pacific region. Biol Psychiatry 1993;
34:59-65.