Attention Deficit/Hyperactivity Disorder (AD/HD) can occur with or without hyperactivity. However, much of the existing research collapses both AD/HD and AD/HD without hyperactivity participants into the AD/HD category, possibly confounding the samples with a heterogeneous population comprised of people with different disorders. The purpose of the present study was to examine the external validity of AD/HD without hyperactivity as a diagnostic category. Quantitative electroencephalogram (EEG) analysis was used to examine possible differences in brain wave activity of the two subtypes of AD/HD while completing the Test of Variables of Attention (TOVA), a computerized task that measures a variety of constructs associated with attention and impulsivity. Although behavioral ratings confirmed differential characteristics of both subtypes of AD/HD, EEG findings did not differentiate between AD/HD with and without hyperactivity. Implications to cognitive models of AD/HD are discussed.
Attention Deficit/Hyperactivity Disorder (AD/HD) is one of the most common childhood behavior disorders and is estimated to affect 3 to 5% of school-age children (DSM-IV; American Psychiatric Association, 1994). However, the disorder has a long and confusing history, having been referred to as the Hyperkinetic Reaction of Childhood Disorder (American Psychiatric Association, 1968), Hyperactivity (Zentall, 1975), Minimal Brain Dysfunction (Bloomingdale & Bloomingdale, 1980), and Childhood Hyperkinesis (Mattes, 1980).
The predominant symptoms of AD/HD are inattention, excessive impulsivity, and/or hyperactivity. Two subtypes of the disorder were included in the Diagnostic and Statistical Manual of Mental Disorders (DSM-III; American Psychiatric Association, 1980): Attention Deficit Disorder with Hyperactivity (ADD/H) and Attention Deficit Disorder without Hyperactivity (ADD/WO). With the publication of the DSM-III-R (American Psychiatric Association, 1987), the distinction between these two subtypes was effectively removed by adopting a unidimensional category referred to as ADHD. Instead, a new diagnosis called Undifferentiated Attention Deficit Disorder (UADD) was created that included some of the disturbances that were previously classified as ADD/WO.
After considerable debate, the publication of the DSM-IV (American Psychiatric Association, 1994) reinstated the diagnosis by establishing three subtypes of AD/HD: the Predominately Inattentive Type, the Combined Type, and the Predominately Hyperactive-Impulsive Type. Most of the evidence supporting the validity of attention deficit without hyperactivity as a subtype of AD/HD comes from observations of overt behavior, comorbid diagnoses, and familial patterns of psychiatric disturbances (see Stewart, 1994). In spite of the specificity in diagnostic criteria the DSM-IV provides, it is still unclear whether attention deficit without hyperactivity warrants recognition as a separate disorder.
From a theoretical and clinical standpoint it is important to establish the external validity of AD/HD without hyperactivity as a disorder that can be clinically differentiated from AD/HD. Presently, most research collapses both subtypes of AD/HD, possibly confounding the samples with a heterogeneous population comprised of people with different disorders (Castellanos, 1999; Hynd et al., 1991).
In recent years, electroencephalogram (EEG) analysis has been used in the diagnosis of AD/HD. Compared to normal controls, children who are described as hyperactive typically exhibit excessive slow wave activity (typically in the theta band) and/or concomitant decrease in fast wave (primarily alpha and beta) activity (Callaway, Halliday, & Naylor, 1983; Dykman, Holcomb, Oglesby, & Ackerman, 1982; Harper, Deering, Cavernos-Gonzales, McNeil, & Ulam, 1996; Mann, Lubar, Zimmerman, Miller, & Muenchen, 1992; Matsuura et al., 1993). Excessive slow wave activity in AD/HD is a neurophysiological response consistent with a hypoarousal hypothesis of hyperactivity (Klove, 1989; Zentall, 1975; Zentall & Zentall, 1983) and provides evidence that AD/HD is a neurophysiological disorder.
Researchers have begun to examine the differences in on-task EEG recordings between AD/HD without hyperactivity versus controls. Results suggest that children diagnosed with AD/HD without hyperactivity may exhibit EEG patterns that are similar to those exhibited by AD/HD with hyperactivity, that is excessive slow wave activity in the theta band and decreased activity in beta bands compared to matched controls (Janzen, Graap, Stephanson, Marshall, & Fitzsimmons, 1995; Mann, Lubar, Zimmerman, Miller, & Muenchen, 1992). However, there has been no comparison of AD/HD with or without hyperactivity. It is possible that these two groups exhibit different patterns of EEG characteristics while completing an attentional task and if so, EEG analysis may be an effective tool in differentiating children with and without the hyperactivity component of AD/HD.
The purpose of the present study was to assess the external validity of AD/RD without hyperactivity as a diagnostic category by comparing a carefully selected group of children who met the DSMIV criteria for AD/HD, predominately inattentive type, to a group of children who met the criteria for AD/HD, combined type. Consistent with the conjectures of Dykman and Ackerman (1993), psychophysiological (EEG) data were collected while the participants completed a computerized task (Test of Variables of Attention: TOVA) that measures a variety of constructs associated with attention and impulsivity. The TOVA was selected as a representative measure of attention and the inhibition of responding. Further, it requires very little movement (only the flexing of the thumb) and no visual scanning, highly desirable characteristics as they minimize the occurrence of movement artifact in the EEG record. As these groups have never been compared on EEG or TOVA variables, this research was exploratory and as such, the primary aim was to identify the nuances between the groups to provide direction for future investigations.
Methods and Materials
Participants with AD/HD were accessed through an Assessment Clinic of a local university or through a local association for attentional disorders. The control sample consisted of children who attended either a local school or church congregation. Participation was voluntary and both the child and one parent were required to sign a consent form, with the understanding that the child could withdraw from the study at any time.
Strict selection criteria were used in order to eliminate confounding factors. All participants were right handed males, of average intelligence or higher, and permitted to be free of any medication used to treat AD/HD on the day the EEG was completed. Those who were diagnosed with comorbid disorders, previous head injuries, learning disabilities, or taking a long-acting medication as a treatment of AD/HD were excluded.
During the first phase of the selection process, mothers were asked to complete a checklist comprised of the symptoms used to diagnose AD/HD according to the DSM-JV (American Psychiatric Association, 1994) and the Parent Report Scale of the Behavior Assessment System for Children (BASC; Reynolds & Kamphaus, 1992). Potential participants for the hyperactive and without hyperactivity groups were identified based on the DSM-IV checklist. Children who received ratings of either pretty much or very much on six or more of the inattention items but less than five of the hyperactivity-impulsivity items were classified as without hyperactivity (AD/HD-). Similarly, those who were rated as either pretty much or very much on six or more of the inattention items and six or more of the hyperactivity-impulsivity items were classified as with hyperactivity (AD/HD+). No one in the control group received ratings of pretty much or very much on more than four of either the inattention or the hyperactivity-impulsivity items. For the purposes of the statistical analysis, values were assigned to reflect the presence or absence of each of the 18 symptoms on the DSM-IV checklist. One point was assigned for a rating of very much or pretty much, while a rating of just a little or not at all received zero points. Children who did not meet these criteria were excluded from the study.
During the second phase of the selection process, all participants were administered the Wechsler Intelligence Scale for Children -- Third Revision (WISC-III; Wechsler, 1991) and the Screening Form of the Wechsler Individual Achievement Test (WIAT; Wechsler, 1992). All children included in the study had a Full Scale Intelligence Quotient that fell at or above 90. In addition, the Reading, Spelling, and Mathematics Reasoning scores from the WIAT were above 85 for all subjects (i.e., within one standard deviation of the mean or higher). No one in the control group exhibited any behavioral or academic difficulties according to the mother's report. Furthermore, none of them had any chronic health problems nor were they taking any prescription medication at the time of the EEG evaluation.
The final sample consisted of 27 males between 10 and 12 years of age, 9 in each group. Previous research indicates that EEG characteristics of boys between the ages of 9 and 13 are stable for up to three years (Fein, Galin, Yingling, Johnstone, & Nelson, 1984). The means for age were 10.85, 11.58, 11.66 years (SD = .53, 1.08, .92) for the AD/HD+, AD-HD-, and control groups, respectively. There were no age differences between the groups, F (2,24) = 2.29, p = .122.
Monopolar recording of the EEG was performed using the Neurosearch-24 system developed by the Lexicor Corporation (Lexicor Medical Technology, 1990). A nylon electrocap with twenty electrodes positioned according the International 10-20 system with linked ear reference was used to collect the EEG data. Settings on the Neurosearch-24 were as follows: Sampling rate = 128 samples per second; Gain = 32000; High Pass Filter = 0.5, Low Pass Filter = 32 Hertz. The EEG recordings were initially saved on a computer hard drive before being transferred to tape.
The EEG recordings took place in a biofeedback training room at the local university between the hours of 9 a.m. and 3 p.m. All participants either washed their hair prior to arriving for the recording session or had it washed and dried in the data collection room immediately before they were fitted with the electrocap. Each site was prepared using Electro-gel until all impedances measured below 5 ohms.
Prior to data collection, participants were permitted to observe their EEG patterns on a computer monitor. They were also given an opportunity to observe the production of unwanted artifact by creating muscle tension and moving their eyes, followed by a request to avoid this type of activity during the data collection process. EEG data was collected under three conditions: resting with eyes open for five minutes, resting with eyes closed for five minutes, and during the completion of the TOVA. A short break of one or two minutes was taken between the recording of the eyes open, eyes closed, and TOVA conditions, but the entire TOVA was administered to each participant without interruption.
The TOVA is a computerized continuous performance test that utilizes simple geometric stimuli and requires no language skill or right-left discrimination (Greenberg, 1991). A square is presented on the computer monitor once every two seconds and remains visible for 100 milliseconds. The square either has a hole near the top or near the bottom edge. The participant is instructed to press a microswitch every time there is a hole near the top of the square (target) and not to respond when there is a hole near the bottom of the square (nontarget). The entire test takes approximately 22 minutes to administer. During the first half (TOVA1) the target is presented randomly once for every three and one half nontarget (infrequent condition). During the second half (TOVA2) the ratio is reversed so that the target is presented three and one half times for every presentation of the nontarget (frequent condition).
The four primary variables that are scored from the TOVA are: errors of omission, which reflect inattention; errors of commission, which reflect impulsivity; mean correct response times, an indication of processing and response time; and variability, a measure of consistency derived from the standard deviations of the response times. Errors of omission and errors of commission are reported in terms of the total percentage of errors that were made while response time and variability are measured in milliseconds. Scores for each of these variables are available for each half of the test, although only the total test scores were analyzed in the present study.
The four TOVA variables and the psychometric variables (WISC and WIAT Reading, Spelling, and Math) were examined using a multivariate analysis of variance (MANOVA). For the purposes of this analysis, the WISC-III scores were broken down into the following sections: Full Scale IQ (FSIQ), Verbal Comprehension Index (VCI), Perceptual Organization Index (POI), Freedom from Distractibility Index (FDI), and Processing Speed Index (PSI).
For the purposes of the EEG recordings, the TOVA was divided into two equal parts: The first 11 minutes of the test (infrequent condition) were designated as TOVA1 while the remaining 11 minutes (frequent condition) were designated TOVA2. This division was necessitated by limitations of the computer system to process the large amount of EEG data collected during the administration of the TOVA. All analyses were conducted using absolute means, referred to as magnitude. Magnitude, which is defined as the voltage of a wave measured peak to peak, is reported in microvolts (uV).
A researcher, who was blind to group membership, visually inspected the EEG records and discarded any epochs containing muscle artifact. The remaining data were then divided into six frequency bands: Delta, 1-4 Hz; theta, 4-8 Hz; alpha, 8-12 Hz; sensory motor rhythm (SMR), 12-16 Hz; beta1, 16-20Hz; beta2, 20-24 Hz. As previous research indicates people with AD/HD show excessive slow wave activity and/or a decrease in fast-wave activity, our analyses were performed on the theta and beta1 frequency (Janzen et al., 1995; Mann et al., 1992).
Given the exploratory nature of the study, it was deemed important to consider all sites of the EEG signal and then use factor analysis to determine which sites were most predominant in explaining the observed variance. Principle component analyses with varimax rotation for the eyes open condition as well as TOVA1 and TOVA2 for both the theta and beta1 frequency bands indicated the first component accounted for large portions of the variance (i.e., between 82% and 87%). In all cases, this component included loadings on the frontal (F3, FZ, F4), and central sites (C3, CZ, and C4).
The EEG data were examined using a separate repeated measures analyses of covariance (ANCOVA) for the frontal and central sites with eyes open condition as the covariate for TOVA1 and TOVA2.
A multivariate analysis of variance on the mean ratings for the DSM-IV checklist and the BASC rating scale verified that participants were appropriately assigned to each of the three groups (approximate F (1,26) = 12.54, p <.0001). The means and standard deviations are presented in Table 1. Post hoc comparisons using the Bonferroni correction indicated that the AD/HD+ group was rated significantly higher than the AD/HD- and control groups on the nine hyperactivity/impulsivity items on the DSM-IV checklist, F = 94.74, p <.00005. There was no difference between the two AD/HD groups on the nine inattention items. The control group was rated significantly lower on the inattention items than the AD/HD+ group, F = 166.53, p <.00005, and the AD/HD- group, F = 177.27, p <.00005.
Post hoc analyses using the Bonferroni correction to compare the three groups on the BASC scores were also completed. The AD/HD+ group was rated significantly higher than the AD/HD-group on the Hyperactivity Scale, F = 9.44, p = .005, but not on the Attention Problems Scale, F = 0.54, p = .47. The only other significant difference between these two groups emerged on the Aggression Scale, with the AD/HD+ group scoring significantly higher in this regard, F = 5.59, p = .03. As expected, there were a number of significant differences between the clinical groups and the control group. In particular, the AD/HD+ participants were rated as being significantly more maladaptive than the control group on all of the 11 BASC Scales using an alpha level of .05. Similarly, the AD/HD-group obtained significantly higher ratings than the control group on the Hyperactivity, Depression, Somatization, Withdrawal, and Attention Problems Scales. This group was also rated lower on the Social Skills and Leadership Scales than was th e control group (the scoring for these two scales is reversed so that lower scores are indicative of maladaptive characteristics). All of these differences were significant at the .05 level.
A multivariate analysis of variance was computed on the mean standard scores for the WISC-III, WIAT, and TOVA variables. As expected, there were no significant differences between the three groups on the WISC-III and WIAT variables. Means are presented in Table 2. Although it did not reach statistical significance, there was a trend for the AD/HD groups to score higher than the control group on all TOVA variables. Notably, the mean variability score for the AD/HD+ group was 90 compared to a mean of 60 for the control group.
The EEG recordings for beta1 and theta bands were examined for differences in mean magnitude across groups at the frontal and central sites. There were no significant group differences for beta1 band at the frontal or central sites in either TOVA1 or TOVA2 conditions. Similarly, there were no group differences for theta band at the frontal sites in either TOVA1 or TOVA2 conditions. However, a repeated measures ANCOVA on theta band indicated a main effect for group on TOVA1, F (2,21) = 4.31, p <.03, at C3, CZ, and C4. Tukey's HSD comparisons of adjusted group means indicated the AD/ND- group had significantly higher mean magnitude than the control group, HSD = 1.38, p <.05. Although the difference in mean magnitude between the AD/HD+ group and the control group did not reach statistical significance, as Figure 1 illustrates, there was a trend for higher magnitude in the theta band for the two clinical groups. There was no significant difference between the AD/HD- and the AD/HD+ groups.
Although group differences for theta band at the central sites were only marginally significant for TOVA2 condition, F (2,2 1) = 2.78, p <.08, Figure 1B demonstrates a similar trend as in the TOVA1 condition. Tukey's post hoc comparisons indicated there were no differences in the mean magnitude between the two AD/HD groups, however, the AD/HD- group had a higher mean magnitude than the control group, HSD 2.14, p <.10, on this conservative test.
Behavior Rating Scales
As anticipated, given the DSM-IV checklist was used to classify the participants as AD/HD-, AD/HD+, or control, there are a number of significant differences between the scores each group obtained on this measure. The AD/HD+ group was rated as being highly inattentive and hyperactive/impulsive, whereas the AD/HD-group only scored high on the inattention items. In comparison, the children in the control group are rated as being relatively free from symptoms of either inattention or hyperactivity/impulsivity.
Comparisons between the two clinical groups on the BASC reveal more similarities than differences. The significantly higher scores for the AD/HD+ group on the Hyperactivity and Aggression scales is consistent with previous research (Dykman & Ackerman, 1991; Edelbrock, Costello, & Kessler, 1984), but none of the other comparisons were statistically significant. The AD/HD+ and AD/HD- groups were both rated as exhibiting similar levels of inattention, anxiety, and physical problems or complaints (somatization), and as possessing relatively limited leadership and social skills.
The AD/HD+ group received the highest mean rating on the Conduct Problems Scale, followed by the AD/HD- and the control groups. However, only the difference between the scores that were obtained by the AD/HD+ group and the control group was significant. It is possible that the sample size utilized in the present study was simply not large enough to replicate the findings of previous studies with respect to higher levels of antisocial behavior in AD/HD+ children as compared to those with AD/HD- (Hynd et al., 1991; King & Young, 1982). The AD/HD- group received a higher mean score than the AD/HD+ group on the Withdrawal scale, and while this is consistent with the findings of other researchers (Edelbrock et al., 1984; Lahey, Schaughency, Strauss, & Frame, 1984) this difference did not reach statistical significance. There was also a nonsignificant trend for the AD/HD+ group to be rated higher than the AD/HD- group with respect to the levels of depression and atypical behavior they display. While the Atypicality scale on the BASC is reported to be a measure of psychotic-type behavior (Reynolds & Kamphaus, 1992), it includes a variety of items that are commonly associated with inattentive and hyperactive behavior (e.g., daydreaming, stares blankly, rocking back and forth in a chair).
It is interesting to note that as a group the children in the AD/HD- group are rated significantly higher than controls on the BASC Hyperactivity scale but not on the hyperactivity/impulsivity items that are included on the DSM-IV checklist. An examination of the individual items that are included on the BASC Hyperactivity scale reveals that many of them do not reflect what is typically thought of as hyperactivity. For instance, the Hyperactivity scale on the BASC Parent Rating Scale for 12 to 18 year olds includes items that reflect the need for a high level of supervision, the presence of temper tantrums, and interrupting parents while talking on the telephone. Such items may well apply to children who do not exhibit hyperactive behavior as defined by DSM-IV criteria.
A critical issue affecting the validity of the present results is the method of identifying AD/HD+ and AD/HD- groups. Using the DSM-IV checklist to select participants appears to have resulted in meaningful differences between the three groups. Support for this conclusion comes from two sources. First, the mean ratings on the BASC Parent Rating Scale are as expected given the composition of the three groups. The AD/HD+ group scored higher on the Hyperactivity and Aggression scales than the AD/HD- group. Both the AD/HD+ and AD/HD- groups are rated as exhibiting significantly more emotional and behavior problems than the control group. Furthermore, the two AD/HD groups are comparable on each of the remaining BASC scales including Attention Problems. Thus, the primary difference between the two clinical groups according the BASC results is the level of hyperactivity and aggression, while they are similar with respect to problems of inattention. In contrast, the mean scores for the control group on all of the BAS C scales fell within the normal range.
The second source of support is found in the number of EEG epochs that it was necessary to discard due to the presence of muscle artifact (which reflects movement, muscle tension, or both). The technician was not aware of each child's group membership when completing the artifacting process. Consequently, the number of rejected EEG epochs can be considered an unbiased indication of the level of activity during the EEG recording session. The number of rejected EEG epochs for the AD/HD+ group during TOVAl is higher than for either the AD/HD-, F (2,16) = 7.38, p = .012, and the control group, F (2,16) = 15.31, p = .001. Similarly, the number of rejected EEG epochs for the AD/HD+ group during the TOVA2 condition is significantly higher than for the AD/HD-, F (2,16) = 7.26, p .0 13, and the control group, F (2,16) = 19.22, p <.0005. The difference between the AD/HD- and control groups is not significant for TOVAl or TOVA2. The higher EEG epoch rejection rate in the AD.HD+ group can likely be attributed to their pr oclivity to be more physically active and fidgety than either normal or AD/HD- children. Consistent with the observations made during the recording session, these results indicate that as a group the AD/HD+ participants are more restless during the collection of the EEG data than either the AD/HD- or the control participants.
As expected, there are no significant differences between the AD/HD- and AD/HD+ groups on any of the WISC-III scores that are considered, and this finding is consistent with the majority of the research that has examined this issue (Ackerman, Anhalt, Dykman, & Holcomb, 1986; Barkley, DuPaul, & McMurray, 1990; Barkley, Grodzinsky, & DuPaul, 1992; Berry et al., 1985; Frank & Ben-Nun, 1988; Hynd et al., 1991; Zagar, Arbit, Hughes, Busell, & Busch, 1989).
Similar findings emerged with respect to the academic test results. Consistent with most previous research (Barkley et al., 1990; Carlson et al., 1986: Dykman & Ackerman, 1991; Rubinstein & Brown, 1984; Zagar et al., 1989) there was no significant difference between groups. Once again this finding should be expected as all participants are required to achieve WIAT reading, spelling, and mathematical reasoning scores that are within one standard deviation of the mean or higher, effectively reducing the vanability in test scores and obscuring any differences that might otherwise exist.
Both clinical groups achieved TOVA Variability scores that are higher than the controls, but only the AD/HD+ group differed significantly from the control group. The Omission, Commission and Response Time variables are not significantly different for the three groups. However, examination of the mean T Scores for each group reveals a consistent trend for both the AD/HD- and the AD/HD+ groups to be more inattentive (as measured by the Omission scores) and impulsive (as measured by the Commission scores), to exhibit greater inconsistency in response times (as measured by the Variability scores), and to have greater response times than the control group. The utility of the TOVA in differentiating between normal and clinical participants is partially supported by the data. According to Greenberg (1991), the Variability score is the most sensitive of the TOVA scores with respect to the presence of AD/HD. This is the case in the present study, particularly with respect to the AD/HD+ group. It is reasonable to assum e that with a larger sample size and a commensurate increase in statistical power significant differences between the clinical groups and the control group may have emerged on other TOVA variables as well.
As there were no differences between the two clinical groups the present findings suggest that the TOVA was not useful in differentiating between AD/HD+ and AD/HD-. However, it should also be noted that in the present study the TOVA was administered in a rather unique context (i.e., while the participants were connected to the EEG equipment) and this may have affected the scores of each group. Consequently, it would be inappropriate to generalize these findings beyond the current study without further examination of this issue under more typical TOVA administration conditions.
As expected, there is a trend for both clinical groups to achieve higher adjusted mean magnitudes than the control group at the central sites on the theta band for both the TOVA1 and TOVA2 conditions. However, the lack of differences between the two clinical groups on the EEG variables is perhaps the most noteworthy finding in the present study. Although the paucity of significant findings may indeed be a consequence of low statistical power due to the small sample size, consistent with previous research comparing AD/HD populations to controls, the AD/HD- group showed greater theta magnitude when compared to the control group (Janzen et al., 1995; Mann et al., 1992).
In an attempt to minimize the number of statistical tests, factor analysis was preferred to identify which sites to analyze. Previous research suggests that children with AD/HD experience a deficit in the parietal areas of the brain (Janzen, 1992; Mann et al., 1992). Also, given the visual nature of the TOVA, one might expect group differences in the occipital sites. Post hoc analyses of P3, PZ, and P4 and P4 and O1 and O2, however, resulted in no significant group differences at these sites among any of the experimental groups. Although further research with large sample sizes might investigate possible differences in these areas, our results support previous research indicating increased theta activity at the central sites for children with AD/HD (Lubar & Lubar, 1999; Matsuura et al., 1993), and provide new evidence that EEG characteristics do not differentiate AD/HD- from AD/ND+.
Considering the substantial body of literature that suggests frontal lobe dysfunction in children with AD/HD (Barkley, 1998; Chelune, Ferguson, Koon, & Dickey, 1986; Crawford & Barabasz, 1996; Mattes, 1980) we were somewhat surprised that no significant group differences emerged at the frontal sites. It is probable that this is a reflection of the task children engaged in while EEG measures were being recorded. Given the TOVA is not a complex cognitive task, which would normally be attributed to frontal lobe activity, it is recommended that further research investigates differential EEG measures using a variety of tasks. As well, the frontal sites are those most susceptible to muscle artifact, hence, the artifacting process may have obscured related activity in this area.
The results of this exploratory study provide little support for the conceptualization of AD/HD without hyperactivity as a separate neurophysiological disorder from AD/HD with hyperactivity. There are no unexpected differences between these groups on the BASC, and while the AD/HD+ group did achieve a higher TOVA Variability score, this difference is not statistically significant. In addition, there are no apparent differences in EEG measures. If both AD/HD- and AD/AD+ children demonstrate similar EEG patterns characterized by the presence of excessive slow wave activity and higher mean magnitude when compared to normal children, we are still left with the question of what could account for the disparate behavioral characteristics.
General cognitive models of AD/HD have not attempted to explain the differential behaviors of subtypes of AD/HD. For example, Sergeant, Oosterlaan, and Van der Meere (1999) emphasized inadequate allocation of cognitive-energetic resources during the motor-output stage of information processing. These researchers proposed that the inhibition deficit (see Barkley, 1998) in AD/HD could be related to the energetics of the psychophysiological system. Douglas (1999) furthered this argument by suggesting that children with AD/HD not only experience motor-output disinhibition, but also deficits in stimulus processing at the encoding stage of information processing. Both approaches are consistent with the hypoarousal hypothesis of AD/HD suggesting that people with AD/HD are in a state of underarousal. However, these models do not explain differential behaviors of AD/HD with hyperactivity compared to AD/HD without hyperactivity. On the other hand, Zentall's (1975) (Zentall & Zentall, 1983) theory of optimal stimulation suggests activity can be seen as an attempt at providing self-stimulation in order to maintain an optimal level of arousal. This theory assumes that for all organisms there exists "a biologically determined optimal level of stimulation. When that level of stimulation is not present, activity can serve as a homeostatic regulator. That is, an organism will initiate stimulation-seeking activity when there is insufficient stimulation" (Zentall & Zentall, 1983, p. 447). As with most psychological and physiological factors, there are individual differences in the level of arousal that may be considered optimal. According to this theory, hyperactive children suffer from a chronic state of underarousal, and their behavior represents an attempt at providing self-stimulation in order to maintain an optimal level of arousal. Essentially their level of activity is thought to be "an attempt to increase insufficient stimulation, rather than being a consequence of overwhelming stimulation" (Zentall, 1975, p. 552). Unfortun ately, while the hyperactive behavior may have the desired effect of increasing the individual's level of arousal, it is often not conducive to learning and frequently interferes with the individual's ability to stay on task. From this theoretical framework, hyperactive behavior is not a consequence of an individual's physiological state, but a reaction to his or her physiological state. Similarly, distractibility may be conceptualized as an attempt to seek stimulation, while inattentiveness is a consequence of insufficient arousal necessary to sustain attention. The present findings and those of numerous previous EEG studies provide strong support for the theory of optimal stimulation as an explanation for hyperactive behavior.
The most obvious limitation of the present study is the small sample size. As well, EEG characteristics are to a large degree a function of the task the participants are engaged in at the time of the recording. As such, generalization from the present results could only be made once these results are replicated using a larger sample size and various tasks during the EEG recordings.
The study reported in this paper was carried out by the first author in partial fulfillment of the requirements for a Doctor of Philosophy, Educational Psycho1ogy, University of Alberta, Edmonton, Alberta.
Ackerman, P. T., Anhalt, J. M., Dykman, R. A., & Holcomb, P. J. (1986). Effortful processing deficits in children with reading and/or attention disorders. Brain and Cognition, 5,22-40.
American Psychiatric Association (1968). Diagnostic and statistical manual of mental disorders (2nd ed.) Washington, DC: American Psychiatric Association.
American Psychiatric Association (1980). Diagnostic and statistical manual of mental disorders (3rd ed.) New York: American Psychiatric Association.
American Psychiatric Association (1987). Diagnostic and statistical manual of mental disorders (3rd ed.) Revised. Washington, DC: American Psychiatric Association.
American Psychiatric Association (1994). Diagnostic and statistical manual of mental disorders (4th ed.) Washington, DC: American Psychiatric Association.
Barkley, R. A. (1990). Attention deficit hyperactivity disorder: A handbook for diagnosis and treatment New York: Guilford.
Barkley, R. A. (1998). Attention-deficit hyperactivity disorder: A handbook for diagnosis and treatment (2nd ed.) New York: Guilford Press.
Barkley, R. A., Grodzinsky, G., & DuPaul, G. P. (1992). Frontal lobe functions in attention deficit disorder with and without hyperactivity: A review and research report. Journal of Abnormal Child Psychology, 20, 163-188.
Berry, C. A., Shaywitz, S. E., & Shaywitz, B. A. (1985). Girls with attention deficit disorder: A silent minority? A report on behavioral and cognitive characteristics. Pediatrics, 76, 801-809.
Bloomingdale, L. M., & Bloomingdale, E. C. (1980). Minimal brain dysfunction: A new screening test and theoretical considerations. The Psychiatric Journal of the University of Ottawa, 5, 295-306.
Callaway, E., Halliday, R., & Naylor, H. (1983). Hyperactive children's event-related potentials fail to support underarousal and maturational lag theories. Archives of General Psychiatry, 40, 1243-1248.
Carlson, C. L. (1986). Attention deficit disorder without hyperactivity: A review of preliminary experimental evidence. In B. B. Lahey & A. E. Kazdin (Eds.), Advances in Clinical Child Psychology, Vol. 9 (pp. 153-175). New York: Plenum.
Castellanos, F. X. (1999). The psychobiology of attention-deficit/hyperactivity disorder. In H. C. Quay & A. E. Hogan (Eds.), Handbook of disruptive behavior disorders (pp. 179-198). New York: Kluwer Academic/Plenum Publishers.
Chelune, G. J., Ferguson, W., Koon, R., & Dickey, T. O. (1986). Frontal lobe disinhibition in attention deficit disorder. Child Psychiatry & Human Development, 16, 221-234.
Crawford, H. J., & Barabasz, M. (1996). Quantitative EEG magnitudes in children with and without attention deficit disorder during neurological screening and cognitive tasks. Child Study Journal, 26, 71-85.
Douglas, V. I. (1999). Cognitive control processes in attention-deficit/hyperactivity disorder. In H. C. Quay & A. E. Hogan (Eds.), Handbook of disruptive behavior disorders (pp. 105-138). New York: Kluwer Academic/Plenum Publishers.
Dykman, R. A., & Ackerman, P. T. (1991). Attention deficit disorder and specific reading disability: Separate but often overlapping disorders. Journal of Learning Disabilities, 24, 96-103.
Dykman, R. A., & Ackerman, P. T. (1993). Behavioral subtypes of attention deficit disorder. Exceptional Children, 60, 132-141.
Dykman, R. A., Holcomb, P. J., Oglesby, D. M., & Ackerman, P. T. (1982). Electrocortical frequencies in hyperactive, learning-disabled, mixed, and normal children. Biological Psychiatry, 17, 675-685.
Edelbrock, C., Costello, A. J., & Kessler, M. D. (1984). Empirical corroboration of attention deficit disorder. Journal of the American Academy of Child Psychiatry, 23, 285-290.
Fein, G., Galin, D., Yingling, C. D., Johnstone, J., & Nelson, M. A. (1984). EEG spectra in 9-13 year old boys are stable over 1-3 years. Electroencephalography and Clinical Neurophysiology, 58, 517-518.
Frank, Y., & Ben-Nun, Y. (1988). Toward a clinical subgrouping of hyperactive and nonhyperactive attention deficit disorder. American Journal of Diseases of Children, 142, 153-155.
Greenberg, L. (1991). TOVA Interpretation Manual: Test of Variables of Attention Computer Program, Versions 5.01, 5.0A. Minneapolis: University of Minnesota.
Harper, H. K., Deering, C. R., Cavernos-Gonzales, C., McNeil, C., & Ulam, F. A. (1996). Quantitative EEG relative power comparisons between ADHD children and matched controls: A replication (Abstract). National Academy of Neuropsychology: Abstracts for the fifteenth annual meeting, San Francisco California, November 1-4, 1995. Archives of Clinical Neuropsychology, 11, 398.
Hynd, G. W., Lorys, A. R., Semrud-Clikeman, M., Nieves, N., Huettner, M., & Lahey, B. B. (1991). Attention deficit disorder with and without hyperactivity: A distinct behavioral and neurocognitive syndrome. Journal of Child Neurology, 6, S37-S43.
Janzen, T. (1992). Comparing the dynamic EEG of successful and attention deficit disordered grade 5 and 6 boys. Unpublished master's thesis, University of Alberta, Edmonton, Alberta.
Janzen, T., Graap, K., Stephanson, S., Marshall, W., & Fitzsimmons, G. (1995). Differences in baseline EEG measures for ADD and normally achieving preadolescent males. Biofeedback and Self-Regulation, 20, 65-82.
King, C., & Young, R. D. (1982). Attentional deficits with and without hyperactivity: Teacher and peer perceptions. Journal of Abnormal Child Psychology, 10, 483-496.
Klove, H. (1989). The hypoarousal hypothesis: What is the evidence? In T. Sagvolden & T. Archer (Eds.), Attentian deficit disorder: Clinical and basic research (pp. 131-136). Hillsdale, NJ: Lawrence Erlbaum Associates.
Lahey, B. B., Schaughency, E. A., Strauss, C. C., & Frame, C. L. (1984). Are attention deficit disorders with and without hyperactivity similar or dissimilar disorders? Journal of the American Academy of Child Psychiatry, 23, 302-309.
Lexicor Medical Technology (1990). Users Manual Neurosearch, 24. Boulder, CO: Lexicor Medical Technology, Inc.
Lubar, J. F., & Lubar, J. 0. (1999). Neurofeedback assessment and treatment for attention deficit/hyperactivity disorders. In J. R. Evans, & A. Abarbanel (Eds.), Introduction to quantitative EEG and neurofeedback (pp. 103-143). San Diego, CA: Academic Press.
Mann, C. A., Lubar, J. F., Zimmerman, A. W., Miller, C. A., & Muenchen, R. A. (1992). Quantitative analysis of EEG in boys with attention-deficit-hyperactivity disorder: Controlled study with clinical implications. Pediatric Neurology, 8, 30-36.
Matsuura, M., Okubo, Y., Toru, M., Kojima, T., He, Y., Yacun, S., & Chung, K. L. (1993). A cross-national EEG study of children with emotional and behavioral problems: A WHO collaborative study in the western pacific region. Biological Psychiatry, 34, 59-65.
Mattes, J. A. (1980). The role of frontal lobe dysfunction in childhood hyperkinesis. Comprehensive Psychiatry, 21, 358-369.
Reynolds, C. R., & Kamphaus, R. W. (1992). Behavior Assessment System for Children. Circle Pines, MN: American Guidance Service.
Rubinstein, R. A., & Brown, R. T. (1984). An evaluation of the validity of the diagnostic category of attention deficit disorder. American Journal of Orthapsychiatry, 54, 388-414.
Sergeant, J. A., Oosterlaan, J., & Van der Meere, J. (1999). Information processing and energetic factors in attention-deficit/hyperactivity disorder. In H. C. Quay & A. E. Hogan (Eds.), Handbook of disruptive behavior disorders (pp. 75-104). New York: Kluwer Academic/Plenum Publishers.
Stewart, G. A. (1994). External validity of attention deficit disorder without hyperactivity: A review of the evidence. Canadian Journal of School Psychalogy, 10, 15-25.
Wechsler, D. (1991). Wechsler Intelligence Scale for Children (WISC-III), 3rd ed. San Antonio, TX: The Psychological Corporation, Harcourt Brace.
Wechsler, D. (1992). Wechsler Individual Achievement Test (WIAT). San Antonio, TX: The Psychological Corporation, Harcourt Brace.
Zagar, R., Arbit, J., Hughes, J. R., Busell, R. E., & Busch, K. (1989). Developmental and disruptive behavior disorders among delinquents. Journal of the American Academy of Child and Adolescent Psychiatry, 28,437-440.
Zentall, S. (1975). Optimal stimulation as a theoretical basis of hyperactivity. American Journal of Orthapsychiatry, 45, 549-563.
Zentall, S. S., & Zentall, T. R. (1983). Optimal stimulation: A model of disordered activity and performance in normal and deviant children. Psychalogical Bulletin, 94,446-471.
[Figure 1 Omitted]
TABLE 1 Behavior Rating Scale Means (and Standard Deviations) Diagnostic and Statistical Manual Of Mental Disorders - 4th edition Group DSM-IV Checklist AD/HD- AD/HD+ Inattention 8.33 (a) (0.71) 8.11 (a) (1.27) Hyperactivity/Impulsivity 1.44 (a) (1.59) 8.11 (b) (1.17) Group DSM-IV Checklist Control Inattention 1.11 (b) (1.36) Hyperactivity/Impulsivity 0.78 (a) (1.56) Behavior Assessment System for Children (BASC) Group DSM-IV Checklist AD/HD- AD/HD+ Hyperactivity 57.56 (a) (11.04) 72.22 (b) (10.58) Aggression 54.89 (a) (14.56) 69.22 (b) (15.31) Conduct Problems 58.22 (15.31) 68.11 (a) (11.40) Anxiety 56.44 (13.31) 58.44 (a) (11.83) Depression 66.00 (a) (13.06) 76.00 (a) (20.31) Somatization 58.11 (a) (8.72) 54.89 (a) (11 88) Atypicality 58.78 (7.17) 69.22 (a) (21.95) Withdrawal 62.67 (a) (15.29) 54.67 (a) (9.85) Attention Problems 71.89 (a) (9.28) 68.56 (a) (8.16) Social Skills 42.33 (a) (8.14) 43.33 (a) (8.94) Leadership 40.77 (a) (8.80) 45.44 (a) (5.29) Group DSM-IV Checklist Control Hyperactivity 43.44 (c) (8.59) Aggression 48.89 (a) (7.03) Conduct Problems 47.67 (b) (6.72) Anxiety 47.89 (b) (4.70) Depression 45.89 (b) (5.28) Somatization 44.56 (b) (9.34) Atypicality 47.44 (b) (7.37) Withdrawal 43.00 (b) (6.27) Attention Problems 46.56 (b) (11.22) Social Skills 54.33 (b) (6.89) Leadership 54.33 (b) (5.96) Note. DSM-IV Checklist means are reported as raw scores, BASC means as TScores. The Adaptability Scale from the 6 to 11 year old form of the BASC was excluded from this analysis because it is not included on the 12 to 18 year old version of the BASC; n = 9 for each group. Means with different superscripts are significantly different (p<.05, df = 2,24). TABLE 2 Mean Wechsler Intelligence Scale fo for Children (WISC), Wechsler Individual Achievement Test (WIAT) and Test of Variables of Attention (TOVA) (and Standard Deviations) Group AD/HD- AD/HD+ WTSC Standard Scares Full Scale Intelligence Quotient 104.89 (9.10) 108.89 (12.33) Verbal Comprehension Index 104.67 (6.06) 110.44 (15.21) Perceptual Organization Index 106.22 (11.36) 111.56 (13.78) Freedom from Distractibility Index 97.33 (7.68) 99.67 (12.45) Processing Speed Index 97.56 (13.80) 103.00 (12.63) WIAT Standard Scores Reading 105.78 (11.95) 104.11 (10.45) Spelling 98.11 (8.05) 98.11 (8.92) Mathematics Reasoning 103.22 (11.67) 101.78 (7.87) TOVA T Scores Omission 62.22 (17.83) 58.11 (4.59) Commission 57.00 (12.27) 61.89 (13.26) Response Time 71.89 (16.00) 69.78 (18.16) Variability 76.33 (17.29) 90.44 (27.73) Group Control WTSC Standard Scares Full Scale Intelligence Quotient 109.89 (10.51) Verbal Comprehension Index 106.33 (11.74) Perceptual Organization Index 110.67 (10.79) Freedom from Distractibility Index 102.78 (5.56) Processing Speed Index 112.56 (11.20) WIAT Standard Scores Reading 104.89 (9.63) Spelling 101.89 (10.89) Mathematics Reasoning 107.89 (7.27) TOVA T Scores Omission 52.22 (5.07) Commission 51.44 (10.46) Response Time 58.78 (11.71) Variability 60.33 (4.58)…