Access the full text.
Sign up today, get DeepDyve free for 14 days.
Loading next page...
/lp/springer-journals/response-inhibition-deficits-in-externalizing-child-psychiatric-YtFcTT0IAf
- Publisher
- Springer Journals
- Copyright
- Copyright © 2005 by Albrecht et al; licensee BioMed Central Ltd.
- Subject
- Biomedicine; Neurosciences; Neurology; Behavioral Therapy; Psychiatry
- eISSN
- 1744-9081
- DOI
- 10.1186/1744-9081-1-22
- pmid
- 16336676
- Publisher site
- See Article on Publisher Site
Abstract
Background: Evidence from behavioural studies suggests that impaired motor response inhibition may be common to several externalizing child psychiatric disorders, although it has been proposed to be the core-deficit in AD/HD. Since similar overt behaviour may be accompanied by different covert brain activity, the aim of this study was to investigate both brain-electric-activity and performance measures in three groups of children with externalizing child psychiatric disorders and a group of normal controls. Methods: A Stop-task was used to measure specific aspects of response inhibition in 10 children with attention-deficit hyperactivity disorder (AD/HD), 8 children with oppositional defiant disorder/conduct disorder (ODD/CD), 11 children with comorbid AD/HD+ODD/CD and 11 normal controls. All children were between 8 and 14 years old. Event-related potentials and behavioural responses were recorded. An initial go-signal related microstate, a subsequent Stop-signal related N200, and performance measures were analyzed using ANCOVA with age as covariate. Results: Groups did not differ in accuracy or reaction time to the Go-stimuli. However, all clinical groups displayed reduced map strength in a microstate related to initial processing of the Go-stimulus compared to normal controls, whereas topography did not differ. Concerning motor response inhibition, the AD/HD-only and the ODD/CD-only groups displayed slower Stop-signal reaction times (SSRT) and Stop-failure reaction time compared to normal controls. In children with comorbid AD/HD+ODD/CD, Stop-failure reaction-time was longer than in controls, but their SSRT was not slowed. Moreover, SSRT in AD/HD+ODD/CD was faster than in AD/HD-only or ODD/CD-only. The AD/HD-only and ODD/CD-only groups displayed reduced Stop-N200 mean amplitude over right-frontal electrodes. This effect reached only a trend for comorbid AD/HD+ODD/CD. Conclusion: Following similar attenuations in initial processing of the Go-signal in all clinical groups compared to controls, distinct Stop-signal related deficits became evident in the clinical groups. Both children with AD/HD and ODD/ CD showed deficits in behavioural response-inhibition accompanied by decreased central conflict signalling or inhibition processes. Neither behavioural nor neural markers of inhibitory deficits as found in AD/HD-only and ODD/CD-only were additive. Instead, children with comorbid AD/HD+ODD/CD showed similar or even less prominent inhibition deficits than the other clinical groups. Hence, the AD/HD+ODD/CD-group may represent a separate clinical entity. Page 1 of 14 (page number not for citation purposes) Behavioral and Brain Functions 2005, 1:22 http://www.behavioralandbrainfunctions.com/content/1/1/22 ing, disinhibition and attention deficits, and thereby to Background Attention-deficit hyperactivity disorder (AD/HD) is char- antisocial behaviour. acterised by symptoms of severe inattention, overactivity and impulsiveness. With its prevalence of 3–5% in The ,Stop-signal paradigm'[25] allows investigating well school-age-children, AD/HD is one of the most common defined response inhibition processes directly. Generally, disorders in child and adolescent psychiatry [1]. Accord- the subjects perform a simple or a two choice reaction ing to Barkley's theory of AD/HD [2,3], deficient behav- task. In some of the trials, a Stop-signal follows the go- ioural inhibition is the core deficit of the disorder, and stimulus at a given delay and requires the inhibition of the may lead to impairments of executive functions. Behav- ongoing response. The longer the Stop-signal-delay ioural inhibition may be separated into three interrelated (SSD), the more difficult it becomes to inhibit the processes called 'inhibition of the initial prepotent response. The „horse race“ model of the Stop-task, which response to an event', 'stopping of an ongoing response' assumes a race between the reaction to the primary task and 'interference control'. and the reaction to the Stop signal, further allows to esti- mate the "virtual" reaction time to the Stop-signal (SSRT) Several behavioural studies reported deficits of response- as a measure for response inhibition performance [25]. In inhibition in children with AD/HD ([4-8]; for a review see a meta-analysis of the Stop-task, Oosterlaan et al. [26] [9]). However, impaired behavioural response inhibition reported that behavioural studies showed consistently is also observed in children with other disruptive disor- slower SSRT for children with ADHD, but also for chil- ders such as ODD/CD [9], which is the most prevalent dren with CD compared to controls, Comparisons comorbidity of AD/HD and poses significant additional between AD/HD and CD as well as between AD/HD+CD clinical and public health problems. In addition, further and AD/HD revealed no differences. However, inferences deficits which are not likely to result from deficient inhi- based on performance data only may have limited valid- bition are present in children with AD/HD, as evident ity, because differences in covert brain mechanisms may from their poor performance in a variety of executive func- lead to similar overt performance [21,27]. tions tasks such as the Continuous Performance Test (CPT) [10,11], Wisconsin Card-Sorting-Task [12-14], A more direct access to brain functions is provided by Tower-of-Hanoi [13,14] and Stroop-Test [12,15]; for a non-invasive methods such as functional magnetic reso- review see [16,17]. nance imaging (fMRI) [28] or event related potentials (ERP) [29]. Briefly, in the blood-oxygenation-level- In a more neurophysiologically oriented theory covering dependent (BOLD) fMRI, changes in cerebral blood-flow both ADHD and ODD/CD, Quay [18,19] following Gray and metabolism related to neuronal activation are meas- [20] argued that the behavioural activation system (BAS, ured with high spatial but low temporal resolution reflect- sensitive to reward) and the behavioural inhibition sys- ing the underlying hemodynamic process. ERPs are tem (BIS, sensitive to punishment) may reflect distinct voltage topographies and fluctuations recorded on the pathways for inhibition deficits. Children with AD/HD scalp which reflect neural activation to an event such as may suffer from an underactive BIS while their BAS seems the presentation of a stimulus or a response. A major to be unimpaired, whereas children with ODD/CD advantage of the ERP technique is the high temporal reso- should have an overactive BAS that dominates their lution in the range of milliseconds which allows to meas- (unimpaired) BIS. Therefore, according to Quay's theory ure brain-electrical correlates of information-processing both AD/HD and ODD/CD groups should display deficits in realtime. A number of studies therefore used electro- in inhibition, but for very different reasons. If comorbid physiological or fMRI measures of response inhibition AD/HD+ODD/CD is an additive combination of AD/HD processes in AD/HD [7,27,30-32]. and ODD/CD, this group should display the worst impairment in response inhibition because an overactive An ERP-study of Brandeis et al. [27] revealed that in AD/ BAS may be combined with a weak BIS. Concerning HD children, successful Stops differed from Stop-failures response control, results from a recent neurophysiological with topographic alterations in a microstate which study with the CPT-task are consistent with this predic- reflected mainly processing of the go-stimulus, whereas tion, and indicate that such deficits are indeed particularly normal controls differed at a slightly later stage of process- pronounced in this comorbid group [21]. Deficits in exec- ing with increased global-field-power (GFP, the spatial utive functioning in general, and inhibition deficits in standard deviation of voltages) in Stop-failures compared particular are also explained by other neurophysiological to correct Stops. Rubia et al. [30] reported that during theories focusing on either AD/HD or ODD/CD. For their fMRI-study, decreased right-inferior-prefrontal acti- ODD/CD [22-24], it has been argued that deficits of the vation in AD/HD occurred solely in the Stop-task, and prefrontal cortex leads to reduced orienting and arousal, thus hypothesized the "brake system of the brain" [30] to both of which predispose individuals to stimulation-seek- be located right-prefrontal. Pliszka et al. [7] reported for Page 2 of 14 (page number not for citation purposes) Table 1: Performance Data Group Controls AD/HD AD/HD+ODD/CD ODD/CD ANCOVA (N) N = 11 (A) N = 10 (AO) N = 11 (O) N = 8 (covariate "age") Measure Mean (SD) Mean (SD) Mean (SD) Mean (SD) F p Planned contrasts (3,35) Go-reaction-time (ms) 598 (81.6) 583 (46.2) 594 (52.7) 649 (109.9) 1.97 .14 SD of Go-reaction- time 161 (43.2) 157.6 (28.4) 155.5 (25.6) 188.5 (49.7) 1.76 .17 Percentage of correct Go-trials 88.4 (.08) 82.8 (.09) 79.9 (.11) 84.0 (.08) 1.43 .25 Stop-failure reaction-time (ms) 450 (37.6) 492 (50.5) 502 (50.2) 477 (57.5) 3.70* .02 N < A*, AO*, O*/A = AO = O SSRT at 250 ms SSD (ms) 245 (33.9) 272 (47.4) 256 (53.1) 274 (49.8) 3.41* .03 N < A*, O*/N = AO/AO < A*, O* Inhibition-function 100 ms SSD 3.9 (5.5) 11.0 (9.9) 9.1 (9.9) 5.8 (5.0) Group: F = 1.50, p = .21 (3,35) (percentage of Stop failures) 250 ms SSD 30.0 (11.1) 40.3 (16.1) 29.9 (9.6) 27.9 (13.2) 700 ms SSD 88.8 (12.0) 89.8 (7.0) 90.8 (6.3) 84.3 (12.5) Group*SSD: F = 1.61, ε = .95, p = .16 (6,70) * one-tailed, p < .05 Behavioral and Brain Functions 2005, 1:22 http://www.behavioralandbrainfunctions.com/content/1/1/22 Page 3 of 14 (page number not for citation purposes) Behavioral and Brain Functions 2005, 1:22 http://www.behavioralandbrainfunctions.com/content/1/1/22 Table 2: Analyses of Microstates Microstate I II III IV V VI Correct Go: GFP 1.25 4.53* C>A*, AO*, O* .88 .88 1.67 b + Correct Go: Topography 1.21 1.15 .66 1.29 1.63 + + Successful Stop: GFP .74 4.28* C>A*, AO*, O* .70 2.33 C>AO , O* .51 .83 Successful Stop: Topography .62 1.06 1.16 1.33 1.18 1.52 * p < .05, for comparisons: one-tailed p < .10, for comparisons: one-tailed F , covariate "age" (3,35) multivariate Pillai's-trace F , covariate "age" (12,102) normal controls a negative wave 200 ms after onset of the erally larger to failed than to successful Stops. Overtoom Stop-signal (Stop-N200) over right inferior frontal elec- et al. [31] found slower SSRT and decreased inhibition trodes which was reduced in ADHD-children. For both performance for AD/HD compared to normal controls. groups, this N200 after successful inhibitions was posi- Interestingly the study showed no N200-effects to the tively correlated with inhibition performance whereas Stop-signal. This could be due to the use of an auditory correlations for Stop-N200 to Stop-failures were not that Stop-signal, as Falkenstein et al. [34] found a Nogo-N2 clear. Following Kok [33], the N200 to the Stop-signal which was smaller for auditory compared to visual stimuli could either reflect a 'red flag' or a subsequent "(action-) despite similar performance in both modalities which inhibitory process, emanating from structures in the pre- could indicate that inhibition is related to a pre-motor frontal cortex" [33]. A second finding was that at right- level. frontal electrode-sites 250–500 ms post Go-signal-onset the control-group displayed greater positivity to failed There is an ongoing debate whether the Nogo-N200 than successful Stop-trials whereas in the ADHD-group reflects inhibitory processes per se [33-37], or conflict successful trials did not differ from failed ones. This pre- monitoring [38-40] which may initiate inhibition. We did paratory activity in failed Stop-trials was more positive in not intend to distinguish between these two models. Both controls than in ADHD patients. Further, Dimoska et al. of them predict that the Stop-N200 is related to inhibition [32] found, despite worse Go-task- and inhibition-per- performance: while the inhibition theory relates dimin- formance in AD/HD compared to controls, different acti- ished Stop-N200 amplitudes directly to an impaired cen- vation-patterns at an early stage of processing the Stop- tral inhibition mechanism, the conflict-signal theory signal. Again, a decreased N200 to the Stop-signal of suc- suggests that impaired triggering of the inhibitory mecha- cessful Stops for AD/HD was found, whereas groups did nisms is responsible. not differ concerning Stop-N2 of Stop-failures. Following Pliszka et al. [7], the authors argued that this N200 would Taken together, studies strongly suggest difficulties in reflect activation of inhibitory processes. However, in con- response inhibition paralleled by neurophysiological trast to Pliszka et al. their auditory evoked N200 was gen- deviances for children with AD/HD compared to normal Table 3: Electrophysiological Data Group Controls AD/HD AD/HD+ODD/ ODD/CD ANCOVA (N) N = 11 (A) N = 10 CD (AO) N = 11 (O) N = 8 (covariate "age") Measure Mean (SD) Mean (SD) Mean (SD) Mean (SD) F p Planned contrasts (3,35) Go-Trial ROI mean -3.20 (1.68) -2.64 (2.10) -2.25 (2.36) -1.98 (2.18) .80 .50 amplitude (µV) Stop-Trial ROI mean -5.36 (1.69) -1.94 (4.00) -2.89 (2.21) -1.90 (4.20) 3.15 .04* N<A*, AO*, O*/A = AO = O amplitude (µV) a + + Stop-N200 ROI mean -2.16 (1.60) .69 (3.43) -.63 (1.76) .07 (2.02) 2.54 .07 N<A*, AO , O*/A = AO = O amplitude (µV) * one-tailed, p < .05 one-tailed, p < .10 Region of interest, mean of electrodes F4 and F8 at 420–500 ms post Go-signal onset Page 4 of 14 (page number not for citation purposes) Behavioral and Brain Functions 2005, 1:22 http://www.behavioralandbrainfunctions.com/content/1/1/22 controls, but to our knowledge there is no such evidence .4). The fourth microstate, related to the Stop-N200, for ODD/CD and comorbid AD/HD+ODD/CD. Thus, the revealed only an overall trend towards group-differences in GFP (F = 2.33, p < .1) with ODD/CD lower than aim of this study was threefold, as we intended (1) to rep- (3,35) licate the neurophysiological finding of Brandeis et al. controls; groups did not differ in topography (Pillai-Spur [27] and of Pliszka et al. [7] concerning both early pre- F = 1.33, p > .2). (12,102) Stop-signal processing and the later Stop-N200-differ- Stop-N200 ences between controls and children with AD/HD; (2) to clarify whether children with ODD/CD and especially In the frontal region of interest, no main-effect of "condi- those with comorbid AD/HD+ODD/CD also display an tion" (F = 1.1, p > .3), but a trend for an interaction- (1,35) inhibitory-deficit as hypothesized according to Quay's effect "condition*group" (F = 2.5, p = .07) was found (3,35) model, i.e. a slower SSRT and slower Stop-failure reaction- at the given time window 170–250 ms post Stop-signal- times paralleled at the neuronal level by a reduced Stop- onset. Separate ANCOVAs for both levels of the "condi- N200-amplitude; and (3) we wanted to test whether an tion"-factor revealed that there were no amplitude differ- additive model of AD/HD and ODD/CD explains ences between the groups for correct Go-trials (F = (3,35) response-inhibition performance of children with comor- .80, p = .50), but significant differences of mean ampli- bid AD/HD+ODD/CD. tude in ROI for successful Stop-trials (F = 3.15, p < (3,35) .04). These differences were reflected by increased negativ- Results ity in controls compared to all clinical groups which did Behavioural data not differ among themselves (see Table 3 and Figures 1, 2, The groups did not differ in terms of correct Go-reaction- and 3). times (F = 1.97, p > .13), standard deviation of Go- (3,35) reaction-time (F = 1.79, p > .17), or accuracy as In order to clarify the interaction "condition*group" (3,35) reflected by percentage of correct Go-trials (F = 1.43, which reflects the Stop-N200, planned comparisons of (3,35) p > .25, Table 1). A significant partial-correlation between the difference between mean amplitude of successful Stop IQ and percentage of correct go-trials was found (r = and correct Go-trials were computed (Figure 4). The (dif- part .45, p < .01). There were also no differences between inhi- ference-) Stop-N200 was increased for normal controls = 1.60, p > .20) and bition-functions (group (F compared to pure AD/HD and ODD/CD, but there was (3,35) group*SSD (F = 1.61, ε = .95, p > .16)). just a trend for increased negativity in controls compared (6,70) to comorbid AD/HD+ODD/CD. Again, clinical groups However, groups differed in their Stop-failure-reaction- did not differ. The Stop-N200 analysed with the 2*2 times (F = 3.70, p = .02) with control children being ANCOVA revealed no main effects AD/HD or ODD/CD (3,35) faster than all clinical groups; no differences were found (F(1,35) = 2.08, p = .16 and F(1,35) = .39, p = .54, respec- among the clinical groups. Stop-failure-reaction-time was tively) but again an interaction AD/HD*ODD/CD correlated with IQ (r = .43, p < .01). There were also (F(1,35) = 4.63, p < .04). For the total sample, this Stop- part group-differences in SSRT (F = 3.41, p > .03) with N200 correlated positively with the speed of the inhibi- (3,35) slower SSRT for the pure AD/HD and ODD/CD groups tion process (rpart = .31, p < .05). compared to controls, but not for the comorbid AD/ HD+ODD/CD which displayed faster SSRT than AD/HD Discussion The Stop-task was used to investigate inhibitory response and ODD/CD. In the 2*2 ANCOVA-design, there were no main effects for AD/HD (F = .14, p > .71) or ODD/CD control in children with AD/HD, ODD/CD and comorbid (1,35) (F = .04, p > .85) on SSRT; but an interaction-effect AD/HD+ODD/CD in comparison to normal controls. (1,35) AD/HD*ODD/CD (F = 10.21, p < .01). While processing the Go-signal, all clinical groups dis- (1,35) played reduced map strength in a microstate attributable Brainmapping to initial orienting, consistent with previous work [7,27]. For correct Go-trials, only the second microstate 200–272 A novel finding was that this Go-signal related reduction ms post go-signal-onset revealed group-differences in GFP occurred on both correct Go-trials and successful Stops (F = 4.53, p < .01) with lower values for all clinical rather than just on Stop-failures, indicating a more gen- (3,35) groups compared to controls (see Table 2). No differences eral deficit than reported in previous work. Moreover, in topography were found (Pillai-Spur F = 1.15, p > these earlier studies had reported a different topography (12,102) .33). of brain electrical activity with frontal positivity whereas in this work particularly in controls frontal negativity In successful Stops, groups again differed in the second emerged. One explanation may be that participants in our microstate in GFP (F = 4.28, p = .01) with higher GFP sample showed less Stop-failures than for instance partic- (3,35) for controls compared to all clinical groups whereas ipants of Brandeis et al. [27] did: In this study, percentages topography did not differ (Pillai-Spur F = 1.06, p > of Stop-failures were 30% for controls and 40% for chil- (12,102) Page 5 of 14 (page number not for citation purposes) Behavioral and Brain Functions 2005, 1:22 http://www.behavioralandbrainfunctions.com/content/1/1/22 'clumsiness' or 'poor motor control' seems not to be valid -1 since there was a lack of group-differences in other per- formance measures not related to behavioural inhibition, such as go-reaction-time or accuracy. -2 The lack of differences between inhibition functions could be due to the adaptive instructions. These were used -3 in this version of the Stop-task to prevent extreme speed- accuracy-tradeoffs at the fixed medium stop signal delay. Such fixed Stop signal delays are advantageous for ERP Group studies, but suboptimal for deriving inhibition functions -4 [41]. Still, we note that it is crucial for any Stop-task to bal- AD/HD ance between the strategies avoiding every Stop-failure or AD/HD+ODD/CD responding so fast that no stopping is possible either -5 implicitly (with standard instructions like "respond as ODD/CD quickly and as accurately as possible to the primary stim- ulus, as well as to inhibit the response on the appearance -6 Controls of the Stop signal" [8]) or explicitly as is done here. The correct Go successful Stop widely reported finding that children with AD/HD display more variable reaction-times could not be replicated here, Condition maybe also because of the adaptive instructions. Mean A Figure 1 mplitudes in the region of interest Inhibition deficits were not limited to children with AD/ Mean Amplitudes in the region of interest. Mean HD, but also characterized children with ODD/CD, as amplitudes in the ROI for correct Go-trials and successful Stops. Normal controls (black), AD/HD (red), AD/ predicted by the model of Quay. Their Stop-N200 was HD+ODD/CD (green) and ODD/CD (blue). also reduced compared to normal controls, and did not differ from AD/HD and comorbid AD/HD+ODD/CD. The latter finding extends the commonality between AD/ dren with AD/HD but 48% and 51% respectively in Bran- HD and ODD/CD to the neurophysiological level, which deis et al. [27]. This could have invoked the same is in contrast to Quay's theory of conduct disorder postu- inhibitory or conflict monitoring mechanism as reflected lating an intact behavioural inhibition system. later on by the Stop-N200. Surprisingly, children with comorbid AD/HD+ODD/CD Normal control children displayed a right anterior nega- tended to be somewhat less impaired than the other clin- tivity to the Stop-signal which could reflect response inhi- ical groups. Their inhibition process (as reflected by SSRT) bition processes in the right prefrontal cortex [7,32,33], or was not significantly slower than in normal controls, and the mechanism triggering such an inhibitory process. was even faster than in the other clinical groups. However Although deciding between these alternatives is beyond their Stop-failure reaction-times were slower compared to the scope of this study, the right-frontal topography of this normal controls and similar to that of the other clinical Stop-N200 [7] slightly favors the inhibition explanation, groups. Hence, inhibition performance was by no means and differs from that of the "conflict" Nogo-N200 at most impaired in comorbid AD/HD+ODD/CD. Although fronto-central electrodes [21,39,40]. there was only a trend for decreased Stop-N200 mean amplitude compared to normal controls, no differences Not only the Stop-N200 effect, but also its attenuation in were found compared to the pure groups, which again AD/HD children as first described by Pliszka et al. [7] stands in contrast to Quay's theory. could be replicated. It can not be attributed to differences in processing the primary-task at that stage, because Consistent with this pattern, the 2*2 ANCOVA with group-amplitudes did not differ in this region of interest between subject factors "AD/HD" and "ODD/CD" in Go-trials. Along with this, children with AD/HD per- revealed no main- but strong interaction-effects for the formed poorer than normal controls in behavioural most important measures of inhibition, indicating that response inhibition. Their Stop-signal reaction-times and effects of AD/HD and ODD/CD symptoms on response their reaction-times in Stop-failures were considerably inhibition were not additive but sub-additive. This sup- slower than those of normal controls, indicating an even ports the conclusion of Banaschewski et al. using the CPT slower inhibitory process, consistent with the majority of [42] who argue against the view that comorbid AD/ previous work [8,9]. Alternative explanations such as HD+ODD/CD is a hybrid or a phenocopy of AD/HD or Page 6 of 14 (page number not for citation purposes) Mean amplitude in the ROI (micro V) Behavioral and Brain Functions 2005, 1:22 http://www.behavioralandbrainfunctions.com/content/1/1/22 ROI – successful Stop ROI – correct Go [µV] [µV] Stop-signal onset -8 -8 -6 -6 -4 -4 -2 -2 0 0 4 4 8 8 0 100 200 300 400 500 600 700 800 900 [ms] 0 100 200 300 400 500 600 700 800 900 [ms] ADHD ADHD Controls AD/HD ODD/CD Controls AD/HD ODD/CD + + ODD/CD ODD/CD (420-500ms) (420-500ms) -10.0 µV 0µV 10.0 µV -10.0 µV 0µV 10.0 µV E Figure 2 RPs for correct Go-trials ERPs for Figure 3 successful Stop-trials ERPs for correct Go-trials. Grand-average waveshapes ERPs for successful Stop-trials. Grand-average waves in from the region of interest (F4/F8), and spline-interpolated the region of interest and spline-interpolated maps for suc- maps for correct Go-trials for normal controls (black), AD/ cessful Stop-trials for normal controls (black), AD/HD (red), HD (red), AD/HD+ODD/CD (green) and ODD/CD (blue). AD/HD+ODD/CD (green) and ODD/CD (blue). Only nor- There were no group-differences and no negative peaks in mal control children display a negative peak approximately the region and time window of interest. 210 ms after onset of the Stop-signal. Conclusion While all clinical groups displayed similarly attenuated neural signs of go-signal processing, the subsequent ODD/CD. The present results suggest that this conclusion response inhibition deficits further separated the clinical is not task specific. groups. Both children with AD/HD and ODD/CD- patients were found to be impaired in behavioural However, since CPT and Stop-task were performed by response inhibition. Also, both groups displayed reduced partly the same sample of children, an independent repli- neuronal inhibition as reflected by smaller right-frontal cation with a larger sample size is needed to further sup- Stop-N200 amplitudes; for AD/HD this is in agreement port this view. with Quay's model of psychopathology whereas for ODD/CD predictions of that model were violated. Hence, Although some evidence for inhibition deficits in AD/HD the inhibition-deficit concerning "stopping of an ongoing has been obtained using the CPT [43,44], it was also response" is by no means specific for AD/HD. In addition, found that neither commission errors nor the Nogo-N200 the comorbid group with AD/HD+ODD/CD which enhancement had differed between the groups in the cued should display the most severe deficits was found to be CPT [21,44]. We note that there are clear differences in even somewhat less impaired than the "pure" groups, what has to be inhibited in these two tasks: In the CPT, indicating that the comorbid condition may represent a participants have to withhold a prepared but not yet initi- separate disorder distinct from AD/HD and ODD/CD. ated response and made only a few false alarms. In the Stop-task, participants have to stop an already ongoing Limitations response which often failed. These two types of inhibition The study is limited by its small sample size and by the have to be differentiated (see e.g., Barkley [2,3]). fact, that another attention test was administered before- Page 7 of 14 (page number not for citation purposes) Behavioral and Brain Functions 2005, 1:22 http://www.behavioralandbrainfunctions.com/content/1/1/22 Datasets of a total of forty participants were thus analysed. ROI – Difference successful Stop minus correct Go [µV] They belonged to one of three clinical subgroups with the Stop-signal onset -4 ICD-10 diagnoses hyperkinetic disorder (F90.0, N = 11), oppositional defiant/conduct disorder (F91, F92, N = 8), -3 hyperkinetic conduct disorder (F90.1 N = 11) or to a -2 group of 11 healthy controls (Table 4). -1 Children of the clinical groups were sequential referrals to the Department of Child and Adolescent Psychiatry of the University of Göttingen who met no other psychiatric diagnoses except reading and/or spelling disorders (N = 15), enuresis (N = 1) or encopresis (N = 1). The diagnosis of a hyperkinetic disorder was concordant with the DSM- IV diagnosis of ADHD-combined type. Control children 0 100 200 300 400 500 600 700 800 900 [ms] met no other psychiatric diagnoses than reading and/or spelling disorders (N = 4). Diagnoses were verified by sen- ADHD Controls AD/HD ODD/CD ior board-certified child psychiatrists. All children under- ODD/CD went standardized IQ-testing with the German versions of the WISC-R [45] or Culture Fair Intelligence Test (CFT [46]). The CFT was used only in 5 cases (for 3 controls, 1 ADHD and 1 ADHD+ODD/CD). (420-500ms) Groups were matched by age but not by IQ, with lower IQs for the psychopathological groups compared to nor- -5.0 µV 0µV 5.0 µV mal controls (F(3/36) = 5.9, p = 0.01). Dif Figure 4 ference ERPs (successful Stop minus correct Go) Difference ERPs (successful Stop minus correct Go). One-way analyses of variance (ANOVAs) were carried out Difference waves and spline-interpolated difference maps to explore group-differences concerning the scales of the between event-related potential grand means of successful parent-rated Child Behaviour Checklist (CBCL [47]). Stop-correct Go-trials in the region of interest for normal There were group differences for all CBCL-scales except controls (black), AD/HD (red), AD/HD+ODD/CD (green) somatic complaints (F(3/36>4.5, p < 0.01), results of and ODD/CD (blue). A clear Stop-signal N200 is present post-hoc Scheffé-Test are shown in Table 4. only for normal controls. Stimuli and task The Stop-task consisted of eight blocks with 40 trials each hand. Valid performance data concerning inhibition per- and was identical to that used by Brandeis et al. [27] and formance was only available for one fixed SSD. Since SSRT Rubia et al [8]. Stimuli were presented in the central 2*2 is only estimated from one fixed SSD with less than the cm square of a VGA monitor at 120 cm viewing distance optimal Stop-failure rate of 50% [41], its reliability and its with fixation marks above and below the scene. Each trial sensitivity to group-differences is decreased compared to started with the presentation of an aeroplane in side view, other strategies to estimate SSRT. suggesting that is was 'flying to' the left or to the right, and the children had to press a button corresponding to the Methods planes flying direction with the index finger of their left or Subjects right hand. They were also told that sometimes a "little Fifty-eight boys aged 8–14 years participated in the study man" with his hands raised would follow, indicating that on the basis of informed consent by child and parent with they should withhold their response. This should be easy approval of the local ethics committee; all had normal or when the "little man" occurred early, but they should no corrected to normal vision, a full scale IQ above 80 and longer be able to stop their prepared response when the understood the Stop-task-instructions. Some datasets man was late. were deleted a priori because of more than 20% omissions of go-trials (for 2 controls, 1 AD/HD, 1 AD/HD+ODD/ Altogether, the "little man" Stop-signal occurred in 50% CD and 2 ODD/CD), and some were lost due to artifacts of the trials. The three fixed Stop-signal-delays (SSD) were in the EEG (1 control, 1 AD/HD, 3 AD/HD+ODD/CD and 100 ms (10% of all trials), 250 ms (30%) or 700 ms 1 ODD/CD) or due to age-matching the groups. (10%). The summed duration of the two signals was in Page 8 of 14 (page number not for citation purposes) Behavioral and Brain Functions 2005, 1:22 http://www.behavioralandbrainfunctions.com/content/1/1/22 Table 4: Sample description Group Controls (N) AD/HD (A) AD/ ODD/CD (O) ANOVA N = 11 N = 10 HD+ODD/CD N = 8 (AO) N = 11 Measure Mean (SD) Mean (SD) Mean (SD) Mean (SD) F p < Scheffé-Tests 3;36 Full-scale-IQ 110,7 (15,1) 94,4 (6,9) 93,0 (9,4) 96,9 (10,5) 5,9 0,01 N > A, AO Age (in months) 130,8 (18,9) 130,1 (18,0) 123,7 (18,5) 131,5 (27,4) 0,3 0,81 CBCL Social withdrawal 52,0 (4,6) 59,8 (6,5) 58,7 (7,7) 63,9 (9,9) 4,5 0,01 N < O Somatic complaints 56,3 (5,0) 56,2 (7,9) 57,2 (9,6) 57,4 (6,1) 0,1 0,98 Anxiety/Depression 50,3 (0,9) 57,2 (7,1) 64,3 (12,0) 67,3 (8,8) 8,4 0,01 N < O, AO Social problems 50.0 (0.0) 62.0 (8,7) 58,2 (8,2) 62,3 (11,5) 5,3 0,01 N < A, O Thought problems 50,6 (2,1) 54,2 (4,9) 61,4 (9,4) 61,8 (10,1) 5,9 0,01 N < O, AO Attention problems 50,3 (0,5) 67,4 (6,6) 67,3 (6,9) 69,1 (6,5) 25,5 0,01 N < A, AO, O 50,8 (2,7) 59,7 (8,7) 71,6 (10,2) 69,9 (10,0) 14,1 0,01 N < AO, O/A < AO Delinquent behaviour Aggressive behaviour 50,3 (0,9) 63,7 (8,7) 78,6 (12,0) 75,8 (14,6) 17,4 0,01 N < A, AO, O/A < AO Internalizing symptoms 44,3 (7,6) 57,4 (10,0) 62,3 (11,2) 65,8 (8,7) 10,0 0,01 N < A, AO, O Externalizing symptoms 38,3 (7,7) 63,1 (7,6) 75,3 (9,1) 73,3 (9,8) 41,9 0,01 N < A, AO, O /A < AO α < 0,1 Child Behaviour Checklist, T-scores every case 800 ms (800+0 ms, 100+700 ms, 250+550 ms, ously from electrodes above and below the left eye and at 700+100 ms) and a trial was presented every 1650 ms. the outer canthi. Impedances were kept below 5 kΩ , fur- ther analyses were computed with the Vision Analyzer Identical instructions were given to all groups before the 1.05 software. practice-block, and were repeated after a block in case the child made more than 25% Stop-failures in the short SSD The EEG was transformed to the average reference of the or less than 75% Stop-failures in the long SSD condition. 10–20 electrodes plus Fpz and Oz. Data were filtered Thus the short and long SSDs aiming at 0% and 100% offline (Butterworth, 0.1 to 30 Hz, 24 dB/oct.). For eye stop failures provided a time frame within which the movement correction the method of Gratton & Coles [48] child's response should occur, therefore only the medium without raw average subtraction was used. Trials with per- SSD was analysed. If there were less than 33% or more formance errors (side-errors, failed Stops and Go-reac- than 66% correct Stops at the medium SSD in a given tion-times faster than 200 ms), amplifier saturation or block, additional instructions were given to slow down or artefacts exceeding +-200 µV amplitude or more than 200 speed up responses, respectively. These adaptive instruc- µV amplitude difference in a segment -100 ms to 1500 ms tions prevent undesired strategies in performing the task, around go-signal-onset were rejected; remaining segments such as extreme speed-accuracy-tradeoffs yielding very fre- were subsequently checked visually. A 100 ms pre-stimu- quent or very rare Stops at the fixed medium SSD [27]. lus baseline (referred to the go-signal-onset) was taken as The inhibitory deficits detected in ADHD children are zero. Averages for successful Stops in the medium SSD comparable when using this Stop task with fixed SSDs, or contain at least 25 sweeps, correct Go-Averages contain at the standard version with adaptive SSDs [8]. least 90 sweeps. Groups did not differ in both numbers of accepted sweeps. ERP recording and processing An ERP was recorded using a Neuroscan recording system Analyses with calibrated technical zero baselines and Nihon Koh- SSRT den Ag/AgCl electrodes attached to the skin with Grass Reaction-times shorter than 200 ms and Go-Trials with EC-2 electrode-cream. Sampling-rate was 250 Hz and cut- side-errors were excluded from all analyses. SSRT was esti- off frequencies were 0.1 and 50 Hz on all 10–20 electrode mated only for the medium SSD because there were too positions using FCz as recording reference and a ground few Stop-failures in Stop-trials with short and too many in electrode placed on the forehead. Vertical and horizontal long SSD. The classic approach to calculate SSRT for a spe- electro-occulograms (EOGs) were recorded simultane- cific SSD is to rank-order reaction-times of the go-trials, Page 9 of 14 (page number not for citation purposes) Behavioral and Brain Functions 2005, 1:22 http://www.behavioralandbrainfunctions.com/content/1/1/22 tion-time slow enough not to yield a Stop-failure, and the Correct Go mean of the two minus their Stop-signal-delay would GFP Diss*10 yield a good estimate for SSRT. This brings into account, [µV] I. II. III. IV. V. 7 that the distribution of Go-reaction-times is discrete rather than continous. Applied to a dataset without Stop-failures, we can only determine one border of the area of reaction times in which correct Stops and Stop-failures occur; we only know 0 100 200 300 400 500 600 700 800 900 [ms] reaction-times which are slow enough not to evoke a Stop-failure. The best we can say therefore is that the SSRT Successful Stop shall be faster than the fastest Go-reaction-time with Stop- GFP Diss*10 [µV] signal-delay subtracted. If a dataset contains no correct I. II. III. IV. V. VI. Stops, we only know that every Go reaction-time was too fast to be stopped, but we do not know anything more; simply taking the fastest Go-reaction-time with SSD sub- tracted as SSRT would be wrong. Because of this indeter- minacy of Stop-signal-reaction-time, participants with no Stop-failures as well as participants with no correct Stops 0 100 200 300 400 500 600 700 800 900 [ms] need to be excluded from analyses. This was not necessary Microstate estimation Figure 5 according to GFP and Diss for the dataset presented. Microstate estimation according to GFP and Diss. Adaptive segmentation of the total groups grand mean from Brain-mapping correct Go (top) and successful Stop (bottom). Microstate Microstates were determined on the total group's grand boarders were determined by relative minima of GFP (black) mean. Borders were set at times with minimal global- together with relative maxima in Diss (red, for better scaling field-power (GFP) indicating low map-strength, plus max- multiplicated with 10) Correct Go-trials revealed five micro- imal dissimilarity (Diss, the GFP of the difference between states (76–196 ms, 200–272 ms, 276–412 ms, 416–504 ms, successive normalized maps) reflecting high topographic 508–640 ms), successful Stops six microstates (76–196 ms, instability [27,50]. In contras, components extracted by 200–272 ms, 276–428 ms, 432–504 ms, 508–592 ms, 596– principal component analysis (PCA) were only statisti- 724 ms). cally defined as sources of variance and may not necessar- ily be grounded by physiological components [51,52]. For correct Go, five microstates were found (76–196 ms, multiply the probability for a Stop-failure with the 200–272 ms, 276–412 ms, 416–504 ms, 508–640 ms), number of go reaction-times which yields n, take the go correct Stops revealed six (76–196 ms, 200–272 ms, 276– reaction-time of the nth rank and subtract the SSD [9,49]. 428 ms, 432–504 ms, 508–592 ms, 596–724 ms, see Fig- This leads to certain difficulties: for instance, if a partici- ure 5). For each microstate a mean map with its GFP and pant makes no Stop-failure in the questioned SSD; the summary measures of topography (centroids) [50] were probability for a Stop-failure will be zero and there is no computed (Figure 6). zero-rank of go reaction-times. But this participant has initiated a quite well-working Stop-process with which Stop-N200 our applied theory can not cope. On the other hand, if a In this version of the Stop-task, processing of the Stop-sig- participant was not able to Stop even once, the algorithm nal is fully time-locked with the preceding go-stimulus would yield a wrong estimate of SSRT. Taken together, the and thus highly confounded with go-signal processing. algorithm stated above could lead to an undefined state Because of this, differences between features of Stop- vs. and to wrong results which makes it susceptible of formal Go-trials were analysed in a repeated measure-design; it is refutation. likely that such differences were caused by processing the additional Stop-signal on Stop- trials. Separate analyses of Hence, we used a slightly different strategy: We took the the conditions and inspection of the segment t-maps of probability for a Stop-failure, multiplied it with the total the group differences in the raw conditions were used to number of correct go-reactions and truncated the result. exclude alternative interpretations. The term "Stop-N200" There we got the rank n (if there is any) of the Go-reac- as used here thus refers to this difference between mean tion-time which was just too fast to be stopped, the (n+1)- negativity in the ROI of successful Stop and correct go-tri- rank (again, if there is any) denotes the fastest Go-reac- als. Page 10 of 14 (page number not for citation purposes) Behavioral and Brain Functions 2005, 1:22 http://www.behavioralandbrainfunctions.com/content/1/1/22 correct Go t-Maps A A-C AO-C AO O-C I II III IV V I II III IV V 76-196ms 200-272ms 276-412ms 416-504ms 508-640ms 76-196ms 200-272ms 276-412ms 416-504ms 508-640ms -10.0µV 0 10.0µV -10.0µV 0 10.0µV successful Stops t-Maps A A-C AO-C AO O-C I II III IV V VI I II III IV V VI 76-196ms 200-272ms 276-428ms 432-504ms 508-592ms 596-724ms 76-196ms 200-272ms 276-428ms 432-504ms 508-592ms 596-724ms -10.0µV 0 10.0µV -10.0 0 10.0 Microstate-maps and t-maps for co Figure 6 rrect Go and successful Stop-trials Microstate-maps and t-maps for correct Go and successful Stop-trials. Spline-interpolated microstate-maps for nor- mal controls (C), children with AD/HD (A), ODD/CD (O) and AD/HD+ODD/CD (AO) and additional exploratory t-maps > 1.7 p < .05. with comparisons of clinical groups vs. controls. Unadjusted two-tailed significance-level is reached at t (17 to 21) Visual inspection of the normal controls' grand mean of almost identical with microstate IV found with the Brain- correct Stops revealed a negativity peaking at about 460 mapping-approach. ms after go-signal-onset (or 210 ms after Stop-signal- onset) at right-frontal electrodes which was absent in go- In order to localize the region of interest in this time-win- trials. For further analyses, the mean amplitude in a time- dow for the sub-sample of normal controls a repeated- window 420–500 ms after onset of the Go-signal was measure-ANCOVA with within-subject-factors "condi- computed separately for correct go- and successful Stop- tion" (successful Stop vs. correct Go) and electrode-sites trials. The time-window used to study this local effect is "anterior-posterior" (3 levels) and "left-right" (5 levels) Page 11 of 14 (page number not for citation purposes) Behavioral and Brain Functions 2005, 1:22 http://www.behavioralandbrainfunctions.com/content/1/1/22 Table 5: Effect size (Cohen's d) for the main dependent variables Controls (N) AD/HD (A) AD/HD+ODD/CD ODD/CD (O) (AO) a a a a Measure effect-size d effect-size d effect-size d effect-size d sc sc sc sc Stop-failure reaction-time (ms) A -1.26* AO -1.22* .03 O -.87* .39 .36 SSRT at 250 ms SSD (ms) A -.91* AO .06 .96* O -1.08* -.18 -1.13* Go-Trial ROI mean amplitude (µV) A -.29 AO -.57 -.28 O -.60 -.31 -.02 Stop-Trial ROI mean amplitude (µV) A -1.16* AO -.93* .22 O -1.15* .01 -.20 Stop-N200 ROI mean amplitude (µV) A -1.15* AO -.66 .49 O -.89+ .26 -.22 d is the standardized mean difference for the sample, based on the ANCOVA model with age as covariate. This is done in order to account for sc developmental effects. Region of interest, mean of electrodes F4 and F8 at 420-500 ms post Go-signal onset * one-tailed, P < 0.05 + one-tailed, P < 0.10 was computed for the vector-length-normalized dataset. Stop was not affected by age (r = .13; all one-tailed tested, Vector-normalization is necessary, because "condi- * p < .05). tion*location" interactions can result from multiplicative changes in source strength between conditions without Therefore, age was taken as a covariate for all comparisons specific differences concerning locations [53]. The to reduce error-variance due to developmental effects and ANOVA revealed a significant interaction-effect "condi- thus increase statistical power. tion*left-right" (F = 5.93, ε = .542, p = .01) and an (4,80) interaction "condition*left-right*anterior-posterior" Statistical tests (F = 3.00, ε = .424, p = .04). Exploratory analyses of Go-reaction-time, Stop-failure reaction-time and SSRT (8,80) repeated measure "condition" for each electrode sepa- were analysed with one-way analyses of covariance rately (without vector normalization) revealed significant (ANCOVAs) with between-subject-factor "group" and differences between conditions only at electrodes F4 covariate "age". In case of overall-differences between (F = 15.1, p = .003) and F8 (F = 15.2, p = .003) groups, planned contrasts were computed in order to test (1,10) (1,10) with increased negativity in trials with successful Stops as the hypothesis that clinical groups display decreased per- well as increased positivity at P3 (F = 20.9, p = .001). formance (slower SSRT, Go- and Stop-failure-reaction- (1,10) Therefore the mean-amplitude of the adjacent right-ante- time) compared to normal controls. The inhibition-func- rior electrodes F4 and F8 in this time-window were used tion of probabilities of Stop-failures for each SSD was ana- as region of interest (ROI) in order to analyse Stop-signal- lysed with a two-way repeated measure ANCOVA with N200, similar to Pliszka et al. [7]. within-subject-factor "SSD", between-subject-factor "group" and covariate "age". To test whether dependent measures were confounded with developmental effects (the higher age, the higher per- All microstates of correct Go and successful Stop were ana- formance and the lower ERP-amplitudes), simple correla- lysed exploratory concerning GFP with one-way ANCO- tions with "age" were computed across all groups. For Go- VAs with between-subject-factor "group" and covariate reaction-time (r = -.40*), Stop-failure reaction-time (r = - "age". Differences in topography as reflected by locations .72*), SSRT (r = -.76*) and mean amplitude in the ROI for of centroids were analyzed with MANCOVAs of depend- correct Go (r = .27*) and successful Stop (r = .27*) devel- ent variables "location of positive and negative centroids" opmental effects occurred, whereas the N200 of the differ- (left to right and anterior to posterior for each), covariate ence-wave in the ROI between correct Go and successful "age" and between-subject-factor "group" [54]. Page 12 of 14 (page number not for citation purposes) Behavioral and Brain Functions 2005, 1:22 http://www.behavioralandbrainfunctions.com/content/1/1/22 2. Barkley RA: Behavioral inhibition, sustained attention, and Group-comparisons of Stop-signal-N200 were analysed executive functions: constructing a unifying theory of with a two-way repeated-measure ANCOVA with within- ADHD. Psychol Bull 1997, 121(1):65-94. subject-factor "condition" (correct Go vs. successful Stop), 3. Barkley RA: The Executive Functions and Self-Regulation: An Evolutionary Neuropsychological Perspective. Neuropsychol between-subject-factor "group" and covariate "age". In Rev 2001, 11(1):1-29. case of significant differences, further one-way ANCOVAs 4. Aman CJ, Roberts RJJ, Pennington BF: A neuropsychological examination of the underlying deficit in attention deficit and additional planned contrasts were computed. In hyperactivity disorder: frontal lobe versus right parietal lobe order to correct results of repeated-measure ANCOVAs theories. Dev Psychol 1998, 34(5):956-969. from violations from sphericity, Greenhouse-Geisser ε 5. Schachar R, Mota VL, Logan GD, Tannock R, Klim P: Confirmation of an inhibitory control deficit in attention-deficit/hyperac- and p-values for corrected degrees of freedom were tivity disorder. J Abnorm Child Psychol 2000, 28(3):227-235. reported. 6. Pliszka SR, Borcherding SH, Spratley K, Leon S, Irick S: Measuring inhibitory control in children. J Dev Behav Pediatr 1997, 18(4):254-259. To test an additive model of effects on response inhibi- 7. Pliszka SR, Liotti M, Woldorff MG: Inhibitory control in children tion, separate 2*2 ANCOVAS with between-subject fac- with attention-deficit/hyperactivity disorder: event-related potentials identify the processing component and timing of tors "AD/HD" and "ODD/CD" and covariate "age" were an impaired right-frontal response-inhibition mechanism. computed for the main dependent variables SSRT and Biol Psychiatry 2000, 48(3):238-246. Stop-N200. 8. Rubia K, Oosterlaan J, Sergeant JA, Brandeis D, v Leeuwen T: Inhib- itory dysfunction in hyperactive boys. Behav Brain Res 1998, 94(1):25-32. Because of small sample size, even trends with p < .10 will 9. Oosterlaan J, Logan GD, Sergeant JA: Response inhibition in AD/ HD, CD, comorbid AD/HD + CD, anxious, and control chil- be reported for hypothesized group and condition differ- dren: a meta-analysis of studies with the stop task. J Child Psy- ences. Cohen's standardized mean difference for the sam- chol Psychiatry 1998, 39(3):411-425. ple with age taken as covariate d were computed (Table sc 10. Losier BJ, McGrath PJ, Klein RM: Error patterns on the continu- ous performance test in non-medicated and medicated sam- 5). ples of children with and without ADHD: a meta-analytic review. J Child Psychol Psychiatry 1996, 37(8):971-987. Since groups were not IQ-matched, influences of IQ on 11. Epstein JN, Erkanli A, Conners CK, Klaric J, Costello JE, Angold A: Relations between Continuous Performance Test perform- dependent measures were analysed using partial correla- ance measures and ADHD behaviors. J Abnorm Child Psychol tion coefficients with covariate "age" and will be reported 2003, 31(5):543-554. 12. Barkley RA, Grodzinsky G, DuPaul GJ: Frontal lobe functions in in case of significance. attention deficit disorder with and without hyperactivity: a review and research report. J Abnorm Child Psychol 1992, Procedure 20(2):163-188. 13. Klorman R, Hazel-Fernandez LA, Shaywitz SE, Fletcher JM, Marchione The psychophysiological experiment took place in a KE, Holahan JM, Stuebing KK, Shaywitz BA: Executive functioning video-controlled, noise-protected and slightly dimmed deficits in attention-deficit/hyperactivity disorder are inde- room at the Department of Child and Adolescent Psychi- pendent of oppositional defiant or reading disorder. J Am Acad Child Adolesc Psychiatry 1999, 38(9):1148-1155. atry at the University of Göttingen. Participants sat in a 14. Pennington BF, Groisser D, Welsh MC: Contrasting cognitive def- dentist-chair during electrode-attachment and task per- icits in attention deficit hyperactivity disorder versus reading disability. Dev Psychology 1993, 29(3):511-523. formance. Throughout the tasks, they could communicate 15. Carter CS, Krener P, Chaderjian M, Northcutt C, Wolfe V: Abnor- with the experimenter via intercom. At first, a CPT lasting mal processing of irrelevant information in attention deficit at about 11 minutes was performed [21,42], followed by hyperactivity disorder. Psychiatry Res 1995, 56(1):59-70. 16. Sergeant JA, Geurts H, Oosterlaan J: How specific is a deficit of this Stop-Task of approximately 10 minutes duration. executive functioning for Attention-Deficit/Hyperactivity Disorder? Behav Brain Res 2002, 130(1-2):3-28. 17. Pennington BF, Ozonoff S: Executive functions and developmen- Competing interests tal psychopathology. J Child Psychol Psychiatry 1996, 37(1):51-87. The author(s) declare that they have no competing inter- 18. Quay HC: Inhibition and attention deficit hyperactivity disor- ests. der. J Abnorm Child Psychol 1997, 25(1):7-13. 19. Quay HC: The psychobiology of undersocialized aggressive conduct disorder: A theoretical perspective. Dev Psychopathol- Authors' contributions ogy 1993, 5(1-2):165-180. BA performed analyses and drafted the manuscript, TB, 20. Gray JA: Neural systems, emotion and personality. In Neorobi- ology of learning, emotion and affect Edited by: Madden J. New York , DB, HH and AR conceived the study and helped to draft Raven Press; 1991:273-306. the manuscript. 21. Banaschewski T, Brandeis D, Heinrich H, Albrecht B, Brunner E, Rothenberger A: Questioning inhibitory control as the specific deficit of ADHD--evidence from brain electrical activity. J Acknowledgements Neural Transm 2004, 111(7):841-864. The authors like to thank all children for participation in the study and 22. Raine A: Annotation: the role of prefrontal deficits, low auto- Susanne Mock for EEG-recording. nomic arousal, and early health factors in the development of antisocial and aggressive behavior in children. J Child Psychol Psychiatry 2002, 43(4):417-434. References 23. Raine A: Biosocial studies of antisocial and violent behavior in 1. American-Psychiatric-Association: Diagnostic and statistical children and adults: a review. J Abnorm Child Psychol 2002, manual of mental disorders. 4th edition. Washington , American 30(4):311-326. Psychiatric Association; 1994. Page 13 of 14 (page number not for citation purposes) Behavioral and Brain Functions 2005, 1:22 http://www.behavioralandbrainfunctions.com/content/1/1/22 24. Raine A: Antisocial behavior and psychophysiology: A bioso- anterior cingulate function in attention-deficit/hyperactivity cial perspective and a prefrontal dysfunction hypothesis. In disorder boys. Clin Neurophysiol 2004, 115(4):973-981. Handbook of antisocial behavior Edited by: D. Stoff JBJDM. New York , 45. Tewes U: Hamburg-Wechsler Intelligenztest für Kinder - Wiley; 1997:289-304. Revision. HAWIK-R. Bern , Hans Huber; 1983. 25. Logan GD, Cowan WB, Davis KA: On the ability to inhibit simple 46. Weiss RH: Grundintelligenztest Skala 2 - (CFT 20), 3 ed. and choice reaction time responses: a model and a method. Braunschweig , Westermann; 1987. J Exp Psychol Hum Percept Perform 1984, 10(2):276-291. 47. Achenbach TM: Manual for Child Behavior Checklist/ 4-18 and 26. Oosterlaan J, Sergeant JA: Response inhibition and response re- 1991 Profile. Burlington, VT , University of Vermont Department engagement in attention-deficit/hyperactivity disorder, dis- of Psychiatry; 1991. ruptive, anxious and normal children. Behav Brain Res 1998, 48. Gratton G, Coles MG, Donchin E: A new method for off-line 94(1):33-43. removal of ocular artifact. Electroencephalogr Clin Neurophysiol 27. Brandeis D, van Leeuwen TH, Rubia K, Vitacco D, Steger J, Pascual- 1983, 55(4):468-484. Marqui RD, Steinhausen HC: Neuroelectric mapping reveals 49. Logan GD, Cowan WB: On the Ability to Inhibit Thought and precursor of stop failures in children with attention deficits. Action: A theory of an Act of Control. Psychological Review 1984, Behav Brain Res 1998, 94(1):111-125. 91(3):295-327. 28. Frahm J, Baudewig J, Dechent P, Merboldt KD: Advances in Func- 50. Brandeis D, Vitacco D, Steinhausen HC: Mapping brain electric tional MRI of the Human Brain. Progr NMR Spectr 2004, 44:1-32. micro-states in dyslexic children during reading. Acta Pae- 29. Picton TW, Bentin S, Berg P, Donchin E, Hillyard SA, Johnson RJ, dopsychiatr 1994, 56(3):239-247. Miller GA, Ritter W, Ruchkin DS, Rugg MD, Taylor MJ: Guidelines 51. Skrandies W: Data reduction of multichannel fields: global for using human event-related potentials to study cognition: field power and principal component analysis. Brain Topogr recording standards and publication criteria. Psychophysiology 1989, 2(1-2):73-80. 2000, 37(2):127-152. 52. Skrandies W: EEG/EP: new techniques. Brain Topogr 1993, 30. Rubia K, Overmeyer S, Taylor E, Brammer M, Williams SC, Simmons 5(4):347-350. A, Bullmore ET: Hypofrontality in attention deficit hyperactiv- 53. McCarthy G, Wood CC: Scalp distributions of event-related ity disorder during higher-order motor control: a study with potentials: an ambiguity associated with analysis of variance functional MRI. Am J Psychiatry 1999, 156(6):891-896. models. Electroencephalogr Clin Neurophysiol 1985, 62(3):203-208. 31. Overtoom CC, Kenemans JL, Verbaten MN, Kemner C, van der 54. Olson CL: On choosing a test statistic in multivariate analysis Molen MW, van Engeland H, Buitelaar JK, Koelega HS: Inhibition in of variance. Psychol Bull 1976, 83:579-586. children with attention-deficit/hyperactivity disorder: a psy- chophysiological study of the stop task. Biol Psychiatry 2002, 51(8):668-676. 32. Dimoska A, Johnstone SJ, Barry RJ, Clarke AR: Inhibitory motor control in children with attention-deficit/hyperactivity disor- der: event-related potentials in the stop-signal paradigm. Biol Psychiatry 2003, 54(12):1345-1354. 33. Kok A: Effects of degradation ov visual stimuli on compo- nents of the event related potential (ERP) in Go/nogo reac- tion tasks. Biol Psychol 1986, 23:21-38. 34. Falkenstein M, Hoormann J, Hohnsbein J: ERP components in Go/ Nogo tasks and their relation to inhibition. Acta Psychol (Amst) 1999, 101(2-3):267-291. 35. Jodo E, Kayama Y: Relation of a negative ERP component to response inhibition in a Go/No-go task. Electroencephalogr Clin Neurophysiol 1992, 82(6):477-482. 36. Kopp B, Mattler U, Goertz R, Rist F: N2, P3 and the lateralized readiness potential in a nogo task involving selective response priming. Electroencephalogr Clin Neurophysiol 1996, 99(1):19-27. 37. Kopp B, Rist F, Mattler U: N200 in the flanker task as a neurobe- havioral tool for investigating executive control. Psychophysiol- ogy 1996, 33(3):282-294. 38. Botvinick MM, Braver TS, Barch DM, Carter CS, Cohen JD: Conflict monitoring and cognitive control. Psychol Rev 2001, 108(3):624-652. 39. Nieuwenhuis S, Yeung N, van den Wildenberg W, Ridderinkhof KR: Electrophysiological correlates of anterior cingulate func- tion in a go/no-go task: effects of response conflict and trial type frequency. Cogn Affect Behav Neurosci 2003, 3(1):17-26. 40. Donkers FC, van Boxtel GJ: The N2 in go/no-go tasks reflects conflict monitoring not response inhibition. Brain Cogn 2004, 56(2):165-176. Publish with Bio Med Central and every 41. Band GP, van der Molen MW, Logan GD: Horse-race model sim- scientist can read your work free of charge ulations of the stop-signal procedure. Acta Psychol (Amst) 2003, 112(2):105-142. "BioMed Central will be the most significant development for 42. Banaschewski T, Brandeis D, Heinrich H, Albrecht B, Brunner E, disseminating the results of biomedical researc h in our lifetime." Rothenberger A: Association of ADHD and conduct disorder-- Sir Paul Nurse, Cancer Research UK brain electrical evidence for the existence of a distinct sub- type. J Child Psychol Psychiatry 2003, 44(3):356-376. Your research papers will be: 43. Brandeis D, Banaschewski T, Baving L, Georgiewa P, Blanz B, Warnke available free of charge to the entire biomedical community A, Steinhausen HC, Rothenberger A, Scheuerpflug P: Multicenter P300 brain mapping of impaired attention to cues in hyper- peer reviewed and published immediately upon acceptance kinetic children. J Am Acad Child Adolesc Psychiatry 2002, cited in PubMed and archived on PubMed Central 41(8):990-998. 44. Fallgatter AJ, Ehlis AC, Seifert J, Strik WK, Scheuerpflug P, Zillessen yours — you keep the copyright KE, Herrmann MJ, Warnke A: Altered response control and BioMedcentral Submit your manuscript here: http://www.biomedcentral.com/info/publishing_adv.asp Page 14 of 14 (page number not for citation purposes)
Journal
Behavioral and Brain Functions – Springer Journals
Published: Dec 9, 2005