Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Increased performance uncertainty in children with ADHD? - Elevated post-imperative negative variation (PINV) over the ventrolateral prefrontal cortex

Increased performance uncertainty in children with ADHD? - Elevated post-imperative negative... Background: We aimed to investigate the influences of attention deficit/hyperactivity disorder (ADHD) on response evaluation, as reflected by the postimperative negative variation (PINV), a slow event-related potential. Methods: We investigated PINV as an indicator of performance uncertainty in an audio-visual contingent negative variation (CNV) paradigm with an interstimulus interval of 3 seconds. A constant, unilateral, quick motor reaction with either the right or the left thumb was required after an auditory forewarned (S1) visual imperative stimulus (S2). We examined 18 ADHD patients (combined or hyperactive-impulsive subtype) aged between 8 and 14 years and an age-, sex and IQ-matched control group of 19 healthy subjects using 64-channel high-density EEG. A first run was recorded drug-free, a second one under methylphenidate (MPH) medication in the ADHD group. Results: We found a significantly increased negativity of the PINV-component over the ventrolateral prefrontal cortex in ADHD children compared to the healthy control group. PINV amplitude was influenced by movement side, most likely due to the slightly more difficult task when left hand responses were required. After the intake of MPH, PINV amplitudes of ADHD children normalized. Conclusions: We conclude that children with ADHD are likely to be more uncertain about the correctness of their performance and interpret the increased PINV as a hint towards compensatory mechanisms for a deficit in the evaluation of contingencies. Further studies are needed to assess the exact extent to which remainders of eye- movement related potentials contribute to PINV amplitude despite the correction for eye-artifacts. Background Certain event-related potentials have been discussed as Attention deficit/hyperactivity disorder (ADHD) is one of markers for the disorder but previous studies have the most common [1] and at the same time still not com- pointed out heterogeneous neurophysiological profiles in pletely pathophysiologically understood child psychiatric ADHD patients [3,4]. diagnoses. Recently, a decreased error related negativity (ERN) over Clear deficits in executive functions like planning, the anterior cingulate cortex (ACC) has been interpreted inhibition and evaluation of movement have been found. as ADHD children’s diminished capacity to monitor their error responses and their failure to predict the likelihood On the other hand, it is still controversially discussed to which extent motivational aspects and deficits in delay that an error occurs in a given context [5-7]. aversion are responsible for the development of ADHD- These findings suggest deficits in the children’s cogni- typical symptoms [2]. tive processing of movement caused by diminished internal monitoring processes [8]. * Correspondence: Stephan.Bender@uniklinikum-dresden.de In the current study we chose the PINV (postimpera- Department of Child and Adolescent Psychiatry, Medical School Carl Gustav tive negative variation) component as another important Carus, Dresden University of Technology, Fetscherstraße 74, D-01307 marker of movement/action monitoring processes, Dresden, Germany Full list of author information is available at the end of the article © 2011 Werner et al; licensee BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. Werner et al. Behavioral and Brain Functions 2011, 7:38 Page 2 of 9 http://www.behavioralandbrainfunctions.com/content/7/1/38 representing the individual’s uncertainty about the cor- we included children treated with both immediate and rectness of a given answer [9], aimed to further investi- extended release MPH in our patient group. gate the disorders’ influence on self-monitoring and the Twelve out of 18 ADHD children were treated with establishment of stable contingencies between stimuli extended release MPH with a mean dosage of 0.85 mg/kg and the corresponding required responses [10]. body weight (0.25 to 1.29 mg/kg), the other six with Enhanced PINV amplitudes have been found in a num- immediate release MPH with a mean dosage of 0.4 mg/kg ber of studies, reflecting a lack of control over aversive (0.15 to 0.74 mg/kg). events, an unexpected change in controllability [11,12] Nineteen right-handed (EHI laterality quotient 97.3 ± 5.2) healthy, age-, gender- and IQ-matched children and and representing contingency reappraisal [13]. Especially schizophrenic [14,15] and depressive [16] adolescents (14 males and 5 females, mean age 11.6 ± 2.1 individuals show elevated PINV amplitudes, representing years, mean IQ 117.4 ± 13.0), who took no psychoactive the uncertainty about the appropriate response [17]. medication and did not suffer from any neurological or PINV amplitude is sensitive to ambiguous contingen- psychiatric symptoms, were recruited as control group at cies and is thought to reflect an unexpected outcome, Heidelberg’s elementary and secondary schools. which causes performance uncertainty [18]. PINV usually In both groups we screened for visual impairments shows a (pre-)frontal maximum, so generators in the pre- (corrected visus ≥ 0.8). frontal cortex have been postulated [18]. A topographic All subjects and their parents provided written informed multi-channel analysis of PINV in ADHD has not been consent according to the Declaration of Helsinki and the performed so far. study was approved by the local ethics committee. We hypothesized that � ADHD children’s contingency evaluation and their Task/recording/data pre-processing cognitive performance monitoring is disturbed, resulting We recorded a CNV paradigm, using an auditory warning in an increased PINV amplitude of ADHD children stimulus S1 (1000 Hz, 90 dB, 50ms duration) and a visual compared to age-matched healthy controls. imperative stimulus S2 (image of a white hand, pointing � Methylphenidate (MPH) has a positive influence on towards the side of the required button press, presented the cognitive evaluation represented by a normalized for 150 ms on a black screen). The interstimulus interval PINV-amplitude after MPH-intake in the ADHD group. was 3 seconds, intertrial intervals varied randomly from 7 to 11 seconds. Methods Subjects were instructed to correctly respond as fast as Subjects possible when S2 occurred on the screen by pressing a We analyzed eighteen right-handed (Edinburgh Handed- button on the STIM response pad (Neuroscan Inc, TX, ness Inventory; EHI; [19]; laterality quotient mean USA) with the thumb of either the right or the left hand value ± standard deviation 94.9 ± 10.2) children between (quick, unilateral motor answer). 8 and 14 years (13 males and 5 females, mean age ± stan- 40 trials per hand were recorded in a counterbalanced dard deviation 11.5 ± 1.9 years, mean IQ ± standard order across subjects. Two runs were recorded: In the deviation 110.5 ± 18.8) who met the criteria of a hyperac- control group both runs, T1 and T2, were drug-free. In tive-impulsive or combined subtype of ADHD according the ADHD group, the first one (T1) was drug-free (after at to the semi-structured interview for DSM, K-SADS [20]. least 24 hours after the last intake of MPH), the second All patients were recruited either in the Child and Ado- one (T2) after 70 minutes after the intake of the individual lescent Psychiatric Department of the University of used dose of MPH. In other studies the same experimental Heidelberg or at a child psychiatrist’s practice, were trea- period of 70 minutes after the intake of MPH was chosen, ted with multilayer-release or immediate-release MPH so comparability is ensured. without other co-medication and suffered from no other Participants fixated a cross on a computer screen in psychiatric diseases. This includes that we assured that order to minimize eye artifacts. Neuroscan Synamp there were no neuropsychiatric disorders such as psy- Amplifiers (Neuroscan Inc., USA) were used to record choses and autism or neurological diseases as epilepsy continuous DC 64-channel EEG with a sampling rate [21], migraine [22] and tic-disorder [10], which are of 250 Hz. An anti-aliasing filter was set at 70 Hz thought to lead to specific changes in contingent negative (low-pass). Surface Ag-AgCl sintered electrodes were variation (CNV) parameters. fixed using an equidistant electrode cap (Easycap, An IQ below 80 (4-subtest short version of HAWIK FMS, Germany) and are named according to an III [23]), led to exclusion from the study. extended international 10-20 system. The vertical and As Quinn et al. [24] found no significant difference horizontal electrooculogram (EOG) was recorded by between the concentration of multilayer- and immediate- electrodes 1 cm next to the outer canthi and above/ release MPH within the first four hours after the intake, below the left eye. Werner et al. Behavioral and Brain Functions 2011, 7:38 Page 3 of 9 http://www.behavioralandbrainfunctions.com/content/7/1/38 Impedances were kept below 5 kΩ. Data were recorded 305 ± 63 ms (right hand button presses) and 320 ± 81 ms against a reference near Cz and transformed offline to (left hand button presses) for healthy control children. average reference. Recordings 1 s before S1 served as base- A repeated measurements ANOVA with the factors line. For the analysis of PINV, the EEG-signal was digitally SIDE of the response movement and GROUP (ADHD filtered (30 Hz high cut-off), segmented into epochs of versus healthy controls) revealed shorter reaction times 7.5 s (1 s pre S1 to 3.5 s post S2), corrected automatically for responses with the dominant right than with the left for DC-drifts by linear regression (Brain Vision Analyzer, hand (F(1;35) = 5.9; p = 0.02). In contrast, GROUP had Brain Products GmbH, Germany), and for eye movements no significant main effect on reaction times (F(1;35) = and blinks (algorithm according to Gratton and Coles as 0.3; p > 0.59) and did not interact with response move- implemented in Brain Vision Analyzer Version 1). ment side either (F(1;35) = 0.0; p > 0.90). Artifacts were rejected automatically if the signal 1.2 EEG data - PINV amplitude exceeded 150 mV. This procedure was con- Group differences of the PINV amplitudes over the ven- firmed by visual inspection; only artifact free trials trolateral prefrontal areas between unmedicated ADHD entered further analysis. Bad channels were interpolated patients and the control group (T1) are presented for using nearest neighbours. Trials were rejected from left (Table 1) and right hand response movements further analysis if subjects responded with the wrong (Table 2). Mean values and standard deviations as well hand or after more than 3.5 s after S2. as the results of the four t-tests are shown. Two ADHD patients and one control child (out of ori- We found significantly elevated PINV amplitudes over ginally n = 20 children in both groups) had to be the right ventrolateral prefrontal cortex (VLPFC) in the excluded from further evaluation due to recording ADHD group in comparison to the healthy control group, errors or excessive artifact-prone data. N = 18 ADHD when the unilateral motor response was given by the left children and n = 19 controls were included for further hand (p = 0.01; t = 2.7). For the ipsilateral, left prefrontal statistical analysis. area there was no significant difference (p = 0.30, t = 1.1). Figure1 shows thetimecourse of the prefrontal PINV Data analysis/statistics amplitudes separately for each hemisphere when the uni- As a first step, planned comparisons for group differences lateral response movement is given with the left hand. between the PINV-amplitudes of unmedicated ADHD For the unilateral response movement with the right versus control children over the left and right ventrolat- hand, there were no significant group differences (cf. eral prefrontal areas (pooled leads AF7, FP1, F9 and AF8, Tables 1 and 2). FP2, F10 during the time interval 2000 to 3000 ms after The topographical analysis of the cortical activation 2000 to 3000 ms after the target stimulus S2 is shown the imperative stimulus S2 in agreement with results of our previous study [18]) were examined for right and left in Figure 2 (reference-free current source density maps), hand button presses by four t-tests. The significance level illustrating the above-described group differences. Irre- was set to p = 0.05/4 = 0.0125 (Bonferroni correction). spective of the side of the response movement, a higher Next, in order to assess the influence of medication on right-sided negativity over ventrolateral prefrontal areas PINV topography in more detail, results were examined during PINV is obvious, although the lateralization of by multivariate analysis of variance (MANOVA), using the the activation is noticeably weaker when the response between subject factor GROUP (ADHD versus healthy movement is given by the right thumb. controls) and the within subject factors SIDE of the response movement (left vs. right hand), HEMISPHERE 2. Separation of PINV and eye movement and blink (left vs. right VLPFC), RUN (T1 vs. T2) and ELECTRO- artifacts DES (AF7/8, FP1/2, F9/10) followed by simpler separate In addition to the eye movement correction, we performed MANOVAs for left and right hand response CNV tasks. a comparison between the time-course of the electroocula- Significant main effects or interactions in the MANOVA gram and the PINV amplitudes, which revealed an inde- were subsequently further examined by post-hoc tests pendent time course as shown in Figure 3. (Newman Keuls). Although quite a lot of blink or eye movement arti- facts occurred during the PINV interval (which made a Results complete removal of all trials with blink artifacts impos- 1. Group differences between drug-free ADHD patients sible), the visual examination of the single trials also and control children (medication-free first run) confirmed that the time-course of the potentials in the 1.1 Behavioral data - reaction times leads over the ventrolateral prefrontal areas could not Mean reaction times (± standard deviation) were 317 ± be explained by remainders of insufficiently corrected 65 ms (right hand button presses) and 333 ± 84 ms (left eye artifacts, as the time-courses differed in single trials hand button presses) for children with ADHD as well as as well. Werner et al. Behavioral and Brain Functions 2011, 7:38 Page 4 of 9 http://www.behavioralandbrainfunctions.com/content/7/1/38 Table 1 Group differences in PINV amplitude for left hand responses right hemisphere mean [μV] ± standard deviation AF8 FP2 F10 right VLPFC Difference right VLPFC ADHD vs CO Control group (N = 19) -2.49 ± 6.72 -0.78 ± 5.21 -1.08 ± 5.90 -1.45 ± 5.21 t = 2.73 p = 0.0099 ADHD group (N = 18) -6.32 ± 5.63 -6.53 ± 7.66 -5.05 ± 6.20 -5.97 ± 4.83 left hemisphere mean [μV] ± standard deviation AF7 FP1 F9 left VLPFC Difference left VLPFC ADHD vs CO Control group (N = 19) 0.26 ±4.84 -0.34 ± 4.60 -0.23 ± 5.81 -0.10 ± 3.66 t = 1.06 p = 0.30 ADHD group (N = 18) -0.70 ± 5.84 -4.49 ± 9.06 -0.60 ± 7.06 -1.93 ± 6.49 PINV amplitudes (mean values and standard deviations) over the right and left ventrolateral prefrontal area (VLPFC) for unilateral response movement with the left hand. CO = healthy control group. Significant differences are presented in bold italics. 3. Medication effects, detailed topographic analysis and hemisphere and at FP1 for the left hemisphere (see comparison of left and right hand response trials Tables 1 and 2). During T2, about 70 minutes after the intake of MPH, Most important, there was an interaction of the factors the PINV amplitude in children with ADHD decreased (RUN x ELECTRODE x GROUP: F(2;34) = 10.1; p = to a normal level (Figure 4). 0.0004). Newman Keuls post-hoc tests showed that this The overall MANOVA model (factors GROUP, SIDE effect has been identified as a consequence of a higher of the response movement, HEMISPHERE, ELECTRODE PINV amplitude especially in FP1/2 in T1 in the ADHD and RUN) yielded an interaction between GROUP, SIDE group. At FP1/2 during T1 there was a difference of the response movement and ELECTRODE (F(2;34) = between the diagnostic groups (p = 0.005). For other 6.0; p = 0.006. This interaction effect was further exam- combinations, e.g. at FP1/2 during T2 (p = 0.34) or at ined in separate MANOVAs for left and right hand AF7/8 during T1 (p = 0.16) the level of significance was response CNV tasks. not reached for any group differences. 3.1 Multivariate analysis of variance (MANOVA) for the Table 3 gives a complete overview over the results unilateral response movement with the left hand with the of the MANOVA for the unilateral response move- factors diagnostic GROUP (ADHD versus healthy control ment with the left and the right hand; showing both children), HEMISPHERE (right versus left), ELECTRODES (AF7/8, significant and non-significant main effects and FP1/2, F9/10) and RUN (T1 versus T2) interactions. A main effect for the factor HEMISPHERE (F(1;35) = 3.2 MANOVA for the unilateral response movement with 12.8, p = 0.001) pointed towards higher PINV ampli- the right hand tudes over the right VLPFC (cf. Tables 1 and 2). A significant interaction of the factors GROUP x ELEC- Furthermore, there was a trend towards an interaction TRODE (F(2;34) = 3.9, p = 0.03) irrespective of the run between GROUP and RUN (F(1;35) = 3.8, p = 0.059). (T1/T2) indicated that the PINV amplitudes of children Newman Keuls post-hoc tests showed that this effect was with ADHD were larger than those of healthy control due to a decrease in PINV amplitude after MPH intake children at FP1/FP2 but not other surrounding electro- in the ADHD group (p = 0.03), which could not be found des: Newman Keuls post-hoc tests showed a significant in the healthy control group (p = 0.85). difference between the two diagnostic groups at the An interaction of the factors HEMISPHERE x ELEC- electrodes FP1/FP2 (p = 0.015), which did not exist for TRODE (F(2;34) = 7.3, p = 0.002) indicated a different the other electrode positions, e.g. at F9/10 (p = 0.99). PINV topography for the left and right hemisphere. The A main effect for the factor RUN (F(1;35) = 6.54, p = strongest negativity was found at AF8 for the right 0.02) pointed towards lower amplitudes at T2. Table 2 Group differences in PINV amplitude for right hand responses right hemisphere mean [μV] ± standard deviation AF8 FP2 F10 right VLPFC Difference right VLPFC ADHD vs CO Control group (N = 19) -3.01 ± 7.25 -3.44 ± 10.14 -3.00 ± 6.35 -3.15 ± 6.62 t = 0.84 p = 0.41 ADHD group (N = 18) -4.76 ± 5.45 -5.42 ± 6.32 -4.02 ± 5.46 -4.73 ± 4.66 left hemisphere mean [μV] ± standard deviation AF7 FP1 F9 left VLPFC Difference left VLPFC ADHD vs CO Control group (N = 19) -1.08 ± 5.83 -0.82 ± 5.20 -2.04 ± 3.79 -1.32 ± 3.83 t = 0.38 p = 0.71 ADHD group (N = 18) -0.66 ± 5.01 -5.81 ± 9.04 0.55 ± 11.12 -1.97 ± 6.48 PINV amplitudes (mean values and standard deviations) over the right and left ventrolateral prefrontal area for unilateral response movements with the right hand. CO = healthy control group. Werner et al. Behavioral and Brain Functions 2011, 7:38 Page 5 of 9 http://www.behavioralandbrainfunctions.com/content/7/1/38 Figure 2 PINV topography: ADHD versus healthy control children. Comparison of the PINV topographies 2000 to 3000 ms after the imperative stimulus S2 in ADHD children versus healthy control children. The activation over the prefrontal cortex is displayed by different shades of grey. Current sinks (negative potential shifts) are presented striped and current sources (positive Figure 1 PINV time course: ADHD versus healthy control potential shifts) without stripes with a scale ranging from -6.5 μVto children. PINV time-course over the left (top) and the right +5 μV. (bottom) ventrolateral prefrontal cortex for unilateral response movement with the left thumb. The potentials of ADHD children are depicted in grey, those of the control-group in black. The drug- right hand responses, there was also an elevated PINV free first run T1 is shown. The vertical dashed line indicates the time amplitude in children with ADHD, but the PINV increase when the auditory warning stimulus S1 occurred, the visual imperative stimulus S2 followed 3 s later. was more limited to leads Fp1/Fp2. Very easy tasks may decrease group differences. The differences in the healthy controls’ PINV amplitudes between left and right hand Another main effect for the factor HEMISPHERE (F (1;35) = 7.11, p = 0.01) pointed towards higher ampli- responses (higher for the right handes) were not statisti- tudes over the right hemisphere. cally significant and thus not further interpreted. b) a normalization of the elevated negativity under Discussion MPH for elevated PINV amplitudes in the left hand Our most important findings were response task, i.e. where the most pronounced group a) a significantly elevated negativity during the PINV differences between unmedicated patients and healthy over the VLPFC in unmedicated children with ADHD in controls had been found. comparison to healthy, age- and gender-matched sub- jects, especially when the unilateral response movement Elevated PINV amplitude as an expression of increased was given with the left hand. Longer reaction times indi- performance uncertainty in children with ADHD cated, that left hand responses represented the slightly Recent electrophysiological studies have suggested defi- more difficult task for our right-handed subjects. For cits in response monitoring and a diminished capacity Werner et al. Behavioral and Brain Functions 2011, 7:38 Page 6 of 9 http://www.behavioralandbrainfunctions.com/content/7/1/38 Figure 3 Evoked EEG response (PINV) versus Electrooculogram (EOG). Comparison of the potential time course over the right ventrolateral prefrontal area in ADHD children versus their electrooculogram (group grand averages). The horizontal and vertical EOG is depicted in black, PINV amplitudes over the right ventrolateral prefrontal cortex (AF8/FP2/F10) in grey. to monitor error responses in children with ADHD, areas interact dynamically with each other and thus represented by decreased amplitudes of the ERN over ensure the permanent self monitoring and adjustment of the ACC, a region which is important for the discrimi- all target-oriented actions [26,27]. nation between stimuli and the monitoring of actions The VLPFC’sroleinparticularisthoughtto be the and errors [25]. processing of negative feedback in order to correct Our finding of a significantly elevated PINV ampli- action with the objective of optimisation of perfor- tude over the VLPFC could be interpreted as a com- mance; it is implicated in contingency detection and in pensatory mechanism in the response monitoring the evaluation of stimuli [28,29]. process. Thus the deficits in error detection could by An overall right-sided preponderance of PINV has been compensated by increased evaluation processing in found also in previous studies [30] and points towards a other brain areas. preferential involvement of the right hemisphere in con- As the ACC and the VLPFC represent important parts tingency evaluation. Apart from differences in task diffi- of a monitoring network, responsible for the evaluation culty, this PINV lateralization could have also played a of the correctness of a given answer and the impairment role for the more pronounced group differences in the of cognitive control in case of failure, the two cortex left hand button press task. Werner et al. Behavioral and Brain Functions 2011, 7:38 Page 7 of 9 http://www.behavioralandbrainfunctions.com/content/7/1/38 Figure 4 Medication effects of methylphenidate. PINV time course over the right ventrolateral prefrontal cortex (AF8/FP2/F10) at T1 and T2 for left hand button press trials. The potentials of the ADHD group are depicted in grey, those of the control-group in black. Since the PINV appears in simple reaction time tasks, found during a stop task in hyperactive adolescents in it seems to be a fundamental mechanism that plays a comparison to healthy peers and it was concluded that role in many goal-directed actions and is not limited to ADHD is associated with subnormal activation of the the two-stimulus situation of the CNV-paradigm. prefrontal systems responsible for higher-order motor The increased compensatory efforts for self-monitoring control [31]. From schizophrenia research, models that and contingency detection, represented by the enhanced include both a compensatory (pre-)frontal hyperactiva- PINV amplitude, may contribute to ADHD children’s tion or a (pre-)frontal hypoactivation depending on the inability to concentrate on relevant stimuli in their difficulty of the examined task, are well established and environment. may explain, why we found an increased PINV over the Using functional magnetic resonance imaging (fMRI), a right VLPFC while fMRI-studies highlight prefrontal pro- lower response in the right mesial prefrontal cortex was cessing deficits in more challenging tasks, despite clear Table 3 MANOVA for left and right hand responses MANOVA for left hand response MANOVA for right hand response Effect F p F p Group 2.42 0.13 1.59 0.22 Run 1.19 0.28 6.54 0.02 Run x Group 3.80 0.06 0.02 0.89 Hemisphere 12.80 0.001 7.11 0.01 Hemisphere x Group 0.98 0.33 0.16 0.69 Electrode 0.65 0.53 1.22 0.31 Electrode x Group 0.16 0.85 3.89 0.03 Run x Hemisphere 0.09 0.77 0.70 0.41 Run x Hemisphere x Group 1.11 0.30 0.09 0.77 Run x Electrode 0.52 0.60 1.34 0.27 Run x Electrode x Group 10.08 0.00037 0.48 0.63 Hemisphere x Electrode 7.31 0.0023 0.56 0.58 Hemisphere x Electrode x Group 0.95 0.40 2.15 0.13 Run x Hemisphere x Electrode 0.69 0.51 1.67 0.32 Run x Hemisphere x Electrode x Group 0.07 0.93 0.94 0.40 Multivariate analysis of variance of PINV amplitude after left and right hand responses examining the factors Hemisphere (left, right), Electrodes (AF7/8, FP1/2, F9/ 10), Run (T1,. T2) and Group (ADHD versus healthy controls). Significant results are presented in bold italics. Werner et al. Behavioral and Brain Functions 2011, 7:38 Page 8 of 9 http://www.behavioralandbrainfunctions.com/content/7/1/38 differences between the deficits in subjects suffering from group could not have shown any change between T1 and schizophrenia and subjects suffering from ADHD. T2 due to a floor effect. It has to be mentioned critically that in the inspection of This interpretation would be consistent with the the single trials we found some examples in which the above-discussed theory of the PINV representing a defi- potentials over the ventrolateral prefrontal area followed cient contingency evaluation. the time-course of the EOG before and after eye artifact In any case, the fact that differences in PINV ampli- correction. In general, PINV occurred independently from tude were reduced and not produced by MPH shows that PINV differences were not due to acute medication the EOG with a different time-course, as illustrated by our effects of MPH (cf. Moll et al. [34]). Further studies may findings in the group grand average findings in Figure 3. After the elimination of all eye artifacts, unfortunately no examine drug-naïve children. reasonable number of trials remained for analysis. There- fore future studies must show the exact extent, to which Conclusions the elevated amplitudes we found are contributed to by In the present study, we examined children with ADHD eye artifacts. ICA (independent component analysis)-based with regard to their slow movement related potentials in ocular correction approaches may yield additional infor- an audio-visual two-stimulus paradigm in comparison to mation. However, our analyses showed that the described age-matched, healthy controls. differences could not be explained sufficiently by eye arti- We found a significant increase of the negative varia- facts in our sample. tion after the target stimulus S2 (PINV) over the right ventrolateral prefrontal cortex area in the ADHD group Interpretation of the normalization of PINV-amplitudes of for left hand responses (the slightly more difficult task). ADHD patients after MPH intake As elaborated above, the detected increase of the pre- MPH, as an indirect dopaminergic agonist, could con- frontal PINV-amplitude can be interpreted as a deficit in ceivably lead to an effect on error awareness, contin- contingency-evaluation representing ADHD children’s gency evaluation and thus to modified PINV-amplitudes: higher uncertainty about the correctness of their own An increased ERN caused by stimulants could be actions caused faulty monitoring processes. found in adult patients [32], however, the intake of The presented results can be used to better understand MPH had no effect on the ERN amplitude in another ADHD children’s specific needs and incertitude. It may study by Groen et al. [6]. help to take another step into creating optimized learning In other studies it was concluded that the inaccurate conditions by reinsurance from the outside by immediate behaviour of ADHD children in conflict tasks might be extremely clear feedback and drawing the child’satten- tion to relevant stimuli to minimize the distraction by related to reduced error-awareness and higher sensitivity to response conflict. The amelioration after the intake of disturbed self-monitoring-processes. MPH was interpreted as its positive influence on brain networks, enabling children with ADHD to allocate more List of abbreviations attention to significant events [33]. ACC: anterior cingulate cortex; ADHD: attention deficit/hyperactivity disorder; MPH’s influence on early error detection, however, CNV: contingent negative variation; EOG: electrooculogram; ERN: error- related negativity; fMRI: functional magnetic resonance imaging; MPH: seems to play a lesser role than the positive effect on methylphenidate; PINV: postimperative negative variation; VLPFC: subsequent processing steps. It has been concluded that ventrolateral prefrontal cortex the effect of MPH on self-monitoring processes is Acknowledgements and Funding mediated rather by the noradrenergic than by the dopa- This work was supported by a grant of the Medical Faculty of the Goethe minergic system [6]. University Frankfurt/Main, Germany to SB (7601002). However, no final statement can be made to which Author details extent MPH has an influence on the amplitude of PINV: Department of Child and Adolescent Psychiatry, University of Heidelberg, Another important point, interpreting the approxima- 2 Blumenstraße 8, D-69115 Heidelberg, Germany. Section for Experimental tion of the PINV-amplitudes in T2, could be learning Psychopathology, Psychiatric Hospital, Voßstraße 4, D-69115 Heidelberg, University of Heidelberg, Germany. SRH-Klinikum Karlsbad-Langensteinbach, effects due to the test repetition. In the second test run Psychiatric Hospital, Guttmannstraße 1, D-76307 Karlsbad, Germany. T2 the paradigm was already familiar to the children. 4 Department of Child and Adolescent Psychiatry, Medical School Carl Gustav They knew the stimulation and what reaction was Carus, Dresden University of Technology, Fetscherstraße 74, D-01307 Dresden, Germany. Department of Child and Adolescent Psychiatry, expected and already practiced it in T1. Moreover, prac- University of Frankfurt, Deutschordenstraße 50, D-60528 Frankfurt am Main, tice effects could differ between ADHD and healthy con- Germany. trol children. Authors’ contributions Accordingly, the increased performance uncertainty JW performed the EEG measurements, conducted large parts of data which we found in T1 could be decreased to a normal analysis and has drafted the manuscript. level (the uncertainty level of controls) in T2. The control Werner et al. Behavioral and Brain Functions 2011, 7:38 Page 9 of 9 http://www.behavioralandbrainfunctions.com/content/7/1/38 MW and VR were involved in data interpretation and manuscript variation) in schizophrenia: their association to the present state and to preparation. Parkinsonian medication effects. Clin Neurophysiol 1999, 110:1175-1192. FR was involved in fund raising, data interpretation and manuscript 16. Diener C, Kuehner C, Brusniak W, Struve M, Flor H: Effects of stressor preparation. controllability on psychophysiological, cognitive and behavioural SB planned the study design, supervised the data acquisition and analysis, responses in patients with major depression and dysthymia. and critically revised the manuscript. Psychological medicine 2009, 39:77-86. All authors have read and approved the final manuscript. 17. Diener C, Kuehner C, Flor H: Loss of control during instrumental learning: a source localization study. NeuroImage 2010, 50:717-726. Competing interests 18. Klein C, Rockstroh B, Cohen R, Berg P, Dressel M: The impact of JW reports no competing interests. performance uncertainty on the postimperative negative variation. MW has recently finished an investigator-initiated trial sponsored by Esparma Psychophysiology 1996, 33:426-433. and signed a contract concerning the development of neuropsychological 19. Oldfield RC: The assessment and analysis of handedness: The Edinburgh diagnostic and training tools with Schuhfried. Inventory. Neuropsychologia 1971, 9:97-113. FR reports no competing interests. 20. Delmo C, Weiffenbach O, Gabriel M, Poustka F: Diagnostisches Interview VR has received honoraries for lectures from Lilly, Medice, Novartis, Shire. He Kiddie-SADS-Present and Lifetime Version. 5. Aufl. der deutschen has been member of an advisory board of Lilly, Novartis. He has received Forschungsversion Frankfurt: Universität Frankfurt; 2000. research support from Novartis. 21. Nagai Y, Critchley HD, Rothwell JC, Duncan JS, Trimble MR: Changes in SB has received honoraries for lectures from Novartis and support for cortical potential associated with modulation of peripheral sympathetic symposia by Shire. activity in patients with epilepsy. Psychosom Med 2009, 71:84-92. 22. Bender S, Weisbrod M, Resch F, Oelkers-Ax R: Stereotyped topography of Received: 23 March 2011 Accepted: 25 August 2011 different elevated contingent negative variation components in children Published: 25 August 2011 with migraine without aura points towards a subcortical dysfunction. Pain 2007, 127:221-233. 23. Schallberger U: Welches sind nach statistischen Kriterien die besten References Unterformen des HAWIK-III ? Zürich: psychologisches Institut der Universität;; 1. Polanczyk G, de Lima MS, Horta BL, Biederman J, Rohde LA: The worldwide prevalence of ADHD: a systematic review and metaregression analysis. 24. Quinn D, Bode T, Reiz JL, Donnelly GA, Darke AC: Single-dose Am J Psychiatry 2007, 164:942-948. pharmacokinetics of multilayer-release methylphenidate and immediate- 2. Sonuga-Barke EJ: Psychological heterogeneity in AD/HD–a dual pathway release methylphenidate in children with attention-deficit/hyperactivity model of behaviour and cognition. Behavioural brain research 2002, disorder. Journal of clinical pharmacology 2007, 47:760-766. 130:29-36. 25. Bush G, Luu P, Posner MI: Cognitive and emotional influences in anterior 3. Barry RJ, Johnstone SJ, Clarke AR: A review of electrophysiology in cingulate cortex. Trends Cogn Sci 2000, 4:215-222. attention-deficit/hyperactivity disorder: II. Event-related potentials. Clin 26. Cavanagh JF, Cohen MX, Allen JJ: Prelude to and resolution of an error: Neurophysiol 2003, 114:184-198. EEG phase synchrony reveals cognitive control dynamics during action 4. Banaschewski T, Brandeis D: Annotation: what electrical brain activity tells monitoring. J Neurosci 2009, 29:98-105. us about brain function that other techniques cannot tell us - a child 27. Botvinick MM, Braver TS, Barch DM, Carter CS, Cohen JD: Conflict psychiatric perspective. Journal of child psychology and psychiatry, and monitoring and cognitive control. Psychological review 2001, 108:624-652. allied disciplines 2007, 48:415-435. 28. Cools R, Clark L, Owen AM, Robbins TW: Defining the neural mechanisms 5. van Meel CS, Heslenfeld DJ, Oosterlaan J, Sergeant JA: Adaptive control of probabilistic reversal learning using event-related functional magnetic deficits in attention-deficit/hyperactivity disorder (ADHD): the role of resonance imaging. J Neurosci 2002, 22:4563-4567. error processing. Psychiatry research 2007, 151:211-220. 29. van Leijenhorst L, Crone EA, Bunge SA: Neural correlates of 6. Groen Y, Wijers AA, Mulder LJ, Waggeveld B, Minderaa RB, Althaus M: Error developmental differences in risk estimation and feedback processing. and feedback processing in children with ADHD and children with Neuropsychologia 2006, 44:2158-2170. Autistic Spectrum Disorder: an EEG event-related potential study. Clin 30. Klein C, Andresen B, Berg P, Krüger H, Rockstroh B: Topography of CNV Neurophysiol 2008, 119:2476-2493. and PINV in Schizotypal Personality. Psychophysiology 1998, 35:272-282. 7. Albrecht B, Brandeis D, Uebel H, Heinrich H, Mueller UC, Hasselhorn M, 31. Rubia K, Overmeyer S, Taylor E, Brammer M, Williams SC, Simmons A, Steinhausen HC, Rothenberger A, Banaschewski T: Action monitoring in Bullmore ET: Hypofrontality in attention deficit hyperactivity disorder boys with attention-deficit/hyperactivity disorder, their nonaffected during higher-order motor control: a study with functional MRI. Am J siblings, and normal control subjects: evidence for an endophenotype. Psychiatry 1999, 156:891-896. Biological psychiatry 2008, 64:615-625. 32. de Bruijn ER, Hulstijn W, Verkes RJ, Ruigt GS, Sabbe BG: Drug-induced 8. Shiels K, Hawk LW Jr: Self-regulation in ADHD: The role of error stimulation and suppression of action monitoring in healthy volunteers. processing. Clin Psychol Rev 2010. Psychopharmacology 2004, 177:151-160. 9. Bender S, Becker D, Oelkers-Ax R, Weisbrod M: Cortical motor areas are 33. Jonkman LM, van Melis JJ, Kemner C, Markus CR: Methylphenidate activated early in a characteristic sequence during post-movement improves deficient error evaluation in children with ADHD: an event- processing. NeuroImage 2006, 32:333-351. related brain potential study. Biological psychology 2007, 76:217-229. 10. Yordanova J, Dumais-Huber C, Rothenberger A, Woerner W: Frontocortical 34. Moll GH, Heinrich H, Rothenberger A: [Transcranial magnetic stimulation activity in children with comorbidity of tic disorder and attention-deficit in child and adolescent psychiatry: excitability of the motor system in hyperactivity disorder. Biological psychiatry 1997, 41:585-594. tic disorders and/or attention deficit hyperactivity disorders]. Z Kinder 11. Elbert T, Rockstroh B, Lutzenberger W, Birbaumer N: Slow brain potentials Jugendpsychiatr Psychother 2001, 29:312-323. after withdrawal of control. Archiv fur Psychiatrie und Nervenkrankheiten 1982, 232:201-214. doi:10.1186/1744-9081-7-38 12. Bolz J, Giedke H: Controllability of an aversive stimulus in depressed Cite this article as: Werner et al.: Increased performance uncertainty in patients and health controls: a study using slow brain potentials. children with ADHD? - Elevated post-imperative negative variation (PINV) Biological psychiatry 1981, 16:441-452. over the ventrolateral prefrontal cortex. Behavioral and Brain Functions 2011 13. Kathmann N, Jonitz L, Engel RR: Cognitive determinants of the 7:38. postimperative negative variation. Psychophysiology 1990, 27:256-263. 14. Klein C, Rockstroh B, Cohen R, Berg P: Contingent negative variation (CNV) and determinants of the post-imperative negative variation (PINV) in schizophrenic patients and healthy controls. Schizophrenia research 1996, 21:97-110. 15. Verleger R, Wascher E, Arolt V, Daase C, Strohm A, Kompf D: Slow EEG potentials (contingent negative variation and post-imperative negative http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Behavioral and Brain Functions Springer Journals

Increased performance uncertainty in children with ADHD? - Elevated post-imperative negative variation (PINV) over the ventrolateral prefrontal cortex

Download PDF
9 pages

Loading next page...
 
/lp/springer-journals/increased-performance-uncertainty-in-children-with-adhd-elevated-post-mIFkmEgTJl

References (74)

Publisher
Springer Journals
Copyright
Copyright © 2011 by Werner et al; licensee BioMed Central Ltd.
Subject
Biomedicine; Neurosciences; Neurology; Behavioral Therapy; Psychiatry
eISSN
1744-9081
DOI
10.1186/1744-9081-7-38
pmid
21867487
Publisher site
See Article on Publisher Site

Abstract

Background: We aimed to investigate the influences of attention deficit/hyperactivity disorder (ADHD) on response evaluation, as reflected by the postimperative negative variation (PINV), a slow event-related potential. Methods: We investigated PINV as an indicator of performance uncertainty in an audio-visual contingent negative variation (CNV) paradigm with an interstimulus interval of 3 seconds. A constant, unilateral, quick motor reaction with either the right or the left thumb was required after an auditory forewarned (S1) visual imperative stimulus (S2). We examined 18 ADHD patients (combined or hyperactive-impulsive subtype) aged between 8 and 14 years and an age-, sex and IQ-matched control group of 19 healthy subjects using 64-channel high-density EEG. A first run was recorded drug-free, a second one under methylphenidate (MPH) medication in the ADHD group. Results: We found a significantly increased negativity of the PINV-component over the ventrolateral prefrontal cortex in ADHD children compared to the healthy control group. PINV amplitude was influenced by movement side, most likely due to the slightly more difficult task when left hand responses were required. After the intake of MPH, PINV amplitudes of ADHD children normalized. Conclusions: We conclude that children with ADHD are likely to be more uncertain about the correctness of their performance and interpret the increased PINV as a hint towards compensatory mechanisms for a deficit in the evaluation of contingencies. Further studies are needed to assess the exact extent to which remainders of eye- movement related potentials contribute to PINV amplitude despite the correction for eye-artifacts. Background Certain event-related potentials have been discussed as Attention deficit/hyperactivity disorder (ADHD) is one of markers for the disorder but previous studies have the most common [1] and at the same time still not com- pointed out heterogeneous neurophysiological profiles in pletely pathophysiologically understood child psychiatric ADHD patients [3,4]. diagnoses. Recently, a decreased error related negativity (ERN) over Clear deficits in executive functions like planning, the anterior cingulate cortex (ACC) has been interpreted inhibition and evaluation of movement have been found. as ADHD children’s diminished capacity to monitor their error responses and their failure to predict the likelihood On the other hand, it is still controversially discussed to which extent motivational aspects and deficits in delay that an error occurs in a given context [5-7]. aversion are responsible for the development of ADHD- These findings suggest deficits in the children’s cogni- typical symptoms [2]. tive processing of movement caused by diminished internal monitoring processes [8]. * Correspondence: Stephan.Bender@uniklinikum-dresden.de In the current study we chose the PINV (postimpera- Department of Child and Adolescent Psychiatry, Medical School Carl Gustav tive negative variation) component as another important Carus, Dresden University of Technology, Fetscherstraße 74, D-01307 marker of movement/action monitoring processes, Dresden, Germany Full list of author information is available at the end of the article © 2011 Werner et al; licensee BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. Werner et al. Behavioral and Brain Functions 2011, 7:38 Page 2 of 9 http://www.behavioralandbrainfunctions.com/content/7/1/38 representing the individual’s uncertainty about the cor- we included children treated with both immediate and rectness of a given answer [9], aimed to further investi- extended release MPH in our patient group. gate the disorders’ influence on self-monitoring and the Twelve out of 18 ADHD children were treated with establishment of stable contingencies between stimuli extended release MPH with a mean dosage of 0.85 mg/kg and the corresponding required responses [10]. body weight (0.25 to 1.29 mg/kg), the other six with Enhanced PINV amplitudes have been found in a num- immediate release MPH with a mean dosage of 0.4 mg/kg ber of studies, reflecting a lack of control over aversive (0.15 to 0.74 mg/kg). events, an unexpected change in controllability [11,12] Nineteen right-handed (EHI laterality quotient 97.3 ± 5.2) healthy, age-, gender- and IQ-matched children and and representing contingency reappraisal [13]. Especially schizophrenic [14,15] and depressive [16] adolescents (14 males and 5 females, mean age 11.6 ± 2.1 individuals show elevated PINV amplitudes, representing years, mean IQ 117.4 ± 13.0), who took no psychoactive the uncertainty about the appropriate response [17]. medication and did not suffer from any neurological or PINV amplitude is sensitive to ambiguous contingen- psychiatric symptoms, were recruited as control group at cies and is thought to reflect an unexpected outcome, Heidelberg’s elementary and secondary schools. which causes performance uncertainty [18]. PINV usually In both groups we screened for visual impairments shows a (pre-)frontal maximum, so generators in the pre- (corrected visus ≥ 0.8). frontal cortex have been postulated [18]. A topographic All subjects and their parents provided written informed multi-channel analysis of PINV in ADHD has not been consent according to the Declaration of Helsinki and the performed so far. study was approved by the local ethics committee. We hypothesized that � ADHD children’s contingency evaluation and their Task/recording/data pre-processing cognitive performance monitoring is disturbed, resulting We recorded a CNV paradigm, using an auditory warning in an increased PINV amplitude of ADHD children stimulus S1 (1000 Hz, 90 dB, 50ms duration) and a visual compared to age-matched healthy controls. imperative stimulus S2 (image of a white hand, pointing � Methylphenidate (MPH) has a positive influence on towards the side of the required button press, presented the cognitive evaluation represented by a normalized for 150 ms on a black screen). The interstimulus interval PINV-amplitude after MPH-intake in the ADHD group. was 3 seconds, intertrial intervals varied randomly from 7 to 11 seconds. Methods Subjects were instructed to correctly respond as fast as Subjects possible when S2 occurred on the screen by pressing a We analyzed eighteen right-handed (Edinburgh Handed- button on the STIM response pad (Neuroscan Inc, TX, ness Inventory; EHI; [19]; laterality quotient mean USA) with the thumb of either the right or the left hand value ± standard deviation 94.9 ± 10.2) children between (quick, unilateral motor answer). 8 and 14 years (13 males and 5 females, mean age ± stan- 40 trials per hand were recorded in a counterbalanced dard deviation 11.5 ± 1.9 years, mean IQ ± standard order across subjects. Two runs were recorded: In the deviation 110.5 ± 18.8) who met the criteria of a hyperac- control group both runs, T1 and T2, were drug-free. In tive-impulsive or combined subtype of ADHD according the ADHD group, the first one (T1) was drug-free (after at to the semi-structured interview for DSM, K-SADS [20]. least 24 hours after the last intake of MPH), the second All patients were recruited either in the Child and Ado- one (T2) after 70 minutes after the intake of the individual lescent Psychiatric Department of the University of used dose of MPH. In other studies the same experimental Heidelberg or at a child psychiatrist’s practice, were trea- period of 70 minutes after the intake of MPH was chosen, ted with multilayer-release or immediate-release MPH so comparability is ensured. without other co-medication and suffered from no other Participants fixated a cross on a computer screen in psychiatric diseases. This includes that we assured that order to minimize eye artifacts. Neuroscan Synamp there were no neuropsychiatric disorders such as psy- Amplifiers (Neuroscan Inc., USA) were used to record choses and autism or neurological diseases as epilepsy continuous DC 64-channel EEG with a sampling rate [21], migraine [22] and tic-disorder [10], which are of 250 Hz. An anti-aliasing filter was set at 70 Hz thought to lead to specific changes in contingent negative (low-pass). Surface Ag-AgCl sintered electrodes were variation (CNV) parameters. fixed using an equidistant electrode cap (Easycap, An IQ below 80 (4-subtest short version of HAWIK FMS, Germany) and are named according to an III [23]), led to exclusion from the study. extended international 10-20 system. The vertical and As Quinn et al. [24] found no significant difference horizontal electrooculogram (EOG) was recorded by between the concentration of multilayer- and immediate- electrodes 1 cm next to the outer canthi and above/ release MPH within the first four hours after the intake, below the left eye. Werner et al. Behavioral and Brain Functions 2011, 7:38 Page 3 of 9 http://www.behavioralandbrainfunctions.com/content/7/1/38 Impedances were kept below 5 kΩ. Data were recorded 305 ± 63 ms (right hand button presses) and 320 ± 81 ms against a reference near Cz and transformed offline to (left hand button presses) for healthy control children. average reference. Recordings 1 s before S1 served as base- A repeated measurements ANOVA with the factors line. For the analysis of PINV, the EEG-signal was digitally SIDE of the response movement and GROUP (ADHD filtered (30 Hz high cut-off), segmented into epochs of versus healthy controls) revealed shorter reaction times 7.5 s (1 s pre S1 to 3.5 s post S2), corrected automatically for responses with the dominant right than with the left for DC-drifts by linear regression (Brain Vision Analyzer, hand (F(1;35) = 5.9; p = 0.02). In contrast, GROUP had Brain Products GmbH, Germany), and for eye movements no significant main effect on reaction times (F(1;35) = and blinks (algorithm according to Gratton and Coles as 0.3; p > 0.59) and did not interact with response move- implemented in Brain Vision Analyzer Version 1). ment side either (F(1;35) = 0.0; p > 0.90). Artifacts were rejected automatically if the signal 1.2 EEG data - PINV amplitude exceeded 150 mV. This procedure was con- Group differences of the PINV amplitudes over the ven- firmed by visual inspection; only artifact free trials trolateral prefrontal areas between unmedicated ADHD entered further analysis. Bad channels were interpolated patients and the control group (T1) are presented for using nearest neighbours. Trials were rejected from left (Table 1) and right hand response movements further analysis if subjects responded with the wrong (Table 2). Mean values and standard deviations as well hand or after more than 3.5 s after S2. as the results of the four t-tests are shown. Two ADHD patients and one control child (out of ori- We found significantly elevated PINV amplitudes over ginally n = 20 children in both groups) had to be the right ventrolateral prefrontal cortex (VLPFC) in the excluded from further evaluation due to recording ADHD group in comparison to the healthy control group, errors or excessive artifact-prone data. N = 18 ADHD when the unilateral motor response was given by the left children and n = 19 controls were included for further hand (p = 0.01; t = 2.7). For the ipsilateral, left prefrontal statistical analysis. area there was no significant difference (p = 0.30, t = 1.1). Figure1 shows thetimecourse of the prefrontal PINV Data analysis/statistics amplitudes separately for each hemisphere when the uni- As a first step, planned comparisons for group differences lateral response movement is given with the left hand. between the PINV-amplitudes of unmedicated ADHD For the unilateral response movement with the right versus control children over the left and right ventrolat- hand, there were no significant group differences (cf. eral prefrontal areas (pooled leads AF7, FP1, F9 and AF8, Tables 1 and 2). FP2, F10 during the time interval 2000 to 3000 ms after The topographical analysis of the cortical activation 2000 to 3000 ms after the target stimulus S2 is shown the imperative stimulus S2 in agreement with results of our previous study [18]) were examined for right and left in Figure 2 (reference-free current source density maps), hand button presses by four t-tests. The significance level illustrating the above-described group differences. Irre- was set to p = 0.05/4 = 0.0125 (Bonferroni correction). spective of the side of the response movement, a higher Next, in order to assess the influence of medication on right-sided negativity over ventrolateral prefrontal areas PINV topography in more detail, results were examined during PINV is obvious, although the lateralization of by multivariate analysis of variance (MANOVA), using the the activation is noticeably weaker when the response between subject factor GROUP (ADHD versus healthy movement is given by the right thumb. controls) and the within subject factors SIDE of the response movement (left vs. right hand), HEMISPHERE 2. Separation of PINV and eye movement and blink (left vs. right VLPFC), RUN (T1 vs. T2) and ELECTRO- artifacts DES (AF7/8, FP1/2, F9/10) followed by simpler separate In addition to the eye movement correction, we performed MANOVAs for left and right hand response CNV tasks. a comparison between the time-course of the electroocula- Significant main effects or interactions in the MANOVA gram and the PINV amplitudes, which revealed an inde- were subsequently further examined by post-hoc tests pendent time course as shown in Figure 3. (Newman Keuls). Although quite a lot of blink or eye movement arti- facts occurred during the PINV interval (which made a Results complete removal of all trials with blink artifacts impos- 1. Group differences between drug-free ADHD patients sible), the visual examination of the single trials also and control children (medication-free first run) confirmed that the time-course of the potentials in the 1.1 Behavioral data - reaction times leads over the ventrolateral prefrontal areas could not Mean reaction times (± standard deviation) were 317 ± be explained by remainders of insufficiently corrected 65 ms (right hand button presses) and 333 ± 84 ms (left eye artifacts, as the time-courses differed in single trials hand button presses) for children with ADHD as well as as well. Werner et al. Behavioral and Brain Functions 2011, 7:38 Page 4 of 9 http://www.behavioralandbrainfunctions.com/content/7/1/38 Table 1 Group differences in PINV amplitude for left hand responses right hemisphere mean [μV] ± standard deviation AF8 FP2 F10 right VLPFC Difference right VLPFC ADHD vs CO Control group (N = 19) -2.49 ± 6.72 -0.78 ± 5.21 -1.08 ± 5.90 -1.45 ± 5.21 t = 2.73 p = 0.0099 ADHD group (N = 18) -6.32 ± 5.63 -6.53 ± 7.66 -5.05 ± 6.20 -5.97 ± 4.83 left hemisphere mean [μV] ± standard deviation AF7 FP1 F9 left VLPFC Difference left VLPFC ADHD vs CO Control group (N = 19) 0.26 ±4.84 -0.34 ± 4.60 -0.23 ± 5.81 -0.10 ± 3.66 t = 1.06 p = 0.30 ADHD group (N = 18) -0.70 ± 5.84 -4.49 ± 9.06 -0.60 ± 7.06 -1.93 ± 6.49 PINV amplitudes (mean values and standard deviations) over the right and left ventrolateral prefrontal area (VLPFC) for unilateral response movement with the left hand. CO = healthy control group. Significant differences are presented in bold italics. 3. Medication effects, detailed topographic analysis and hemisphere and at FP1 for the left hemisphere (see comparison of left and right hand response trials Tables 1 and 2). During T2, about 70 minutes after the intake of MPH, Most important, there was an interaction of the factors the PINV amplitude in children with ADHD decreased (RUN x ELECTRODE x GROUP: F(2;34) = 10.1; p = to a normal level (Figure 4). 0.0004). Newman Keuls post-hoc tests showed that this The overall MANOVA model (factors GROUP, SIDE effect has been identified as a consequence of a higher of the response movement, HEMISPHERE, ELECTRODE PINV amplitude especially in FP1/2 in T1 in the ADHD and RUN) yielded an interaction between GROUP, SIDE group. At FP1/2 during T1 there was a difference of the response movement and ELECTRODE (F(2;34) = between the diagnostic groups (p = 0.005). For other 6.0; p = 0.006. This interaction effect was further exam- combinations, e.g. at FP1/2 during T2 (p = 0.34) or at ined in separate MANOVAs for left and right hand AF7/8 during T1 (p = 0.16) the level of significance was response CNV tasks. not reached for any group differences. 3.1 Multivariate analysis of variance (MANOVA) for the Table 3 gives a complete overview over the results unilateral response movement with the left hand with the of the MANOVA for the unilateral response move- factors diagnostic GROUP (ADHD versus healthy control ment with the left and the right hand; showing both children), HEMISPHERE (right versus left), ELECTRODES (AF7/8, significant and non-significant main effects and FP1/2, F9/10) and RUN (T1 versus T2) interactions. A main effect for the factor HEMISPHERE (F(1;35) = 3.2 MANOVA for the unilateral response movement with 12.8, p = 0.001) pointed towards higher PINV ampli- the right hand tudes over the right VLPFC (cf. Tables 1 and 2). A significant interaction of the factors GROUP x ELEC- Furthermore, there was a trend towards an interaction TRODE (F(2;34) = 3.9, p = 0.03) irrespective of the run between GROUP and RUN (F(1;35) = 3.8, p = 0.059). (T1/T2) indicated that the PINV amplitudes of children Newman Keuls post-hoc tests showed that this effect was with ADHD were larger than those of healthy control due to a decrease in PINV amplitude after MPH intake children at FP1/FP2 but not other surrounding electro- in the ADHD group (p = 0.03), which could not be found des: Newman Keuls post-hoc tests showed a significant in the healthy control group (p = 0.85). difference between the two diagnostic groups at the An interaction of the factors HEMISPHERE x ELEC- electrodes FP1/FP2 (p = 0.015), which did not exist for TRODE (F(2;34) = 7.3, p = 0.002) indicated a different the other electrode positions, e.g. at F9/10 (p = 0.99). PINV topography for the left and right hemisphere. The A main effect for the factor RUN (F(1;35) = 6.54, p = strongest negativity was found at AF8 for the right 0.02) pointed towards lower amplitudes at T2. Table 2 Group differences in PINV amplitude for right hand responses right hemisphere mean [μV] ± standard deviation AF8 FP2 F10 right VLPFC Difference right VLPFC ADHD vs CO Control group (N = 19) -3.01 ± 7.25 -3.44 ± 10.14 -3.00 ± 6.35 -3.15 ± 6.62 t = 0.84 p = 0.41 ADHD group (N = 18) -4.76 ± 5.45 -5.42 ± 6.32 -4.02 ± 5.46 -4.73 ± 4.66 left hemisphere mean [μV] ± standard deviation AF7 FP1 F9 left VLPFC Difference left VLPFC ADHD vs CO Control group (N = 19) -1.08 ± 5.83 -0.82 ± 5.20 -2.04 ± 3.79 -1.32 ± 3.83 t = 0.38 p = 0.71 ADHD group (N = 18) -0.66 ± 5.01 -5.81 ± 9.04 0.55 ± 11.12 -1.97 ± 6.48 PINV amplitudes (mean values and standard deviations) over the right and left ventrolateral prefrontal area for unilateral response movements with the right hand. CO = healthy control group. Werner et al. Behavioral and Brain Functions 2011, 7:38 Page 5 of 9 http://www.behavioralandbrainfunctions.com/content/7/1/38 Figure 2 PINV topography: ADHD versus healthy control children. Comparison of the PINV topographies 2000 to 3000 ms after the imperative stimulus S2 in ADHD children versus healthy control children. The activation over the prefrontal cortex is displayed by different shades of grey. Current sinks (negative potential shifts) are presented striped and current sources (positive Figure 1 PINV time course: ADHD versus healthy control potential shifts) without stripes with a scale ranging from -6.5 μVto children. PINV time-course over the left (top) and the right +5 μV. (bottom) ventrolateral prefrontal cortex for unilateral response movement with the left thumb. The potentials of ADHD children are depicted in grey, those of the control-group in black. The drug- right hand responses, there was also an elevated PINV free first run T1 is shown. The vertical dashed line indicates the time amplitude in children with ADHD, but the PINV increase when the auditory warning stimulus S1 occurred, the visual imperative stimulus S2 followed 3 s later. was more limited to leads Fp1/Fp2. Very easy tasks may decrease group differences. The differences in the healthy controls’ PINV amplitudes between left and right hand Another main effect for the factor HEMISPHERE (F (1;35) = 7.11, p = 0.01) pointed towards higher ampli- responses (higher for the right handes) were not statisti- tudes over the right hemisphere. cally significant and thus not further interpreted. b) a normalization of the elevated negativity under Discussion MPH for elevated PINV amplitudes in the left hand Our most important findings were response task, i.e. where the most pronounced group a) a significantly elevated negativity during the PINV differences between unmedicated patients and healthy over the VLPFC in unmedicated children with ADHD in controls had been found. comparison to healthy, age- and gender-matched sub- jects, especially when the unilateral response movement Elevated PINV amplitude as an expression of increased was given with the left hand. Longer reaction times indi- performance uncertainty in children with ADHD cated, that left hand responses represented the slightly Recent electrophysiological studies have suggested defi- more difficult task for our right-handed subjects. For cits in response monitoring and a diminished capacity Werner et al. Behavioral and Brain Functions 2011, 7:38 Page 6 of 9 http://www.behavioralandbrainfunctions.com/content/7/1/38 Figure 3 Evoked EEG response (PINV) versus Electrooculogram (EOG). Comparison of the potential time course over the right ventrolateral prefrontal area in ADHD children versus their electrooculogram (group grand averages). The horizontal and vertical EOG is depicted in black, PINV amplitudes over the right ventrolateral prefrontal cortex (AF8/FP2/F10) in grey. to monitor error responses in children with ADHD, areas interact dynamically with each other and thus represented by decreased amplitudes of the ERN over ensure the permanent self monitoring and adjustment of the ACC, a region which is important for the discrimi- all target-oriented actions [26,27]. nation between stimuli and the monitoring of actions The VLPFC’sroleinparticularisthoughtto be the and errors [25]. processing of negative feedback in order to correct Our finding of a significantly elevated PINV ampli- action with the objective of optimisation of perfor- tude over the VLPFC could be interpreted as a com- mance; it is implicated in contingency detection and in pensatory mechanism in the response monitoring the evaluation of stimuli [28,29]. process. Thus the deficits in error detection could by An overall right-sided preponderance of PINV has been compensated by increased evaluation processing in found also in previous studies [30] and points towards a other brain areas. preferential involvement of the right hemisphere in con- As the ACC and the VLPFC represent important parts tingency evaluation. Apart from differences in task diffi- of a monitoring network, responsible for the evaluation culty, this PINV lateralization could have also played a of the correctness of a given answer and the impairment role for the more pronounced group differences in the of cognitive control in case of failure, the two cortex left hand button press task. Werner et al. Behavioral and Brain Functions 2011, 7:38 Page 7 of 9 http://www.behavioralandbrainfunctions.com/content/7/1/38 Figure 4 Medication effects of methylphenidate. PINV time course over the right ventrolateral prefrontal cortex (AF8/FP2/F10) at T1 and T2 for left hand button press trials. The potentials of the ADHD group are depicted in grey, those of the control-group in black. Since the PINV appears in simple reaction time tasks, found during a stop task in hyperactive adolescents in it seems to be a fundamental mechanism that plays a comparison to healthy peers and it was concluded that role in many goal-directed actions and is not limited to ADHD is associated with subnormal activation of the the two-stimulus situation of the CNV-paradigm. prefrontal systems responsible for higher-order motor The increased compensatory efforts for self-monitoring control [31]. From schizophrenia research, models that and contingency detection, represented by the enhanced include both a compensatory (pre-)frontal hyperactiva- PINV amplitude, may contribute to ADHD children’s tion or a (pre-)frontal hypoactivation depending on the inability to concentrate on relevant stimuli in their difficulty of the examined task, are well established and environment. may explain, why we found an increased PINV over the Using functional magnetic resonance imaging (fMRI), a right VLPFC while fMRI-studies highlight prefrontal pro- lower response in the right mesial prefrontal cortex was cessing deficits in more challenging tasks, despite clear Table 3 MANOVA for left and right hand responses MANOVA for left hand response MANOVA for right hand response Effect F p F p Group 2.42 0.13 1.59 0.22 Run 1.19 0.28 6.54 0.02 Run x Group 3.80 0.06 0.02 0.89 Hemisphere 12.80 0.001 7.11 0.01 Hemisphere x Group 0.98 0.33 0.16 0.69 Electrode 0.65 0.53 1.22 0.31 Electrode x Group 0.16 0.85 3.89 0.03 Run x Hemisphere 0.09 0.77 0.70 0.41 Run x Hemisphere x Group 1.11 0.30 0.09 0.77 Run x Electrode 0.52 0.60 1.34 0.27 Run x Electrode x Group 10.08 0.00037 0.48 0.63 Hemisphere x Electrode 7.31 0.0023 0.56 0.58 Hemisphere x Electrode x Group 0.95 0.40 2.15 0.13 Run x Hemisphere x Electrode 0.69 0.51 1.67 0.32 Run x Hemisphere x Electrode x Group 0.07 0.93 0.94 0.40 Multivariate analysis of variance of PINV amplitude after left and right hand responses examining the factors Hemisphere (left, right), Electrodes (AF7/8, FP1/2, F9/ 10), Run (T1,. T2) and Group (ADHD versus healthy controls). Significant results are presented in bold italics. Werner et al. Behavioral and Brain Functions 2011, 7:38 Page 8 of 9 http://www.behavioralandbrainfunctions.com/content/7/1/38 differences between the deficits in subjects suffering from group could not have shown any change between T1 and schizophrenia and subjects suffering from ADHD. T2 due to a floor effect. It has to be mentioned critically that in the inspection of This interpretation would be consistent with the the single trials we found some examples in which the above-discussed theory of the PINV representing a defi- potentials over the ventrolateral prefrontal area followed cient contingency evaluation. the time-course of the EOG before and after eye artifact In any case, the fact that differences in PINV ampli- correction. In general, PINV occurred independently from tude were reduced and not produced by MPH shows that PINV differences were not due to acute medication the EOG with a different time-course, as illustrated by our effects of MPH (cf. Moll et al. [34]). Further studies may findings in the group grand average findings in Figure 3. After the elimination of all eye artifacts, unfortunately no examine drug-naïve children. reasonable number of trials remained for analysis. There- fore future studies must show the exact extent, to which Conclusions the elevated amplitudes we found are contributed to by In the present study, we examined children with ADHD eye artifacts. ICA (independent component analysis)-based with regard to their slow movement related potentials in ocular correction approaches may yield additional infor- an audio-visual two-stimulus paradigm in comparison to mation. However, our analyses showed that the described age-matched, healthy controls. differences could not be explained sufficiently by eye arti- We found a significant increase of the negative varia- facts in our sample. tion after the target stimulus S2 (PINV) over the right ventrolateral prefrontal cortex area in the ADHD group Interpretation of the normalization of PINV-amplitudes of for left hand responses (the slightly more difficult task). ADHD patients after MPH intake As elaborated above, the detected increase of the pre- MPH, as an indirect dopaminergic agonist, could con- frontal PINV-amplitude can be interpreted as a deficit in ceivably lead to an effect on error awareness, contin- contingency-evaluation representing ADHD children’s gency evaluation and thus to modified PINV-amplitudes: higher uncertainty about the correctness of their own An increased ERN caused by stimulants could be actions caused faulty monitoring processes. found in adult patients [32], however, the intake of The presented results can be used to better understand MPH had no effect on the ERN amplitude in another ADHD children’s specific needs and incertitude. It may study by Groen et al. [6]. help to take another step into creating optimized learning In other studies it was concluded that the inaccurate conditions by reinsurance from the outside by immediate behaviour of ADHD children in conflict tasks might be extremely clear feedback and drawing the child’satten- tion to relevant stimuli to minimize the distraction by related to reduced error-awareness and higher sensitivity to response conflict. The amelioration after the intake of disturbed self-monitoring-processes. MPH was interpreted as its positive influence on brain networks, enabling children with ADHD to allocate more List of abbreviations attention to significant events [33]. ACC: anterior cingulate cortex; ADHD: attention deficit/hyperactivity disorder; MPH’s influence on early error detection, however, CNV: contingent negative variation; EOG: electrooculogram; ERN: error- related negativity; fMRI: functional magnetic resonance imaging; MPH: seems to play a lesser role than the positive effect on methylphenidate; PINV: postimperative negative variation; VLPFC: subsequent processing steps. It has been concluded that ventrolateral prefrontal cortex the effect of MPH on self-monitoring processes is Acknowledgements and Funding mediated rather by the noradrenergic than by the dopa- This work was supported by a grant of the Medical Faculty of the Goethe minergic system [6]. University Frankfurt/Main, Germany to SB (7601002). However, no final statement can be made to which Author details extent MPH has an influence on the amplitude of PINV: Department of Child and Adolescent Psychiatry, University of Heidelberg, Another important point, interpreting the approxima- 2 Blumenstraße 8, D-69115 Heidelberg, Germany. Section for Experimental tion of the PINV-amplitudes in T2, could be learning Psychopathology, Psychiatric Hospital, Voßstraße 4, D-69115 Heidelberg, University of Heidelberg, Germany. SRH-Klinikum Karlsbad-Langensteinbach, effects due to the test repetition. In the second test run Psychiatric Hospital, Guttmannstraße 1, D-76307 Karlsbad, Germany. T2 the paradigm was already familiar to the children. 4 Department of Child and Adolescent Psychiatry, Medical School Carl Gustav They knew the stimulation and what reaction was Carus, Dresden University of Technology, Fetscherstraße 74, D-01307 Dresden, Germany. Department of Child and Adolescent Psychiatry, expected and already practiced it in T1. Moreover, prac- University of Frankfurt, Deutschordenstraße 50, D-60528 Frankfurt am Main, tice effects could differ between ADHD and healthy con- Germany. trol children. Authors’ contributions Accordingly, the increased performance uncertainty JW performed the EEG measurements, conducted large parts of data which we found in T1 could be decreased to a normal analysis and has drafted the manuscript. level (the uncertainty level of controls) in T2. The control Werner et al. Behavioral and Brain Functions 2011, 7:38 Page 9 of 9 http://www.behavioralandbrainfunctions.com/content/7/1/38 MW and VR were involved in data interpretation and manuscript variation) in schizophrenia: their association to the present state and to preparation. Parkinsonian medication effects. Clin Neurophysiol 1999, 110:1175-1192. FR was involved in fund raising, data interpretation and manuscript 16. Diener C, Kuehner C, Brusniak W, Struve M, Flor H: Effects of stressor preparation. controllability on psychophysiological, cognitive and behavioural SB planned the study design, supervised the data acquisition and analysis, responses in patients with major depression and dysthymia. and critically revised the manuscript. Psychological medicine 2009, 39:77-86. All authors have read and approved the final manuscript. 17. Diener C, Kuehner C, Flor H: Loss of control during instrumental learning: a source localization study. NeuroImage 2010, 50:717-726. Competing interests 18. Klein C, Rockstroh B, Cohen R, Berg P, Dressel M: The impact of JW reports no competing interests. performance uncertainty on the postimperative negative variation. MW has recently finished an investigator-initiated trial sponsored by Esparma Psychophysiology 1996, 33:426-433. and signed a contract concerning the development of neuropsychological 19. Oldfield RC: The assessment and analysis of handedness: The Edinburgh diagnostic and training tools with Schuhfried. Inventory. Neuropsychologia 1971, 9:97-113. FR reports no competing interests. 20. Delmo C, Weiffenbach O, Gabriel M, Poustka F: Diagnostisches Interview VR has received honoraries for lectures from Lilly, Medice, Novartis, Shire. He Kiddie-SADS-Present and Lifetime Version. 5. Aufl. der deutschen has been member of an advisory board of Lilly, Novartis. He has received Forschungsversion Frankfurt: Universität Frankfurt; 2000. research support from Novartis. 21. Nagai Y, Critchley HD, Rothwell JC, Duncan JS, Trimble MR: Changes in SB has received honoraries for lectures from Novartis and support for cortical potential associated with modulation of peripheral sympathetic symposia by Shire. activity in patients with epilepsy. Psychosom Med 2009, 71:84-92. 22. Bender S, Weisbrod M, Resch F, Oelkers-Ax R: Stereotyped topography of Received: 23 March 2011 Accepted: 25 August 2011 different elevated contingent negative variation components in children Published: 25 August 2011 with migraine without aura points towards a subcortical dysfunction. Pain 2007, 127:221-233. 23. Schallberger U: Welches sind nach statistischen Kriterien die besten References Unterformen des HAWIK-III ? Zürich: psychologisches Institut der Universität;; 1. Polanczyk G, de Lima MS, Horta BL, Biederman J, Rohde LA: The worldwide prevalence of ADHD: a systematic review and metaregression analysis. 24. Quinn D, Bode T, Reiz JL, Donnelly GA, Darke AC: Single-dose Am J Psychiatry 2007, 164:942-948. pharmacokinetics of multilayer-release methylphenidate and immediate- 2. Sonuga-Barke EJ: Psychological heterogeneity in AD/HD–a dual pathway release methylphenidate in children with attention-deficit/hyperactivity model of behaviour and cognition. Behavioural brain research 2002, disorder. Journal of clinical pharmacology 2007, 47:760-766. 130:29-36. 25. Bush G, Luu P, Posner MI: Cognitive and emotional influences in anterior 3. Barry RJ, Johnstone SJ, Clarke AR: A review of electrophysiology in cingulate cortex. Trends Cogn Sci 2000, 4:215-222. attention-deficit/hyperactivity disorder: II. Event-related potentials. Clin 26. Cavanagh JF, Cohen MX, Allen JJ: Prelude to and resolution of an error: Neurophysiol 2003, 114:184-198. EEG phase synchrony reveals cognitive control dynamics during action 4. Banaschewski T, Brandeis D: Annotation: what electrical brain activity tells monitoring. J Neurosci 2009, 29:98-105. us about brain function that other techniques cannot tell us - a child 27. Botvinick MM, Braver TS, Barch DM, Carter CS, Cohen JD: Conflict psychiatric perspective. Journal of child psychology and psychiatry, and monitoring and cognitive control. Psychological review 2001, 108:624-652. allied disciplines 2007, 48:415-435. 28. Cools R, Clark L, Owen AM, Robbins TW: Defining the neural mechanisms 5. van Meel CS, Heslenfeld DJ, Oosterlaan J, Sergeant JA: Adaptive control of probabilistic reversal learning using event-related functional magnetic deficits in attention-deficit/hyperactivity disorder (ADHD): the role of resonance imaging. J Neurosci 2002, 22:4563-4567. error processing. Psychiatry research 2007, 151:211-220. 29. van Leijenhorst L, Crone EA, Bunge SA: Neural correlates of 6. Groen Y, Wijers AA, Mulder LJ, Waggeveld B, Minderaa RB, Althaus M: Error developmental differences in risk estimation and feedback processing. and feedback processing in children with ADHD and children with Neuropsychologia 2006, 44:2158-2170. Autistic Spectrum Disorder: an EEG event-related potential study. Clin 30. Klein C, Andresen B, Berg P, Krüger H, Rockstroh B: Topography of CNV Neurophysiol 2008, 119:2476-2493. and PINV in Schizotypal Personality. Psychophysiology 1998, 35:272-282. 7. Albrecht B, Brandeis D, Uebel H, Heinrich H, Mueller UC, Hasselhorn M, 31. Rubia K, Overmeyer S, Taylor E, Brammer M, Williams SC, Simmons A, Steinhausen HC, Rothenberger A, Banaschewski T: Action monitoring in Bullmore ET: Hypofrontality in attention deficit hyperactivity disorder boys with attention-deficit/hyperactivity disorder, their nonaffected during higher-order motor control: a study with functional MRI. Am J siblings, and normal control subjects: evidence for an endophenotype. Psychiatry 1999, 156:891-896. Biological psychiatry 2008, 64:615-625. 32. de Bruijn ER, Hulstijn W, Verkes RJ, Ruigt GS, Sabbe BG: Drug-induced 8. Shiels K, Hawk LW Jr: Self-regulation in ADHD: The role of error stimulation and suppression of action monitoring in healthy volunteers. processing. Clin Psychol Rev 2010. Psychopharmacology 2004, 177:151-160. 9. Bender S, Becker D, Oelkers-Ax R, Weisbrod M: Cortical motor areas are 33. Jonkman LM, van Melis JJ, Kemner C, Markus CR: Methylphenidate activated early in a characteristic sequence during post-movement improves deficient error evaluation in children with ADHD: an event- processing. NeuroImage 2006, 32:333-351. related brain potential study. Biological psychology 2007, 76:217-229. 10. Yordanova J, Dumais-Huber C, Rothenberger A, Woerner W: Frontocortical 34. Moll GH, Heinrich H, Rothenberger A: [Transcranial magnetic stimulation activity in children with comorbidity of tic disorder and attention-deficit in child and adolescent psychiatry: excitability of the motor system in hyperactivity disorder. Biological psychiatry 1997, 41:585-594. tic disorders and/or attention deficit hyperactivity disorders]. Z Kinder 11. Elbert T, Rockstroh B, Lutzenberger W, Birbaumer N: Slow brain potentials Jugendpsychiatr Psychother 2001, 29:312-323. after withdrawal of control. Archiv fur Psychiatrie und Nervenkrankheiten 1982, 232:201-214. doi:10.1186/1744-9081-7-38 12. Bolz J, Giedke H: Controllability of an aversive stimulus in depressed Cite this article as: Werner et al.: Increased performance uncertainty in patients and health controls: a study using slow brain potentials. children with ADHD? - Elevated post-imperative negative variation (PINV) Biological psychiatry 1981, 16:441-452. over the ventrolateral prefrontal cortex. Behavioral and Brain Functions 2011 13. Kathmann N, Jonitz L, Engel RR: Cognitive determinants of the 7:38. postimperative negative variation. Psychophysiology 1990, 27:256-263. 14. Klein C, Rockstroh B, Cohen R, Berg P: Contingent negative variation (CNV) and determinants of the post-imperative negative variation (PINV) in schizophrenic patients and healthy controls. Schizophrenia research 1996, 21:97-110. 15. Verleger R, Wascher E, Arolt V, Daase C, Strohm A, Kompf D: Slow EEG potentials (contingent negative variation and post-imperative negative

Journal

Behavioral and Brain Functions – Springer Journals

Published: Aug 25, 2011

There are no references for this article.