Article

Parietal Attentional System Aberrations During Target Detection in Adolescents With Attention Deficit Hyperactivity Disorder: Event-Related fMRI Evidence

Leanne Tamm, Ph.D. Objective: Directed attention, the ability Results: Individuals with ADHD made to allocate and direct attention toward a significantly more errors of commission Vinod Menon, Ph.D. salient stimulus, is impaired in attention than comparison subjects. Further, rela- deficit hyperactivity disorder (ADHD). This tive to comparison subjects, individuals Allan L. Reiss, M.D. construct is often assessed with target de- with ADHD showed significantly less acti- tection or oddball tasks, and individuals vation in the bilateral parietal lobes (in- with ADHD perform poorly on such tasks. cluding the superior parietal and However, to date, the specific struc- supramarginal and angular gyri of the in- tures or neural mechanisms underlying ferior ), right , and thalamus. target detection dysfunction in individu- als with ADHD have not been identified. Conclusions: Adolescents with ADHD The authors’ goal was to investigate neu- demonstrated significant impairments in ral correlates of target detection dysfunc- their ability to direct and allocate atten- tion in ADHD using event-related fMRI. tional resources. These difficulties were associated with significant aberrations in Method: Behavioral and brain activation the parietal attentional system, which is data were collected while subjects per- known to play a significant role in atten- formed a visual oddball task. Participants tion shifting and detecting specific or sa- included 14 right-handed male adoles- lient targets. Thus, dysfunction in the pa- cents with ADHD (combined type) and 12 rietal attentional system may play a typically developing age- and handed- significant role in the behavioral pheno- ness-matched male comparison subjects. type of ADHD.

(Am J Psychiatry 2006; 163:1033–1043)

Attention deficit hyperactivity disorder (ADHD) is a ADHD commit more errors of commission (failure to behavioral disorder characterized by marked problems withhold a response to frequent or standard stimuli) and with inattention, impulsivity, and hyperactivity that often have slower or more variable reaction times than causes significant functional impairment in multiple set- comparison subjects (4, 5). Further, individuals with tings (e.g., school and home). Although behavioral and ADHD show aberrant event-related potentials (i.e., de- cognitive deficits in ADHD are well described, there is little creased P300 latency and amplitude) in response to odd- consensus regarding the specific brain structures or neu- ball or rare target stimuli in conjunction with poorer be- ral mechanisms underlying dysfunction in individuals havioral performance relative to comparison subjects (4– with this disorder (1). To further elucidate this issue, a va- 8). Thus, it appears that impairments in directed attention riety of laboratory paradigms have been used, including are characteristic of ADHD. A clearer understanding of the tasks involving target detection and response inhibition. underlying neural correlates of these deficits in ADHD will One such paradigm is the “oddball task,” which involves be important in diagnosis and treatment planning. detection and response to infrequent (oddball) target Although event-related potential studies index the tim- events embedded in a series of repetitive events. Following initial sensory processing, target detection involves the ing and sequence of neuronal activity underlying cogni- process of bringing a salient stimulus into conscious tive processes, this technique is limited because of poor awareness for further evaluation, categorization, and re- spatial resolution (i.e., a number of current distributions sponse. This task is thought to capture the construct of di- in the brain can give rise to identical field distributions on rected attention and reflects an individual’s ability to the scalp [2]). Thus, complementary methods are needed monitor the environment for change and decide on a to determine the neural correlates of target detection. In course of action, a critical survival skill (2, 3). particular, the advent of noninvasive functional MRI Studies that have investigated target detection or odd- (fMRI) has allowed investigation of the neural correlates of ball task performance have reported that individuals with such functions as target detection, thus contributing to a

Am J Psychiatry 163:6, June 2006 ajp.psychiatryonline.org 1033 NEURAL CORRELATES IN ADHD better understanding of executive dysfunction observed group reported current use of stimulant medication; all five were in neuropsychiatric disorders such as ADHD. subjected to an 18-hour washout of their stimulant medication To date, there have been no fMRI studies investigating prior to the scan. In addition, all participants did not have caf- feine at least 2 hours before the scan, and of those reporting drug target detection or oddball task performance in an ADHD use (two in the ADHD group, one in the comparison group) and population. However, five fMRI studies have investigated alcohol use (three in the ADHD group, six in the comparison response inhibition in ADHD populations using go/no go group), a 3-week washout period for drugs (typically marijuana) type tasks and block (9–11) and event-related designs (12, and a 1-week washout period for alcohol were observed. 13). The go/no go and oddball tasks are somewhat like Typically developing comparison subjects were screened (via telephone interview with their primary caretaker and a follow-up mirror images of one another, with go/no go type tasks in- written questionnaire) for absence of neurological, developmen- volving response inhibition in the context of frequent re- tal, and psychiatric disorders and no family history of specific sponse generation, and oddball tasks involving response psychiatric disorders (i.e., ADHD, bipolar disorder, substance generation in the context of frequent response inhibition abuse/dependence, and conduct disorder/antisocial personality (14). It is unclear whether response inhibition (the act of disorder). All individuals in the comparison group scored in the normative range (<65) on the Child Behavior Checklist (25). withholding or suppressing a response) differs from that ADHD diagnosis was determined through a structured clinical of selecting an alternative response. interview conducted with the primary caregiver of each partici- To examine these processes and associated dysfunction pant during the initial telephone screening and was verified dur- in ADHD, we designed two event-related fMRI tasks, a go/ ing the initial visit (Diagnostic Interview for Children and Adoles- no go task controlling for novelty (13) and an oddball task. cents–Version IV [DICA-IV] completed by primary caregiver). In addition, each primary caregiver completed the Conner ADHD/ In this study, we used a standard visual-based oddball task DSM-IV scale–Parent Version about their adolescent (ADHD In- similar to that used in other fMRI studies (15), with a mod- dex Score: mean=75.5, SD=10.0). Individuals with a family history ification that controls for motor responding. This type of of bipolar disorder were excluded from participation. One partic- task also has been referred to as a “change task” (16) or a ipant in the ADHD group met diagnostic criteria for current ma- response selection task (14). Individuals with ADHD are jor depression and was excluded from analyses to avoid con- founds related to psychomotor retardation. known to perform poorly on these tasks (16). To familiarize subjects with the MRI environment and to mini- We predicted that individuals with ADHD would make mize artifacts related to movement, a protocol using an MRI sim- more errors of commission than typically developing ulator was administered to subjects in the ADHD group. Specifi- healthy comparison subjects on the oddball task. Al- cally, a movie viewed by the subject during the simulated scan though no previous fMRI studies have investigated odd- would cutoff for 3 seconds whenever the subject moved beyond a set criterion. Increasingly stringent movement criteria (decreas- ball task performance in ADHD subjects, previous fMRI ing from 3 to 1 mm) and increasing time periods (2 minutes, 5 studies of normal subject performance on oddball para- minutes, 10 minutes) were selected until all subjects met a crite- digms have reported activation to target stimuli in the in- rion of less than three movements exceeding 1 mm within a 10- ferior and middle frontal gyri (15, 17–19) as well as tempo- minute period. Exclusion criteria of head movement greater than ° ral regions ( [Brodmann’s area 22] 3 mm and rotation less than 3 during fMRI scanning were estab- lished; no participants exceeded this movement threshold. and ) and parietal regions (inferior Cognitive functioning was assessed with the Wechsler Abbrevi- parietal lobules [including ] and su- ated Intelligence Scale (WAIS), and academic functioning (i.e., perior parietal lobe) (2, 15, 17–20). On the basis of mor- reading, spelling, and arithmetic) was assessed with the Wide phological studies (21) and functional imaging studies of Range Achievement Test, Third Edition (WRAT-III). response inhibition (9–13) implicating these regions in Experimental Task ADHD, we predicted that activation might be aberrant in In the event-related oddball task, stimuli were presented using the frontal and striatal regions in the ADHD group. Fur- a fast event-related design (2, 26–28). A jittered stimulus presen- ther, since the parietal regions and frontal lobes play a crit- tation was used with mean inter-oddball interval of 10.9 sec (SD= ical role in sustained attention or vigilance (22–24) (also 9.8) (28, 29). Such a jittered stimulus presentation has been known to be impaired in ADHD [24]), we also predicted shown to be efficient for separating brain activation to individual aberrant activation in these regions. events (28). To ensure that events could be statistically separated, we further verified that the predicted responses for each pair of stimuli were mutually orthogonal. Expected waveforms for each Method of the two conditions were computed after convolution with a he- modynamic response function (30). The correlation between ex- Participants pected responses to each pair of conditions was less than 0.01, al- Male participants diagnosed with ADHD (N=14) and 12 typi- lowing us to independently assess brain activation to each cally developing male comparison subjects participated in the condition (31). study after giving written informed consent and assent in accor- The event-related oddball experiment consisted of a 20-second dance with Stanford Human Subjects Institutional Review Board rest epoch at the beginning and end of the task, during which requirements. All participants were paid $100 for participating in subjects passively viewed a screen showing the word “rest.” Fol- the study. Participants were right-handed and ranged in age from lowing the initial rest period, green circles (standard stimulus, 14 to 18 years (group-matched; mean age=15.64 years, SD=1.15). 80% of trials) or green triangles (target or oddball stimulus, 20% of Participants were predominantly of Caucasian ethnicity (80%), trials) were presented sequentially for 250 msec with an inter- with the remaining participants identifying themselves as mixed- stimulus interval of 1750 msec. Total number of trials was 225, Asian (12%) and Hispanic (8%). Five participants in the ADHD which were presented in a single run. Participants were in-

1034 ajp.psychiatryonline.org Am J Psychiatry 163:6, June 2006 TAMM, MENON, AND REISS

TABLE 1. Cognitive and Academic Functioning of Male Adolescents With ADHD and an Age-Matched Comparison Group of Healthy Male Adolescents ADHD Subjects (N=14) Healthy Comparison Subjects (N=12) Functioning Type and Measure Mean SD Mean SD Cognitive functioninga Verbal IQ 110.15 13.75 111.33 14.84 Performance IQ 105.23 12.58 103.92 12.30 Full scale IQ 108.38 13.28 111.58 11.73 Academic functioningb Reading 105.53 7.95 111.00 5.67 Spelling 97.71 13.57 106.00 10.44 Arithmetic 103.57 14.97 106.25 15.52 a Assessed with the Wechsler Abbreviated Intelligence Scale. b Assessed with the Wide Range Achievement Test, 3rd ed.

TABLE 2. Oddball Taska Performance Data for Male Adolescents With ADHD and a Matched Comparison Group of Healthy Male Adolescents ADHD (N=14) Healthy Comparison Subjects (N=12) Variable Mean SD Mean SD Errors of commission Pressed 1 instead of 2 3.15 2.03 1.50* 1.51 Expressed as percent of novel stimuli 0.09 0.05 0.04* 0.04 Pressed 2 instead of 1 13.00 20.12 2.33* 5.68 Expressed as percent of standard stimuli 0.08 0.14 0.02* 0.04 Total number correct 155.36 6.54 172.58* 2.89 Reaction time to novel stimuli (msec) 188.16 76.83 216.14 87.09 Reaction time to standard stimuli (msec) 158.37 89.98 177.93 99.78 a Subjects viewed a series of triangles (novel stimuli, N=45) or circles (standard stimuli, N=180), pressing “1” with their forefinger when a tri- angle appeared and pressing “2” with their middle finger when a circle appeared. *p<0.05. structed to respond to the standard stimuli (circles) by pressing built magnet compatible projection system (Resonance Technol- one button on the response box and to target stimuli (triangles) ogy, Calif.). by pressing a second button on the response box. Specifically, the instructions read, “For this task, the computer will show you cir- Image Preprocessing cles and triangles. Your job is to press 1 for triangles or press 2 for Images were reconstructed, by inverse Fourier transform, for circles, as quickly as you can. Remember you want to be quick but each of the 120 time points into 64×64×18 image matrices (voxel also accurate, so do not go too fast.” Participants responded using size: 3.75×3.75×7 mm). fMRI data were preprocessed using the forefinger and middle finger of the right hand. Errors of com- SPM99. Images were corrected for movement using least square mission (pressing 1 instead of 2/pressing 2 instead of 1) and reac- minimization without higher-order corrections for spin history, tion times to standard and target stimuli were recorded. and normalized to Montreal Neurological Institute (MNI) tem- plate provided with SPM. Images were then resampled every 2 Image Acquisition mm using sinc interpolation. Images were acquired on a 1.5-T GE Signa scanner with echo Statistical Analysis speed gradients using a custom-built whole head coil that provides a 50% advantage in signal-to-noise ratio over that of the standard Statistical analysis was performed on group data by using a GE coil (32). A custom-built headholder was used to prevent head random effects model (34) along with the theory of Gaussian ran- movement. Eighteen axial slices (6 mm thick, 1 mm skip) parallel to dom fields as implemented in SPM99. This method takes advan- the anterior and posterior commissures covering the whole brain tage of multivariate regression analysis and corrects for temporal were imaged with a temporal resolution of 2 seconds using a T2* and spatial autocorrelations in the fMRI data (35). weighted gradient echo spiral pulse sequence (TR=2000 msec, TE= Confounding effects of fluctuations in global mean were re- 30 msec, flip angle=89o and 1 interleave) (33). The field of view was moved by proportional scaling where, for each time point, each 240 mm, and the effective in-plane spatial resolution was 3.75 mm. voxel was scaled by the global mean at that time point. Low-fre- To aid in localization of functional data, high-resolution T1- quency noise was removed with a high-pass filter (0.5 cycles/ weighted inversion recovery spoiled grass gradient recalled (SPGR) minute) applied to the fMRI time series at each voxel. A temporal three-dimensional MRI sequence with the following parameters smoothing function (4 mm Gaussian kernel corresponding to dis- was used: TR=9, TE=1.9, flip angle=15o, 124 slices in coronal plane, persion of 8 seconds) was applied to the fMRI time series to en- 256×192 matrix, acquired resolution=1.5×0.9×1.2 mm. The images hance the temporal signal-to-noise ratio. Voxel-wise t statistics were reconstructed as a 124×256×256 matrix with a 1.5×0.9×0.9 mm were computed using the random effects model and normalized spatial resolution. to z scores to provide a statistical measure of activation indepen- The task was programmed using PsyScope (http://psy- dent of sample size. Finally, to determine the presence of signifi- scope.psy.cmu.edu) on a Macintosh notebook computer. Initia- cant clusters of activation, a joint expected probability distribu- tion of scan and task was synchronized using a TTL pulse deliv- tion of height (z>1.67, p<0.05) and extent (p<0.05) threshold (36) ered to the scanner timing microprocessor board from a “CMU was used to correct for spatial correlations in the data. Button Box” microprocessor connected to the Macintosh. Stimuli For group analysis, a random effects model was used to deter- were presented visually at the center of a screen using a custom- mine voxel-wise t statistics contrasting specific events of interest

Am J Psychiatry 163:6, June 2006 ajp.psychiatryonline.org 1035 NEURAL CORRELATES IN ADHD

TABLE 3. Brain Areas Showing Significant (p<0.01) Activation Within Each Group for the Oddball Event Number of Voxels Maximum z Peak Location Region Activated During Oddball Event Activated Score (Talaraich coordinates) Healthy comparison subjects Precuneus (Brodmann’s area 7) extending to and supramarginal gyrus, also to post-central gyrus 3256 4.57 8, -68, 30 Superior parietal lobe extending to angular gyrus (Brodmann’s area 40) and intraparietal and post-central gyrus 3266 4.45 -26, -70, 40 Putamen extending to claustrum, posterior/medial orbital gyrus, and insula 2279 3.82 -24, 14, -8 – (Brodmann’s area 6)/pre- central gyrus extending to 519 3.73 -20, -8, 68 (Brodmann’s area 47/44) extending to middle frontal gyrus 905 3.59 34, 24, 2 Superior frontal gyrus – supplementary motor area (Brodmann’s area 6), extending to anterior cingulate gyrus (Brodmann’s area 24) 808 3.49 4, 8, 54 Posterior (Brodmann’s area 23/31) adjacent to splenium/isthmus 667 3.42 0, -28, 32 Male adolescents with ADHD Inferior frontal gyrus extending to the lateral/posterior orbital gyrus 993 3.82 46, 12, 14 Right anterior/mid-cingulate cortex extending to superior frontal gyrus 680 3.21 2, 16, 42

TABLE 4. Brain Areas Showing Significant (p<0.01) Between-Group Activationa Number of Voxels Maximum z Peak Location Region Activated During Oddball Event Activated Score (Talaraich coordinates) Right inferior parietal gyrus (Brodmann’s area 40) including angular gyrus and su- pramarginal gyrus, extending to superior parietal lobe 580 3.37 40, -38, 38 Left inferior parietal gyrus (Brodmann’s area 40) including angular gyrus and su- pramarginal gyrus 1137 3.34 -44, -28, 38 Right precuneus (Brodmann’s area 7/31) extending to and subparietal sulcus. 414 2.88 4, -50, 36 Thalamus extending to mid-cingulate cortex 419 2.79 0, -22, 20 a All significant between-group activation differences represent regions in which the comparison subjects exhibited significantly greater activa- tion than the ADHD subjects. ADHD subjects had no regions in which they exhibited greater activation than the comparison subjects during the oddball event.

(i.e., oddball versus standard stimuli). This model estimates the that there were no significant between-group movement error variance for each condition of interest across subjects rather differences by comparing the groups’ average displace- than across scans (34). The random effects model provides better ment from the mean head position in six dimensions generalization to the subject population, albeit with some loss in power due to averaging in the time domain and small N. This (translation in x, y, and z, and rotation around the x, y, and analysis proceeded in two steps. In the first step, adjusted images z axes). No significant group differences emerged. corresponding to the conditions/events of interest were deter- mined. For each condition, a weighted average of the images was IQ and Achievement computed, taking into account the hemodynamic response. In The two groups did not significantly differ with regard to the second step, these condition-specific images were contrasted verbal, performance, or full scale IQ scores or on reading, in a general linear model to determine appropriate t statistics. The t statistics were normalized to z scores to determine signifi- spelling, or arithmetic scores. Table 1 includes descriptive cant clusters of activation. We examined activation to oddball data for these variables. Two subjects in the ADHD group versus standard events. met DSM-IV criteria for a learning disability in spelling. Neuroanatomical locations of activation were first determined using the standard Talairach atlas and then refined using the Behavioral Data more detailed and thorough Duvernoy atlas (37). An initial investigation of the dependent variables (both Functional ROI Analysis errors of commission types, reaction time to triangles Following these analyses, exploratory analyses were conducted [oddballs], reaction time to circles [standards], and total to assess the relationship between brain activation and perfor- number correct) using stem-and-leaf plots in SPSS re- mance. Specifically, regions in which the comparison subjects ex- vealed the presence of outliers in both groups for the er- hibited significantly greater activation than the ADHD group, i.e., rors of commission variables. To prevent any confounds in functional regions of interest, were identified. Pearson correla- the data resulting from non-normality of the data, non- tions between percentage of voxels activated (height threshold: z>1.67) between the functional regions of interest and errors of parametric Mann-Whitney tests were conducted to com- commission were then computed for each group. pare the groups on the errors of commission variables. In- dependent sample t tests were conducted for the reaction Results time variables. The results of these analyses revealed significant group Movement Analysis differences for both errors of commission variables, with Although no participant exceeded minimum movement the ADHD group making significantly more errors of com- thresholds of 3 mm in any direction, we further ensured mission than comparison subjects, pressing 1 instead of 2

1036 ajp.psychiatryonline.org Am J Psychiatry 163:6, June 2006 TAMM, MENON, AND REISS

FIGURE 1. Regions of Significant Activation in Healthy (N=12) and ADHD (N=13) Male Adolescents During Oddball Versus Standard Events

Left Right

63 59 55 51 47

43 39 35 31 27

23 19 15 11 7

3–1–5–9–13

–17–21 –25 –29 –33

–37–41 –45 –49 –53

–57–61 –65 –69 –73 5 5

0 0 –77–81 –85 Comparison ADHD group group (N=12) (N=13)

(Mann-Whitney U=38.0, p<0.05) and pressing 2 instead of Brain Activation < 1 (Mann-Whitney U=24.5, p 0.01). The two groups did not Within-group activations to oddball versus standard differ in reaction time to either novel triangles or standard events were examined for each group. A between-group circles. Means and standard deviations for these variables analysis was then conducted to examine group differences can be found in Table 2. in activation. To control for the potentially confounding Observation of the data reported in Table 2 suggests a higher influences of deactivation in the ADHD group, which has number of errors in which 2 was pressed instead of 1 (pressing been shown to significantly influence between-group standard button when novel stimulus appears) than errors in analyses, a mask of deactivation (activation to standard which 1 was pressed instead of 2 (pressing novel button when versus oddball events) was created for the between-group standard stimulus appears); however, a paired samples t test analyses. For a detailed discussion about deactivation in- comparing the two types of errors was not significant. fluences on between-group activation comparisons, see

Am J Psychiatry 163:6, June 2006 ajp.psychiatryonline.org 1037 NEURAL CORRELATES IN ADHD

FIGURE 2. Regions of Significantly Greater Activation in the Comparison Group (N=12) Relative to the ADHD Group (N=13) for Oddball Versus Standard Events

Left Right

63 59 55 51 47

43 39 35 31 27

23 19 15 11 7

3–1–5–9–13

–17–21 –25 –29 –33

–37–41 –45 –49 –53

–57–61 –65 –69 –73

4

2

0 –77–81 –85 Comparison – ADHD a Regions mapped onto an average T1-weighted image of healthy and ADHD .

Tamm et al. (38). Only data that survived masking with de- tion was observed in the bilateral parietal lobe, activation are reported for the between-group analyses. including the angular gyrus, supramarginal gyrus, ex- ADHD group. Male subjects with ADHD showed signifi- tending to the . Significant activation cant activation for the oddball versus standard events in also was observed in the left putamen extending to the the right inferior frontal gyrus extending to the orbital gy- claustrum, insula, and posterior orbital gyrus. Bilateral rus, and in the right anterior to mid-cingulate cortex ex- activation in the superior frontal gyrus/supplementary tending to the superior frontal gyrus (Table 3, Figure 1). motor area also emerged, which extended ventrally to Comparison group. Comparison subjects showed sev- the middle frontal gyrus. Finally, activation was also ob- eral peaks of activation associated with the oddball ver- served in the right inferior frontal gyrus and right poste- sus standard events (Table 3, Figure 1). Significant activa- rior cingulate cortex.

1038 ajp.psychiatryonline.org Am J Psychiatry 163:6, June 2006 TAMM, MENON, AND REISS

Group differences. The ADHD group did not show FIGURE 3. Surface Rendering of Regions of Significantly greater activation to oddball versus standard events rela- Greater Activation in Healthy Subjects Relative to the ADHD Group tive to healthy subjects on the between-group compari- sons. In contrast, for the oddball versus standard events, the comparison group showed significantly more activa- tion than the ADHD group in the bilateral parietal associ- ation cortex (Brodmann’s area 40/5/7), including the an- gular and supramarginal gyri, as well as the right precuneus/cuneus and thalamus/mid-cingulate (Table 4, Figure 2, Figure 3).

Exploratory Functional Region of Interest Analysis late, and frontal lobes is likely associated with the process Bivariate correlations were conducted between the two of target detection/directed attention (including activa- errors of commission variables, number correct, and per- tion related to response inhibition, vigilance, and working cent voxels activated in functional regions of interest (i.e., memory), since these regions are consistently activated in regions in which comparison subjects exhibited greater visual oddball tasks (19, 44). activation than the ADHD group). A significant negative More specifically, anterior cingulate activation to odd- correlation was observed in comparison subjects between ball stimuli observed in the comparison group likely re- percent voxels activated in the left parietal cortex and the flects response conflict (i.e., simultaneous activation of in- commission error of pressing the standard button when compatible response tendencies [14]), and inferior frontal the novel stimulus appeared (Table 5). Activation in the gyrus activation is thought to be related to the need to right parietal cortex, precuneus, and thalamus was not “stop” prepotent motor responses (26, 45, 46). Putamen significantly correlated with errors of commission. No sig- activation during target detection in normal subjects has nificant correlations were observed in the ADHD group. been reported previously (47) and may be associated with response selection during controlled processing (48) and Discussion response inhibition (9). Overall, those regions in which comparison subjects showed significant activation to This study is the first to examine the performance and oddball stimuli in our study are similar to those reported brain activation of individuals with ADHD relative to com- by target detection studies, suggesting that the task used parison subjects on an oddball task using fMRI. Results in- in this study captured the construct of directed attention. dicated significant behavioral differences between the two In contrast to the comparison group, individuals with groups accompanied by significant differences in brain ADHD showed fewer peaks of activation. Within-group activation. Adolescents in the ADHD group made signifi- analyses revealed activation in the right inferior frontal gy- cantly more errors of commission than comparison sub- rus and in the right cingulate gyrus extending to the supe- jects on the oddball task, but the two groups did not differ rior frontal gyrus. Again, activation in the right inferior significantly on the reaction time variables. Individuals frontal gyrus likely reflects response inhibition (26, 45, 46), with ADHD exhibited significantly less activation than and activation in the cingulate gyrus likely reflects re- comparison subjects in the bilateral parietal association sponse conflict (14). Notably, activation in the cingulate cortex, in the right precuneus, and in the thalamus. These region was more anterior in the ADHD group than that ob- findings suggest a critical role for these brain regions in ac- served in the comparison group. It has been suggested curate target detection and task performance, which may that the cingulate can be subdivided in terms of function underlie known deficits in directed attention in individu- into the caudal cingulate zone, the rostral cingulate zone, als with ADHD. and the anterior rostral cingulate zone (49), with the ante- Within-group analyses revealed that the comparison rior rostral cingulate zone thought to be an “error-sensi- group activated several regions during task performance, tive” region (14). Because individuals in the ADHD group including peaks of activation in the bilateral parietal lobe, made significantly more errors of commission than com- left putamen, bilateral superior frontal gyrus/supplemen- parison subjects, activation in this region may reflect re- tary motor area, left middle frontal gyrus, right posterior sponse to errors or increased conflict resulting in errors of cingulate gyrus, and right inferior frontal gyrus. These re- commission (14). However, to assess the validity of this hy- gions of activation are consistent with those reported by pothesis, continued investigation is warranted. It was not other fMRI studies of oddball tasks in normal comparison possible to investigate brain activation specifically related subjects (2, 12, 13, 15, 17–20, 39–41). It is probable that ac- to errors of commission because of insufficient power. tivation in the supplementary motor area reflects higher- Adolescents with ADHD did not show more activation order motor planning (14, 42, 43) and potentially atten- than comparison subjects in any region on the between- tional shifting (43). Activation in the parietal lobes, cingu- group analyses. However, the ADHD group did make sig-

Am J Psychiatry 163:6, June 2006 ajp.psychiatryonline.org 1039 NEURAL CORRELATES IN ADHD

TABLE 5. Correlations Between Oddball Taska Performance Data and Percentage of Voxels Activated in the Four Regions of Significantly Greater Activation in Healthy Subjects Relative to the ADHD Group Errors of Commission Pressed 1 Instead of 2 Pressed 2 Instead of 1 Total Correct Functional Regions of Interest Healthy (N=12) ADHD (N=13) Healthy (N=12) ADHD (N=13) Healthy (N=12) ADHD (N=13) Right inferior parietal gyrus 0.057 0.210 0.007 –0.047 0.168 –0.012 Left inferior parietal gyrus –0.433 0.182 –0.521* –0.071 0.357 –0.084 Right precuneus 0.177 0.346 0.064 0.251 –0.184 –0.292 Right thalamus 0.025 –0.151 –0.088 –0.299 –0.175 0.260 a Subjects viewed a series of triangles (novel stimuli, N=45) or circles (standard stimuli, N=180), pressing “1” with their forefinger when a tri- angle appeared and pressing “2” with their middle finger when a circle appeared. *p<0.05. nificantly more errors of commission and showed signifi- switched between objects presented in the same location cantly less activation in several posterior brain regions. Ex- [e.g., the oddball stimuli (53)]. ploratory functional regions of interest analyses indicated Group differences in angular gyrus activation were that activation in the left parietal lobe was significantly somewhat unexpected, since this region is most fre- negatively correlated with errors for the comparison quently reported as involved in language (62) and number group, supporting the hypothesis that for comparison processing (63, 64). However, angular gyrus activation also subjects, the left parietal cortex may be directly involved has been observed in response to visuospatial functioning in improving behavioral performance. The fact that no (65), visual pattern recognition (66), and perception of significant correlations were observed between errors of spatial features of objects (67), which are clearly relevant commission and other regions with group differences sug- for the visual oddball task. Activation in the thalamus has gests that the right parietal cortex and right cingulate are been reported in previous oddball studies (2, 17, 18, 41) either related to performance indirectly or via a mecha- and is thought to reflect engagement of the hippocampal- nism not measured in this study. Regardless, it does ap- hypothalamic network in mediating the initial orienting pear that activation in these regions plays a role in task response (2, 3). performance (i.e., target detection) and may be associated Failure of the ADHD group to activate to the same degree with improving performance in the comparison group. as comparison subjects in parietal/precuneus regions sug- Male adolescents with ADHD activated the bilateral su- gests that they had difficulties in shifting attention and al- perior parietal lobe, angular and supramarginal gyri, the ternating motor responses to target stimuli, possibly one reason why this group made more errors than comparison right precuneus, and the thalamus significantly less than subjects. A hypothesis that individuals with ADHD have comparison subjects. There is strong evidence to suggest impaired attention shifting is supported by our finding that that activation in the superior and inferior parietal lobes is this group made more of both errors of commission types. related to sustained attention/vigilance and attentional or That is, they appeared to have trouble inhibiting a prepo- set shifting (50, 51), or more generally, effortful attention tent response and shifting to make an alternative response (52, 53). Similarly, the precuneus has been shown to acti- to target stimuli (i.e., pressing 1 instead of 2), but they also vate in response to relevancy and context (54, 55), as well had difficulty shifting back to making the prepotent re- as attentional shifting (43, 51, 56, 57). The majority of fMRI sponse (i.e., pressing 2 instead of 1). Thus, the difficulties studies investigating a number of oddball paradigm vari- expressed by individuals with ADHD appear to reflect ants report activation in the bilateral supramarginal gyrus more than impaired response inhibition; they also reflect (most typically in Brodmann’s area 40), which suggests a fundamental impairments in the ability to shift attention critical role for this region in target detection (2, 15, 19–20, and make alternative responses. In addition, errors of com- 58). Other studies suggest a role for the left parietal cortex, mission signal impulsivity, which also is indicative of im- and in particular the supramarginal gyrus, in motor atten- paired directed attention in ADHD. tion (59, 60), and the right supramarginal gyrus in atten- Limitations of this study include the relatively small study tional shifting (61). Thus, the confluence of evidence sug- group size. In addition, despite the fact that we excluded one gests that the bilateral supramarginal gyrus plays a role in subject who met criteria for current major depression, a postsensory processing of salient stimuli for evaluation, small percentage of individuals in the ADHD group met di- categorization, response, and decision making (2). It is in- agnostic criteria for other disorders, which may have con- teresting that the pattern of results for the ADHD group is founded the findings. It should be noted that interpretation consistent with impairments in the regions thought to be of fMRI group differences may reflect only performance im- associated with goal-directed attention (i.e., a top-down pairments, as opposed to a deficit specific to diagnosis with process for attention to features of a stimulus) including ADHD. Future studies with sufficient power and errors to the fronto-parietal network and in particular the posterior conduct a covariance analysis with error rates will be parietal region, which activates in response to attention needed to disentangle this important question.

1040 ajp.psychiatryonline.org Am J Psychiatry 163:6, June 2006 TAMM, MENON, AND REISS

Overall, the results of this study demonstrate a relation- 7. Lazzaro I, Gordon E, Whitmont S, Meares R, Clarke S: The mod- ship between impaired target detection and brain func- ulation of late component event related potentials by pre- stimulus EEG theta activity in ADHD. Int J Neurosci 2001; 107: tion in ADHD; more generally, these results contribute to 247–264 our understanding of the neural correlates of impaired 8. Jonkman LM, Kemner C, Verbaten MN, Van Engeland H, Camf- ability to direct attention in ADHD. Our findings suggest ferman G, Buitelaar JK, Koelega HS: Attentional capacity, a that individuals with ADHD have significant aberrations probe ERP study: differences between children with attention- in the posterior parietal attentional system, which is deficit hyperactivity disorder and normal control children and effects of methylphenidate. Psychophysiology 2000; 37:334– known to play a critical role in maintaining and shifting at- 346 tention. Prior imaging studies of ADHD have primarily fo- 9. Teicher MH, Anderson CM, Polcari A, Glod CA, Maas LC, Ren- cused on the role of frontal and striatal regions in execu- shaw PF: Functional deficits in basal ganglia of children with tive functions, such as response inhibition (9–11), since attention- deficit/hyperactivity disorder shown with functional morphological abnormalities of these regions are com- magnetic resonance imaging relaxometry. Nat Med 2000; 6: monly reported (21). Our results, however, suggest that it 470–473 10. Rubia K, Overmeyer S, Taylor E, Brammer M, Williams SC, Sim- is important to reconsider the notion of ADHD as prima- mons A, Andrew C, Bullmore ET: Functional frontalisation with rily a disorder of frontal-striatal function, and consider the age: mapping neurodevelopmental trajectories with fMRI. role of the parietal attentional system in the behavioral Neurosci Biobehav Rev 2000; 24:13–19 phenotype of ADHD. Additional studies are warranted to 11. Vaidya CJ, Austin G, Kirkorian G, Ridlehuber HW, Desmond JE, investigate the relationship and interconnections be- Glover GH, Gabrieli JD: Selective effects of methylphenidate in attention deficit hyperactivity disorder: a functional magnetic tween frontal, striatal, and parietal regions in executive resonance study. Proc Natl Acad Sci USA 1998; 95:14494– dysfunction in ADHD. 14499 12. Schulz KP, Fan J, Yang CK, Newcorn JH, Buchsbaum MS, Cheung Received April 24, 2003; revision received Feb. 22, 2005; accepted AM, Halperin JM: Response inhibition in adolescents diag- March 18, 2005. From the Department of Psychiatry and Behavioral nosed with attention deficit hyperactivity disorder during Sciences, Program in , and Stanford Brain Research childhood: an event-related fMRI study. Am J Psychiatry 2004; Center, Stanford University School of Medicine, Stanford, Calif. Ad- 161:1650–1657 dress correspondence and reprint requests to Dr. Reiss, Department 13. Tamm L, Menon V, Ringel J, Reiss AL: Event-related fMRI evi- of Psychiatry and Behavioral Sciences, Stanford University School of dence for fronto-temporal involvement in aberrant response Medicine, 401 Quarry Rd., Stanford, CA 94305-5717; reiss@stan- inhibition evidenced by adolescents with attention deficit hy- ford.edu (e-mail). peractivity disorder. J Am Acad Child Adolesc Psychiatry 2004; Supported by NIH grants (HD-40761, HD-31715, MH-01142, MH- 43:1430–1440 50047, MH-19908, and MH-62430) and a fellowship from the Con- stance Bultman Wilson Foundation. 14. Braver TS, Barch DM, Gray JR, Molfese DL, Snyder A: Anterior cingulate cortex and response conflict: effects of frequency, in- The authors thank Anamika Banerji and Ben Krasnow for assis- tance with data processing and analysis, Jessica Ringel for assistance hibition and errors. Cereb Cortex 2001; 11:825–836 with fMRI scanning, Cindy Johnston for assistance with psychological 15. Stevens AA, Skudlarski P, Gatenby JC, Gore JC: Event-related testing, and Dean Mobbs for assistance with the mock simulator. fMRI of auditory and visual oddball tasks. Magn Reson Imaging 2000; 18:495–502 16. Sergeant JA, Oosterlaan J, van der Meere J: Information pro- References cessing and energetic factors in attention-deficit/hyperactivity disorder, in Handbook of Disruptive Behavior Disorders. Edited 1. Stefanatos GA, Wasserstein J: Attention deficit/hyperactivity dis- by Hogan HCQAE. New York, Plenum Publishers, 1999, pp 75– order as a right hemisphere syndrome. Selective literature re- 104 view and detailed neuropsychological case studies. Ann N Y 17. Kiehl KA, Laurens KR, Duty TL, Forster BB, Liddle PF: Neural Acad Sci 2001; 931:172–195 sources involved in auditory target detection and novelty pro- cessing: an event-related fMRI study. Psychophysiology 2001; 2. Menon V, Ford JM, Lim KO, Glover GH, Pfefferbaum A: Com- 38:133–142 bined event-related fMRI and EEG evidence for temporal-pari- etal cortex activation during target detection. Neuroreport 18. Clark VP, Fannon S, Lai S, Benson R, Bauer L: Responses to rare 1997; 8:3029–3037 visual target and distractor stimuli using event- related fMRI. J Neurophysiol 2000; 83:3133–3139 3. Knight R: Contribution of human hippocampal region to nov- 19.McCarthy G, Luby M, Gore J, Goldman-Rakic P: Infrequent elty detection. Nature 1996; 383:256–259 events transiently activate human prefrontal and parietal cor- 4. Idiaz bal Alecha M, Palencia Taboada AB, Sangorr n J, Espad- tex as measured by functional MRI. J Neurophysiol 1997; 77: aler Gamissans JM: [Cognitive evoked potentials in the hyper- 1630–634 activity attention deficit disorder]. Rev Neurol 2002; 34:301– 20. Linden DE, Prvulovic D, Formisano E, Vollinger M, Zanella FE, 305 Goebel R, Dierks T: The functional neuroanatomy of target de- 5. Kemner C, Verbaten MN, Koelega HS, Buitelaar JK, van der tection: an fMRI study of visual and auditory oddball tasks. Gaag RJ, Camfferman G, van Engeland H: Event-related brain Cereb Cortex 1999; 9:815–823 potentials in children with attention-deficit and hyperactivity 21. Hendren RL, De Backer I, Pandina GJ: Review of neuroimaging disorder: effects of stimulus deviancy and task relevance in the studies of child and adolescent psychiatric disorders from the visual and auditory modality. Biol Psychiatry 1996; 40:522–534 past 10 years. J Am Acad Child Adolesc Psychiatry 2000; 39: 6. Lazzaro I, Anderson J, Gordon E, Clarke S, Leong J, Meares R: 815–828 Single trial variability within the P300 (250–500 ms) processing 22. Lewin JS, Friedman L, Wu D, Miller DA, Thompson LA, Klein SK, window in adolescents with attention deficit hyperactivity dis- Wise AL, Hedera P, Buckley P, Meltzer H, Friedland RP, Duerk JL: order. Psychiatry Res 1997; 73:91–101 Cortical localization of human sustained attention: detection

Am J Psychiatry 163:6, June 2006 ajp.psychiatryonline.org 1041 NEURAL CORRELATES IN ADHD

with functional MR using a visual vigilance paradigm. J Comput 43. Nagahama Y, Okada T, Katsumi Y, Hayashi T, Yamauchi H, Assist Tomogr 1996; 20:695–701 Sawamoto N, Toma K, Nakamura K, Hanakawa T, Konishi J, 23. Coull JT, Frackowiak RS, Frith CD: Monitoring for target objects: Fukuyama H, Shibasaki H: Transient neural activity in the me- activation of right frontal and parietal cortices with increasing dial superior frontal gyrus and precuneus time locked with at- time on task. Neuropsychologia 1998; 36:1325–1334 tention shift between object features. Neuroimage 1999; 10: 24. Sunshine JL, Lewin JS, Wu DH, Miller DA, Findling RL, Manos MJ, 193–199 Schwartz MA: Functional MR to localize sustained visual atten- 44. Herrmann CS, Knight RT: Mechanisms of human attention: tion activation in patients with attention deficit hyperactivity event-related potentials and oscillations. Neurosci Biobehav disorder: a pilot study. AJNR Am J Neuroradiol 1997; 18:633– Rev 2001; 25:465–476 637 45. Garavan H, Ross TJ, Stein EA: Right hemispheric dominance of 25. Achenbach TM: Manual for the Child Behavior Checklist/4–18 inhibitory control: an event-related functional MRI study. Proc and 1991 Profile. Burlington, University of Vermont Depart- Natl Acad Sci U S A 1999; 96:8301–8306 ment of Psychiatry Press, 1991 46. Liddle PF, Kiehl KA, Smith AM: Event-related fMRI study of re- 26. Menon V, Adleman NE, White CD, Glover GH, Reiss AL: Error-re- sponse inhibition. Hum Brain Mapp 2001; 12:100–109 lated brain activation during a Go/NoGo response inhibition 47. Tulving E, Markowitsch HJ, Kapur S, Habib R, Houle S: Novelty task. Hum Brain Mapp 2001; 12:131–143 encoding networks in the : positron emission to- 27. Burock MA, Buckner RL, Woldorff MG, Rosen BR, Dale AM: Ran- mography data. Neuroreport 1994; 5:2525–2528 domized event-related experimental designs allow for ex- 48. Rossell SL, Bullmore ET, Williams SC, David AS: Brain activation tremely rapid presentation rates using functional MRI. Neu- during automatic and controlled processing of semantic rela- roreport 1998; 9:3735–3739 tions: a priming experiment using lexical-decision. Neuropsy- 28. Friston KJ, Zarahn E, Josephs O, Henson RN, Dale AM: Stochas- chologia 2001; 39:1167–1176 tic designs in event-related fMRI. Neuroimage 1999; 10:607– 49. Picard N, Strick PL: Motor areas of the medial wall: a review of 619 their location and functional activation. Cereb Cortex 1996; 6: 29. Dale AM: Optimal experimental design for event-related fMRI. 342–353 Hum Brain Mapp 1999; 8:109–114 50. Nagahama Y, Okada T, Katsumi Y, Hayashi T, Yamauchi H, Oy- anagi C, Konishi J, Fukuyama H, Shibasaki H: Dissociable mech- 30. Kruggel F, von Cramon DY: Temporal properties of the hemo- anisms of attentional control within the human prefrontal cor- dynamic response in functional MRI. Hum Brain Mapp 1999; 8: tex. Cereb Cortex 2001; 11:85–92 259–271 51. Le TH, Pardo JV, Hu X: 4 T-fMRI study of nonspatial shifting of 31. Clark VP, Maisog JM, Haxby JV: An fMRI study of face perception selective attention: cerebellar and parietal contributions. J and memory using random stimulus sequences. J Neurophys- Neurophysiol 1998; 79:1535–1548 iol 1998; 79:3257–3265 52. Ford JM, Sullivan EV, Marsh L, White PM, Lim KO, Pfefferbaum 32. Hayes C, Mathias C: Improved brain coil for fMRI and high res- A: The relationship between P300 amplitude and regional gray olution imaging, in ISMRM 4th Annual Meeting Proceedings. matter volumes depends upon the attentional system en- New York, ISMRM, 1996 gaged. Electroencephalogr Clin Neurophysiol 1994; 90:214– 33. Glover GH, Lai S: Self-navigated spiral fMRI: interleaved versus 228 single-shot. Magn Reson Med 1998; 39:361–368 53. Corbetta M, Shulman GL: Control of goal-directed and stimu- 34. Holmes AP, Friston KJ: Generalizability, random effects, and lus-driven attention in the brain. Nat Rev Neurosci 2002; 3: population inference. Neuroimage 1998; 7:754 201–215 35. Friston KJ, Holmes AP, Worsley KJ, Poline JP, Frith CD, Frackow- 54. Downar J, Crawley AP, Mikulis DJ, Davis KD: The effect of task iak RSJ: Statistical parametric maps in functional imaging: a relevance on the cortical response to changes in visual and au- general linear approach. Hum Brain Mapp 1995; 2:189–210 ditory stimuli: an event-related fMRI study. Neuroimage 2001; 36. Poline JB, Worsley KJ, Evans AC, Friston KJ: Combining spatial 14:1256–1267 extent and peak intensity to test for activations in functional 55. Banich MT, Milham MP, Jacobson BL, Webb A, Wszalek T, Cohen imaging. Neuroimage 1997; 5:83–96 NJ, Kramer AF: Attentional selection and the processing of task- 37. Duvernoy HM, Bourgouin P, Cabanis EA, Cattin F: The Human irrelevant information: insights from fMRI examinations of the Brain: Surface, Three-Dimensional Sectional With Stroop task. Prog Brain Res 2001; 134:459–470 MRI, and Blood Supply. New York, Springer-Verlag, 1999 56. Brass M, Zysset S, von Cramon DY: The inhibition of imitative 38. Tamm L, Menon V, Johnston CK, Hessl DR, Reiss AL: fMRI study response tendencies. Neuroimage 2001; 14:1416–1423 of cognitive interference processing in females with fragile X 57. Dove A, Pollmann S, Schubert T, Wiggins CJ, von Cramon DY: syndrome. J Cognitive Neurosci 2002; 14:160–171 activation in task switching: an event-related 39. Kruggel F, Herrmann CS, Wiggins CJ, von Cramon DY: Hemody- fMRI study. Brain Res Cogn Brain Res 2000; 9:103–109 namic and electroencephalographic responses to illusory fig- 58. Clark VP, Fannon S, Lai S, Benson R: Paradigm-dependent ures: recording of the evoked potentials during functional MRI. modulation of event-related fMRI activity evoked by the odd- Neuroimage 2001; 14:1327–1336 ball task. Hum Brain Mapp 2001; 14:116–127 40. Kirino E, Belger A, Goldman-Rakic P, McCarthy G: Prefrontal ac- 59. Rushworth MF, Krams M, Passingham RE: The attentional role tivation evoked by infrequent target and novel stimuli in a vi- of the left parietal cortex: the distinct lateralization and local- sual target detection task: an event-related functional mag- ization of motor attention in the human brain. J Cogn Neurosci netic resonance imaging study. J Neurosci 2000; 20:6612–6618 2001; 13:698–710 41. Yoshiura T, Zhong J, Shibata DK, Kwok WE, Shrier DA, Numagu- 60. Rushworth MF, Nixon PD, Renowden S, Wade DT, Passingham chi Y: Functional MRI study of auditory and visual oddball RE: The left parietal cortex and motor attention. Neuropsycho- tasks. Neuroreport 1999; 10:1683–1688 logia 1997; 35:1261–1273 42. Kawashima R, Satoh K, Itoh H, Ono S, Furumoto S, Gotoh R, 61. Perry RJ, Zeki S: The neurology of saccades and covert shifts in Koyama M, Yoshioka S, Takahashi T, Takahashi K, Yanagisawa spatial attention: an event-related fMRI study. Brain 2000; 123: T, Fukuda H: Functional anatomy of GO/NO-GO discrimination 2273–2288 and response selection: a PET study in man. Brain Res 1996; 62. Clark CR, Moores KA, Lewis A, Weber DL, Fitzgibbon S, Greenb- 728:79–89 latt R, Brown G, Taylor J: Cortical network dynamics during ver-

1042 ajp.psychiatryonline.org Am J Psychiatry 163:6, June 2006 TAMM, MENON, AND REISS

bal working memory function. Int J Psychophysiol 2001; 42: 65. Gobel S, Walsh V, Rushworth MF: The mental number line and 161–176 the human angular gyrus. Neuroimage 2001; 14:1278–1289 63. Gruber O, Indefrey P, Steinmetz H, Kleinschmidt A: Dissociating 66. Herath P, Kinomura S, Roland PE: Visual recognition: evidence neural correlates of cognitive components in mental calcula- for two distinctive mechanisms from a PET study. Hum Brain tion. Cereb Cortex 2001; 11:350–359 Mapp 2001; 12:110–119 64. Menon V, Rivera SM, White CD, Glover GH, Reiss AL: Dissociating 67. Faillenot I, Decety J, Jeannerod M: Human brain activity related prefrontal and parietal cortex activation during arithmetic pro- to the perception of spatial features of objects. Neuroimage cessing. Neuroimage 2000; 12:357–365 1999; 10:114–124

Am J Psychiatry 163:6, June 2006 ajp.psychiatryonline.org 1043