Neurofunctional Effects of Methylphenidate and
Atomoxetine in Boys with Attention-Deficit/
Hyperactivity Disorder During Time Discrimination
Anna Smith, Ana Cubillo, Nadia Barrett, Vincent Giampietro, Andrew Simmons,
Mick Brammer, and Katya Rubia
Background: The catecholamine agonists methylphenidate and atomoxetine effectively treat attention-deficit/hyperactivity disorder
(ADHD). Furthermore, dopamine agonists have shown to improve time estimation in ADHD, a core cognitive deficit. However, few have
compared the effects of methylphenidate and atomoxetine on brain function in ADHD, and none during time estimation. Using single
dose challenges, we investigated shared and drug-specific effects in ADHD adolescents on the neural substrates of time
Methods: Twenty ADHD adolescent male subjects were compared in a randomized double-blind cross-over design after single doses of
methylphenidate, atomoxetine, and placebo in functional magnetic resonance imaging during TD. Normalization effects were assessed
by comparing brain activation under each drug condition with that of 20 healthy age-matched control subjects.
Results: Relative to control subjects, patients under placebo showed TD deficits and reduced activation of ventrolateral prefrontal cortex
(VLPFC)/insula, inferior frontal cortex, and supplementary motor area. Performance differences were normalized only by methylphe-
nidate, relative to both atomoxetine and placebo. Both medications, however, significantly upregulated right VLPFC/insula activation
within patients and normalized its underactivation in ADHD boys under placebo relative to control subjects. The supplementary motor
area and inferior frontal cortex activation differences that were observed under placebo were reduced by methylphenidate and
atomoxetine, respectively, but neither survived rigorous testing for normalization.
Conclusions: While only methylphenidate had a drug-specific effect of improving TD performance deficits, both drugs significantly
upregulated and normalized right VLPFC underactivation in ADHD boys under placebo relative to control subjects, suggesting shared
effects of stimulants and nonstimulants on a key prefrontal dysfunction during timing.
Key Words: Attention-deficit/hyperactivity disorder, atomoxetine,
supplementary motor area, time discrimination
functions (1) but also in temporal processes (2,3). They are
particularly impaired in fine temporal discrimination (TD), i.e.,
the discrimination of intervals that differ in the millisecond range
(3–5), which has been shown to be one of the best discriminatory
measures for ADHD among a large battery of tasks (5). Using
functional magnetic resonance imaging (fMRI), we have shown
that TD deficits in ADHD adolescents are underpinned by fronto-
striato-cerebellar activation deficits, including right inferior frontal
cortex (IFC), dorsolateral prefrontal cortex (DLPFC) supplementary
ttention-deficit/hyperactivity disorder (ADHD) is defined by
problems with inattention, impulsivity, and hyperactivity
(DSM-IV). Children with ADHD are impaired in executive
motor area (SMA), anterior cingulate cortex (ACC), the basal
ganglia, and cerebellum (2,6,7).
One of the most frequently prescribed medications for ADHD
is the stimulant methylphenidate. Methylphenidate blocks dop-
amine transporters in the striatum and norepinephrine trans-
porters (NET) in NET-rich cortical regions, including prefrontal
cortex, where it increases concentrations of both dopamine (DA)
and norepinephrine (NE) (8). There is a strong association
between DA, the striatum, and fine temporal processes (9): the
striatal DA receptor agonist methylphenidate has been shown to
improve motor timing and time estimation deficits in children
with ADHD in the millisecond (10,11) and second ranges (11,12).
fMRI studies in ADHD patients have shown that single doses of
methylphenidate consistently upregulate and normalize frontos-
triatal activation during cognition (2,13–15). The only fMRI study
investigating the influence of methylphenidate during TD showed
that a single dose of methylphenidate significantly upregulated
and normalized all underactivations observed in ADHD patients
relative to control subjects during placebo in DLPFC, ventrolateral
prefrontal cortex (VLPFC), ACC, and cerebellum (2).
The only other licensed medication for patients with ADHD
is the nonstimulant atomoxetine. While methylphenidate has
immediate effects on behavior, atomoxetine takes up to 6 to 8
weeks to show clinical effects (16). Some studies comparing
chronic dose effects show that methylphenidate is more effective
than atomoxetine (17), while meta-analyses show either compa-
rable efficacy rates in reducing ADHD symptoms or superior
effects of the long-acting methylphenidate preparations only
(18–20). Atomoxetine is a selective presynaptic blocker of NETs
(21), leading to enhanced NE and DA in prefrontal cortex, but also
influences other regions including ACC, thalamus, locus coeruleus,
Authors ASm and AC contributed equally to this work.
Address correspondence to Anna Smith, Ph.D., Institute of Psychiatry,
Department of Child and Adolescent Psychiatry, 16 De Crespigny Park,
Camberwell, London, SE5 8AF, United Kingdom; E-mail: anna.smith@
Received Mar 20, 2012; revised Mar 4, 2013; accepted Mar 11, 2013.
From the Department of Child Psychiatry (ASm, AC, NB, KR), Department
of Neuroimaging (VG, ASi, MB), Centre for Neurodegeneration
Research (ASi), and National Institute for Health Research Biomedical
Research Centre for Mental Health at South London and Maudsley
National Health Service Trust (ASi), Institute of Psychiatry, King’s
College London, United Kingdom.
BIOL PSYCHIATRY 2013;]:]]]–]]]
& 2013 Published by Elsevier Inc
on behalf of Society of Biological Psychiatry
and cerebellum (22). Compared with methylphenidate, however,
atomoxetine has no direct effect upon basal ganglia (22).
In healthy adults, a single dose of atomoxetine increased right
VLPFC/superior temporal lobe activation during inhibitory control
(23,24). We recently compared single dose effects of atomoxetine
with methylphenidate during motor inhibition in ADHD and
showed that both medications upregulated and normalized left
VLPFC, with drug-specific effects of methylphenidate on normal-
izing right VLPFC and cerebellum (25). In this study, we wanted to
test the effects of a single clinical dose of atomoxetine and of
methylphenidate in fMRI during a time discrimination task (6),
measuring another key disorder-sensitive function (2–5) shown to
be mediated by frontostriatal networks (26–28) and modified by
DA agonists in ADHD (10,12).
Given the strong association between DA and frontostriatal
networks in TD (26) and evidence for positive effects of
methylphenidate on time estimation (2,10,12) and its underlying
frontostriatal networks in ADHD (2), we hypothesized that
methylphenidate would enhance TD performance and its asso-
ciated frontostriatal correlates. However, we proposed that atom-
oxetine would also increase ventrolateral prefrontal activation, as
observed in healthy adults during tasks of cognitive control
(23,24) and in children with ADHD during motor inhibition (25).
Methods and Materials
Twenty-eight primarily medication-naive, right-handed adoles-
cent boys between 10 and 17 years old (mean age of final
sample: 12 years, 11 months [SD: 1 year, 7 months]) with a clinical
diagnosis of inattentive/hyperactive-impulsive combined ADHD,
as assessed by an experienced child psychiatrist using the
standardized Maudsley diagnostic interview (29), which assesses
ADHD according to DSM-IV-Text Revision criteria (30), were
recruited from clinics. One patient had a brief medication trial
of methylphenidate 9 months before participation. Eight patients
were excluded due to neurological abnormalities detected during
the scan (n ¼ 1), left handedness (n ¼ 1), braces (n ¼ 1), loss
of data (n ¼ 1), or intolerance to the scanning situation (n ¼ 4).
No participant was excluded due to intolerance of medication
or increased movement. In line with their diagnoses, all patients
scored above clinical cutoff for hyperactive/inattentive symptoms
on the parental Strengths and Difficulties Questionnaire (SDQ)
(31) and the Conners’ Parent Rating Scale-Revised (CPRS-R) (32)
and below clinical cutoff on the Social Communication Ques-
tionnaire (33) to ensure lack of comorbidity with autism spectrum
disorder. They were scanned every Monday between 5:30 PM and
7:30 PM over 3 consecutive weeks using a double-blind, pseudor-
andomized, crossover drug design, receiving a single dose of
either placebo (vitamin C, 50 mg), methylphenidate (Equasym
.3 mg/kg: range 5–20 mg), or atomoxetine (Strattera 1 mg/kg:
range 16–66 mg), all over-encapsulated. Dosages were deter-
mined following National Institute for Health and Clinical Excel-
lence guidelines at the time of the study for typical clinical
efficacious dosages with minimal side effects (http://www.nice.
org.uk/CG72). As suggested by evidence from pharmacokinetics
studies, both medications were administered 1.5 hours before the
scan to allow for maximum absorption (34,35).
Twenty-one right-handed, healthy boys between 10 and 17
years old (mean age of final sample: 13 years, 10 months [SD: 2
years, 4 months]) were recruited through advertisements. They
scored below clinical cutoff for the SDQ, Social Communication
Questionnaire, and CPRS-R. One participant was excluded due to
CPRS-R and SDQ scores above clinical threshold. Control subjects
were scanned once, unmedicated, for feasibility and ethical
reasons. The final subject numbers were therefore 20 ADHD
and 20 control subjects.
Participants were excluded if they had comorbid psychiatric
disorders (except for conduct disorder and oppositional defiant
disorder in the ADHD group: n = 2), including learning disability;
reading, speech, or language disorder; neurological abnormalities;
epilepsy; and drug or substance abuse. Student t tests showed no
group differences for age (t = 1.5; p = ns). All participants had an
IQ ?70 on the Wechsler Abbreviated Scale of Intelligence (36),
but significant group differences were observed (mean control
subjects: 113 ; mean ADHD: 91 ; t ¼ 6.3; p ? .0001), not
unexpected given that low IQ scores are typical in ADHD (37,38).
IQ was therefore included as a covariate in case-control analyses.
Participants were paid £50 for each visit. Written informed
consent and assent were obtained and the study was approved
by the local ethics committee.
fMRI Time Discrimination Task
After one practice session outside the scanner, the TD task was
visually presented in the magnetic resonance imaging scanner via
a prism from a liquid crystal diode projector. The 5-minute block-
design task consisted of 5 ? 30-sec alternated blocks for two
conditions: TD (active condition) and temporal order judgment
(TOJ) (control condition), which was always presented first. The
TD condition began with the appearance of a centrally located
grey circle (5 cm in diameter) with the letter L for 3 sec. This was
followed by two equally sized red (left side of screen) and green
(right side of screen) circles, appearing consecutively with no
intermittent pause and in random order. One circle was randomly
presented for 1 sec, and the comparison circle was presented for
either 1.3 sec, 1.4 sec, or 1.5 sec, with two trials for each compari-
son and with 2.1 sec response time for each trial. The subjects
were told that in this experimental condition indicated by the
letter L, they had to decide which circle stayed on the screen for
the longest time by responding with a left-sided button if the red
circle (displayed left) lasted longest or a right-sided button if the
green circle (displayed right) lasted longest.
The TOJ (control) condition was presented identically. The
only difference was that these blocks began with the presenta-
tion of the number 2 and required subjects to indicate which
circle came second using the same response buttons as descri-
Task Performance Analysis. Repeated measures analyses of
variance (ANOVAs) were conducted to test for within-group
effects of drug condition on TD and TOJ errors and to test for
practice effects. Case-control ANOVAs were conducted to com-
pare healthy control subjects and patients for each of the three
medication conditions and for each task condition (TD, TOJ) (6).
Magnetic Resonance Imaging Acquisition and Analysis
Gradient-echo echo-planar imaging data were acquired on a
GE Signa 3T Horizon DHx system (General Electric, Milwaukee,
Wisconsin) at the Centre for Neuroimaging Sciences, Institute of
Psychiatry, Kings’ College London, United Kingdom. A semiauto-
mated quality control procedure ensured consistent image
quality (39). A quadrature birdcage head coil was used for radio
frequency transmission and reception. In each of 48 noncontig-
uous planes parallel to the anterior-posterior commissure line,
100 T2*-weighted magnetic resonance images depicting blood
oxygen level–dependent (BOLD) contrast covering the whole
BIOL PSYCHIATRY 2013;]:]]]–]]]
A. Smith et al.
brain were acquired with echo time ¼ 30 msec, repetition
time ¼ 3 sec, flip angle ¼ 90º, in-plane resolution ¼ 3.1 mm,
slice thickness ¼ 3.0 mm, and slice skip ¼ .3 mm. This echo-planar
imaging dataset provided complete coverage.
We used the nonparametric XBAM software (http://www.
brainmap.co.uk) developed at the Institute of Psychiatry, Kings
College, London (40–42), which uses median statistics to control
outlier effects and permutation rather than normal theory based
inference, recommended for fMRI (43). Furthermore, the most
common test statistic is computed by standardizing for individual
difference in residual noise before embarking on second-level,
multisubject testing using robust permutation-based methods.
This allows a mixed-effects approach to analysis recommended
After preprocessing (Supplement 1), time series analysis for
each individual subject was based on a previously published
wavelet-based data resampling method for fMRI data (41,42).
After first-level analysis, the individual statistical maps were then
normalized into Talairach standard space (42). A group brain
activation map was then produced for the experimental condition
(TD–TOJ) and hypothesis testing was carried out at the cluster level.
The detection of activated voxels is extended from voxel to
cluster level using a two-pass method (41). We first used a voxel-
level threshold of p ? .05 to give maximum sensitivity and to
avoid type II errors. Three-dimensional clusters were then built by
joining together adjacent significant voxels. Cluster mass (rather
than a cluster extent) threshold was used as second-pass cluster
statistic, to minimize discrimination against possible small,
strongly responding foci of activation (41). The cluster-level
threshold was then computed in such a way as to ensure that
the final expected number of type I error clusters was less than
one per whole brain.
Between-Group Analyses between Control Subjects and
ADHD Under Each Drug Condition
For case-control comparisons, a series of three analyses of
covariance (with IQ as covariate) were carried out comparing
control subjects with 1) ADHD under placebo; 2) ADHD under
methylphenidate; and 3) ADHD under atomoxetine. We used a
region of interest (ROI) analysis where we selected brain regions
that have consistently been found in previous research to be
underactivated in children with ADHD during TD (2,6), including
our recent meta-analysis of fMRI studies of timing (7). For this
purpose, we applied the Talairach Client (Research Imaging
Institute, University of Texas Health Science Center, San Antonio)
(44,45) to determine a predefined mask of these regions that
included the frontal lobes, including ACC/SMA; the cerebellum;
and the striatum (2,6,7). Statistical measures of BOLD response
were then extracted for each participant in each of the clusters of
between-group differences and post hoc t tests (corrected using
least significant difference) were conducted to clarify the direc-
tion of the differences.
Within-ADHD Repeated Measures Analyses of Drug
To test for upregulation effects of both drugs on brain areas
related to TD within the ADHD group, a repeated measures
ANOVA was carried out to test for effects of drug condition
(placebo, methylphenidate, atomoxetine) on brain activation
within patients using the same predefined regions in the same
ROI mask, as described above.
To test for potential order effects, repeated measures ANOVAs
were then conducted within patients on the extracted BOLD
response measures of the resulting activations.
To test whether the between-group and within-group differ-
ences in brain activation were related to performance, statistical
measures of BOLD response (sum of squares ratios) were
extracted for each participant in each of the clusters of
between-group and within-group differences.
Repeated measures ANOVAs within ADHD under each medi-
cation condition were significant for TD errors (F2,39 = 4.8;
p ? .02), due to fewer TD errors in ADHD patients during the
methylphenidate condition relative to placebo (p ? .06) and
atomoxetine (p ? .03), but not for TOJ errors (F2,39= .1, p = .48)
A series of ANOVAs comparing errors between control and
ADHD boys under each medication condition revealed for the
ADHD placebo-control comparison, a significant effect of con-
dition (F1,38¼ 52; p ? .0001) and a group by condition interaction
(F1,38¼ 6.8; p ? .013) and for the ADHD atomoxetine-control
comparison, a significant effect of condition (F1,38¼ 57; p ? .0001)
and a group by condition interaction (F1,38 ¼ 6.6; p ? .014).
However, under methylphenidate, there was a main effect of
condition (F1,38¼ 40; p ? .0001) but no significant group by
condition interaction (F1,38¼ 2.7; p ? .107). The significant effect
of condition was explained in each case by superior performance
for TOJ compared with TD for all subjects. The significant group
by condition interactions were attributable to ADHD patients
making more errors than control subjects in TD but not TOJ under
placebo (p ? .02) and under atomoxetine (p ? .02) but not under
methylphenidate (p ? .16) (Table 1).
Motion. Multivariate analyses of variance showed no signifi-
cant effects in the extent of three-dimensional motion as
measured by translation (voxels) for x, y, and z axes for the
comparison of control subjects and ADHD children under each
Within-Group Activation. Within-group
results are described in the text and Figure S1 in Supplement 1.
Table 1. Mean Percentage of Errors for Temporal Discrimination and Temporal Order Judgment for Control Subjects and Children with ADHD During
Placebo, Methylphenidate, and Atomoxetine
Control SubjectsADHD (Placebo)ADHD (Methylphenidate)ADHD (Atomoxetine)
Errors in Time Discrimination Trials (%)
Errors in Order Discrimination Trials (%)
ADHD, attention-deficit/hyperactivity disorder; ANOVA, analysis of variance.
aSignificant group by condition interaction effect during ANOVA comparison of control subjects and patients, showing enhanced time discrimination
errors in patients relative to control subjects.
A. Smith et al.
BIOL PSYCHIATRY 2013;]:]]]–]]]
ROI Analyses Results for Between-Group Comparisons of
Healthy Control Subjects and ADHD Boys Under Placebo,
Methylphenidate, or Atomoxetine
Under placebo, ADHD boys compared with control subjects
showed reduced activation in right VLPFC (Brodmann area [BA]
47)/insula, right IFC (BA 45), and SMA/ACC (Figure 1, Table 2).
Under methylphenidate, ADHD boys compared with control
subjects showed reduced activation in right IFC but no longer in
right VLPFC/insula. The activation differences in the ACC cluster
were still observed but reduced in size and no longer included
the SMA (Figure 1, Table 2). No areas were enhanced in activation
for the ADHD boys compared with control subjects.
Under atomoxetine, activation differences between ADHD
boys and control subjects were no longer observed in either
right VLPFC/insula or right IFC, but SMA/ACC activation was still
reduced compared with control subjects (as observed during
placebo) (Figure 1, Table 2). No areas of enhanced activation
for the ADHD group compared with healthy subjects were
To test for the statistical significance of these apparent normal-
ization effects of each drug on case-control activation differences
observed under placebo, we used nonparametric Friedman two-
way analysis of variance by ranks on the extracted BOLD responses
during each medication condition for each of the three clusters
shown to be significantly different in the comparison between
control subjects and children with ADHD during placebo. We
conducted this test only within patients, given that control subjects
were only tested once and hence constant across comparisons.
We found that the BOLD response within the right VLPFC/insula
cluster was significantly different for each medication condition
(p ? .043). Post hoc analysis using Wilcoxon signed-rank tests
showed that this significant difference was explained by signifi-
cantly lower BOLD response for patients during placebo compared
with methylphenidate (p ? .011) and atomoxetine (p ? .013).
Figure 1. Transversal images of the between-group analysis of variance comparison between healthy control boys and boys with attention-deficit/
hyperactivity disorder on (A) placebo, (B) methylphenidate, and (C) atomoxetine during time discrimination. Statistical threshold selected at p ? .05 for
voxel and p ? .01 for cluster levels. Slices are marked with the z coordinate as distance in millimeters from the anterior-posterior commissure.
Table 2. Between-Group ANOVA Results Showing Differences between Control Subjects and Boys with ADHD
Under Either Placebo, Methylphenidate, or Atomoxetine for the Contrast of TD Versus Order Judgment
Number of Voxelsxyz
Control Subjects ? ADHD Under Placebo
Control Subjects ? ADHD Under Methylphenidate
Control Subjects ? ADHD Under Atomoxetine
R and L SMA/anterior cingulate cortex
R inferior frontal cortexa
32R and L anterior cingulate cortex
R inferior frontal cortex45/44
32/6R and L SMA/anterior cingulate cortex
R and L SMA76
ADHD, attention-deficit/hyperactivity disorder; ANOVA, analysis of variance; BA, Brodmann area; L, left;
R, right; SMA, supplementary motor area; TD, temporal discrimination; VLPFC, ventrolateral prefrontal cortex.
aAreas that were no longer observed to be abnormal by atomoxetine. Only the cluster in VLPFC, which was
no longer observed under either drug, reached significance after rigorous normalization testing.
bAreas that were no longer observed to be abnormal by methylphenidate.
BIOL PSYCHIATRY 2013;]:]]]–]]]
A. Smith et al.
ROI Results for Within-Patients Comparison between Placebo,
Methylphenidate, and Atomoxetine
Friedman’s two-way ANOVA on the preselected ROI regions
showed a significant effect of drug condition in right VLPFC/insula
activation that reached into the head of putamen, in a similar
location to the cluster that was reduced in ADHD under placebo
(11 voxels; peak Talairach coordinates [x, y, z]: 40, 19, 4; BA 45/47;
p ? .02) (Figure 2). Within-subjects contrasts revealed that this
was attributable to enhanced activation when ADHD boys were
under methylphenidate (p ? .004) and atomoxetine (p ? .02),
relative to placebo. No differences were observed between
activations during methylphenidate and atomoxetine.
Correlations between Brain Activation and Performance
There were no significant correlations between TD errors and
the statistical measure of brain activation extracted for the cluster
of VLPFC activation that showed differences across the three
medication conditions in the within-group analysis or for any of
the clusters that differed between ADHD and control subjects
under either medication condition.
Practice Effects. A repeated measures ANOVA within patients
showed that the order of drug administration had no significant
effect on TD performance or BOLD response in the VLPFC cluster of
activation difference in the within-group analysis (Table S1 in
Supplement 1). Although we could not directly measure practice
effects in the between-group analyses due to low power, given that
there were no differences in the within-group analysis, we assume
that they were unlikely to have contributed to performance or brain
activation differences between patients and control subjects.
This study demonstrates a relatively superior effect of a single
dose of methylphenidate compared with placebo and atom-
oxetine on TD performance in ADHD boys but shared effects on
the underlying neurofunctional networks of TD. Compared with
control subjects, ADHD boys under placebo made significantly
more TD but not TOJ errors and had reduced activation in typical
areas of TD in right VLPFC/insula, right IFC, and SMA/ACC.
Methylphenidate relative to atomoxetine and placebo signifi-
cantly decreased TD errors within patients, while only methyl-
phenidate, but not atomoxetine, normalized the TD deficits
relative to control subjects. Within-ADHD comparisons showed
that both medications significantly upregulated right VLPFC/
insula activation compared with placebo. In line with this, case-
control comparisons showed that both medications significantly
normalized right VLPFC/insula underactivation that was observed
in ADHD adolescents under placebo relative to control subjects.
The findings show that both methylphenidate and atomoxetine
upregulate and normalize a key right ventrolateral fronto-insular
area for TD in ADHD.
Attention-deficit/hyperactivity disorder patients had reduced
activation relative to control subjects under placebo in key areas
that have consistently been found to be involved in time
perception in adults (27,28,46–48) and adolescents (48), i.e., in
right VLPFC, IFC, insula, SMA, and ACC. The findings replicate, in
a larger sample, previous findings of right VLPFC and DLPFC
underactivation during the same task in ADHD adolescents (2,6).
The finding of a significant upregulation and normalization
effect of methylphenidate on the underactivation in right VLPFC/
insula in ADHD boys during placebo replicates previous findings
of neurofunctional upregulation and normalization with methyl-
phenidate in this region during the same TD task (2). However, in
this study, we show for the first time that the upregulation and
normalization effects of methylphenidate on right VLPFC are
shared with atomoxetine. Right VLPFC and the insula are key
areas of temporal discrimination (3,27), as well as key regions that
have been shown to be consistently underactivated in ADHD
children during TD (2,6,49) and during other functions such as
attention and inhibition (49–53). It has been argued that VLPFC
is part of a more generic cognitive control network that subserves
several cognitive functions mediated by different VLPFC-striatal
neural networks (54), including both executive and timing
The normalization effect of right VLPFC underactivation in
ADHD relative to control subjects by atomoxetine echoes pre-
vious evidence for upregulation with atomoxetine of this region
in healthy adults during motor inhibition and performance
monitoring (23,24), suggesting similar mechanisms of action of
atomoxetine on upregulating right VLPFC activation in healthy
subjects and ADHD patients.
The shared normalization effects of both drugs on right VLPFC
underactivation in ADHD is intriguing, given the superior per-
formance effects of acute methylphenidate, which parallel the
typically faster behavioral effects of this stimulant over atom-
oxetine (16,55). The findings are important, as they suggest
shared mechanisms of action on a key area that has been shown
to be consistently underactivated in ADHD patients during timing,
as well as during other cognitive functions (7,14,52,53), and that
furthermore has been shown to be disorder-specifically under-
activated in ADHD relative specifically to other child psychiatric
disorders (51,56,57). They extend our recent findings of shared
upregulation and normalization effects of both drugs on left
VLPFC underactivation in ADHD relative to control boys during a
motor inhibition task (25) and may be the underlying mechanism
of action for the relatively comparable efficacy rates of both drugs
on ADHD behaviors (17,18). These shared effects on ventrolateral
frontal regions likely reflect both noradrenergic and dopaminer-
gic mechanisms, given that in frontal regions, both drugs
upregulate NE equally or more than DA via reuptake inhibition
of NETs that clear DA and NE (8,58–61). These shared neurofunc-
tional modulation effects in VLPFC cortex are hence in line with
Figure 2. Transversal images of the within-group analysis of variance,
showing areas of increased brain activation in boys with attention-deficit/
hyperactivity disorder with an acute dose of methylphenidate compared
with placebo and atomoxetine during time discrimination contrasted with
order judgment. Statistical threshold set at p ? .05 for voxel-wise and p ?
.01 for cluster-wise analysis. Slices are marked with the z coordinate as
distance in millimeters from the anterior-posterior commissure. Mean
statistical blood oxygen level–dependent response is shown for each drug
condition within ventrolateral prefrontal cortex (Brodmann area 45/47).
A. Smith et al.
BIOL PSYCHIATRY 2013;]:]]]–]]]
the relatively similar catecholaminergic mechanisms of action of
both drugs on frontal regions.
It is of note that SMA underactivation was no longer observed
under methylphenidate, while right IFC underactivation was no
longer observed under atomoxetine. However, neither region
survived our rigorous testing for normalization, and hence, unless
replicated in future studies, these potential drug-specific modu-
lation effects need to be considered with caution.
The drug-specific enhanced performance effects of methyl-
phenidate are in line with our hypothesis. While the positive
effects of dopamine agonists on TD are well documented (26), the
relationship between NE and timing functions is less studied,
although there is some evidence for improved TD in healthy adults
under a single dose of the NE reuptake inhibitor reboxetine (62).
These drug-specific performance findings parallel evidence for the
immediate behavioral effect of methylphenidate (55) relative to
atomoxetine, which takes 6 to 8 weeks to show clinical effects (16).
We have previously shown that brain activation is more sensitive
than behavior to catecholaminergic drug effects (2,13–15,25),
which could explain why brain upregulation effects under atom-
oxetine appear before behavioral effects. It is possible that
methylphenidate operates within a wider DA-innervated network
of timing, involving the right VLPFC as well as the DA- innervated
striatum and the SMA, which is a crucial region mediating TD (27),
whereas the effects of atomoxetine were restricted to right fronto-
cortical parts of the network, which may not have been sufficient to
elicit performance benefits. However, given that the amelioration
effects on SMA with methylphenidate were not significant in our
rigorous testing for normalization, this has to remain a speculation.
Future studies should test for comparative effects of both
drugs on brain activation and performance under longer chronic
administration to adjust for differences in time delays to behav-
A limitation of the study is the investigation of single rather than
chronic doses of each drug. As a single dose challenge, this design
reduces long-term confounds such as symptomatic improvement,
side effects, or chronic effects on brain activation. However, it may
be biased toward methylphenidate, given the immediacy of
behavioral effects, while atomoxetine takes maximum clinical
efficacy after 6 to 8 weeks of chronic administration (16), possibly
explaining normalization of TD performance by methylphenidate
only (see above). Future studies should compare long-term effects
to accommodate for differences in time to maximum efficacy.
Another limitation is that for ethical reasons, control subjects
were only tested once, while patients were tested three times.
However, for the within-subjects analysis, potential practice
effects were controlled for by the counterbalanced design, and
furthermore, no practice effects were observed.
To conclude, to our knowledge, this is one of the first studies
to directly compare neurofunctional effects of atomoxetine and
methylphenidate in ADHD boys and the first to do so during TD.
The study is strengthened by the recruitment of medication-naïve
ADHD boys, thus controlling for the confounding long-term
stimulant medication effects on brain activation and structure
(7,53,63,64). We found that while only methylphenidate normal-
ized TD performance deficits, both drugs showed significant
upregulation and normalization effects in a key area of TD in
right VLPFC. The findings show that both drugs are equally
efficient in upregulating and normalizing right fronto-cortical
areas of timing in ADHD.
This work was supported by grants by the National Institute of
Health Research Biomedical Research Centre for Mental Health at
South London and Maudsley National Health Service Foundation
Trust and Institute of Psychiatry, Kings College London, and Lilly
Pharmaceuticals. Dr. Anna Smith, Ms. Ana Cubillo, and Dr. Andrew
Simmons are supported by the National Institute of Health Research
Biomedical Research Centre. Lilly Pharmaceuticals had no input into
the design, analysis, data interpretation, or write-up.
KR has received speaker’s honoraria from Lilly, Shire, Novartis,
and Medice. The other authors report no biomedical financial
interests or potential conflicts of interest.
Supplementary material cited in this article is available online at
1. Willcutt E, Doyle A, Nigg J, Faraone S, Pennington B (2005): Validity of
the executive function theory of attention-deficit/hyperactivity dis-
order: A meta-analytic review. Biol Psychiatry 57:1336–1346.
2. Rubia K, Halari R, Christakou A, Taylor E (2009): Impulsiveness as a
timing disturbance: Neurocognitive abnormalities in attention-deficit
hyperactivity disorder during temporal processes and normalization
with methylphenidate. Phil Trans R Soc Lond B Biol Sci 364:1919–1931.
3. Noreika V, Falter CM, Rubia K (2012): Timing deficits in attention-
deficit/hyperactivity disorder (ADHD): Evidence from neurocognitive
and neuroimaging studies. Neuropsycholia 51:235–266.
4. Smith A, Taylor E, Warner Rogers J, Newman S, Rubia K (2002):
Temporal processing in children with ADHD: Evidence for a pure time
perception deficit in ADHD. J Child Psychol Psychiatry 43:529–542.
5. Rubia K, Smith A, Taylor E (2007): Performance of children with ADHD
on a test battery of impulsiveness. Child Neuropsychol 13:276–304.
6. Smith AB, Taylor E, Brammer M, Halari R, Rubia K (2008): Reduced
activation in right lateral prefrontal cortex and anterior cingulate
cortex in medication-naïve adolescents with attention deficit hyper-
activity disorder during time discrimination. J Child Psychol Psychiatry
7. Hart H, Radua J, Mataix-Cols D, Rubia K (2012): Meta-analysis of fMRI
studies of timing in attention-deficit hyperactivity disorder (ADHD).
Neurosci Biobehav Rev 36:2248–2256.
8. Hannestad J, Gallezot J-D, Planeta-Wilson B, Lin S-F, Williams WA, van
Dyck CH, et al. (2010): Clinically relevant doses of methylphenidate
significantly occupy norepinephrine transporters in humans in vivo.
Biol Psychiatry 68:854–860.
9. Takahashi T (2007): Hyperbolic discounting may be reduced to
electrical coupling in dopaminergic neural circuits. Med Hypotheses 69:
10. Rubia K, Smith AB, Noorlos J, Gunning W, Sergeant JA (2003): Motor
timing deficits in community and clinical boys with hyperactive
behaviour: The effect of methylphenidate on motor timing. J Abnorm
Child Psychol 31:301–313.
11. Ben-Pazi H, Shalev RS, Gross-Tsur V, Bergman H (2006): Age and
medication effects on rhythmic responses in ADHD: Possible oscil-
latory mechanisms? Neuropsychology 44:412–416.
12. Baldwin RL, Chelonis JJ, Flake RA, Edwards MC, Feild CR, Meaux JB,
Paule MG (2004): Effect of methylphenidate on time perception in
children with attention-deficit/hyperactivity disorder. Exp Clin Psycho-
13. Rubia K, Halari R, Cubillo A, Mohammad A-M, Brammer M, Taylor E
(2009): Methylphenidate normalises activation and functional con-
nectivity deficits in attention and motivation networks in medication-
naïve children with ADHD during a rewarded continuous performance
task. Neuropharmacology 57:640–652.
14. Rubia K, Halari R, Cubillo A, Smith AB, Mohammad A-M, Brammer M,
Taylor E (2011): Methylphenidate normalizes fronto-striatal under-
activation during interference inhibition in medication-naive boys
with attention-deficit hyperactivity disorder. Neuropsychopharmacol-
15. Rubia K, Halari R, Mohammad A, Taylor E, Brammer M (2011):
Methylphenidate normalises fronto-cingulate underactivation during
error processing in attention-deficit hyperactivity disorder. Biol Psy-
16. Montoya A, Hervas A, Cardo E, Artigas J, Mardomingo MJ, Alda JA,
et al. (2009): Evaluation of atomoxetine for first-line treatment of
newly diagnosed, treatment-naive children and adolescents with
BIOL PSYCHIATRY 2013;]:]]]–]]]
A. Smith et al.
attention deficit/hyperactivity disorder. Curr Med Res Opin 25:
17. Newcorn JH, Kratochvil CJ, Allen AJ, Casat CD, Ruff DD, Moore RJ,
Michelson D (2008): Atomoxetine and osmotically released methyl-
phenidate for the treatment of attention deficit hyperactivity disorder:
Acute comparison and differential response. Am J Psychiatry 165:
18. Hazell PL, Kohn MR, Dickson R, Walton RJ, Granger RE, van Wyk GW
(2011): Core ADHD symptom improvement with atomoxetine versus
methylphenidate: A direct comparison meta-analysis. J Atten Disord
19. Hanwella R, Senanayake M, de Silva V (2011): Comparative efficacy
and acceptability of methylphenidate and atomoxetine in treatment
of attention deficit hyperactivity disorder in children and adolescents:
A meta-analysis. BMC Psychiatry 11:176.
20. Cottrell S, Tilden D, Robinson P, Bae J, Arellano J, Edgell E, et al. (2008):
A modeled economic evaluation comparing atomoxetine with stimu-
lant therapy in the treatment of children with attention-deficit/
hyperactivity disorder in the United Kingdom. Value Health 11:
21. Bymaster FP, Katner JS, Nelson DL, Hemrick-Luecke SK, Threlkeld PG,
Heiligenstein JH, et al. (2002): Atomoxetine increases extracellular
levels of norepinephrine and dopamine in prefrontal cortex of rat: A
potential mechanism for efficacy in attention deficit/hyperactivity
disorder. Neuropsychopharmacology 27:699–711.
22. Takano A, Gulyas B, Varrone A, Maguire RP, Halldin C (2009): Saturated
norepinephrine transporter occupancy by atomoxetine relevant to
clinical doses: A rhesus monkey study with (S,S)-[(18)F]FMeNER-D (2).
Eur J Nucl Med Mol Imaging 36:1308–1314.
23. Chamberlain SR, Hampshire A, Müller U, Rubia K, del Campo N, Craig K,
et al. (2009): Atomoxetine modulates right inferior frontal activation
during inhibitory control: A pharmacological functional magnetic
resonance imaging study. Biol Psychiatry 65:550–555.
24. Graf H, Abler B, Freudenmann R, Beschoner P, Schaeffeler E, Spitzer M,
et al. (2011): Neural correlates of error monitoring modulated by
atomoxetine in healthy volunteers. Biol Psychiatry 69:890–897.
25. Cubillo A, Smith AB, Barrett N, Giampietro V, Brammer MJ, Simmons A,
Rubia K (2012): Shared and drug-specific effects of atomoxetine and
methylphenidate on inhibitory brain dysfunction in medication-naive
ADHD boys [published online ahead of print October 9]. Cereb Cortex.
26. Rubia K (2006): The neural correlates of timing functions. In: Glicksohn J,
Myslobodsky MS, editors. Timing the Future: The Case for a Time-Based
Prospective Memory. Hackensack, NJ: World Scientific Publishing.
27. Wiener M, Turkeltaub P, Coslett HB (2010): The image of time: A voxel-
wise meta-analysis. Neuroimage 49:1728–1740.
28. Rubia K, Smith A (2004): The neural correlates of cognitive time
management: A review. Acta Neurobiol Exp (Wars) 64:329–340.
29. Goldberg D, Murrey R (2002): Maudsley Handbook of Practical
Psychiatry. Oxford, UK: Oxford University Press.
30. American Psychiatric Association (1994): Diagnostic and Statistical
Manual of Mental Disorders, 4th ed. Washington DC: American
31. Goodman R, Ford T, Simmons H, Gatward R, Meltzer H (2000): Using
the Strengths and Difficulties Questionnaire (SDQ) to screen for child
psychiatric disorders in a community sample. Br J Psychiatry 177:
32. Conners CK, Sitarenios G, Parker JDA, Epstein JN (1998): The Revised
Conners' Parent Rating Scale (CPRS-R): Factor structure, reliability, and
criterion validity. J Abnorm Child Psychol 26:257–268.
33. Rutter M, Bailey A, Lord C (2003): Social Communication Questionnaire.
Los Angeles: Western Psychological Services.
34. Witcher JW, Long A, Smith B, Sauer JM, Heilgenstein JH, Wilens T, et al.
(2003): Atomoxetine pharmacokinetics in children and adolescents
with attention deficit hyperactivity disorder. J Child Adolesc Psycho-
35. Swanson JM, Volkow ND (2002): Pharmacokinetic and pharmacody-
namic properties of stimulants: Implications for the design of new
treatments for ADHD. Behav Brain Res 130:73–78.
36. Wechsler D (1999): Wechsler Abbreviated Scale of Intelligence. San
Antonio, TX: The Psychological Corporation.
37. Kuntsi J, Eley TC, Taylor A, Hughes C, Asherson P, Caspi A, Moffitt TE
(2004): Co-occurrence of ADHD and low IQ has genetic origins. Am J
Med Genet B Neuropsychiatr Genet 124B:41–47.
38. Polderman TJC, Gosso MF, Posthuma D, van Beijsterveldt T, Heutink P,
Verhurst F, Boomsma DI (2006): A longitudinal twin study on IQ,
executive functioning, and attention problems during childhood and
early adolescence. Acta Neurol Belg 106:191–207.
39. Simmons A, Moore E, Williams SCR (1999): Quality control for func-
tional MRI using automated data analysis and Shewhart charting.
Magn Reson Med 41:1274–1278.
40. Brammer MJ, Bullmore ET, Simmons A, Williams SC, Grasby PM,
Howard RJ, et al. (1997): Generic brain activation mapping in func-
tional magnetic resonance imaging: A nonparametric approach. Magn
Reson Imaging 15:763–770.
41. Bullmore ET, Brammer MJ, Rabe-Hesketh S (1999): Methods for
diagnosis and treatment of stimulus-correlated motion in generic
brain activation studies using fMRI. Hum Brain Mapp 7:38–48.
42. Bullmore E, Long C, Suckling J, Fadili J, Calvert G, Zelaya F, et al. (2001):
Colored noise and computational inference in neurophysiological
(fMRI) time series analysis: Resampling methods in time and wavelet
domains. Hum Brain Mapp 12:61–78.
43. Thirion B, Pinel P, Meriaux S, Roche A, Dehaene S, Poline J-B (2007):
Analysis of a large fMRI cohort: Statistical and methodological issues
for group analysis. Neuroimage 35:105–120.
44. Lancaster JL, Woldorff MG, Parsons LM, Liotti M, Freitas CS, Rainey L,
et al. (2000): Automated Talairach atlas labels for functional brain
mapping. Hum Brain Mapp 10:120–131.
45. Lancaster JL, Rainey LH, Summerlin JL, Freitas CS, Fox PT, Evans AC,
et al. (1997): Automated labeling of the human brain: A preliminary
report on the development and evaluation of a forward-transform
method. Hum Brain Mapp 5:238–242.
46. Wittmann M, van Wassenhove V (2009): The experience of time:
Neural mechanisms and the interplay of emotion, cognition and
embodiment. Philos Trans R Soc Lond B Biol Sci 364:1809–1813.
47. Smith AB, Taylor E, Lidzba K, Rubia K (2003): A right hemispheric
frontocerebellar network for time discrimination of several hundreds
of milliseconds. Neuroimage 20:344–350.
48. Smith A, Giampietro V, Brammer M, Taylor E, Simmons AN, Rubia K
(2011): Functional development of fronto-striato-parietal networks
associated with time perception. Front Hum Neurosci 5:136.
49. Rubia K (2011): “Cool” inferior frontostriatal dysfunction in attention deficit
hyperactivity disorder (ADHD) versus “hot” ventromedial orbitofronto-
limbic dysfunction in conduct disorder. Biol Psychiatry 69:e69–e87.
50. Smith A, Taylor E, Brammer M, Toone B, Rubia K (2006): Task specific
hypoactivation in prefrontal and temporoparietal brain regions during
motor inhibition and task switching in medication-naïve children and
adolescents with ADHD. Am J Psychiatry 163:1044–1051.
51. Rubia K, Halari R, Smith A, Majeed M, Scott S, Brammer M (2009):
Shared and disorder-specific prefrontal abnormalities in boys with
pure attention-deficit/hyperactivity disorder compared to boys with
pure CD during interference inhibition and attention allocation. J Child
Psychol Psychiatry 50:669–678.
52. Rubia K, Smith A, Brammer M, Toone B, Taylor E (2005): Abnormal brain
activation during inhibition and error detection in medication-naive
adolescents with ADHD. Am J Psychiatry 162:1067–1075.
53. Hart H, Radua J, Nakao T, Mataix-Cols D, Rubia K (2013): Meta-analysis
of functional magnetic resonance imaging studies of inhibition and
attention in Attention-deficit/Hyperactivity Disorder: Exploring task-
specific, stimulant medication, and age effects. JAMA Psychiatry 70:
54. Cole MW, Schneider W (2007): The cognitive control network:
Integrated cortical regions with dissociable functions. Neuroimage 37:
55. Greenhill LL, Swanson JM, Vitiello B, Davies M, Clevenger W, Wu M,
et al. (2001): Impairment and deportment responses to different
methylphenidate doses in children with ADHD: The MTA titration trial.
J Am Acad Child Adolesc Psychiatry 40:180–187.
56. Rubia K, Halari R, Smith AB, Mohammed M, Scott S, Gampietro V, et al.
(2008): Dissociated functional brain abnormalities of inhibition in
boys with pure conduct disorder and in boys with pure attention
deficit hyperactivity disorder. Am J Psychiatry 165:889–897.
57. Rubia K, Cubillo A, Smith AB, Woolley J, Heyman I, Brammer MJ (2010):
Disorder-specific dysfunction in right inferior prefrontal cortex during
two inhibition tasks in boys with attention-deficit hyperactivity
disorder compared to boys with obsessive-compulsive disorder.
Hum Brain Mapp 31:287–299.
A. Smith et al.
BIOL PSYCHIATRY 2013;]:]]]–]]]
58. Balcioglu A, Ren J-Q, McCarthy D, Spencer TJ, Biederman J, Bhide PG Download full-text
(2009): Plasma and brain concentrations of oral therapeutic doses of
methylphenidate and their impact on brain monoamine content in
mice. Neuropharmacology 57:687–693.
59. Berridge CW, Devilbiss DM, Andrzejewski ME, Arnsten AFT, Kelley AE,
Schmeichel B, et al. (2006): Methylphenidate preferentially increases
catecholamine neurotransmission within the prefrontal cortex at
low doses that enhance cognitive function. Biol Psychiatry 60:
60. Moron JA, Brockington A, Wise R, Rocha BA, Hope BT (2002):
Dopamine uptake through the norepinephrine transporter in brain
regions with low levels of the dopamine transporter: Evidence from
knock-out mouse lines. J Neurosci 22:389–395.
61. Yamamoto BK, Novotney S (1998): Regulation of extracellular dop-
amine by the norepinephrine transporter. J Neurochem 71:274–280.
62. Rammsayer TH, Hennig J, Haag A, Lange N (2001): Effects of
noradrenergic activity on temporal information processing in humans.
Q J Exp Psychol B 54:247–258.
63. Nakao T, Radua J, Rubia K, Mataix-Cols D (2011): Gray matter volume
abnormalities in ADHD: Voxel-based meta-analysis exploring the
effects of age and stimulant medication. Am J Psychiatry 168:
64. Konrad K, Neufang S, Fink GR, Herpertz-Dahlmann B (2007): Long-term
effects of methylphenidate on neural networks associated with execu-
tive attention in children with ADHD: Results from a longitudinal
functional MRI study. J Am Acad Child Adolesc Psychiatry 46:1633–1641.
BIOL PSYCHIATRY 2013;]:]]]–]]]
A. Smith et al.