Reduced Interhemispheric Resting State FunctionalConnectivity in Cocaine AddictionClare Kelly, Xi-Nian Zuo, Kristin Gotimer, Christine L. Cox, Lauren Lynch, Dylan Brock, Davide Imperati,Hugh Garavan, John Rotrosen, F. Xavier Castellanos, and Michael P. Milham
Background: Models of cocaine addiction emphasize the role of disrupted frontal circuitry supporting cognitive control processes.However, addiction-related alterations in functional interactions among brain regions, especially between the cerebral hemispheres, arerarely examined directly. Resting-state functional magnetic resonance imaging (fMRI) approaches, which reveal patterns of coherentspontaneous fluctuations in the fMRI signal, offer a means to quantify directly functional interactions between the hemispheres. Weexamined interhemispheric resting-state functional connectivity (RSFC) in cocaine dependence using a recently validated approach,voxel-mirrored homotopic connectivity.
Methods: We compared interhemispheric RSFC between 25 adults (aged 35.0 � 8.8) meeting DSM-IV criteria for cocaine dependencewithin the past 12 months but currently abstaining (�2 weeks) from cocaine and 24 healthy comparisons (35.1 � 7.5), group-matched onage, sex, education, and employment status.
Results: We observed reduced prefrontal interhemispheric RSFC in cocaine-dependent participants relative to control subjects. Furtheranalyses demonstrated a striking cocaine-dependence-related reduction in interhemispheric RSFC among nodes of the dorsal attentionnetwork, comprising bilateral lateral frontal, medial premotor, and posterior parietal areas. Further, within the cocaine-dependent group,RSFC within the dorsal attention network was associated with self-reported attentional lapses.
Conclusions: Our findings provide further evidence of an association between chronic exposure to cocaine and disruptions withinlarge-scale brain circuitry supporting cognitive control. We did not detect group differences in diffusion tensor imaging measures,suggesting that alterations in the brain’s functional architecture associated with cocaine exposure can be observed in the absence of
detectable abnormalities in the white matter microstructure supporting that architecture.
C ocaine addiction profoundly alters the integrity of prefrontalbrain regions supporting cognitive control processes (1-3).Addicted individuals’ inability to exert volitional control over
their behavior, and the consequent dominance of motivationalprocesses that drive reflexive drug use, are commonly attributed tothese disturbances in prefrontal function (1,2,4,5). Acknowledgingthat frontal dysfunction alone cannot explain the pathophysiologyof addiction, recent work (2,4,6-13) emphasizes the role of dis-rupted functional circuitry. However, addiction-related alterationsin functional interactions between the cerebral hemispheres arerarely examined directly.
Diffusion tensor and structural imaging studies suggest thatchronic exposure to cocaine affects the integrity of white mattertracts connecting distal brain regions. Cocaine dependence is asso-ciated with decreased fractional anisotropy (FA), suggesting alteredwhite matter microstructure, in inferior and orbital frontal regions(14-16) and anterior (17) and posterior (15,18) corpus callosum,although regions of increased FA have also been reported (16).
From the Phyllis Green and Randolph Co�wen Institute for Pediatric Neuro-science at the New York University Child Study Center (CK, X-NZ, KG, CLC,DI, FXC, MM) and Department of Psychiatry, New York University Schoolof Medicine (LL, DB, JR), VA New York Harbor Healthcare System, NewYork, New York; Trinity College Institute for Neuroscience and Depart-ment of Psychology (HG), Trinity College Dublin, Ireland; and NathanKline Institute for Psychiatric Research (HG, FXC, MPM), Orangeburg,New York, New York.
Address correspondence to Clare Kelly, Ph.D., Phyllis Green and RandolphCo�wen Institute for Pediatric Neuroscience, NYU Child Study Center,New York, New York 10016. E-mail: [email protected].
sReceived Jul 22, 2010; revised Nov 18, 2010; accepted Nov 19, 2010.
uch white matter abnormalities correlate with cocaine use dura-ion (15) and measures of cognitive and behavioral control (15-17).ecrements in white matter integrity, particularly in corpus callo-
um, may affect interhemispheric functional interactions funda-ental to integrative attentional processing and cognitive control
19-21). Supporting this, electroencephalogram (EEG) and func-ional magnetic resonance imaging (fMRI) studies have demon-trated altered interhemispheric EEG coherence (22-24) and abnor-
al bilateral task-related activation (25,26) in cocaine users.Resting state fMRI (R-fMRI) approaches, which reveal patterns of
oherent spontaneous fluctuations in the fMRI signal, offer a meanso quantify directly interhemispheric functional interactions. Func-ional homotopy—the high degree of correlated activity betweenomotopic interhemispheric counterparts—is one of the most sa-
ient aspects of the brain’s intrinsic functional architecture (27). Theast majority of large-scale functional networks detected usingoth task-based and R-fMRI are bilateral (28,29), and strong resting-tate functional connectivity (RSFC) is even observable betweenomotopic regions with few monosynaptic callosal connections
30-32), suggesting that functional homotopy reflects an essentialspect of brain function. Consistent with this conclusion, homo-opic RSFC exhibits regional variation congruent with the brain’sunctional hierarchy (33). Further, the developmental trajectories ofomotopic RSFC show regional and hierarchical specificity across
he life span (34), and homotopic RSFC is disrupted in autism (35).A conspicuous feature of the brain’s functional architecture,
omotopic RSFC may provide a sensitive indicator of the effects ofocaine exposure on the integrity of cognitive control circuitry. Inn initial demonstration of the sensitivity of RSFC approaches to the
mpact of cocaine on interhemispheric connectivity (36), cocainedministration decreased RSFC within bilateral primary visual andotor cortex in cocaine addicts. Other studies suggest effects of
ubstance dependence on RSFC within and between functional
C. Kelly et al. BIOL PSYCHIATRY 2011;69:684–692 685
networks, including those associated with cognitive control pro-cesses (8-13). Here, we directly examined interhemispheric RSFC incocaine dependence using a recently validated approach, voxel-mirrored homotopic connectivity (VMHC) (34). VMHC quantifies theRSFC between each voxel in one hemisphere and its mirrored coun-terpart in the opposite hemisphere. We compared VMHC betweencocaine-dependent individuals (COC) and matched control sub-jects, hypothesizing that COC would show reduced interhemi-spheric RSFC. Given evidence for frontal lobe dysfunction associ-ated with cocaine addiction (1,3,37,38), we expected frontal areasto be particularly affected. We also hypothesized reduced FA in COCrelative to control subjects. Finally, we explored whether, withinCOC, RSFC was related to cognitive control function, using theCognitive Failures Questionnaire (CFQ) (39). High CFQ scores indi-cate frequent “cognitive slips and errors,” which may reflect ineffec-tive cognitive control and a predominance of automatic, stimulus-driven responses over “controlled behavior” (40). Such behavioraltendencies are relevant in addiction, in which environmental cuesdrive reflexive behaviors (substance seeking and use), and there is afailure or inability to exert cognitive control over behavior (1,5).
self report and the Edinburgh Inventory) (41) adults (36.6 � 9 years)meeting DSM-IV criteria for cocaine dependence within the past 12months but currently abstaining (� 2 weeks) were scanned. Tenparticipants were excluded from analyses: five tested positive forcocaine at the scan, one exhibited an incidental brain abnormality,and four exhibited excessive movement (displacement/rotation�2.5 mm/degrees). Twenty-five of the remaining 29 participants,for whom healthy comparison participants group-matched on age,sex, education level, and employment status were available, wereincluded in the analyses (see Table 1 for demographics; see FigureS4 in Supplement 1 for supplementary analyses including all 29cocaine-dependent participants).
COC were recruited through the research program at the NYUSchool of Medicine and the VA New York Harbor Healthcare System(NYHHS). Written informed consent was obtained from all partici-pants, who received monetary compensation for their participa-tion. VA NYHHS, New York University (NYU), and NYU School ofMedicine Internal Review Boards approved study procedures.
Cocaine dependence and comorbid Axis I diagnoses were as-sessed with the Structured Clinical Interview for DSM-IV. Partici-pants were included if they met criteria for cocaine dependencewithin the previous 12 months but reported abstaining from co-caine use for the past 2 weeks. Ten-drug (including cocaine, am-phetamines, marijuana, and opiates) urine screens at assessmentand scan confirmed abstinence. Most participants reported usingcocaine intranasally and consuming 1 to 2 g per use (Table 1).Cocaine withdrawal severity was assessed using the Cocaine Selec-tive Severity Assessment (CSSA) (42). CSSA scores greater than 20indicate more severe withdrawal symptoms (43,44). The meanCSSA score was 12.8 � 8.4 (range 2-28). Only four participants hadCSSA scores greater than 20, suggesting that overall, our partici-pants were not in acute cocaine withdrawal.
Current or lifetime history of neurological disorders, autism,schizophrenia, suicidality, psychosis, mania, current psychotropicuse, and current substance abuse or dependence (other than co-caine and nicotine) were exclusionary. Four participants reported alifetime history of major depressive disorder (three in full remission,
one reported mild symptoms during the past month). One partici- v
ant met criteria for posttraumatic stress disorder (mild symptomsn the past month). One participant reported a history of panicisorder and one of bulimia nervosa; neither reported currentymptoms. Eleven participants had a lifetime history of alcoholependence, and one a history of alcohol abuse; none met criteria
or dependence within the past month. Six participants had lifetimeistories of cannabis dependence, and four of cannabis abuse;one met criteria for dependence within the past month. Smokingtatus was available for 20 of the 25 participants; 13 were smokersSupplement 1).
Control Group. We selected 24 right-handed adults from aool of adults participating in ongoing studies in our laboratory toatch the COC group on age, sex, education, employment and
andedness (per the Edinburgh Inventory). Only participants withegligible movement (displacement/rotation �2.5 mm/degrees)ere included. The Structured Clinical Interview for DSM-IV was
dministered to all control participants; inclusion required absencef current or past DSM-IV Axis I psychiatric diagnoses, includingubstance abuse or dependence. Smoking status was available for6 of the 24 control participants; 3 were smokers.
ata Acquisition
Imaging data were acquired using a Siemens Allegra 3T (NYU Cen-er for Brain Imaging). A T1-weighted image (magnetization preparedapid acquisition gradient-echo, repetition time [TR] � 2530 msec;cho time [TE] � 3.25 msec; inversion time [TI] � 1100 msec; flip angle
7°; 128 slices; field of view�256 mm; voxel-size�1�1.3�1.3 mm)nd one 6-min resting state scan (multiecho echo planar imaging (EPI)equence; 180 time points; TR � 2000 msec; flip angle � 90°; 33 slices;
able 1. Participant Demographics
Cocaine Controls
umber of Participants 25 24ean (� SD) Age (years) 35.0 (8.8) 35.1 (7.5)umber (%) Male 23 (92%) 20 (83%)evel of Education
Some high school 0 1High school graduate or GED 7 3Some college 10 10College graduate 7 8Advanced graduate/professional degree 1 2
mploymentFull time 8 8Part time 7 4Unemployed 8 6Student 1 6Disability/retired 1 0ean (� SD) Years Since Initiation of
Cocaine Usea11.43 (8.5) —
ean Age of Initiation of Cocaine Usea 22 —requency of Cocaine Use
Daily 8 —5 times per week 2 —3–4 times per week 13 —Twice per week 2 —
SSA, Mean (� SD) 12.48 (8.4) —FQ, Mean (� SD) 28.68 (13.0) —
Age, sex distributions, educational attainment, and employment statusid not differ significantly (p � .20).
oxel size � 3 � 3 � 4 mm) were acquired. For most participants, a
www.sobp.org/journal
dfit
R
PmwtdflKtia
V
rrselpfa
(ht
Gb(mccVoscon
Vy
cF
ssmpvtfIedv
ctpim
acaic(ifn
s
slHaawtadVabia
D
bcspfiieartpe
686 BIOL PSYCHIATRY 2011;69:684–692 C. Kelly et al.
w
second resting state scan was acquired later in the session (Supple-ment 1). A field map and short TE EPI scan were also acquired toimprove functional-to-anatomic coregistration.
All COC and 11 control subjects were scanned with eyes open.The remaining 13 control subjects were scanned with eyes openand closed, counterbalanced. For six participants, the order was“eyes open” first, and for the remaining seven, the order wasreversed.
Diffusion Tensor Imaging DataTwo diffusion tensor imaging (DTI) scans were acquired per
participant using a twice-refocused diffusion-weighted EPI se-quence (45) (TR � 5200 msec; TE � 78 msec; 50 slices; matrix 64 �64; field of view � 192 mm; acquisition voxel size � 3 � 3 � 3 mm;64 diffusion directions with b value 1000 s/mm2; one image with no
iffusion weighting; bandwidth � 3720 Hz/pixel). A gradient echoeld map (TR � 834 msec; TE � 5.23/7.69 msec) acquisition usedhe same slice positioning and resolution as the DTI scans.
esting State Functional ConnectivityData processing was performed using AFNI (46) and FSL (47).
reprocessing comprised slice time correction; three-dimensionalotion correction; temporal despiking; spatial smoothing (fullidth at half maximum � 6 mm); mean-based intensity normaliza-
ion; temporal band-pass filtering (.009 –.1 Hz); linear and quadraticetrending; nuisance signal removal (white matter, cerebrospinaluid, global signal, motion parameters) via multiple regression (seeelly et al. (48) for details); linear registration of functional to struc-ural images (with intermediate registration to a low-resolutionmage and b0 unwarping); nonlinear registration of structural im-ges to the MNI152 template (49,50).
oxel-Mirrored Homotopic ConnectivityTo account for geometric differences between hemispheres, we
efined the registration from individual anatomic to Montreal Neu-ological Institute (MNI) 152 template space using a group-specificymmetrical template. All 49 registered structural images were av-raged to create a mean image, which was then averaged with its
eft–right mirror to generate a group-specific symmetrical tem-late. Nonlinear registration to this symmetrical template was per-
ormed for each participant, and the resultant transformation waspplied to each participant’s preprocessed functional data.
Homotopic RSFC was computed as the Pearson correlationFisher Z-transformed) between every pair of symmetrical inter-emispheric voxels’ time series. The resultant correlations consti-
ute VMHC.We examined global and regional group differences in VMHC.
lobal VMHC was calculated by averaging VMHC values across allrain voxels within a unilateral hemispheric gray matter mask
there is only one correlation for each pair of homotopic voxels). Theask was created using the MNI152 gray matter tissue prior in-
luded with FSL (threshold � 25% tissue-type probability). We ex-luded voxels medial of x � �4, to minimize artifactually increasedMHC because of blurring across the midline. Group comparisonsf global VMHC were performed using t tests. For this and all sub-equent analyses, only “eyes-open” scans were included. To ac-ount for control subjects whose “eyes-open” scan was their sec-nd scan of the imaging session, scan order was modeled as auisance covariate, along with age and sex.
To test for regional group differences in VMHC, individual-levelHMC maps were entered into a group-level voxelwise t test anal-
sis using a mixed-effects ordinary least squares model. Multiple e
ww.sobp.org/journal
omparisons corrections were performed using Gaussian Randomield theory (min Z �2.3, cluster significance: p � .05, corrected).
To address potential confounds associated with interhemi-pheric structural asymmetry, we repeated our analyses with 1) nomoothing and 2) exaggerated smoothing (10 mm full width at half
aximum). Further, we calculated voxel-mirrored homotopic mor-hometry, a measure of left–right differences in gray matter (GM)olume for each participant, using FSL’s voxel-based morphome-ry-style analysis pipeline (51,52) (see Supplement 1 for details andor group comparisons of whole-brain voxel-based morphometry).n supplementary group comparisons of VMHC, left–right differ-nces in GM volume were controlled for by including the three-imensional voxel mirrored homotopic morphometry volumes as aoxel-dependent covariate.
Finally, because the COC and control groups differed signifi-antly in smoking and psychiatric history, we performed two addi-ional group-level analyses that controlled for whether a partici-ant 1) smoked and 2) had a history of any psychiatric disorder,
Seed-Based RSFC. We examined the RSFC associated withreas exhibiting significantly different VMHC between groups. Spe-ifically, we computed whole-brain voxelwise correlations associ-ted with mean time series derived separately for two regions of
nterest (ROIs), comprising all voxels within the inferior frontal sul-us (IFS) area exhibiting greater VMHC for controls, relative to COCFigure 1C). Fisher Z–transformed correlation maps were enterednto a group-level voxelwise t test analysis. Whole-brain correctionor multiple comparisons was performed (min Z � 2.3; cluster sig-ificance: p � .05, corrected).
Within-Sample Replication. See Supplement 1 for within-ses-ion stability analyses.
Brain–Behavior Relationships. The CFQ (39), a 25-item mea-ure of self-reported attentional lapses with good construct va-idity and test–retest reliability (53,54), was provided by 23 COC.igher scores indicate more frequent lapses of attention and aressociated with poorer performance on task-based measures ofttentional function (40,55). Scores on the CFQ correlate wellith deficits in cognitive control (39,40) as well as anxiety symp-
oms (39). First, we explored the relationship between the CFQnd mean prefrontal interhemispheric RSFC within a 4-mm-ra-ius sphere centered on the peak of the group difference inMHC, after regressing out age and sex from both the RSFC datand CFQ scores. Second, we examined the voxelwise associationetween the CFQ and RSFC by entering CFQ scores as a covariate
n the group-level seed-based RSFC analysis, controlling for agend sex.
TIDTI data analysis was performed using the FSL program tract-
ased spatial statistics (56). Each participant’s pair of DTI scans wereoncatenated and corrected for eddy currents and motion. Diffu-ion gradients were rotated to improve consistency with motionarameters. Magnetic field inhomogeneities were accounted for byeld map reconstruction, and data were averaged across scans to
mprove the signal-to-noise ratio. Diffusion tensors were fitted forach voxel to obtain images containing FA, first diffusion eigenvari-te (L1), mean diffusivity and radial diffusivity values, and wereegistered to the FMRIB58-FA standard space image (1 mm3 resolu-ion) using FNIRT. A group-mean FA skeleton was created, and eacharticipant’s standard space FA data were projected onto the skel-ton. Resultant skeletonized FA images were entered into vox-
lwise nonparametric group comparisons, performed using ran-
uta
ppAV
(t
waC
fifd(ppeVtGVaseRF
C. Kelly et al. BIOL PSYCHIATRY 2011;69:684–692 687
domize. The mean FA skeleton (thresholded at FA � .2) was usedas a mask, and 5000 permutations were performed, covarying forage and sex. We corrected for multiple comparisons (p � .05)
sing threshold-free cluster enhancement, controlling for spa-ial nonstationarity (57). See Supplement 1 for additional DTInalyses.
ResultsVoxel-Mirrored Homotopic Connectivity. Consistent with
revious studies (27,33), homotopic RSFC was a robust global brainhenomenon, with regional differences in strength (Figure 1A).lthough the control and cocaine groups did not differ on globalMHC (control subjects � .33 � .07; cocaine � .34 � .07; p � .50),
group comparisons, controlling for age, sex, and scan order, re-vealed one region in which controls exhibited stronger VMHC thanCOC. This region extended from deep within the IFS into the middlefrontal gyrus and ventral premotor cortex, anteriorly from the infe-rior frontal junction, along the superior part of pars opercularis andtriangularis, and into the frontal operculum (Figure 1C). This groupdifference in VMHC remained significant 1) in the secondary scan(Figure 1D); 2) when different levels of smoothing were applied(Figures S3A and S3B in Supplement 1); and when 3) smoking(Figure S4B in Supplement 1), 4) psychiatric history (Figure S4C inSupplement 1) and 5) left-right differences in grey matter volume
Figure 1. (A) Group-level voxel-mirrored homotopic connectivity (VMHC)corrected; images are axial slices at z � 5, 28, 51). Although there is only onhemispheres to minimize confusion regarding the laterality of the results. (B(IFS) area exhibiting significant group differences in the primary VMHC analr � .587, p � .01; all participants: r � .69; p � .0001). See Supplement 1 for fullgroup exhibited significantly stronger VMHC than the cocaine-dependent gnonindependence in voxelwise analyses of group differences necessarily proacross the IFS area exhibiting significant group differences in the primary VM2) data. The group difference in VMHC for the secondary scan is significant [c� .09; t(38) � 4.13, p � .001].
(Figures S3C and S4D in Supplement 1) were taken into account (
Table S1 in Supplement 1). No regions exhibited stronger VMHC inhe cocaine, relative to the control group.
Seed-Based RSFC. We examined whole-brain RSFC associatedith two ROIs (one per hemisphere; Figure 1C), comprising the IFS
rea that exhibited greater VMHC for control subjects, relative toOC.
Both left and right IFS exhibited RSFC with a large bilateral dorsalrontoparietal network comprising lateral prefrontal cortex (primar-ly middle frontal gyrus, but also portions of inferior and superiorrontal gyri), dorsal premotor cortex, including the frontal eye fields,orsal paracingulate cortex and presupplementary motor areas
preSMA), posterior parietal cortex and intraparietal sulcus (IPS),osterior middle temporal cortex, and caudate (Figure S1 in Sup-lement 1). This network is commonly identified as the frontopari-tal or dorsal attention network (DAN) (58-61). Consistent with theMHC analysis, control subjects exhibited stronger RSFC between
he right IFS seed and left lateral prefrontal and premotor cortex.roup differences in left IFS RSFC were similarly consistent with theMHC analysis, but also revealed reduced RSFC between left IFSnd right posterior parietal cortex and IPS in COC, relative to controlubjects (Figure S1, Table S2 in Supplement 1). These group differ-nces remained significant 1) in the secondary scan (except left IFSSFC with right IFS/MFG, which just failed to reach significance; seeigure S2 and text in Supplement 1) and when 2) GM volume
e control and cocaine-dependent groups (Z � 2.3, cluster level p � .05,relation for each pair of homotopic voxels, results are projected onto bothss-scan consistency of mean VMHC values across the inferior frontal sulcus.e., the area shown in Figure 1C; control subjects: r � .495, p � .05; cocaine:ls of the within-sample replication analysis. (C) IFS area for which the control(Z � 2.3, cluster-level p � .05, corrected). (D) In recognition of the fact thatinflated estimates of effect sizes (89 –91), the plots show mean VMHC valuesalysis (shown in panel B) computed on the basis of the secondary scan (scanl subjects mean Scan 2 VMHC � .38 � .10, cocaine mean Scan 2 VMHC � .26
for the cor) Cro
ysis (idetairoupvidesHC anontro
Figure S4D in Supplement 1), 3) smoking (Figure S4B in Supple-
www.sobp.org/journal
R[
�.r.rb
B
(tfw
hh
dhs
area;
688 BIOL PSYCHIATRY 2011;69:684–692 C. Kelly et al.
w
ment 1), and 4) psychiatric history were taken into account (FigureS4C, Table S2 in Supplement 1). No regions exhibited stronger RSFCin COC, relative to the control group.
To assess further the nature of cocaine-related decrements inRSFC within the DAN, we 1) identified peak regions for right and leftIFS RSFC maps, revealing eight unique unilateral nodes; 2) createdspherical ROIs (4-mm radius), centered on each node and its homo-topic interhemispheric counterpart (see Table S3 in Supplement 1for coordinates); 3) extracted the mean time series for each ROI; and4) computed all pairwise correlations between nodes for each sub-ject. We then sorted correlations according to whether they werebetween intrahemispheric (e.g., right IFS and right IPS), heterotopicinterhemispheric (e.g., right IFS and left IPS), or homotopic inter-hemispheric (e.g., right and left IFS) node pairs and tested for groupdifferences in mean RSFC for each of these correlation types (62).
elative to control subjects, COC exhibited reduced homotopic
Figure 2. Cocaine-dependent participants exhibit reduced interhemisphericorsal attention network (DAN). We tested for group differences in intrahemeterotopic interhemispheric (e.g., right IFS and left IPS), and homotopic intechematic in top-left panel). Relative to control subjects, the cocaine-depe
interhemispheric RSFC but not reduced intrahemispheric RSFC. antIFS, anterposterior IFS; MT, middle temporal area; preSMA, presupplementary motor
control subjects mean � .45 � .11; COC � .39 � .09; t (47) � 2.66, p W
ww.sobp.org/journal
.05] and heterotopic [control subjects mean � .16 � .07; COC �11 � .07; t (47) � 2.68, p � .05] interhemispheric RSFC, but noteduced intrahemispheric RSFC [control subjects left–left mean �25 � .07; COC � .23 � .08; t (47) � .76, p � .45; control subjectsight–right mean � .25 � .1; COC � .21 � .08; t (47) � 1.44, p � .16]etween nodes of the DAN (Figure 2).
rain–Behavior RelationshipsThe DAN subserves top-down control of attention and behavior
58,59). As such, we tested whether DAN RSFC was associated withhe CFQ. CFQ scores were available for 23 COC participants but notor control subjects; one score, greater than 3 SD above the mean,
as excluded.First, we examined the relationship between the CFQ and inter-
emispheric RSFC (VMHC) associated with the IFS region that ex-ibited significantly weaker RSFC in COC, relative to controls.
not intrahemispheric resting-state functional connectivity (RSFC) within theric (e.g., right inferior frontal sulcus [IFS] and right intraparietal sulcus [IPS]),ispheric (e.g., right and left IFS) RSFC between all pairs of 16 DAN nodes (seet group exhibited reduced homotopic [p � .05] and heterotopic [p � .05]erior frontal sulcus; FEF, frontal eye fields; L, left; midIFC, middle IFS; postIFS,R, right.
, butispherhemndenior inf
ithin COC, we observed a significant negative correlation be-
pqpC.
D
Faaiogt
D
argAp6is
ijrRtRpttithc
dtd
e weo neu
V
S
C. Kelly et al. BIOL PSYCHIATRY 2011;69:684–692 689
tween CFQ score and mean VMHC within a spherical 4-mm-radiusROI centered on the VMHC group difference peak (reported inFigure 3A, Table 2; r � –.43, n � 22, p � .05), suggesting that weaker
refrontal interhemispheric RSFC was associated with more fre-uent attentional failures. However, this finding should be inter-reted with caution, because it did not remain significant whenSSA withdrawal symptoms were taken into account (r � –.36, p �
1), and it was absent in the second scan (r � .27, n � 17, p � .27).The voxelwise analysis revealed a more robust brain– behavior
relationship. Higher CFQ scores were associated with weaker posi-tive RSFC between the right IFS seed and bilateral preSMA (Figure3B). This relationship just escaped significance in the secondaryscan data (n � 17; r � –.46, p � .056); but was significant afterpartialing out CSSA scores (Scan 1: r � –.66, p � .001; Scan 2: r �–.50, p � .05). No relationship was observed for the left IFS seed.
Figure 3. Region of interest and voxelwise brain– behavior relationships. (A)homotopic connectivity [VMHC]) within a 4-mm-radius sphere centered on tfailures, as measured by the Cognitive Failures Questionnaire (CFQ; r � –.43interhemispheric RSFC reported experiencing more frequent attentional faisulcus [IFS] RSFC and self-reported cognitive failures. Shown in green is thnegative relationship with the CFQ. Cocaine-dependent participants with thattentional failures. Axial slices (z � 5, 28, and 51) are displayed according t
Table 2. Significant Group Differences in VMHC and Seed-Based RSFC
TIVoxelwise nonparametric group comparisons of skeletonized
A data revealed no significant group differences, even when agend sex covariates were removed and when a parametric voxelwisenalysis of all (nonskeletonized) white matter FA, L1, mean diffusiv-
ty, and radial diffusivity values was performed. ROI-based analysesf FA within specific white matter tracts similarly did not reveal anyroup differences or any correlations between FA and VMHC. Addi-
ional exploratory analyses are presented in Supplement 1.
iscussion
Homotopic functional connectivity is one of the most salient char-cteristics of the brain’s intrinsic functional architecture (27,33), likelyeflecting the importance of interhemispheric communication to inte-rated brain function underlying coherent cognition and behavior.ltered interhemispheric functional interactions have been found insychiatric and clinical disorders (35,63-66) and in normal aging (67-9). With a few exceptions (22-26), electrophysiologic and neuroimag-
ng studies have less frequently paid direct attention to interhemi-pheric interactions in substance dependence.
Here, we demonstrated reduced prefrontal interhemispheric RSFCn abstinent cocaine-dependent participants, relative to control sub-ects. Follow-up analyses demonstrated striking cocaine dependence-elated reductions in homotopic and heterotopic interhemisphericSFC among the nodes of a dorsal frontoparietal network implicated inop-down control of attention and behavior by both task-based and-fMRI studies (28,60,61), although the extensive pattern of RSFC overrefrontal areas suggests that the network also encompasses regions
ypically considered part of the ventral attention network (58,59). In-erestingly, the IFS area that exhibited robust group differences innterhemispheric RSFC has been identified as a site of interaction be-ween the dorsal and ventral attention networks (59,66,70) and thusas been postulated to subserve a general role in both attentionalontrol and awareness (70).
The brain’s intrinsic functional architecture constitutes the foun-ation on which momentary neuronal responses underlying cogni-
ion and behavior are built (71,72). Thus, although we detectedecrements in interhemispheric RSFC among nodes of the dorsal
hemispheric resting-state functional connectivity (RSFC; i.e., voxel-mirroredak of the group difference in VMHC correlated with self-reported cognitive23, p � .05). Cocaine-dependent participants with the weakest prefrontal
(B) Voxelwise analyses revealed relationships between right inferior frontaldial/superior lateral premotor area whose RSFC with right IFS exhibited aakest RSFC between these two areas reported experiencing more frequentrological convention (right is right).
Interhe pe, n �
lures.e me
ttention network while participants were at rest, those decre-
www.sobp.org/journal
bmcprwti
fw
shsarftfdfcswgstmes
awc
tCtRtFP
mddaNdSSspnlnicDjt
c
690 BIOL PSYCHIATRY 2011;69:684–692 C. Kelly et al.
w
ments likely contribute to impaired functioning when that networkis called on to support goal-directed thought and action. Consistentwith this proposal, two studies in patients with unilateral strokedemonstrated that reduced interhemispheric RSFC between ho-motopic and heterotopic nodes of the DAN was associated withimpaired behavioral performance on a target detection task (66,73).Further, there was a correlated restoration of both interhemisphericRSFC and behavior with time (66), an observation confirmed in therat sensorimotor system (74). Interestingly, studies with healthycontrol subjects have shown that disrupting the balance of prefron-tal interhemispheric interaction with transcranial stimulation alterssubsequent risk-taking behavior during gambling tasks (75,76).These findings suggest the importance of homotopic and hetero-topic interaction to cognition and behavior. Although we did notdirectly assess attentional control, RSFC within the dorsal attentionnetwork (between right IFS and bilateral preSMA) was related toself-reported attentional lapses, and a less robust relationship wasobserved between self-reported attentional lapses and IFS homo-topic interhemispheric RSFC. Future work is required not only toreplicate our findings and their generalizability to other substancesof abuse but also to identify the behavioral correlates of impairedinterhemispheric RSFC in substance dependence.
Our cocaine-dependent participants were abstaining from co-caine use (� 2 weeks) and were not in acute withdrawal (indicated
y the low group mean CSSA score). Accordingly, the RSFC decre-ents we observed cannot be attributed to recent cocaine intoxi-
ation nor are they likely to reflect group differences in smoking orsychiatric history (Figure S4 in Supplement 1). Instead, they likely
eflect enduring effects of chronic cocaine exposure. Nonetheless,e cannot rule out the possibility that such decrements preceded
he initiation of cocaine addiction and may constitute a vulnerabil-ty to the development of substance-related disorders (77). Studies
of animals experimentally exposed to cocaine are beginning totease apart these possibilities (78,79). In humans, this question maybest be examined in longitudinal and familial/twin studies (80).Alternatively, reversal of decrements with maintenance of absti-nence may help differentiate direct effects of cocaine use frompreexisting conditions. We plan to examine these issues in fol-low-up studies.
Despite robust effects of cocaine dependence on RSFC and pre-vious studies demonstrating significant cocaine-related alterationsin FA (14-18), we observed no significant group differences in DTImeasures. We did observe a significant correlation between FA inthe posterior limb of the internal capsule and reported duration ofcocaine dependence (Supplement 1). Although a complete under-standing of this observation awaits further studies, Xu et al. (81)observed that FA in this area correlated with duration of abstinencein cocaine-dependent individuals.
Optimistically, preservation of structural connectivity encour-ages hope for recovery of function with abstinence from use or dueto cognitive interventions. Studies of cognitive training interven-tions suggest that functional deficits are sometimes reversible (82);future studies should investigate whether cognitive training cannormalize attentional function and RSFC within attentional net-works. On the other hand, our failure to detect significant groupdifferences in DTI measures may reflect either weakness in ourprotocol (e.g., low resolution) or in DTI FA measures more broadly—for example, crossing fibers, integral to interhemispheric connec-tivity, are known to affect DTI measures such as FA (83,84). A num-ber of studies have demonstrated associations between chroniccocaine exposure and reduced cortical thickness (77) and GM vol-ume (85-87). Although we did not detect group differences in GM
volume or GM volume asymmetries, an important future direction
ww.sobp.org/journal
or this dataset is to examine cortical thickness and its relationshipith R-fMRI measures.
Several other limitations should be noted. First, the brain is notymmetrical. To improve the functional correspondence betweenomotopic regions, we used a symmetrical standard template andmoothed the functional data. We also performed supplementarynalyses to ensure that potential group differences in morphomet-ic asymmetry could not account for our findings. Nonetheless,uture studies should consider data-driven approaches (e.g., clus-ering interhemispheric voxels on the basis of their RSFC) to defineunctional homotopy. Second, as is typical of studies in cocaine-ependent populations, the sample was predominantly male. R-
MRI studies have only begun to understand sex differences, in theontext of exceptionally large samples (88). The potential role ofex-differences in our findings should be examined in future studiesith considerably larger samples. Third, the cocaine and controlroups differed in smoking status. Although supplementary analy-es suggested that the RSFC decrements we observed are unlikelyo reflect group differences in smoking, future studies should
atch groups on smoking status and history. Finally, studies shouldndeavor to control for the potential effects of participants’ currenttate (e.g., anxiety, arousal) on R-fMRI measures.
In conclusion, this work suggests that measures of homotopicnd heterotopic interhemispheric RSFC may provide sensitive toolsith which to study the large-scale circuitry supporting cognitive
ontrol and its disruption in addictive disorders.
Financial support for this project was provided by grants fromhe National Institute on Drug Abuse (Grant Nos. R03DA024775 toK, R01DA016979 to FXC, and 2T32DA007254-16A2 for CLC), the Na-
ional Institute of Mental Health (Grant Nos. R01MH083246 and01MH081218 to FXC and MPM), and Autism Speaks, as well as gifts to
he New York University Child Study Center from the Stavros Niarchosoundation, Leon Levy Foundation, and an endowment provided byhyllis Green and Randolph Co�wen.
We thank Dr. Souheil Inati and Dr. Pablo Valesco for their work onultiecho echo planar imaging and diffusion tensor imaging sequence
evelopment, Saroja Bangaru and Devika Jutagir for their assistance inata analysis and study management, Dr. Maarten Mennes and Dr. Adri-na Di Martino for their helpful editorial and esthetic advice, the Stavrosiarchos Foundation, Leon Levy Foundation and Phyllis Green and Ran-olph Co�wen for their generous support of the New York University Childtudy Center, and all our participants for their time and cooperation.cripts containing the processing commands employed here to computeeed-based resting state functional connectivity have been released asart of the 1000 Functional Connectomes Project (88) (http://www.itrc.org/projects/fcon_1000). The data used in this study have been re-
eased for unrestricted non-commercial use (http://fcon_1000.projects.itrc.org/indi/retro/nyuCocaine.html) via the International Neuroimag-
ng Data-Sharing Initiative (http://fcon_1000.projects.nitrc.org). X-NZ isurrently affiliated with the Laboratory for Functional Connectome andevelopment, Institute of Psychology, Chinese Academy of Sciences, Bei-
ing, and the Division of Cognitive and Developmental Psychology, Insti-ute of Psychology, Chinese Academy of Sciences, Beijing, China.
All authors report no biomedical financial interests or potentialonflicts of interest.
Supplementary material cited in this article is available online.
1. Garavan H, Hester R (2007): The role of cognitive control in cocainedependence. Neuropsychol Rev 17:337–345.
2. Goldstein RZ, Volkow ND (2002): Drug addiction and its underlying
neurobiological basis: Neuroimaging evidence for the involvement ofthe frontal cortex. Am J Psychiatry 159:1642–1652.
7. Everitt BJ, Belin D, Economidou D, Pelloux Y, Dalley JW, Robbins TW(2008): Review. Neural mechanisms underlying the vulnerability to de-velop compulsive drug-seeking habits and addiction. Philos Trans R SocLond B Biol Sci 363:3125–3135.
8. Tomasi D, Volkow ND, Wang R, Carrillo JH, Maloney T, Alia-Klein N, et al.(2010): Disrupted functional connectivity with dopaminergic midbrainin cocaine abusers. PLoS ONE 5:e10815.
9. Gu H, Salmeronl BJ, Ross TJ, Geng X, Zhanl W, Stein EA, et al. (2010):Mesocorticolimbic circuits are impaired in chronic cocaine users asdemonstrated by resting state functional connectivity. Neuroimage 53:593– 601.
10. Yuan K, Qin W, Dong M, Liu J, Liu P, Zhang Y, et al. (2010): Combiningspatial and temporal information to explore resting-state networkschanges in abstinent heroin-dependent individuals. Neurosci Lett 475:20 –24.
11. Ma N, Liu Y, Li N, Wang CX, Zhang H, Jiang XF, et al. (2010): Addictionrelated alteration in resting-state brain connectivity. Neuroimage 49:738 –744.
2. Cole DM, Beckmann CF, Long CJ, Matthews PM, Durcan MJ, Beaver JD(2010): Nicotine replacement in abstinent smokers improves cognitivewithdrawal symptoms with modulation of resting brain network dy-namics. Neuroimage 52:590 –599.
13. Hong LE, Gu H, Yang Y, Ross TJ, Salmeron BJ, Buchholz B, et al. (2009):Association of nicotine addiction and nicotine’s actions with separatecingulate cortex functional circuits. Arch Gen Psychiatry 66:431– 441.
4. Lim KO, Choi SJ, Pomara N, Wolkin A, Rotrosen JP (2002): Reducedfrontal white matter integrity in cocaine dependence: A controlled dif-fusion tensor imaging study. Biol Psychiatry 51:890 – 895.
15. Lim KO, Wozniak JR, Mueller BA, Franc DT, Specker SM, Rodriguez CP, etal. (2008): Brain macrostructural and microstructural abnormalities incocaine dependence. Drug Alcohol Depend 92:164 –172.
6. Romero MJ, Asensio S, Palau C, Sanchez A, Romero FJ (2010): Cocaineaddiction: Diffusion tensor imaging study of the inferior frontal andanterior cingulate white matter. Psychiatry Res 181:57– 63.
17. Moeller FG, Hasan KM, Steinberg JL, Kramer LA, Dougherty DM, SantosRM, et al. (2005): Reduced anterior corpus callosum white matter integ-rity is related to increased impulsivity and reduced discriminability incocaine-dependent subjects: Diffusion tensor imaging. Neuropsychop-harmacology 30:610 – 617.
18. Ma L, Hasan KM, Steinberg JL, Narayana PA, Lane SD, Zuniga EA, et al.(2009): Diffusion tensor imaging in cocaine dependence: Regional ef-fects of cocaine on corpus callosum and effect of cocaine administrationroute. Drug Alcohol Depend 104:262–267.
19. Banich MT (1998): The missing link: The role of interhemispheric inter-action in attentional processing. Brain Cogn 36:128 –157.
20. Gazzaniga MS (2000): Cerebral specialization and interhemisphericcommunication: Does the corpus callosum enable the human condi-tion? Brain 123:1293–1326.
21. Hoptman MJ, Davidson RJ (1994): How and why do the two cerebralhemispheres interact? Psychol Bull 116:195–219.
22. Reid MS, Flammino F, Howard B, Nilsen D, Prichep LS (2006): Topo-graphic imaging of quantitative EEG in response to smoked cocaineself-administration in humans. Neuropsychopharmacology 31:872– 884.
23. Reid MS, Flammino F, Howard B, Nilsen D, Prichep LS (2008): Cocaine cueversus cocaine dosing in humans: Evidence for distinct neurophysiolog-ical response profiles. Pharmacol Biochem Behav 91:155–164.
24. Roemer RA, Cornwell A, Dewart D, Jackson P, Ercegovac DV (1995):Quantitative electroencephalographic analyses in cocaine-preferringpolysubstance abusers during abstinence. Psychiatry Res 58:247–257.
25. Hanlon CA, Wesley MJ, Roth AJ, Miller MD, Porrino LJ (2010): Loss of
laterality in chronic cocaine users: An fMRI investigation of sensorimo-tor control. Psychiatry Res 181:15–23.
6. Hanlon CA, Wesley MJ, Porrino LJ (2009): Loss of functional specificity inthe dorsal striatum of chronic cocaine users. Drug Alcohol Depend 102:88 –94.
7. Salvador R, Suckling J, Schwarzbauer C, Bullmore E (2005): Undirectedgraphs of frequency-dependent functional connectivity in whole brainnetworks. Philos Trans R Soc Lond B Biol Sci 360:937–946.
8. Smith SM, Fox PT, Miller KL, Glahn DC, Fox PM, Mackay CE, et al. (2009):Correspondence of the brain’s functional architecture during activationand rest. Proc Natl Acad Sci U S A 106:13040 –13045.
9. Toro R, Fox PT, Paus T (2008): Functional coactivation map of the humanbrain. Cereb Cortex 18:2553–2559.
0. Vincent JL, Patel GH, Fox MD, Snyder AZ, Baker JT, Van Essen DC, et al.(2007): Intrinsic functional architecture in the anaesthetized monkeybrain. Nature 447:83– 86.
1. Roy AK, Shehzad Z, Margulies DS, Kelly AM, Uddin LQ, Gotimer K, et al.(2009): Functional connectivity of the human amygdala using restingstate fMRI. Neuroimage 45:614 – 626.
2. Di Martino A, Scheres A, Margulies DS, Kelly AM, Uddin LQ, Shehzad Z, etal. (2008): Functional connectivity of human striatum: A resting stateFMRI study. Cereb Cortex 18:2735–2747.
3. Stark DE, Margulies DS, Shehzad ZE, Reiss P, Kelly AM, Uddin LQ, et al.(2008): Regional variation in interhemispheric coordination of intrinsichemodynamic fluctuations. J Neurosci 28:13754 –13764.
4. Zuo XN, Kelly C, Di Martino A, Mennes M, Margulies DS, Bangaru S, et al.(2010): Growing together and growing apart: Regional and sex differ-ences in the lifespan developmental trajectories of functional homo-topy. J Neurosci 30:15034 –15043.
5. Anderson JS, Druzgal TJ, Froehlich A, Dubray MB, Lange N, AlexanderAL, et al. (2010): Decreased interhemispheric functional connectivity inautism. Cereb Cortex.
6. Li SJ, Biswal B, Li Z, Risinger R, Rainey C, Cho JK, et al. (2000): Cocaineadministration decreases functional connectivity in human primary vi-sual and motor cortex as detected by functional MRI. Magn Reson Med43:45–51.
7. Volkow ND, Li TK (2004): Drug addiction: The neurobiology of behaviourgone awry. Nat Rev Neurosci 5:963–970.
9. Broadbent DE, Cooper PF, FitzGerald P, Parkes KR (1982): The CognitiveFailures Questionnaire (CFQ) and its correlates. Br J Clin Psychol 21:1–16.
0. Manly T, Robertson IH, Galloway M, Hawkins K (1999): The absent mind: Fur-ther investigations of sustained attention to response. Neuropsychologia 37:661–670.
1. Oldfield RC (1971): The assessment and analysis of handedness: TheEdinburgh Inventory. Neuropsychologia 9:97–113.
2. Kampman KM, Volpicelli JR, McGinnis DE, Alterman AI, Weinrieb RM,D’Angelo L, et al. (1998): Reliability and validity of the Cocaine SelectiveSeverity Assessment. Addict Behav 23:449 – 461.
3. Mulvaney FD, Alterman AI, Boardman CR, Kampman K (1999): Cocaineabstinence symptomatology and treatment attrition. J Subst Abus Treat16:129 –135.
4. Kampman KM, Pettinati HM, Volpicelli JR, Oslin DM, Lipkin C, Spark-man T, O’Brien CP, (2004): Cocaine dependence severity predictsoutcome in outpatient detoxification from cocaine and alcohol. Am JAddict 13:74 – 82.
5. Reese TG, Heid O, Weisskoff RM, Wedeen VJ (2003): Reduction of eddy-current-induced distortion in diffusion MRI using a twice-refocused spinecho. Magn Reson Med 49:177–182.
6. Cox RW (1996): AFNI: Software for analysis and visualization of func-tional magnetic resonance neuroimages. Comput Biomed Res 29:162–173.
7. Smith SM, Jenkinson M, Woolrich MW, Beckmann CF, Behrens TE, Johan-sen-Berg H, et al. (2004): Advances in functional and structural MR imageanalysis and implementation as FSL. Neuroimage 23(suppl 1):S208 –S219.
8. Kelly C, de Zubicaray G, Di Martino A, Copland DA, Reiss PT, Klein DF,et al. (2009): L-dopa modulates functional connectivity in striatal cogni-tive and motor networks: A double-blind placebo-controlled study.J Neurosci 29:7364 –7378.
9. Andersson JLR, Jenkinson M, Smith SM (2007): Non-linear optimisation
(FMRIB Technical Report TR07JA1). Oxford, United Kingdom: FMRIBCentre.
www.sobp.org/journal
7
7
7
7
7
7
7
8
8
8
8
8
8
8
8
8
8
9
9
692 BIOL PSYCHIATRY 2011;69:684–692 C. Kelly et al.
w
50. Andersson JLR, Jenkinson M, Smith SM (2007): Non-linear registration,a.k.a. Spatial normalisation (FMRIB Technical Report TR07JA2). Oxford,United Kingdom: FMRIB Centre.
51. Good CD, Johnsrude IS, Ashburner J, Henson RN, Friston KJ, FrackowiakRS (2001): A voxel-based morphometric study of ageing in 465 normaladult human brains. Neuroimage 14:21–36.
52. Zhang Y, Brady M, Smith S (2001): Segmentation of brain MR imagesthrough a hidden Markov random field model and the expectation–maximization algorithm. IEEE Trans Med Imaging 20:45–57.
53. Vom Hofe A, Mainemarre G, Vannier L (1998): Sensitivity to everydayfailures and cognitive inhibition: Are they related? Europ Jour App Psy-chol 48:49 –55.
54. Rast P, Zimprich D, Van Boxtel M, Jolles J (2009): Factor structure andmeasurement invariance of the cognitive failures questionnaire acrossthe adult life span. Assessment 16:145–158.
55. Robertson IH, Manly T, Andrade J, Baddeley BT, Yiend J (1997): “Oops!”:Performance correlates of everyday attentional failures in traumaticbrain injured and normal subjects. Neuropsychologia 35:747–758.
56. Smith SM, Jenkinson M, Johansen-Berg H, Rueckert D, Nichols TE, Mac-kay CE, et al. (2006): Tract-based spatial statistics: Voxelwise analysis ofmulti-subject diffusion data. Neuroimage 31:1487–1505.
57. Edden RA, DKJ (2010): Skeleton Thickness Biases Statistical Power in Skel-eton-Based Analyses of Diffusion MRI Data. Stockholm: International So-ciety for Magnetic Resonance in Medicine.
58. Corbetta M, Shulman GL (2002): Control of goal-directed and stimulus-driven attention in the brain. Nat Rev Neurosci 3:201–215.
59. Fox MD, Corbetta M, Snyder AZ, Vincent JL, Raichle ME (2006): Sponta-neous neuronal activity distinguishes human dorsal and ventral atten-tion systems. Proc Natl Acad Sci U S A 103:10046 –10051.
60. Dosenbach NU, Fair DA, Cohen AL, Schlaggar BL, Petersen SE (2008): Adual-networks architecture of top-down control. Trends Cogn Sci 12:99 –105.
61. Dosenbach NU, Fair DA, Miezin FM, Cohen AL, Wenger KK, DosenbachRA, et al. (2007): Distinct brain networks for adaptive and stable taskcontrol in humans. Proc Natl Acad Sci U S A 104:11073–11078.
62. Gee DG, Biswal BB, Kelly C, Stark DE, Margulies DS, Shehzad Z, et al.(2010): Low frequency fluctuations reveal integrated and segregatedprocessing among the cerebral hemispheres. Neuroimage 54:517–527.
63. Spencer KM, Nestor PG, Niznikiewicz MA, Salisbury DF, Shenton ME,McCarley RW (2003): Abnormal neural synchrony in schizophrenia.J Neurosci 23:7407–7411.
64. Pettigrew JD, Miller SM (1998): A “sticky” interhemispheric switch inbipolar disorder? Proc Biol Sci 265:2141–2148.
65. Clarke AR, Barry RJ, Heaven PC, McCarthy R, Selikowitz M, Byrne MK(2008): EEG coherence in adults with attention-deficit/hyperactivity dis-order. Int J Psychophysiol 67:35– 40.
66. He BJ, Snyder AZ, Vincent JL, Epstein A, Shulman GL, Corbetta M (2007):Breakdown of functional connectivity in frontoparietal networks under-lies behavioral deficits in spatial neglect. Neuron 53:905–918.
67. Li Z, Moore AB, Tyner C, Hu X (2009): Asymmetric connectivity reductionand its relationship to “HAROLD” in aging brain. Brain Res 1295:149 –158.
68. Cabeza R (2002): Hemispheric asymmetry reduction in older adults: TheHAROLD model. Psychol Aging 17:85–100.
69. Park DC, Reuter-Lorenz P (2009): The adaptive brain: Aging and neuro-cognitive scaffolding. Annu Rev Psychol 60:173–196.
70. Asplund CL, Todd JJ, Snyder AP, Marois R (2010): A central role for thelateral prefrontal cortex in goal-directed and stimulus-driven attention.Nat Neurosci 13:507–512.
71. Raichle ME (2010): Two views of brain function. Trends Cogn Sci 14:180 –190.
72. Mennes M, Kelly C, Zuo XN, Di Martino A, Biswal BB, Castellanos FX,Milham MP (2010): Inter-individual differences in resting-state func-tional connectivity predict task-induced BOLD activity. Neuroimage 50:
1690 –1701.
ww.sobp.org/journal
3. Carter AR, Astafiev SV, Lang CE, Connor LT, Rengachary J, Strube MJ, etal. (2010): Resting interhemispheric functional magnetic resonance im-aging connectivity predicts performance after stroke. Ann Neurol 67:365–375.
4. van Meer MP, van der Marel K, Wang K, Otte WM, El Bouazati S, RoelingTA, et al. (2010): Recovery of sensorimotor function after experimentalstroke correlates with restoration of resting-state interhemisphericfunctional connectivity. J Neurosci 30:3964 –3972.
5. Fecteau S, Knoch D, Fregni F, Sultani N, Boggio P, Pascual-Leone A(2007): Diminishing risk-taking behavior by modulating activity in theprefrontal cortex: A direct current stimulation study. J Neurosci 27:12500 –12505.
6. Knoch D, Gianotti LR, Pascual-Leone A, Treyer V, Regard M, Hohmann M,Brugger P (2006): Disruption of right prefrontal cortex by low-frequencyrepetitive transcranial magnetic stimulation induces risk-taking behav-ior. J Neurosci 26:6469 – 6472.
7. Makris N, Gasic GP, Kennedy DN, Hodge SM, Kaiser JR, Lee MJ, et al.(2008): Cortical thickness abnormalities in cocaine addiction—A reflec-tion of both drug use and a pre-existing disposition to drug abuse?Neuron 60:174 –188.
9. Beveridge TJ, Gill KE, Hanlon CA, Porrino LJ (2008): Review. Parallel studiesof cocaine-related neural and cognitive impairment in humans and mon-keys. Philos Trans R Soc Lond B Biol Sci 363:3257–3266.
0. Agrawal A, Lynskey MT (2008): Are there genetic influences on addic-tion: Evidence from family, adoption and twin studies. Addiction 103:1069 –1081.
1. Xu J, DeVito EE, Worhunsky PD, Carroll KM, Rounsaville BJ, Potenza MN(2010): White matter integrity is associated with treatment outcomemeasures in cocaine dependence. Neuropsychopharmacology 35:1541–1549.
2. Kelly C, Foxe JJ, Garavan H (2006): Patterns of normal human brainplasticity after practice and their implications for neurorehabilitation.Arch Phys Med Rehabil 87:S20 –S29.
3. Pfefferbaum A, Sullivan EV, Hedehus M, Lim KO, Adalsteinsson E, Mose-ley M (2000): Age-related decline in brain white matter anisotropy mea-sured with spatially corrected echo-planar diffusion tensor imaging.Magn Reson Med 44:259 –268.
5. Franklin TR, Acton PD, Maldjian JA, Gray JD, Croft JR, Dackis CA, et al.(2002): Decreased gray matter concentration in the insular, orbitofron-tal, cingulate, and temporal cortices of cocaine patients. Biol Psychiatry51:134 –142.
6. Sim ME, Lyoo IK, Streeter CC, Covell J, Sarid-Segal O, Ciraulo DA, et al.(2007): Cerebellar gray matter volume correlates with duration of co-caine use in cocaine-dependent subjects. Neuropsychopharmacology32:2229 –2237.
7. Matochik JA, London ED, Eldreth DA, Cadet JL, Bolla KI (2003): Frontalcortical tissue composition in abstinent cocaine abusers: A magneticresonance imaging study. Neuroimage 19:1095–1102.
8. Biswal BB, Mennes M, Zuo XN, Gohel S, Kelly C, Smith SM, et al. (2010):Toward discovery science of human brain function. Proc Natl Acad Sci U S A107:4734–4739.
9. Kriegeskorte N, Simmons WK, Bellgowan PS, Baker CI (2009): Circularanalysis in systems neuroscience: The dangers of double dipping. NatNeurosci 12:535–540.
0. Vul E, Harris C, Winkielman P, Pashler H (2009): Puzzlingly high correla-tions in fMRI studies of emotion, personality and social cognition. PerspPsychol Sci 4:274 –290.
1. Poldrack RA, Mumford JA (2009): Independence in ROI analysis: Where is
