1

2

3Q1

4

5

6

7

89101112131415

1 6

171819

202122232425

42

4344

45

46

47

48

49Q2

50

51

52

53

NeuroImage xxx (2014) xxx–xxx

YNIMG-11691; No. of pages: 8; 4C: 4, 5

Contents lists available at ScienceDirect

NeuroImage

j ourna l homepage: www.e lsev ie r .com/ locate /yn img

Decreased centrality of subcortical regions during the transition toadolescence: A functional connectivity study

PRO

OFJoão Ricardo Sato a,c,h,⁎, Giovanni Abrahão Salum b,h, Ary Gadelha c,h, Gilson Vieira e,g, André Zugman c,h,

Felipe Almeida Picon b,h, Pedro Mario Pan c,h, Marcelo Queiroz Hoexter c,d,h, Mauricio Anés b,h,Luciana Monteiro Moura c,h, Marco Antonio Gomes Del’Aquilla c,h, Nicolas Crossley f, Edson Amaro Junior g,Philip Mcguire f, Acioly L.T. Lacerda c,h, Luis Augusto Rohde b,h, Euripedes Constantino Miguel d,h,Andrea Parolin Jackowski c,h, Rodrigo Affonseca Bressan c,h

a Center of Mathematics, Computation and Cognition, Universidade Federal do ABC, Santo Andre, Brazilb Department of Psychiatry, Federal University of Rio Grande do Sul, Porto Alegre, Brazilc Interdisciplinary Lab for Clinical Neurosciences (LiNC), Universidade Federal de Sao Paulo (UNIFESP), Sao Paulo, Brazild Department of Psychiatry, School of Medicine, University of Sao Paulo, Brazile Bioinformatics Program, Institute of Mathematics and Statistics, University of Sao Paulo, Brazilf Institute of Psychiatry, King’s College London, United Kingdomg Department of Radiology, School of Medicine, University of Sao Paulo, Brazilh National Institute of Developmental Psychiatry for Children and Adolescents, CNPq, Brazil

O

⁎ Corresponding author at: Av. dos Estados, 5001, BairrCEP 09210-580.

E-mail address: joao.sato@ufabc.edu.br (J.R. Sato).

http://dx.doi.org/10.1016/j.neuroimage.2014.09.0631053-8119/© 2014 Elsevier Inc. All rights reserved.

Please cite this article as: Sato, J.R., et al., Decrstudy, NeuroImage (2014), http://dx.doi.org

ED

a b s t r a c t

a r t i c l e i n f o

26

27

28

29

30

31

32

33

34

Article history:Accepted 26 September 2014Available online xxxx

Keywords:NeurodevelopmentNeuroimagingNetworksChildrenGraph

35

36

37

38

39

40

41

RRECTInvestigations of brainmaturation processes are a key step to understand the cognitive and emotional changes of

adolescence. Although structural imagingfindingshavedelineated clear brain developmental trajectories for typ-ically developing individuals, less is known about the functional changes of this sensitive development period.Developmental changes, such as abstract thought, complex reasoning, and emotional and inhibitory control,have been associatedwithmore prominent cortical control. The aim of this study is to assess brain networks con-nectivity changes in a large sample of 7- to 15-year-old subjects, testing the hypothesis that cortical regions willpresent an increasing relevance in commanding the global network. Functional magnetic resonance imaging(fMRI) data were collected in a sample of 447 typically developing children from a Brazilian community samplewhowere submitted to a resting state acquisition protocol. The fMRI datawere used to build a functional weight-ed graph fromwhich eigenvector centrality (EVC)was extracted. For each brain region (a node of the graph), theage-dependent effect on EVC was statistically tested and the developmental trajectories were estimated usingpolynomial functions.Our findings show that angular gyrus become more central during this maturation period, while the caudate;cerebellar tonsils, pyramis, thalamus; fusiform, parahippocampal and inferior semilunar lobe become less central.In conclusion, we report a novel finding of an increasing centrality of the angular gyrus during the transition toadolescence, with a decreasing centrality of many subcortical and cerebellar regions.

© 2014 Elsevier Inc. All rights reserved.

C

54

55

56

57

58

59

60

61

UNIntroduction

The transition from childhood to adolescence involves majorchanges in cognitive and emotional functions (Paus, 2005; Rubia,2014). Evidence supports that such changes are a result of importantrefinements in complex neural dynamics and may ultimately reflectthe organization of the human brain (Fair et al., 2009; Kelly et al.,2009; Stevens et al., 2009; Dosenbach et al., 2010; Blakemore, 2012).

62

63

64

65

66

o Bangu, Santo André, SP, Brazil,

eased centrality of subcortical/10.1016/j.neuroimage.2014.0

The brain can be conceptualized as a set of partially segregated net-works emerging froma complex relationship between a variety of nervecells organized into circuits and functional areas (Power et al., 2010).Neurodevelopmental studies investigating the trajectories of thesecomplex networks may shed light on the hierarchical structure of suchcircuitries during development. These types of studies are especially im-portant in transitional periods, such as adolescence, as they can advanceour understanding of this window of vulnerability for disruptions intypical development and increased incidence of mental disorders(Kim-Cohen et al., 2003; Ernst et al., 2006; Insel, 2009; Salum et al.,2010).

Several brain structural changes occur during this period. Althoughsome regions exhibit linear changes in cortical thickness and volume,

regions during the transition to adolescence: A functional connectivity9.063

T

67

68

69

70

71

72

73

74

75

76

77

78

79

80

81

82

83

84

85

86

87

88Q3

89

90

91

92

93

94

95

96

97

98

99

100

101

102

103

104

105

106

107

108

109

110

111

112

113

114

115

116

117

118

119

120

121

122

123

124

125

126

127

128

129

130

131

132

133

134

135

136

137

138

139

140

141

142

143

144

145

146

147

148

149

150

151

152

153

154

155

156

157

158

159

160

161

162

163

164

165

166

167

168

169

170

171

172

173

174

175

176

177

178

179

180

181

182

183

184

185

186

187

188

189

190

191

192

193

2 J.R. Sato et al. / NeuroImage xxx (2014) xxx–xxx

UNCO

RREC

others appear to follow amaturational course, with an initial childhoodincrease until approximately 7–10 years of age and a subsequent ado-lescent decline (Giedd, 2004; Gogtay et al., 2004; Shaw et al., 2008).This decline in thickness during adolescence is followed by a period ofslower decline, stabilizing in adulthood (Shaw et al., 2008; Sowellet al., 2003).

Intrinsic functional connectivity (Biswal et al., 1995; Buckner et al.,2013) based on fMRI is sensitive to coupling dynamics and has theability to broadly survey the whole brain, providing information aboutrelationships between networks (Buckner et al., 2013). The first studiesinvestigating the development of brain networks (Fair et al., 2007, 2008,2009; Dosenbach et al., 2010) focused on the organizing principles ofsuch development and found evidence of functional segregation ofnearby functional areas across time, which occur through the weaken-ing of short-range functional connections. These studies also detectedthe integration of distant regions in a functional network, which occurby strengthening long-range functional connections. Subsequentstudies more specifically investigated the meaning of such connectionsand addressed important questions, such as communication betweencortical and subcortical areas that are thought to undergo dramaticchanges over time, especially in adolescence (Paus, 2005; Lebel et al.,2008; Supekar et al., 2009; Blakemore, 2012; Rubia, 2014). Studies havesuggested that children have stronger subcortical-cortical and weakercortico-cortical connectivity compared to young adults (Supekar et al.,2009). Some research (Power et al., 2012; Satterthwaite et al., 2012;Van Dijk et al., 2012) has shown that earlier developmental studiescould have been influenced by head movement artifacts. These maybias functional connectivity by decreasing long-distance correlations (ofBOLD signal) and increasing short-distance correlations. Interestingly,many initial findings were replicated in a recent study, even afteradjusting for movement (Satterthwaite et al., 2013a).

A method that is increasingly being used to investigate the rele-vance of brain network nodes comes from a branch of mathematicscalled graph theory. In graph theory, a network is a collection ofitems (nodes) that possesses pairwise relationships (edges). Graphsrepresent the collection of nodes and edges. There are a variety ofmethods to characterize how important or central a node is to thenetwork, and each seems to capture independent aspects of brainnetworks (Bullmore and Sporns, 2009; Rubinov and Sporns, 2010;Sporns, 2011; Zuo et al., 2012). Investigating a variety of measuresof centrality can reveal how brain regions communicate and charac-terize network properties that support inter-regional interactions(Hwang et al., 2013).

Nevertheless, few studies have investigated developmental changesof these brain networks in periods of high incidence (or onset) of manypsychiatric disorders. Moreover, studies so far have included sampleswith potential selection bias, such as those composed of voluntaryresearch subjects or specific populations (Uher and Rutter, 2012).Therefore, it is unclear if such developmental effects can be easily trans-lated to population samples. In addition,most studies relied on relative-ly small sample sizes (fewer than one hundred) and concentrate solelyon American and European populations (Satterthwaite et al., 2013b),which are limited on miscegenation (genetic admixture) and environ-mental factors (e.g., education and violence). Lastly, all available studieshave been performed in high-income countries, and to the best of ourknowledge, there are no neuroimaging development studies in low-and middle-income countries based on large samples.

This study examines a unique large community sample fromadevel-oping country (Brazil) with high levels of genetic admixture. It is impor-tant to highlight that, due to environmental factors (e.g., poverty,violence, poor education and medical assistance, etc.), children livingin developing countries are more vulnerable to mental disorders andrepresent the majority of such cases worldwide (Kieling et al., 2009).Thus, the investigation of children and adolescents under these environ-mental and genetic factors is crucial to enhance the comprehension ofthe typical neurodevelopment.

Please cite this article as: Sato, J.R., et al., Decreased centrality of subcorticalstudy, NeuroImage (2014), http://dx.doi.org/10.1016/j.neuroimage.2014.0

ED P

RO

OF

Here, we performed a whole-brain graph analysis aiming to identifybrain regions with age-effects on eigenvector centrality (EVC)measure.Given previous empirical evidence demonstrating a process of corticalmaturation (Fair et al., 2007, 2008, 2009; Supekar et al., 2009; Shawet al., 2008; Satterthwaite et al., 2013a), the existence of developmentalchanges in global network properties (Hwang et al., 2013) and the pres-ence of hub regions in adults (Power et al., 2013), we hypothesize thatage-related changes occur in the central regions of brain networks,with an increase in cortical integrative regions.

Materials and methods

Participants

The participants of this study are a subsample (N = 447, healthychildren) of the High Risk Cohort Study for Psychiatric Disorders inChildhood (HRC, N = 2512 children). For a detailed description of thiscohort, see Salum et al. (2013, in press). The current investigation wasbased on healthy children from two different Brazilian centers in the cit-ies of São Paulo and Porto Alegre. The study was approved by the localethics committee (University of Sao Paulo, IORG0004884, 1138/08)and required written consent from all parents and verbal assent fromall children. From the total cohort of 2512 subjects, 1004 childrenwere invited to participate in the MRI scanning. Seven hundred forty-one subjects (and the parent/guardian) accepted the invitation andwere scanned. From these, 447 children were healthy and presentedtypical development (229 males; mean age ± standard deviation:9.51 ± 1.92, from 7 to 15 years old; 248 scanned at São Paulo and199 at PortoAlegre). None of the volunteers fulfilled any criteria for psy-chiatric diagnoses (further details on Salum et al., 2013, in press) basedon the Development and Well-Being Assessment (DAWBA, Goodmanet al., 2000) interview (answered by the parents, guardian or caregiver).On the day of neuroimaging data acquisition, parents/caregiver alsofilled the Child Behavior Checklist (CBCL; Achenbach and Rescorla,2001). The estimation of intelligence quotient (IQ) was carried outusing the vocabulary and block design subtests of the Weschler Intelli-gence Scale for Children, using the method proposed by Tellegen andBriggs (1967). The socioeconomic status was assessed using theBrazilian rating scale (ABIMEP) and categorized into very low/low (Eand D classes), medium (C and B classes) and comfortable (A class)groups. Fig. 1 depicts the summary of demographical information.

In São Paulo, the scans were performed at the Institute of Radiology(INRAD) of the Hospital das Clínicas and in Porto Alegre at Santa CasaFoundation. The volunteers and parents did not receive any financialcompensation for participating in this study, according to the lawsestablished by the Brazilian Federal Ethics Committee for experimentsinvolving humans.

On the day of the examinations, recreational techniques were usedas a desensitization method to improve adherence and decrease move-ment during MRI scans. In addition, a simulation of the MRI acquisitionprocedure using a child’s fabric play tunnel. Before the start of the acqui-sition of the resting-state fMRI data, the technician asked the child tolook at the black dot painted on the magnet and not to sleep. Anotherverbal contact was performed after the acquisition to ensure that thechild was awake.

Image acquisition and resting state protocol

The image acquisition was carried out in two 1.5-T MRI scanners(models Signa HDX and HD from G.E., United States of America) usingthe same parameters and protocols. Functional MRI acquisitionconsisted of 180 EPI volumes (TR = 2000 ms, TE = 30 ms, slicethickness = 4 mm, gap = 0.5 mm, flip angle = 80°, matrix size =80 × 80, NEX = 1, slices = 26, total acquisition time = 6 min). ThefMRI images were obtained during a resting state protocol with theeyes open and a fixation point (a small black dot painted inside the

regions during the transition to adolescence: A functional connectivity9.063

RECTED P

RO

OF

194

195

196

197

198

199

200

201

202

203

204

205

206

207

208

209

210

211

212

213

214

215

216

217

218

219

220

221

222

223

224

225

226

227

228Q4

229

230

231

232

233

234

235

236

237

238

239

Fig. 1. Demographical information of this sample.

3J.R. Sato et al. / NeuroImage xxx (2014) xxx–xxx

UNCO

Rmagnet). Anatomical T1-weighted scans were acquired in a maximumof 156 axial slices (TR = 10.916 ms, TE = ms 4.2, thickness =1.2 mm, flip angle = 15°, matrix size = 256 × 192, FOV = 245 mm,NEX = 1, bandwidth = 122.109).

Brain imaging and connectivity analysis

Functional and structural data processingwas carried out using AFNI(version 2011_12_21_1014; Cox, 1996) and FSL (version 5.0; Jenkinsonet al., 2012; Woolrich et al., 2009) packages. The preprocessing stepswere carried out in the following order: EPI images had their first fourvolumes discarded to achieve steady state and the skull stripped to im-prove registration. The images were subsequently motion corrected,despiked and normalized to a grand-mean of 10,000. Next, the datawere band-pass filtered to frequencies between 0.01 Hz and 0.1 Hzand detrended using first- and second-order polynomials. The datawere spatially smoothed using a Gaussian kernel (FWHM = 8 mm)and linearly registered to their respective structural scan. Structural im-ageswere also skull-stripped and registered to standard space using theMontreal Neurological Institute (MNI) template. This transformationwas used to create a map from the volunteers’ functional native spaceto the standard space by nonlinear registration procedures (FNIRT).Nuisance covariates (averaged time series of voxels inside the CSF andwhite matter, global signal as well as the six linear parameters of

Please cite this article as: Sato, J.R., et al., Decreased centrality of subcorticalstudy, NeuroImage (2014), http://dx.doi.org/10.1016/j.neuroimage.2014.0

motion) were extracted from functional images using brain maskscreated by FAST based on the volunteers’ structural scan. Lastly, the re-siduals after regressing out the nuisance covariates from the functionaldata were linearly resampled to standard space using FEAT.

It has been alleged that head motion is one of the main sources ofvariation in functional connectivity measures and can have a largeimpact in studies involving patients with psychiatric disorders and/orchildren. The effects of such artifacts on functional connectivity analysesmight be reduced using the scrubbingmethod proposed by Power et al.(2012). Volumes were excluded (flagged scan, one before and twoafter) if the respective frame-wise displacement (FD, based on equation9 of Yan et al., 2013) or the temporal derivative of the RMS variance overthe voxels (DVARS, Power et al., 2011) was larger than the 95% percen-tile of the whole sample. Our choice to use the percentile-based cutoffsfor DVARS and FD was made to avoid scaling problems (mainly forDVARS). The rationale of using a cutoff for discarding the “worst”scans was to preserve the relative motionless data of the sample byremoving the outliers.

A whole-brain analysis was carried out based on the average BOLDsignal of 325 similarly sized and functionally homogenous regions of in-terest (ROIs), defined in accordance to the AT325 atlas (Craddock et al.,2012; see Supplementary Material). Pearson's correlations between thesignals of pairs of regions were used as the measure of functional con-nectivity. As short-distance correlations may arise due to smoothing,

regions during the transition to adolescence: A functional connectivity9.063

T

240

241

242

243

244

245

246

247

248

249

250

251Q5

252

253

254

255

256

257

258

259

260

261

262

263

264

265

266

267

268

269

270

271

272

273

274

275

276

277

278

279

280

281

282

283

284

285

286

287

288

289

290

291

292

293

294

295

296

297

298

299

300

301

302

303

304

305

t1:1 Table 1t1:2 Regions of interest with significant age-dependent effects on eigenvector centrality (corrected p-value b 5%). The sign column represents an increase or decrease in EVC with age.

t1:3 Center of mass—MNI (mm)

t1:4 Brain Region ID X Y Z p-value Ajusted R2 Sign

t1:5 Left angular gyrus 110 −44.89 −66.46 38.51 0.042 0.057 +t1:6 Left caudate nucleus 115 −12.34 −5.77 22.27 0.004 0.065 -t1:7 Right inferior semilunar lobe 156 16.91 −75.77 −44.3 0.008 0.06 -t1:8 Left parahippocampal gyrus 169 −21.15 −7.56 −17.13 0.018 0.04 -t1:9 Right pyramis 208 9.53 −76.44 −32.01 0.002 0.056 -t1:10 Left cerebellar tonsil 219 −8.97 −62.37 −43.89 0.035 0.058 -t1:11 Left cerebellum posterior lobe 85 −27.95 −56.04 −38.48 0.033 0.058 -t1:12 Left caudate 236 −12.88 12.96 12.84 0.008 0.04 -t1:13 Right fusiform gyrus 240 33.71 −37.59 −13.21 0.018 0.039 -t1:14 Right thalamus 257 12.97 −17.72 15.88 0.037 0.058 -t1:15 Left parietal subgyral 315 −34.49 −28.84 47.38 0.017 0.05 -

Fig. 2. Eigenvector centrality maps—the brain network nodes with significant (correctedp b 0.05) age-dependent effects on EVC are highlighted in red-yellow scales (EVCincreases with age) and in blue (EVC decreases with age).

4 J.R. Sato et al. / NeuroImage xxx (2014) xxx–xxx

UNCO

RREC

slicing and head movement (Power et al., 2013; Grayson et al., 2014),we removed the short-distance connections (less than 20 mm) fromthe correlation matrix.

For each subject, eigenvector centrality (EVC)was extracted for eachregion based on the undirected weighted graph defined by the individ-ual correlation matrix (adjacency matrix). The EVC of a node is propor-tional to the eigenvector centrality of the nodes in its neighbors(Bonacich, 1987). Thus, EVC is a “global” descriptor of centrality, mea-suring how the neighbors of a node are connected to the network.This metric was chosen because it quantifies the hierarchical relevanceof a node and is an established method in social network analysis(Wasserman and Faust, 2004) for investigating global (as opposed tolocal) features of the graph. Moreover, EVC is a common form of fMRIanalysis (Lohmann et al., 2010). As our focus is the investigation of“global” hierarchical organization and considering that the removal ofshort-distance connections may hinder the interpretation of somegraph metrics, we did not used other measures, such as degree, close-ness or betweenness. There are several different graph descriptors,and our decision to concentrate on EVCwas based on the aims, hypoth-esis and interpretability of the findings.

An age curve was then fitted for EVC at each node using quadraticpolynomials. This was achieved using a general linear model that alsotook into account the site of the acquisition and gender as covariates.The Wald test was used to statistically test for age effects. The Type Ierror was set at 5%, considering Bonferroni correction (325 multiplecomparisons, the number of ROIs). The brain regions were labeledbased on the MNI coordinates of the ROIs’ center of gravity and theSchaltenbrand atlas (Schaltenbrand et al., 1977).

Lastly, an exploratory analysis of the connectivity patterns of the re-gions with significant age effects in EVC was carried out. A whole brainfunctional connectivity mapping between the seed-ROIs and the otherROIswas performedwith a general linearmodel. Thismodel consideredthe Fisher transform of the correlation coefficient between the BOLDsignals as the response variable, age (and age squared) as main regres-sor, and both site and gender as covariates. The significance level ofthese exploratory analyses was 5% using the Bonferroni correction (in324 multiple tests, i.e., the number of ROIs minus one, the seed).

Results

The results of the age effects analysis revealed a statistically signifi-cant (corrected p-value b 0.05, Bonferroni method considering 325multiple comparisons, the number of ROIs) increase in the EVC of theangular gyrus. In addition, significant decreases in EVC measures werefound in the caudate, thalamus; cerebellar tonsils and pyramis; andthe fusiform, parahippocampal, parietal sub-gyral and semilunar lobule.The statistical information of significant age-dependent regions ishighlighted in Table 1 and Fig. 2 (red/blue scales depict increases/decreases in these measures over time). Although statistically signifi-cant, the percentage of the variance of EVC explained by age changes

Please cite this article as: Sato, J.R., et al., Decreased centrality of subcorticalstudy, NeuroImage (2014), http://dx.doi.org/10.1016/j.neuroimage.2014.0

ED P

RO

OF

is very small (all adjusted-R2 coefficients below 7%). The estimatedage curves of these regions are depicted in Fig. 3. Lastly, correlationanalysis between the mean EVC of the node (across subjects) and theROI size was significant (r = 0.180; p-value = 0.001), indicating anassociation between centrality and cluster extension.

To extract further information regarding age-related changes insome regions, we performed exploratory analyses focusing on ageeffects on the functional connectivity between seed-ROI and otherROIs. We selected left angular gyrus as a seed ROIs due to its relevanceas a cortical integrative region. The right thalamus and left caudatewere also selected as seed ROIs as they are subcortical regions thatsend and receive several anatomical connections and their EVCs werefound to be negatively correlated with age. Fig. 4 depicts the brainmapping of areas in which the functional connectivity with the seed isinfluenced by age. The angular gyrus functional connectivity mapshows several positive correlations with associative regions that arepart of the default mode network, such as the ventral and anterior me-dial prefrontal cortex and negative correlations with the supramarginal

regions during the transition to adolescence: A functional connectivity9.063

T

PRO

OF

306

307

308

309

310

311

312

313

314

315

316

317

318

319

Fig. 3. Age curves for eigenvector centrality of brain regions with significant age effects on this measure (corrected p b 0.05).

5J.R. Sato et al. / NeuroImage xxx (2014) xxx–xxx

gyrus. Caudate map depicts negative correlations with bilateral insula.The thalamus map highlights positive correlations with the insula andputamen and negative correlations with the occipital and dorsal anteri-or cingulate regions.

Regarding head movement, the cutoffs (95% percentile) for FD andDVARS were 0.45 mm and 46.36, which is very similar to those fre-quently used in previous studies. The mean frame displacement (FD)

UNCO

RREC

Fig. 4. Seed-ROI functional connectivity maps: exploratory analyses of age-related changesfunctional connectivity (with the seed ROI) over time are highlighted in red, and regions with

Please cite this article as: Sato, J.R., et al., Decreased centrality of subcorticalstudy, NeuroImage (2014), http://dx.doi.org/10.1016/j.neuroimage.2014.0

EDacross all subjects was 0.14 mm (standard deviation of 0.21), and the

mean DVARS was 25.25 mm (SD = 8.72). The mean (across subjects)of the percentage of discarded volumes was 11.97% (SD = 18.28%). Asexpected, the mean FD was negatively correlated with age (Pearsoncorrelation r = −0.19, p b 0.001), as was the mean DVARS (Pearsoncorrelation r = −0.21, p b 0.001) given that older children presentless head motion than younger ones.

in functional connectivity (p b 0.05, Bonferroni correction). The regions with increaseddecreased functional activity are in blue.

regions during the transition to adolescence: A functional connectivity9.063

T

320

321

322

323

324

325

326

327

328

329

330

331

332

333

334

335

336

337

338

339

340

341

342

343

344

345

346

347

348

349

350

351

352

353

354Q6

355Q7

356

357

358

359

360

361

362

363

364

365

366

367

368

369

370

371

372

373

374

375

376

377

378

379

380

381

382

383

384

385

386

387

388

389

390

391

392

393

394

395

396

397

398

399

400

401

402

403

404

405

406

407

408

409

410

411

412

413

414

415

416

417

418

419

420

421

422

423

424

425

426

427

428

429

430

431

432

433

434

435

436

437

438

439

440

441

442

443

444

445

446

447

448

449

450

6 J.R. Sato et al. / NeuroImage xxx (2014) xxx–xxx

UNCO

RREC

Discussion

In this study, we investigated the neurodevelopmental changes infunctional connectivity in a large sample of 447 healthy children andadolescents in an age range from 7 to 15 years old. Themainmotivationof characterizing typical development during this period is that mostpsychiatric disorders, which are deviations from the normal develop-mental trajectory, have an early onset in youth. Fifty percent of DSM-IV disorders have an onset before 14 years old, and 75% have an onsetbefore the age of 24 years (Kessler et al., 2005). Thus, the early identifi-cation of brain network changes can improve the understanding of thepathophysiology of mental disorders and provide awindow of opportu-nity to early neuroprotective interventions. This information can alsohelp clinicians by providing more precise diagnostic and prognosticinformation,which are fundamental to reduce the impact of these debil-itating conditions. Based on a data-driven and graph theory approach,our results point toward a reduction in EVC values in cerebellar andsubcortical regions and an increase of these values in the angular gyrus.

The observed shift in EVCvalueswith decreased relevance of subcor-tical areas in the ages investigated in our study occurs immediatelyfollowing the thickness peak described by Shaw et al. (2008). This resultsuggests an interplay between functional development and structuralmaturation. The EVC quantifies the relevance of a node in the contextof its neighbors and the entire network, being a relative measure. Ourfindings suggest that changes in the hierarchical reorganization of thenetwork over time are more significant than the changes in the func-tional integration between each node and the network. This resultmight suggest that the hierarchical reconfiguration and the integrationof different systems during development are more prominent thannode-to-network integration. In terms of functional development, somebasic configurations also appear to occur in the initial developmentalphases. Wu et al. (2013) suggested a small-world organization in func-tional brain networks in children. Other studies reported convergingresults (Gao et al., 2011; Fransson et al., 2011). Interestingly, EVC central-ity was found to be associated with connectivity flexibility induced bytask modulation in the aging brain (Geerligs et al., in press), cognitionand CSF biomarkers of Alzheimer’s disease (Binnewijzend et al., inpress). Furthermore, Zuo et al. (2012) have shown that adults exhibit adecrease in the degree of centrality across age, but such a change is notobserved in EVC. This finding suggests that while local connectivitydecreases with age in adults, connections with hub-like regions withinthe brain remain stable with age at a global level.

Regarding the relationship between our developmentalfindings andbehavior and cognition, Fair et al. (2008) highlighted that children al-ready present several introspective abilities, such as mentalizing,envisioning the future, and autobiographical memory, which are facul-ties associated with the DMN. In addition, childhood is a crucial periodin the development of social cognition and theory of mind, also relatedto the DMN (Jack et al., 2012). Attention modulation is also improvedduring childhood and adolescence (Wendelken et al., 2011). In ourstudy, the angular gyrus, a well-established integrative region that ispart of DMN, was found to become more central during development.The exploratory analysis also revealed that the functional connectivitybetween the angular gyrus and the ventral and medial prefrontalcortices, also part of DMN, increases with age. A decrease in the EVC inthe caudate and parahippocampal cortex (limbic system) and othernon-cortical regions, such as the cerebellum and thalamus, may beassociated with the development of voluntary modulation of behaviorssuch as fear, inhibitory control and sensitivity to reward delay, whichare observed in adults but are unstable in children.

Moreover, the exploratory analysis of our findings revealed that thesubcortical identified regions, such as the caudate and thalamus, exhibitage effects with respect to functional connectivity with other regions ofthe limbic system.We found a negative correlation between age and theconnectivity between the caudate and insula and a positive correlationbetween the thalamus and the insula. While the caudate is classically

Please cite this article as: Sato, J.R., et al., Decreased centrality of subcorticalstudy, NeuroImage (2014), http://dx.doi.org/10.1016/j.neuroimage.2014.0

ED P

RO

OF

associated with the reward system, the insula is involved in pain, fearand disgust (Eisenberger, 2012; Ipser et al., 2013; Stark et al., 2007). Itis well established that the thalamus is an important hub in the process-ing of sensory stimuli. Interestingly, we found a negative correlationwith age between the functional connectivity of the thalamus andboth the occipital cortex and cingulate gyrus. This result may suggesta reorganization of thalamic interactionswith the sensory and emotion-al systems during late childhood. Indeed, the main contributions of thecurrent study are the findings regarding the hierarchical reorganizationof subcortical regions in the context of the whole brain network. Themajority of the findings from previous studies concentrate on corticalhub regions or specific networks, such as the DMN (Fair et al. (2008).

Regarding comparisons of our sample with those used in previousstudies, a pioneer maturational study was conducted by Fair et al.(2008), based on anAmerican population, andwasmore focused on dif-ferences between adults and children. Hwang et al. (2013) investigatedan American population at a different age range (10–20 years). Theyfound that hub architectures were detectable in late childhood andstable from adolescence to early adulthood. Our results add to thesefindings by considering children from 7 to 10 years of age and by show-ing the curves estimates of EVC over time. Moreover, the data in Fig. 3suggests U-shaped EVC curves for some regions (e.g., the caudate andparahippocampal cortex) after 10 years of age, similar to the cortical de-velopmental process described by Shaw et al. (2008). In a very detailedinvestigation, Satterthwaite et al. (2013a) present the results of a largeAmerican sample with a similar age range of our sample. However,EVC was not explored in this previous study. In summary, our findingsare in line with and complement the findings of previous studies. Thediverse and large sample allowed a more detailed exploration of devel-opmental changes at the examined age span not only by increasing thestatistical power but also by allowing more accurate estimates of EVCcurves over time.

The strengths of our study include the following: (i) the use of atopological data-driven analysis of whole-brain networks in a largecommunity sample of children and pre-adolescents; (ii) the subjectswere extensively evaluated and did not present mental disorders; and(iii) the sample comes from a low and middle-income third-worldcountry. Some limitations must also be considered. Our data are cross-sectional rather than longitudinal, but our sample size allows us consid-er different ages with a large number of individuals. In addition, thesmall age range undermines our ability to assess what is uniquely im-mature during adolescence because no comparisons with adults werepossible. Although an association between EVC and cluster extensionwas found, we believe this did not strongly influence the results giventhat the ROI sizes were the same for all of the subjects. Moreover, ourmain results focus on changes in EVC over time and not identifying(or ranking) the most central regions. In addition, we used a conven-tional pipeline of temporal band-pass filtering and nuisance variableregression in two separate steps, not simultaneously, which may haveinfluenced age effects. Lastly, as previously mentioned, a recent study(Power et al., 2013) has shown that regions belonging to largernetworks tend to have high degree values. This fact may be a possibleconfounder in the analysis of functional brain networks when usingdegree as a descriptor. However, the effects of this confounder on EVChave not been investigated.

Studies addressing the impact of brain circuit conformations in be-havior and cognitive are highly warranted. Understanding typicalbrain development is an essential step because it is a primary strategyto identify deviant trajectories in structural and functional networksand, consequently, obtain more insights on the behavioral conse-quences of these trajectories. Detecting deviant trajectories may aidthe early identification of children and adolescents at risk for future cog-nitive and psychiatric disorders, and preventive interventions could betested in such subjects. Although the current study is cross-sectional,our results add to current literature by reinforcing a central role fornetwork reconfiguration and integration over development. This is

regions during the transition to adolescence: A functional connectivity9.063

T

451

452

453

454

455

456

457

458

459

460

461

462

463

464

465

466

467

468

469

470

471

472

473

474

475

476

477

478

479

480

481

482

483

484

485

486

487

488

489

490

491492493494Q9495496497498499500501502503504505506507508509510511512513514515516

517518519520521522523524525526527528529530531532533534535536537538Q10539540541542543544545546547548549550551552553554555556557558559560561562563564565566567568569570571572573574575576577578579580581582583584585586587588589590591592593594595596597598599600601Q11602

7J.R. Sato et al. / NeuroImage xxx (2014) xxx–xxx

UNCO

RREC

particularly true for the observed increase in relevance of cortical oversubcortical regions during specific periods of development. In addition,the data used in this study belong to an ongoing project in which thechildren are re-assessed and re-scanned after 3 years. Longitudinalstudies including individuals with and without psychiatric symptomswill be conducted to further corroborate these findings and to investi-gate abnormal trajectories associated with the onset of psychiatricsymptoms. Studies addressing the impact of changes in these braincircuit conformations in behavior and cognition are highly warranted.

Disclosure

Dr. Luis Augusto Rohde LAR has been on the speakers’ bureau/advisory board and/or acted as consultant for Eli-Lilly, Janssen-Cilag,Novartis and Shire in the last 3 years. The ADHD and Juvenile BipolarDisorder Outpatient Programs chaired by him received unrestricted ed-ucational and research support from the following pharmaceutical com-panies in the last 3 years: Eli-Lilly, Janssen-Cilag, Novartis, and Shire. Hereceives authorship royalties fromOxford Press and ArtMed. Dr. RodrigoAffonseca Bressan has been on the speakers’ bureau/advisory board ofAstraZeneca, Bristol, Janssen and Lundbeck; he has received researchgrants from Janssen, Eli Lilly, Lundbeck, Novartis, Roche, FAPESP, CNPq,CAPES, Fundação E.J. Safra, and Fundação ABAHDS. He is a shareholderof Biomolecular Technology Ltda. Dr. Edson Amaro Jr has receivedresearch grants from FAPESP, CNPq, CAPES, Fundação E.J. Safra, andFundação ABAHDS. Dr. Pedro Pan has received payment for the develop-ment of educational material for Janssen-Cilag and Astra-Zeneca.

Supplementary data to this article can be found online at http://dx.doi.org/10.1016/j.neuroimage.2014.09.063.

Acknowledgements

The opinions, hypothesis, conclusions, and recommendations of thisstudy are under the responsibility of the authors, not necessarily repre-sentative of the opinion of the funding agencies. The authors are gratefulto Sao Paulo Research Foundation–FAPESP (J.R.S. grant nos. 2013/10498-6 and 2013/00506-1), CAPES, and CNPq, Brazil, for funding thisresearch. This is a study from the National Institutes of Science andTechnology for Developmental Psychiatry of Children and Adolescents(INPD) supported by CNPq (573974/2008-0) and FAPESP (2008/57896-8) N.C. is supported by the Wellcome Trust. A.Z. receives a stu-dentship from CAPES-Brazil. We are also thankful to Eduardo JoaquimLopes Alho for the support in creating Table 1.

References

Achenbach, T.M., Rescorla, L.A., 2001. Manual for the ASEBA School-Age Forms andProfiles. University of Vermont, Research Center for Children, Youth, and Families,Burlington, VT.

Binnewijzend, M.A., Adriaanse, S.M., Van der Flier, W.M., Teunissen, C.E., de Munck, J.C.,Stam, C.J., Scheltens, P., van Berckel, B.N., Barkhof, F., Wink, A.M., 2014. Brain networkalterations in Alzheimer's disease measured by Eigenvector centrality in fMRI are re-lated to cognition and CSF biomarkers. Hum. Brain Mapp. http://dx.doi.org/10.1002/hbm.22335 (in press).

Biswal, B., Yetkin, F.Z., Haughton, V.M., Hyde, J.S., 1995. Functional connectivity in themotor cortex of resting human brain using echo-planar MRI. Magn. Reson. Med. 34(4), 537–541.

Blakemore, S.J., 2012. Imaging brain development: the adolescent brain. NeuroImage 61(2), 397–406.

Bonacich, P., 1987. Power and centrality: a family of measures. Am. J. Sociol. 92,1170–1182.

Buckner, R.L., Krienen, F.M., Yeo, B.T., 2013. Opportunities and limitations of intrinsicfunctional connectivity MRI. Nat. Neurosci. 16 (7), 832–837.

Bullmore, E., Sporns, O., 2009. Complex brain networks: graph theoretical analysis ofstructural and functional systems. Nat. Rev. Neurosci. 10 (3), 186–198.

Cox, R.W., 1996. AFNI: software for analysis and visualization of functional magneticresonance neuroimages. Comput. Biomed. Res. 29, 162–173.

Craddock, R.C., James, G.A., Holtzheimer III, P.E., Hu, X.P., Mayberg, H.S., 2012. A wholebrain fMRI atlas generated via spatially constrained spectral clustering. Hum. BrainMapp. 33 (8), 1914–1928.

Dosenbach, N.U., Nardos, B., Cohen, A.L., Fair, D.A., Power, J.D., Church, J.A., Nelson, S.M.,Wig, G.S., Vogel, A.C., Lessov-Schlaggar, C.N., Barnes, K.A., Dubis, J.W., Feczko, E.,

Please cite this article as: Sato, J.R., et al., Decreased centrality of subcorticalstudy, NeuroImage (2014), http://dx.doi.org/10.1016/j.neuroimage.2014.0

ED P

RO

OF

Coalson, R.S., Pruett, J.R., Barch, D.M., Petersen, S.E., Schlaggar, B.L., 2010. Predictionof individual brain maturity using fMRI. Science 329 (5997), 1358–1361.

Eisenberger, N.I., 2012. The neural bases of social pain: evidence for shared representa-tions with physical pain. Psychosom. Med. 74 (2), 126–135 (Review).

Ernst, M., Pine, D.S., Hardin, M., 2006. Triadic model of the neurobiology of motivatedbehavior in adolescence. Psychol. Med. 36 (3), 299–312.

Fair, D.A., Dosenbach, N.U., Church, J.A., Cohen, A.L., Brahmbhatt, S., Miezin, F.M., Barch,D.M., Raichle, M.E., Petersen, S.E., Schlaggar, B.L., 2007. Development of distinctcontrol networks through segregation and integration. Proc. Natl. Acad. Sci. U. S. A.104 (33), 13507–13512.

Fair, D.A., Cohen, A.L., Dosenbach, N.U., Church, J.A., Miezin, F.M., Barch, Raichle, M.E.,Petersen, S.E., Schlaggar, B.L., 2008. The maturing architecture of the brain's defaultnetwork. Proc. Natl. Acad. Sci. U. S. A. 105 (10), 4028–4032.

Fair, D.A., Cohen, A.L., Power, J.D., Dosenbach, N.U., Church, J.A., Miezin, F.M., Schlaggar,B.L., Petersen, S.E., 2009. Functional brain networks develop from a local to distribut-ed organization. PLoS Comput. Biol. 5 (5), e1000381.

Fransson, P., Aden, U., Blennow, M., Lagercrantz, H., 2011. The functional architecture ofthe infant brain as revealed by resting-state fMRI. Cereb. Cortex 21, 145–154.

Gao,W., Gilmore, J.H., Giovanello, K.S., Smith, J.K., Shen, D., Zhu, H., Lin,W., 2011. Temporaland spatial evolution of brain network topology during the first two years of life. PLoSOne 6 (9), e25278.

Geerligs, L., Saliasi, E., Renken, R.J., Maurits, N.M., Lorist, M.M., 2014. Flexible connectivityin the aging brain revealed by task modulations. Hum. Brain Mapp. http://dx.doi.org/10.1002/hbm.22437 (in press).

Giedd, J.N., 2004. Structural magnetic resonance imaging of the adolescent brain. Ann. N.Y. Acad. Sci. 1021, 77–85.

Gogtay, N., Giedd, J.N., Lusk, L., Hayashi, K.M., Greenstein, D., Vaituzis, A.C., Nugent III, T.F.,Herman, D.H., Clasen, L.S., Toga, A.W., Rapoport, J.L., Thompson, P.M., 2004. Dynamicmapping of human cortical development during childhood through early adulthood.Proc. Natl. Acad. Sci. U. S. A. 101 (21), 8174–8179.

Goodman, R., Ford, T., Richards, H., Gatward, R., Meltzer, H., 2000. The Development andWell-Being Assessment: description and initial validation of an integrated assess-ment of child and adolescent psychopathology. J. Child Psychol. Psychiatry 41 (5),645–655.

Grayson, D.S., Ray, S., Carpenter, S., Iyer, S., Dias, T.G., Stevens, C., Nigg, J.T., Fair, D.A., 2014.Structural and functional rich club organization of the brain in children and adults.PLoS One 9 (2), e88297.

Hwang, K., Hallquist, M.N., Luna, B., 2013. The development of hub architecture in thehuman functional brain network. Cereb. Cortex 23 (10), 2380–2393.

Insel, T.R., 2009. Disruptive insights in psychiatry: transforming a clinical discipline. J. Clin.Invest. 119 (4), 700–705.

Ipser, J.C., Singh, L., Stein, D.J., 2013. Meta-analysis of functional brain imaging in specificphobia. Psychiatry Clin. Neurosci. 67 (5), 311–322 (Review).

Jack, A.I., Dawson, A.J., Begany, K.L., Leckie, R.L., Barry, K.P., Ciccia, A.H., Snyder, A.Z., 2012.fMRI reveals reciprocal inhibition between social and physical cognitive domains.NeuroImage 66C, 385–401.

Jenkinson, M., Beckmann, C.F., Behrens, T.E.J., Woolrich, M.W., Smith, S.M., 2012. FSL.NeuroImage 62, 782–790.

Kelly, A.M., Di Martino, A., Uddin, L.Q., Shehzad, Z., Gee, D.G., Reiss, P.T., Margulies, D.S.,Castellanos, F.X., Milham, M.P., 2009. Development of anterior cingulate functionalconnectivity from late childhood to early adulthood. Cereb. Cortex 19 (3), 640–657.

Kessler, R.C., Berglund, P., Demler, O., Jin, R., Merikangas, K.R., Walters, E.E., 2005. Lifetimeprevalence and age-of-onset distributions of DSM-IV disorders in the NationalComorbidity Survey Replication. Arch. Gen. Psychiatry 62 (6), 593–602.

Kieling, C., Herrman, H., Patel, V., Tyrer, P., Mari, J.J., 2009. A global perspective on thedissemination of mental health research. Lancet 374 (9700), 1500.

Kim-Cohen, J., Caspi, A., Moffitt, T.E., Harrington, H., Milne, B.J., Poulton, R., 2003. Priorjuvenile diagnoses in adults with mental disorder: developmental follow-back of aprospective-longitudinal cohort. Arch. Gen. Psychiatry 60 (7), 709–717.

Lebel, C., Walker, L., Leemans, A., Phillips, L., Beaulieu, C., 2008.Microstructural maturationof the human brain from childhood to adulthood. NeuroImage 40 (3), 1044–1055.

Lohmann, G., Margulies, D.S., Horstmann, A., Pleger, B., Lepsien, J., Goldhahn, D., Schloegl,H., Stumvoll, M., Villringer, A., Turner, R., 2010. Eigenvector centrality mapping foranalyzing connectivity patterns in fMRI data of the human brain. PLoS One 5 (4),e10232.

Paus, T., 2005. Mapping brainmaturation and cognitive development during adolescence.Trends Cogn. Sci. 9 (2), 60–68.

Power, J.D., Fair, D.A., Schlaggar, B.L., Petersen, S.E., 2010. The development of humanfunctional brain networks. Neuron 67 (5), 735–748.

Power, J.D., Barnes, K.A., Snyder, A.Z., Schlaggar, B.L., Petersen, S.E., 2012. Spurious butsystematic correlations in functional connectivity MRI networks arise from subjectmotion. NeuroImage 59 (3), 2142–2154.

Power, J.D., Schlaggar, B.L., Lessov-Schlaggar, C.N., Petersen, S.E., 2013. Evidence for hubsin human functional brain networks. Neuron 79 (4), 798–813.

Rubia, K., 2014. Functional brain imaging across development. Eur. Child Adolesc. Psychiatry22 (12), 719–731.

Rubinov, M., Sporns, O., 2010. Complex networkmeasures of brain connectivity: uses andinterpretations. NeuroImage 52 (3), 1059–1069.

Salum, G.A., Polanczyk, G.V., Miguel, E.C., Rohde, L.A., 2010. Effects of childhood develop-ment on late-life mental disorders. Curr. Opin. Psychiatry 23 (6), 498–503.

Salum, G.A., Mogg, K., Bradley, B.P., Gadelha, A., Pan, P., Tamanaha, A.C., Moriyama, T.,Graeff-Martins, A.S., Jarros, R.B., Polanczyk, G., do Rosário, M.C., Leibenluft, E.,Rohde, L.A., Manfro, G.G., Pine, D.S., 2013. Threat bias in attention orienting: evidenceof specificity in a large community-based study. Psychol. Med. 43, 733–745.

Salum, G.A., Gadelha, A., Pan, P.M., Moriyama, T.S., Graeff-Martins, A.S., Tamanaha, A.C.,Alvarenga, P., Krieger, F.V., Fleitlich-Bilyk, B., Jackowski, A., Sato, J.R., Brietzke, E.,

regions during the transition to adolescence: A functional connectivity9.063

603604605606607608609610611612613614615616617618619620621622623624625626627628629630631

632633634635636637638639640641642643644645646647648649650651652653654655656657658659660

661

8 J.R. Sato et al. / NeuroImage xxx (2014) xxx–xxx

Polanczyk, G.V., Brentani, H., Mari, J.J., Rosário, M.C., Manfro, G.G., Affonseca-Bressan,R., Mercadante, M.T., Miguel, E.C., Rohde, L.A., 2014. High risk cohort study for psychi-atric disorders in childhood: rationale, design, methods and preliminary results. Int. J.Methods Psychiatr. Res. (in press).

Satterthwaite, T.D., Wolf, D.H., Loughead, J., Ruparel, K., Elliott, M.A., Hakonarson, H., Gur,R.C., Gur, R.E., 2012. Impact of in-scanner head motion onmultiple measures of func-tional connectivity: relevance for studies of neurodevelopment in youth. NeuroImage60 (1), 623–632.

Satterthwaite, T.D., Wolf, D.H., Ruparel, K., Erus, G., Elliott, M.A., Eickhoff, S.B., Gennatas,E.D., Jackson, C., Prabhakaran, K., Smith, A., Hakonarson, H., Verma, R., Davatzikos,C., Gur, R.E., Gur, R.C., 2013a. Heterogeneous impact of motion on fundamental pat-terns of developmental changes in functional connectivity during youth. NeuroImage83C, 45–57.

Satterthwaite, T.D., Elliott, M.A., Ruparel, K., Loughead, J., Prabhakaran, K., Calkins, M.E.,Hopson, R., Jackson, C., Keefe, J., Riley, M., Mentch, F.D., Sleiman, P., Verma, R.,Davatzikos, C., Hakonarson, H., Gur, R.C., Gur, R.E., 2013b. Neuroimaging of thePhiladelphia neurodevelopmental cohort. NeuroImage 86, 544–553.

Schaltenbrand, G., Hassler, R., Wahren, W., 1977. Atlas for Stereotaxy of the Human Brain:With an Accompanying Guide. Thieme, Stuttgart.

Shaw, P., Kabani, N.J., Lerch, J.P., Eckstrand, K., Lenroot, R., Gogtay, N., Greenstein, D.,Clasen, L., Evans, A., Rapoport, J.L., Giedd, J.N., Wise, S.P., 2008. Neurodevelopmentaltrajectories of the human cerebral cortex. J. Neurosci. 28, 3586–3594.

Sowell, E.R., Peterson, B.S., Thompson, P.M., Welcome, S.E., Henkenius, A.L., Toga, A.W., 2003.Mapping cortical change across the human life span. Nat. Neurosci. 6 (3), 309–315.

Sporns, O., 2011. The human connectome: a complex network. Ann. N. Y. Acad. Sci. 1224,109–125.

Stark, R., Zimmermann, M., Kagerer, S., Schienle, A., Walter, B., Weygandt, M., Vaitl, D.,2007. Hemodynamic brain correlates of disgust and fear ratings. NeuroImage 37(2), 663–673.

UNCO

RRECT

Please cite this article as: Sato, J.R., et al., Decreased centrality of subcorticalstudy, NeuroImage (2014), http://dx.doi.org/10.1016/j.neuroimage.2014.0

OO

F

Stevens, M.C., Pearlson, G.D., Calhoun, V.D., 2009. Changes in the interaction of resting-state neural networks from adolescence to adulthood. Hum. Brain Mapp. 30 (8),2356–2366.

Supekar, K., Musen, M., Menon, V., 2009. Development of large-scale functional brainnetworks in children. PLoS Biol. 7 (7), e1000157.

Tellegen, A., Briggs, P.F., 1967. Old wine in new skins: grouping Wechsler subtests intonew scales. J. Consult. Psychol. 31 (5), 499–506.

Uher, R., Rutter, M., 2012. Basing psychiatric classification on scientific foundation:problems and prospects. Int. Rev. Psychiatry 24 (6), 591–605.

VanDijk, K.R., Sabuncu,M.R., Buckner, R.L., 2012. The influence of headmotion on intrinsicfunctional connectivity MRI. NeuroImage 59 (1), 431–438.

Wasserman, S., Faust, K., 2004. Social Network Analysis: Methods and Applications,Second edition. Cambridge University Press, New York and Cambridge, ENG.

Wendelken, C., Baym, C.L., Gazzaley, A., Bunge, S.A., 2011. Neural indices of improvedattentional modulation over middle childhood. Dev. Cogn. Neurosci. 1 (2), 175–186.

Woolrich, M.W., Jbabdi, S., Patenaude, B., Chappell, M., Makni, S., Behrens, T., Beckmann, C., Jenkinson, M., Smith, S.M., 2009. Bayesian analysis of neuroimaging data in FSL.NeuroImage 45, S173–S186.

Wu, K., Taki, Y., Sato, K., Hashizume, H., Sassa, Y., Takeuchi, H., Thyreau, B., He, Y., Evans,A.C., Li, X., Kawashima, R., Fukuda, H., 2013. Topological organization of functionalbrain networks in healthy children: differences in relation to age, sex, and intelli-gence. PLoS One 8 (2), e55347.

Yan, C.-G., Cheung, B., Kelly, C., Colcombe, S., Craddock, R.C., DiMartino, A., Li, Q., Zuo, X.-N.,Castellanos, F.X., Milham, M.P., 2013. A comprehensive assessment of regional varia-tion in the impact of headmicromovements on functional connectomics. NeuroImage76, 183–201.

Zuo, X.N., Ehmke, R., Mennes, M., Imperati, D., Castellanos, F.X., Sporns, O., Milham, M.P.,2012. Network centrality in the human functional connectome. Cereb. Cortex 22 (8),1862–1875.

R

ED P

regions during the transition to adolescence: A functional connectivity9.063