Associations of physical activity and screen time with white matter microstructure in children from the general population

Associa ions o physical ac i i y and sc een ime wi h whi e ma e mic os uc u e in child en om he gene al popula ion Ma ía Rod iguez-Ayllon a , I onne P.M. De ks b , c , Michiel A. an den D ies b , c , I ene Es eban-Co nejo a , d , Je emy A. Lab ecque e , Junwen Yang-Huang c , , Hein Raa , Meike W. Ve nooij e , g , Tonya Whi e b , g , F ancisco B. O ega a , Henning Tiemeie b , h , * , Ryan L. Mue zel b , e a PROFITH “PROmo ing FITness and Heal h h ough physical ac i i y” esea ch g oup, Spo and Heal h Uni e si y Resea ch Ins i u e (iMUDS), Depa men o Physical and Spo s Educa ion, Facul y o Spo Sciences, Uni e si y o G anada, G anada, Spain b Depa men o Child and Adolescen Psychia y, E asmus MC –Sophia Child en’s Hospi al, Ro e dam, he Ne he lands c Gene a ion R S udy G oup, E asmus MC –Sophia Child en’s Hospi al, Ro e dam, he Ne he lands d Cen e o Cogni i e and B ain Heal h, Depa men o Psychology, No heas e n Uni e si y, Bos on, MA, USA e Depa men o Epidemiology, E asmus MC, Ro e dam, he Ne he lands Depa men o Public Heal h, E asmus MC, Ro e dam, he Ne he lands g Depa men o Radiology and Nuclea Medicine, E asmus MC, Ro e dam, he Ne he lands h Depa men o Social and Beha io al Science, Ha a d T.H. Chan School o Public Heal h, Bos on, MA, Uni ed S a es ARTICLE INFO Keywo ds: MRI Di usion enso imaging B ain de elopmen Seden a y beha io Tele ision iewing Video games Ac i e commu ing ABSTRACT Physical ac i i y and seden a y beha io s ha e been linked o a a ie y o gene al heal h benefi s and p oblems. Howe e , ew s udies ha e examined how physical ac i i y du ing childhood is ela ed o b ain de elopmen , wi h he majo i y o wo k o da e ocusing on ca dio-me abolic heal h. This s udy examines he associa ion be- ween physical ac i i y and sc een ime wi h whi e ma e mic os uc u e in he gene al pedia ic popula ion. In a sample o 2532 child en (10.12 0.58 yea s; 50.04% boys) om he Gene a ion R S udy, a popula ion-based coho in Ro e dam, he Ne he lands, we assessed physical ac i i y and sc een ime using pa en - epo ed ques ionnai es. Magne ic esonance imaging o whi e ma e mic os uc u e was conduc ed using di usion enso imaging. To al physical ac i i y was posi i ely associa ed wi h global ac ional aniso opy (β¼0.057, 95% CI ¼0.016, 0.098, p ¼0.007) and nega i ely associa ed wi h global mean di usi i y (β¼0.079, 95% CI ¼0.120, 0.038, p <0.001), wo commonly de i ed scala measu es o whi e ma e mic os uc u e. Two componen s o o al physical ac i i y, ou doo play and spo pa icipa ion, we e posi i ely associa ed wi h global ac ional aniso opy (β¼0.041, 95% CI¼(0.000, 0.083), p ¼0.047; β¼0.053, 95% CI¼(0.010, 0.096), p¼0.015, espec i ely) and in e sely associa ed wi h global mean di usi i y (β¼0.074, 95% CI¼(0.114, 0.033), p <0.001; β¼0.043, 95% CI¼(-0.086, 0.000), p ¼0.049, espec i ely). No associa ions we e obse ed be ween sc een ime and whi e ma e mic os uc u e (p >0.05). This s udy p o ides new e idence ha physical ac i i y is modes ly associa ed wi h whi e ma e mic os uc u e in child en. In con as , complemen ing o he ecen e idence on cogni ion, sc een ime was no associa ed wi h whi e ma e mic os uc u e. Causal in e ences om hese modes associa ions mus be in e p e ed cau iously in he absence o longi udinal da a. Howe e , hese da a s ill o e a p omising a enue o u u e wo k o explo e o wha ex en physical ac i i y may p omo e heal hy whi e ma e de elopmen . Abb e ia ions: MRI, magne ic esonance imaging; METs, me abolic equi alen s; DTI, di usion enso imaging; FA, ac ional aniso opy; MD, mean di usi i y; RCT, andomized con ol ial; EPI, di ec ion echo plana imaging; AD, axial di usi i y; RD, adial di usi i y; BMI, body mass index; IQ, Non- e bal in elligence quo ien ; SON-R 2.5–7, Snijde s-Oomen Nie - e bale in elligen ie Tes - Re isie; CBCL/6–18, Child Beha io Checklis school-age e sion; FDR, alse disco e y a e; ZonMw, Ne he lands O ganiza ion o Heal h Resea ch and De elopmen . * Co esponding au ho . Depa men o Social and Beha io al Science, Ha a d T.H. Chan School o Public Heal h Bos on, Massachuse s, Uni ed S a es. E-mail add ess: [email p o ec ed]a a d.edu (H. Tiemeie ). Con en s lis s a ailable a ScienceDi ec Neu oImage jou nal homepage: www.else ie .com/loca e/neu oimage h ps://doi.o g/10.1016/j.neu oimage.2019.116258 Recei ed 4 Feb ua y 2019; Recei ed in e ised o m 1 Oc obe 2019; Accep ed 7 Oc obe 2019 A ailable online 9 Oc obe 2019 1053-8119/©2019 Else ie Inc. This is an open access a icle unde he CC BY-NC-ND license (h p://c ea i ecommons.o g/licenses/by-nc-nd/4.0/). Neu oImage 205 (2020) 116258 1. In oduc ion Despi e compelling e idence sugges ing physical ac i i y plays an impo an ole in gene al heal h (Lee e al., 2012), a subs an ial p o- po ion o you h do no mee he ecommended le els o physical ac i i y (Hallal e al., 2012). Du ing childhood and adolescence, he b ain un- de goes a a ie y o de elopmen al changes in whi e and g ay ma e s uc u e (Len oo and Giedd, 2006;Schmi ho s and Yuan, 2010), which also coincide wi h changes in beha io and cogni ion. Du ing his sen- si i e pe iod o ma u a ion, a numbe o en i onmen al ac o s ha e been shown o be ela ed o he b ain’s whi e ma e mac os uc u e and mic os uc u e (F odl e al., 2012;Hanson e al., 2013;Bick and Nelson, 2016), ye li le wo k has explo ed o wha ex en physical ac i i y and seden a y beha io , associa e wi h whi e ma e in child en. Physical ac i i y and seden a y beha io a e he wo independen beha io s ha occupy all waking hou s o a day. Physical ac i i y, defined as any body mo emen ha inc eases ene gy expendi u e (e.g. ac i e commu ing, ou doo play, o spo pa icipa ion) (Caspe sen e al., 1985), has been consis en ly linked o men al heal h benefi s ac oss a pe son’s li espan (Biddle e al., 2018;Chek oud e al., 2018). On he o he hand, seden a y beha io , defined as any waking beha io cha - ac e ized by an ene gy expendi u e 1.5 me abolic equi alen s (METs), while in a si ing, eclining o lying pos u e (T emblay e al., 2017), has become a cen al componen o he daily li es o child en. The e o e, he deba e abou whe he seden a y beha io , specifically ec ea ional sc een ime, nega i ely impac s child en men al heal h is becoming inc easingly impo an . Recen wo k has sugges ed ha while nega i e associa ions a e obse ed be ween sc een ime (i.e., echnology use and playing ideogames) and men al heal h and cogni ion in young people, hey a e likely oo small in magni ude o wa an policy change (Fe gu- son, 2017;O ben and P zybylski, 2019). Though p e ious wo k has s udied he associa ion o hese wo beha io s wi h indica o s o men al heal h in childhood, li le wo k has examined any neu al co ela es. Magne ic esonance imaging (MRI) o e s an in i o iew in o he de eloping b ain. Whi e ma e neu onal issue consis s o axons w ap- ped in he lipid- ich myelin shea h and is esponsible o p o iding as and e ficien connec ions h oughou co ex and subco ex. Di usion enso imaging (DTI) is able o sample ea u es o he mic os uc u al a chi ec u e o whi e ma e . F ac ional aniso opy (FA) and mean di usi i y (MD) a e wo commonly de i ed scala me ics om DTI. FA ep esen s he deg ee o which wa e di uses p e e en ially along one axis (i.e., he di usion is hinde ed by s uc u es such as myelin and igh ly packed axons), and has shown o inc ease wi h age du ing child and adolescen b ain de elopmen (Gio gio e al., 2010;Schmi ho s and Yuan, 2010;Tamnes e al., 2018) and in many cases o be lowe in he con ex o a ious neu ological and psychia ic diseases (Hul e sho n e al., 2011;Ayling e al., 2012;Aoki e al., 2018). MD is he simple a e age di usion, wi h highe le els indica ing ela i ely unimpeded di usion (i.e., i is nega i ely co ela ed wi h FA). Nume ous s udies ha e demons a ed ha de elopmen al changes in whi e ma e mic o- s uc u e occu h oughou childhood (Schmi ho s and Yuan, 2010; S iles and Je nigan, 2010), and he echnique has been shown o be sensi i e o sub le ea u es o psychopa hology (Mue zel e al., 2017). Howe e , li le wo k has explo ed whe he physical ac i i y o sc een ime a e associa ed wi h whi e ma e mic os uc u e. In one andomized con ol ial (RCT) o o e weigh 8–11 yea -old child en, he e ec o an 8-mon h exe cise in e en ion on whi e ma - e mic os uc u e was examined (K a e al., 2014;Schae e e al., 2014). This s udy showed ha FA inc eased in he uncina e and supe io longi udinal asciculus in he in e en ion g oup. Simila ly, ano he RCT examined he e ec s o an a e -school exe cise p og am on he mic o- s uc u e o whi e ma e ac s in 7- o 9-yea -old child en. This s udy ound ha child en who pa icipa ed in he exe cise p og am showed inc eased whi e ma e mic os uc u e in he genu o he co pus callosum (Chaddock-Heyman e al., 2018). In addi ion, a c oss-sec ional s udy demons a ed ha , in 9- o-10 yea -old child en, highe le els o ca dio- espi a o y fi ness we e associa ed wi h g ea e FA in sec ions o he co pus callosum, co ona adia a, and supe io longi udinal asciculus (Chaddock-Heyman e al., 2014). Howe e , g ea e ca dio espi a o y fi ness was also associa ed wi h lowe FA in he co icospinal ac (He ing e al., 2014). Collec i ely, i is di ficul o d aw conclusions om his limi ed li e a u e, which la gely ocuses on he ela ion be ween ca dio espi a o y fi ness, exe cise and whi e ma e mic os uc u e, omi ing he cons uc s o physical ac i i y (e.g., ou doo play, ac i e commu ing, and spo pa icipa ion). Fu he , o ou knowledge, no o he s udies ha e examined whe he sc een ime is associa ed wi h whi e ma e mic os uc u e in child en. The p esen s udy examined he associa ions o physical ac i i y and sc een ime wi h whi e ma e mic os uc u e in 10-yea old child en om he gene al popula ion. Based on p e ious li e a u e, we hypo he- sized highe le els o physical ac i i y would be ela ed o highe FA and lowe MD. We hypo hesized inc eased sc een ime would be ela ed o lowe FA and highe MD, howe e , in line wi h p e ious li e a u e (Fe guson, 2015,2017;O ben and P zybylski, 2019), we expec hese ela ionships o be ela i ely small. 2. Ma e ials and me hods 2.1. S udy design and pa icipan s The s udy was pa o he Gene a ion R S udy, a p ena al popula ion- based coho . A de ailed desc ip ion o he s udy design and me hods has been published elsewhe e (Kooijman e al., 2016). In o al, 3992 child en isi ed ou s udy-dedica ed imaging acili y o an MRI scan a he mean age o 10 (Whi e e al., 2018). O he 3992 child en, a o al o 942 we e excluded due o missing a comple e DTI scan (n ¼285), image a i ac s (n ¼37), poo image quali y (n ¼406), an inciden al finding (n ¼13) o hei da a we e collec ed using di e en pa ame e s on he MRI sys em (n ¼201), lea ing 3050 pa icipan s o analysis. A e excluding chil- d en wi h missing in o ma ion on physical ac i i y (n ¼518), he s udy sample included 2532 child en (mean age 10.12 0.58 yea s; 50.04% boys). Supplemen al Table S1 shows he sample cha ac e is ics o chil- d en included in analyses o sc een ime (N ¼2346), which is highly simila o hose included in analyses o physical ac i i y. A flow cha illus a ing he exclusion o da a is depic ed in Fig. 1. The Medical E hics Commi ee o he E asmus Medical Cen e app o ed he s udy p ocedu es and pa icipan s p o ided w i en in o med consen . 2.2. Physical ac i i y and sc een ime assessmen s In o ma ion on he le el o physical ac i i y and sc een ime we e ob ained h ough a pa en - epo ed ques ionnai e adminis e ed when child en we e 10 yea s old. The ques ionnai es we e in ended o he p ima y ca egi e who was mos o en he mo he (97%). To assess he le el o physical ac i i y, esponden s indica ed bo h he numbe o days pe week and du a ion pe day hei child engages in: (i) walking o cycling o/ om school, (ii) ou doo play, and (iii) spo s. Time pe week spen on each ac i i y was calcula ed by using he ollowing o mula: weekly ime spen on he ac i i y ¼(days pe week) * (hou s pe day). A o al physical ac i i y sco e was calcula ed by adding he hou s o ac i e commu ing, ou doo play, and spo pa icipa ion pe week. Responden s we e also asked o indica e he numbe o days and hou s pe day hei child: (i) wa ches ele ision (including ideos/DVDs) and (ii) uses a compu e o simila de ice (including ideo games). Sc een imes we e assessed sepa a ely o weekdays and weekend days bu we e combined o es ima e he o al hou s pe week spen in each ac i i y. A o al weekly sc een ime sco e was calcula ed by adding he hou s o playing ideo games and wa ching ele ision. M. Rod iguez-Ayllon e al. Neu oImage 205 (2020) 116258 2 2.3. MRI 2.3.1. Image acquisi ion MRI da a we e acqui ed wi h a 3 Tesla GE MR-750W sys em (Gene al Elec ic, Milwaukee, WI). The DTI sequence consis ed o a 35 di ec ion echo plana imaging (EPI) sequence using he ollowing sequence pa- ame e s: TR ¼12,500 m s, TE ¼72 m s, flip angle ¼90, ma- ix ¼120 120, FOV ¼240 mm 240 mm, slice hickness ¼2 mm, numbe o slices ¼65 and oxel esolu ion ¼222mm 3 (Mue zel e al., 2017). 2.3.2. Image p ep ocessing Image p ep ocessing was conduc ed using he Func ional MRI o he B ain’s So wa e Lib a y (Jenkinson e al., 2012) and he CAMINO oolki (Cook e al., 2006) h ough py hon in e aces (h ps:// sl. m ib.ox.ac.uk) (Go golewski e al., 2011). Images we e fi s adjus ed o mino head mo ion and eddy-cu en induced a i ac s (Haselg o e and Moo e, 1996; Jenkinson and Smi h, 2001). In o de o accoun o o a ions applied o he image da a (Leemans and Jones, 2009;Jones and Ce cignani, 2010), he esul ing ans o ma ion ma ices we e used o o a e he di usion g adien di ec ion able. Non-b ain was emo ed using he FSL B ain Ex ac ion Tool (Smi h, 2002). The di usion enso was fi using he RESTORE me hod (Chang e al., 2005;Cook e al., 2006), and common scala maps (i.e., FA, MD, Axial di usi i y (AD), adial di usi i y (RD)) we e subsequen ly compu ed. Fully au oma ed p obabilis ic ac og- aphy was un using a p edefined se ies o seed- and a ge -masks om he FSL Au oP x plugin (De G oo e al., 2015;Mue zel e al., 2017). B iefly, ollowing a nonlinea egis a ion o he FA map o s anda d space (FMRIB58 1 mm), a p edefined lib a y o seed, a ge , and exclu- sion masks we e wa ped o each indi idual’s na i e space FA map. FSL’s Bayesian Es ima ion o Di usion Pa ame e s using Sampling Techniques (BEDPOSTx) and P ob ackx we e used o conduc p obabilis ic fibe ac og aphy, accoun ing o wo fibe o ien a ions (Beh ens e al., 2003). Resul ing connec i i y dis ibu ions we e no malized based on he o al numbe o success ul seed- o- a ge a emp s, and we e h esh- olded based on es ablished alues o emo e oxels wi h a low p oba- bili y (Mue zel e al., 2017). A e age DTI scala me ics we e ex ac ed o each ac (see Supplemen al Ma e ial), wi h each oxel being weigh ed by he no malized connec i i y alue om p obabilis ic ac- og aphy. Global, whole b ain es ima es o DTI me ics (i.e., FA, MD, RD, AD) we e calcula ed using a confi ma o y ac o analysis ac oss mul iple ac s desc ibed in de ail elsewhe e (Mue zel e al., 2017). 2.3.3. Image quali y assu ance Raw image quali y was assessed wi h bo h an au oma ed so wa e and wi h a isual inspec ion (Mue zel e al., 2017). B iefly, slice-wise a ia- ions in signal in ensi y we e examined using he DTIp ep package (h ps://www.ni c.o g/p ojec s/d ip ep). The sum-o -squa es e o (SSE) Fig. 1. Flow cha indica ion da a inclusion/exclusion. No e: Box wi h do ed lines indica e he sample used in he cu en s udys. QC ¼quali y con ol, IF ¼inciden al o finding, PA ¼physical ac i i y, ST ¼sc een ime, MRI ¼magne ic esonance imaging, DTI ¼di usion enso imaging. M. Rod iguez-Ayllon e al. Neu oImage 205 (2020) 116258 3 maps om he enso es ima ion we e also calcula ed and isually inspec ed o s uc u ed noise. Las ly, nonlinea egis a ion o s anda d space was inspec ed o accu acy o ensu e seed and a ge masks we e p ope ly aligned o na i e space. Da ase s de e mined o be o insu ficien quali y o s a is ical analyses we e excluded (n ¼285 om au oma ed quali y con ol o slice-wise a ia ion in signal and n ¼121 om an addi ional isual inspec ion o SSE maps and egis a ion quali y). Example images o FA maps om child en ep esen a i e o he sample a e p esen ed in he Supplemen a y Ma e ial (see Fig. S1). 2.4. Co a ia es Age a MRI scanning and sex we e included as co a ia es. Fu he , ma e nal educa ion le el and e hnici y we e assessed by ques ionnai es. E hnici y was based on he coun y o bi h o he mo he and mo he ’s pa en s. Ma e nal educa ion le el was defined by he highes comple ed educa ion and di ided in o 3 ca ego ies anging om low ( om no ed- uca ion o p ima y school), middle (high school o oca ional aining) o high educa ion le el ( om highe oca ional educa ion o uni e si y). Heigh and weigh we e measu ed a he age o 10 yea s a he esea ch cen e and body mass index (BMI) was calcula ed using s anda dized sco es acco ding o he Du ch e e ence g ow h cu es (h ps://g ow ha nalyse .o g)(F ed iks e al., 2000). A non- e bal in elligence quo ien (IQ) was assessed a app oxima ely 6 yea s o age using he Snijde s-Oomen Nie - e bale in elligen ie Tes - Re isie (SON-R 2.5–7) (Tellegen e al., 2005). Emo ional and beha io al p oblems we e assessed a age o 10 yea s by pa en - epo using he alida ed Child Beha io Checklis school-age e sion (CBCL/6–18) (Achenbach and Resco la, 2003). Apa om non- e bal IQ (12%), he pe cen age o missing alues did no exceed 10%. 2.5. S a is ical analysis All analyses we e pe o med using he S a is ical Package o Social Sciences (IBM SPSS S a is ics o Windows, e sion 22.0, A monk, NY, p se a <0.05). Mul iple impu a ion was used o impu e missing da a o co a ia es. In o al, 10 impu ed da ase s we e c ea ed and pooled esul s a e p esen ed. Mul iple linea eg ession analyses we e pe o med wi h physical ac i i y o sc een ime measu es as independen a iables and DTI scala me ics (i.e., FA, and MD) as dependen a iables. Indepen- den ly, as hey p o ide added in o ma ion in he in e p e a ion o he unde lying whi e ma e mic os uc u e, we also es ed he associa ion o physical ac i i y and sc een ime a iables wi h global RD and global AD. Linea eg ession analyses we e hie a chically adjus ed o co a ia es by c ea ing wo models. Model 1 included age a he ime o scanning, sex, e hnici y, ma e nal educa ion, and BMI. Model 2 was addi ionally adjus ed o beha io p oblems and non- e bal IQ. In o de o de e mine whe he he associa ion wi h whi e ma e mic os uc u e was indeed only global o es ic ed o a pa icula se o whi e ma e bundles, as- socia ions wi h FA and MD wi hin indi idual ac s we e also es ed i exposu es showed an associa ion wi h global DTI me ics. Fo ac - specific analyses, alse disco e y a e (FDR) co ec ion was applied o accoun o he numbe o es s pe o med. Addi ionally, a numbe o sensi i i y analyses we e un. Fi s , o al physical ac i i y and o al sc een ime we e en e ed in o he model simul aneously o adjus es ima es o one ano he . Second, we di ided he o al sample in ou subg oups o pa icipan s acco ding o hei physical ac i i y and sc een ime le els o compa e global FA and MD amongs he g oups. Physical ac i i y and sc een ime g oups we e defined acco ding o he sample median (high/low physical ac i i y/ sc een ime). Thi d, di e en in e ac ion e ms we e es ed o asce ain whe he associa ions we e di e en ac oss sub-g oups o subjec s (i.e., sex, pa en al educa ional le el, ances al backg ound). Las ly, we es ed quad a ic and cubic age e ms, as well as an age-by-sex in e ac ion e m (using a quad a ic age e m). We also es ed quad a ic sc een ime and physical ac i i y e m, as well as he in e ac ions be ween quad a ic o al physical ac i i y a iable and quad a ic o al sc een ime a iable in as- socia ion wi h whi e ma e mic os uc u e. 3. Resul s Table 1 p esen s pa icipan demog aphic cha ac e is ics. Di e ences be ween physical ac i i y and sc een ime le els be ween boys and gi ls a e shown in Table S2. Cha ac e is ics o included and excluded pa ic- ipan s a e shown in Table S3. 3.1. Associa ion o physical ac i i y and sc een ime wi h global whi e ma e mic os uc u e To al physical ac i i y was posi i ely associa ed wi h global FA (β¼0.051, 95% CI¼(0.010, 0.092), p ¼0.016) when he model was adjus ed o sociodemog aphic ac o s. Fu he , he posi i e associa ion be ween physical and global FA emained a e addi ional adjus men o beha io p oblems and non- e bal IQ (β¼0.057, 95% CI¼(0.016, 0.098), p ¼0.007). Resul s om analyses o global FA and MD a e p e- sen ed in Table 2, and ollow-up analyses wi h global RD and AD o be e desc ibe he di usion p ofile a e p esen ed in Table S4. To de e mine whe he a pa icula ype o physical ac i i y (i.e. ac i e commu ing, ou doo play and spo pa icipa ion) was esponsible o he o al PA associa ion wi h global FA, each ype o physical ac i i y was examined sepa a ely. In he ully adjus ed model (model 2), posi i e associa ions be ween ou doo play and global FA (β¼0.041, 95% CI¼(0.000, 0.083), p¼0.047) and be ween spo pa icipa ion and global FA we e obse ed (β¼0.053, 95% CI¼(0.010, 0.096), p ¼0.015). To al physical ac i i y was in e sely associa ed wi h global MD (β¼0.079, 95% CI¼(0.119, 0.038), p <0.001) when he model was adjus ed o sociodemog aphic co a ia es (model 1). A e , addi- ional adjus men o beha io al p oblems and non- e bal IQ he asso- cia ion emained simila (model 2) (β¼0.079, 95% CI¼(0.120, 0.038), p <0.001). Each ype o physical ac i i y was examined sepa- a ely o de e mine whe he all o a subse o ac i i ies we e esponsible o he associa ions wi h o al physical ac i i y. In he ully adjus ed Table 1 Sample cha ac e is ics. N Mean/% SD Sex Boys,% 1267 50.04 Age a MRI assessmen ,yea s 2532 10.12 0.58 Age a CBCL assessmen , yea s 2495 9.71 0.28 E hnici y, % Du ch 1671 66.63 O he Wes e n 213 8.49 Non-Wes e n 624 24.88 BMI, kg/m 2 2526 17.36 2.50 Beha io p oblems, sum sco e CBCL 2475 16.76 14.63 Non- e bal IQ, sum sco e SON-R 2218 104.32 14.59 Ma e nal educa ion,% Low 187 8.06 Medium 618 26.65 High 1514 65.29 Exposu es cha ac e is ics To al physical ac i i y, h/w 2532 9.06 5.10 To al sc een ime, h/w 2346 17.10 11.63 *No e: 86 pa icipan s had no da a on physical ac i i y bu hey we e included in analyses o sc een ime. MRI ¼magne ic esonance imaging, CBCL¼Child Beha io Checklis school-age, BMI¼Body mass index, IQ ¼in elligence quo- ien , SON-R¼ he Snijde s-Oomen Nie - e bale in elligen ie Tes - Re isie, h/w ¼hou s pe week. To al physical ac i i y ¼hou s o ac i e commu ing, ou doo play, and spo pa icipa ion pe week. To al sc een ime ¼hou s o compu e ime, playing ideo games, and wa ching ele ision. Ma e nal educa ion le el was defined by he highes comple ed educa ion and di ided in o h ee ca e- go ies anging om low ( om no educa ion o p ima y school), middle (high school o oca ional aining) o high educa ion le el ( om highe oca ional educa ion o uni e si y). M. Rod iguez-Ayllon e al. Neu oImage 205 (2020) 116258 4 model (model 2), ou doo play and spo pa icipa ion we e in e sely associa ed wi h MD (β¼0.074, 95% CI¼(0.114, 0.033), p <0.001; β¼0.043, 95% CI¼(0.086, 0.000), p ¼0.049, espec i ely). In o de o be e desc ibe hese associa ions wi h FA and MD, u he analyses explo ed which unde lying di usi i y me ics (i.e., RD and/o AD) we e associa ed wi h physical ac i i y. Bo h RD and AD we e significan ly associa ed wi h o al physical ac i i y (β¼0.079, 95% CI¼(0.120, 0.039), p <0.001; β¼0.054, 95% CI¼(0.094, 0.014), p ¼0.008, espec i ely). No associa ion was ound be ween indi idual o agg ega e sc een ime a iables and global DTI me ics (all p >0.05) (Table 2). 3.2. Associa ion be ween physical ac i i y, sc een ime and ac -specific FA and MD Associa ions be ween o al physical ac i i y and ac -specific FA and MD a e shown in Table S5. B iefly, no associa ion was ound be ween o al physical ac i i y and FA wi hin indi idual ac s when significance le els we e adjus ed o mul iple es ing. Howe e , o al physical ac i i y was associa ed wi h MD in nea ly all ac s when adjus ed o mul iple compa isons. Fig. 2 g aphically ep esen s he significan nega i e asso- cia ions be ween o al physical ac i i y and MD ac oss indi idual ac s. Las ly, explo a o y analyses es ing he associa ion be ween o al sc een ime and ac -specific FA and MD a e shown in Table S6. 3.3. Sensi i i y analyses Resul s emained simila when models o physical ac i i y and global whi e ma e mic os uc u e we e addi ionally adjus ed o o al sc een ime, sugges ing ha physical ac i i y is posi i ely associa ed wi h whi e ma e mic os uc u e, independen o o al sc een ime (Table S7). Sc een ime was s ill un ela ed o global DTI me ics a e addi ional adjus men o o al physical ac i i y (Table S7). Nex , when ca ego ically subdi iding he sample based on physical ac i i y and sc een ime le els (e.g., high sc een ime, low physical ac- i i y), he subg oup wi h high physical ac i i y (>8.4 h/week) and high sc een ime (>14.8 h/week seem) showed lowe MD compa ed o he es o subg oups (Fig. S2 and Fig. S3, espec i ely). Then, di e en in e ac ion e ms we e es ed o asce ain whe he associa ions we e di e en ac oss sub-g oups o subjec s. The associa ion be ween physical ac i i y and whi e ma e mic os uc u e was consis- en ac oss pa en s’educa ion ca ego ies. Simila ly, no in e ac ion be- ween sex o e hnici y and physical ac i i y was obse ed. Las ly, we ha e included in o he model nonlinea age e ms, spe- cifically a quad a ic age e m (Table S8) and also included a cubic age e m and esul s emain highly simila (Table S9). We also es ed age-by- sex in e ac ion e ec s wi h hese a iables (i.e., age 2 and age 3 ) and he in e ac ion e m was no significan (all P >0.470). Impo an ly, he model es ima es o he associa ion wi h physical ac i i y and sc een ime also emained highly simila . Finally, nonlinea sc een ime and physical ac i i y e ms we e no associa ed wi h FA o MD (See Supplemen al Ma e ial). 4. Discussion Using neu oimaging da a om o e 2500 child en, we ound ha physical ac i i y is associa ed wi h whi e ma e mic os uc u e, speci - ically ou doo play and spo pa icipa ion ime. On he o he hand, no associa ion was obse ed be ween sc een ime and whi e ma e mic o- s uc u e. In he con ex o he li e a u e, his s udy expands upon he compelling e idence o gene al heal h benefi s o physical ac i i y by demons a ing a new associa ion wi h whi e ma e mic os uc u e in child en. P e ious wo k on ypical b ain de elopmen has shown a posi i e associa ion be ween FA and age, as well as a nega i e associa ion wi h MD (Ba nea-Go aly e al., 2005;Elu a hingal e al., 2007;Asa o e al., 2010;Gio gio e al., 2010;Schmi ho s and Yuan, 2010;B ouwe e al., 2012;Tamnes e al., 2018). Fu he , a simila pa e n o associa ion has been shown wi h cogni i e unc ion, whe e whi e ma e mic os uc u e is associa ed wi h be e neu opsychological pe o mance (Mue zel e al., 2015). Wi hin his con ex , he posi i e link be ween physical ac i i y and whi e ma e mic os uc u e shown in he p esen s udy could explain he imp o emen s obse ed in cogni ion (Hillman e al., 2008, 2015). Impo an ly, specific aspec s o physical ac i i y, namely ou doo Table 2 Associa ion be ween physical ac i i y and sc een ime wi h global ac ional aniso opy and mean di usi i y. F ac ional aniso opy Model 1 Model 2 β95% CI p βCI p Physical ac i i y (n ¼2532) To al physical ac i i y, h/w 0.051 (0.010, 0.092) 0.016 0.057 (0.016, 0.098) 0.007 Ac i e commu ing, h/w 0.021 (-0.020, 0.062) 0.325 0.023 (-0.018, 0.064) 0.272 Ou doo play, h/w 0.035 (-0.006, 0.076) 0.093 0.041 (0.000, 0.083) 0.047 Spo pa icipa ion, h/w 0.052 (0.009, 0.094) 0.017 0.053 (0.010, 0.096) 0.015 Sc een ime (n ¼2346) To al sc een ime, h/w 0.021 (-0.065, 0.023) 0.347 0.010 (-0.054, 0.035) 0.668 Playing compu e games, h/w 0.003 (-0.045, 0.039) 0.880 0.000 (-0.041, 0.042) 0.985 Wa ching ele ision,h/w 0.028 (-0.073, 0.017) 0.219 0.014 (-0.060, 0.031) 0.532 Mean di usi i y Model 1 Model 2 β95% CI p βCI p Physical ac i i y (n ¼2532) To al physical ac i i y, h/w 0.079 (-0.119, 0.038) <0.001 0.079 (-0.120, 0.038) <0.001 Ac i e commu ing, h/w 0.005 (-0.045, 0.036) 0.821 0.005 (-0.046, 0.036) 0.812 Ou doo play, h/w 0.073 (-0.114, 0.032) <0.001 0.074 (-0.114, 0.033) <0.001 Spo pa icipa ion,h/w 0.043 (-0.086, 0.001) 0.045 0.043 (-0.086, 0.000) 0.049 Sc een ime (n ¼2346) To al sc een ime, h/w 0.028 (-0.072, 0.015) 0.203 0.031 (-0.075, 0.013) 0.170 Playing compu e games, h/w 0.024 (-0.066, 0.017) 0.250 0.026 (-0.067, 0.016) 0.230 Wa ching ele ision, h/w 0.022 (-0.067, 0.022) 0.325 0.025 (-0.070, 0.020) 0.280 *No e: Model 1 was adjus ed o sex, age a he ime o scanning, ances al backg ound, body mass index and ma e nal educa ion. Model 2 was addi ionally adjus ed o emo ional and beha io p oblems and non- e bal in elligence quo ien . h/w ¼hou s pe week. M. Rod iguez-Ayllon e al. Neu oImage 205 (2020) 116258 5 play and spo pa icipa ion, we e associa ed wi h whi e ma e mic o- s uc u e. The associa ion be ween physical ac i i y and whi e ma e mic os uc u e has no been s udied p e iously, which hampe s u he compa isons wi h simila s udies. None heless, his wo k complimen s p e ious s udies ocused on he associa ion be ween ca dio espi a o y fi ness and exe cise wi h whi e ma e mic os uc u e. Fo ins ance, p e ious s udies demons a ed ha , in 9- o-10 yea -old child en, highe le els o ca dio espi a o y fi ness we e associa ed wi h highe FA in sec ions o he co pus callosum, co ona adia a, and supe io longi udinal asciculus (Chaddock-Heyman e al., 2014). In con as , adolescen s wi h be e ca dio espi a o y fi ness showed g ea e numbe o s eamlines, especially in he co icospinal ac and o ceps mino (He ing e al., 2014), bu also lowe FA in he co icospinal ac (He ing e al., 2014). Addi ionally, an 8-mon h exe cise in e en ion showed inc eased FA in sec ions o he uncina e asciculus and supe io longi udinal asciculus in o e weigh child en in he exe cise g oup when compa ed o con ols (K a e al., 2014;Schae e e al., 2014). Mo eo e , an 9-mon h exe - cise in e en ion showed inc eased whi e ma e mic os uc u e in he genu o he co pus callosum in 7- o 9-yea -old child en in he exe cise g oup when compa ed o con ols (Chaddock-Heyman e al., 2018). In he p esen s udy, no associa ion was ound be ween ac i e commu ing and whi e ma e mic os uc u e. P e ious wo k has also shown ha ac i e commu ing was no associa ed wi h cogni i e pe o - mance in child en (Van Dijk e al., 2014). Thus, esul s a e in line wi h an inc easing li e a u e sugges ing ha inc eases in cogni i e unc ioning in childhood due o physical ac i i y a e mos clea ly obse ed in asks ha in ol e execu i e unc ioning (e.g., spo pa icipa ion) (Tompo owski e al., 2008). P e ious wo k has demons a ed ha inc eased le els o physical ac i i y migh imp o e ce eb al blood flow (Hillman e al., 2008), and DTI has p o en o be sensi i e o whi e ma e mic os uc u al changes ha may esul om ca dio- and neu o- ascula isk ac o s (Lee e al., 2009;She e e al., 2013). A possible physiological mechanism could be he posi i e influence physical ac i i y has on ca dio-me abolic isk ac o s, such as insulin esis ance, blood lipids, blood p essu e and in- flamma o y p o eins (Eisenmann, 2004). Mo eo e , p e ious li e a u e showed highe le els physical ac i i y a e associa ed wi h be e men al heal h in child en (Biddle and Asa e, 2011;Biddle e al., 2018). As childhood is a pe iod whe e many physiological and psychological changes occu , public heal h ad oca ion o no el app oaches o inc easing physical ac i i y in child en could be use ul o imp o ed (neu o) physiological and psychological heal h. In his line, a ecen la ge Fig. 2. Associa ion be ween o al physical ac i i y and ac -specific MD. Colou ba ep esen s s anda dized be a coe ficien s wi h ligh e blue e e ing o nega i e associa ion. Ligh e blue e e ing o s onge associ- a ion be ween o al physical ac i i y and MD. Co i- cospinal ac (CST), acous ic adia ion (AR), an e io halamic adia ion (ATR), supe io longi udinal asciculus (SLF), in e io longi udinal asciculus (ILF), in e io on o-occipi al asciculus (IFOF), uncina e asciculus (UNC), cingula e gy us pa o cingulum (CGC), hippocampal pa o he cingulum (CGH), o - ceps mino (FMI), and supe io halamic adia ions (STR). M. Rod iguez-Ayllon e al. Neu oImage 205 (2020) 116258 6 s udy o physical ac i i y and cogni ion showed highe le els o physical ac i i y we e associa ed wi h be e global cogni ion (Walsh e al., 2018). Ou esul s sugges ha hese imp o emen s in cogni ion could be acili a ed by be e unde lying whi e ma e mic os uc u e. Ano he impo an finding was ha sc een ime was no associa ed wi h whi e ma e mic os uc u e. To he bes o ou knowledge, his is he fi s s udy o examine he associa ion be ween sc een ime and whi e ma e mic os uc u e in child en. These esul s u he suppo he wo k om Fe guson e al. and O ben e al. (Fe guson, 2017;O ben and P zy- bylski, 2019) who showed ha he ela ionship be ween sc een ime and di e en men al heal h ou comes (e.g., ex e nalizing p oblems, in e - nalizing p oblems and academic pe o mance) was null o e y small. O e all, he associa ion be ween sc een ime and men al heal h a e likely nonlinea , and mode a e engagemen in ce ain sc een ime ac i i ies may no lead o beha io al o emo ional p oblems (Fe guson, 2017). Fo ins ance, p e ious esea ch ound ha mee ing he 2 h o less ec ea- ional sc een ime pe day ecommenda ions was associa ed wi h supe- io global cogni ion (Walsh e al., 2018). In con as , equen weekly use o ideo games was associa ed wi h conduc p oblems, as well as, inc eased le els o ele ision iewing we e nega i ely associa ed wi h neu ocogni i e de elopmen o child en (Takeuchi e al., 2015). In addi ion o hese disc epancies in he li e a u e, ad ances in echnology (e.g., a ailabili y o hand-held media de ices, and social media) compa ed o he cons uc o pu ely passi e ele ision iewing, u he unde sco e he impo ance o main aining well-cha ac e ized measu es o sc een ime, as di e ing amoun s/ ypes o sc een ime could influence he b ain di e en ly. In e es ingly, associa ions be ween physical ac i i y and whi e ma e emained a e adjus ing o le els o sc een ime. Thus, hough pu ely specula i e, he benefi s o physical ac i i y may no be nega i ely influenced by playing compu e games and wa ching ele i- sion mode a ely, which has been epo ed p e iously o o he gene al ca diome abolic ou comes (Rezende e al., 2014). Key s eng hs o he cu en s udy a e he la ge sample size, and inco po a ion o popula ion-based neu oimaging. The na u e and size o he coho alloca es us wi h he da a and powe o adjus o mul iple con ounding ac o s. Fu he , he popula ion-based sampling o e s inc eased gene alizabili y o he findings compa ed o p e ious s udies. Howe e , se e al limi a ions equi e discussion. Fi s , his is a c oss- sec ional s udy, limi ing in e ences abou causali y and di ec ionali y o any o he associa ed ac o s. Fu he , hough esidual con ounding e- mains a possibili y, e ec es ima es we e no subs an ially changed a e adjus ing o a numbe o po en ial con ounding ac o s. Fu u e wo k should conside o he en i onmen al con ounde s including well- cha ac e ized measu es o socioeconomic s a us. Second, physical ac i - i y and sc een ime le els we e assessed by pa en al- epo ed ques ion- nai es, lending o he possibili y o unde - o o e es ima ions o he beha io s. The e o e, objec i e measu es such as accele ome y and expe ience sampling should be u ilized in u u e s udies. Las ly, he e ec sizes o he associa ion be ween physical ac i i y and whi e ma e mic os uc u e a e ela i ely small. Howe e , simila e ec sizes a e obse ed in s uc u e- unc ion associa ions wi h cogni ion and whi e ma e mic os uc u e (Mue zel e al., 2015), and he clinical ele ance o such e ec sizes in he con ex o DTI has ye o be o mally e alua ed. 5. Conclusion This s udy demons a es ha highe le els o physical ac i i y a e associa ed wi h g ea e whi e ma e mic os uc u e in child en ages 9- o-10 yea s old om he gene al popula ion. No associa ion was obse ed be ween sc een ime and whi e ma e mic os uc u e. Fu u e wo k should con inue o explo e longi udinal da a in o de o mo e conc e ely deciphe he empo ali y o he associa ions, so ha educa- ional and heal h ins i u ions can conside whe he o p omo e physical ac i i y du ing childhood as a po en ial (modifiable) p o ec i e ac o in he con ex o b ain heal h. Funding This s udy was suppo ed by he Sophia Founda ion (S18-20 awa ded o RM) and Ne he lands O ganiza ion o Heal h Resea ch and De el- opmen (ZonMw) (p ojec 016.VICI.170.200, awa ded o HT and TOP 91211021, awa ded o TW). Supe compu ing esou ces we e suppo ed by he Ne he lands O ganiza ion o Scien ific Resea ch (Exac e We en- schappen) and SURFsa a (Ca esius Compu e Clus e , www.su sa a.nl). The Gene a ion R S udy is conduc ed by he E asmus Medical Cen e in close collabo a ion wi h he School o Law and he Facul y o Social Sciences o he E asmus Uni e si y Ro e dam, he Municipal Heal h Se ice Ro e dam a ea, and he S ich ing T ombosediens en A senla- bo a o ium Rijnmond. The gene al design o he Gene a ion R S udy was made possible by financial suppo om he E asmus Medical Cen e , Ro e dam, ZonMw, he Ne he lands O ganiza ion o Scien ific Resea ch, and he Minis y o Heal h, Wel a e, and Spo . The au ho s g a e ully acknowledge he con ibu ions o he pa icipa ing child en and pa en s, gene al p ac i ione s, hospi als, midwi es, and pha macies in Ro e dam. MRA ecei ed a schola ship om he Uni e si y o G an- ada om he PhD In e na ional Mobili y P og am o a b ie s ay in he E asmus Medical Cen e in Ro e dam (The Ne he lands). IEC is sup- po ed by a g an om he Alicia Koplowi z Founda ion. This wo k is pa o he Ma ía Rod iguez-Ayllon Ph.D. Thesis conduc ed in he Biomedi- cine Doc o al S udies o he Uni e si y o G anada, Spain. Acknowledgmen s The au ho s hank child en and pa en s who pa icipa ed in his s udy. The au ho s epo no biomedical financial in e es s o po en ial conflic s o in e es . Appendix A. Supplemen a y da a Supplemen a y da a o his a icle can be ound online a h ps://doi. o g/10.1016/j.neu oimage.2019.116258. Re e ences Achenback, T.M., Resco la, L a, 2003. Reliabili y, in e nal consis ency, c oss-in o man ag eemen , and s abili y. Man. ASEBA Sch. Age Fo ms P ofiles 99–107. Aoki, Y., Co ese, S., Cas ellanos, F.X., 2018. Resea ch Re iew: di usion enso imaging s udies o a en ion-defici /hype ac i i y diso de : me a-analyses and eflec ions on head mo ion. J. Child Psychol. Psychia y 59, 193–202. Asa o, M.R., Te willige , R., Woo, J., Luna, B., 2010. Whi e ma e de elopmen in adolescence: a DTI s udy. Ce eb . Co ex 20, 2122–2131. Ayling, E., Aghajani, M., Fouche, J.P., Van De Wee, N., 2012. Di usion enso imaging in anxie y diso de s. Cu . Psychia . Rep. 14, 197–202. Ba nea-Go aly, N., Menon, V., Ecke , M., Tamm, L., Bamme , R., Ka chemskiy, A., e al., 2005. Whi e ma e de elopmen du ing childhood and adolescence: a c oss-sec ional di usion enso imaging s udy. Ce eb . Co ex 15, 1848–1854. Beh ens, T.E.J., Wool ich, M.W., Jenkinson, M., Johansen-Be g, H., Nunes, R.G., Cla e, S., e al., 2003. Cha ac e iza ion and p opaga ion o unce ain y in di usion-weigh ed MR imaging. Magn. Reson. Med. 50, 1077–1088. Bick, J., Nelson, C.A., 2016. Ea ly ad e se expe iences and he de eloping b ain. Neu opsychopha macology 41, 177–196. Biddle, S.J.H., Asa e, M., 2011. Physical ac i i y and men al heal h in child en and adolescen s: a e iew o e iews. B . J. Spo s Med. 45, 886–895. Biddle, S.J.H., Ciaccioni, S., Thomas, G., Ve gee , I., 2018. Physical ac i i y and men al heal h in child en and adolescen s: an upda ed e iew o e iews and an analysis o causali y. Psychol. Spo Exe c. 1–10. B ouwe , R.M., Mandl, R.C.W., Schnack, H.G., Soelen, ILC a, Baal, GC an, Pepe , J.S., e al., 2012. Whi e ma e de elopmen in ea ly pube y: a longi udinal olume ic and di usion enso imaging win s udy. PLoS One 7, e32316. Caspe sen, C.J., Powell, K.E., Ch is enson, G.M., 1985. Physical ac i i y, exe cise, and physical fi ness: defini ions and dis inc ions o heal h- ela ed esea ch. Public Heal h Rep. 100, 126–131. Chaddock-Heyman, L., E ickson, K.I., Hol op, J.L., Voss, M.W., Pon i ex, M.B., Raine, L.B., e al., 2014. Ae obic fi ness is associa ed wi h g ea e whi e ma e in eg i y in child en. F on . Hum. Neu osci. 8, 584. Chaddock-Heyman, L., E ickson, K.I., Kienzle , C., D olle e, E., Raine, L., Kao, S.-C., e al., 2018. Physical ac i i y inc eases whi e ma e mic os uc u e in child en. F on . Neu osci. 12, 950. M. Rod iguez-Ayllon e al. Neu oImage 205 (2020) 116258 7 Chang, L.C., Jones, D.K., Pie paoli, C., 2005. RESTORE: obus es ima ion o enso s by ou lie ejec ion. Magn. Reson. Med. 53, 1088–1095. Chek oud, S.R., Gueo guie a, R., Zheu lin, A.B., Paulus, M., K umholz, H.M., K ys al, J.H., e al., 2018. Associa ion be ween physical exe cise and men al heal h in 1⋅2 million indi iduals in he USA be ween 2011 and 2015: a c oss-sec ional s udy. The Lance Psychia . 5, 739–746. Cook, P a, Bai, Y., Seuna ine, K.K., Hall, M.G., Pa ke , G.J., Alexande , D.C., 2006. Camino: open- sou ce di usion-MRI econs uc ion and p ocessing. 14 h Sci. Mee . In . Soc. Magn. Reson. Med. 14, 2759. Van Dijk, M.L., De G oo , R.H., Van Acke , F., Sa elbe g, H.H., Ki schne , P.A., 2014. Ac i e commu ing o school, cogni i e pe o mance, and academic achie emen : an obse a ional s udy in Du ch adolescen s using accele ome e s. BMC Public Heal h 14, 799. Eisenmann, J.C., 2004. Physical ac i i y and ca dio ascula disease isk ac o s in child en and adolescen s: an o e iew. Can. J. Ca diol. 20, 295–301. Elu a hingal, T.J., Hasan, K.M., K ame , L., Fle che , J.M., Ewing-Cobbs, L., 2007. Quan i a i e di usion enso ac og aphy o associa ion and p ojec ion fibe s in no mally de eloping child en and adolescen s. Ce eb . Co ex 17, 2760–2768. Fe guson, C.J., 2015. Do ang y bi ds make o ang y child en? A me a-analysis o ideo game influences on child en’s and adolescen s’agg ession, men al heal h, p osocial beha io , and academic pe o mance. Pe spec . Psychol. Sci. 10, 646–666. Fe guson, C.J., 2017. E e y hing in mode a ion: mode a e use o sc eens unassocia ed wi h child beha io p oblems. Psychia . Q. 88, 797–805. F ed iks, A.M., an Buu en, S., Bu gmeije , R.J., Meulmees e , J.F., Beuke , R.J., B ugman, E., e al., 2000. Con inuing posi i e secula g ow h change in The Ne he lands 1955-1997. Pedia . Res. 47, 316–323. F odl, T., Ca balledo, A., Fagan, A.J., Lisiecka, D., Fe guson, Y., Meaney, J.F., 2012. E ec s o ea ly- li e ad e si y on whi e ma e di usi i y changes in pa ien s a isk o majo dep ession. J. Psychia y Neu osci. 37, 37–45. Gio gio, A., Wa kins, K.E., Chadwick, M., James, S., Winmill, L., Douaud, G., e al., 2010. Longi udinal changes in g ey and whi e ma e du ing adolescence. Neu oimage 49, 94–103. Go golewski, K., Bu ns, C.D., Madison, C., Cla k, D., Halchenko, Y.O., Waskom, M.L., e al., 2011. Nipype: a flexible, ligh weigh and ex ensible neu oimaging da a p ocessing amewo k in Py hon. F on . Neu oin . 5, 13. De G oo , M., Ik am, M.A., Akoudad, S., K es in, G.P., Ho man, A., Van De Lug , A., e al., 2015. T ac -specific whi e ma e degene a ion in aging: he Ro e dam s udy. Alzheime ’s Demen ia 11, 321–330. Hallal, P.C., Ande sen, L.B., Bull, F.C., Gu hold, R., Haskell, W., Ekelund, U., e al., 2012. Global physical ac i i y le els: su eillance p og ess, pi alls, and p ospec s. Lance 380, 247–257. Hanson, J.L., Adlu u, N., Chung, M.K., Alexande , A.L., Da idson, R.J., Pollak, S.D., 2013. Ea ly neglec is associa ed wi h al e a ions in whi e ma e in eg i y and cogni i e unc ioning. Child De . 84, 1566–1578. Haselg o e, J.C., Moo e, J.R., 1996. Co ec ion o dis o ion o echo-plana images used o calcula e he appa en di usion coe ficien . Magn. Reson. Med. 36, 960–964. He ing, M.M., Colby, J.B., Sowell, E.R., Nagel, B.J., 2014. Whi e ma e connec i i y and ae obic fi ness in male adolescen s. De . Cogn. Neu osci. 7, 65–75. Hillman, C.H., E ickson, K.I., K ame , A.F., 2008. Be sma , exe cise you hea : exe cise e ec s on b ain and cogni ion. Na . Re . Neu osci. 9, 58–65. Hillman, C.H., Khan, N.A., Kao, S.-C., 2015. The ela ionship o heal h beha io s o childhood cogni ion and b ain heal h. Ann. Nu . Me ab. 66 (Suppl. 3), 1–4. Hul e sho n, L.A., Cullen, K., Anand, A., 2011. Towa d dys unc ional connec i i y: a e iew o neu oimaging findings in pedia ic majo dep essi e diso de . B ain Imaging Beha . 5, 307–328. Jenkinson, M., Beckmann, C.F., Beh ens, T.E.J., Wool ich, M.W., Smi h, S.M., 2012. FSL. Neu oimage 62, 782–790. Jenkinson, M., Smi h, S., 2001. A global op imisa ion me hod o obus a fine egis a ion o b ain images. Med. Image Anal. 5, 143–156. Jones, D.K., Ce cignani, M., 2010. Twen y-fi e pi alls in he analysis o di usion MRI da a. NMR Biomed. 23, 803–820. Kooijman, M.N., K ui ho , C.J., an Duijn, C.M., Duij s, L., F anco, O.H., an IJzendoo n, M.H., e al., 2016. The Gene a ion R S udy: design and coho upda e 2017. Eu . J. Epidemiol. 31, 1243–1264. K a , C.E., Schae e , D.J., Schwa z, N.F., Chi, L., Weinbe ge , A.L., Pie ce, J.E., e al., 2014. Imp o ed on opa ie al whi e ma e in eg i y in o e weigh child en is associa ed wi h a endance a an a e -school exe cise p og am. De . Neu osci. 36, 1–9. Lee, D.Y., Fle che , E., Ma inez, O., O ega, M., Zozulya, N., Kim, J., e al., 2009. Regional pa e n o whi e ma e mic os uc u al changes in no mal aging, MCI, and AD. Neu ology 73, 1722–1728. Lee, I.-M., Shi oma, E.J., Lobelo, F., Puska, P., Blai , S.N., Ka zma zyk, P.T., e al., 2012. E ec o physical inac i i y on majo non-communicable diseases wo ldwide: an analysis o bu den o disease and li e expec ancy. Lance 380, 219–229. Leemans, A., Jones, D.K., 2009. The B-ma ix mus be o a ed when co ec ing o subjec mo ion in DTI da a. Magn. Reson. Med. 61, 1336–1349. Len oo , R.K., Giedd, J.N., 2006. B ain de elopmen in child en and adolescen s: insigh s om ana omical magne ic esonance imaging. Neu osci. Biobeha . Re . 30, 718–729. Mue zel, R.L., Blanken, L.M.E., an de Ende, J., El Ma oun, H., Shaw, P., Sud e, G., e al., 2017. T acking b ain de elopmen and dimensional psychia ic symp oms in child en: a longi udinal popula ion-based neu oimaging s udy. Am. J. Psychia y 175, 54–62. Mue zel, R.L., Mous, S.E., an de Ende, J., Blanken, L.M.E., an de Lug , A., Jaddoe, V.W.V., e al., 2015. Whi e ma e in eg i y and cogni i e pe o mance in school-age child en: a popula ion-based neu oimaging s udy. Neu oimage 119, 119–128. O ben, A., P zybylski, A.K., 2019. The associa ion be ween adolescen well-being and digi al echnology use. Na . Hum. Beha . 3, 173–182. Rezende, LFM de, Rod igues Lopes, M., Rey-L opez, J.P., Ma sudo, V.K.R., Luiz, O. do C., 2014. Seden a y beha io and heal h ou comes: an o e iew o sys ema ic e iews. PLoS One 9, e105620. Schae e , D.J., K a , C.E., Schwa z, N.F., Chi, L., Rod igue, A.L., Pie ce, J.E., e al., 2014. An 8- mon h exe cise in e en ion al e s on o empo al whi e ma e in eg i y in o e weigh child en. Psychophysiology 51, 728–733. Schmi ho s , V.J., Yuan, W., 2010. Whi e ma e de elopmen du ing adolescence as shown by di usion MRI. B ain Cogn. 72, 16–25. She e , G., Ma cus, Y., S e n, N., 2013. Is obesi y a b ain disease? Neu osci. Biobeha . Re . 37, 2489–2503. Smi h, S.M., 2002. Fas obus au oma ed b ain ex ac ion. Hum. B ain Mapp. 17, 143–155. S iles, J., Je nigan, T.L., 2010. The Basics o B ain De elopmen , pp. 327–348. Takeuchi, H., Taki, Y., Hashizume, H., Asano, K., Asano, M., Sassa, Y., e al., 2015. The impac o ele ision iewing on b ain s uc u es: c oss-sec ional and longi udinal analyses. Ce eb . Co ex 25, 1188–1197. Tamnes, C.K., Roal , D.R., Goddings, A.-L., Lebel, C., 2018. Di usion MRI o whi e ma e mic os uc u e de elopmen in childhood and adolescence: me hods, challenges and p og ess. De . Cogn. Neu osci. 33, 161–175. Tellegen, P.J., Winkel, M., Wijnbe g-Williams, B., La os, J.A., 2005. Snijde s-Oomen Nie - Ve bale In elligen ie es : SON-R 2-1/2 - o-7. Ans e dam. Boom Tes ui ge e s. Tompo owski, P.D., Da is, C.L., Mille , P.H., Naglie i, J.A., 2008. Exe cise and child en’s in elligence, cogni ion, and academic achie emen . Educ. Psychol. Re . 20, 111–131. T emblay, M.S., Aube , S., Ba nes, J.D., Saunde s, T.J., Ca son, V., La ime -Cheung, A.E., e al., 2017. Seden a y beha io esea ch ne wo k (SBRN) – e minology consensus p ojec p ocess and ou come. In . J. Beha . Nu . Phys. Ac . 14, 75. Walsh, J.J., Ba nes, J.D., Came on, J.D., Goldfield, G.S., Chapu , J.-P., Gunnell, K.E., e al., 2018. Associa ions be ween 24 hou mo emen beha iou s and global cogni ion in US child en: a c oss-sec ional obse a ional s udy. Lance . Child Adolesc. Heal h 2, 783–791. Whi e, T., Mue zel, R.L., El Ma oun, H., Blanken, L.M.E., Jansen, P., Bolhuis, K., e al., 2018. Paedia ic popula ion neu oimaging and he Gene a ion R S udy: he second wa e. Eu . J. Epidemiol. 33, 99–125. M. Rod iguez-Ayllon e al. Neu oImage 205 (2020) 116258 8