SAGE Open
Oc obe -Decembe 2016: 1 –18
© The Au ho (s) 2016
DOI: 10.1177/2158244016681393
sgo.sagepub.com
C ea i e Commons CC-BY: This a icle is dis ibu ed unde he e ms o he C ea i e Commons A ibu ion 3.0 License
(h p://www.c ea i ecommons.o g/licenses/by/3.0/) which pe mi s any use, ep oduc ion and dis ibu ion o
he wo k wi hou u he pe mission p o ided he o iginal wo k is a ibu ed as speci ied on he SAGE and Open Access pages
(h ps://us.sagepub.com/en-us/nam/open-access-a -sage).
A icle
Lea ning o egula e emo ional expe iences and impulses is
one o he mos impo an de elopmen al asks du ing child’s
ea ly yea s. The e is obus e idence ha bo h emo ion egu-
la ion (Aldao, Nolen-Hoeksema, & Schweize , 2010) and
de ense mechanisms (Bond, 2004) a e impo an o one’s
men al heal h and socioemo ional unc ioning. Howe e ,
ew empi ical s udies ha e conside ed emo ion egula ion
and de ense mechanisms oge he , p obably because o hei
di e en o igins in cogni i e e sus psychodynamic and
clinical esea ch adi ions (Sala, Tes a, Pons, & Molina,
2015). Fu he mo e, whe eas subs an ial empi ical esea ch
is a ailable on he de elopmen o emo ion egula ion (e.g.,
Calkins & Hill, 2007; Eisenbe g, Spin ad, & Eggum, 2010;
Kopp & Neu eld, 2003), esea ch on child en’s de ense
mechanisms and hei ea ly p edic o s is mo e sca ce (see
C ame , 2006). Thus, he p esen s udy aims o inc ease
knowledge conce ning he di e ences and simila i ies
be ween ea ly p edic o s o emo ion egula ion and de ense
mechanisms. We analyze how he quali y o amily ela ion-
ships du ing in ancy p edic s child en’s emo ion egula ion
and de ense mechanisms in middle childhood. We expand
ea lie esea ch by conside ing emo ion egula ion and
de ense mechanisms oge he , and by es ing he impo ance
o iming o he amily ela ionships du ing ea ly and la e
in ancy.
Emo ion Regula ion and De ense
Mechanisms
Regula ion o a ec i e s a es, such as emo ions, mood, s ess,
and mo i a ional impulses, in ol es mul iple p ocesses which
help o main ain, o example, goal-di ec ed beha io s, posi-
i e mood, and sense o secu i y (G oss & Thompson, 2007;
Ha , 2014; Koole, 2009). Acco ding o G oss and Thompson,
such a ec egula ion p ocesses in ol e bo h emo ion egula-
ion and de ense mechanisms (G oss, 1998; G oss &
Thompson, 2007). Al hough he di e ences and simila i ies
681393SGOXXX10.1177/2158244016681393SAGE OpenLindblom e al.
esea ch-a icle2016
1Uni e si y o Tampe e, Finland
2Uni e si y o Tu ku, Finland
3Uni e si y o Helsinki, Finland
4Helsinki Uni e si y Cen al Hospi al, Finland
Co esponding Au ho :
Jallu Lindblom, School o Social Sciences and Humani ies / Psychology,
Uni e si y o Tampe e, FIN-33014, Finland.
Email: [email p o ec ed]
Ea ly Family Rela ionships P edic
Child en’s Emo ion Regula ion and
De ense Mechanisms
Jallu Lindblom1, Raija-Leena Punamäki1, Ma jo Flyk 1,
Me i Vänskä1, Tapio Nummi1, Ja i Sinkkonen2,
Aila Tii inen3,4, and Maija Tulppala3
Abs ac
Ea ly amily ela ionships ha e been sugges ed o in luence he de elopmen o child en’s a ec egula ion, in ol ing bo h
emo ion egula ion and de ense mechanisms. Howe e , we lack esea ch on he speci ic amily p edic o s o hese wo
o ms o a ec egula ion, which ha e been concep ualized o di e in hei unc ions and accessibili y o consciousness.
Acco dingly, we examine how he (a) quali y and (b) iming o amily ela ionships du ing in ancy p edic child’s la e emo ion
egula ion and de ense mechanisms. Pa en s (N = 703) epo ed au onomy and in imacy in ma i al and pa en ing ela ionships
a he child’s ages o 2 and 12 mon hs, and he child’s use o emo ion egula ion and imma u e and neu o ic de enses a 7 o
8 yea s. As hypo hesized, he esul s showed ha unc ional ea ly amily ela ionships p edic ed child en’s e icien emo ion
egula ion, whe eas dys unc ional ela ionships p edic ed eliance on de ense mechanisms in middle childhood. Fu he ,
esul s showed a iming e ec o neu o ic de enses, pa ially con i ming ou hypo hesis o ea ly in ancy being an especially
impo an pe iod o he de elopmen o de ense mechanisms. The indings a e discussed om he iewpoin s o a achmen
and amily dynamics, emo ional sel -awa eness, and sense o secu i y.
Keywo ds
amily ela ionships, emo ion egula ion, de ense mechanisms, in ancy, ca egi ing
2 SAGE Open
be ween hese wo a e inadequa ely unde s ood, emo ion eg-
ula ion is concep ualized o ocus on managing disc e e emo-
ional s a es (G oss, 1998; G oss & Thompson, 2007), and
de ense mechanisms on managing mo i a ional impulses and
needs (Ha , 2014; Vaillan , 1995). Fu he mo e, whe eas
emo ion egula ion has been sugges ed o ope a e bo h con-
sciously and unconsciously (G oss & Thompson, 2007;
Mauss, Bunge, & G oss, 2007), de ense mechanisms a e
hough o ope a e unconsciously (C ame , 2008; G oss &
Thompson, 2007; Vaillan , 1995).
Emo ion egula ion e e s o he p ocesses indi iduals use
o in luence which emo ions hey expe ience and when and
how hey expe ience and exp ess hem (G oss, 1998; G oss &
Thompson, 2007). Volun a y emo ion egula ion ypically
occu s when an indi idual becomes awa e o own emo ional
s a es and shapes hem acco ding o si ua ional demands and
pe sonal goals (G oss, Richa ds, & John, 2006). Emo ions
can be egula ed, o example, by a ending owa d less emo-
ion-p o oking aspec s o he si ua ion o by cogni i ely eap-
p aising he meaning o he si ua ion. Sel -awa eness o one’s
own emo ions os e s e icien emo ion egula ion, likely
because his allows in e nal s a es o be be e unde s ood and
modi ied (He wig, Ka enbe ge , Jäncke, & B ühl, 2010;
Subic-W ana e al., 2014). In e es ingly, howe e , esea ch
sugges s ha emo ion egula ion can also occu au oma ically
ou side o awa eness (Ba gh, Schwade , Hailey, Dye , &
Boo hby, 2012; Mauss, Bunge, & G oss, 2007). Fo example,
Mauss, Cook, and G oss (2007) demons a ed ha sublimi-
nally p iming emo ion egula ion (by p esen ing wo ds ela ed
o emo ional con ol) helped pa icipan s o down egula e
hei emo ional esponses o ange p o oca ion. Such au o-
ma ic emo ion egula ion p obably e lec s he ac i a ion o
p e iously lea ned and ou inized emo ion egula ion s a e-
gies (Mauss, Bunge, & G oss, 2007).
De ense mechanisms modula e emo ional expe iences
unconsciously, wi hou being consciously accessible (C ame
& B illian , 2001; G oss & Thompson, 2007). They aim o
main ain psychological sense o secu i y by p oducing cog-
ni i e dis o ions and by limi ing he conscious expe ience o
nega i e emo ions (Ha , 2014; S eine , A aujo, & Koopman,
2001). Resea ch sugges s ha de ensi e sel -decep ion,
in ol ing biased a en ion and memo y, is inhe en ly uncon-
scious because awa eness o i would impede i s e ec i e-
ness ( on Hippel & T i e s, 2011). Fo example, conscious
and delibe a e a emp s o supp ess unwan ed hough s o en
esul in hei eappea ance in o consciousness (Ab amowi z,
Tolin, & S ee , 2001), whe eas unconscious and au oma ic
ep ession o unwan ed hough s is mo e e icien (Ge ae s,
D i schel, K eplin, Miyagawa, & Wadding on, 2012; Lambie
& Ke in, 2003). Thus, i seems essen ial o de ense mecha-
nisms o ul ill hei unc ion by ope a ing unconsciously.
F om de elopmen al pe spec i e, C ame and B illian
(2001) ound ha child en’s unde s anding o de ense mech-
anisms inc eased om 7 o 11 yea s. Impo an ly, hose chil-
d en who unde s ood he de ensi e unc ion o denial and
p ojec ion in igne es we e less likely o use hese de enses
hemsel es. Such indings suppo he iew ha child en
ypically ely on p imi i e and cogni i ely simple de ense
mechanisms (e.g., denial) du ing in ancy and ea ly child-
hood (F aibe g, 1982), bu p og ess owa d mo e complex
de enses (e.g., p ojec ion and iden i ica ion) as hei cogni-
i e abili ies and sel -awa eness de elop in middle childhood
and beyond (C ame , 2006).
Vaillan (1971, 1995) ca ego ized de enses acco ding o
hei de elopmen al ma u i y and men al complexi y.
Empi ical s udies ha e con i med he exis ence o wo o
h ee de ense dimensions among adul s (e.g., And ews,
Pollock, & S ewa , 1989; Bond, 1995, 2004) and child en
and adolescen s (A aujo, Medic, Yasno sky, & S eine ,
2006; S eine e al., 2001). Imma u e de enses p oduce
se e e cogni i e dis o ions abou sel and o he s. Fo
example, in p ojec ion, unaccep able emo ions a e a ib-
u ed o emancipa e om o he s, and in omnipo ence, sel is
pe cei ed supe io in compa ison wi h o he s. Neu o ic
de enses ypically al e subjec i e expe iences by dissocia -
ing emo ional and cogni i e men al con en s. Fo example,
in ep ession, a h ea ening hough is shu ou o con-
sciousness, and in eac ion o ma ion, i is ans o med in o
i s opposi e. In con as , ma u e de enses ypically cause
only mino cogni i e dis o ions (Vaillan , 2000) and hey
ha e been sugges ed o be mo e conscious and delibe a e
(Con e & Plu chik, 1993; C ame , 2006; Vaillan , 2000).
Indeed, eliance on ma u e de enses has been ound o asso-
cia e wi h high emo ional sel -awa eness and e icien emo-
ion egula ion (Besha a & Khaja i, 2013; Sala e al.,
2015). As ma u e de enses canno be clea ly di e en ia ed
om emo ion egula ion, he p esen s udy ocuses on
imma u e and neu o ic de enses.
Acco ding o Ha ’s (2014) in eg a i e de ense heo y,
eliance on sel -decep i e de ense mechanisms is p ima ily
mo i a ed by sense o insecu i y, cha ac e ized by expe i-
ences o ulne abili y and lack o con idence abou one’s
abili y o cope wi h h ea s. Insecu i y can s em om a i-
ous sou ces, in ol ing a achmen ela ionships, sel -
es eem, and con lic s in belie s. In line wi h he heo y,
s udies ha e shown eliance on de ense mechanisms o
inc ease a e being ejec ed by an impo an pee among 9-
o 11-yea -old gi ls (Sands om & C ame , 2003) and o
associa e wi h low emo ional upse a e auma ic e en
among 10- o 13-yea -old boys (Dollinge & C ame , 1990).
Fu he mo e, e ospec i e and c oss-sec ional adul s udies
ha e shown eliance on imma u e and neu o ic de enses o
associa e wi h childhood expe iences o ha sh pa en ing
(Finzi-Do an & Ka u, 2006), a achmen insecu i y
(Besha a & Khaja i, 2013), and belie s o abandonmen
(Walbu g & Chia amello, 2015) and one’s own emo ions
being unaccep able (Sala e al., 2015). While such associa-
ions ha e no been s udied among child en, hey sugges
ha eliance on de ense mechanisms can be shaped by ea ly
social and emo ional expe iences.
Lindblom e al. 3
Ea ly Family Rela ionships and
De elopmen o A ec Regula ion
Du ing in ancy, child en ely hea ily on hei mo he s and
a he s o aid o egula e hei a ousal and dis ess (Ekas,
B aunga -Rieke , Lickenb ock, Zen all, & Maxwell, 2011;
Kopp & Neu eld, 2003). Sensi i e and suppo i e ca egi ing
acili a es child en’s cogni i e de elopmen , such as a en-
ion and execu i e skills (Be nie , Ca lson, Deschênes, &
Ma e-Gagné, 2012; E ans & Po e , 2009), which in u n
p omo es e icien emo ion egula ion (Eisenbe g e al.,
2010). Fu he mo e, as depic ed by a achmen heo y, sensi-
i e ca egi e s os e child en’s sense o secu i y by p o id-
ing emo ional accep ance, expe ise, and con inuous suppo
on which he child can ely on (Cassidy, 1994; Thompson &
Meye , 2007). This helps child en o de elop emo ional
awa eness and e ec i e emo ion egula ion wi h po en ially
long- e m posi i e impac on la e de elopmen (Dykas &
Cassidy, 2011; Mou siana e al., 2015).
In con as , insensi i e ca egi ing impai s child en’s cog-
ni i e de elopmen and sense o secu i y (Be nie e al.,
2012; Cassidy, 1994). Child en wi h insecu e a achmen
s yle end o ei he exagge a e o supp ess hei emo ional
exp essions o ensu e pa en al p oximi y and p o ec ion
(Ainswo h, Bleha , Wa e s, & Wall, 1978). A achmen
esea ch sugges s ha insecu ely a ached child en and adul s
p ocess a achmen - ela ed in o ma ion de ensi ely, o a oid
psychological pain ( o a e iew, see Dykas & Cassidy,
2011). Fo example, Ki sh and Cassidy (1997) ound ha
insecu ely a ached 3-yea -old child en showed a en ional
a oidance o a achmen cues (e.g., mo he -child d awings)
and de ici s in emembe ing h ea ening in o ma ion (e.g.,
s o y abou ma e nal ejec ion). Howe e , o he bes o ou
knowledge, p e ious s udies ha e no examined whe he
insensi i e ca egi ing p edic s child en’s eliance on de ense
mechanisms du ing la e de elopmen .
To gain a comp ehensi e unde s anding abou child en’s
de elopmen al en i onmen , i is impo an o b oaden he
ocus om dyadic ela ionships o he wide amily sys em
(Mo is, Silk, S einbe g, Mye s, & Robinson, 2007;
Thompson & Meye , 2007). Acco ding o emo ional secu i y
heo y, child en adap hei egula o y s a egies o i he
quali y o amily ela ionships, in ol ing in e pa en al con-
lic s and in e ac ions (Da ies & Ma in, 2013). Child en
may, o example, supp ess o exagge a e hei exp ession o
emo ional dis ess o de use o a oid in e pa en al con lic s.
In line wi h his, in an s udies ha e demons a ed ha in e -
pa en al con lic s inc ease child en’s emo ion dys egula ion
and a en ional a oidance o s ess-p o oking s imuli
(C ockenbe g, Lee kes, & Lekka, 2007; Du Roche
Schudlich, Whi e, Fleischhaue , & Fi zge ald, 2011). In ou
p e ious s udy, we ound ha dys unc ional amily sys ems
du ing in ancy, in ol ing bo h pa en al and ma i al subsys-
ems, p edic ed bo h child en’s a en ional a oidance and
a en ional bias owa d h ea s (i.e., ang y acial exp essions)
in middle childhood (Lindblom, e al., 2015). Al oge he ,
hese s udies sugges ha child en’s de elopmen o emo ion
egula ion and de ense mechanism is malleable by he ea ly
in e pe sonal expe iences wi hin he amily.
Resul s o a achmen and amily s udies concu wi h
Ha ’s (2014) in eg a i e de ense heo y, by demons a ing
ha insecu i y-p o oking amily ela ionships heigh en chil-
d en’s de ensi eness, as indica ed by hei sel -p o ec i e
beha io s and in o ma ion p ocessing biases. Su p isingly,
howe e , p e ious p ospec i e s udies ha e no examined
how he quali y o amily ela ionships du ing in ancy p e-
dic s child en’s a ec egula ion du ing middle childhood o
beyond. Fu he mo e, mos esea ch abou a ec egula ion
has ocused on he quali y o ma e nal ca egi ing, excluding
a he s and in e pa en al ela ionships. Thus, in he cu en
p ospec i e s udy, we assess he quali y o child en’s in e ac-
ional en i onmen a he amily le el, including bo h he
pa en ing and he ma i al subsys ems. Au onomy and in i-
macy, wo o he e y mos basic ela ional dimensions
(Byng-Hall, 1999), a e applied o de ine he ela ionship
quali y in ma i al and pa en ing subsys ems. In ela ionship
wi h he spouse o child, au onomy e e s o indi iduali y
and a sense o agency, e lec ing unc ional amily bounda -
ies, and in imacy e e s o eelings o lo e and sha ing o
emo ions (Ma eja & Scholz, 1994). Low au onomy and
in imacy in he ma i al subsys em indica e poo ela ionship
quali y and associa e wi h in e pa en al con lic s and e bal
agg ession (Ga azzi, McKen y, Jacobson, Julian, & Lohman,
2000; Rankin-Esque , Bu ne , Baucom, & Eps ein, 1997),
whe eas low au onomy and in imacy in pa en ing indica e
insensi i e pa en ing and associa e wi h o e ly hesi an ,
in usi e, and emo ionally disengaged ca egi ing (Leung,
Mille , Lumeng, Kaci o i, & Rosenblum, 2015; Sokolowski,
Hans, Be ns ein, & Cox, 2007).
Age-Speci ic De elopmen Du ing
In ancy
Neu ode elopmen al and beha io al s udies sugges ha sen-
si i e pe iods exis in child de elopmen wi h po en ial long-
e m consequences on la e unc ioning (Pech el & Pizzagalli,
2011; To enham & She idan, 2010). As an example o sensi-
i e pe iod, animal s udies ha e demons a ed ha e y ea ly
dis up ions in ma e nal ca e can pe manen ly al e a ach-
men - ela ed neu al unc ioning and emo ional lea ning
(Rincón-Co és & Sulli an, 2014). Fu he mo e, he s udies
o ins i u ionalized and hen adop ed child en sugges ha he
i s 2 yea s o li e a e especially impo an o social skills
(Almas e al., 2012), and he second yea o li e is especially
impo an o execu i e unc ions (Me z, McCall, W igh , &
Luna, 2013). Howe e , esea ch is lacking abou he mo e
no ma i e social expe iences wi hin he in ancy pe iod. Fo
example, one s udy ound ha high amoun o ma e nal s ok-
ing a he in an ’s age o 2 mon hs was bene icial on chil-
d en’s emo ional well-being a he age o 2.5 yea s (Sha p,
4 SAGE Open
Hill, Hellie , & Pickles, 2015). Ye , he s udy did no es
whe he he iming o ma e nal s oking (e.g., ea ly s. la e
in ancy) would mode a e he e ec . To ou knowledge, only
one s udy has explici ly es ed o he exis ence o age-spe-
ci ic iming e ec s du ing in ancy. The pe son-o ien ed s udy
analyzed he cou se o ma e nal psychological dis ess ac oss
he p e- and pos pa um pe iod, and ound ha child en o
mo he s who we e symp oma ic a he child’s age o 2 mon hs
(bu no du ing he p egnancy o he child’s age o 12 mon hs)
showed inc eased in e nalizing symp oms a 7 o 8 yea s,
compa ed wi h child en o mo he s who we e symp oma ic
only a p egnancy o a he child’s age o 12 mon hs (Vänskä
e al., 2011). The esul s a e indica i e o po en ial age-spe-
ci ic e ec s du ing in ancy on he de elopmen o a ec
egula ion.
Neu ode elopmen al esea ch sugges s ha simple and
in olun a y unc ions, such as implici emo ional lea ning,
de elop ea lie han complex and olun a ily con olled
unc ions, such as execu i e skills (Pech el & Pizzagalli,
2011). Conside ing he unc ional di e ences be ween emo-
ion egula ion and de ense mechanisms, i can be hypo he-
sized ha he au oma ized p ocesses ela ed o de ense
mechanisms de elop ea lie han hose o he mo e cogni-
i ely complex emo ion egula ion. Rega ding de ense
mechanisms, i has been sugges ed ha ep ession, ha is,
de ensi e exclusion o h ea ening hough s om conscious-
ness, is ela ed o impai ed memo y o ma ion and ecall,
in ol ing al e ed amygdala and hippocampus unc ion
(Axmache , Do Lam, Kessle , & Fell, 2010). In e es ingly,
de elopmen al esea ch sugges s ha he de elopmen o
hese b ain s uc u es is malleable o s ess-induced al e a-
ions al eady du ing ea ly in ancy (To enham & She idan,
2010). Rega ding emo ion egula ion, he e is some e idence
ha he unde lying b ain s uc u es, ela ed o conscious
moni o ing o and con olling own emo ions (e.g., o bi o-
on al co ex and an e io cingula e gy us), a e malleable o
expe iences du ing he la e in ancy and la e on (Mou siana
e al, 2014; Zelazo, Qu, & Kesek, 2010).
Resea ch Tasks and Hypo heses
We examined how he quali y and iming o ea ly amily
ela ionships p edic child en’s la e emo ion egula ion and
de ense mechanisms. Au onomy and in imacy in ma i al and
pa en al amily subsys ems we e assessed a he child’s age
o 2 mon hs and 12 mon hs. Emo ion egula ion, imma u e
de enses, and neu o ic de enses we e assessed when he chil-
d en we e 7 o 8 yea s old.
As ou i s esea ch ask, we es ed a di e en amily p e-
condi ions hypo hesis ha he quali y o amily ela ionships
du ing in ancy p edic s child en’s la e emo ion egula ion
and de ense mechanisms. We hypo hesized ha well- unc-
ioning amily ela ionships, in ol ing high le els o au on-
omy and in imacy, would p edic child en’s e icien emo ion
egula ion, and dys unc ional amily ela ionships, in ol ing
low le el o au onomy and in imacy, would p edic chil-
d en’s eliance on imma u e and neu o ic de enses.
As ou second esea ch ask, we es ed an age-speci ic
hypo hesis ha he iming o amily ela ionship quali y a
he ages o 2 mon hs and 12 mon hs would di e en ly p edic
emo ion egula ion and de ense mechanisms. We hypo he-
sized ha ela ionship quali y a 12 mon hs would p edic
e ec i eness o emo ion egula ion mo e s ongly han a 2
mon hs. Fu he , we hypo hesized ha amily- ela ionship
quali y a 2 mon hs would p edic eliance on imma u e and
neu o ic de enses mo e s ongly han a 12 mon hs.
Child’s ea ly cha ac e is ics, in ol ing empe amen al
ai s and de elopmen al achie emen s, can in luence he
quali y o ea ly pa en ing (e.g., Bi ingen, Emde, Campos, &
Appelbaum, 1995) and child’s sel - egula ion de elopmen
(e.g., U sache, Blai , S i e , Voeg line, & Family Li e P ojec
In es iga o s, 2013). Fu he mo e, hal o he couples pa -
icipa ing in ou sample had achie ed pa en hood h ough
assis ed ep oduc i e ea men (ART). Thus, we con olled
o he po en ially con ounding e ec s o he i s yea de el-
opmen al achie emen s and ART s a us in addi ion o o he
backg ound a iables. Finally, when modeling age-speci ic
e ec s, we con olled o he e ec s o concu en amily
ela ionships a he age o 7 o 8 yea s.
Me hod
Pa icipan s and P ocedu e
The s udy sample consis ed o 703 Finnish ma ied o cohab-
i an couples. O he pa icipa ing couples, 56% had ecei ed
success ul ART (n = 392) and we e ec ui ed om i e in e -
ili y clinics in Finland, whe eas 44% we e na u ally con-
cei ing couples (NC; n = 311) and we e ec ui ed a Helsinki
Uni e si y Cen al Hospi al while pa icipa ing in ou ine
ul asonog aphic examina ion. Couples wi h mul iple p eg-
nancies we e excluded om he s udy sample and only
women abo e he age o 25 yea s we e included in he NC
g oup. The pa icipan s p o ided in o med consen a he
beginning o he s udy and a T3. The e hics commi ees o
he esponsible clinics app o ed he s udy a each s age o
he da a collec ion.
The ART couples we e mo e o en p imipa ous (65.70%)
han he NC couples (34.30%), χ2(1, N = 703) = 49.91, p <
.001. In he whole sample, he educa ion le el was ela i ely
high: 29% o he mo he s and 31% o he a he s had a uni-
e si y-le el educa ion, 40% o he mo he s and 27% o he
a he s had a college-le el educa ion, 14% o he mo he s
and 23% o he a he s had oca ional aining, and 17% o
he mo he s and 19% o he a he s had basic educa ion o
we e s uden s. The mean age o he mo he s was 33.19 yea s
(SD = 3.73) and a he s 34.60 (SD = 4.95).
This s udy is based on ques ionnai es comple ed sepa-
a ely by mo he s and a he s when hei child was 2 mon hs
(T1), 12 mon hs (T2), and 7 o 8 yea s (T3) o age. Response
Lindblom e al. 5
a es (a leas one pa en pa icipa ing) we e 94% a T1 (n =
656), 78% a T2 (n = 547), and 60% a T3 (n = 420). A i ion
a T2 and T3 was independen o he mo he s’ o a he s’
le el o educa ion, he leng h o hei ela ionship, he pa -
en s’ ages, and he child’s gende . A i ion a T2 and T3 was
also independen o he ART s a us o he mo he s, bu i was
g ea e among NC (8.9%) han ART (5.4%) a he s a T2,
χ2(1, N = 656) = 3.18, p = .052.
Measu es
Family ela ionships. Family ela ionships we e measu ed by
he Subjec i e Family Pic u e Tes (SFPT; Ma eja &
Scholz, 1994) a he child’s ages o 2 mon hs (T1), 12 mon hs
(T2), and 7 o 8 yea s (T3). Bo h pa en s a ed he quali y o
ou amily ela ionships: wi e- o-husband, husband- o-wi e,
mo he - o-child, and a he - o-child. These ela ionships
we e a ed in e ms o au onomy ( ou pai s o i ems, e.g.,
de e mined–indecisi e, shy–sel -assu ed) and emo ional
in imacy ( ou pai s o i ems, e.g., ejec ing–lo ing, wa m–
cool) using a 7-poin scale. Highe sco es on au onomy indi-
ca e ela ional sel -assu ance, agency, and independence,
whe eas high sco es on in imacy indica e emo ional a ach-
men , in e es , and accep ance. To measu e au onomy o in i-
macy in he ma i al subsys em and in he amily subsys em,
he co esponding wo i ems we e a e aged ac oss pa en ing
ela ionships (mo he - o-child and a he - o-child) and ac oss
ma i al ela ionships (wi e- o-husband and husband- o-wi e)
sepa a ely o bo h pa en s’ epo s. This a e aging was jus-
i ied as he e we e medium-sized co ela ions be ween he
i ems (a e age = .44, anging om .27 o .79, all p < .001).
Howe e , he co ela ion was small o he i em de e mined–
indecisi e be ween mo he s’ epo s o wi e- o-husband and
husband- o-wi e ela ionships a T3, = .15, p < .001. The
esul ing a iables we e used as indica o a iables o la en
cons uc s o ma i al au onomy, ma i al in imacy, pa en al
au onomy, and pa en al in imacy, sepa a ely o a he s’ and
mo he s’ epo s, a T1, T2, and T3. The alidi y and eli-
abili y o SFPT scales ha e been demons a ed, o example,
by la ge co ela ions (a e age = .60) wi h o he amily
diagnos ic ques ionnai es, and accep able e es eliabili y
o e 2 weeks in e al ( = .77; Ma eja & Scholz, 1994).
Child’s emo ion egula ion. The child’s emo ional sel - egula-
ion a he age o 7 o 8 yea s (T3) was assessed by he sel -
egula ion subscale o he Emo ion Ques ionnai e (EQ;
Rydell, Be lin, & Bohlin, 2003). EQ consis s o igne es
desc ibing emo ion-e oking si ua ions o di e en emo-
ions. Fo each igne e, pa en s es ima ed on a 5-poin Lik-
e scale how easily he child was able o calm down by
him- o he sel (1 = doesn’ apply a all, 5 = applies e y well
o my child). We used nine igne es o nega i e emo ions o
ange (e.g., My child ge s in o a con lic wi h a pee ), sadness
(e.g., A oy is los o b oken), and ea (e.g., My child ge s
igh ened and wo ied). To include mo e complex emo ions
as well, we added six igne es o shame/guil (e.g., My
child ge s caugh doing some hing o bidden). Fi s , i ems
we e a e aged sepa a ely o each emo ion o ange ( h ee
i ems), sadness ( h ee i ems), ea ( h ee i ems), and shame/
guil (six i ems). Second, hese ou a iables we e a e aged
o ep esen e iciency o emo ion egula ion sepa a ely o
mo he s (α = .84) and a he s (α = .93). These wo a iables
we e used as indica o a iables in s uc u al equa ion mod-
els. The alidi y and eliabili y o EQ has been demons a ed
by showing high sco es on EQ a he age o 6 yea s o p edic
low beha io al and emo ional p oblems and p osocial beha -
io s a he age o 8 yea s, and accep able e es eliabili y
o e 5 weeks in e al ( anging om = .74 o .97; Rydell
e al., 2003).
Child’s de ense mechanisms. The child’s de ense mechanisms
we e measu ed a he age o 7 o 8 yea s (T3) wi h he pa en
e sion o he Response E alua ion Measu e o Pa en s
(REM-P: S eine e al., 2001; Yasno sky e al., 2003).
REM-P is based on Vaillan ’s (1971, 1995) model o de ense
mechanisms and is simila o he widely s udied De ense
S yle Ques ionnai e (And ews e al., 1989; Bond, 1995).
Howe e , REM-P is modi ied o be sui able o adolescen s
and child en and o a oid o e ly pa hological wo ding. I
comp ises 71 i ems ha desc ibe 21 de enses anging om
imma u e o neu o ic and ma u e de ense mechanisms, such
as ep ession ( h ee i ems; e.g., My child doesn’ show his/
he ue eelings), p ojec ion ( h ee i ems; e.g., My child eels
ha s/he is always ea ed un ai ly), and in ellec ualiza ion
( ou i ems; e.g., My Child uses eason and logic, no eel-
ings, o unde s and people). Bo h pa en s independen ly es i-
ma ed he child’s ypical de ensi e beha io s on a 5-poin
Like scale (1 = o ally disag ee, 9 = o ally ag ee). Al hough
de ense mechanisms a e conside ed o ope a e uncon-
sciously, hei ope a ion can be assessed h ough hei esidu-
als in beha io and emo ional esponding (Bond, 1995).
Fi s , o ob ain de ense sco es, 21 sum a iables we e
compu ed by a e aging he i ems ep esen ing each de ense
mechanism ( h ee o ou i ems pe de ense mechanism).
Second, o examine he ac o s uc u e o pa en al epo s o
de ense mechanisms in his age g oup o child en, we pe -
o med explo a o y ac o analyses (using a e aged alues
be ween he pa en s’ epo s and he p incipal ex ac ion
me hod wi h oblimin o a ion). The analysis yielded a h ee-
ac o solu ion: (a) imma u e de enses (22.19% a iance
explained; e.g., ac ing ou , p ojec ion, displacemen , omnip-
o ence, passi e agg ession), (b) ma u e de enses (16.12%;
e.g., humo , in ellec ualiza ion, sublima ion, eac ion o ma-
ion, al uism), and (c) neu o ic de enses (8.17%; e.g., ep es-
sion, denial, dissocia ion, wi hd awal, supp ession). Two
sum a iables o indi idual de ense mechanisms had o be
excluded om he analyses because o low a iabili y (con-
e sion) o low ini ial eigen alues in ac o analysis (<0.20
o soma iza ion). Thi d, based on his h ee- ac o solu ion,
de ense-s yle sco es we e compu ed by a e aging he

6 SAGE Open
co esponding sum a iables o ep esen he child’s eliance
on imma u e de enses ( i e a iables; mo he , α = .74; a he ,
α = .72) and neu o ic de enses (six a iables; mo he , α =
.64; a he , α = .67) sepa a ely o bo h pa en s’ epo s. The
esul ing ou a iables we e used as indica o a iables in
s uc u al equa ion models. Ma u e de enses we e excluded
om he main analyses because ou hypo heses did no con-
ce n hem. The alidi y o he sel - epo e sion o he ques-
ionnai e has been demons a ed, o example, by showing
co ela ions wi h anxie y and psychosocial unc ioning
among 8- o 15-yea -old child en (A aujo e al., 2006).
Howe e , he pa en e sion has been p e iously used only
in one s udy o 34 mo he s and hei 7- o 10-yea -old chil-
d en (Yasno sky e al., 2003). The s udy showed accep able
e es eliabili y o e 2 weeks in e al ( = .81), bu only
modes con e gence wi h child en’s sel - epo s o de ense
mechanisms ( = .36; imma u e and neu o ic de enses we e
conside ed as a join ac o ).
Ea ly de elopmen al achie emen s. Ea ly de elopmen al
achie emen s (o delays) we e measu ed wi h pa en al
epo s. A child’s age o 2 mon hs (T2), pa en s epo ed he
eme gence o he child’s con ac smile (0 = no, 1 = yes), eye
con ac (0 = no, 1 = yes), and egula i y o ea ing and sleep-
ing hy hms (0 = no, 1 = yes). A he age o 12 mon hs (T3),
pa en s epo ed he child’s abili y o walk wi hou suppo (0
= no, 1 = yes), abili y o s and (0 = no, 1 = yes), and egula -
i y o sleeping hy hms (0 = no, 1 = yes). The six i ems we e
s anda dized and a e aged o o m a de elopmen al achie e-
men index. The eliabili y o his index was poo (α = .53),
indica ing ha he de elopmen al domains we e independen
o each o he . Ye , o ob ain a balanced assessmen o de el-
opmen al achie emen s du ing he i s yea , we decided o
use his a iable as a ough cumula i e index ( o a simila
app oach, see Appleya d, Egeland, an Dulmen, & S ou e,
2005). P o iding some alidi y o he index, we ound in
ou p e ious s udy a highly simila index o associa e nega-
i ely wi h bi h complica ions and poo neona al heal h
(e.g., low agpa sco es) (Punamäki e al., 2006).
Backg ound a iables. Backg ound a iables we e child’s
gende , mo he s’ age, ART s a us, pa en s’ a e age educa ion
le el (academic le el, college le el, oca ional aining,
basic educa ion/s uden ), and numbe o p e ious child en
(p imi- o mul ipa i y).
S a is ical Analyses
S a is ical analyses we e ca ied ou using s uc u al equa ion
modeling wi h he Mplus 5 p og am (Mu hén & Mu hén,
1998-2007) using maximum likelihood es ima ion wi h
obus s anda d e o s. This es ima ion me hod handles miss-
ing da a using ull in o ma ion maximum likelihood. The
o e all i o he models was e alua ed wi h he compa a i e
i index (CFI), he Tucke –Lewis index (TLI), he oo mean
squa e e o s o app oxima ion (RMSEA), and he chi-squa e
(χ2). As a c i e ion o accep able i , we used alues o >0.95
o CFI and TLI (Hu & Ben le , 1999), and <0.08 o RMSEA
(B owne & Cudeck, 1993). R-squa ed alues we e epo ed
o indica e absolu e (R2) and inc emen al (ΔR2) a iance
accoun ed by he independen a iables o e and abo e he
con ol a iables. Benjamini-Hochbe g p ocedu e was used
o p o ec signi icance le els agains alse posi i e disco e -
ies (Benjamini & Hochbe g, 1995). In all models, child’s
de elopmen al achie emen s and backg ound a iables we e
used as co a ia es.
The di e en amily p econdi ions hypo hesis was es ed
by eg essing he la en a iables o ea ly amily- ela ionship
quali y on emo ion egula ion and de ense mechanisms. To
examine whe he he quali y o amily ela ionships
dec eased o inc eased he child’s e icacy o emo ion egu-
la ion and eliance on he neu o ic and imma u e de enses,
eg ession coe icien s we e es ima ed sepa a ely o bo h
assessmen s a 2 mon hs (T1) and a 12 mon hs (T2).
To es he age-speci ic hypo hesis, age-speci ic models
we e buil sepa a ely o each amily- ela ionship dimension
p edic ing child en’s a ec egula ion, ha is, emo ion egu-
la ion, neu o ic de enses, and imma u e de enses ( o a con-
cep ual depic ion, see Figu e 1). Two c i e ia, adap ed om
Budescu (1993), we e used o compa e he ela i e impo -
ance o amily ela ionships a 2 mon hs and a 12 mon hs.
The age-speci ic p edic o is mo e impo an han ano he
p edic o i i bo h (a) explains a la ge p opo ion o he
dependen a iable when examined wi hou he ano he p e-
dic o and (b) explains a unique p opo ion o he dependen
a iable when sha ed a iance wi h ano he p edic o is
aken in o accoun . As shown in Figu e 1, he concu en
e ec s o amily ela ionships a he child’s age o 7 o 8
yea s (T3) we e con olled in all age-speci ic models.
Fi s , we used Akaike’s in o ma ion c i e ion (AIC;
Akaike, 1973) o es which o he wo age-speci ic models
(Model T1 o Model T2 in Figu e 1) explained a la ge p o-
po ion o emo ion egula ion and neu o ic and imma u e
de enses. A di e ence o ≥ |2.00| in AIC was used as a ule o
humb o indica e meaning ul signi icance in he explana o y
powe be ween he non-nes ed age-speci ic models (Bu nham
& Ande son, 2002). Nega i e alues indica e g ea e p edic-
i e powe o amily ela ionships a 2 mon hs (Model T1)
o e 12 mon hs (Model T2). Con e sely, posi i e alues indi-
ca e g ea e p edic i e powe o amily ela ionships a 12
mon hs (Model T2) o e 2 mon hs (Model T1).
Second, we used he Sa o a-Ben le adjus ed chi-squa e
es (Δχ2; Sa o a & Ben le , 2001) o es whe he each o he
age-speci ic model had unique p edic i e powe o e and
abo e he o he age-speci ic model. This was achie ed by
nes ed compa isons be ween he age-speci ic models (Model
T1 and Model T2 in Figu e 1) and he baseline model
(Models T1 and T2). To es he unique con ibu ion o 2
mon hs o e 12 mon hs, he i o Model T2 was es ed
agains ha o Models T1 and T2. Con e sely, o es he
Lindblom e al. 7
unique con ibu ion o amily ela ionships quali y a 12
mon hs o e 2 mon hs, he i o Model T1 was es ed agains
ha o Models T1 and T2. Poo e i o he age-speci ic (e.g.,
Model T1) model compa ed wi h ha o he baseline model
(Models T1 and T2) indica es unique p edic i e powe o he
excluded age-speci ic p edic o (e.g., 12 mon hs).
Bo h non-nes ed and nes ed compa isons o age-speci ic
models we e based on he o al i o he models because his
is no in luenced by he possible p oblem o mul icollinea i y
biasing indi idual pa h coe icien s (e.g., Ma sh, Dowson,
Pie sch, & Walke , 2004). In o he wo ds, he compa isons
e lec he combined o al e ec s o bo h pa en s’ epo s o
amily ela ionships.
Resul s
Measu emen and S uc u al Model
Family ela ionships. Figu es 2 o 5 p esen he measu emen
models o ma i al au onomy, pa en al au onomy, ma i al
in imacy, and pa en al in imacy. All models showed good i ,
and es s o longi udinal ac o ial in a iance con i med s a-
bili y o e ime, indica ing ha he la en cons uc s cap u ed
iden ical con en ac oss T1, T2, and T3. Howe e , we had o
exclude one indica o a iable (independen –dependen ) due
o low ac o loading om models o au onomy. Tes s o ac-
o ial in a iance be ween mo he s’ and a he s’ epo s
showed simila i y in epo s o pa en al au onomy, ma i al
au onomy, and ma i al in imacy, bu no in epo s o pa en-
al in imacy. The lack o in e pa en al ac o ial in a iance o
pa en al in imacy indica es ha mo he s and a he s pe -
cei ed he la en concep o pa en al in imacy di e en ly. In
all models, e o e ms we e co ela ed ac oss ime (T1-T2,
T2-T3, and T1-T3) wi hin each esponden (mo he o a he )
o con ol o i em- ela ed biases. These e o co ela ions
we e cons ained o be he same when his did no impai he
model i .
A ec egula ion. The measu emen model o he child’s
emo ion egula ion and de ense mechanisms, p esen ed in
Figu e 6, showed good i , CFI = .99; TLI = .99; RMSEA =
.00, 90% CI = [.00, .08]; χ2(3) = 2.83, p = .860. E icien
emo ion egula ion co ela ed nega i ely wi h use o bo h
neu o ic, = −.40, p < .001, and imma u e, = −.71, p <
.001, de enses. Only a ma ginally signi ican posi i e co e-
la ion was ound be ween imma u e and neu o ic de enses,
= .21, p = .070. The e was some luc ua ion o ac o load-
ings be ween mo he s (.54-.81) and a he s (.66-.78). Thus,
o ensu e in subsequen analyses ha bo h he a he s’ and he
mo he s’ epo s con ibu ed equally o a ec egula ion a i-
ables, he ac o loadings o indica o a iables we e ixed a
one. Despi e his echnical es ic ion, he esul ing model
showed good i , CFI = .99; TLI = .99; RMSEA = .02, 90%
CI = [.00, .09]; χ2(3) = 3.29, p = .907, and p ac ically epli-
ca ed he co ela ions be ween emo ion egula ion and
de ense mechanisms.
Equali y o s uc u al models be ween subg oups. Be o e es -
ing he esea ch hypo heses, we examined whe he modeling
should be done sepa a ely o he mo he s’ and a he s’
epo s o amily ela ionships, sepa a ely o amilies wi h a
boy o a gi l as he a ge child, o sepa a ely o amilies
wi h o wi hou e ili y ea men his o y (ART o NC). The
simila i y o he la en co ela ions, ha is, s uc u al equali y
assump ion, was es ed in models combining he amily ela-
ionships (Figu es 2-5) and a ec egula ion (Figu e 6). In
hese combined models, amily ela ionships a T1, T2, and
T3 we e allowed o co ela e wi h he child’s emo ion egula-
ion, neu o ic de enses, and imma u e de enses a T3.
Chi-squa e di e ence es s showed simila co ela ions
be ween he mo he s’ and a he s’ epo s o amily ela ion-
ships and emo ion egula ion, neu o ic de enses, and
imma u e de enses ega ding pa en al au onomy, χ2(9) =
12.76, p = .174; ma i al in imacy, χ2(9) = 10.02, p = .349;
and ma i al au onomy, χ2(9) = 9.76, p = .371. Howe e , he
Figu e 1. Concep ualiza ion o he models used o es age-speci ic e ec s o amily ela ionships on a ec egula ion.
No e. T1, T2, and T3 e e espec i ely o child’s ages o 2 mon hs, 12 mon hs, and 7 o 8 yea s. Family ela ionships a T3 (ma ked wi h dashed lines) a e
included in all age-speci ic models as a co a ia e. T = ime.
8 SAGE Open
co ela ions di e ed be ween mo he s’ and a he s’ epo s o
pa en al in imacy and emo ion egula ion, neu o ic de enses,
and imma u e de enses, χ2(9) = 17.06, p = .047. Thus, subse-
quen analyses we e done sepa a ely o mo he s’ and a he s’
epo s o pa en al in imacy.
The mul i-g oup compa ison be ween boys and gi ls
showed simila co ela ions o all amily ela ionships and
emo ion egula ion, neu o ic de enses, and imma u e
de enses: pa en al au onomy, χ2(18) = 16.64, p = .549;
pa en al in imacy, χ2(18) = 21.99, p = .233; ma i al au on-
omy, χ2(18) = 26.06, p = .100; ma i al in imacy, χ2(18) =
20.74, p = .293. Fu he mo e, he mul i-g oup compa ison
be ween ART and NC amilies showed simila co ela ions
o all amily ela ionships and emo ion egula ion, neu o ic
de enses, and imma u e de enses: pa en al au onomy, χ2(18)
= 8.82, p = .946; pa en al in imacy, χ2(18) = 24.29, p = .150;
ma i al au onomy, χ2(18) = 9.26, p = .987; ma i al in imacy,
χ2(18) = 18.36, p = .433. The e o e, hese subg oups we e
analyzed oge he in subsequen analyses.
E ec s o Backg ound Va iables and
De elopmen al Achie emen s
Be o e es ing ou esea ch hypo heses, we examined he
in luence o backg ound a iables and ea ly de elopmen al
achie emen s on a ec egula ion. The model showed ha
child en om mul ipa ous amilies had mo e e icien emo-
ion egula ion, B = −0.20, SE = 0.07, p = .006, and used
ewe neu o ic de enses, B = −0.20, SE = 0.07, p = .005, and
ewe imma u e de enses, B = −0.19, SE = 0.09, p = .024,
han child en in p imipa ous amilies. Boys used mo e
imma u e de enses han gi ls, B = 0.27, SE = 0.08, p = .001,
and he e was also a non-signi ican end o boys o ha e
poo e emo ion egula ion, B = 0.10, SE = 0.07, p = .054,
han gi ls. Highe ea ly de elopmen al achie emen s p e-
dic ed mo e e icien emo ion egula ion, B = −0.13, SE =
0.08, p = .001, and less use o imma u e de enses, B = −0.21,
SE = 0.10, p = .025. These a iables accoun ed o 7.0% o
he a iance o emo ion egula ion, 10.6% o imma u e
de enses, and 9.1% o neu o ic de enses. ART s a us, mo h-
e s’ age, and pa en s’ le el o educa ion did no p edic a ec
egula ion. The model had accep able i , CFI = .973; TLI =
.934; RMSEA = .03, 90% CI = [.02, .05]; χ2(21) = 36.45,
p = .019.
Family Rela ionships P edic ing Emo ion
Regula ion and De ense Mechanisms
Table 1 p esen s he eg ession coe icien s o each amily
ela ionship dimension sepa a ely p edic ing emo ion egula-
ion and de ense mechanisms. The esul s, o he mos pa ,
con i med ou di e en amily p econdi ions hypo hesis. As
Figu e 2. Measu emen model o mo he ’s and a he ’s epo s o ma i al au onomy.
No e. Mma1-mmal9 = mo he s’ epo s o ma i al au onomy; ma1- ma9 = a he s’ epo s o ma i al au onomy. E o co ela ions a e no shown. Model
i : χ2(129) = 200.861, p < .001, CFI = 0.98, TLI = 0.97, RMSEA = 0.03, 90% CI = [0.02, 0.04]. T = ime; CFI = compa a i e i index; TLI = Tucke –Lewis
index; RMSEA = oo mean squa e e o s o app oxima ion; CI = con idence in e al.
*p < .05. ***p < .001.
Lindblom e al. 9
hypo hesized, high le els o ma i al au onomy and pa en al
au onomy a he child’s ages o 2 and 12 mon hs p edic ed
child en’s e icien emo ion egula ion (column “Emo ion eg-
ula ion” in Table 1). Fu he , as hypo hesized, low le els o
ma i al au onomy and pa en al au onomy a he child’s ages o
2 and 12 mon hs p edic ed child en’s high eliance on neu o ic
de enses and imma u e de enses (columns “Neu o ic de enses”
and “Imma u e de enses” in Table 1). Finally, low ma i al in i-
macy a he child’s age o 12 mon hs p edic ed child en’s eli-
ance on imma u e de enses and on neu o ic de enses. Ma i al
au onomy accoun ed o an a e age o 4% o a ec egula ion,
pa en al au onomy accoun ed o an a e age o 6% o a ec
egula ion, and ma i al in imacy accoun ed o an a e age o
2% o a ec egula ion o e and abo e child en’s de elop-
men al achie emen s and backg ound a iables.
Table 1 ( ows wi h “Pa en al in imacy”) p esen s he eg es-
sion coe icien s o pa en al in imacy, analyzed sepa a ely o
mo he s’ and a he s’ epo s because ini ial analyses indica ed
s uc u al inequali y be ween he pa en s’ epo s. In line wi h
ou hypo heses, a he ’s epo s o low pa en al in imacy a he
child’s age o 12 mon hs p edic ed eliance on imma u e
de enses (accoun ing o 4% o he a iance o e and abo e
he con ol a iables). Howe e , agains ou hypo hesis, mo h-
e s’ o a he s’ epo s o pa en al in imacy a he child’s age o
2 mon hs did no p edic emo ion egula ion, neu o ic de enses,
o imma u e de enses.
Age Speci ici y in Family Rela ionships P edic ing
Emo ion Regula ion and De ense Mechanisms
Table 2 shows he esul s o non-nes ed and nes ed compa i-
sons o de e mine he ela i e impo ance o amily ela ion-
ships a he child’s ages o 2 mon hs and 12 mon hs in
p edic ing emo ion egula ion and de ense mechanisms ( o
a concep ual depic ion, see Figu e 1). In all age-speci ic
models, we con olled o he e ec s o concu en amily
ela ionships a he age o 7 o 8 yea s (T3), child en’s de el-
opmen al achie emen s, and backg ound a iables.
The esul s con i med he age-speci ic hypo hesis only
ega ding neu o ic de enses (column “Neu o ic de enses” in
Table 2). Non-nes ed compa isons showed ha pa en al
au onomy a 2 mon hs (T1) explained a la ge p opo ion o
neu o ic de enses, ΔAIC = –6.42, han pa en al au onomy a
12 mon hs (T2). Fu he , nes ed compa isons showed ha
pa en al au onomy a 2 mon hs (T1) explained a unique p o-
po ion o neu o ic de enses, Δχ2 = 15.52, p < .001, o e
pa en al au onomy a 12 mon hs (T2) and 7 o 8 yea s (T3).
Thus, we concluded ha pa en al au onomy a he age o 2
mon hs was mo e impo an p edic o o child en’s eliance
on neu o ic de enses han pa en al au onomy a he age o 12
mon hs.
Agains ou hypo heses, non-nes ed compa isons sug-
ges ed ha bo h ma i al au onomy, ΔAIC = 6.38, and ma i al
Figu e 3. Measu emen model o mo he ’s and a he ’s epo s o pa en al au onomy.
No e. Mpa1-mpa9 = mo he s’ epo s o pa en al au onomy; pa1- pa9 = a he s’ epo s o pa en al au onomy. E o co ela ions a e no shown. Model
i : χ2(127) = 164.61, p = .014, CFI = 0.99, TLI = 0.99, RMSEA = 0.02, 90% CI = [0.01, 0.03]. T = ime; CFI = compa a i e i index; TLI = Tucke –Lewis
index; RMSEA = oo mean squa e e o s o app oxima ion; CI = con idence in e al.
***p < .001.
16 SAGE Open
du ing ea ly childhood. Jou nal o Ea ly Childhood and In an
Psychology, 7, 1-24.
Du Roche Schudlich, T. D., Whi e, C. R., Fleischhaue , E. A., &
Fi zge ald, K. A. (2011). Obse ed in an eac ions du ing li e
in e pa en al con lic . Jou nal o Ma iage and Family, 73,
221-235. doi:10.1111/j.1741-3737.2010.00800.x
Dykas, M. J., & Cassidy, J. (2011). A achmen and he p ocessing
o social in o ma ion ac oss he li e span: Theo y and e idence.
Psychological Bulle in, 137, 19-46. doi:10.1037/a0021367
Eisenbe g, N., Spin ad, T. L., & Eggum, N. D. (2010). Emo ion-
ela ed sel - egula ion and i s ela ion o child en’s malad-
jus men . Annual Re iew o Clinical Psychology, 6, 495-525.
doi:10.1146/annu e .clinpsy.121208.131208
Ekas, N. V., B aunga -Rieke , J. M., Lickenb ock, D. M., Zen all,
S. R., & Maxwell, S. M. (2011). Toddle emo ion egula ion
wi h mo he s and a he s: Tempo al associa ions be ween
nega i e a ec and beha io al s a egies. In ancy, 16, 266-294.
doi:10.1111/j.1532-7078.2010.00042.x
Ellis on, D., McHale, J., Talbo , J., Pa mley, M., & Kue s en-
Hogan, R. (2008). Wi hd awal om copa en ing in e ac-
ions du ing ea ly in ancy. Family P ocess, 47, 481-499.
doi:10.1111/j.1545-5300.2008.00267.x
E ans, C. A., & Po e , C. L. (2009). The eme gence o mo he -
in an co- egula ion du ing he i s yea : Links o in an s’
de elopmen al s a us and a achmen . In an Beha io &
De elopmen , 32, 147-158. doi:10.1016/j.in beh.2008.12.005
Finge , B., Hans, S. L., Be ns ein, V. J., & Cox, S. M. (2009).
Pa en ela ionship quali y and in an -mo he a ach-
men . A achmen & Human De elopmen , 11, 285-306.
doi:10.1080/14616730902814960
Finzi-Do an, R., & Ka u, T. (2006). F om emo ional abuse in
childhood o psychopa hology in adul hood: A pa h media ed
by imma u e de ense mechanisms and sel -es eem. Jou nal o
Ne ous and Men al Disease, 194, 616-621. doi:10.1097/01.
nmd.0000230654.49933.23
F aibe g, S. (1982). Pa hological de enses in in ancy. The
Psychoanaly ic Qua e ly, 51, 612-635.
Ga azzi, S. M., McKen y, P. C., Jacobson, J. A., Julian, T. W., &
Lohman, B. (2000). Modeling he e ec s o exp essed emo-
ion, psychia ic symp omology, and ma i al quali y le els on
male and emale e bal agg ession. Jou nal o Ma iage and
Family, 62, 669-682. doi:10.1111/j.1741-3737.2000.00669.x
Ge ae s, E., D i schel, B., K eplin, U., Miyagawa, L., &
Wadding on, J. (2012). Reduced speci ici y o nega i e
au obiog aphical memo ies in ep essi e coping. Jou nal o
Beha io The apy and Expe imen al Psychia y, 43, S32-S36.
doi:10.1016/j.jb ep.2011.05.007
Ge gely, G., & Wa son, J. S. (1996). The social bio eedback heo y
o pa en al a ec -mi o ing: The de elopmen o emo ional
sel -awa eness and sel -con ol in in ancy. The In e na ional
Jou nal o Psychoanalysis, 77, 1181-1212.
G oss, J. J. (1998). The eme ging ield o emo ion egula ion: An
in eg a i e e iew. Re iew o Gene al Psychology, 2, 271-299.
doi:10.1037/1089-2680.2.3.271
G oss, J. J., Richa ds, J. M., & John, O. P. (2006). Emo ion eg-
ula ion in e e yday li e. In Emo ion egula ion in couples
and amilies: Pa hways o dys unc ion and heal h (pp. 13-
35). Washing on, DC: Ame ican Psychological Associa ion.
doi:10.1037/11468-001
G oss, J. J., & Thompson, R. A. (2007). Emo ion egula ion:
Concep ual ounda ions. In J. J. G oss (Ed.), Handbook o emo-
ion egula ion (pp. 3-24). New Yo k, NY: Guild o d P ess.
doi:10.1080/00140130600971135
Ha , J. (2014). Towa d an in eg a i e heo y o psychological
de ense. Pe spec i es on Psychological Science, 9, 19-39.
doi:10.1177/1745691613506018
He wig, I., Ka enbe ge , T., Jäncke, L., & B ühl, A. B. (2010).
Sel - ela ed awa eness and emo ion egula ion. Neu oImage,
50, 734-741. doi:10.1016/j.neu oimage.2009.12.089
Hoehl, S. (2014). Emo ion p ocessing in in ancy. In K. H. Laga u a
(Ed.), Child en and emo ion. New insigh s in o de elopmen-
al a ec i e sciences (pp. 1-12). Basel, Swi ze land: Ka ge .
doi:10.1159/000354346
Hu, L. T., & Ben le , P. M. (1999). Cu o c i e ia o i indexes
in co a iance s uc u e analysis: Con en ional c i e ia e sus
new al e na i es. S uc u al Equa ion Modeling, 6(1), 1-55.
doi:10.1080/10705519909540118
Ki sh, S., & Cassidy, J. (1997). P eschoole s’ a en ion o and mem-
o y o a achmen ele an in o ma ion. Child De elopmen ,
68, 1143-1153. doi:10.2307/1132297
Koole, S. L. (2009). The psychology o emo ion egula ion:
An in eg a i e e iew. Cogni ion & Emo ion, 23, 4-41.
doi:10.1080/02699930802619031
Kopp, C. B., & Neu eld, S. J. (2003). Emo ional de elopmen du -
ing in ancy. In R. J. Da idson, L. Sche e , & H. Goldsmi h
(Eds.), Handbook o a ec i e sciences (pp. 347-374). Ox o d,
UK: Ox o d Uni e si y P ess.
Lambie, B., & Ke in, L. (2003). In en ional a oidance and social
unde s anding in ep esso s and non ep esso s: Two unc ions
o emo ion expe ience? Consciousness and Emo ion, 4, 17-42.
doi:10.1075/ce.4.1.03lam
Leung, C. Y. Y., Mille , A. L., Lumeng, J. C., Kaci o i, N. A., &
Rosenblum, K. L. (2015). Ma e nal ep esen a ions o hei
child en in ela ion o eeding belie s and p ac ices among
low-income mo he s o young child en. Appe i e, 95, 176-181.
doi:10.1016/j.appe .2015.06.021
Lindblom, J., Pel ola, M. J., Vänskä, M., Hie anen, J. K., Laakso,
A., Tii inen, A., . . . Punamäki, R-L. (2015). Ea ly amily
sys em ypes p edic child en’s emo ional a en ion biases a
school age. In e na ional Jou nal o Beha io al De elopmen .
Ad ance online publica ion. doi:10.1177/0165025415620856
Lyons-Ru h, K. (1999). The wo-pe son unconscious: In e subjec i e
dialogue, enac i e ela ional ep esen a ion, and he eme -
gence o new o ms o ela ional o ganiza ion. Psychoanaly ic
Inqui y, 19, 576-617. doi:10.1080/07351699909534267
Ma sh, H. W., Dowson, M., Pie sch, J., & Walke , R. (2004).
Why mul icollinea i y ma e s: A eexamina ion o ela ions
be ween sel -e icacy, sel -concep , and achie emen . Jou nal
o Educa ional Psychology, 96, 518-522. doi:10.1037/0022-
0663.96.3.518
Ma eja , F., & Scholz, M. (1994). Das subjek i e Familienbild
(SFB) [Subjec i e Family Image]. Gö ingen, Ge many:
Hog e e.
Mauss, I. B., Bunge, S. A., & G oss, J. J. (2007). Au oma ic emo-
ion egula ion. Social & Pe sonali y Psychology Compass, 1,
146-167. doi:10.1111/j.1751-9004.2007.00005.x
Mauss, I. B., Cook, C. L., & G oss, J. J. (2007). Au oma ic
emo ion egula ion du ing ange p o oca ion. Jou nal o

Lindblom e al. 17
Expe imen al Social Psychology, 43, 698-711. doi:10.1016/j.
jesp.2006.07.003
Me z, E. C., McCall, R. B., W igh , A. J., & Luna, B. (2013).
Inhibi o y con ol and wo king memo y in pos -ins i u ional-
ized child en. Jou nal o Abno mal Child Psychology, 41, 879-
890. doi:10.1007/s10802-013-9737-9
Mo is, A. S., Silk, J. S., S einbe g, L., Mye s, S. S., & Robinson,
L. R. (2007). The ole o he amily con ex in he de elop-
men o emo ion egula ion. Social De elopmen , 16, 361-388.
doi:10.1111/j.1467-9507.2007.00389.x
Mou siana, C., Fea on, P., Mu ay, L., Coope , P., Goodye , I.,
Johns one, T., & Halligan, S. (2014). Making an e o o eel
posi i e: Insecu e a achmen in in ancy p edic s he neu al
unde pinnings o emo ion egula ion in adul hood. Jou nal o
Child Psychology and Psychia y, 55, 999-1008. doi:10.1111/
jcpp.12198
Mou siana, C., Johns oke, T., Mu ay, L., Fea on, P., Coope , P. J.,
Plia sikas, C., . . . Halligan, S. L. (2015). Insecu e a achmen
du ing in ancy p edic s g ea e amygdala olumes in ea ly
adul hood. Jou nal o Child Psychology and Psychia y, 56,
540-548. doi:10.1111/jcpp.12317
Mu hén, L. K., & Mu hén, B. O. (1998-2007). Mplus use ’s guide
(5 h ed.). Los Angeles, CA: Au ho .
Pech el, P., & Pizzagalli, D. A. (2011). E ec s o ea ly li e s ess
on cogni i e and a ec i e unc ion: An in eg a ed e iew
o human li e a u e. Psychopha macology, 214, 55-70.
doi:10.1007/s00213-010-2009-2
Punamäki, R-L., Repoka i, L., Vilska, S., Poikkeus, P., Tii inen, A.,
Sinkkonen, J., & Tulppala, M. (2006). Ma e nal men al heal h
and medical p edic o s o in an de elopmen and heal h p ob-
lems om p egnancy o one yea : Does o me in e ili y ma -
e ? In an Beha io and De elopmen , 29, 230-242.
Rankin-esque , L. A., Bu ne , C. K., Baucom, D. H., & Eps ein,
N. (1997). Au onomy and ela edness in ma i al unc ion-
ing. Jou nal o Ma i al & Family The apy, 23, 175-190.
doi:10.1111/j.1752-0606.1997. b00242.x
Rincón-Co és, M., & Sulli an, R. M. (2014). Ea ly li e auma and
a achmen : Immedia e and endu ing e ec s on neu obeha -
io al and s ess axis de elopmen . F on ie s in Endoc inology,
5, 1-15. doi:10.3389/ endo.2014.00033
Rydell, A., Be lin, L., & Bohlin, G. (2003). Emo ionali y, emo ion
egula ion, and adap a ion among 5- o 8-yea -old child en.
Emo ion, 3, 30-47. doi:10.1037/1528-3542.3.1.30
Sala, M. N., Tes a, S., Pons, F., & Molina, P. (2015). Emo ion
egula ion and de ense mechanisms. Jou nal o Indi idual
Di e ences, 36, 19-29. doi:10.1027/1614-0001/a000151
Sands om, M. J., & C ame , P. (2003). Gi ls’ use o de ense mech-
anisms ollowing pee ejec ion. Jou nal o Pe sonali y, 71,
605-627. doi:10.1111/1467-6494.7104004
Sa o a, A., & Ben le , P. M. (2001). A scaled di e ence chi-squa e
es s a is ic o momen s uc u e analysis. Psychome ika, 66,
507-514. doi:10.1007/BF02296192
Sha p, H., Hill, J., Hellie , J., & Pickles, A. (2015). Ma e nal an e-
na al anxie y, pos na al s oking and emo ional p oblems in
child en: Ou comes p edic ed om p e- and pos na al p o-
g amming hypo heses. Psychological Medicine, 45, 269-283.
doi:doi:10.1017/S0033291714001342.
Sokolowski, M. S., Hans, S. L., Be ns ein, V. J., & Cox, S. M. (2007).
Mo he s’ ep esen a ions o hei in an s and pa en ing beha io :
Associa ions wi h pe sonal and social-con ex ual a iables in a
high- isk sample. In an Men al Heal h Jou nal, 28, 344-365.
doi:10.1002/imhj.20140
S eine , H., A aujo, K. B., & Koopman, C. (2001). The esponse e al-
ua ion measu e (REM-71): A new ins umen o he measu e-
men o de enses in adul s and adolescen s. Ame ican Jou nal o
Psychia y, 158, 467-473. doi:10.1176/appi.ajp.158.3.467
S oud, C. B., Du bin, C. E., Wilson, S., & Mendelsohn, K. A.
(2011). Spillo e o iadic and dyadic sys ems in amilies wi h
young child en. Jou nal o Family Psychology, 25, 919-930.
doi:10.1037/a0025443
Subic-W ana, C., Beu el, M. E., B ähle , E., S öbel-Rich e , Y.,
Knebel, A., Lane, R. D., & Wil ink, J. (2014). How is emo-
ional awa eness ela ed o emo ion egula ion s a egies and
sel - epo ed nega i e a ec in he gene al popula ion? PloS
ONE, 9(3), e91846. doi:10.1371/jou nal.pone.0091846
Thompson, R. A., & Meye , S. (2007). Socializa ion o emo ion
egula ion in he amily. In J. J. G oss (Ed.), Handbook o emo-
ion egula ion (pp. 249-268). New Yo k, NY: Guild o d P ess.
To enham, N., & She idan, M. A. (2010). A e iew o ad e si y,
he amygdala and he hippocampus: A conside a ion o de el-
opmen al iming. F on ie s in Human Neu oscience, 3(68), 1-
18. doi:10.3389/neu o.09.068.2009
U sache, A., Blai , C., S i e , C., Voeg line, K, & Family Li e
P ojec In es iga o s. (2013). Emo ional eac i i y and egu-
la ion in in ancy in e ac o p edic execu i e unc ioning in
ea ly childhood. De elopmen al Psychology, 49, 127-137.
doi:10.1037/a0027728
Vaillan , G. E. (1971). Theo e ical hie a chy o adap i e ego mech-
anisms: A 30-yea ollow-up o 30 men selec ed o psycho-
logical heal h. A chi es o Gene al Psychia y, 24, 107-118.
doi:10.1001/a chpsyc.1971.01750080011003
Vaillan , G. E. (1995). The wisdom o he ego. Camb idge, MA:
Ha a d Uni e si y P ess.
Vaillan , G. E. (2000). Adap i e men al mechanisms: Thei ole
in posi i e psychology. Ame ican Psychologis , 55, 89-98.
doi:10.1037//0003-066X.55.1.89
on Hippel, W., & T i e s, R. (2011). The e olu ion and psychol-
ogy o sel -decep ion. The Beha io al and B ain Sciences, 34,
1-56. doi:10.1017/S0140525X10001354
Vänskä, M., Punamäki, R-L., Tol anen, A., Lindblom, J., Flyk ,
M., Unkila-Kallio, L., . . . Tulppala, M. (2011). Ma e nal p e-
and pos na al men al heal h ajec o ies and child men al heal h
and de elopmen : P ospec i e s udy in a no ma i e and o -
me ly in e ile sample. In e na ional Jou nal o Beha io al
De elopmen , 36(5), 517-531.
Walbu g, V., & Chia amello, S. (2015). Link be ween ea ly
maladap i e schemas and de ense mechanisms. Eu opean
Re iew o Applied Psychology, 65, 221-226. doi:10.1016/j.
e ap.2015.07.003
Yasno sky, J., A aujo, K., King, M., Mason, M., Pa elski, R., Shaw,
R., & S eine , H. (2003). De enses in school age child en’s e -
sus pa en s’ epo . Child Psychia y & Human De elopmen ,
33, 307-323. doi:10.1023/A:1023088313470
Zelazo, P. D., Qu, L., & Kesek, A. C. (2010). Ho execu i e unc-
ion: Emo ion and he de elopmen o cogni i e con ol. In S. D.
Calkins & M. A. Bell (Eds.), Child de elopmen a he in e sec-
ion o emo ion and cogni ion (pp. 97-111). Washing on, DC:
Ame ican Psychological Associa ion. doi:10.1037/12059-006
18 SAGE Open
Au ho Biog aphies
Jallu Lindblom, MA, is a uni e si y eache a he School o Social
Sciences / Psychology, Uni e si y o Tampe e, Finland. He is cu -
en ly inalizing his disse a ion abou ea ly amily en i onmen s
and hei signi icance on child en’s emo ional de elopmen .
Raija-Leena Punamäki, PhD, is p o esso o psychology a he
School o Social Sciences/Psychology, Uni e si y o Tampe e,
Finland. He esea ch opics a e child and adolescen de elopmen
and men al heal h, impac s o isk and auma ic en i onmen s and
psychosocial in e en ions.
Ma jo Flyk , PhD, psycho he apis , is pos -doc o al esea che a
he School o Social Sciences and Humani ies / Psychology,
Uni e si y o Tampe e, Finland. He esea ch is abou amilies a
psychosocial and medical isk.
Me i Vänskä, MA, is a esea che a he School o Social Sciences
/ Psychology, Uni e si y o Tampe e, Finland. She is cu en ly
inalizing he disse a ion abou pa en al p e- and pos na al men al
heal h and i s impac on child de elopmen .
Tapio Nummi, PhD, is an adjunc p o esso a he School o
In o ma ion Sciences, Uni e si y o Tampe e. His esea ch in e es
in s a is ics includes me hods o he analysis o longi udinal da a,
g ow h cu es, mixed models, nonpa ame ic- and mix u e
eg ession.
Ja i Sinkkonen, MD, is an adjunc p o esso o child psychia y a
he Uni e si y o Tu ku, Finland. He has w i en, edi ed, and co-
edi ed se en een books on child de elopmen . His esea ch in e -
es s include a achmen and adop ions.
Aila Tii inen, PhD, is a p o esso o Rep oduc i e Medicine a
Uni e si y o Helsinki and Head o Depa men a Helsinki
Uni e si y Cen al Hospi al.
Maija Tulppala, MD, is a gynecologis wi h a subspeciali y in
ep oduc i e endoc inology and a psycho he apis (indi idual,
couple and amily he apis ) a he Uni e si y o Helsinki,
Finland.He esea ch in e es is in he ield o ep oduc i e endo-
c inology and pa icula ly in he psychological aspec s o assis ed
ep oduc ion.