Content uploaded by Karin Ensink
Author content
All content in this area was uploaded by Karin Ensink on Jan 25, 2016
Content may be subject to copyright.
Intergenerational Pathways From Reflective Functioning to Infant
Attachment Through Parenting
Karin Ensink, Lina Normandin,
and André Plamondon
Université Laval
Nicolas Berthelot
Université du Québec Trois-Rivières
Peter Fonagy
University College London
The aim of this prospective study was to examine temporal pathways from mothers’ reflective function-
ing (RF) through parenting to infant attachment measured more than 16 months later. Participants were
88 mother–infant dyads from demographically diverse backgrounds and included a group of mothers
with histories of childhood maltreatment. RF was assessed using the RF rating of the Adult Attachment
Interview before the birth of the baby. Parenting was assessed when the infants were 6 months old using
the Maternal Sensitivity scale, as well as when they were 16 months using the Disconnected and
Extremely Insensitive Parenting scale. Infant attachment was assessed when the infants were 16 months
old using the Strange Situation. As hypothesised, the study findings showed that mothers’ mentalization
regarding their own early attachment relationships was associated with later parenting and infant
attachment. Negative parenting behaviours explained the link between mothers’ RF about their own
attachment relationships and infant attachment disorganization. The findings suggest that mothers’
mentalization about their early attachment relationships has important implications in the transition to
becoming parents themselves. Mentalization appears to be particularly important in helping mothers
screen and inhibit negative parenting behaviours that would otherwise undermine infant attachment
security and organization.
Keywords: reflective functioning, mentalization, maternal sensitivity, parenting, infant attachment
In a now classic study, Fonagy, Steele, Steele, Moran, and
Higgitt (1991) showed that mothers’ capacity to mentalize
about their own early attachment experiences, measured using
the Adult Attachment Interview (AAI), predicted infant attach-
ment security more than 16 months later. Based on these
findings, Fonagy and colleagues (1991) proposed that mental-
ization, the uniquely human capacity to be cognizant of one’s
own affects and imagine the feelings and thoughts of others,
plays an important role in the intergenerational transmission of
attachment. An intergenerational pattern is proposed in which
the parent’s mentalization and capacity to imagine what the
infant’s behaviour says about what they feel, need, and want,
underlies sensitive responding and promotes attachment secu-
rity (Fonagy & Target, 1997). The development of mentaliza-
tion in turn is seen as rooted in attachment, with infants dis-
covering their minds through being in a relationship with
someone who treats them as someone with a mind. However,
the findings of Fonagy and colleagues (Fonagy et al., 1991)of
an intergenerational pattern in which mothers’ mentalization
about their own early attachment relationships is linked to
infant attachment have not been replicated. Also surprisingly,
no study has examined the link between mothers’ mentalization
regarding their own attachment relationships and parenting, and
most research has focused on parental reflective functioning
(RF) about the child using the Parent Development Interview
(PDI; Slade, Grienenberger, Bernbach, Levy, & Locker, 2005).
Mothers’ mentalization about their own parents and attachment
relationships is likely to be particularly important in the tran-
sition to becoming parents themselves, and it would seem a
matter of priority to examine this empirically. The aim of this
study was to examine evidence for the theoretically assumed
intergenerational pathway from mothers’ RF regarding attach-
ment through parenting to infant attachment.
RF, Parenting, and Infant Attachment
RF is seen as underlying sensitive responding by helping moth-
ers to mentally put themselves in the place of the infant and
imagine the infant’s experience (Fonagy & Target, 1997). By the
same token, RF can be expected to play a role in inhibiting
Karin Ensink and Lina Normandin, École de psychologie, Université
Laval; André Plamondon, Dép. des fondements et pratiques en éducation,
Université Laval; Nicolas Berthelot, Département des Sciences Infirmières,
Université du Québec Trois-Rivières; Peter Fonagy, Research Department
of Clinical, Educational and Health Psychology, University College Lon-
don.
This research was supported in part by grants to Karin Ensink and Lina
Normandin from the Fonds québécois de la recherche sur la société et la
culture, from the Social Sciences and Humanities Research Council of
Canada, and from the Centre de recherche interdisciplinaire sur les
problèmes conjugaux et les agressions sexuelles.
Correspondence concerning this article should be addressed to Karin
Ensink, École de psychologie, 2325, rue des Bibliothèques, Québec, QC
Canada G1V 0A6. E-mail: karin.ensink@psy.ulaval.ca
Canadian Journal of Behavioural Science / Revue canadienne des sciences du comportement © 2016 Canadian Psychological Association
2016, Vol. 48, No. 1, 9 –18 0008-400X/16/$12.00 http://dx.doi.org/10.1037/cbs0000030
9
negative parenting by helping parents mentally step back and be
cognizant of their own affects and their impact on the infant. To
date, only two studies have examined the relationship between
parental RF about the child using the PDI, parenting, as well as
infant attachment. Grienenberger, Kelly, and Slade (2005) dem-
onstrated in a high-income sample (n ⫽ 48) that higher parental
RF was related to less disrupted affective communication in
interaction with their infants as assessed with the Atypical Mater-
nal Behaviour Instrument for Assessment and Classification
(AMBIANCE; Lyons-Ruth, Bronfman, & Parsons, 1999). These
parenting behaviours mediated the relationship between parental
RF and infant attachment, considered ordinally as secure, insecure,
and disorganized. More recently, Stacks et al. (2014), in a demo-
graphically diverse sample (n ⫽ 83) including mothers with his-
tories of abuse and neglect, found that parental RF about the child,
and not history of abuse, was associated with parenting sensitivity,
as well as negative parenting and infant attachment security. How-
ever, neither parental RF nor parenting was related to infant
attachment disorganization. Furthermore, the relationship between
parental RF and infant attachment security was mediated by par-
enting sensitivity, but a similar path was not found for disorgani-
zation. The findings of other studies in which parental RF and
parenting were examined are not all convergent. Schechter et al.
(2008) found no relationship between parental RF and atypical
behaviour in a referred sample of mothers exposed to violence, and
neither did Pajulo and colleagues (2012) in a sample of drug-
dependent mothers in treatment. Suchman, DeCoste, Leigh, and
Borelli (2010) also did not find a relationship between parental RF
and maternal sensitivity, but mentalization regarding self was
related to both sensitivity and socioemotional and cognitive
growth fostering. In addition, mothers’ RF about past trauma was
found to predict infant attachment disorganization (Berthelot et al.,
2015) in a sample in which all mothers had histories of abuse and
neglect. Although definite conclusions cannot be made based on
these studies, given that the majority had small sample sizes, it
does suggest that the links between other dimensions of RF,
parenting, and attachment should be considered.
Sensitive and Negative Maternal Behaviours and
Infant Attachment
In their seminal work on attachment assessment, Ainsworth,
Blehar, Waters, and Wall (1978) proposed that the mother’s
awareness of the infant’s communication and capacity to empathi-
cally identify and put herself “in the infant’s shoes” is central to
understanding the infant and responding in a well-timed, appro-
priate manner. Consistent with this, there is converging evidence
that mothers of secure infants are more responsive to infant’s
signals (Ainsworth et al., 1978; Isabella, 1993). By comparison,
the interactional antecedents of disorganization, characterised as a
“contradiction or an inhibition of action as it is being undertaken”
(Main & Hesse, 1990, p. 173), remain less clear (Beebe & Steele,
2013). Initially high rates of disorganized attachment were ob-
served in contexts of child abuse, and were linked to threatening
and fear inducing behaviours, thought to place the infant in an
impossible situation of fear without resolution, given that the only
potential source of comfort is also the source of fear (Hesse &
Main, 2000; Main & Hesse, 1990). Subsequently, it was found
that approximately 16% of infants in low-risk samples also
manifest disorganized attachment, and this was associated with
unresolved maternal states of mind regarding loss or trauma and
threatening, frightened, dissociative, deferential sexualized, and
disorganized/disorientated maternal behaviours (Main & Hesse,
1990; van IJzendoorn, 1995; Van IJzendoorn & Bakermans-
Kranenburg, 2006). Expanding on this system, Lyons-Ruth and
colleagues (Lyons-Ruth et al., 1999; Lyons-Ruth & Spielman,
2004) identified a range of atypical maternal behaviours that
contribute to disorganization, including affective communica-
tion errors, intrusiveness or negative behaviours, and with-
drawal (measured with the AMBIANCE). Disrupted affective
communication, for example, when the mother laughs when an
infant cries or fails to respond when he falls, predicted attach-
ment disorganization in three studies (Beebe et al., 2010;
Lyons-Ruth et al., 1999; Madigan, Moran, & Pederson, 2006),
suggesting that it may be as important as the intrusion of the
mother’s aggression, fear, and anxiety.
Reviewing the origins of disorganization, Bernier and Meins
(2008) conclude that both disconnected (i.e., threatening and atyp-
ical) and insensitive behaviours contribute to disorganization, and
that insensitive behaviours alone may predict disorganization es-
pecially in contexts with multiple risk factors, depending on the
individual susceptibility of children. Recently, Out, Bakermans-
Kranenberg, and Van Ijzendoorn (2009) developed the Discon-
nected and Extremely Insensitive Parenting (DIP) scale with the
explicit purpose of facilitating the assessment of both disconnected
behaviours and extremely insensitive behaviours considered to
represent complementary pathways to infant attachment disorga-
nization.
Histories of Abuse and Parenting
Parental history of abuse and neglect is a risk factor associated
with compromised parenting capacity (Koren-Karie, Oppenheim,
& Getzler-Yosef, 2008). However, there are few studies of
mother–infant interactions in samples in which mothers have ex-
perienced childhood abuse, and some findings are somewhat un-
expected. For example, Stacks and colleagues (2014) recently
failed to find a relationship between abuse and the quality of
parenting in interaction with infants. Abuse was not related to RF,
and although abuse was associated with posttraumatic stress dis-
order and depression, it was not related to the quality of mother–
infant interactions. This suggests that many pieces of the puzzle
regarding the relationship between history of abuse and parenting
remain unsolved.
The first aim of the current study was to examine relationships
between RF, parenting, and infant attachment. Because we were
interested in associations with personal risk factors such as child-
hood history of abuse, and demographic risk factors such as low
education, a relatively heterogeneous sample was recruited in
terms of the presence of history of abuse, education, and income.
To address this aim, RF was assessed prenatally, and then parent-
ing and attachment were assessed at 6 and 16 months postpartum.
We hypothesised that higher RF would be related to more sensitive
parenting and less parenting negativity, and to secure and orga-
nized infant attachment. Furthermore, we expected that abuse and
higher demographic risk would be associated with lower RF, lower
sensitivity, higher parental negative behaviours, and a higher prev-
alence of infant attachment disorganization.
10
ENSINK ET AL.
The second aim was to examine pathways from mothers’ men-
talization via parental behaviours to infant attachment security and
organization, controlling for personal and demographic risk fac-
tors. We predicted that whereas RF associated maternal sensitivity
would promote attachment security in the infant, low RF would be
associated with more disconnected and intrusive aggressive behav-
iours, which in turn would be associated with an increased risk of
attachment disorganization.
Method
Participants and Procedure
Mothers were between 21 and 43 years old (M ⫽ 30.82, SD ⫽
4.00), and 64% were primiparous mothers. All participants were
Caucasian, largely reflecting the ethnic composition of the French-
Canadian city in which the study was conducted. In terms of
marital status, 66% were in common-law relationships, 32% were
married, and 2% were single. Regarding education, 14% had high
school diplomas, 30% had college degrees, and 56% had univer-
sity degrees. In terms of annual income, 21% of families had
household incomes ranging from $30,000 to $60,000, 47% had
incomes ranging from $60,000 to $99,999, and 32% had incomes
over $100,000 annually. Infant gender was equally presented and
50% of the infants in the sample were male.
In terms of history of abuse, 30% of mothers reported histories
of physical, sexual, or emotional abuse: 11 reported physical
abuse; two reported sexual abuse; seven reported psychological
abuse; six reported both sexual and physical abuse; and one
reported sexual, physical, and psychological abuse.
Data reported here were collected as part of a larger longitudinal
study designed to examine the transition from pregnancy to ma-
ternity. A total of 115 mothers participated in the current study at
the prenatal assessment; 99 mother–infant dyads participated in the
laboratory visit when their infants were 6 months old, and 88
participate when the infants were 16 months old. Analysis in the
current study is based on the 88 mother–infant dyads who partic-
ipated in the 16-month assessment.
Pregnant women were recruited through flyers at community
health services, at gynecological and obstetric clinics, as well as at
commercial outlets for mothers and babies. Women interested in
participating in the study were screened via telephone for histories
of childhood maltreatment in order to recruit a ratio of approxi-
mately 30% of women exposed to childhood abuse so that the
sample would be broadly representative of the reported prevalence
of childhood maltreatment in the general population (Afifi et al.,
2011; Scher, Forde, McQuaid, & Stein, 2004). Exclusion criteria
included age below 19 years, abnormal pregnancy, major psycho-
sis, drug or alcohol abuse, and illiteracy or intellectual limitations.
Mothers with complications during the last phases of the preg-
nancy, and with infants born very prematurely or with severe
health problems, were excluded from the study; 16 dyads were
therefore excluded from the initial sample.
Measures and Procedure
During the third trimester of their pregnancy, women completed
the AAI (George, Kaplan, & Main, 1985) to assess their RF
regarding their early attachment experiences. They subsequently
returned to the laboratory with their infants twice. When the
infants were 6 months old, maternal sensitivity was assessed with
the Sensitivity Scale (Ainsworth, Bell, & Stayton, 1974) during
filmed mother–infant interactions in which mothers played with
their infants with and without toys, fed their infants, changed their
nappies, filled out a questionnaire, and engaged in an interview
regarding the infant. When the infants were 16 months old, moth-
ers and infants participated in the Strange Situation (SS) paradigm
(Ainsworth et al., 1978) to assess mother–infant attachment as
well as negative parenting using the DIP scale (Out et al., 2009).
Measures
RF coding of AAI. RF was rated from transcribed AAI’s
(George et al., 1985). The AAI is a semistructured interview
designed to elicit the participant’s current state of mind regarding
attachment experiences with parents and other significant caregiv-
ers during childhood. The interview includes questions about the
quality of childhood experiences with parents; the participant’s
responses to experiences of rejection, separation, loss, and trauma
during childhood; and the participant’s evaluation of the effects of
those childhood experiences on his or her current functioning.
The AAI was coded for RF using the RF manual (Fonagy,
Target, Steele, & Steele, 1998). The psychometric properties de-
tailed in the manual include high interrater reliability as well as
good discriminant validity and predictive validity across a number
of samples. RF is rated on a scale of ⫺1 to 9, with every score
representing a different level of mental state explanation. Ratings
of ⫺1 indicate an attack on mentalization. Ratings of 0 indicate a
refusal to engage in mentalization. Ratings of 1 indicate the ab-
sence of any recognition of mental states so that interpersonal
reactions are described only in behavioural terms, or individuals
are described only in terms of physical or global characteristics.
Ratings of 3 indicate limited ability to identify and acknowledge
simple mental states and affects at a very general level, but without
understanding how mental states function. Ratings of 5 indicate a
basic capacity to consider how mental states influence behaviour
and perceptions, and are in turn evoked in interpersonal interac-
tions. Higher ratings indicate increasingly full and sophisticated
mental state accounts of subjective experience underlying interac-
tions, and ratings of 9 are reserved for exceptional accounts that
reflect nuanced understanding of motivations underlying behav-
iour or unusual insights. RF is typically rated based on all AAI
questions that explicitly demand and appreciation of mental states
(e.g., “Why did your parents behave as they did during your
childhood?”). An overall RF score that represents respondents’
characteristic level of RF is derived based on the individual scores
using a decision algorithm outlined in the manual and that takes
into account the frequency of low scores or failures in mentaliza-
tion as well as the frequency of high scores.
Two graduate students who were trained by the study authors
rated the AAI protocols for RF. Interrater reliability was assessed
on 19 coding protocols and the intraclass correlation coefficient
(ICC) was high (ICC ⫽ .80), indicative of satisfactory interrater
reliability.
Assessment of reported history of abuse. Potential partici-
pants were asked screening questions regarding the presence of
childhood experiences of abuse during a telephone interview to
assure that the sample would be composed of approximately 30%
11
MATERNAL REFLECTIVE FUNCTIONING AND INFANT ATTACHMENT
of mothers with such histories. To further assess the presence of
reported histories of abuse, the information provided by the moth-
ers during the AAI in response to questions regarding past trau-
matic experiences was used. Abuse history was defined as the
mother’s report of abuse on the AAI. This procedure has been used
in previous studies and has been shown to be reliable in studies
using the AAI as well as other independent measures of childhood
abuse and neglect (Ensink, Berthelot, Bernazzani, Normandin, &
Fonagy, 2014).
SS procedure. The SS (Ainsworth et al., 1978) was used to
assess infant attachment security and organization when infants
were approximately 16 months old. The SS is widely used and its
validity has been demonstrated in major long-term studies (Sroufe,
Egeland, Carlson, & Collins, 2005). It consists of eight 3-min
episodes during which the mother leaves the infant (separation
episodes) and reunites with the infant (reunion episodes). Two
trained coders rated SSs for the three organized attachment clas-
sifications (secure, anxious-resistant-insecure, and anxious-
avoidant-insecure) using the criteria of Ainsworth et al. (1978),
and attachment disorganization using the criteria of Main and
Solomon (1990). In this sample, 50% were double coded with near
full agreement for the organized attachment classifications, and
97% agreement for infant attachment disorganization.
Maternal sensitivity. The Sensitivity Scale of Ainsworth et
al. (1974) was used to code maternal sensitivity during mother
infant interactions in the laboratory when the infants were 6
months old. Mothers were filmed while playing with their infants
with and without toys, breast- or bottle-fed their infants, changed
their nappies, filled out a questionnaire, and engaged in an inter-
view regarding the infant. The first 5 min of each activity was
coded, for a total of 30 min. The tasks were selected to be broadly
representative of a range of everyday contexts in which mothers
interact with their infants, with the last two contexts thought to
resemble more stressful everyday contexts in which the mother is
required to divide her attention and keep the infant in mind while
addressing the competing demands of another task or interacting
with someone else. Sensitivity was rated on a continuous scale
from 1 to 9, with a 9 representing a very sensitive mother. Four
dimensions of sensitivity are assessed, namely, (a) awareness of
the infant’s communication, (b) an accurate interpretation of the
signals, (c) appropriateness of reactions, and (d) rapidity of the
response to the infant’s signals.
Sensitivity was rated by three trained graduate students who
demonstrated high interrater reliability (ICCs from .90 to .98) on
25% of the sample. To deal with the negative skew in this variable,
scores were reflected and log transformed.
DIP scale. The DIP scale was developed to assess both dis-
connected behaviour and extremely insensitive behaviour. For
disconnected behaviour, all items from Main & Hesse’s (1998)
coding instrument for frightening, frightened, dissociated, sexual-
ized, and disorganized parental behaviour were adapted to assess
the following five dimensions of parental behaviour: (a) frighten-
ing and threatening behaviour; (b) behaviours indicating fear of the
child; (c) dissociative behaviours indicative of absorption (freezing
or handling as though the child is an inanimate object) or intrusion
of an altered state of awareness (such as inexplicable shifts in
mood or sudden fear regarding the environment); (d) interacting
with the child in a timid, submissive, and/or deferential manner,
sexualized/romantic behaviours; and (e) disoriented/disorganized
behaviours (such as contradictions in behaviour or vocalizations,
disorientation, and other anomalous movements and postures). For
disconnected behaviours, their sudden occurrence, lack of meta
signals indicating play or affection (such as smiling), and lack of
explanation for their occurrence are important considerations, and
scores are assigned for specific behaviours using clearly specified
criteria (Out et al., 2009).
For the second dimension, that of extreme parental insensitivity,
two types are rated separately: (a) parental withdrawal and neglect,
and (b) intrusive, negative, aggressive, or otherwise harsh parental
behaviours (Out et al., 2009). This dimension is an adaptation of
items from the AMBIANCE measure developed by Lyons-Ruth
and colleagues (1999). The AMBIANCE items referring to ex-
tremely insensitive behaviour were combined into more general
categories of parental behaviours. In the DIP scale, parental with-
drawal/neglect is scored when the parent (repeatedly) fails to show
responsive behaviour when the child is in distress, seeks contact,
or approaches the parent (Out et al., 2009). For example, the parent
does not respond to the child’s repeated vocalizations or ignores a
crying child, or does not intervene when the child engages in
dangerous behaviour. Physical intrusiveness is scored when the
parent gets too close to the child and intrusively overrides the
child’s cues (e.g., ignores the child’s evident interest in a particular
toy and attempt to physically move closer to the toy and engage
with it, by either removing the toy, introducing another toy, or
picking the child up and physically changing their direction or
position so they cannot engage with the toy), or engages in
interactions that are too intense and vigorous. The distinction
between extreme insensitivity and ordinary insensitive responses
lies in the duration, frequency, quality, and severity of the behav-
iours (e.g., aggressive behaviours), as well as the context in which
the behaviour occurs (e.g., when the child is in distress).
The DIP scale has been shown to have good psychometric
properties (Out et al., 2009). Discrete disconnected and extremely
insensitive behaviours were rated on a 9-point scale each time they
occurred. For all dimensions, a final score was assigned, equal to
the highest individual score or 1 point higher when the parental
behaviour was severe or occurred frequently. In the present study,
the DIP scale was rated by two graduate students who were trained
in consultation with the developers of the measure, and the ICC for
the global scores was high, both for disconnected (r ⫽ .92) and
insensitive (r ⫽ .79) behaviours.
We use the term negative maternal behaviours when referring to
insensitive behaviours as assessed with the DIP scale so as not to
contribute to confusion regarding the differences between sensi-
tivity and insensitivity
Analytic Plan
To address the first research objective, correlations between
variables were examined and regressions and t tests were used. To
address the second research objective, a path analysis was con-
ducted in Mplus 7.0 (L. K. Muthén & Muthén, 1998–2012). Using
this method, it is possible to test the hypothesis that RF is associ-
ated with outcomes through maternal behaviours while controlling
for factors that might account for these associations. The models
were estimated using a statistical method allowing for binary
outcomes, means, and variance adjusted weighted least squares
(B. O. Muthén, 2010). Regressions predicting the binary outcomes
12
ENSINK ET AL.
(i.e., secure/insecure [including disorganised] as well as organiza-
tion/disorganization) were conducted using the probit link. This
estimator was chosen because it can be used to derive indirect
effects when using binary outcomes (L. K. Muthén & Muthén,
1998–2012). The hypothesised indirect association between RF
and child outcomes (attachment organization and attachment se-
curity) through maternal behaviour was tested by investigating the
significance of the indirect pathways via both sensitive behaviours
and negative (intrusive/aggressive and withdrawing and neglect-
ful) behaviours simultaneously. Estimates of indirect effects were
obtained using bootstrapping, the gold standard for testing indirect
effects (Preacher & Hayes, 2008). An indirect effect is significant
if the 95% confidence interval (CI) does not include zero. Model
fit can be assessed using various indicators (Marsh, Hau, & Wen,
2004; Raykov & Marcoulides, 2006). Generally, the chi square
should be small and nonsignificant. The comparative fit index and
Tucker-Lewis index should be above .90, whereas the root mean
square error of approximation should be below .05.
Results
Preliminary Analysis
First we examined the distributions of infant attachment classi-
fications. The four-way distribution is reported in Table 1 and
indicates that of the 88 infants in the sample, 48 (55%) were
secure, 14 (16%) were avoidant, 10 (11%) were resistant, and 16
(18%) were disorganized.
The three-way distributions (secure, insecure, disorganized)
show that 48 (55%) infants were classified as secure, 24 (27%)
infants were classified insecure, and 16 (18%) infants were clas-
sified as showing disorganized attachment strategies. In Table 1,
we also report means of mothers’ RF, mean sensitivity scores, as
well as their mean scores on the disconnected and insensitive
scales of the DIB for each type of infant attachment. Although we
used the dimensional scores for disconnected and extremely in-
sensitive behaviours in the path analyses, using the criteria of Out
and colleagues (2009), 25% of the mothers in our sample were
classified as insensitive and 27% mothers were classified as dis-
connected.
Correlational analyses (see Table 2) show significant associa-
tions between maternal RF and both sensitivity and negative
parenting, but not disconnected behaviours. Disconnected behav-
iours were only associated with insensitive behaviours. Infant
attachment disorganization was associated with child sex, maternal
education, maternal RF, and maternal insensitive behaviours. In-
fant attachment security was associated with maternal education
and negative parenting.
To examine whether disconnected and extremely insensitive
behaviours were associated with attachment, regression analyses
using probit link were conducted. The results indicate that insen-
sitive behaviours predicted attachment organization and security
(b ⫽⫺.63, SE ⫽ .27, p ⬍ .05, and b ⫽⫺.49, SE ⫽ .21, p ⬍ .05,
respectively). No association was found between disconnected
behaviours and infant attachment organization or security
(b ⫽⫺.07, SE ⫽ .05, p ⬎ .05, and b ⫽⫺.07, SE ⫽ .06, p ⬎ .05,
respectively). For this reason, we only focus on insensitive behav-
iours in the subsequent analyses.
There were no significant differences between mothers with
histories of abuse and others in terms of RF, sensitivity, negative
parenting, or disconnected behaviours. Probit regressions also in-
dicated no association between maternal history of abuse and
infant attachment organization or security.
Path Analysis
We then fitted a path analysis in which RF was allowed to
predict both maternal behaviours and child attachment outcomes,
whereas maternal behaviours predicted child attachment outcomes.
A visual depiction of the model is presented in Figure 1.Inthe
model, covariates (maternal education, maternal history of abuse
and infant gender) were allowed to predict all variables in the
model. Because the model is saturated, there is no test of model fit.
Parameter estimates are reported in Tables 3 and 4. Maternal
sensitivity was predicted by maternal education, whereas maternal
insensitivity was only predicted by RF. RF was predicted by
maternal education. Infant attachment security was predicted by
maternal education as well as maternal insensitivity, whereas at-
tachment organization was predicted by infant gender (boys being
more likely to be disorganized; 28% of boys vs. 9% of girls) and
maternal insensitivity. The odds ratios indicate that an increase of
1 on the measure of negative behaviours increases the odds of
attachment insecurity by 2.22 and attachment disorganization by
2.56.
Next, we focused our attention on the results pertaining to the
indirect association between mothers’ RF and child attachment
outcomes via maternal behaviours. Although the results are the
same as those in Tables 3 and 4, we present them in Figure 1 for
Table 1
Maternal Reflective Functioning and Maternal Behaviours According to Child
Attachment Categories
Variable
Secure
n ⫽ 48 (55%)
Avoidant
n ⫽ 14 (16%)
Resistant/
Ambivalent
n ⫽ 10 (11%)
Disorganized
n ⫽ 16 (18%)
MSDMSDMSDMSD
Maternal RF 4.62 1.16 4.71 .99 4.60 1.51 3.94 1.34
Disconnected behaviours 1.98 2.46 1.79 2.58 3.50 3.14 3.06 2.98
Negative behaviours 2.11 2.27 2.29 2.34 3.80 2.53 4.31 2.39
Sensitivity 5.59 1.72 5.38 2.60 4.50 2.93 4.88 2.68
Note. RF ⫽ reflective functioning.
13
MATERNAL REFLECTIVE FUNCTIONING AND INFANT ATTACHMENT
the sake of clarity. RF was associated with maternal insensitivity
but not maternal sensitivity. In turn, maternal insensitivity pre-
dicted both attachment security and attachment organization, con-
sistent with potential indirect effects. We conducted a formal test
of these indirect effects by running the same model with 5,000
bootstrap samples. Results showed that the confidence intervals of
the indirect effects between RF and both attachment security and
attachment organization via maternal insensitivity did not include
zero, b ⫽ .06, 95% CI [.002, .18] and b ⫽ .07, 95% CI [.01, .21],
respectively. This indicates that the indirect effects are significant.
In contrast, the confidence intervals of the indirect effects between
RF and both attachment security and attachment organization via
maternal sensitivity included zero, b ⫽⫺.01, 95% CI [⫺1.60, .03]
and b ⫽ .00, 95% CI [-.09, .06], respectively. This indicates that
these indirect effects are not significant.
Discussion
The study findings showed that mothers RF about attachment
measured before the birth of the baby predicted their subsequent
sensitivity and negative behaviours (intrusive aggressive and with-
drawn). There was a pathway from mothers’ RF to later infant
attachment (both infant attachment security and infant attachment
disorganization) through mothers’ negative behaviours. These
findings extend that of Fonagy et al. (1991) that mothers’ RF (on
the AAI) was associated with infant attachment security, and show
that mother’s RF is also associated with infant attachment disor-
ganization. The findings that better mentalization is associated
with maternal sensitivity is in line with the theory that there is an
intergenerational pattern of transmission in which mothers’ RF,
developed in the context of their own early attachment relation-
ships, helps to make infant behaviour meaningful and underlies
sensitivity. In addition, our findings suggest that better mentaliza-
tion also helps mothers to screen their own negative intrusive,
aggressive, and withdrawn responses that undermine the develop-
ment of attachment security and organization. Mothers with higher
RF may be better at filtering their own affects of aggression,
anxiety, and fear, and screen their infants from negative behav-
iours, because they are more aware of their own affects and can see
themselves from the outside and imagine the infants’ distress.
Mentalization also implies an implicit understanding that affects
become less intense with time and can be changed through think-
ing and seeing situations differently, and this may help mothers
with higher RF tolerating difficult feelings. At the same time,
mothers who are not in touch with mental states may be more at
Table 2
Correlational Statistics for the Main Study Variables
Variable 1 2 3 4 5 6 7 8 MSD
1. Child gender (1 ⫽ boys, 2 ⫽ girls)
2. Abuse group (0 ⫽ no abuse, 1 ⫽ abuse) .05
3. Maternal education .12 ⫺.25
ⴱⴱ
4.51 1.17
4. Reflective functioning .05 ⫺.14 .25
ⴱⴱ
4.52 1.26
5. Negative parenting .09 ⫺.03 ⫺.18 ⫺.24
ⴱⴱ
1.60 1.66
6. Disconnected behaviours ⫺.01 ⫺.20 ⫺.06 ⫺.03 .22
ⴱⴱ
2.32 2.68
7. Sensitivity .04 ⫺.18
ⴱ
.40
ⴱⴱ
.26
ⴱⴱ
⫺.32
ⴱⴱ
⫺.08 5.92 1.42
8. Attachment organization (0 ⫽ disorganized, 1 ⫽ organized) .25
ⴱⴱ
⫺.11 .30
ⴱⴱ
.22
ⴱⴱ
⫺.27
ⴱⴱ
⫺.11 .17
9. Attachment security (0 ⫽ insecure, 1 ⫽ secure) .20
ⴱ
⫺.09 .32
ⴱⴱ
.10 ⫺.25
ⴱⴱ
⫺.10 .14 .52
ⴱⴱ
Note. Scores have been reflected and log transformed.
ⴱ
p ⬍ .10.
ⴱⴱ
p ⬍ .05.
Figure 1. Path analysis describing the associations between reflective functioning, maternal behavior, and child
attachment outcomes (attachment security and attachment disorganization). Covariates were allowed to predict
all variables in the model (parameter estimates are presented in Tables 3 and 4). Parameters are standardized,
except those predicting attachment security and attachment organization, which are unstandardized estimates
obtained using probit link.
ⴱ
p ⬍ .10.
ⴱⴱ
p ⬍ .05.
14
ENSINK ET AL.
risk of resorting to nonmentalizing, teleological actions in which
mental states are crudely translated into behaviours that ignore the
infant’s intentionality. Such behaviours may disrupt the infant’s
intentional actions, and disorientate the infant and undermine their
developing self-organization, whereas aggressive reactions and
failure to respond may have a disorganizing impact though in-
creasing anxious arousal.
We did not find an association between disconnected maternal
behaviours and infant attachment disorganization. The lack of
significant findings when comparing mothers of disorganized in-
fants and others appears to be due to the fact that both mothers of
insecure resistant/ambivalent infants had even higher rates of dis-
connected behaviours than mothers of disorganized infants. By
comparison, mothers of secure infants and insecure avoidant in-
fants displayed much lower rates of disconnected behaviours.
Disconnected behaviours were also not significantly related to RF,
which may in part be because the study lacked power. It may also
be because these behaviours—for example, sexualized behav-
iours—operate in a way that is too far outside conscious aware-
ness. Furthermore, disconnected behaviours have been shown to
be specifically linked to lack of resolution of loss and trauma,
involving dissociative processes (Hesse & Main, 1999), and fol-
lows that it is more likely to be associated with RF regarding
trauma and loss rather than RF in general (Berthelot et al., 2015).
In addition, contrary to expectation, reported maternal histories
of abuse were not associated with lower RF, or related to maternal
sensitivity or negativity, or associated with an increased likelihood
of their infant developing insecure or disorganized attachment.
This is consistent with the findings of the Stacks et al., (2014)
study, in which history of abuse was also not associated with
maternal RF about the child, parenting sensitivity or negativity, or
infant attachment organization. In the present study, RF and ma-
ternal education, rather than abuse, were related to maternal sen-
sitivity and maternal negative behaviours, as well as infant attach-
ment security and organization. This extends previous findings that
it is not trauma per se, but the capacity of the mother to mentalize
about her early experiences, that has implication for infant attach-
ment security (Fonagy, Steele, Steele, Higgitt, & Target, 1994)or
her capacity to resolve or mentalize trauma, that has implications
for infant attachment disorganization (Berthelot et al., 2015). Al-
though there is overwhelming evidence showing that childhood
experiences of abuse is a risk factor associated with compromised
parenting (Koren-Karie et al., 2008), poor mental health, height-
ened stress reactivity, and poor affect regulation (Heim, Shugart,
Craighead, & Nemeroff, 2010), there is also evidence of resilience
and individual and contextual protective factors (Dixon, Browne,
& Hamilton-Giachritsis, 2009; Jaffee et al., 2013). Our findings
and that of Stacks et al., (2014) suggest that in less stressful
contexts they may be able to use their resources and relationships
to limit the impact of trauma. This is also in line with the findings
in the study by Fonagy et al. (1994), in which mothers with
histories of deprivation, but who had high RF, had securely at-
tached infants. However, the way we assessed abuse in the current
study as present or absent, and using the AAI rather than an
validated measure of trauma, and the fact that we did not examine
the severity of abuse and the type of trauma because of the
relatively small subsample who reported abuse, may have contrib-
uted to the lack of significant findings regarding abuse. Further-
more, it is possible that mothers with histories of abuse experience
more difficulties later, for example, when confronted by the tod-
dler’s increasing strives for independence and self-assertion.
An unanticipated finding was that boys were at a significantly
elevated risk of developing disorganized attachment styles: 28% of
boys and 9% of girls in the sample displayed behaviours consid-
ered indicative of attachment disorganization during the SS. This
adds to some previous observations that boys may be at higher risk
for manifesting disorganized attachment patterns (Beebe & Steele,
2013; Lyons-Ruth et al., 1999; van IJzendoorn, 1995). However,
David and Lyons-Ruth (2005) have suggested that boys are more
likely to be classified as disorganized because they react with
Table 3
Parameter Estimates of the Associations Between Covariates
Included in the Path Analysis and Child Attachment Security
and Child Attachment Disorganization
Variable Estimate SE p value
Attachment organization regressed on
Child sex .84
ⴱⴱ
.38 .03
Maternal education .25 .17 .14
Abuse history ⫺.13 .47 .77
Reflective functioning .12 .14 .40
Negative parenting ⫺.59
ⴱⴱ
.25 .02
Sensitivity ⫺.05 .36 .89
Attachment security regressed on
Child sex .51
ⴱ
.27 .06
Maternal education .33 .14 .02
Abuse history ⫺.19 .32 .55
Reflective functioning ⫺.02 .12 .84
Negative parenting ⫺.51
ⴱⴱ
.19 .02
Sensitivity ⫺.13 .35 .71
Note. Parameter estimates are unstandardized regressions using probit
link. The regressions were estimated in the path analysis shown in Figure
1 but are reported here to avoid visual clutter. SE ⫽ standard error.
ⴱ
p ⬍ .10.
ⴱⴱ
p ⬍ .05.
Table 4
Parameter Estimates of the Associations Between Covariates
Included in the Path Analysis and Reflective Functioning
and Parenting
Variable Estimate SE p value
Sensitivity regressed on
Child sex ⫺.02 .19 .93
Maternal education .34
ⴱ
.10 .00
Abuse history ⫺.07 .10 .49
Reflective functioning .17
ⴱ
.09 .05
Negative parenting regressed on
Child sex .26 .21 .22
Maternal education ⫺.16 .11 .18
Abuse history ⫺.11 .12 .29
Reflective functioning ⫺.24
ⴱ
.10 .02
Reflective functioning regressed on
Child sex .13 .21 .56
Maternal education .23
ⴱ
.09 .01
Abuse history ⫺.07 .11 .50
Note. We report standardized estimates obtained using linear regressions.
The regressions were estimated in the path analysis shown in Figure 1 but
are reported here to avoid visual clutter. SE ⫽ standard error.
ⴱ
p ⬍ .05.
15
MATERNAL REFLECTIVE FUNCTIONING AND INFANT ATTACHMENT
fight-and-flight responses when mothers behave in frightened or
frightening ways, whereas girls show more “tend and befriend”
reactions. Further longitudinal research is needed to ascertain
whether such gender-based differences in response styles are as-
sociated with risk over time or contribute to misclassification.
With regard to mothers’ RF and infant attachment, the findings
of this study are broadly in line with that of Grienenberger and
colleagues (2005) and Berthelot et al. (2015) that RF was associ-
ated with infant attachment organization. However, Grienenberger
and colleagues found a significant association only when infant
attachment was considered ordinally as secure, insecure, and dis-
organized, whereas we found clear associations with both attach-
ment security and disorganization. Furthermore, whereas Grienen-
berger and colleagues assessed mothers’ mentalization about the
child using the PDI (Slade et al., 2005), we assessed mentalization
regarding the mother’s own early attachment relationships. It has
been suggested that mentalization about the child is more proximal
and directly relevant to the caregiving experience. However, our
findings show that current mentalization about past attachment
relationships and the parenting received is not necessarily more
distal, and may be particularly relevant when becoming a parent
for the mother, and has important implications for mother–infant
interaction, infant attachment security, and organization. Like
Grienenberger and colleagues, we also found evidence for medi-
ation, but in the present study, negative intrusive and withdrawn
behaviours mediated the relationship between mothers’ RF and
infant attachment disorganization, whereas in the Grienenberger
study, mediation was via affective communication errors. We did
not assess affective communication errors, and the different me-
diating mechanisms may also reflect sample differences. Our sam-
ple was more heterogeneous in terms of income, education, and
reported abuse than the small and highly educated group of moth-
ers in the Grienenberger et al. study, and in our population, we may
have been able to observe a wider range of aggressive and intru-
sive behaviours. Contrary to our findings, Berthelot et al. did not
report an association between RF about early attachment experi-
ence and infant secure or disorganized attachment in a sample of
abused and neglected mothers. However, they found that RF
regarding trauma assessed before the birth of the child predicted
infant disorganized attachment. These different findings may also
be related to the sample composition: Whereas the sample in our
study was composed of mothers from the community at low to
moderate risk, the mothers in the Berthelot et al. study all had
histories of trauma.
As predicted, mothers’ RF was also associated with sensitivity,
and this is in line with the findings of Stacks and colleagues
(2014), although we focused on RF about early attachment rela-
tionships, whereas they focused on RF about the child. We also
predicted that mothers’ RF, by helping to make infant behaviour
meaningful, would be linked to sensitivity, which would in turn
facilitate secure attachment. However, there was no evidence of a
path through sensitivity and later infant attachment security in our
sample, but this may be due to the fact that we assessed sensitivity
at 6 months rather than concurrently. Our findings are thus diver-
gent from that of Stacks et al., in which sensitivity, rather than
parental negativity, mediated the relationship between maternal RF
about attachment security, and in which they found no association
between RF and disorganization. Differences in the measures used
to assess parenting sensitivity and negativity may have contributed
to these divergent results, given that the measure used in the
present study to assess negative parenting behaviours, the DIP
scale, was developed especially to assess behaviours that have
previously been demonstrated to be linked with disorganization.
Both mother’s RF and sensitivity were associated with educa-
tion, in line with previous findings that maternal RF about the child
was associated with education (Pajulo et al., 2012) and that edu-
cation was positively associated with maternal sensitivity (Tamis-
LeMonda, Briggs, McClowry, & Snow, 2009; Wendland et al.,
2014), and negatively associated with extremely insensitive be-
haviours (Out et al., 2009). Education may be especially important
when children do not have access in their own families to rela-
tionships that foster the development of mentalization, and may be
associated with the development of RF indirectly through access to
relationships with teachers and peers outside the family that stim-
ulate the development of RF.
The study has several strengths, such as the inclusion of a group
of mothers who experienced childhood abuse, and the use of
objective ratings of maternal behaviours maternal RF, and infant
attachment. However, the findings should be interpreted the find-
ings should be interpreted in light of certain limitations. First, the
sample size is relatively small, and replication with a larger sample
is required before drawing definite conclusions regarding the as-
sociations between RF, maternal behaviours, and infant attach-
ment. Second, a weakness of the study is the fact that negative
parenting was measured at the same time as infant attachment,
whereas maternal sensitivity was measured when the infants were
6 months old, and may have contributed to finding that maternal
insensitivity was relatively more important for infant attachment.
The lack of significant associations between maternal histories of
abuse and disconnected and insensitive behaviours was an unex-
pected finding, but replication with different and larger samples is
necessary to investigate the implications of different types of abuse
and abuse severity. Nearly all the mothers in the study were in
common law relationships or married, and this is likely to have
made an important contribution that was not considered in the
present study. Further research is also needed to examine the
infant’s contribution to the interaction. Although more attention
has traditionally been paid to the mother’s role, there is evidence
suggesting that infants also have an important role in maintaining
or improving the quality of the interaction, for example, by elic-
iting more positive responses from the mother (Kivijärvi et al.,
2001). This may be due to genetic factors (Raby et al., 2012) and
neurobiological differences (Tharner et al., 2011) that have been
shown to increase the risk of attachment disorganization and which
have not been assessed in the present study.
Conclusion
This study provides new evidence of a path from mothers’ RF
about their own early attachment relationships to infant attachment
through parenting. This offers support for a model of intergenera-
tional transmission in which mothers’ quality of mentalization in
relation to early attachment relationships have implications for
their ability to filter negative behaviours and develop and promote
relationships that foster attachment security and organization.
These findings further underscore the importance of a mentalizing
stance for parenting. We propose that in the process of becoming
a mother, it is not just the parent that the mother wishes to be that
16
ENSINK ET AL.
is important but also the quality of mentalizing about her own early
attachment relationships. The mother’s ability to access her expe-
riences in mental state terms, to think about how it has been or
should have been, also promotes her capacity to monitor her
behaviour and its impact on the infant. Difficulties in adopting a
mentalizing stance regarding past and present relationships might
explain why mothers who manifest negative behaviours appear
unaware of this even when it is evident to observers. In sum, the
study provides evidence that mothers’ mentalization about attach-
ment relationships has important implications for their capacity to
establish relationships in which infants feel safe to explore and
seek comfort.
Résumé
L’objectif de cette étude prospective était d’examiner le chemine-
ment dans le temps des mères, au moyen de leur fonctionnement
réflexif (FR), du parentage et de l’attachement de l’enfant, évalué
plus de 16 mois plus tard. Les participants étaient 88 dyades
mère-bébé, issues de milieux démographiques diversifiés, parmi
lesquelles certaines mères avaient des antécédents de maltraitance
durant l’enfance. Le FR a été établi au moyen de l’Adult Attach-
ment Interview, avant la naissance du bébé. Le parentage a été
évalué lorsque les bébés avaient 6 mois, au moyen de l’échelle de
la sensibilité maternelle, puis de nouveau a
`
16 mois, au moyen de
l’échelle Disconnected and Extremely Insensitive Parenting.
L’attachement de l’enfant a été évalué a
`
16 mois, au moyen de la
situation étrangère. Conformément a
`
l’hypothèse de départ, les
résultats ont révélé que la mentalisation des mères au sujet de leurs
propres relations d’attachement durant l’enfance était associée a
`
leur type de parentage et a
`
l’attachement de l’enfant. Les com-
portements parentaux négatifs expliquaient le lien entre le FR des
mères au sujet de leurs propres relations d’attachement et a
`
la fois
l’attachement désorganisé et l’attachement non sécure de l’enfant.
Les résultats suggèrent que la mentalisation des mères au sujet de
leurs propres relations d’attachement durant leur enfance avait
d’importantes répercussions au cours de leur transition vers le rôle
de parents. La mentalisation semble particulièrement aider les
mères a
`
évaluer et a
`
éviter des comportements parentaux négatifs,
qui nuiraient a
`
l’attachement sécure de l’enfant et a
`
son organisa-
tion.
Mots-clés : fonctionnement réflexif, mentalisation, sensibilité ma-
ternelle, parentage, attachement de l’enfant.
References
Afifi, T. O., Mather, A., Boman, J., Fleisher, W., Enns, M. W., Macmillan,
H., & Sareen, J. (2011). Childhood adversity and personality disorders:
Results from a nationally representative population-based study. Journal
of Psychiatric Research, 45, 814 – 822. http://dx.doi.org/10.1016/j
.jpsychires.2010.11.008
Ainsworth, M. D. S., Bell, S. M., & Stayton, D. J. (1974). Infant-mother
attachment and social development: “Socialization” as a product of
reciprocal responsiveness to signal. In M. P. M. Richards (Ed.), The
integration of a child into a social world (pp. 99–135). London, UK:
Cambridge University Press.
Ainsworth, M. D. S., Blehar, M., Waters, E., & Wall, S. (1978). Patterns
of attachment: A psychological study of the strange situation. Hillsdale,
NJ: Erlbaum.
Beebe, B., Jaffe, J., Markese, S., Buck, K., Chen, H., Cohen, P.,...
Feldstein, S. (2010). The origins of 12-month attachment: A microanal-
ysis of 4-month mother-infant interaction. Attachment & Human Devel-
opment, 12, 3–141. http://dx.doi.org/10.1080/14616730903338985
Beebe, B., & Steele, M. (2013). How does microanalysis of mother-infant
communication inform maternal sensitivity and infant attachment? At-
tachment & Human Development, 15, 583–602. http://dx.doi.org/10
.1080/14616734.2013.841050
Bernier, A., & Meins, E. (2008). A threshold approach to understanding the
origins of attachment disorganization. Developmental Psychology, 44,
969–982. http://dx.doi.org/10.1037/0012-1649.44.4.969
Berthelot, N., Ensink, K., Bernazzani, O., Normandin, L., Luyten, P., &
Fonagy, P. (2015). Intergenerational transmission of attachment in
abused and neglected mothers: The role of trauma-specific reflective
functioning. Infant Mental Health Journal, 36, 200 –212. http://dx.doi
.org/10.1002/imhj.21499
David, D. H., & Lyons-Ruth, K. (2005). differential attachment responses
of male and female infants to frightening maternal behavior: Tend or
befriend versus fight or flight? Infant Mental Health Journal, 21, 1–18.
http://dx.doi.org/10.1002/imhj.20033
Dixon, L., Browne, K., & Hamilton-Giachritsis, C. (2009). Patterns of risk
and protective factors in the intergenerational cycle of maltreatment.
Journal of Family Violence, 24, 111–122. http://dx.doi.org/10.1007/
s10896-008-9215-2
Ensink, K., Berthelot, N., Bernazzani, O., Normandin, L., & Fonagy, P.
(2014). Another step closer to measuring the ghosts in the nursery:
Preliminary validation of the Trauma Reflective Functioning Scale.
Frontiers in Psychology, 5, 1471.
Fonagy, P., Steele, M., Steele, H., Higgitt, A., & Target, M. (1994). The
Emanuel Miller Memorial Lecture 1992. The theory and practice of
resilience. Journal of Child Psychology and Psychiatry, and Allied
Disciplines, 35, 231–257. http://dx.doi.org/10.1111/j.1469-7610.1994
.tb01160.x
Fonagy, P., Steele, M., Steele, H., Moran, G. S., & Higgitt, A. C. (1991).
The capacity for understanding mental states: The reflective self in
parent and child and its significance for security of attachment. Infant
Mental Health Journal, 12, 201–218. http://dx.doi.org/10.1002/1097-
0355(199123)12:3⬍201::AID-IMHJ2280120307⬎3.0.CO;2-7
Fonagy, P., & Target, M. (1997). Attachment and reflective function: Their
role in self-organization. Development and Psychopathology, 9, 679–
700. http://dx.doi.org/10.1017/S0954579497001399
Fonagy, P., Target, M., Steele, H., & Steele, M. (1998). Reflective-
functioning manual, version 5.0, for application to adult attachment
interviews. London, UK: University College London.
George, C., Kaplan, M., & Main, M. (1985). Adult Attachment Interview.
Berkeley, CA: University of California.
Grienenberger, J. F., Kelly, K., & Slade, A. (2005). Maternal reflective
functioning, mother-infant affective communication, and infant attach-
ment: Exploring the link between mental states and observed caregiving
behavior in the intergenerational transmission of attachment. Attachment
& Human Development, 7, 299–311. http://dx.doi.org/10.1080/
14616730500245963
Heim, C., Shugart, M., Craighead, W. E., & Nemeroff, C. B. (2010).
Neurobiological and psychiatric consequences of child abuse and ne-
glect. Developmental Psychobiology, 52, 671–690. http://dx.doi.org/10
.1002/dev.20494
Hesse, E., & Main, M. (1999). Second-generation effects of unresolved
trauma in nonmaltreating parents: Dissociated, frightened and threaten-
ing parental behavior. Psychoanalytic Inquiry, 19, 481–540. http://dx
.doi.org/10.1080/07351699909534265
Hesse, E., & Main, M. (2000). Disorganized infant, child, and adult
attachment: Collapse in behavioral and attentional strategies. Journal of
the American Psychoanalytic Association, 48, 1097–1127. http://dx.doi
.org/10.1177/00030651000480041101
17
MATERNAL REFLECTIVE FUNCTIONING AND INFANT ATTACHMENT
Isabella, R. A. (1993). Origins of attachment: Maternal interactive behavior
across the first year. Child Development, 64, 605–621. http://dx.doi.org/
10.2307/1131272
Jaffee, S. R., Bowes, L., Ouellet-Morin, I., Fisher, H. L., Moffitt, T. E.,
Merrick, M. T., & Arseneault, L. (2013). Safe, stable, nurturing rela-
tionships break the intergenerational cycle of abuse: A prospective
nationally representative cohort of children in the United Kingdom.
Journal of Adolescent Health, 53(4, Suppl.), S4–S10. http://dx.doi.org/
10.1016/j.jadohealth.2013.04.007
Kivijärvi, M., Voeten, M. J. M., Niemelä, P., Räihä, H., Lertola, K., &
Piha, J. (2001). Maternal sensitivity behavior and infant behavior in
early interaction. Infant Mental Health Journal, 22, 627– 640. http://dx
.doi.org/10.1002/imhj.1023
Koren-Karie, N., Oppenheim, D., & Getzler-Yosef, R. (2008). Shaping
children’s internal working models through mother-child dialogues: The
importance of resolving past maternal trauma. Attachment & Human
Development, 10, 465–483. http://dx.doi.org/10.1080/146167
30802461482
Lyons-Ruth, K., Bronfman, E., & Parsons, E. (1999). Chapter IV. Maternal
frightened, frightening, or atypical behavior and disorganized infant
attachment patterns. Monographs of the Society for Research in Child
Development, 64, 67–96. http://dx.doi.org/10.1111/1540-5834.00034
Lyons-Ruth, K., & Spielman, E. (2004). Disorganized infant attachment
strategies and helpless-fearful profiles of parenting: Integrating attach-
ment research with clinical intervention. Infant Mental Health Journal,
25, 318–335. http://dx.doi.org/10.1002/imhj.20008
Madigan, S., Moran, G., & Pederson, D. R. (2006). Unresolved states of
mind, disorganized attachment relationships, and disrupted interactions
of adolescent mothers and their infants. Developmental Psychology, 42,
293–304. http://dx.doi.org/10.1037/0012-1649.42.2.293
Main, M., & Hesse, E. (1990). Parents’ unresolved traumatic experiences
are related to infant disorganized attachment status: Is frightened and/or
frightening parental behavior the linking mechanism? In M. T. Green-
berg, D. Cicchetti, & E. M. Cummings (Eds.), Attachment in the pre-
school years: Theory, research, and intervention (pp. 161–182). Chi-
cago, IL: University of Chicago Press.
Main, M., & Hesse, E. (1998). Frightening, frightened, dissociated, def-
erential, sexualized and disorganized parental behavior: A coding sys-
tem for parent–infant interactions (6th ed.). Berkeley, CA: University of
California at Berkeley.
Main, M., & Solomon, J. (1990). Procedures for identifying disorganized/
disoriented infants during the Ainsworth Strange Situation. In M. Green-
berg, D. Cicchetti, & M. Cummings (Eds.), Attachment in the preschool
years (pp. 121–160). Chicago, IL: University of Chicago Press.
Marsh, H. W., Hau, K.-T., & Wen, Z. (2004). In search of golden rules:
Comment on hypothesis-testing approaches to setting cutoff values for
fit indexes and dangers in overgeneralizing Hu and Bentler’s (1999).
findings. Structural Equation Modeling, 11, 320–341. http://dx.doi.org/
10.1207/s15328007sem1103_2
Muthén, B. O. (2010). Bayesian analysis in Mplus: A brief intro-
duction. Retrieved from http://www.statmodel.com/download/
IntroBayesVersion%203.pdf
Muthén, L. K., & Muthén, B. O. (1998–2012). Mplus user’s guide (7th
ed.). Los Angeles, CA: Author.
Out, D., Bakermans-Kranenburg, M. J., & Van Ijzendoorn, M. H. (2009).
The role of disconnected and extremely insensitive parenting in the
development of disorganized attachment: Validation of a new measure.
Attachment & Human Development, 11, 419–443. http://dx.doi.org/10
.1080/14616730903132289
Pajulo, M., Pyykkönen, N., Kalland, M., Sinkkonen, J., Helenius, H.,
Punamäki, R. L., & Suchman, N. (2012). Substance-abusing mothers in
residential treatment with their babies: Importance of pre- and postnatal
maternal reflective functioning. Infant Mental Health Journal, 33, 70–
81. http://dx.doi.org/10.1002/imhj.20342
Preacher, K. J., & Hayes, A. F. (2008). Asymptotic and resampling
strategies for assessing and comparing indirect effects in multiple me-
diator models. Behavior Research Methods, 40, 879 –891. http://dx.doi
.org/10.3758/BRM.40.3.879
Raby, K. L., Cicchetti, D., Carlson, E. A., Cutuli, J. J., Englund, M. M., &
Egeland, B. (2012). Genetic and caregiving-based contributions to infant
attachment: Unique associations with distress reactivity and attachment
security. Psychological Science, 23, 1016–1023. http://dx.doi.org/10
.1177/0956797612438265
Raykov, T., & Marcoulides, G. A. (2006). A first course in structural
equation modeling. Mahwah, NJ: Erlbaum.
Schechter, D. S., Coates, S. W., Kaminer, T., Coots, T., Zeanah, C. H., Jr.,
Davies, M.,...Myers, M. M. (2008). Distorted maternal mental
representations and atypical behavior in a clinical sample of violence-
exposed mothers and their toddlers. Journal of Trauma & Dissociation,
9, 123–147. http://dx.doi.org/10.1080/15299730802045666
Scher, C. D., Forde, D. R., McQuaid, J. R., & Stein, M. B. (2004).
Prevalence and demographic correlates of childhood maltreatment in an
adult community sample. Child Abuse & Neglect, 28, 167–180. http://
dx.doi.org/10.1016/j.chiabu.2003.09.012
Slade, A., Grienenberger, J., Bernbach, E., Levy, D., & Locker, A. (2005).
Maternal reflective functioning, attachment, and the transmission gap: A
preliminary study. Attachment & Human Development, 7, 283–298.
http://dx.doi.org/10.1080/14616730500245880
Sroufe, L. A., Egeland, B., Carlson, E., & Collins, W. A. (2005). The
development of the person: The Minnesota study of risk and adaptation
from birth to adulthood. New York, NY: Guilford Press.
Stacks, A. M., Muzik, M., Wong, K., Beeghly, M., Huth-Bocks, A., Irwin,
J. L., & Rosenblum, K. L. (2014). Maternal reflective functioning among
mothers with childhood maltreatment histories: Links to sensitive par-
enting and infant attachment security. Attachment & Human Develop-
ment, 16, 515–533. http://dx.doi.org/10.1080/14616734.2014.935452
Suchman, N. E., DeCoste, C., Leigh, D., & Borelli, J. (2010). Reflective
functioning in mothers with drug use disorders: Implications for dyadic
interactions with infants and toddlers. Attachment & Human Develop-
ment, 12, 567–585. http://dx.doi.org/10.1080/14616734.2010.501988
Tamis-LeMonda, C. S., Briggs, R. D., McClowry, S. G., & Snow, D. L.
(2009). Maternal control and sensitivity, child gender, and maternal
education in relation to children’s behavioral outcomes in African Amer-
ican families. Journal of Applied Developmental Psychology, 30, 321–
331. http://dx.doi.org/10.1016/j.appdev.2008.12.018
Tharner, A., Herba, C. M., Luijk, M. P., van Ijzendoorn, M. H.,
Bakermans-Kranenburg, M. J., Govaert, P. P.,...Tiemeier, H. (2011).
Subcortical structures and the neurobiology of infant attachment disor-
ganization: A longitudinal ultrasound imaging study. Social Neurosci-
ence, 6, 336 –347. http://dx.doi.org/10.1080/17470919.2010.538219
van IJzendoorn, M. H. (1995). Adult attachment representations, parental
responsiveness, and infant attachment: A meta-analysis on the predictive
validity of the Adult Attachment Interview. Psychological Bulletin, 117,
387–403. http://dx.doi.org/10.1037/0033-2909.117.3.387
Van IJzendoorn, M. H., & Bakermans-Kranenburg, M. J. (2006). DRD4
7-repeat polymorphism moderates the association between maternal
unresolved loss or trauma and infant disorganization. Attachment &
Human Development, 8, 291–307. http://dx.doi.org/10.1080/
14616730601048159
Wendland, B. E., Atkinson, L., Steiner, M., Fleming, A. S., Pencharz, P.,
Moss, E.,...MAVAN Study Team. (2014). Low maternal sensitivity at
6 months of age predicts higher BMI in 48 month old girls but not boys.
Appetite, 82, 97–102. http://dx.doi.org/10.1016/j.appet.2014.07.012
Received March 3, 2015
Revision received October 5, 2015
Accepted October 6, 2015 䡲
18
ENSINK ET AL.
A preview of this full-text is provided by American Psychological Association.
Content available from Canadian Journal of Behavioural Science/Revue canadienne des sciences du comportement
This content is subject to copyright. Terms and conditions apply.
