This article was downloaded by: [85.241.28.

67]
On: 19 February 2013, At: 06:49
Publisher: Routledge
Informa Ltd Registered in England and Wales Registered Number: 1072954 Registered
office: Mortimer House, 37-41 Mortimer Street, London W1T 3JH, UK

Attachment & Human Development

Publication details, including instructions for authors and
subscription information:
http://www.tandfonline.com/loi/rahd20

Continuity and discontinuity of

attachment patterns: A short-term
longitudinal pilot study using a sample
of late-adopted children and their
adoptive mothers
a a a
C.S. Pace , G.C. Zavattini & M. D'Alessio
a
Department of Dynamic and Clinical Psychology, Sapienza
University of Rome, Italy
Version of record first published: 23 Dec 2011.

To cite this article: C.S. Pace , G.C. Zavattini & M. D'Alessio (2012): Continuity and discontinuity of
attachment patterns: A short-term longitudinal pilot study using a sample of late-adopted children
and their adoptive mothers, Attachment & Human Development, 14:1, 45-61

To link to this article: http://dx.doi.org/10.1080/14616734.2012.636658

PLEASE SCROLL DOWN FOR ARTICLE

Full terms and conditions of use: http://www.tandfonline.com/page/terms-and-

conditions

This article may be used for research, teaching, and private study purposes. Any
substantial or systematic reproduction, redistribution, reselling, loan, sub-licensing,
systematic supply, or distribution in any form to anyone is expressly forbidden.
The publisher does not give any warranty express or implied or make any representation
that the contents will be complete or accurate or up to date. The accuracy of any
instructions, formulae, and drug doses should be independently verified with primary
sources. The publisher shall not be liable for any loss, actions, claims, proceedings,
demand, or costs or damages whatsoever or howsoever caused arising directly or
indirectly in connection with or arising out of the use of this material.
 Attachment & Human Development
Vol. 14, No. 1, January 2012, 45–61

Continuity and discontinuity of attachment patterns: A short-term

longitudinal pilot study using a sample of late-adopted children
and their adoptive mothers
C.S. Pace, G.C. Zavattini* and M. D’Alessio

Department of Dynamic and Clinical Psychology, Sapienza University of Rome, Italy

(Received 15 July 2010; final version received 20 July 2011)

This study analysed the attachment patterns of 28 late-adopted children (placed

Downloaded by [85.241.28.67] at 06:49 19 February 2013

when they were between four and seven years of age) and their adoptive mothers.
The change in the children’s internal working models (IWMs) within seven to
eight months of their placement was evaluated. In addition, we wanted to observe
the influence of a secure-autonomous maternal state of mind in facilitating the
change in the children’s IWMs and the possible associations between the maternal
IWMs and the children’s IWMs in the adoptive dyads. The separation-reunion
procedure (SRP) was used for the late-adopted children in order to assess their
attachment behavioural patterns, and the Manchester Child Attachment Story
Task (MCAST) was used to evaluate their attachment narrative patterns. The
adoptive mothers completed the Adult Attachment Interview (AAI) in order to
classify their state of mind with regard to attachment. The results showed a
significant change in the attachment behavioural patterns of late-adopted
children, from insecure to secure (p ¼ .002). Furthermore, the children who
presented this change were predominantly placed with secure-autonomous
adoptive mothers (p ¼ .047), although the link between the adoptive mothers’
representations of their attachment history and their adopted children’s
completed narratives was not significant. In conclusion, it seems possible to
revise the attachment behaviour of late-adopted children but, for about one-third
of children, the adverse history will persist at a narrative/representational level.
Keywords: attachment; continuity/discontinuity; internal working models
(IWMs); late-adopted children; adoptive mothers

Introduction
At the beginning of research based on attachment theory (Bowlby, 1969, 1973, 1980),
many studies assessed the continuity and stability of internal working models
(IWMs) from infancy to adulthood, particularly if an individual’s life conditions –
both positive and unfavourable or maladaptive – were stable over time (Hamilton,
2000; Waters, Merrick, Treboux, Crowell, & Albersheim, 2000). However, in more
recent years, the literature on attachment has been considering a more complex
theoretical concept that encompasses the discontinuity of attachment representa-
tions. Although adults’ internal models are conceived as being resistant to change,

*Corresponding author. Email: [email protected]

ISSN 1461-6734 print/ISSN 1469-2988 online

Ó 2012 Taylor & Francis
http://dx.doi.org/10.1080/14616734.2012.636658
http://www.tandfonline.com
 46 C.S. Pace et al.

some researchers have demonstrated the potential for IWMs to be revised over time
in relation to a new environment (Grossman, Grossman, & Waters, 2005; Howes &
Spieker, 2008; Oppenheim & Goldsmith, 2007; Steele, Hodges, Kaniuk, & Steele,
2010; Steele & Steele, 2008).
In this context, adoptive and foster families have been the subject of a great deal of
interest, as adoption has often been described as a ‘‘natural experiment’’ which is able
– more than other types of social interventions – to affect the lives of children who
have often suffered traumas in their infancy and childhood (Dozier & Rutter, 2008;
Rutter & O’Connor, 2004; van IJzendoorn & Juffer, 2005, 2006). The question is
whether children’s previous negative experiences – such as abandonment, neglect and
abuse – can have a negative effect on their IWMs and therefore prevent any potential
change, and whether these insecure and/or disorganised patterns could be processed
into secure models thanks to new and accessible caregivers (Steele, Henderson,
Hodges, Kaniuk, Hillman, & Steele, 2007; Steele, Hodges, Kaniuk, Hillman, &
Henderson, 2003; Steele, Hodges, Kaniuk, Steele, Hillman, & Asquith, 2008).
Downloaded by [85.241.28.67] at 06:49 19 February 2013

The literature on adoption (Howe, 1997, 1998, 2003) tends to distinguish between
‘‘early-adopted’’ (or ‘‘baby-adopted’’) and ‘‘late-adopted’’ (or ‘‘older-adopted’’) child-
ren. Early-adopted children are those who were placed as babies (before the age of 12
months old) and therefore could form their first attachment relationship with their
biologically unrelated parents. Many studies have found that baby-adopted children,
both in national and international adoptions, managed to almost completely catch up
with non-adopted children in terms of their cognitive, behavioural, relational and
affective development and attachment security (Juffer & van IJzendoorn, 2009; Van
den Dries, Juffer, van IJzendoorn, & Bakermans-Kranenburg, 2009; van IJzendoorn,
Juffer, & Klein Poelhuis, 2005; Van Londen, Juffer, & van IJzendoorn, 2007).
Late-adopted children are those who were placed in their adoptive families after
the age of 12 months old, and therefore experienced either at least one rupture in their
primary attachment or the total absence of a stable figure in the period of attachment
formation. Research on older-adopted children have shown both difficulties (Howe,
2001; Hushion, Sherman, & Siskind, 2006; Rutter et al., 2007) and the potential to
catch-up at cognitive and emotional levels (Bick & Dozier, 2008; Dozier, 2005; Juffer
& van IJzendoorn, 2007). The majority of late-adopted children, and particularly
children who were placed at pre-school- or school-age, have had a prolonged expo-
sure to deleterious experiences with their biological parents, ranging from neglect,
rejection and role-reversal to physical, psychological and sexual abuse (Howe, 1998).
Given the children’s background, they may develop negative attachment representa-
tions and their trust in adult attachment figures and their ability to build attachment
bonds could be notably compromised (Goldsmith, 2007). The most extreme
situations, characterised by serious deprivation and ‘‘cumulative traumas’’, such as
early, long and severe periods of institutionalisation, prolonged maltreatment and
multiple placements, could have notably damaged the organisation of the attachment
systems of these children who may subsequently present with pathologies such as
‘‘secure base distortion’’ (Howe, 2003; O’Connor & Zeanah, 2003) and symptoms of
‘‘reactive attachment disorder’’, such as a behavioural pattern of indiscriminate
friendliness (Balbernie, 2010; Dozier, 2003; Juffer, Hoksbergen, Riksen-Walraven, &
Kohnstamm, 1997; van IJzendoorn & Bakermans-Kranenburg, 2003).
Once adopted, older children, and particularly those considered to be ‘‘special
needs adoptions’’ (Clark, Thigpen, & Yates, 2006), meaning children with special
features such as severe emotional and behavioural disorders or organic disabilities
 Attachment & Human Development 47

after the age of four years old, frequently show contradictory behaviours towards their
adoptive parents, oscillating between compliance, dependence, passivity, withdrawal,
rejection, hostility and provocation (Steele, Hodges, Kaniuk, & Steele, 2010). These
behaviours can make it hard to build an attachment relationship based on trust and a
sense of belonging and protection in the adoptive families (Slade, 2007; Steele et al.,
2007a).
Some authors have underlined the importance of high quality adoptive parenting
in order to manage the conflicting and vulnerable representations that late-adopted
children bring with them into their adoptive placements (Levy & Orlans, 2003;
Schofield & Beek, 2005). According to attachment theory, this high level of parenting
is strongly associated with the secure adults’ attachment state of mind which stems
from their own attachment histories. Some studies on foster care or adoptive
placements have also taken into consideration the attachment state of mind of the
new caregiver, particularly maternal representations, as a central component in the
older-placed child’s development of new attachment relationships.
Downloaded by [85.241.28.67] at 06:49 19 February 2013

Dozier, Stovall, Albus, and Bates (2001) have examined the correlation between
foster mothers’ attachment state of mind, measured using the AAI (George, Kaplan,
& Main, 1985), and foster infants’ attachment quality, assessed using the Strange
Situation Procedure (SSP; Ainsworth, Blehar, Waters, & Wall, 1978), for 50 foster
mother–infant dyads. The results found that the degree of secure-insecure
correspondence between the maternal state of mind and the infant’s attachment
quality was 72% (k ¼ .43), which is similar to the level found among biologically
intact mother–infant dyads. Contrary to the expectations, variations in the child’s age
at placement were not associated with infant security status. Instead, a correlation
between maternal state of mind and infant attachment was found for relatively late-
placed babies (between 12–20 months of age), as well as early placed babies (within 12
months of age). Based on these findings, the author proposed that foster children may
organise their attachment behaviour around the availability of their foster parent.
A further study on adoption (Caspers, Yucuis, Troutman, Arndt, & Langbehn,
2007) investigated sibling concordance on attachment by examining concordance for
adult attachment in a sample of 126 genetically unrelated sibling pairs. The results
showed that the distribution of attachment classifications was independent of
adoptive status and attachment concordance rates were unassociated with gender
concordance and sibling age difference. The concordance for secure/non-secure
classifications was significant at 61%, as was the concordance for primary
classifications at 53%, demonstrating the similarity of working models of attachment
between siblings, independent of their genetic relatedness.
Verissimo and Salvaterra (2006) evaluated the relationship between adoptive
mothers’ attachment representations as measured using attachment script representa-
tions (Waters & Waters, 2006), and children’s quality of attachment as measured
using the Attachment Behaviour Q-Set (AQS; Waters, 1995) in a sample of
Portuguese children adopted between three weeks and 47 months of age. The results
showed that the AQS security scores of adopted children did not differ significantly
from those of biological dyads. In addition, neither the child’s age at the time of
adoption nor the age of the child at the time of assessment significantly predicted their
AQS security score. However, scores reflecting the presence and quality of maternal
secure base scripts did predict AQS security (R Spearman ¼ 0.38, p 5 0.01).
All participants in the studies mentioned above were children who were adopted
before they reached four years of age, and only a few studies have examined children
 48 C.S. Pace et al.

who were placed after this age. The ‘‘Attachment representations and adoption
outcome study’’, which was a longitudinal study coordinated by Miriam Steele and
conducted at London’s Anna Freud Centre, explored the development of
attachments in a group of 63 children aged between four and eight at the time of
placement, all of whom had a history of separation, abandonment, maltreatment
and multiple placements. In addition, this study aimed to test whether or not secure
mothers (as coded by the AAI) transfer a sense of security to their recently adopted
children when compared to insecure and/or unresolved mothers. The children were
assessed using the story stem assessment technique, which provides them with a way
of displaying their expectations and perceptions of attachment figures using both
verbal and nonverbal means of representation. The results highlighted that late-
adopted children showed a progressive improvement in their inner representations of
self, others and relationships, which were described as being more positive and
benevolent, and an increase in internal consistency (Hodges, Steele, Hillman, &
Henderson, 2003; Hodges, Steele, Hillman, Henderson, & Kaniuk, 2003; Hodges,
Downloaded by [85.241.28.67] at 06:49 19 February 2013

Steele, Hillman, Henderson, & Neil, 2000; Kaniuk, Steele, & Hodges, 2004).
Mothers who were judged to be insecure (either dismissing or preoccupied)
according to their AAIs were likely to have adopted children who, three months after
their placement, provided story-completions with higher levels of aggressiveness
when compared to the stories provided by children adopted by mothers with secure-
autonomous AAIs. Children whose adoptive mothers’ AAIs were indicative of
unresolved (as opposed to resolved) mourning regarding past loss or trauma
provided story completions with higher scores for emotional themes such as ‘‘parent
appearing child-like’’ and ‘‘throwing out or throwing away’’ (Steele et al., 2003).
These results were confirmed by evaluations conducted two years after adoption.
Children’s insecurity correlated significantly with the parental AAI variable (R
Spearman ¼ 70.29, p 5 .05), in that the presence of one or more secure parents in
the children’s lives made the presence of insecure themes in the children’s stories
significantly less likely. Similarly, and more significantly, the children’s disorganisa-
tion correlated negatively with parental security (R Spearman ¼ 70.36, p 5 .01), in
that parental insecurity was strongly linked to elevated levels of disorganisation in
their adopted children. In other words, when neither parent’s AAI was secure at the
time of adoptive placement, at the time of the two-year follow-up, 86% of their
children were in the highest scoring group for disorganisation (Steele et al., 2008). In
addition, children adopted by unresolved mothers had failed in their attempts to
establish secure attachments, and they did not manage to build up a positive image
of their parents (Steele et al., 2007a, 2007b, 2010).
All of these findings support a non-genetic mechanism for the intergenerational
transmission of attachment, supporting the notion that the transmission of
attachment security across generations implicates a shared environment and involves
mutual exchanges and learning by the child, and that the exchanges leading to secure
attachment need not begin at birth. These results have extended previous research by
examining the critical influence of maternal representations of attachment on the
quality of children’s attachment security.

Adoption in Italy
Adoption in Italy is regulated by Law n. 183 (1984), which was supplemented by
successive modifications in 1998 (Law n. 476) and 2001 (Law n. 149). In Italy,
 Attachment & Human Development 49

children are placed exclusively in the context of married couples. Couples wishing to
adopt must first undergo a medical-psychiatric screening in order to assess whether
they have any psychopathological disorders. Then they are involved in a selection-
training course, which is conducted by an adoption team, a psychologist and a social
worker from the Social Health Services. The course comprises individual interviews,
couple interviews, home-visits and occasionally involves standardised tests. The
adoptive parents are selected if they are free from psychiatric disorders and if they
show enough sensitivity, empathy, personal autonomy, flexibility and the ability to
adapt to new circumstances. If the adoption team’s report is positive, the couples can
be entered into the Juvenile Court databank for national adoption (NA) and/or they
could approach an authorised international adoption (IA) Agency.
In recent years, three changes in the features of adoption in Italy have been noted.
First, there has been a drastic reduction in the number of NAs since social
interventions for the protection and support of parents at risk of psycho-social
disorders have improved. Second, there has been a sharp increase in the number of
Downloaded by [85.241.28.67] at 06:49 19 February 2013

IAs, currently totalling approximately 75% of adopted children. Finally, within these
IAs, the age of the children at the time of placement has increased, and currently
children are adopted predominantly between three and eight years of age. This change
is probably due to an increasingly high number of countries adhering to the Aja
Convention of 1993, which described IA as an ‘‘extra’’ intervention to be used
exclusively when all other attempts to help children to remain in their country of
origin have failed (e.g. home visiting, foster or adoptive placements with biologically
related persons – such as grandparents or uncles – and national adoption). This,
therefore, makes adoption an intervention which is often realised later on.
As the practice of adoption of children older than three years of age is relatively
recent in Italy, it would be interesting to explore the continuity/discontinuity of
IWMs and to examine the attachment models of families that have adopted pre-
school and school-age children who had suffered negative experiences in their
pre-adoption lives. The present study, therefore, proposes to assess the IWMs of
late-adopted children and their adoptive mothers during the first period after their
placement. Our first objective is to investigate whether a positive revision – intended
as a change from insecurity towards security – in attachment behaviour patterns can
be seen in late-adopted children within seven to eight months of their adoption. Our
second objective is to investigate whether such a change is seen predominantly in
children placed with mothers who display secure states of mind. Finally, and for
exploratory purposes only, we wanted to test the impact of adoptive mothers’
attachment representations on the attachment narrative patterns presented by their
late-adopted children in a story completion task.

Methodology
Participants
The main sample included 48 participants, comprising late-adopted children aged
between four and seven years old (N ¼ 28) and their adoptive mothers (N ¼ 20), all
recruited via the Social Health Services and authorised IA agencies. Participation in
this research was voluntary. The sample of children had a mean age of 70 months
(SD ¼ 12.75); 46% were male (N ¼ 13) and 54% were female (N ¼ 15). Five
children were born in Italy, while 23 (82%) children came from IAs. The pre-
adoption histories of the children generally presented with multiple difficulties in
 50 C.S. Pace et al.

the family of origin, including serious neglect, abuse and maltreatment, as well as
psychiatric problems, alcoholism and psycho-social problems in the biological
parents. In addition, 89% of the children (N ¼ 25) had experienced prolonged
institutionalisation before being adopted (M ¼ 29.12 months and SD ¼ 18.37).
Children with moderate or high level of physical disorders, mental retardation and
psychiatric diagnoses were excluded from the research. The adoptive mothers
(N ¼ 20) were aged between 38 and 52 years old (M ¼ 44.50 years old and
SD ¼ 4.37) and had between eight and 18 years of formal education (M ¼ 15.65
and SD ¼ 2.98). All of them had been married, with a mean of 10 years of marriage
(SD ¼ 4.83), and none of the mothers presented with psychiatric symptoms.

Measures
The separation-reunion procedure (SRP; Cassidy, 1988; Main & Cassidy, 1988) is a
laboratory procedure based on the SSP model and designed for classifying the
Downloaded by [85.241.28.67] at 06:49 19 February 2013

behavioural attachment patterns shown by children at pre-school and school-age. In the

SRP, the child experiences two separations from the mother and two reunions with her,
both of which are coded. The first separation lasts for 10–15 minutes, the second for
approximately 45–60 minutes. The first coding comprises the assigning of scores on the
scales of avoidance and security. The second coding classifies reunion behaviours into
one of the patterns of attachment. It is possible to assign one of the following three main
categories: secure (B), when the child appears relaxed and shows pleasure in the dyadic
interaction; insecure/avoidant (A), when the child tends to ignore the mother’s presence
and remains concentrated on their own activities; and insecure/ambivalent (C), when
the child shows signs of passivity-immaturity and/or anger. In addition, reunion
behaviour could also be assigned to one of the two secondary categories: insecure/
controlling (D), when the child seems to want to actively direct the caregiver’s attention
and behaviour, and insecure-unclassifiable (U), when the child appears extremely
depressed or scared by the parent. Studies using the SRP have shown test-retest relia-
bility of 85% for the insecure/secure classifications at one month of distance (excluding
the D classification, which is the most unstable) and a 78–90% predictability for the
three main categories on the basis of the SSP classification at one year of age (Cassidy,
1988; Kerns, 2008; Main & Cassidy, 1988; Speranza, 1994). The two SRP reunions were
coded by two independent raters trained by A.M. Speranza, an expert in the coding
system set out by Main and Cassidy (1988), and in the SSP by Ainsworth et al. (1978),
who directly supervised the first coding. One of the raters was completely blind to other
data. The degree of agreement between the raters was 87.5% for the categories of
secure/insecure (k ¼ .71) and 82% (k ¼ .73) for the three-way system (A, B and C).
The Manchester Child Attachment Story Task (MCAST; Barone, Del Giudice,
Fossati, Manaresi, Actis Perinetti, Colle et al., 2009; Goldwyn, Stanley, Smith, &
Green, 2000; Green, Stanley, Smith, & Goldwyn, 2000) is a story completion task
which is used to classify the attachment narrative representations of children between
the ages of four and eight years old. The interviewer shows the child a doll’s house and
a set of dolls, and the child has to choose the doll which best represents himself (or
herself) and the doll which best represents his or her mother. In our pilot study, the
MCAST was intentionally set up without specifying birth or adoptive mother, in
order to leave the child free to decide on the caregiver him/herself, without asking the
child direct questions which might cause him or her undue conflict or anxiety (Steele
et al., 2010). The assessment is organised into six stories to be completed. The four
central stories are designed to activate the motivational system of attachment and
 Attachment & Human Development 51

they concern: nightmares, a knee injury, a stomach ache, and getting lost in a
shopping centre. Scores of 1–9 on 33 scales and a final classification (secure/avoidant/
ambivalent/disorganised) are then assigned to each story. From these codes, it is
possible to add more scores from three global scales – coherence of mind, menta-
lisation, and disorganisation – and a final classification to the whole MCAST. The
pattern is classified as secure (B) when the strategy of attachment is characterised by
the representation of an interaction of proximity with the caregiver that brings an
assuagement of the discomfort. The pattern is rated as avoidant (A) when a restriction
is seen in the story content, or there is a representation of self-soothing or
displacement. The strategy is assessed as resistant-ambivalent (C) when contact with
the caregiver causes an increase in and prolonging of unease. The disorganised
classification (D) can be attributed if the score for episodic disorganisation,
(disorganised, bizarre and unusual themes in the child’s narrative and behaviours
displayed in the SSP episodes) is higher than five, or when pervasive disorganisation
emerges that represents a complete absence of strategy or the presence of multiple
Downloaded by [85.241.28.67] at 06:49 19 February 2013

internally contradictory strategies. Studies on the MCAST have shown good content
validity, a short term stability of 75% after approximately 5.5 months (Green et al.,
2000) and a concordance of 80% with respect to the separation anxiety test (SAT;
k ¼ .41, p 5 .01) (Goldwyn et al., 2000; Green, Stanley, & Peters, 2007). Our
MCASTs were independently coded by two raters trained by Jonathan Green, and
one of them was completely blind to other data. Inter-rater agreement was 89%
(k ¼ .82) for four-way (A, B, C, and D) classification.
The Adult Attachment Interview (AAI; Bakermans-Kranenburg & van
IJzendoorn, 2009; Main, Goldwyn, & Hesse, 2008) is an hour-long semi-structured
interview composed of 20 questions capable of evaluating and classifying adults’
state of mind with respect to attachment with regard to their early attachment
experiences. The interviews are valued on 17 ordinal scales of nine points each (Main
et al., 2008) and then assigned to one of the following classifications: secure/
autonomous (F/A); dismissing (Ds); preoccupied-entangled (E); unresolved/dis-
organised (U); and cannot classify (CC). The AAIs were coded by two independent
and skilled raters, trained by Debra Jacobvitz and Nino Dazzi, of which one was
blind to all the other data. Inter-rater agreement was 85% (k ¼ .73) for four-way (F,
Ds, E, and U) and 90% (k ¼ .79) for two-way (secure vs. insecure) classifications. In
order to guarantee the accuracy of the measurements and in order to minimise bias,
each of the attachment classification instruments – SRP, AAI, MCAST – were
codified by two independent, double-blinded and well-qualified raters, each of whom
was unaware of the coding assigned by the other to the same measure.
The Leiter International Performance Scale - revised (Roid & Miller, 1997) is a
test to measure nonverbal IQ (conceptualisation, inductive reasoning and visualisa-
tion) in individuals ranging in age from two to 20 years. It is useful for late-adopted
children, who cannot be reliably evaluated with traditional intelligence tests that
predominantly utilise verbal components that are subject to cultural, social and
educative influences. From validation studies, nonverbal IQ has been shown to be
highly correlated with the total IQ (.85), performance IQ (.85) and verbal IQ (between
.77 and .80) scores of the Wechsler Intelligence Scale for Children-III (WISC-III).
The Peabody Picture Vocabulary Test - revised (PPVT-R; Dunn & Dunn, 1981;
Stella, Pizzoli, & Tressoldi, 2000) is a vocabulary test designed to assess children’s
comprehension level of standard Italian. The PPVT-III has been shown to be highly
correlated with other validated instruments including the WISC-III (.82 to .92). The
internal validity (Cronbach’s a) is equal to .93.
 52 C.S. Pace et al.

Research design and procedure

A short-term longitudinal research design was used in order to achieve the previously
described objectives. The procedure was composed of two steps of data collection:

(1) The first (T1) was completed at the beginning of the adoption (approximately
40 days after the children’s placement) to classify the attachment patterns
which the children exhibited at their meeting with their adoptive parents,
which were probably still highly affected by their previous adverse experi-
ences. At the first session, the children were assessed using the SRP (SRP-T1)
and the Leiter Scale. The adoptive mothers were asked about their children’s
background history and were assessed using the AAI, which was recorded
and subsequently transcribed verbatim;
(2) The second step (T2) occurred six months after the first, within approxi-
mately seven to eight months of adoption. In this session, the children were
assessed using the SRP-T2, the PPVT-R and the MCAST.
Downloaded by [85.241.28.67] at 06:49 19 February 2013

The two assessment sessions with mother and child occurred in a well-equipped
video recording laboratory. Each session lasted approximately an hour and a half
and was entirely video-recorded. All of the mothers signed an informed consent sheet
for themselves and their children at the beginning of the study, written on the basis
of the rights and limits set out in Law 675/96 ‘‘Privacy of the person and other
aspects with respect to the handling of personal data’’.

Data analysis
The results were analysed using the Statistical Package for the Social Science (SPSS,
Version 12.0). We decided to use primarily non-parametric tests – appropriate for
this type of variable – because they do not require the population from which the
sample is taken to be assumed to be normal. Moreover, they are appropriate for
statistically testing small samples, as in this pilot study (Siegel & Castellan, 1988).
The following tests were used for the quantitative variables on an ordinal scale:
the U-Mann Whitney test for independent samples, the Wilcoxon test for dependent
samples, and Spearman’s correlation coefficient and linear regression statistic. The
data analysis was carried out by categorising the children’s behaviours in the
Separation Reunion Procedure (SRP) and Manchester Attachment Story Task
(MCAST), and the maternal AAI classifications into secure and insecure groups in
preparation for using more powerful statistical tests. The following tests were used:
Fisher’s exact test for independent samples, McNemar’s test for dependent samples,
the phi correlation coefficient and binary logistic regression. The level of significance
for all analyses was p 5 .05.

Results
Preliminary analyses
Distribution of attachment classifications
Table 1 shows the distribution of the children’s classifications in the SRP-T1, SRP-
T2, MCAST and the adoptive mothers’ AAI categories. No child was classified as D
according to the SRP and no adoptive mother was classified as CC. It is perhaps
 Attachment & Human Development 53

Table 1. Classifications according to the SRP-T1, SRP-T2, MCAST and AAI.

Classification of attachment – frequency (%)

B (%) A (%) C (%) D (%) Total
Adopted children
SRP-T1 4 (14%) 14 (50%) 10 (36%) 0 28
SRP-T2 14 (50%) 9 (32%) 5 (18%) 0 281
MCAST 13 (48%) 3 (11%) 1 (4%) 10 (37%) 27

F (%) Ds (%) E (%) U (%) Total

Adoptive mothers
AAI 12 (60%) 3 (15%) 2 (10%) 3 (15%) 20
1
One child did not fully complete the story completion task.

surprising that the baseline SRP showed no D children, given their backgrounds, but
Downloaded by [85.241.28.67] at 06:49 19 February 2013

we can assume that, at the beginning of the placement, it could be hard for children
to display humiliating and embarrassing or solicitous and protective behaviours
towards their ‘‘new’’ parents. The MCAST distribution, instead, presented 37% of
the children as D, perhaps revealing that the MCAST is a more sensitive test for
capturing disorganisation. The distribution of secure, dismissing, preoccupied, and
unresolved mothers was in line with samples of nonclinical populations (Bakermans-
Kranenburg & van IJzendoorn, 2009), although our sample of well-vetted adopters
showed a higher proportion of interviews that were classified as autonomous-secure.

Classifications of attachment and descriptive variables

The SRP-T1, SRP-T2 and MCAST secure/insecure classifications did not present any
significant links (the p values were always between .14 and .92) with any of the following
demographic variables: gender; presence/absence of siblings; level of schooling; age at
placement; NA/IA; nonverbal IQ level (M ¼ 83.8 and SD ¼ 10); and language
comprehension (M ¼ 78 and SD ¼ 8.80). The secure/insecure maternal states of mind
were not associated with their level of education (U-Mann Whitney: 69.500, p ¼ .27).

Revision of attachment behavioural patterns shown by late-adopted children

Ten (42%) out of the 24 children classified as insecure in the SRP-T1 were secure
according to the SRP-T2, showing a statistically significant discontinuity (36%
McNemar Test, p ¼ .002). All of the children classified as secure in the SRP-T1
(N ¼ 4) remained secure in the SRP-T2. The children who had moved from being
classified as insecure to secure (named ‘‘earned secure’’) had noticeably lower scores on
the avoidance scale (U Mann-Whitney ¼ 43.500, p ¼ .024) and significantly higher
scores on the security scale (U Mann-Whitney ¼ 38.500, p ¼ .012) between SRP-T1
and SRP-T2, compared to the children who remained insecure in both T1 and T2.

Impact of secure maternal states of mind on the revision of the children’s attachment
behavioural patterns
Considering the group of insecure children in T1 (N ¼ 24), the results suggest that
the adoptive mothers who displayed a secure-autonomous state of mind in the AAI
 54 C.S. Pace et al.

were more likely to have children who became ‘‘earned secure’’ between T1 and T2,
compared to mothers classified as insecure following the AAI, whose children all
‘‘remained insecure’’ (Fisher’s exact, p ¼ .047).
In order to fully explore the possible influence of maternal state of mind on the
change in children’s responses from insecure to secure (dependent variable),
controlling for the effects of baseline SRP and the age of the children, we used a
hierarchical binary logistic regression. In the first block, we included SRP-T1 and the
age of the children as independent variables, while in the second block we added the
AAI of the mothers. The analysis showed that the introduction of the first two
variables was not significant (w2(2) ¼ 3.989, p ¼ . 136), while the subsequent inclusion
of the AAI produced a significant increase in the model fit (w2(1) ¼ 6.237, p ¼ . 013).
No other potentially confounding variable were included given low and not
significant correlations with the predictors and small size of the sample.
Upon examining the correlations between the scores on the AAI and SRP scales,
it emerged that adoptive mothers with high scores on the coherence of transcript
Downloaded by [85.241.28.67] at 06:49 19 February 2013

scale had children whose scores on the security scale increased significantly between
SRP-T1 and SRP-T2 (Spearman’s r ¼ .344, p 5 .05). In addition, mothers with
high scores on the insistence on lack of recall scale had children who scored lower on
the security scale in SRP-T2 (Spearman’s r ¼ 7.428, p 5 .05).

Relation between the adoptive mothers’ state of mind and the children’s attachment
narrative patterns
The correspondence between the mothers’ AAI classifications and the attachment
representations of the children, as expressed at a narrative level and measured by
the MCAST, was 55.6% (k ¼ .12) for a two-way classification (secure vs. insecure)
and did not show any statistical significance (phi ¼ .126, p ¼ .516 n.s.). Of the
children, 37% (N ¼ 10) were classified as disorganised according to the MCAST,
and among them, 40% were adopted by mothers with dismissing states of mind and
10% were placed with unresolved mothers. Significant correlations were reported
from the analyses of scores on the ordinal scales of the AAI and the MCAST (Table
2).
We ran a linear regression statistic with ‘‘global disorganisation’’ as the dependent
measure, and ‘‘idealisation of mother’’ and ‘‘anger toward father’’ as the independent
blocks. The regression summary shown in Table 3 indicates that ‘‘idealisation of
mother’’ alone makes a significant contribution to the prediction of high ‘‘global
disorganisation’’ and low ‘‘coherence of mind’’, while ‘‘anger toward father’’ in the
AAI did not produce a significant result in the linear regression model. None of the
other independent contributors correlated with the children’s global D score.

Table 2. Correlations between the AAI scales of actual states of mind and the MCAST
scales.

AAI states of mind MCAST global scale Spearman’s Rho

Idealisation of mother Coherence of mind 7.63**
Idealisation of mother Global disorganisation score .60**
Anger towards father Global disorganisation score .37*

*p 5 .05; **p 5 .01.

 Attachment & Human Development 55

Table 3. Regression model predicting ‘‘global disorganisation’’ and ‘‘coherence of mind’’ in

children according to the MCAST from maternal ‘‘idealisation of mother’’ and ‘‘anger
towards father’’ in the AAI.

Factor Standardised beta t p

Global disorganisation
Idealisation of mother .635 3.977 .001
Anger towards father .072 .449 .658
Coherence of mind
Idealisation of mother 7.494 72.877 .008

Discussion
Our first objective was to find a significant and positive change in late-adopted
children’s attachment behavioural patterns within seven to eight months of their
Downloaded by [85.241.28.67] at 06:49 19 February 2013

adoption. At the beginning of their adoptive placements (T1), almost all of the
children in our sample, aged between four and seven years old, had shown insecure
IWMs. It is probable that these late-adopted children, who had almost certainly
experienced poor quality care prior to being placed, had internalised confusing,
incoherent and conflicting models from unpredictable and often violent caregivers.
When they started to live with their adoptive parents, these negative representations
were sometimes translated into insecure relational strategies predominantly
characterised by avoidance, passive withdrawal or angry ambivalence. Within seven
to eight months of their adoption, a significant percentage of the insecure children had
shown a positive change in their attachment behavioural patterns and were classified
as (earned) secure, demonstrating more pleasure and responsiveness in their
interaction with their adoptive mothers. These results indicate that once inserted
into a new, stable and nurturing environment, late-adopted children have a good
chance of revising their IWMs, changing them from insecure to secure, and building
new positive representations of attachment relationships. These results also show that
late adoption does not seem to be a factor that can in itself impede children’s IWM
revision, as has been previously reported in some studies (Steele et al., 2007b, 2008;
Verissimo & Salvaterra, 2006), although other studies have highlighted that, if
children are older, this can have a partially negative effect on the success of the
adoption (Howe, 2001; van IJzendoorn, Bakermans-Kranenburg, & Juffer, 2007).
Although 42% of the children in our pilot study changed from insecure to secure
behavioural patterns, their narrative attachment representations, measured by the
MCAST after seven to eight months of adoption, revealed that 37% of the narratives
were classified as disorganised, meaning that they were characterised mainly by
‘‘bizarre themes’’ and ‘‘complete chaos’’. This is despite the fact, surprisingly, that
48% of the narratives showed secure patterns. In accordance with researchers in the
field of attachment, we may suggest that although many secure attachments and
positive representations are achieved with the new parents, in a significant number of
cases the old, disorganised and traumatic representations continued to be present
and were triggered at stressful points in the relationship. The previous and negative
representations are not ‘‘extinguished’’, but continue to exert an influence, and the
change in behaviour seems to precede a change in the children’s narratives.
On the basis of our results, we could infer that a faster (than MCAST-indexed)
changing relationship could be captured by the Separation Reunion Procedure
 56 C.S. Pace et al.

(SRP-T1 and T2) with first 14% then 50% showing security, pointing to the rapid
development (for half the sample) of a new (more positive) attachment relationship
to the adoptive caregiver. In other words, the 37% of the children at T2 presenting
story-completions judged disorganised by the MCAST rating procedure suggests
that the doll play and story beginnings elicited attachment representations linked to
early (troubled) experiences for a little more than a one-third of the sample. This
persistence of trauma-based thoughts and feelings for some children was correlated
with having an adoptive mother highly prone to idealise her own mother, a marker
of being less effective (than other adoptive mothers) at promoting a new and
organised representational world in her late-adopted child. High idealisation may,
accordingly, be an AAI-based indicator of a need for post-adoption support, adding
to the AAI and story-stem based clues as to who may be most in need of therapeutic
intervention or support following adoption (Steele et al., 2007a, 2007b, 2008).
With respect to the 63% of children who were not disorganised in response to the
MCAST and the full sample of children who showed no disorganisation in the SRP
Downloaded by [85.241.28.67] at 06:49 19 February 2013

observations, this may be taken as evidence of marked and positive change achieved
soon after placement. Yet we cannot be sure that this change would be stable, or that
it is actually a sign of lasting psychological well-being. Attachment researchers and
clinicians have noticed that older-adopted children all have in common experiences
of having lost birth parents or having been abandoned and, in many instances, the
same caregiver was the perpetrator of abuse and neglect. These deleterious
experiences obviously make them more vulnerable to negative outcomes in terms
of their mental health and psychological functioning. The overrepresentation of late-
adopted children in mental health and special educational services is reflected in
lower levels of achievement at school and self esteem, and higher rates of
externalising and internalising behavioural problems in childhood, adolescence
and early adulthood (Howe, 1997; Juffer et al., 1997; Juffer & van IJzendoorn, 2009;
Steele et al., 2010).
The second objective was to investigate whether the secure-autonomous IWMs of
adoptive mothers were able to positively influence the change in attachment patterns
(from insecure to secure) of their late-adopted children. In our sample, all of the
‘‘earned secure’’ children were adopted by mothers who were shown to be secure in
their AAI, while the children placed with insecure mothers were also classified as
insecure at both data collection points. Secure maternal states of mind are usually
associated with a style of parenting which emphasises affective relationships and
values attachment experiences. These results suggest that secure adoptive mothers
could enable insecure children to mobilise affect and positively activate needs,
feelings and behaviours linked to their attachment systems. Secure-autonomous
mothers seems to produce positive effects in both avoidant children, enabling them
to abandon their defensive behaviours that are directed towards the de-activation of
attachment needs in favour of a stronger expression of feelings, and ambivalent-
resistant children, permitting them to free themselves from dependent-passive or
hostile behaviours that display a hyper-activation of the attachment system and
towards a more balanced and clear expression of their needs.
These results, therefore, highlight that it could be not only adoption in itself, but
also the opportunity to be adopted by mothers with secure IWMs who are capable of
balanced emotional regulation, that guarantees late-adopted children the possibility
of revising their insecure IWMs. Secure-autonomous adoptive caregivers, who have
an attentive, sensitive and responsive style of parenting, are probably able to look
 Attachment & Human Development 57

beyond what may otherwise be seen as hostility, defensiveness and rejection. They
may tend to respond in a ‘‘counter-complementary’’ way (Lieberman, 2003), without
showing retaliatory rejection, and therefore challenging the insecure and/or
disorganised representations of their children.
Although in our work we did not use instruments to assess parent–child
interaction, we may also hypothesise, as suggested by Steele et al. (2007a), that
mothers with secure states of mind tend to spontaneously demonstrate ‘‘attachment
facilitating behaviours’’ (Steele et al., 2007a, p. 68), involving the use of the child’s
name, references to we or us when interacting with their child, and references to
shared past experiences. These aspects appear to be associated with the development
of positive relationships in adoptive families.
In addition, it is always reassuring to find that the proportion of secure-
autonomous adopters according to their AAIs is high (60%), and that very few are
unresolved (15%), giving some credit to adoption services workers and their ability
to prepare and select substitute parents. While not surprising, this is very pleasing, as
Downloaded by [85.241.28.67] at 06:49 19 February 2013

it offers implicit evidence of the careful work performed by the social workers who
undertook the assessment of the adopters (Steele et al., 2007a). Given the influences
upon the internal world of children exerted by their adoptive mothers’ state of mind,
those assessing the adopters will need to take their final selection of adopters very
seriously in the hope of facilitating the difficult task of matching child with adopter
and reducing the amount of disruption in placements.
Finally, the third objective of our study was to examine the possible association
between adoptive mothers’ state of mind and the attachment narrative representa-
tions of their children. The level of concordance between the secure/insecure AAI
classifications and the results of the MCAST (55.6%) in our sample did not reach
either the level of significance of 75% indicated by van IJzendoorn’s (1995) meta-
analysis conducted on the AAI and SSP results of 854 ‘‘normal’’ dyads, nor the level
of significance of 72% found by Dozier et al. (2001) in a sample of 50 foster dyads.
On the other hand, analysis of the dimensional scale scores of the two instruments
showed that idealisation scores in the mothers’ AAIs significantly and inversely
correlated with scores for general coherence of mind and positively correlated with
the disorganisation scores on the MCAST. In other words, it was more probable that
a mother who tended to speak about her own childhood and past relationships in a
generic, unbelievable and untruthful manner would have an adopted child who
completed the story completion task in an incoherent, confused and irrelevant way,
and therefore showed aspects of disorganisation such as bizarre themes and chaotic
behaviours.
In general, it is noteworthy that only two scales – insistence on lack of recall and
idealisation of the mother – were connected with the dismissing classification and
with difficulty on the part of the mother in connecting present and past and semantic
and specific memories. These were linked with the worst outcomes in late-adopted
children, who have to deal with the hard task of integrating the past and the present.
The placement of late-adopted children who have suffered adverse and
dysfunctional experiences represents a context of affective ties and of care that
challenges the maladaptive representations of the children. Steele et al. (2007b)
hypothesised that it is easier for traumatised children to take on new and positive
representations rather than change the early negative representations that continue
to exert their influence. The security of the adoptive parents’ IWMs, along with the
capacity to offer adopted children the opportunity to slowly revise their negative
 58 C.S. Pace et al.

representations of the self and others, can be considered as a factor that is able to
facilitate the revision of the children’s IWMs.

Limitations and future developments

This work has some evident methodological limitations. First, the sample size is
small and a bigger sample size could be useful in order to further elaborate our data,
confirming or disconfirming these initial results. On the other hand, our small sample
has the advantage of having been observed from the beginning of their placement. In
fact, the few studies conducted on samples of adopted children in Italy, which are
always quite small in terms of size, have tended to focus on pre-adolescent or
adolescent children who have been adopted for a long time. This pilot study is the
first contribution to examine ‘‘newly’’ adoptive dyads and to enable the measuring of
possible changes to late-adopted children’s IWMs. Collecting data from a sample of
newly formed adoptive families has been quite difficult because often, just after
Downloaded by [85.241.28.67] at 06:49 19 February 2013

placement, adopters tend to shut themselves away and are not particularly amenable
to external interference, such as participation in research.
Second, the T2 session was approximately eight months after the placements
began and, although it was possible to observe a shift towards security, the period of
time between the two assessment sessions was not enough long to establish whether
the modifications were really stable. The follow-up is probably too soon for any
definitive view on the changes made in placements, even if some research has found
associations between the state of mind of adoptive mothers and emotional narratives
in previously maltreated children within only three months of adoption (Steele et al.,
2003). It would therefore be interesting to carry out a follow-up longitudinal study in
the medium to long term in order to explore the development of the IWMs of late-
adopted children throughout their childhood, adolescence and adulthood.

Acknowledgements
Our thanks to Professor Pietro San Martini for his methodological supervision, and also to all
the workers and adoptive families for their involvement in this study.

References
Ainsworth, M.D.S., Blehar, M.C., Waters, E., & Wall, S. (1978). Patterns of attachment: A
psychological study of Strange Situation. Hillsdale, NJ: Erlbaum.
Bakermans-Kranenburg, M.J., & van IJzendoorn, M.H. (2009). The first 10,000 Adult
Attachment Interviews: Distributions of adult attachment representations in clinical and
non-clinical groups. Attachment & Human Development, 11, 223–263.
Balbernie, R. (2010). Reactive attachment disorder as an evolutionary adaptation. Attachment
& Human Development, 12(3), 265–381.
Barone, L., Del Giudice, M., Fossati, A., Manaresi, F., Actis Perinetti, B., Colle, L., et al.
(2009). Psychometric properties of the Manchester Child Attachment Story Task: An
Italian multicentre study. International Journal of Behaviour Development, 33, 185–190.
Bick, J., & Dozier, M. (2008). Helping foster parents change: The role of parental state of
mind. In H. Steele & M. Steele (Eds.), Clinical applications of the Adult Attachment
Interview (pp. 452–470). New York, NY: Guilford Press.
Bowlby, J. (1969). Attachment and loss. Vol. 1. Attachment. New York, NY: Basic Books.
Bowlby, J. (1973). Attachment and loss. Vol. 2. Separation: Anxiety and anger. New York, NY:
Basic Books.
Bowlby, J. (1980). Attachment and loss. Vol. 3. Loss: Sadness and depression. New York, NY:
Basic Books.
 Attachment & Human Development 59

Caspers, K., Yucuis, R., Troutman, B., Arndt, S., & Langbehn, D. (2007). A sibling adoption
study of adult attachment: The influence of shared environment on attachment states of
mind. Attachment & Human Development, 9(4), 375–391.
Cassidy, J. (1988). Child–mother attachment and the self in six-year-olds. Child Development,
59, 121–134.
Clark, P., Thigpen, S., & Yates, A.M. (2006). Integrating the older/special needs adoptive
child into the family. Journal of Marital and Family Therapy, 32(2), 181–194.
Dozier, M. (2003). Attachment-based treatment for vulnerable children. Attachment & Human
Development, 5(3), 253–257.
Dozier, M. (2005). Challenges of foster care. Attachment & Human Development, 7(1), 27–30.
Dozier, M., Stovall, K.C., Albus, K.E., & Bates, B. (2001). Attachment for infants in foster
care: The role of caregiver state of mind. Child Development, 72(5), 1467–1477.
Dozier, M., & Rutter, M. (2008). Challenges to the development of attachment relationships
faced by young children in foster and adoptive care. In J. Cassidy & P.R. Shaver (Eds.),
Handbook of attachment. Theory, research, and clinical applications (2nd ed., pp. 102–129).
New York, NY: Guilford Press.
Dunn, L.M., & Dunn, L. (1981). Peabody Picture Vocabulary Test - Revised manual for forms
L and M. Circle Pines, MN: American Guidance Service.
Downloaded by [85.241.28.67] at 06:49 19 February 2013

George, C., Kaplan, N., & Main, M. (1985). Adult Attachment Interview Protocol (2nd ed.).
Unpublished manuscript, University of California at Berkeley.
Goldsmith, D.F. (2007). Challenging children’s negative internal working models: Utilizing
attachment-based treatment strategies in a therapeutic preschool. In D. Oppenheim &
D.F. Goldsmith (Eds.), Attachment theory in clinical work with children (pp. 203–225).
New York, NY: Guilford Press.
Goldwyn, R., Stanley, C., Smith, V., & Green, J. (2000). The Manchester Child Attachment
Story Task: Relationship with parental AAI, SAT, and child behaviour. Attachment &
Human Development, 2(1), 71–84.
Green, J., Stanley, C., & Peters, S. (2007). Disorganized attachment representation and
atypical parenting in young school age children with externalizing disorder. Attachment &
Human Development, 9(3), 207–222.
Green, J., Stanley, C., Smith, V., & Goldwyn, R. (2000). A new method of evaluating
attachment representations in young school-age children: The Manchester Child
Attachment Story Task. Attachment & Human Development, 2(1), 48–70.
Grossman, K.E., Grossman, K., & Waters, E. (Eds.) (2005). Attachment from infancy to
adulthood. The Major Longitudinal Studies. New York, NY: Guilford Press.
Hamilton, C.E. (2000). Continuity and discontinuity of attachment from infancy through
adolescence. Child Development, 71, 690–694.
Hodges, J., Steele, M., Hillman, S., & Henderson, K. (2003). Mental representations and
defences in severely maltreated children: A story stem battery and rating System for
clinical assessment and research applications. In R. Emde, D. Wolf, & D. Oppenheim
(Eds.), Revealing the inner worlds of young children: The MacArthur Story Stem Battery
and parent–child narratives (pp. 240–267). Oxford: Oxford University Press.
Hodges, J., Steele, M., Hillman, S., Henderson, K., & Kaniuk, J. (2003). Change in
attachment representations over the first year of adoptive placement; narratives of
maltreated children. Journal of Child Clinical Psychology, 8, 351–367.
Hodges, J., Steele, M., Hillman, S., Henderson, K., & Neil, M. (2000). Effects of abuse on
attachment representations: Narrative assessments of abused children. Journal of Child
Psychotherapy, 26(3), 433–455.
Howe, D. (1997), Parent-reported problems in 211 adopted children: Some risk and protective
factors. Journal of Child Psychology and Psychiatry, 38(4), 401–411.
Howe, D. (1998). Patterns of adoption. Nature, nurture and psychosocial development. Oxford:
Blackwell Science.
Howe, D. (2001). Age at placement, adoption experience and adult adopt people’s contact
with their adoptive and birth mothers: An attachment prospective. Attachment & Human
Development, 3(2), 222–237.
Howe, D. (2003). Attachment disorders: Disinhibited attachment behaviours and secure base
distortions with special reference to adopted children, Attachment & Human and
Development, 5(3), 265–270.
 60 C.S. Pace et al.

Howes, C., & Spieker, S. (2008). Attachment relationships in the context of multiple
caregivers. In J. Cassidy & P.R. Shaver (Eds.), Handbook of attachment. Theory, research,
and clinical applications (2nd ed., pp. 317–332). New York, NY: Guilford Press.
Hushion, K., Sherman, S.B., & Siskind, D. (2006). Understanding adoption. Clinical work with
adults, children and parents. Lanham, Maryland: Jason Aronson.
Juffer, F., Hoksbergen, R.A.C., Riksen-Walraven, J.M., & Kohnstamm, G.A. (1997). Early
intervention in adoptive families: Supporting maternal sensitive responsiveness, infant–
mother attachment, and infant competence. Journal of Child Psychology and Psychiatry,
38(8), 1039–1050.
Juffer, F., & van IJzendoorn, M.H. (2007). Adoptees do not lack self-esteem: A meta-analysis
of studies on self-esteem of transracial, international and domestic adoptees. Psychological
Bulletin, 133, 1067–1083.
Juffer, F., & van IJzendoorn, M.H. (2009). International adoption comes of age: Development
of international adoptees from a longitudinal and meta-analytical perspective. In G.M.
Wrobel & E. Neil (Eds.), International advances in adoption research for practice (pp. 169–
192). London: John Wiley and Sons.
Kaniuk, J., Steele, M., & Hodges, J. (2004). Report on a longitudinal research project,
exploring the development of attachments between older, hard-to-place children and
Downloaded by [85.241.28.67] at 06:49 19 February 2013

their adopters over the first two years of placement. Adoption & Fostering, 28,
61–67.
Kerns, K.A. (2008). Attachment in middle childhood. In J. Cassidy & P.R. Shaver (Eds.),
Handbook of attachment. Theory, research, and clinical applications (2nd ed., pp. 366–382).
New York, NY: Guilford Press.
Levy, T.M., & Orlans, M. (2003). Creating and repairing attachments in biological, foster, and
adoptive families. In S.M. Johnson & E. Wiffen (Eds.), Attachment process in couple and
family therapy (pp. 165–190). New York, NY: Guilford Press.
Lieberman, A.F. (2003). The treatment of attachment disorder in infancy and early childhood:
Reflections from clinical intervention with later adopted foster care children. Attachment
& Human Development, 5(3), 279–282.
Main, M., & Cassidy, J. (1988). Categories of response to reunion with the parent at age six:
Predicted from attachment classifications and stable over a one-month period.
Developmental Psychology, 24, 415–426.
Main, M., Goldwyn, R., & Hesse, E. (2008). Adult Attachment Scoring and Classification
System. Unpublished manuscript, Department of Psychology, University of California at
Berkeley.
O’Connor, T., & Zeanah, C.H. (2003). Attachment disorders: Assessment strategies and
treatment approaches. Attachment & Human Development, 5(3), 223–244.
Oppenheim, D., & Goldsmith, D.F. (Eds.) (2007). Attachment theory in clinical work with
children. New York, NY: Guilford Press.
Roid, G.H., & Miller, L.J. (1997). Leiter International Performance Scale – Revised. Wood
Dale, IL: Stoelting Co. Italian translation (2002) International Performance Scale –
Revised. Florence: O.S. Organizzazioni Speciali.
Rutter, M., Colvert, E., Kreppner, J., Beckett, C., Castle, J., Groothues, C., et al. (2007). Early
adolescent outcomes for institutionally deprived and non-deprived adoptees: Disinhibited
attachment. Journal of Child Psychology and Psychiatry, 48(1), 17–30.
Rutter, M., O’Connor, T.G., & the English and Romanian Adoptees Study Team. (2004). Are
there biological programming effects for psychological development? Findings from a
study of Romanian adoptees. Developmental Psychology, 40, 81–94.
Schofield, G., & Beek, M. (2005). Providing a secure base: Parenting children in long-term
foster family care. Attachment & Human Development, 7(1), 3–26.
Siegel, S., & Castellan, N.J., Jr. (1988). Nonparametric statistics for the behavioural sciences
(2nd ed.). New York: McGraw-Hill.
Slade, A. (2007). Disorganized mother, disorganized child: The mentalization of affective
dysregulation and therapeutic change. In D. Oppenheim & D.F. Goldsmith (Eds.),
Attachment theory in clinical work with children (pp. 226–250). New York, NY: Guilford
Press.
Speranza, A.M. (1994). Infant attachment stability and intergenerational transmission of
attachment patterns (Unpublished doctoral dissertation). La Sapienza, University of
Rome.
 Attachment & Human Development 61

Steele, M., Henderson, K., Hodges, J., Kaniuk, J., Hillman, S., & Steele, H. (2007b). In the
best interests of the late-placed child: A report from Attachment Representations and
Adoption Outcome study. In L. Mayes, P. Fonagy, & M. Target (Eds.), Developmental
science and psychoanalysis. Integration and innovation (pp. 159–182). London: Karnac.
Steele, M., Hodges, J., Kaniuk, J., Hillman, S., & Henderson, K. (2003). Attachment
representations and adoption: Associations between maternal states of mind and emotion
narratives in previously maltreated children. Journal of Child Psychotherapy, 29(2), 187–205.
Steele, M., Hodges, J., Kaniuk, J., & Steele, H. (2010). Mental representation and change:
Developing attachment relationships in an adoption context. Psychoanalytic Inquiry, 30,
25–40.
Steele, M., Hodges, J., Kaniuk, J., Steele, H., D’Agostino, D., Blom, I., et al. (2007a).
Intervening with maltreated children and their adoptive families: Identifying attachment-
facilitative behaviors. In D. Oppenheim & D.F. Goldsmith (Eds.), Attachment theory in
clinical work with children (pp. 58–89). New York, NY: Guilford Press.
Steele, M., Hodges, J., Kaniuk, J., Steele, H., Hillman, S., & Asquith, K. (2008). Forecasting
outcomes in previously maltreated children. The use of the AAI in a longitudinal adoption
study. In H. Steele & M. Steele (Eds.), Clinical applications of the Adult Attachment
Interview (pp. 427–451). New York, NY: Guilford Press.
Downloaded by [85.241.28.67] at 06:49 19 February 2013

Steele, H., & Steele, M. (Eds.) (2008). Clinical applications of the Adult Attachment Interview.
New York, NY: Guilford Press.
Stella, G., Pizzoli, C., & Tressoldi, P.E. (2000). Peabody. Tests of receptive vocabulary. Turin:
Omega Eds.
Van den Dries, L., Juffer, F., van IJzendoorn, M.H., & Bakermans-Kranenburg, M.J. (2009).
Fostering security? A meta-analysis of attachment in adopted children. Children and Youth
Services Review, 31, 410–421.
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.
van IJzendoorn, M.H., & Bakermans-Kranenburg, M.J. (2003). Attachment disorders and
disorganized attachment: Similar and different. Attachment & Human Development, 5(3),
313–320.
van IJzendoorn, M.H., Bakermans-Kranenburg, M.J., & Juffer, F. (2007). Plasticity of growth
in height, weight and head circumference: Meta-analytic evidence of massive catch-up after
international adoption. Journal of Developmental and Behavioral Pediatrics, 28, 334–343.
van IJzendoorn, M.H., & Juffer, F. (2005). Adoption is a successful natural intervention
enhancing adopted children’s IQ and school performance. Current Directions in
Psychological Science, 14, 326–330.
van IJzendoorn, M.H., & Juffer, F. (2006). The Emanuel Miller Memorial Lecture 2006:
Adoption as intervention. Meta-analytic evidence for massive catch-up and plasticity in
physical, socio-emotional, and cognitive development. Journal of Child Psychology and
Psychiatry, 47, 1228–1245.
van IJzendoorn, M.H., Juffer, F., & Klein Poelhuis, C.W. (2005). Adoption and cognitive
development: A meta-analytic comparison of adopted and non-adopted children’s IQ and
school performance. Psychological Bulletin, 131(2), 301–316.
Van Londen, W.M., Juffer, F., & van IJzendoorn, M.H. (2007). Attachment, cognitive and
motor development in adopted children: Short-term outcomes after international
adoption. Journal of Pediatric Psychology, 32, 1259–1263.
Verissimo, M., & Salvaterra, F. (2006). Maternal secure base scripts and children’s attachment
security in an adopted sample. Attachment & Human Development, 8(3), 261–273.
Waters, E. (1995). The Attachment Q-set (Version 3). In E. Waters, B.E. Vaughn, G. Posada, &
K. Kondo-Ikemura (Eds.), Caregiving, cultural, and cognitive perspectives on secure-base
behavior and working models: New growing points of attachment theory and research. Mono-
graphs of the Society for Research in Child Development (pp. 234–246), 60 (2-3, Serial No. 244).
Waters, E., Merrick, S.K., Treboux, D., Crowell, J., & Albersheim, L. (2000). Attachment
security from infancy to early adulthood: A 20 year longitudinal study. Child Development,
71, 684–689.
Waters, H.S., & Waters, E. (2006). The attachment working models concept: Among other
things, we build script-like representations of secure base experiences. Attachment &
Human Development, 8(3), 185–197.