Attachment Science

Unfortunately, there is lots of confusion and misinformation about attachment theory and research circulating at the moment. Dr Vrticka has recently written about it in VICE and in The Conversation UK.

On this page, Dr Pascal Vrticka provides summaries of recent studies and theoretical perspectives on attachment to help disseminating evidence-based attachment science.

All summaries will also be posted on Dr Vrticka’s Instagram account.

Please use the question / feedback form below to get in touch and make sure to check out Dr Vrticka’s two other series:


List of Topics

Attachment Science Topics #01 – #10
#01 – Child-Mother Attachment & Language and Cognitive Child Outcomes#06 – Attachment & Adoption
#02 – Child-Mother and -Father Attachment & Child Behavioural Problems#07 – Disorganised Attachment: Practical Implications
#03 – Attachment & Brain Structure in Adolescence#08 – Attachment in Children with Disability or Developmental Delay
#04 – Attachment & Epigenetics: Current State of the Field#09 – Attachment & Parent-Child Neural Synchrony
#05 – Attachment & Parental Sensitivity#10 – Parental Attachment, Reflective Functioning & Sensitivity
Attachment Science Topics #11 – #20
#11 – Pros and cons of brain changes in fathers

#1: Child-Mother Attachment & Language and Cognitive Child Outcomes

Meta-Analysis

Does child-mother attachment predict and mediate language and cognitive outcomes? A series of meta-analyses

Deneault et al. (2023)


Summary

In a large sample of over 9000 children, a significant association between child-mother attachment #security and child #language and #cognitive outcomes was observed. Furthermore, there was a small but significant indirect effect of maternal #sensitivity on this association.

Below are some specification questions & answers to better understand the results.

1) Why do children who form a secure attachment to their mother have better cognitive and language skills?
More secure children may be less likely to be distracted by the #relationship, and could therefore more readily focus on #exploration and #learning.

2) How strong is this association?
The magnitude of the effect sizes suggests that child-mother attachment security is influential, but not deterministic of children’s cognitive and language skills. However, even small gains in cognitive and language skills can set children up on a developmental pathway of success.

3) What about the contribution of maternal sensitivity?
The present results support the view that attachment predicts cognition and language outcomes beyond the direct and indirect contributions of maternal sensitivity. In fact, both maternal sensitivity and child-mother attachment predicted cognition and child language directly. Maternal sensitivity and child-mother attachment security may thus have distinct contributions to fostering language and cognitive skills in children.

4) But then, what is the distinct contribution of attachment & sensitivity?
The current analysis does not provide answers regarding this question, but it confirms the importance of explicitly testing both caregiver sensitivity and child-caregiver attachment.

Post on Instagram

#2: Child-Mother and -Father Attachment & Child Behavioural Problems

Meta-Analysis

Configurations of mother-child and father-child attachment as predictors of internalizing and externalizing behavioral problems: An individual participant data (IPD) meta-analysis

Dagan et al. (2022)


Summary

In a large sample of almost 1100 #children, Dagan et al. (2022) assessed whether early #attachment networks with #mothers and #fathers are associated with children’s #internalising & #externalising #behavioural problems

The authors had two main research questions:
1) Is the number of #insecure or #disorganised attachments important in predicting #child behavioural problems?
2) Are child behavioural problems better predicted by attachment to the #mother or the #father?

What did the authors find?
For internalisation, the number of organised secure versus insecure attachment relationships within an early attachment network was important. Children with either no or only one single secure attachment had higher internalising scores compared with children with two secure attachment relationships. Furthermore, there was no difference in children’s internalising scores as a function of secure (versus insecure) attachment to father vs mother.

For externalisation, the number of organised secure versus disorganised attachment relationships within an early attachment network was important. Children who were disorganised with both parents had higher externalising scores than children who had one or no disorganised attachments within their network. Again, no difference was observed regarding attachment to mother vs father.

Take-home messages:
1) It appears that it takes two – and not merely one – secure attachment to primary caregivers to buffer children from increased vulnerability to experiencing internalising behavioural problems.
2) Conversely, it seems that organised attachment to one parent is enough to offset the otherwise unfavourable heightened externalising behavioural problems that children with two disorganised attachments experience.
3) Finally, mother-child and father-child attachment relationships may not differ in the roles they play in children’s development of internalising and externalising behavioural problems. The quality of attachment patterns to mother and father seems to be equally important.

Post on Instagram

#3: Attachment & Brain Structure in Adolescence

Original Research

Longitudinal associations between self-reported attachment dimensions and neurostructural development from adolescence to early adulthood

Puhlmann et al. (2023)


Summary

In a sample of 95 typically developing adolescents (12–19 years old at study baseline), the authors used structural magnetic resonance imaging (sMRI) for up to four times during up to four years. Teenagers’ attachment was assessed with a self-report questionnaire (Relationship Scales Questionnaire – RSQ – with two dimensions: avoidance and anxiety) at study baseline and then related to changes in cortical thickness (CT) over time.

Finding 1: Change in CT over time
CT decreased as adolescents grew older, and cortical thinning was more pronounced when they were younger at study baseline. These overall findings of CT decrease correspond to a normal pattern of adolescent brain development.

Finding 2: Change in CT x attachment
Both self-reported anxious and avoidant attachment dimensions were related to cortical thinning in several brain areas (prefrontal cortical and cortical midline structures and anterior temporal cortex), and there was an additional relation with adolescent age. Specifically, more insecurely attached teenagers showed faster cortical thinning, and especially so if they were younger at study baseline.

What do these findings mean? 
The observed CT patterns point to a possible acceleration of brain – and particularly cortical – development in adolescents with insecure attachment. These findings are consistent with another study on grey matter volume change in children from age 18 months to 10-11 years, in which the authors reported that children more securely attached to their mother in infancy had larger grey matter volumes ten years later. Interestingly, in both studies, structural brain changes were predominantly found in areas involved in social cognition and emotion regulation, and may thus be related to adolescents’ socio-emotional skills.

The authors also assessed subcortical volume but their findings were less consistent. They also mention some limitations, including a lack of child & adult data, possible confounding factors and the use of self-reports. 

See here for more.

Post on Instagram

#4: Attachment & Epigenetics: Current State of the Field

Two Review Papers

(1) Attachment and Epigenetics: A Scoping Review of Recent Research and Current Knowledge

(2) Attachment-related dimensions in the epigenetic era: A systematic review of the human research


(1) Darling Rasmussen et al. (2021)
(2) Craig et al. (2021)


Summary

About 20 years ago, Meaney et al. reported links between maternal care, gene expression, and the transmission of individual differences in stress reactivity across generations by using a very clever cross-fostering experimental design in rats. Since then, the interest in such a gene by environment – or nature by nurture – interaction in humans has steadily grown, including within the field of attachment science.

Despite this growing interest, the number of studies on the epigenetics of human attachment is still very limited. Two recent reviews by Darling Rasmussen et al. (2021) and Craig et al. (2021) summarise the results of 11 and 13 studies, respectively. 

Some common themes start emerging for:
(1) Oxytocin receptor gene (OXTR) methylation. However, findings are mixed about the direction of OXTR methylation – i.e., both hypo- and hyper-methylation were associated with anxious and avoidant attachment.

(2) HPA stress axis with glucocorticoid receptor gene (NR3C1) and the FK506-binding protein 5 (FKBP5) methylation. Yet, results are mixed and linked to anxious/resistant and avoidant attachment in different ways.

(3) Serotonin signalling, including SLC6A4 and HTR3A methylation – sometimes in association with 5HTTLPR genotype – and particularly in relation to early life adversity, loss and trauma. But, once more, it is too early to draw any firm conclusions.

(4) Genome-wide DNA methylation and genetic variation in children. Attachment accounted for 11.9% of methylation, and this effect was most apparent in girls when comparing secure versus disorganised attachment.

Conclusions
Findings expand attachment theory by pointing to a complex nature-nurture interplay involving epigenetic processes for attachment development. However, what both reviews note is that in the so far available studies, there is strong diversity of terminological and methodological approaches used to assess attachment, study design, analysis approaches and included participant populations. We therefore need to be very careful when interpreting these data. 

Also see here for more.

Post on Instagram

#5: Attachment & Parental Sensitivity

Meta-Analysis

A meta-analysis on observed paternal and maternal sensitivity

Deneault et al. (2022)


Summary

One of attachment theory’s central tenets is that parental sensitivity is associated with children’s (secure) attachment, language development, positive relationships and socio-emotional skills/functioning. Several large meta-analyses confirm the above links. 

Attachment theory suggests that this is because sensitive parents pay attention, interpret, and respond to their children’s signals in an appropriate and timely manner, and take their children’s perspective when considering children’s needs.

The present meta-analysis examined two main research questions.


(1) Does sensitivity differ in partnered mothers versus fathers?

Yes – partnered mothers’ (as compared to fathers’) sensitivity was significantly higher. However, this difference was small in magnitude.

These findings support the notion that parenting behaviours are more similar than may be expected, and that the impact of maternal and paternal sensitivity on child outcomes is similar, too.

Very interestingly, the difference in parents’ sensitivity was dependent on: (a) geography – because larger in North America and the Middle East as compared to Europe; (b) time – because smaller in more recent studies; and (c) context – because smaller if fathers’ sensitivity was assessed during dyadic father-child interaction as compared to triadic mother-child-father interaction. These effects reflect recent changes in the perception of fathers’ roles as caregivers and attachment figures.


(2) Is partnered mothers’ and fathers’ sensitivity related? 

Yes – there was a positive correlation, but only with a moderate effect size. This means that other factors (e.g., parents’  & children’s personal  characteristics,  socio-economic status, etc.) may be important in explaining variance in parental sensitivity, too.

Again, geography played a role, as the correlation was strongest in the Middle East, where families tend to be more convergent in their views and share more egalitarian ideas about the frequency of engagement in daily care.

Post on Instagram

#6: Attachment & Adoption

Opinion & Meta-Analysis

1) Attachment across the lifespan: insights from adoptive families
2) Fostering security? A meta-analysis of attachment in adopted children

Raby & Dozier (2019) & van den Dries et al. (2009)


Summary

How can research with adoptive families provide novel insights about attachment processes across the lifespan? Two papers may provide some answers. 

Secure attachment in adopted children
Adopted children’s attachment seems to depend on their age at adoption. 
No difference in attachment security between children adopted before their first birthday and non-adopted children was found. 
However, children adopted after their first birthday showed significantly less attachment security than non-adopted children. 
Prolonged early adversity can have lingering consequences for later attachment. As van den Dries et al. nicely summarise: “It may be easier to prevent insecure attachment than to change insecure attachment”.

Disorganised attachment in adopted children
Adopted children showed more disorganised attachment than non-adopted children (regardless of age at adoption). 
However, the proportion of disorganised attachment was much higher in institutionalised and maltreated children. 
Adopted children thus showed an impressive – although incomplete – catch-up after their placement.

Different facets of secure attachment in adopted children
Experiencing early adversity appears to have limited effects on adopted children’s sense of security with their adoptive parents. 
However, there are long-term consequences regarding more general attachment representations as assessed by the AAI. 
Positive attachment representations to adoptive parents may thus exist alongside, rather than in place of, more abstract and global attachment representations, which appear to be more resistant to change.

Adoptive parents’ and children’s attachment
An association between adoptive parents’ and their children’s attachment behaviours and representations was found. 
This shows that adoptive parents’ attachment shapes their children’s attachment and can support recovery in attachment quality for children with histories of early adversity.

Post on Instagram

#7: Disorganised Attachment: Practical Implications

Consensus Statement

Disorganized attachment in infancy: a review of the phenomenon and its implications for clinicians and policy-makers

Granqvist et al. (2017)


Summary

In 2017, 40 child development and mental health experts released a joint statement calling for caution when referring to the concept of disorganised attachment. A very informative media statement was released alongside. 

Disorganised attachment has seen widespread interest from policy makers, practitioners, and clinicians in recent years. Unfortunately, some of this interest is based on false and misleading assumptions.

These include:
(1) Attachment measures can be used as definitive assessments of individuals in forensic/child protection settings.
(2) Disorganised attachment reliably indicates child maltreatment.
(3) Disorganised attachment is a strong predictor of pathology.
(4) Disorganised attachment represents a fixed or static “trait” of the child, impervious to development or help.

There has been a steady growth in recent years to use the assessment of disorganised attachment to screen for child abuse. Crucially, however, although disorganised attachment is more common among maltreated infants, it does not necessarily indicate maltreatment. According to a series of meta-analyses (2010), if families have five or more socioeconomic risk factors – like living in poverty, drug addiction, or not graduating from high school – rates of disorganised infant attachment are comparable to those in families where there is known to be abuse.

Important additional considerations include:
(a) Disorganised infant attachment is not a validated individual-level clinical diagnosis – as opposed to attachment disorders included in the DSM/ICD diagnostic systems. Disorganised attachment is not an infant’s fixed property but is relationship specific. 
(b) Misapplications are likely to selectively harm already underprivileged families (e.g., those raised by parents in socioeconomic adversity or with functional impairments).
(c) Attachment-based interventions and naturalistically occurring reparative experiences can break intergenerational cycles of abuse and lower the number of children with disorganised attachment.

Post on Instagram



Important Additional Considerations
Disorganised attachment refers to a specific pattern of child behaviour (usually observed during the Strange Situation Procedure). It is different from both organised attachment (i.e., secure, insecure-avoidant, insecure-anxious attachment) and attachment disorders (which necessitate a clinical diagnosis).
Disorganised attachment is also different from a fearful(-avoidant) attachment style (sometimes said to comprise elements of both attachment avoidance and anxiety), which is derived from adult self-report questionnaires in the context of romantic relationships.
There is no such thing as a (insecure-)disorganised attachment style in adults (see also here).

#8: Attachment in Children with Disability or Developmental Delay

Meta-Analysis & Review

Attachment security in children with disability or developmental delay: Systematic review of quality and interventions


Alexander et al. (2023)


Summary

A recent systematic review explored evidence regarding the attachment security of children with Disability or Developmental Delay (DD), the nature of interventions that have been employed to improve secure child-parent attachment in this population, and the findings’ implications for early childhood intervention (ECI) professionals.



Research Question 1 – What is the available evidence regarding the attachment quality of children with DD?

Overall, the data examined in this systematic review indicate that children with DD are at significantly greater risk of forming a disorganised attachment and that fewer than half will develop a secure attachment towards their parents.

DD included Autism Spectrum Disorder (ASD), nonspecific intellectual disability, neurological conditions such as cerebral palsy or epilepsy, Down syndrome, pervasive developmental disorder and hearing impairment.



Research Question 2 – What is the available evidence regarding the effectiveness of interventions used to improve attachment security of children with DD?

The systematic review found emerging evidence that interventions can improve attachment towards parents for children with DD.

There are two main implications from these findings for ECI professionals.

(1) The findings that fewer than half of children with DD were classified as secure and that well over a quarter were classified as disorganised should raise concern. ECI aims to enhance the development, wellbeing, and participation of children with DD and insecure and disorganised attachment work against these aims. ECI professionals should therefore look into ways of enhancing parent-child attachment security.

(2) The best ECI strategies to improve parent-child attachment security appear to be cognitive and behavioural strategies involving coaching and video feedback. ECI professionals should be supported to develop their capacity to promote sensitive and responsive parental caregiving and to identify when referrals for more intensive support are required.

Post on Instagram

#9: Attachment & Parent-Child Neural Synchrony

Original Research

Visualizing the invisible tie: Linking parent–child neural synchrony to parents’ and children’s attachment representations

Nguyen, Kungl et al. (2024)


Summary

In a large study now published online, we for the first time examined the links between parent-child interpersonal neural synchrony (INS) and both parents’ and children’s attachment representations. 

We asked participants to collaboratively solve Tangram puzzles. Attachment was derived from the Adult Attachment Interview (AAI) in parents and a developmentally appropriate counterpart (Picture Story Stem Battery – PSSB) in their preschool-aged children. To assess INS, we employed functional near-infrared spectroscopy (fNIRS) hyperscanning in bilateral prefrontal (PFC) and temporo-parietal (TPJ) brain regions. 

Our main two findings are briefly summarised below.

1. Attachment representations and INS
We observed higher INS in the right TPJ in daughter-parent dyads in which daughters had more secure attachment representations. We also observed higher INS in the left PFC in mother-child dyads in which mothers had more insecure attachment representations. Interestingly, such patterns were not present in behavioural synchrony coded from video-recordings.

Our data show that INS may relate to both more secure and more insecure attachment representations. More synchrony may thus not always be better – it could also reflect interaction or relationship difficulties. One explanation is that higher INS related to more insecure attachment representations may be necessary to compensate for certain behavioural interaction difficulties.

2. Parental biological sex and INS
Although we found that both mothers’ and fathers’ brain activations were synchronised with their children’s brain activations during puzzle play, father-child dyads showed higher INS in right PFC and left TPJ. Interestingly, behavioural synchrony showed the opposite pattern, being stronger in mother-child as compared to father-child dyads.

In line with our interpretation of higher INS in mother-child dyads with insecurely attached mothers, our finding of increased INS in father-child dyads may point to compensatory neural mechanisms to overcome a relative lack of behavioural synchrony.

Post on Instagram

#10: Parental Attachment, Reflective Functioning & Sensitivity

Original Research

Precursors and Effects of Self-reported Parental Reflective Functioning: Links to Parental Attachment Representations and Behavioral Sensitivity

Kungl et al. (2024)


Summary

Parental reflective functioning (PRF) is thought to provide a missing link between caregivers’ attachment histories and their parenting behaviours. Parents with high PRF can reflect upon their children’s mental experiences, giving meaning to children’s behaviour and their own caregiving experiences.

Evidence clearly supports PRF’s key role in children’s secure attachment development, which, in turn, has repeatedly been found to be crucial for children’s social-emotional development and psychosocial adjustment.

PRF’s influence on child secure attachment and social-emotional development is thought to work through its impact on parental responsiveness to their children’s affective needs – also known as parental sensitivity. Parental sensitivity involves the rapid perception, correct interpretation as well as prompt and adequate response to children’s signals.

In a new study, we assessed the links between parental attachment, PRF and parental sensitivity. We for the first time did so in parents of 5-to-6-year-old preschool children including a large sample of fathers. We obtained parental attachment representations with the Adult Attachment Interview, PRF with a self-report questionnaire and parental sensitivity coded behaviourally from video-recordings.

Our findings revealed significant relations between parental attachment, PRF and parental sensitivity. Parents with more insecure (vs secure) attachment representations scored lower on PRF, particularly if they were classified as insecure-dismissive/avoidant. A similar pattern emerged for parental sensitivity, where scores were lowest for insecure-dismissive (vs secure and preoccupied/anxious) parents. Very interestingly, these results were largely comparable between mothers and fathers as we did not find any significant sex differences.

Our findings highlight the crucial role PRF plays for parenting behaviour quality. This link is particularly relevant for practitioners working with families by emphasising the need to target PRF in prevention and intervention programs – i.e., raising parents’ interest and curiosity in their children’s mental states.

Post on Instagram

#11: Pros and Cons of Brain Changes in Fathers

Original Research

Cortical volume reductions in men transitioning to first-time fatherhood reflect both parenting engagement and mental health risk

Darby & Martínez-García (2024)


Summary

In a new longitudinal study, researchers scanned 38 first-time fathers’ brains during their partner’s pregnancy and 6 months postpartum, and collected self-report data prenatally and 3, 6, and 12 months postpartum. 

Self-report data included data on paternal antenatal attachment, paternal postnatal attachment, paternal involvement in childcare activities, as well as paternal stress, sleep, psychopathology and mental health symptoms, depression and anxiety.



Main Findings:
In men transitioning to first-time fatherhood, the researchers found significant grey matter volume reductions across the entire cortex. 

These brain changes were linked with parental involvement: men who reported during pregnancy that they felt more strongly bonded to their unborn infant and planned to take more time off after the baby’s birth subsequently showed brain changes.

Grey matter volume reductions were also found in men who spent more time with their infants after birth and who reported lower parenting stress and greater feelings of bonding with their infant. 

These results suggest that grey matter brain changes in new fathers reflect successful adaptation to parenthood. 

Importantly, however, findings also revealed a potential downside of perinatal brain change in men. Larger perinatal grey matter volume reductions predicted worse sleep and mental health after childbirth. 

Men’s brain changes may thus reflect both adaptation and vulnerability, predicting both positive parenting outcomes but also negative consequences associated with the transition to parenthood, such as insomnia and mental health challenges.



Conclusions:

By providing evidence that men’s brains change over the transition to parenthood in ways that resemble mothers’ brains, these results underscore the message that neural adaptation to parenthood does not require direct experience of pregnancy. 

And by highlighting both vulnerability and adaptation, they reinforce the importance of reducing stress and protecting men’s time during new parenthood through policies such as paid paternity leave and advocacy efforts to normalise men’s fatherhood in the workplace.

Post on Instagram


Question & Feedback Form