Since its first description by John Bowlby and Mary Ainsworth about seventy years ago, attachment theory has become one of the most comprehensive current psychology frameworks. Attachment theory describes how we initiate and maintain close social bonds with significant others and how these bonds in turn influence our social, cognitive and affective development as well as our bodily and mental health across the life span.

Traditionally, attachment is assessed by behavioural observations, interviews and self-reports. Classical examples are the Strange Situation Procedure (SSP), the Adult Attachment Interview (AAI) and the Experiences in Close Relationships (ECR) self-report. Recently, however, there has been growing interest in also better understanding the neurobiological underpinnings of human attachment. This growing interest fuelled heightened research efforts at the interface of psychology and neuroscience, which lead to the emergence of a new area of research: the social neuroscience of human attachment (SoNeAt).

SoNeAt proposes that an in-depth understanding of attachment requires experimental approaches combining many different methods so that attachment can be assessed at various observational levels. This means that SoNeAt is advocating for studies using not only the above-mentioned traditional attachment assessment tools but also state-of-the-art social neuroscience methods. Yet, why is such an approach important and what can it reveal about the nature of attachment beyond what is already known?

In their seminal 1993 study, Spangler and Grossmann for the first time studied 12-month-old infants during the SSP. They not only used behavioural observation but also measured infants’ objective physiological and endocrine responses – i.e., infants’ heart rate and stress hormone (cortisol) levels from saliva samples. In doing so, the authors could show that despite less overt distress after short separations from the mother, insecure-avoidant infants’ heart rate during separation indicated increased arousal patterns similar to secure infants. Furthermore, the authors found higher stress hormone levels in insecure-avoidant infants across the SSP. Taken together, these results indicate that although exhibiting low behavioural distress, insecure-avoidant infants were negatively affected by the SSP and that they lacked an appropriate coping strategy to deal with their distress.

The above-described study by Spangler and Grossmann is a very nice demonstration of the added value of objective physiological and endocrine measures to more comprehensively characterising attachment. But SoNeAt goes one step further. Besides physiological (i.e., heart rate) and endocrine (i.e., stress hormone) responses, it is nowadays also considering neuroimaging measures of brain activity, structure, and connectivity, as well as genetics and epigenetics. Moreover, SoNeAt urges these combined social neuroscience methods to no longer be regarded as solely providing added value but as independent and integral research tools that are necessary to fully capture the complex nature of attachment.

The benefits of combining social neuroscience methods with traditional attachment assessment tools are manyfold, especially since the attachment field currently faces a series of central questions that have important implications for future theory and research. These questions are very nicely outlined in a recent book edited by Ross Thompson, Jeffry Simpson and Lisa Berlin. They include considerations of what constitutes an attachment relationship, the best ways to measure attachment security, how internal working models operate, the importance of early attachment relationships for later behaviour, challenges in cross-cultural research, how attachment-based interventions work, and more. It is SoNeAt’s understanding that several of these questions can only be answered if also including social neuroscience methods – of course, always in combination traditional attachment assessment tools.

A very good starting point for learning about SoNeAt’s considerations is to check out the first neuro-anatomical model of human attachment (NAMA) that lies at its core.

NAMA starts by establishing prototypical attachment pathways delineating the most important components of attachment interactions as well as their sequence. In other words, NAMA asks when and why attachment interactions are initiated and which sequence of behaviours they involve – from a stressor leading to the primary attachment strategy of proximity seeking to social co-regulation to the formation of predictions and expectations as part of internal working models (IWMs).

NAMA then maps these components to activity, structure and connectivity patterns within and between four brain modules. This process is performed several times. In a first step, it is done for early / initial attachment interactions to illustrate the general principle of the behavioural sequence. It is then repeated for each (organised) attachment orientation – i.e., secure, insecure-avoidant, insecure-anxious – separately. Finally, a similar model is suggested for attachment disruption and disorganisation as part of NAMDA.

Both NAMA and NAMDA provide unique insights into the putative neurobiological basis of attachment by summarising and explaining the to date available scientific evidence.

In a nutshell, and in accordance with another very useful social neuroscience account known as the “Social Baseline Theory”, SoNeAt’s current view of attachment can be briefly summarised as follows. From the moment we are born and across our entire life span, it is our brain’s (and body’s) baseline or default state to assume the presence of others. We simply cannot survive and thrive without others’ help and resources, mainly provided by means of co-regulation or social allostasis. Our primary survival- and attachment-based strategy under distress therefore is the one of proximity seeking. If we feel unwell or threatened, it is our natural tendency to ask and seek for others’ help and support.

But what happens if our social resources are removed or if we feel as if they were removed? Under such circumstances, compensatory strategies will kick in. These compensatory strategies are aimed at enhancing our chances of survival in the short term and involve heightened energy and capacity for self-regulation and/or the intensification of support-seeking. In attachment terms, these compensatory strategies are nothing else than the secondary (insecure) attachment strategies associated with avoidance and anxiety, respectively.

Importantly, these compensatory strategies are initially beneficial as they represent meaningful and appropriate adaptations to specific environmental demands. Unfortunately, if employed chronically and/or generalised out of the specific context within which they emerged, these compensatory strategies will begin to constitute risk factors for the development of attachment disorganisation as well as physical and mental health issues.

SoNeAt’s NAMA and NAMDA further explain the above considerations in terms of the involved brain modules and neurotransmitter / hormone systems. 

As informative as the scientific insights generated by NAMA and NAMDA are, the final question of course is how they can be translated for the use of as many people as possible. A very nice first example of such translation is the BabyGro Book based on SoNeAt’s NAMA that has just been released by the UK Charity BabyGro. The BabyGro Book illustrates – in pictures and with minimal words – how babies’ brains develop, and how responsive communication between parent and baby leads to later life (mental) health and wellbeing.

It is SoNeAt’s hope that NAMA – and ultimately also NAMDA – can soon be translated into additional useful resources for parents, teachers, practitioners, and many more, and thereby help informing the community about the usefulness and necessity of understanding attachment from a social neuroscience perspective.