Recent years have seen a resurgence in interest in Attachment Theory and its applications in therapeutic practice.
Developed by John Bowlby and Mary Ainsworth, Attachment Theory explains that our relationship with our primary caregiver in early years continues to influence us in later life.
Problems in attachment are often the root cause of many common mental health problems, such as depression, anxiety and addiction. Understanding the dynamics of attachment is therefore critical for anyone working in the helping professions today.
This post will introduce some of the foundational ideas and research that Attachment Theory is built upon.
Note: this post was made possible thanks to the work of Professor Jeremy Holmes – a leading expert in Attachment Theory and its implications in therapeutic practice. You can see Prof Holmes’ latest book here, and his TWU lecture here.
The Secure Base
A key concept in attachment theory is the ‘secure base’.
This is the idea that children use their caregivers as a kind of secure base from which to explore the world.
In the 1950s, early attachment researchers observed nannies and toddlers in Regents Park (London) and noticed a common pattern. The children would explore their environment up to approximately one hundred metres away from their caregivers. Then, once they reached this imaginary boundary, they would get anxious, look back to check if their nanny was still there, toddle back for reassurance, and then go back out exploring again.
Who we derive our secure base from changes throughout the lifespan.
For example, as a young child it might be your parents, as a teen, it might be your closest friends, and as an adult, it could be your spouse.
A good illustration of how strongly our secure base affects us can be found in veterans returning from war. During wartime, a soldier’s safety is dependent upon other members of their troop – so their squad becomes their new secure base. Although adaptive during combat, this shift in attachment can cause problems when they return home. In other words, because their secure base had switched to their fellow soldiers, and they now no longer have them around, this can cause returning veterans to experience high levels of anxiety in normal day-to-day life – where, paradoxically, they are much safer than they were during wartime.
The Neuroscience of The Secure Base
Researchers have discovered that the presence of a secure base alters our brain chemistry.
Psychologist James Coan was struggling to make progress with a client suffering from PTSD, as he was unable to open up about his experiences. During one of the sessions, the client asked if he could bring his wife into therapy too, which Coan agreed to. Although it didn’t make any difference initially, everything changed when the client’s wife reached out and held her husband’s hand during the session – at which point he began opening up about his experiences.
Coan was intrigued enough to set up an experiment which would measure how holding hands with a secure base affected how people would experience feelings of threat.
So, in a groundbreaking 2006 study, he recruited ‘happily married’ couples, and put the wives in an fMRI scanner under 3 conditions: (1.) Alone, (2.) Holding hands with a stranger, and (3.) Holding hands with their husbands.
Experimenters then informed the wives they would receive an electric shock to their leg in 20 seconds, at which point Coan measured the arousal levels (blood flow to the hypothalamus and HPA axis).
Threat levels were:
- Highest when the women were alone
- Second highest when holding hands with a stranger
- Lowest when holding hands with their husband
Coan’s study provides neurological evidence for Mary Ainsworth’s concept of “the secure base”, and demonstrates just how vital it is – even in adults.
Can you think who your secure base might be?
If you’re not sure, a simple way to work this out is to imagine a major fire breaks out in the building you are working in, and you just about make it out alive.
Who is the first person you would call once you reach safety?
The chances are, this person is your ‘secure base’.
Implications for Psychotherapy
In the same way that the children in Regents Park needed a secure base from which to explore their outer environment, often when a client comes to therapy, they need a secure base from which they can explore their inner world.
Therefore, in attachment-informed psychotherapy, it’s vital that therapists first create conditions of trust, safety, and security, before they begin therapeutic work.
Mary Ainsworth, ‘The Strange Situation’ & The Three Types of Attachment
Mary Ainsworth, the co-developer of Attachment Theory, devised the ‘Strange Situation’ experiment to measure differences in attachment styles.
In this experiment, a young child and their caregiver are invited into a laboratory room which has toys for the child. Once the child is settled and playing, the caregiver suddenly leaves the room and doesn’t return for three minutes.
As you can imagine, this is extremely stressful for the child, as they have essentially been abandoned in a strange environment.
After three minutes, the caregiver returns, and Ainsworth observed the infants’ responses.
Over time, three broad patterns began to emerge:
- Secure
- Avoidant
- Anxious
In the secure pattern, the children would protest, show outrage and display how upset they were, expressing some version of the sentiment: “Are you crazy!? How could you leave me alone like that?! Don’t you know how dangerous that was?!” Then, the caregiver would soothe them, which would heal the rupture, and after some time, these infants would return to playing.
In the avoidant pattern, the children didn’t protest much, and maintained a kind of distance from their caregiver when they returned. Unlike the secure pattern however, avoidants infants did not return to playing – indicating their capacity for exploration had been inhibited.
In the anxious pattern, children would show outrage (similar to the secure pattern), and the caregiver would attempt to soothe them. However, these children would not respond to the soothing, and maintained the protest indefinitely. Exploration was also inhibited in this pattern.
A key takeaway from the experiment was that exploration had been inhibited in both the avoidant and anxious patterns.
This is important for mental health professionals to be aware of, because it indicates that clients who are insecurely attached may find it more difficult to explore their inner worlds (e.g. thoughts and feelings) than those who are securely attached.
Mary Main and Disorganised Attachment
At this stage, it’s important to note that both of the insecure patterns identified so far (avoidant and anxious), are in some sense ‘organised’ in that they enable the children to stay safe and maintain a safe bond with their caregiver – either by damping down their emotions (avoidant pattern) or by excessively expressing them (anxious pattern).
In other words, these are adaptive strategies that serve the purpose of keeping the caregiver close.
However, in later years, Mary Main (one of Ainsworth’s PhD students), noticed a fourth, less common type of pattern – disorganised attachment.
These children didn’t fall into any of the previously identified categories, and would behave in bizarre, seemingly unintelligible ways when their caregiver would return. For example, they might go to the corner of the room, rock backwards and forwards, curl themselves up into a ball, or look ‘spaced out’ and dissociated. These children would have no way of ‘activating’ their caregiver so that they would be cared for.
Main believed that they were in a kind of approach/avoid dilemma, perhaps caused by previous experiences with the caregiver. For instance, if the caregiver had abused the child, then the source of safety was also a source of threat, so their desire for seeking safety was in direct conflict with the motivation to avoid harm.
Although this pattern is uncommon in non-clinic populations (approx 5%), the incidence of disorganised attachment is very high among those suffering with severe mental health conditions – sometimes as high as 80%. Therefore, it’s extremely important for mental health professionals to be aware of, and take disorganised attachment into consideration, when working with clients.
The Attachment Dynamic
Ainsworth wanted to understand what caused these patterns, so she began observing families with newborn babies in the first year of the child’s life, to identify if there were any behavioural differences that could account for the different attachment styles.
To do this, Ainsworth visited families in their homes, and observed how the mothers would interact with their children. Then, when these children reached between the ages of 9 and 18 months, she would conduct the Strange Situation Experiment, to see if there were any links between the behavioural differences she observed in the first year, and the different types of attachment.
She discovered three important things.
First, securely attached infants had sensitive caregivers, who were responsive and attuned to the child’s needs in the moment. For example, if the mum was doing the dishes, and the child started crying, she would immediately drop what she was doing, and tend to her infant.
Secondly, avoidant infants had “fair-weather” caregivers. These mothers happily interacted when their child was in a positive mood, but distanced themselves when the infant showed distress. Interestingly, neuroimaging studies of insecurely attached mothers show heightened levels of activity in brain areas associated with disgust when they see images of distressed children, whereas this is not found in securely attached mothers – which may help explain the distancing behaviour.
Thirdly, anxious infants had inconsistent caregivers, who were ‘out of touch’ with their child’s needs. For example, the infant might be playing happily, and in no real need of anything, and the mother would come over, interrupt, and start soothing. Conversely, at other times when the child was in distress, and needed soothing, the caregiver might not respond. So, the anxious infants had no way of ‘predicting’ when they would receive attention.
We can see how both the avoidant and anxious patterns might be adaptive from the baby’s perspective. If a mother distances herself when her child shows negative emotions, then it makes perfect sense for the infant to dampen down these emotions when they occur – otherwise they’ll push the caregiver away.
Or, if the child has an inconsistent caregiver, and can’t predict when they will receive attention, then making a lot of fuss (and not being easily soothed), is the best strategy for keeping the caregiver closeby.
Thus, Ainsworth’s observational research was able to offer an explanation for the different attachment patterns she had discovered in the ‘Strange Situation’ experiment.
Developmental Pathways

Attachment research confirms the psychoanalytic idea that early childhood experiences play an important role in shaping who we become. For example, securely attached children are better learners, and more socially adept in their first years of school. Conversely, avoidant children tend to be bullies, while anxious children are often their victims. Therefore, there are important, long-term developmental consequences of early attachment patterns, as they can set you off on self-reinforcing trajectories for the rest of life.
Transference involves bringing relational expectations from the past into the present. Anxiously attached individuals tend to assume that others will not be sensitive to their needs unless they make a huge fuss and a lot of drama, whereas avoidant types may struggle with intimacy and opening up emotionally to others. A good example of this is the finding that anxiously attached people tend to over-visit their doctor, whereas avoidant types are less likely to go; putting them at increased risk of harm.
Importantly, research has also found that attachment patterns are intergenerational. In other words, securely attached caregivers are more likely to raise securely attached children, and insecurely attached parents are more likely to have insecurely attached children. Therefore, not only does our relationship with our primary caregiver influence our own quality of life, it can affect the lives of the generations that follow.
Although attachment patterns can be difficult to change, the good news is that they are not entirely ‘set in stone’ and can be changed.
Under the right circumstances, we can move from insecure to more secure forms of attachment.
Conclusion
Our theory about the world limits what we can see.
To use a metaphor, it is like we are all walking around in the dark, each with our own flashlight (theory).
Most flashlights are narrow and limited, meaning we can only see tiny fragments of the world at any given time. However, there are some that are like floodlights, that can help us see the ‘full picture’ of a scene, and how the different fragments are connected.
Attachment theory, in some sense, can be thought of as a ‘floodlight’ theory, because it can help illuminate the dynamics of almost any human relationship, and cast a whole range of psychological problems in a new light.
Understanding the basic mechanics of attachment then, is vital for anyone working in the helping professions today.
Here’s a brief summary of some of the key points we’ve covered:
First, human beings have a basic need for a ‘secure base’ from which to explore the world; both inner and outer. Therefore, it’s vital to first create conditions of safety and security, before beginning therapeutic work.
Second, Mary Ainsworth identified three main types of attachment: secure, avoidant, and anxious. Avoidant and anxious patterns are in some sense ‘adaptive’ in early life as they keep the caregiver close to the infant, but become maladaptive later in life. Importantly, exploration is inhibited in insecure attachment patterns, meaning therapy may progress more slowly in these cases.
Third, disorganised attachment was later identified by Mary Main as a fourth attachment style, which is very low in the general population (approx. 5%), but quite common in those suffering from severe mental health problems (sometimes as high as 80%). In these cases, children may be in an ‘approach/avoid’ dilemma, as they perceive that their source of safety is a source of threat too.
Fourth, the origins of attachment styles can be traced to the first year of life and are often attributed to how a mother interacts with her infant. Caregivers who are sensitive and attuned to their child’s needs are likely to raise secure children. Conversely, those who distance themselves when the child is in distress tend to raise avoidant children, while inconsistent caregivers usually have anxious children.
Finally, these patterns can have developmental consequences throughout the lifespan, and can cross generations.
The good news is that although attachment styles can be difficult to change, they can be changed, and an attachment-informed approach to psychotherapy is one of the most effective vehicles for doing that.
Follow Up Resources
- Please click here for a ‘mini-lecture’ which summarises the key ideas from this post.
- Attachment Theory and Psychotherapy (Online Course)
- John Bowlby, Attachment Theory and Psychotherapy – Prof Jeremy Holmes (TWU Lecture)
- Attachment in Therapeutic Practice – Jeremy Holmes & Arietta Slade
- Attachment Theory in Practice: Emotionally Focused Therapy (EFT) with Individuals, Couples, and Families – Sue Johnson
- The Brain has a Mind of its Own: Attachment, Neurobiology and the New Science of Psychotherapy – Jeremy Holmes
- Attachment-Based Therapy: What It Is, Techniques & Effectiveness