Skip to content

Attachment in Adulthood: Structure, Dynamics, and Change


Useful.  In short, the concept of attachment that started with Bowlby so long ago is useful.  It’s useful for shaping parents’ responsiveness, public policy, and even understanding how adults respond in relationships.  It’s even important to understand what we can do to change our relationships in ways that are more enriching for us and for others.  Attachment in Adulthood: Structure, Dynamics, and Change walks through the world of attachment and the rich research that improves the way we understand and respond to others.


Before we start, it’s important to note that this book is over 1,000 pages, and it is permeated with research.  That’s amazing for someone who wants to get to the details and understand the full picture.  However, it doesn’t make for an easy or quick read.  If you want details, this is definitely a place to find them.  If you’re looking for something more digestible, you may want to start with Attached.


Certainly, we understand that parents pass their genes to their children, and from these, children may win or lose at the genetic lottery.  (See The Selfish Gene for more on genetics.)  However, there are also many spooky situations where the vulnerability (or strength) conveyed from a parent to a child may operate outside of the bounds of genetics.

Fetal onset of adult disease (FOAD) is one way that we know something happening to the parents impacts a child.  (See Why Zebras Don’t Get Ulcers.)  We also know that the mother’s attachment style predicts the child’s attachment style.  If a mother is securely attached, she is more likely to be able to be appropriately responsive to her child’s needs.  Conversely, if she has an insecure attachment, she’s likely to not always respond reliably to her child’s needs; through the lack of – or poor – response, she conveys a new generation of insecure attachments.

Beyond these two extremes are epigenetics.  (See In an Unspoken Voice for more.)  Switches in our genetic code become activated with stress and trauma, allowing for changes that are neither wholly environmental nor wholly genetic.

It’s also important to realize that the change can come across further generations.  A responsive grandfather or grandmother can tilt the scales back towards secure if they themselves are secure.

Mature Autonomy

Attachment is presumed to be a built-in survival system, with its origins in the need for children to be protected after birth.  The presumption is that avoidant insecurely-attached people received negative reinforcement when they indicated a need.  Anxious insecurely-attached people are presumed to have felt neglected during a time of need.  Attachment security, therefore, should come when we feel as if our needs will be met in a positive way.

The extension into adult attachment allows us to fulfil some of our needs on our own.  This may be in the independent way that we first suppose.  We can make money, get food, provide shelter and clothing, and so on.  However, there’s another, more subtle, mechanism that may also be in play.

Thankfully, most of us can recall people in our lives who were helpful to us.  Whether they offered sage advice or they materially supported us, we can find some people who directly or indirectly met our needs for psychological or physical safety.  (See The Fearless Organization for more on psychological safety.)  The presumption is that we can internalize these helpful people, turning them into the voices inside our head.  Consistent with Internal Family Systems (IFS), we have parts of ourselves that aren’t us at all.  Instead, they’re echoes of the people whom we’ve met.  (See No Bad Parts for more.)

Intrapsychic activation allows us to call these people to mind and internally simulate what we believe they would do for us – whether or not these people are still living and could respond this way.

The once-popular WWJD – What Would Jesus Do – bracelets are a tangible example.  They call on people to internalize the voice of Jesus in asking what he would do.  This may not seem like a supportive thing, but decisional anxiety is real.  (See The Paradox of Choice.)

Distance and Autonomy

At a glance, separation from others – one’s autonomy – would seem incompatible with attachment.  However, the concepts aren’t incompatible.  Consider friends you can call at a moment’s notice, and they’ll be there to help.  Maybe it’s a childhood friend you’ve not spoken to in years.  Certainly, you have autonomy with respect to them, yet somewhere in the recesses of your mind, you know that when you’re in need, they’ll be there for you.

Deactivating Strategies

There’s a natural anxiety when attachment figures are unavailable.  It was evolutionarily dangerous.  Those who’ve developed secure attachment leverage a series of deactivating strategies that allow them to address, cope, or avoid anxiety.  Some of these strategies can be adaptive – and some maladaptive.  In other words, some of the strategies make things worse and some don’t.  Avoidant insecurely-attached types use a more global avoidance strategy, trying to push down the feelings and avoiding a connection to others.  More secure strategies rely on their internal chorus and awareness that separations from attachment figures are temporary.

The problem with the maladaptive types of deactivations is that they often result in misunderstandings.  They basically take the least charitable interpretation of something.  The result is that both sides feel misunderstood and unappreciated.

John Gottman in The Science of Trust explains the opportunities we have to connect as “sliding door moments.”  They’re brief opportunities to attune to the other person – and connect.  When a deactivating strategy is in place, these opportunities aren’t squandered, they’re thwarted.

Attachment Hierarchy

Intuitively, we know it.  When there’s a critical emergency, there’s a call list.  You call someone to tell them, then you call the next person on the list and so on.  You may or may not talk to everyone at the end of the list.  You’ll process the emergency, and then stop reaching out to folks – or will do so differently.

It appears that our attachment system has a similar strategy.  We have primary connections, and we have others when our primary connections aren’t available.  Richo suggests in How to Be an Adult in Relationships that we get no more than 25% of our nurturance from one other person – thus we have built-in backups.

Having “attachments in reserve” gives us the chance to get our needs met even if those who are our current attachment figures are unavailable.  It’s a way to minimize the chances that our needs won’t be met – and thus it’s a way to protect ourselves from an attachment-related trauma.

Attachment-Related Traumas

A person who has all their needs met always should, by definition, be securely attached.  They believe that there will always be others to help them.  However, reality is rarely that simple.  There are times when parents are sleeping that a child remains hungry for some time.  There are times when parents are having a bad day and are short with a child who asks for a glass of milk.

These traumas – big or little – impact our ability to become and remain securely attached.  There is no malice in either of the above gaps in responsiveness.  Both are reasonable, if not desirable, responses to life.  However, from the point of view of the child, they’re a temporary tragedy.

We encounter similar things as adults.  Our spouse is uninterruptable, because they’re giving a presentation at a conference (hypothetically).  The result is that our needs need to be redirected to someone else, like a close friend, or need to be temporarily deferred.  (See The Marshmallow Test.)

There are, of course, more malicious conditions as well.  We see situations where our trust is betrayed.  (See Trust => Vulnerability => Intimacy, Revisited for more on trust and betrayal.)  The are also cases of abuse.  While some abuse clearly crosses moral boundaries and social norms, much of what constitutes abuse isn’t so clear.  (See Moral Disengagement and How Good People Make Tough Choices.)

Perhaps one of the most curious aspects of those who have experienced relational-attachment trauma is their tendency to play it out over and over again in their lives.  The daughter of an abusive father gets into relationships with abusive men.  Several researchers have suggested that this might be so they can change the outcome.  If they can make their boyfriend less abusive, then perhaps it could be done for their father, too.  Others are suspicious of this claim and propose that it’s just what they know.  Either way, there’s a recognizable pattern of creating the situations that led to prior abuse in a person’s current life.

Service to Heal Ourselves

There’s a curious tightrope of service when it comes to our health and the development of secure attachment.  On the one hand, when we serve, we make it more likely that we expect that others will serve us in our time of need.  Because of the availability of thoughts, it’s easier to think of times when we’ve helped others and when others help us when we’re serving others.  If this were all there was, then encouraging service to others would have no downside.

However, there is a downside.  The downside is that sometimes people can become overly involved in helping others.  It’s all about trying to heal their wounds in a way that doesn’t work.  Consider the situation of an abused child (from the child’s perspective).  They grow up longing for parents who love them and support them in the way they want.  As a service to others, they join a child advocacy group as a volunteer who teaches parents how to better respond to their children.

One evening ,they’re working with a family and the father says something nearly identical to what the volunteer’s father says – and they’re triggered in a moment.  The thing that the father said doesn’t need to rise to the level of abuse – or even be a dysfunctional response.  The volunteer, suddenly triggered, becomes the protector for the child – who doesn’t even need protection.

What happened in that moment was that the child was no longer a client child.  The child became them – and they were going to protect and rescue the younger version of themselves.  They’ve moved from the helpful perspective of knowing that there are “helpers” (see The Good Neighbor and The Mister Rogers Effect).

We serve others, because we want to believe that others will be there when we’re in need.  We wanted someone to come to our aid back then, so by serving others in some small way, we might visualize someone providing that support we needed.  It’s important to find the right balance of this so that it doesn’t trigger or become about the person serving.  It is easier access thoughts of helpfulness in memory when you’re serving than when you’re not – so serving others may provide better healing.  That’s great as long as it doesn’t become the focus.

Tolerably Accurate Working Models

When Gary Klein was studying fire captains and their ability to make good decisions while commanding firemen in each new fire, he didn’t discover that they had perfect working models of how fires worked.  (See Sources of Power for more.)  What he discovered is that they had “good enough” working models of fires.  Good enough for them to predict, plan, and direct.  He makes the point that, when they encountered a fire that didn’t do what they expected, they often pulled teams back so they could reevaluate and adjust their model – to keep the men safe.

What Klein found was tolerably accurate working models.  When their expectations were violated, the models were no longer tolerably accurate, and they required revision.  Even in the less impactful, day-to-day life, we deal with tolerably accurate working models – and use them until they’re no longer tolerably accurate.

Anomalous Attachment

Attachment is, at its heart, a form of trust.  It’s a trust that people will be there for you when you need them.  It is not unexpected, then, that the degree to which people show attachment behaviors might differ from person-to-person just like trust varies person-to-person.  Trust is a measure of the degree to which we believe we can predict another person’s behavior.  (See Trust=>Vulnerability=>Intimacy, Revisited for more on trust.)  In the case of attachment, it’s a prediction of the degree to which they will be available to meet our needs.

In an adult sense, we describe a sense of attachment as secure or various forms of insecure.  That represents the baseline attachment that someone has in general – separate from their specific attachment relationships, which may exhibit more- or less-secure attachments than their baseline.

Moving to Secure Attachment

Perhaps the most important part of studying attachment is the capacity to change attachment from insecure to secure, because secure attachment is associated with more positive, helpful behaviors.  There are some clues to how we can do that.

The first clue comes from studies of foster children and children who were adopted.  When placed with securely attached adults, the children started to move from their insecure starting points towards secure attachment.

Similarly, research indicates that insecure people in secure relationships will tend towards security.  It seems that placing people in secure relationships causes their attachment to move towards security.

Anti-Spoiling Children

There’s a tendency towards coddling children in today’s society.  It’s been that way for decades.  Spock proposed a more permissive approach to child care in The Common Sense Book of Baby and Child Care, a perspective that Mihaly Csikszentmihalyi says he regretted later in life.  (See Finding Flow.)  Maybe that’s why college professors Jonathan Haidt and Greg Lukianoff were so critical of the expectations of college age children in The Coddling of the American Mind.  The concern is that our focus on not allowing children to struggle or be challenged may have been the wrong thing.  In Posttraumatic Growth, we discovered that chicks that are “helped” out of their shell often die, because they presumably need the lessons learned from breaking free from their shells.  Similarly, sea turtles that are “helped” to the sea die, because they need the struggle to the water to calibrate their sense of direction.

One might think that John Bowlby, with the focus on attachment and people feeling secure that their needs would be met, would believe in a permissive parenting strategy.  However, he felt that some struggle was good and even necessary.

Many recall Fred Rogers from watching Mister Rogers’ Neighborhood as a child – or as an adult with a child.  Given his attentiveness to children, one might think that he believed children should always get what they want.  However, this is not the case.  He, too, believed in limits, boundaries, and discipline.

It’s important to recognize that it’s not necessary to spoil a child nor to never say no to create a securely attached child.  Even Creative Confidence acknowledges that, sometimes, constraints are helpful and even necessary.

Insecure Feedback

It’s odd to think that insecurely attached people prefer negative feedback more than securely attached people – but it makes sense from the point of view of importance.  To the insecurely attached individual, they either didn’t get their needs met – or they were met with an attack.  For them, misreading or failing to read the situation had a consequence.  Their ability to read a situation could have become a matter of life and death.

Abused and neglected individuals must believe that there’s something they can do to resolve the situation, because the alternative – that the world is really awful – is too difficult to bear.  (See Trauma and Recovery.)  As a result, they become hypersensitive to signals that they can use to adjust their behavior to get their needs met.  The kind of feedback most likely to help them identify the problem?  Negative.

Another explanation is, of course, that they are used to negative feedback, so it’s the kind of feedback that most closely matches their internal perception.  The feedback that you get that matches your internal perspective is the feedback you’ll appreciate most.

Strangely, if you want to get better, negative feedback is far more helpful.  (See Peak and The Psychology of Not Holding Children Accountable.)

Attachment and Hostility

If attachment is good, how can bad come from of it?  The answer comes from the “us versus them” mentality that sometimes develops.  Attachment in Adulthood says, “attachment to one’s ingroup does not require hostility toward outgroups.”  Whether it’s required or not, it occurs frequently.

Fear and Attachment

The concept of attachment is necessarily and inseparably linked to safety.  We feel secure attachment because of the safety we feel in others’ willingness and capacity to come help us when we need it.  Fear, conceptually, is about the probability of an outcome multiplied by its impact divided by our capacity to cope.  (See Emotion and Adaptation.)  Our capacity to cope is somewhat about our internal capacities but more broadly about the help that others can and will bring.

Consider the college-age child of a wealthy family who knows their parents will always take care of them, compared to a child for whom there is little or no safety net.  How will be they behave differently and in ways that demonstrate secure attachment or not?  (See Never Broken for more.)

Insecure attachment is then a state where the outcomes we fear are larger than the capacities that we believe we have, internally and externally, to cope.  While it may not be certain, it’s a concern that we have.

The Limits of Compartmentalization

There are natural and normal defenses to trauma (overwhelming experiences).  One of those is compartmentalization.  (See Trauma and Recovery for more.)  However, even Bowlby acknowledges that compartmentalization has limits, saying “people’s segregated mental systems cannot be hidden from conscious awareness indefinitely and that traumatic events can resurrect distress that had been sealed off from consciousness.”  This is similar to the expectation in No Bad Parts that we need to reintegrate parts of ourselves to be “whole.”

Emotional Recovery

In Dialogue, Richard Moon is quoted as saying, “It is not … that the great masters of aikido never lose their center. They only discover it sooner and recover it faster than novices.”  While the idea that we’d never be emotionally disturbed is ideal, if we’re going to be disturbed, it would be great to discover it faster and recover faster.  Leadership and Self-Deception expresses it differently.  There, it’s about discovering you’ve got yourself in a “box” – and learning how to get out sooner instead of inviting others in.

What’s interesting about securely attached individuals is that they do emotional recovery faster than insecurely attached individuals – and this includes situations that require them to adapt their thinking.  Being safe is a foundation for learning – even learning to work through our emotions.  (See Helping Children Succeed.)

Active Avoidance

Works like The Invisible Gorilla explain how focus on one thing can leave you vulnerable to unintentionally ignoring another – like a gorilla.  However, there’s a more active system at work when avoidant insecurely-attached people are confronted with what they perceive to be threatening material.  It’s not like they’re ignoring it, it’s like they’re actively blocking it.  It’s like the subjects in Asch’s experiment, when it was replicated, who seemed to really believe that an incorrect line was the same length as another.  (See Unthink.)  It’s not that they’re lying to us – it’s that their perception has been altered.  (See Telling Lies for more about lies.)

Incognito makes the powerful point that our brain can make up answers.  Equally spookily, our brains can actively resist the information that it doesn’t want to hear.  That’s one of the reasons that people get caught Going to Extremes.  They literally cannot see disconfirming – or in this case, threatening – information.

Breakdown of Behavioral Corrections

People compare what is currently happening in the environment and themselves and compare it to a goal state.  In short, they evaluate whether they’re being effective at achieving their goals.  This is at the core of the work we did on Extinguish Burnout.  We said that burnout is the result of the comparison between expectations and reality.  When you didn’t have as much efficacy as you expected, it was burnout.  This same force seems to be at work in attachment as well.

Cognitive Control

Ultimately, secure attachment conveys benefits.  A person who is securely attached is less impacted by stereotypes.  (See A Class Divided for more on stereotypes and discrimination.)  It’s as if the securely attached person has an integrated self-image that’s resistant to being pushed.  (See Braving the Wilderness for more on integrated self-image.)

Additionally, they appear to be able to loosen cognitive control, making them more able to make unusual associations.  These are the kinds of associations that people would call creative (see Creative Confidence) or innovative (see The Innovator’s DNA).

There is a richness that is expected and unexpected that’s worth exploring in Attachment in Adulthood.