You’ve taken three online quizzes and gotten three different results. You see yourself in multiple categories. Your attachment style seems to change depending on who you’re with. And every article confidently tells you different things.
You’re starting to think attachment theory is just astrology for therapy people.
It’s not. But the internet has turned seventy years of developmental research into clickbait quizzes and oversimplified four-box models. Here’s what the research actually shows.
Bowlby’s Attachment Theory: Where It All Started
Attachment theory began with John Bowlby, a British psychologist working in the 1950s. He observed that infants who were separated from their mothers showed consistent patterns of distress. This wasn’t just crying. It was a predictable sequence of protest, despair, and eventually detachment.
Bowlby proposed that humans are biologically wired to form attachments to caregivers as a survival mechanism. These early bonds create what he called “internal working models” – templates for how we expect relationships to function throughout our lives.
In the 1970s, developmental psychologist Mary Ainsworth expanded Bowlby’s work with the Strange Situation experiment. She observed how infants responded when briefly separated from their mothers, then reunited. The patterns she identified became the foundation for the attachment styles we discuss today.
Originally developed to understand infant-caregiver relationships, attachment theory was later adapted to explain adult romantic relationships, friendships, and even workplace dynamics. The patterns formed in childhood don’t disappear. They evolve.
Secure vs Insecure Attachment: The Foundation
Before diving into the four types, understand the fundamental split. Attachment patterns fall into two broad categories: secure and insecure.
Secure attachment develops when caregivers are consistently responsive and available. The child learns that seeking comfort works. That expressing needs leads to having them met. That relationships are generally safe and reliable.
As adults, securely attached people are comfortable with both intimacy and independence. They trust that others will be available when needed. They can handle conflict without spiraling into panic, aggression or shutting down entirely. They don’t cling or distance reflexively.
Insecure attachment develops when caregiving is inconsistent, dismissive, or frightening. The child adapts. They develop strategies to manage the unpredictability or absence of support. These strategies made sense at the time. They were survival mechanisms.
In adulthood, these strategies persist. They show up as three distinct insecure patterns: anxious-preoccupied, dismissive avoidant, and fearful avoidant.
Research by Mickelson and colleagues using a nationally representative U.S. sample found that approximately 50-60% of adults are securely attached, while 40-50% fall into one of the insecure categories. These percentages vary by measurement method, population, and culture, but the pattern is consistent: secure attachment is typically the largest group, followed by dismissive avoidant, anxious-preoccupied, and fearful avoidant.
The Four Attachment Styles Explained
Here are the four main patterns identified in research. Most people don’t fit neatly into one box. You’re working with two continuous dimensions – anxiety about abandonment and avoidance of intimacy – not four rigid categories. But understanding each pattern helps you recognise your tendencies.
1. Secure Attachment (50-60% of adults)
Comfortable with closeness and autonomy. Trust others and communicate needs directly. Can handle conflict without catastrophizing or withdrawing. Feel stable in relationships without needing constant reassurance. This is the attachment style most people are working toward, and it’s possible to develop at any age.
2. Anxious-Preoccupied Attachment (10-20% of adults)
Fear of abandonment and need for constant reassurance. Hyper-vigilant to signs of rejection or distance. Often feel “too much” in relationships. Struggle to trust that partners truly care, even when given evidence. Read more about anxious-preoccupied attachment.
3. Dismissive Avoidant Attachment (20-25% of adults)
Excessive self-reliance and discomfort with emotional closeness. Pull away when intimacy increases. Value independence over connection. Appear emotionally unavailable or shut down when partners seek deeper engagement. Read more about dismissive avoidant attachment.
4. Fearful Avoidant Attachment (5-15% of adults)
Want closeness but fear it intensely. Swing between pursuing connection and pushing people away. Send confusing mixed signals. Both crave and are terrified of intimacy. This is often considered the most challenging pattern to navigate. Read more about fearful avoidant attachment.
These percentages come from large-scale population studies and vary depending on culture, age, and measurement method. The exact numbers matter less than understanding that insecure attachment is common, not rare, and most people don’t exist at the extremes of any one category.
What Is the Unhealthiest Attachment Style?
People often ask which attachment style is “worst” or most harmful. The research suggests fearful avoidant (also called disorganised attachment) creates the most internal conflict, disharmony and relationship instability.
Fearful avoidants want intimacy but associate closeness with pain or danger. This creates a perpetual approach-avoidance conflict. They move toward connection, then panic and withdraw. Their partners experience whiplash. They experience constant internal chaos.
But framing any attachment style as “unhealthy” or “toxic” misses the point. These patterns developed as adaptations to difficult circumstances. They were survival strategies. The question isn’t which is worse. It’s whether your attachment pattern is causing you genuine suffering or preventing the relationships you want.
From Childhood to Adulthood: How Attachment Develops
Attachment patterns begin forming in the first year of life and consolidate through early childhood, based on how caregivers respond to a child’s needs. Consistent, sensitive responses foster security. Inconsistent, dismissive, or frightening responses create insecurity.
These early experiences shape expectations about relationships and self-worth. A securely attached child learns: “When I’m distressed, help comes. I am worthy of care.” An insecurely attached child learns: “Help is unpredictable or unavailable. I must manage alone” or “I must amplify my distress to get attention.”
Longitudinal research by Fraley and colleagues shows moderate stability in attachment patterns from infancy to adulthood – roughly 60-70% of people show consistency in their patterns. But that means 30-40% do shift, usually based on significant relational experiences – supportive partnerships, effective therapy, or unfortunately, trauma that disrupts previously secure patterns.
In adulthood, attachment patterns influence far more than romantic relationships. They affect friendships, workplace dynamics, parenting, and even how you relate to yourself. The template formed with early caregivers becomes the lens through which you interpret all close relationships.
A 2024 paper in Nature Reviews Neuroscience frames attachment patterns as plastic, modifiable adaptations rather than fixed traits. Your attachment style isn’t your destiny. It’s your starting point.
Why Your Attachment Style Seems to Change (And Why That’s Normal)
One of the most common sources of confusion: “I’m secure with my best friend but anxious with my partner. Which is my real attachment style?”
Both. And neither. Attachment is relational, not a fixed personality trait.
Your nervous system responds differently to different people based on the specific cues they provide, who they remind you of from your past, and how safe you actually are with them. You might be dismissive avoidant with a partner who pursues you anxiously, triggering your need for space. But secure with a friend who respects your boundaries and doesn’t demand constant emotional availability.
This doesn’t mean attachment theory is invalid. It means your attachment response is adaptive and context-sensitive. Think of it as a nervous system map. Your body remembers where the relational potholes are and routes you differently depending on the terrain.
The secure parts of you that emerge with safe people prove your nervous system knows how to attach differently. You’re not stuck in one pattern forever.
Secure Doesn’t Mean the Same Thing for Everyone
There’s another layer to this that rarely gets discussed. Remember that attachment exists on two continuous dimensions: anxiety and avoidance. Most people classified as “secure” aren’t sitting perfectly at zero on both scales. They score low enough to fall below the clinical threshold, but they still lean in one direction. Some secure people lean slightly toward the anxious end. Others lean slightly toward the avoidant end. Under normal conditions, this lean is barely noticeable. But under significant stress, it becomes more visible.
Think of it this way: a person who tests as secure might, under relationship strain, become more clingy and reassurance-seeking (their anxious lean showing), while another secure person under the same pressure might withdraw and shut down emotionally (their avoidant lean showing). Both were “secure” on paper. But the direction they move under pressure reveals something about their underlying architecture. This is why the dimensional model matters so much more than the four-box model. Security isn’t a single destination. It’s a region on the map, and where you sit within that region predicts how you’ll respond when things get difficult.
What Attachment Styles Actually Measure (And Why Online Quizzes Fall Short)
The gold standard for measuring attachment in research is the Adult Attachment Interview, a semi-structured clinical interview that takes about an hour. Trained assessors evaluate not just what you say about your relationships, but how you talk about them. The coherence of your narrative. Your ability to reflect on difficult experiences. Whether you idealise or dismiss past relationships.
Research tools like the Experiences in Close Relationships-Revised (ECR-R) measure attachment on two continuous dimensions: anxiety and avoidance. You get scores on each, not a category assignment. This captures the reality that most people exist somewhere in the middle of the spectrum.
Most online quizzes lack psychometric validity. They’re based on simplified self-report questions that can’t capture context or nuance. They categorise you into neat boxes when the research shows most people are dimensional and mixed. They give you different results because you’re answering based on different relationships or different emotional states.
If professional researchers use hour-long interviews and dimensional scales to assess attachment, a ten-question internet quiz isn’t going to nail it. The quiz might give you a starting point for reflection. It won’t give you a diagnosis.
What the Internet Gets Wrong About Attachment
Attachment Misconception 1: “My attachment style is my personality”
Reality: It’s a relational pattern shaped by experience, not an inherent trait. Personality is relatively stable across contexts. Attachment responses shift based on who you’re with and how safe you feel.
Attachment Misconception 2: “If I can’t pick one category, something’s wrong with me”
Reality: Most people are mixed or intermediate. That’s normal. Attachment exists on a spectrum. If you score moderately on both anxiety and avoidance, you’re not broken. You’re human.
Attachment Misconception 3: “I just need to find a secure partner and I’ll be fine”
Reality: A secure partner helps, but it’s not magic. If you’re insecurely attached, you’ll still bring your patterns into the relationship. A secure partner can buffer some of the impact and provide a corrective relational experience, but you still need to do your own work.
Attachment Misconception 4: “Anxious + avoidant relationships are doomed”
Reality: This is statistically the most challenging pairing, but not impossible. A 2019 meta-analysis by Li and Chan found that anxious-avoidant couples report lower relationship satisfaction than other pairings. But with therapy and willingness from both partners to work on their patterns, these relationships can function. It requires conscious effort, not just hoping things will improve. If you’re struggling with relationship difficulties rooted in attachment patterns, professional support can help.
Attachment Misconception 5: “If I had good parents, I’m secure”
Reality: About 40% of people with relatively healthy childhoods still develop insecure attachment. Trauma outside the family, temperament differences, peer relationships, and later relational experiences all influence attachment. Good-enough parenting doesn’t guarantee secure attachment, though it certainly helps.
Attachment Misconception 6: “Once you’re secure, you’re always secure”
Reality: Security isn’t a permanent achievement. It’s partly maintained by your current relational network. Research on the “secure base” concept by Brooke Feeney and others shows that attachment security functions more like an ecosystem than a fixed trait. Your sense of security is supported by multiple attachment figures: a partner, close family members, trusted friends. These are the pillars holding up your felt sense of safety in the world.
Remove enough of those pillars and security can erode. Someone who has always functioned as securely attached might lose a parent, go through a divorce, and have a close friendship break down in quick succession. With their relational support network significantly reduced, their underlying attachment vulnerabilities — the anxious or avoidant lean that was always there but never activated — can surface. They start behaving in ways they don’t recognise. They become clingy or withdrawn in ways that feel foreign to them. This doesn’t mean their security was fake. It means security is something you maintain, not something you simply have.
Attachment Misconception 7: “My attachment style changes with different people, so attachment theory must be wrong”
Reality: That’s exactly how attachment is supposed to work. It’s context-dependent and relational. The fact that you respond differently to different people doesn’t invalidate the theory. It confirms it.
Attachment Style Pairings: What Actually Happens
The anxious-avoidant pairing creates what’s known as the pursuer-distancer dynamic. One partner (anxious) seeks reassurance and closeness. The other (avoidant) withdraws to protect their sense of autonomy. The pursuit triggers more withdrawal. The withdrawal triggers more pursuit. Both partners end up more entrenched in their patterns.
This is the most common problematic pairing seen in therapy. Research consistently shows it’s associated with the lowest relationship satisfaction. It can work, but it requires both partners recognising the cycle and actively working to change their responses. Couples therapy that addresses attachment patterns can help break these cycles.
When one partner is secure and the other insecure, the secure partner provides a buffering effect. They don’t respond to bids for reassurance with irritation or to requests for space with panic. This stability can gradually help the insecure partner feel safer. But it’s not a given. The insecure partner still needs to engage with their own patterns.
Two anxiously attached people can create high-intensity relationships with lots of reassurance-seeking. This can work if both partners enjoy that level of emotional engagement and can meet each other’s needs. Two avoidant people may feel comfortable but risk creating relationships that lack emotional depth or vulnerability.
The Neuroscience of Attachment Styles
Attachment isn’t just psychological. Brain imaging studies suggest that different attachment styles process social information differently at a neural level, though this field is still evolving.
Research by Vrtička and colleagues found that securely attached individuals tend to show balanced activation in both reward and threat processing regions when viewing social cues. They can enjoy closeness without excessive anxiety about rejection.
Anxiously attached individuals often show heightened amygdala responses to social threats and elevated cortisol under relationship stress. Their nervous systems appear hypervigilant to signs of rejection or abandonment.
Dismissive avoidants tend to show dampened activation in social reward circuits. Their brains appear to turn down the signal that closeness feels rewarding. They also show increased prefrontal cortex activity, suggesting they’re cognitively managing emotions rather than feeling them directly.
Fearful avoidants show more variable patterns in both oxytocin and cortisol – the systems involved in bonding and stress management appear dysregulated.
Here’s the hopeful part: security priming research by Mikulincer and Shaver shows these patterns are modifiable. When people are asked to visualise supportive relationships or think about times they felt safe, their defensive neural patterns temporarily reduce. With repeated exposure to secure relational experiences, these changes can become more lasting.
Can Your Attachment Style Change?
Yes. But not quickly or easily, and not completely.
The moderate stability found in longitudinal studies suggests core attachment patterns tend to persist for many people. But 30-40% show meaningful shifts, usually in response to significant relational experiences – secure partnerships, effective therapy, or sometimes trauma that disrupts previously secure patterns.
Change Goes Both Ways
That last point deserves more attention. Change doesn’t only go in the direction of becoming more secure. It can go the other way too. When previously secure people experience significant relational losses such as the death of key family members, the end of a long partnership, the collapse of a close friendship group etc. they can shift toward insecurity. This happens because their security was being actively maintained by those relationships. The internal working model that says “people are reliable, the world is safe, I am worthy of love” was being continuously reinforced by people who showed up for them. Remove those people, and the model stops getting the evidence it needs to sustain itself.
There is Hope
Research on “earned security” shows that people who had difficult childhoods can become securely attached as adults. But some studies by Roisman and colleagues suggest these individuals may still carry subtle vulnerabilities, like slightly higher depression risk compared to those who were secure from childhood. The evidence here is mixed, but change appears to be layered rather than a complete transformation. Those with a long history of consistent security from childhood do tend to have more resilient internal working models – they bounce back faster after loss or disruption. But no one is immune. Security exists on a continuum of resilience, not as a binary state you either have or don’t.
What actually helps? Attachment-based therapy, emotionally focused therapy for couples, and relationships with securely attached partners who provide consistent, safe experiences over time. Security priming exercises – deliberately reflecting on times you felt supported and safe – can also gradually shift patterns.
The important distinction: attachment wounding is not your fault. It happened to you based on circumstances outside your control. But healing is your responsibility. You can’t wait for someone else to fix it for you.
When to Actually Worry About Your Attachment Style
Not everyone needs therapy for their attachment patterns. Many people with mild insecure attachment navigate relationships successfully without professional help.
Seek help if your patterns are causing repeated relationship failures that follow the same script. If you can’t sustain intimacy or constantly need reassurance to the point where it’s creating distress for you or your partners. If you recognise the pattern but can’t change it on your own despite genuine effort.
Therapy that addresses attachment doesn’t try to erase your history. It helps you understand your specific triggers, build capacity for emotional regulation, and gradually increase your tolerance for either intimacy or independence, depending on your pattern. It creates new relational experiences that update old expectations.
The goal isn’t perfection. It’s developing enough security to have the relationships you want without your patterns sabotaging them.
Attachment styles aren’t personality types or permanent labels. They’re patterns learned early and reinforced throughout life – but they’re also modifiable with the right experiences and support. Understanding your attachment pattern is the first step. Doing something about it when it’s causing problems is the second.
If you want to learn more about attachment styles and how they show up in relationships, Attached by Amir Levine and Rachel Heller is an accessible, research-based guide that explains the science without the jargon.
How we can help
At The Private Therapy Clinic, we specialise in attachment-focused therapy and couples therapy for relationship difficulties. We also offer comprehensive autism assessments and ADHD assessments.
If you think you might benefit from speaking to someone about the issues in this article, we offer a FREE 15-MINUTE CONSULTATION with one of our specialists to help you find the best way to move forward.
