Anxious vs Avoidant Attachment: What Each One Is Actually Afraid Of

If you've read a few articles on this, you know the cartoon. Anxious people chase. Avoidant people run. Everyone ends up in a group chat asking whether "k" is a bad sign.

The cartoon gets the behavior right and skips the reason. Chasing and running are what you can see. Underneath each one is a fear, and the two fears are closer to each other than the behavior suggests.

So here are the two of them side by side, with what the research says (and where it's shakier than the internet admits), and then what you can do about your half. One thing first: nobody can diagnose him from a screenshot, me included. Use this to understand the pattern. Don't use it to build a case.

The short version

Anxious attachment turns the alarm up when someone feels far away. Avoidant attachment turns it down when someone feels close. Everything else follows from that.

These are tendencies, not boxes. Most people land somewhere in between, and plenty of us recognize both columns depending on the week. If you're not sure which one is yours, the Attachment Style Quiz is a decent place to start.

What the anxious side is afraid of

The fear is being left, and under that sits a quiet suspicion that you'd be the reason.

Phillip Shaver and Mario Mikulincer, the psychologists whose work most of this rests on, describe anxious attachment as doubt about whether a partner will be there, mixed with doubt about whether you're lovable enough to keep. That pairing is why a slow reply doesn't register as "he's busy." It registers as evidence.

They call the resulting strategy hyperactivating. The attachment system never fully stands down. You scan for signs of distance, the threshold for spotting one drops, and once you find something your mind keeps chewing on it. Think of a smoke detector set so sensitively that toast sets it off.

Then there's the part nobody warns you about. The thing that quiets the alarm is the thing that keeps it loud. Relationship researchers have a name for asking for the same comfort again and again even after you've gotten it: excessive reassurance seeking, best known from James Coyne's work on depression. Shaver and Mikulincer note that partners can come to find it intrusive and demanding. In my experience the relief itself lasts a couple of hours, which is part of why the asking never ends.

None of this means you're doing it wrong. The reaching comes from a real need to know you matter. It just works better when some of the settling happens inside you, instead of living entirely in his replies.

What the avoidant side is afraid of

The fear is being swallowed, or being let down the second you finally lean on someone. Often both.

In Shaver and Mikulincer's model, avoidance starts with doubt that other people come through when it counts. The strategy that follows is the mirror image of the anxious one. They call it deactivating: keep the attachment system quiet by playing down needs, keeping some distance, and pushing attachment-related thoughts out of the room. John Bowlby, the British psychiatrist who started attachment theory, called the adult version compulsive self-reliance, which is a more honest phrase than "independent."

You can see it in the lab. In stressful-task studies, avoidant partners were more reluctant to seek comfort from the person they were dating. In week-long diary studies, avoidant participants described feeling bored and unengaged in social interactions where anxious participants were busy worrying about being rejected. Same room, very different weather.

One myth is worth retiring, which is that avoidant people just don't feel much. The earliest evidence comes from babies, because that's where Mary Ainsworth first spotted the pattern. In her lab procedure, about 15 to 20 percent of American infants looked away and played with toys when a parent left and came back, instead of reaching for comfort. For a while the assumption was that they simply weren't bothered. Then researchers measured. In a 1993 study, avoidant babies showed less visible distress during separation, yet their heart rates looked much like the secure babies', and cortisol was higher in the avoidant and disorganized babies. Attachment researcher and clinician Judith Solomon puts it plainly: the babies who looked unruffled were hiding their distress. Ainsworth also found that their caregivers tended to be less sensitive and more rejecting when the baby was upset, and that at home these same babies cried and got angry more than you'd expect.

Those are infants, and nobody has strapped a heart monitor to your ex. Treat this as a reason to doubt "quiet means calm," not as proof of what he's feeling. But the next time someone tells you avoidant people don't care, the lab would like a word.

What the research says, and where it gets shakier

The two-style story is a useful map, and the researchers who drew it will tell you it's a simplified one.

Cindy Hazan and Phillip Shaver took Ainsworth's infant patterns and applied them to adult romantic love in a 1987 paper that set off decades of research. Kim Bartholomew then split avoidance into two kinds: dismissing, where closeness feels unnecessary, and fearful, where closeness is wanted and scary. The fearful version is the person who wants you near and doesn't want you any closer, which I wrote about in Can Someone Have Both Anxious and Avoidant Attachment Styles.

The finding that changes how you hold all of this came next. Chris Fraley and Niels Waller analyzed a large body of adult attachment data in 1998 and found no evidence of true attachment types. What exists is two continuous dimensions, anxiety and avoidance, and the "styles" are regions on a map. Shaver himself warns that the shorthand can mislead people into thinking in types. So when a quiz says you're anxious, it means you scored high on one dimension. You haven't been sorted into a species.

Two more caveats worth knowing. Attachment isn't fixed: a meta-analysis of 127 papers on security with parents, tracked from infancy into early adulthood, found moderate stability (r = .39) and no significant stability over gaps longer than 15 years. And it isn't one setting. Shaver and Mikulincer's review notes that reports of attachment orientation can shift with context and recent experience, and that a supportive or unsupportive person can switch on the matching pattern in you even when it isn't your usual one. That's why you can be the one chasing with one partner and the one pulling back with another.

Why these two keep finding each other

The research on who ends up with whom is thinner than the internet implies. One often-cited study, by Lee Kirkpatrick and Keith Davis, followed 354 dating couples and found the pairings weren't random: there were no anxious-anxious couples and no avoidant-avoidant couples at all. They also found that relationships between avoidant men and anxious women stayed together unexpectedly well over three years, even though both partners rated them poorly at the start. That's one study of heterosexual couples from the nineties, so take it as a clue and not a law.

What's better documented is what happens once the two meet. When conflict arrives, Shaver and Mikulincer's review describes anxious people's hyperactivating strategies as intensifying it, while avoidant people's deactivating strategies pull them away from it. Each move is the other person's trigger. Two people doing exactly what their own nervous system recommends, in sequence, indefinitely.

Couples therapists have been naming this for decades. Sue Johnson, who developed Emotionally Focused Therapy, calls it the Protest Polka: one partner reaches for connection, even if it comes out sideways, and the other steps back, which makes the reaching louder. In that framing the protest is a bid for connection and not an attack, and the withdrawing partner is usually feeling shut down and like a failure while trying not to make things worse. Researchers who study couples call the same dance demand-withdraw, and Shaver and Mikulincer describe it as destructive to relationship stability and satisfaction.

If you want to watch it unfold step by step, the Push Pull Cycle Visualizer and the page on anxious avoidant relationships walk through the four stages.

What to do about your half

You can't change his column. Yours is the only one that responds to you.

If you're the anxious one, settle first and ask second. When the alarm goes off, the pull is to fix it through him. The research on reassurance seeking points to a loop: the asking repeats even after the answer comes, and the person on the other end tends to tire out. In my experience the comfort wears off within hours, too. So get your body steady before you reach for the phone. Feet on the floor, a longer exhale, something to do with your hands that isn't scrolling. The 90 second move is the quick version, and the nervous system regulation page explains why it works. Then, if you still need something from him, ask once and ask plainly. Needing something is fine. Asking five times from the middle of a spike is what tends to land as pressure.

I'm pro self-soothing and wary of living on reassurance, and I know that's not the popular take. Needing comfort is fine. I just want some of it to be comfort you can give yourself, because that kind doesn't depend on anyone's reply time.

One exercise borrowed from EFT counselors: fill in this sentence about your last argument. "The more I ___, the more you ___, and then the more I ___." Writing the loop down turns it into something the two of you are in, instead of something one of you is doing to the other.

If you're the avoidant one, or you can't tell which column you're in, the best-supported tool comes from the Gottmans' lab. When your heart rate spikes mid-conflict (they use roughly 100 beats a minute as the signal), Julie Gottman, a clinical psychologist, recommends saying you need a break, framed around yourself and not your partner's state, and telling your partner when you'll be back. The return time is what tells them they aren't being rejected or abandoned. She suggests 20 to 30 minutes, 24 hours at the outside, and the break only works if you spend it calming down instead of rehearsing the fight. "I need some space, I'll check in tonight" does more than any amount of silence.

Can this change?

Yes, with an honest asterisk. Researchers use the term earned secure attachment for people who began insecure and later became secure, and the evidence is encouraging but still being worked out. I went through what's known, and what isn't, on the Rewire page. What both the research and my own work point to is repetition: patterns shift through repeated, felt experiences of being met differently, and one good conversation doesn't do it.

If you read both columns and saw yourself in each, that's common enough to have a name. Bartholomew called it fearful avoidant, and many clinicians call it disorganized. I wrote about it in Can Someone Have Both Anxious and Avoidant Attachment Styles, and the disorganized attachment page has the full picture.

If you want help with your half

Reading about the pattern helps, and it tends to stall at the moment the alarm goes off at 9pm and none of the reading is in the room with you. That's the part I work on with women, in the body, using the Healing Hearts Method. An Attachment Clarity Call is a free conversation where we look at what's actually happening for you and whether working together makes sense. No pitch.

Sources

• Shaver and Mikulincer, Attachment Theory, Individual Psychodynamics, and Relationship Functioning (draft chapter, UC Davis)
• Shaver and Fraley, Self-Report Measures of Adult Attachment (types versus dimensions, Fraley and Waller 1998)
• Spangler and Grossmann, Biobehavioral Organization in Securely and Insecurely Attached Infants (Child Development, 1993, abstract)
• Judith Solomon interview, The Psychologist (British Psychological Society, 2017)
• Attachment stability meta-analysis, 127 papers (abstract)
• Kirkpatrick and Davis, Attachment Style, Gender, and Relationship Stability (Journal of Personality and Social Psychology, 1994)
• Stopping the negative cycle of protest-withdraw with Emotionally Focused Couple Therapy (Counselling Directory)
• How to Fight Without Hurting Your Relationship (Greater Good, interview with John and Julie Gottman, January 2024)
• Reassurance Seeking (Encyclopedia of Human Relationships, SAGE)