In short
Anxious attachment in friendships describes the same high attachment anxiety that gets written about in romantic relationships, showing up where there is no label, no agreed frequency of contact and no accepted way to raise a problem. It looks like ranking friends, reading meaning into reply times, fearing you are the one who cares more, and treating one friendship as the only reliable place to go. The Psychology.com Attachment Style Test asks mainly about close and romantic relationships, so applying a result to friendship is an extension rather than a measurement. A result describes a pattern and is not a diagnosis.
Across our Attachment Style Quiz, here is which of the 4 results people land on.
- Core pattern
- High attachment anxiety, low attachment avoidance
- Friendship version
- Ranking, reply-time reading, fear of caring more
- Why it hides
- Friendship has no agreed terms to compare yourself against
- Biggest cost
- A base of one, carrying weight it cannot hold
- What helps first
- Widening the base before fixing any single friendship
- Test
- Attachment Style Test
Why attachment shows up in friendship at all
Attachment is not a romance-only system. It describes how a person handles needing someone: whether you turn to them when something goes wrong, whether their availability steadies you, and what happens in you when they are not reachable. Fraley and Davis found in 1997 that close friendships in young adults take on exactly those functions, with peers gradually becoming the people turned to for comfort and used as a base to go out from. Grabill and Kerns found partial support in 2000 for a link between attachment style and intimacy in friendship, clear in a questionnaire study and only partly borne out in a lab-based one.
What friendship does not have is structure. A romantic relationship eventually acquires terms: a label, an implied frequency of contact, an expectation that an unresolved argument gets returned to, a shared sense of what counts as a lot or a little. Friendship has almost none of that. There is no moment where two friends define the relationship, no agreed reply time, no accepted way of saying you want to see each other more.
For someone with high attachment anxiety, that absence is the whole problem. The system is built to monitor availability, and in friendship there is nothing to measure availability against, so everything becomes evidence. A four-hour reply means something. Being the one who suggests meeting three times running means something. Not being told about the party until afterward means something. None of it has a fixed value, so the interpretation defaults to the one the system already expects.
This is also why the pattern goes unnamed for years. Almost everything written about anxious attachment is written about partners, so people recognize themselves in a romantic description and never connect it to the far more constant low-level work they are doing to keep track of their friends.
The patterns people actually describe
Ranking. A private league table of friends, with an acute awareness of where you sit on theirs. The painful part is rarely the ranking itself, it is the asymmetry: being sure that someone is in your top three and almost sure you are in their top eight, with no way to check and no acceptable way to ask.
Reading the reply. Not just the delay but the shape of it. A short message, a reaction instead of words, a reply to one part of a message and not the other. Each one gets a reading, and the readings accumulate into a verdict that the friend has no idea is being written.
The fear of caring more. This is the one people name first when they finally talk about it. It usually produces a specific strategy: stop initiating, see what happens, and treat the result as the answer. It almost never is the answer, because the other person may simply be someone who does not initiate much with anyone.
Withdrawal after an unanswered message. A quiet protest, and the friendship equivalent of the protest behavior described in romantic attachment. It is usually invisible: no confrontation, just a stepping back that the other person might register weeks later as you having gone a bit distant.
And the base of one. One friendship that holds everything, is contacted every day, and is the first place every difficult feeling goes. It is often a genuinely good friendship. The problem is load, not quality.
What it costs, on both sides
For you, the obvious cost is the mental hours. Composing and recomposing, checking, rereading, and deciding what a silence meant. Most people in this pattern underestimate the total until they try to count a week of it.
The subtler cost is that the evidence is unfalsifiable. Because nothing in friendship is defined, no amount of warmth settles the question for long, and reassurance you have to extract yourself never counts. So the friendship can be going well and the experience of it can still be anxious, which is both exhausting and confusing.
For the friend, the main cost is weight. Being somebody's primary source of steadiness is flattering for a while and heavy over years, especially when it is unspoken. People rarely say they feel it; they tend to become slightly less available instead, which the anxious reading registers as the thing it was already afraid of.
The second cost on their side is the sense of being tracked. Not in a dramatic way, but in the small accuracy of being asked why you did not reply on Tuesday, or being told that you have been quiet lately. It is often kindly meant and it still communicates that the friendship is under observation, and friendships rarely survive feeling audited.
Worth saying plainly: none of this makes anyone a bad friend. People with high attachment anxiety are often the most attentive, most loyal and most present friends in a group, and the same sensitivity that generates the monitoring is what makes them notice when someone is struggling. The pattern is a cost sitting alongside a real strength, not a replacement for one.
Why friendship repair is not couple repair
Most of the standard advice for anxious attachment assumes a partner: ask for reassurance directly, agree how you will handle contact, name the fear, return to the conversation. All of it rests on a relationship that has a format for serious conversations.
Friendship does not, and ignoring that is why so many attempts land badly. Sitting a friend down to discuss the friendship reads as an escalation, because the format itself signals that something has gone seriously wrong. The content might be mild and the framing will be heard first.
The second difference is that friendship tolerates dormancy in a way a relationship does not. Adult friendships routinely go quiet for two months and resume unchanged, and the friend who went quiet usually did not experience those two months as anything. The anxious reading treats dormancy as the beginning of an ending, and acts accordingly, which is how a perfectly healthy gap turns into an actual distance.
So the moves have to be lighter and smaller. Not less honest, lighter: embedded in ordinary contact rather than staged, specific rather than general, and asked once rather than returned to.
The repair moves that work with friends
Ask small and directly. Not are we okay, which asks the other person to reassure you about something abstract and tends to produce a reflexive yes. Ask for a thing: do you want to make this a monthly thing, I would like that. A specific request gets a real answer, and a real answer is what actually settles the system.
Space contact on purpose. If you message every day, try every other day for a month, not as a test but as an experiment in finding out what the friendship looks like when you are not holding it up. Most people discover the friendship is roughly the same, which is information no amount of reassurance could have given them.
Let one message go unanswered without doing anything. This is the hard one. Pick a specific gap, forty-eight hours, and commit to not following up, not withdrawing and not reinterpreting. The point is not patience as a virtue. It is to collect evidence that an unanswered message is an unanswered message.
Widen the base before you work on any single friendship. Three people you contact regularly changes the mathematics completely, because no one of them is carrying the whole weight and no single silence can mean very much. This is the change with the largest effect and the slowest payoff, and it is usually the one people skip.
Name the pattern once, lightly, to the friend who already knows you well. I read into reply times, it is a me thing, if I ever go quiet you are allowed to just ask. A sentence in passing does more than a conversation about the friendship, and it gives them something to do that is not reassurance.
And drop the ledger. Counting who initiated last is the single habit that most reliably turns an uneven friendship into a resentful one. If you want to see someone, ask. If you have asked three times in a row and the answer keeps being a vague later, that is the information, and it is available without counting.
When a friendship actually is one-sided
This page has spent a lot of words on misreading, so it should say the obvious thing too: sometimes the reading is correct. Some friendships are genuinely unequal, some people are happy to be contacted and never contact, and some friendships have quietly ended without anyone saying so.
The way to tell is not the feeling, because the feeling is the same either way. It is the result of an actual request. Ask for a specific, small, reasonable thing twice over a couple of months. Someone who wants the friendship and is simply unavailable will take one of the offers, or propose another. Someone for whom the friendship has become an obligation will be warm and will not.
If that is the answer, letting it become a lighter friendship is allowed and is usually what happens. Adult friendships change weight all the time. The grief is real and it is a different thing from a verdict on whether you are easy to be close to.
What the research supports, and where it stops
Adult attachment is measured as two continuous dimensions, anxiety and avoidance, in a tradition running from Hazan and Shaver's 1987 extension of attachment theory to adult relationships through Bartholomew and Horowitz's 1991 model. Attachment anxiety describes worry about availability and responsiveness; it is a leaning, not a category.
The friendship evidence is real but narrower than the romantic literature. Fraley and Davis in 1997 showed that peer relationships take on attachment functions, and Grabill and Kerns in 2000 found partial support for a link between attachment style and intimacy in friendship. That is enough to say the system operates in friendship. It is not enough to make confident claims about how any particular friendship will go.
There is a limit worth being explicit about, because it affects how you should read your own result. Psychology.com's Attachment Style Test asks how you think and feel in close relationships, especially romantic ones. Reading that result across to friendship is a reasonable extension and not a measurement of your friendships, and your pattern can genuinely differ between a romantic relationship and a long friendship.
Patterns also move. Arriaga and colleagues' 2018 review describes attachment security growing through repeated ordinary experiences that contradict the old expectation, which in friendship means many unremarkable gaps that turn out to mean nothing. A result on a self-report test is a prompt for reflection and is not a diagnosis of anything.
How to find out if this is you
The free Attachment Style Test asks how you usually think, feel and act in close relationships, using items in the spirit of the widely used Experiences in Close Relationships framework, and returns a leaning toward secure, anxious-preoccupied, dismissive-avoidant or fearful-avoidant with a plain-language explanation. If you want to use it for friendship, answer it twice: once with a relationship in mind and once with your closest friendship in mind, and notice whether the picture changes.
If the result leans anxious-preoccupied, the full description of the pattern sits on the anxious-preoccupied page and goes further on where it comes from and what tends to help generally. If what you recognize is wanting the closeness and also finding it uncomfortable, with a pull away after a good conversation, the fearful-avoidant pattern is the closer read. An optional in-depth report is available after the test for a closer look at how the four styles show up in your answers.
| Feature | What the silence probably means | What the anxious reading supplies |
|---|---|---|
| A four-hour reply | They were at work | The tone has changed |
| Two quiet months | An ordinary adult dormant stretch | The friendship is being wound down |
| You initiated three times running | They are not someone who initiates | You are the one who cares more |
| Hearing about the party afterward | It was arranged in a different group | You are on the outer ring |
| A short, flat message | They were typing with one hand | Something is wrong and unnamed |
| Not being asked for advice this time | The problem suited someone else better | You have been replaced |
Key takeaways
- Friendship activates the same attachment system as romantic relationships but supplies none of the structure, so there is nothing to measure availability against and everything becomes evidence.
- The recognizable patterns are ranking, reading reply times, the fear of caring more, quiet withdrawal after an unanswered message, and treating one friendship as the only reliable base.
- The cost to a friend is weight and the sense of being tracked; the cost to you is that the question can never be settled, because no amount of warmth is conclusive.
- Couple repair moves misfire here: a staged conversation about the friendship reads as an escalation, and adult friendships normally go dormant without meaning anything.
- Widening the base is the change with the largest effect, because no single silence can carry much weight once three people are in regular contact.
- The attachment test asks mainly about close and romantic relationships, so reading a result across to friendship is a reasonable extension rather than a measurement, and it is not a diagnosis.
Curious which pattern your answers lean toward?
Take the free Attachment Style Test for a plain-language read on which of the four attachment styles your answers lean toward.
Cite this source
Fontane Pennock, S. (2026, November 8). Anxious Attachment in Friendships: Patterns, Repair and What Helps. Psychology.com. https://psychology.com/types/anxious-friendships
Frequently asked questions
Can you have anxious attachment with friends but not with partners?
Yes, and the reverse happens too. Attachment patterns vary by relationship, and a long, stable friendship can feel entirely safe while a newer or more precarious romantic relationship brings the anxiety out, or the other way around. Note which relationship you had in mind when you answered the test, because the result will reflect that one more than the others.
Why do I feel like I always care more about my friends than they do about me?
Two things usually combine. Attachment anxiety makes asymmetry the default interpretation of any ambiguous signal, and friendship genuinely does not supply the information that would settle it. You can see your own effort completely and only see a fraction of theirs. The route out is asking for a specific thing and reading the answer, rather than trying to measure feeling from behavior.
Is it bad to have one best friend I rely on for everything?
It is not bad, and it can become fragile. The issue is load rather than quality: when one friendship holds every difficult feeling, every ordinary gap in their availability becomes high-stakes, and the friendship ends up carrying more than friendship usually does. Adding two or three people you contact regularly changes how much any single silence can mean.
How do I tell a friend I need more contact without being too much?
Ask for a thing rather than for reassurance. A monthly standing plan, a walk on Sundays, a call on the way home from work: all of these are easy to say yes or no to. Asking whether everything is okay between you is harder to answer honestly and usually gets a reflexive yes, which settles nothing and tends to repeat.
What is protest behavior in a friendship?
The same indirect bid described in romantic attachment, scaled down. Going quiet after an unanswered message, replying noticeably briefly, or dropping out of a group chat to see whether it is remarked on. It is usually invisible to the other person for weeks, which is what makes it ineffective: the message it is sending never arrives, and the distance it creates is real.
My friend has gone quiet. Should I reach out or give them space?
Reach out once, plainly and without a subtext, and then leave it. Thinking of you, no reply needed, works because it does not require anything. Repeated check-ins read as pressure, and disappearing to see whether they notice is a test rather than a question. If two or three light approaches over a couple of months get nothing, that is information you can act on.
Does anxious attachment in friendships mean something went wrong in childhood?
An adult self-report result cannot establish that. Attachment anxiety can reflect earlier experiences, later relationships, current circumstances and individual differences, and a questionnaire cannot separate them or reconstruct your history. Treat origin explanations you read online as possibilities to examine rather than conclusions, and remember that a result is a description of a pattern and not a diagnosis.
Related types
References
- Fraley RC, Davis KE. Attachment formation and transfer in young adults' close friendships and romantic relationships. Personal Relationships. 1997;4(2):131-144. doi.org
- Grabill CM, Kerns KA. Attachment style and intimacy in friendship. Personal Relationships. 2000;7(4):363-378. doi.org
- Hazan C, Shaver P. Romantic love conceptualized as an attachment process. Journal of Personality and Social Psychology. 1987;52(3):511-524. doi.org
- Bartholomew K, Horowitz LM. Attachment styles among young adults: a test of a four-category model. Journal of Personality and Social Psychology. 1991;61(2):226-244. doi.org
- Arriaga XB, Kumashiro M, Simpson JA, Overall NC. Revising working models across time: relationship situations that enhance attachment security. Personality and Social Psychology Review. 2018;22(1):71-96. doi.org
- Mikulincer M, Shaver PR. Attachment in Adulthood: Structure, Dynamics, and Change. 2nd ed. Guilford Press; 2016. guilford.com
