Is My Ex Just Anxious? What Social Media Gets Wrong
By Mark Ryan, BACP Accredited Psychotherapist and NCPS Accredited Relationship Therapist
I've written before about why labels like this one tend to get in the way of accountability rather than help it, and separately about what anxious attachment looks like in dating, difficult conversations, and conflict. This piece sits between the two. It's about the specific version of anxious that shows up when you're describing an ex, and it works differently from the other labels in this series. Anxious rarely gets used as an insult the way narcissist does, or as a villain label the way avoidant does. It gets used as an explanation, and a gentle one. That gentleness is exactly what makes it worth a closer look, since gentle explanations are usually the ones nobody thinks to question.
The label that lets you off the hook too
Most of the reductive labelling this series has covered works by making someone else the problem. Anxious does something slightly different when it's aimed at an ex rather than claimed for yourself. It tends to arrive with a note of sympathy, she was just anxious, he had a lot of anxiety around commitment, which sounds kinder than narcissist or avoidant. But the sympathy is doing a specific piece of work. It locates the difficulty entirely inside the other person's wiring, and quietly closes off a harder question, which is whether anything you did, or didn't do, helped produce the anxiety being described.
It helps to be specific about who tends to use this framing and when. It shows up most often from the partner who felt burdened by the anxiety rather than the partner who experienced it, which makes sense, since that's the person who needed an explanation for why the relationship felt so effortful. It's less often used by anxious partners about themselves, who more commonly describe their own experience in terms of specific fears rather than a diagnostic label. That asymmetry, in who reaches for the word and why, is itself a clue about what the label is doing.
This matters because attachment anxiety isn't a fixed, self-contained trait that switches on regardless of context. It's frequently a response to something, inconsistency, mixed signals, a partner who's warm one week and distant the next. Calling an ex anxious, full stop, without asking what that anxiety was responding to, tends to treat a relational pattern as though it were a solo performance.
There's a version of this that shows up specifically in how the phrase gets used after a relationship ends. Anxious rarely appears as a standalone accusation. It usually arrives attached to a small, sympathetic shrug, she had a lot going on, he came from a difficult family, as though the anxiety explains itself without needing anything from the person on the other end of it. That framing isn't necessarily wrong. It's just conveniently incomplete, since almost every account of someone's anxious behaviour with a specific partner also involves that partner's specific responses, and those tend to disappear from the version of the story told afterward.
Why this isn't a fixed trait the way it sounds
There's a specific, well-established distinction in the research that online content almost never makes. Attachment researchers separate a person's global attachment style, their general pattern across relationships and over a lifetime, from their relationship-specific attachment, how they relate to one partner. A large longitudinal study following nearly five thousand adults over repeated assessments found these two things move somewhat independently, someone's attachment security with a specific partner can shift meaningfully even while their broader, global attachment pattern stays comparatively stable (Dugan, Fraley, Gillath and Deboeck, 2022). In plain terms, someone can be anxious with you specifically, in this relationship specifically, without that meaning anxious is simply who they are.
This cuts against the tidy, fixed-trait version of the label that gets used after a breakup. If attachment security to a specific partner can move independently of someone's overall pattern, then an ex's anxiety while they were with you isn't automatically proof of a stable personal disorder waiting to be diagnosed. It might be. It might also have been a specific, situational response to something happening between the two of you, checking out when things felt uncertain, going quiet during exactly the moments reassurance would have helped, that produced more anxious behaviour than either of you would have predicted from that person in a different relationship.
This isn't a claim that every anxious ex was made anxious by you specifically, or that attachment patterns don't have real, stable roots that predate any relationship. Plenty of people bring a consistent anxious pattern into every relationship they're in, largely independent of how any given partner treats them. The point is narrower than that. It's that the fixed-trait explanation is one possibility among several, not the automatic default, and reaching for it without checking which one applied tends to foreclose a question worth answering, usually before either of you has properly asked it.
Why the label feels satisfying to reach for
There's a real emotional logic to why anxious lands so easily after a relationship ends, and it works a little differently from the other labels in this series. Research on how people regulate lingering feelings for an ex has found that focusing on someone's negative qualities is a common and effective way of reducing residual attachment (Langeslag and Sanchez, 2018). Anxious does this while also doing something extra. It doesn't just help you stop missing someone. It quietly excuses whatever role your own unavailability, inconsistency, or eventual withdrawal may have played, by framing the whole dynamic as something that lived inside them rather than between you.
This is a comfortable place to land, particularly if you were the one who pulled away, went quiet, or ended things. Their anxiety becomes the reason it didn't work, rather than one half of a pattern the two of you built together. The same research found this kind of reappraisal doesn't reliably make people feel better overall, even as it succeeds at reducing attachment, so the comfort tends to be partial and short-lived rather than the resolution it can feel like in the moment.
There's also a specific version of guilt this label is well suited to managing. If you sensed, even faintly, that your own inconsistency or unavailability was contributing to how anxious your partner became, filing them as simply an anxious person retroactively removes that contribution from the story. It's a more comfortable account to carry forward, and comfort isn't a small thing after a difficult ending. But it tends to trade long-term accuracy for short-term relief, which usually means the same dynamic has a real chance of repeating with the next partner, since the part of it that involved you never got looked at directly.
A specific version of this applies for anyone who ended a relationship because a partner's anxiety felt like too much to manage. That feeling is real and leaving a relationship that felt exhausting or unsustainable isn't something that needs justifying after the fact. But there's a difference between deciding a relationship wasn't right for you and deciding the other person was simply broken in a way that had nothing to do with the relationship itself. The first is a decision about fit. The second is a diagnosis you're probably not qualified to make, and one that tends to make the ending easier to live with at the cost of a less accurate account of what happened between you.
The fixed-trait explanation is one possibility among several, not the automatic default
What the label costs you
Trait labels applied to other people behave differently from labels applied to yourself. Research tracking how impressions of others change over time found that negative trait attributions become more fixed rather than less, resistant to being revised even by new information (Sanbonmatsu, Adams and White, 2022). Once an ex is filed as anxious, that filing tends to hold, and it tends to quietly absorb your own part in the story along with it. Their anxiety becomes the entire explanation, and the question of what it would have taken from you to make them feel secure stops needing an answer.
Anxious attachment often develops, or intensifies, in response to a specific kind of scarcity too. A particular flavour of loneliness that can exist inside an otherwise intact relationship often sits underneath the checking-in, the need for reassurance, the difficulty trusting that things are fine when they haven't been told so directly. Naming an ex as anxious without asking what they were lonely for tends to end the conversation exactly where it would have gotten most useful.
This loneliness doesn't always look like obvious neglect from the outside. A relationship can involve regular contact, affection, and shared time, and still leave one partner quietly certain that none of it can be trusted to last, particularly where reassurance was inconsistent, where affection arrived unpredictably rather than reliably, or where the relationship's status itself stayed ambiguous for longer than felt comfortable. None of this is offered as an accusation against anyone specific. It's offered as a reminder that anxious behaviour usually has an ecology around it and understanding that ecology tends to matter more than naming the behaviour on its own.
What to look at instead
Rather than reaching for the trait, it tends to be more useful to describe specifically what happened. Not she was so anxious, but she'd ask where I was most nights, and I never worked out whether that was about me or about something from before me. Not he needed constant reassurance, but he'd go quiet for days after any disagreement, and I never knew whether reaching out helped or made it worse. Specific behaviour holds up to scrutiny in a way a label doesn't, and it leaves room for a more complete account than either fixed diagnosis or blanket dismissal, which is usually what you want out of looking back at all.
This kind of account tends to reveal something the label skips over, that the anxious behaviour often had a shape to it rather than being constant background noise. If it clustered around specific moments, after time apart, following an unanswered message, during any conversation that touched on the relationship's future, that pattern points somewhere useful. It suggests the anxiety was responding to something identifiable and specific, even if neither of you named it clearly at the time or had the language for it back then. If it showed up everywhere, regardless of what either of you did, that's a different and more stable, trait-like picture. The shape of it matters more than the fact of it.
A question that's easy to skip when the label already feels like an answer is what your own pattern was in response. Did reassurance offered early and often tend to settle things or did it never quite land. Did kinds of distance on your side reliably precede the anxious behaviour on theirs. These questions don't excuse a relationship that was difficult to be in, and they're not about assigning yourself blame for someone else's attachment history. They're about getting an account of what happened that's accurate enough to learn from, rather than one that closes the file too early.
How therapy can help
I’d rather you didn't try to have a settled account of what your ex was before we've talked. What's more useful is looking at the specific pattern between you, what the anxiety looked like in practice, what tended to trigger it, and how you responded when it showed up. That work tends to surface a fuller picture than either the label or its absence would on their own.
Sometimes this work confirms that the anxiety was largely independent of anything happening in the relationship, a pattern that would likely have shown up with almost any partner. Other times it surfaces something closer to a specific, situational response to real inconsistency, one that might have settled with more consistency and directness rather than reflecting a fixed and permanent trait. Both are useful things to understand clearly, and neither one requires deciding in advance which you'll find or arriving already certain of the answer.
Where this work happens with both partners present, whether the relationship is continuing or you're trying to understand one that's already ended, the same principle applies. What matters isn't agreeing on a shared diagnosis of who was anxious and who wasn't. It's building an honest, detailed account of what happened between two specific people, which tends to be more useful for the next relationship than any label either of you settles on for the last one.
This is also part of a short series looking at how social media flattens clinical language into something closer to name-calling, covering narcissist, avoidant, and anxiously attached in turn. If you've read this far wondering whether the label fits, the more useful question is usually what it's doing for you right now, whether it's helping you understand what happened, or just letting you close the book on your own part in it.
If you feel like now might be the right time to talk this through with a therapist, I offer a free 30-minute consultation in person (in London) or online. You're welcome to arrive without a settled account of what happened, for either of you involved.
About the Author
Mark Ryan is a BACP Accredited Psychotherapist and NCPS Accredited Relationship Therapist working in person across central London, from rooms in Pimlico, Kensington, and Angel. He works with couples, individuals, and polycules on relationship issues including jealousy, affairs and betrayals, sex problems, and considered separation. His practice is integrative, sex-positive, and affirming of all relationship structures and identities.
References
Dugan, K. A., Fraley, R. C., Gillath, O., & Deboeck, P. R. (2022). Changes in global and relationship-specific attachment working models. Journal of Social and Personal Relationships, 39(10), 3015–3043. https://doi.org/10.1177/02654075211051408
Langeslag, S. J. E., & Sanchez, M. E. (2018). Down-regulation of love feelings after a romantic break-up: Self-report and electrophysiological data. Journal of Experimental Psychology: General, 147(5), 720–733. https://doi.org/10.1037/xge0000360
Sanbonmatsu, D. M., Adams, T., & White, P. H. (2022). Perceived changes in trait attributions to others and the self. The Journal of Social Psychology, 164(5), 588–606. https://doi.org/10.1080/00224545.2022.2136060
