One of us feels secure and the other doesn’t
A secure person and a partner who is anxious or avoidant raise a common question about whether the secure one settles the other. Studies point two ways. Questionnaire ratings show little sign of it, while a filmed disagreement shows avoidant partners easing at moments when the other softens what they ask for. What matters most is what happens in a hard moment.
The first thing to look at is whom the page is about. A person can feel secure in one relationship and anxious in another, and a check result describes one day’s answers. Both people’s answers matter here, and neither is a type.
If you feel afraid of them, or of what happens when you don’t go along with what they want, that matters more than any pattern on this page. Domestic abuse services talk to people who aren’t sure yet whether what’s happening counts.
If you or someone you know is having a difficult time, free support is available.
In an emergency, call your local emergency number.
Take the attachment style check together
Patterns that could explain it
Fits if one partner is at ease with closeness and with being depended on, and signals distress directly.
One partner is at ease with closeness and with being depended on, and says so when upset. A disagreement is less likely to feel like a threat to the relationship. This describes a lean on the day and does not make a person the steady one in every moment.
Dismissive-avoidant attachment
Check whether the other partner handles strain alone and bristles when asked to change.
The other partner handles strain alone and may go silent or turn defensive when asked to change. Anger and pulling back can both rise in that moment. They may be softer in a calm hour, and the pattern can show with others too.
Anxious-preoccupied attachment
Check whether the other partner’s worry rises when the first is busy or hard to reach.
The other partner’s worry rises when the first is busy or hard to reach, and eases with contact. A calm partner can seem reassuring or, to a worried one, distant. Which it is may depend on the day.
Fits if one person’s calm seems to steady the other in a hard moment.
One person’s steadier state seems to help the other’s. The influence runs both ways, so a calmer partner can also be pulled into the other’s state. The library describes this as two people shaping each other’s state.
Check whether either partner had difficult early care and describes it in a balanced way now.
Someone whose early care was hard can still describe it in a coherent, balanced way and function securely in adult life. A secure-looking partner may be this, and the result on a check can’t tell the two apart.
Fits if disagreements break the connection and the open question is how it gets restored.
Closeness breaks often, and the restoring matters more than the break. Here the question is what each person does after a disagreement, and whether the connection comes back.
Understand it better
An example: Hana and Tom, one conversation about plans
Hana and Tom have been together four years. Hana says she has always felt easy about the relationship. Tom says he likes to sort things out on his own. On a Sunday she asks him to tell her when plans change.
The first minute Hana says, “You never tell me when plans change. I find out from your sister.” Tom goes still. “It was one time.” His voice is tight. He says he has to get something from the garage.
The sixth minute Tom comes back and stands in the doorway. Hana says she can see he likes to decide things himself, and that she still wants to know when something changes, even if it is only a text. Tom exhales and sits down. “I don’t think about it,” he says.
The twelfth minute Tom says the first minute felt like being told off. Hana says she was hurt about the sister. They agree on nothing in particular. Tom says, “I’ll try,” and Hana says she is not sure what that will look like.
The room is easier in the sixth minute than in the first. By then Tom has had time in the garage and Hana has said the same thing another way, and the story can’t tell which mattered. Nothing is settled at the end, and “I’ll try” covers a lot.
The story fits the filmed-discussion study on this page. In it, avoidant partners showed more anger and withdrawal when asked to change, and less of it at moments when the partner softened what they asked for. That study can’t show that the softening caused the easing, and one evening can’t either. The evening also shows what the questionnaire studies leave open. A calmer moment is not a settled one, and nothing in it shows that one partner’s steadiness has made the other secure. Tom still says he does not think about it, and Hana still isn’t sure.
Other details would point elsewhere. If Tom were quick to talk about it with everyone, avoidance would describe him less well. If Hana’s calm covered worry she kept hidden, the pairing would be different from the one it looks like. If Tom had sat down before Hana said anything new, the time away would explain the change better than the way she asked. One conversation can’t settle which. What happens over many, on other subjects, says more.
What the research shows
The research is split by what it measures. Questionnaire ratings of satisfaction show little sign of a secure partner lifting the other’s ratings, and one filmed disagreement shows a link between a partner’s softening and an avoidant partner’s reactions. None of it tests whether anyone’s attachment changed.
Newlywed couples with two secure partners reported better adjustment than couples with one or two insecure partners. A married-couples study found a partner’s attachment could strengthen or soften the effect of a person’s own, while a re-analysis of dating couples found no sign that one secure partner lifts the other’s satisfaction. In a filmed disagreement, avoidant partners’ anger and withdrawal were lower at moments when the partner softened what they asked for.
Among newlyweds, couples with two secure partners had better marital adjustment than couples with one or two insecure partners
In 1992, Marilyn Senchak and Kenneth Leonard studied 322 young newlywed couples taking part in a longitudinal study of alcohol use and marital functioning. Each partner completed a three-item measure of attachment style and measures of marital intimacy, marital functioning and conflict resolution behavior.
Couples in which both partners were secure showed better overall marital adjustment than couples in which one or both were insecure. Husbands and wives also tended to pair on similarity of security.
Doesn’t show Newlyweds recruited for a study of alcohol use, with a short measure of attachment. The abstract contrasts two secure partners with the rest and doesn’t give a result for a secure partner with an anxious one, or with an avoidant one.
In married couples, a partner’s attachment could strengthen or soften the effect of a person’s own
In 2004, Rainer Banse asked 333 married couples to rate how well four descriptions of attachment fit them in their partnership: secure and three insecure kinds. Each person also rated their satisfaction with the marriage.
Satisfaction could be predicted from a person’s own ratings, from their partner’s ratings and from the interaction between the two. In general, secure attachment went with higher satisfaction and insecure attachment with lower. In some pairings, the lift that goes with secure attachment, or the drop that goes with insecure attachment, grew or shrank depending on the spouse’s attachment.
Doesn’t show The abstract doesn’t say which pairings changed the effect, or in which direction, and the full paper was not opened for this page. Married couples at one point in time. It can’t show that the pairing causes the satisfaction.
In two sets of dating couples, a secure partner showed no sign of lifting the other partner’s satisfaction
In 2021, Elizabeth Lozano, Wing Yan Sze, R. Chris Fraley and Jia Chong re-analyzed data from two earlier studies of heterosexual dating couples, 172 couples in one and 194 in the other. Each person reported their own attachment style and their satisfaction with the relationship. The authors tested the idea that a couple with at least one secure partner is as satisfied as a couple with two.
Contrary to what they expected, secure partners did not appear to buffer insecurely attached partners in these data. A partner’s attachment also explained little of a person’s satisfaction beyond the person’s own.
Doesn’t show Self-reports of satisfaction at one point in time, in dating couples, from a re-analysis of earlier data. Satisfaction ratings aren’t what happens in a hard conversation. The next study looks at that.
In a filmed disagreement, avoidant partners’ anger and withdrawal were lower at moments when the partner softened what they asked for
In 2013, Nickola Overall, Jeffry Simpson and Helena Struthers filmed 180 heterosexual couples in two discussions, in each of which one partner wanted the other to change in some way. Afterward each person rated how angry they had felt in every 30-second stretch. Observers coded withdrawal by the partner being asked to change, and softening by the partner asking, meaning being sensitive to the other’s need for autonomy and showing that the other is valued.
Partners high in attachment avoidance showed more anger and withdrawal when they were the one being asked to change, and those discussions went less well. Within a discussion, their anger and withdrawal were lower at moments when the partner softened more. Across couples, avoidant partners whose partners softened more felt less anger, and their discussions went better.
Doesn’t show One lab discussion per couple, in heterosexual couples. It shows softening and easing going together, and can’t show that the softening caused the easing. The abstract doesn’t say whether the softening partners were secure, and it reports nothing on anxious partners.
Try it: the first minute
Take the last hard conversation and write a few lines about its first minute and what came after. Keep the notes somewhere only you can open.
- The first sentence, and who said it
- What each of you did in the next minute
- A moment when the tone changed, and what changed it
- What was still open when it ended
The third line is where the filmed-discussion study looked, at moments inside one conversation. The fourth is where the questionnaire studies leave off: a better moment is not the same as a settled one.
Questions people ask
Can a secure person be with an avoidant partner?
Nothing in this research says they can’t. Newlywed couples with two secure partners reported better adjustment than couples with one or two insecure partners, and a re-analysis of dating couples found no sign that a secure partner lifts the other’s satisfaction. Those are averages over many couples and say nothing about one.
Does a secure partner make an anxious or avoidant partner more secure?
None of the studies read for this page shows that. The re-analysis of dating couples found no buffering in satisfaction ratings. The filmed-discussion study looked at avoidant partners within a single conversation. None tested whether anyone’s attachment changed over time.
What is “softening” in the research?
In the filmed-discussion study, observers coded a partner as softening when they were sensitive to the other’s need for autonomy and showed that the other was valued while asking for a change. The study measured it. It did not test it as something to do, and it doesn’t say that it works for any one couple.
Does it matter which of us is the secure one?
The studies here don’t settle it. The married-couples study found that a partner’s attachment could strengthen or soften the effect of a person’s own, in ways the abstract doesn’t spell out. The dating-couples re-analysis found a partner’s attachment explained little beyond a person’s own. The two used different samples and different ratings.
Can someone be secure in one relationship and anxious in another?
The library’s attachment check notes that attachment can differ by relationship: a person can be at ease with a friend and anxious with a partner. A result describes one person’s answers on the day.
Can two people take the attachment check together?
Each person takes the same short check, and both results appear side by side on one drawing, with a note on what the library’s research says about that pairing. A result is a person’s own answers on the day. It is not a type, a label or a diagnosis of either person.
Close to this
Go deeper
Other ways people say it
“secure with avoidant partner”
“secure attachment with avoidant partner”
“secure and avoidant relationship”
“can a secure person be with an avoidant”
“secure partner with an anxious partner”
“does a secure partner help an insecure partner”
“dating someone avoidant when I’m secure”
“my partner is secure and I’m anxious”
“my partner is secure and I’m avoidant”
“secure anxious relationship”
Questions to carry
- In a hard moment between us, what does each of us do in the first minute?
- When one of us is upset, what does the other one’s tone do to the next ten minutes?
How this page was made. Built with Conscious Living AI and led by Andrew Fayad. How the library is made · Report an issue · Last changed November 11, 2026.