How to love a high-functioning partner

The competence was never the problem

In May 2026, Thomas Curran and Andrew Hill published an update to their work on perfectionism in Psychological Bulletin — 307 samples, 82,939 college students in Canada, the US and Britain, between 1989 and 2024. Personal standards rose over that period, modestly: d = 0.45. Concerns over mistakes rose by d = 0.80, roughly twice as much, and by far the largest change they measured.

Two different things. Wanting to do well went up a little. Dread of falling short went up a lot. If you are living with someone who works past midnight on something already finished, you are not living with high standards. You are living with a person who cannot afford to be seen falling short, and the difference matters because only one of those is treatable in a marriage.

Paul Hewitt and Gordon Flett measured what that dread does socially. Their work on perfectionistic self-presentation separates three moves: promoting your successes, refusing to display imperfection, and refusing to disclose it. In a clinical sample of 468 people, nondisclosure of imperfection uniquely predicted depression (r = .44) even after controlling for perfectionism itself. Nondisplay predicted social phobia at r = .51.

The drive to look flawless is measurably a drive to conceal, and the concealing is what does the damage. Which means the thing standing between you and your partner is not their competence. It is the machinery they built to keep anyone from seeing underneath it — and you live inside the blast radius of that machinery without having agreed to.

They are less happy than you think

Ovidiu Candel and Maria Turliuc pooled 148 independent samples and 71,011 people to separate two questions that usually get muddled: does your attachment style predict your satisfaction, or your partner's?

	Predicts own dissatisfaction	Predicts partner's dissatisfaction
Avoidant	r = −.44	r = −.14
Anxious	r = −.32	r = −.12

The gap is not a rounding error. Formally tested, the actor effect beat the partner effect at z = −10.75. The self-sufficient one is roughly three times more miserable about the relationship than they are making you.

I want to be careful here, because “high-functioning” and “avoidant” are not the same thing and plenty of accomplished people are genuinely, easily close to their partners. But when the pattern is avoidance — the self-reliance, the deflected question, the door that closes quietly — the research says something most partners have exactly backwards. The flatness is not contentment. It is what dissatisfaction looks like when a person has no practised way to say so.

There is a second cost, and it lands on you. In 2025, Robert Körner and colleagues studied 1,256 couples across Israel, Germany and New Zealand. One study put couples in a lab and had observers code the conflict discussion; another ran a 21-day diary. One person's avoidance predicted their partner's lower sense of power in the relationship, and observer-coded withdrawal was the mechanism carrying it. Withdrawal does not just remove them from the room. It removes your ability to affect anything at all.

Half-measures make it worse. This is the finding I would tape to the fridge.

Yuthika Girme, Nickola Overall, Jeffry Simpson and Garth Fletcher ran four studies with 298 couples, three of them observing actual support conversations, and found something that reverses ordinary intuition. As practical support moved from low to moderate, highly avoidant recipients got worse on every measure: more distress, more perception that their partner was controlling or critical, more distancing, lower confidence in themselves. As support moved from moderate to high, all of it reversed.

Their explanation is that high levels of practical support overcome an avoidant person's defences by offering evidence of availability that is too obvious to argue with. A tentative offer can be read as interference. An unmistakable one cannot.

Lucia Vedelago and colleagues found the matching result in 184 cohabiting couples: for avoidantly attached partners, solution-focused coping improved relationship satisfaction, while emotionally intimate strategies provided no protective benefit at all. For anxiously attached partners it ran exactly the other way, and problem-solving actively backfired.

The move: stop making tentative emotional offers to a self-reliant partner and expecting them to land. Cancel something real. Take the whole thing off their plate, not the easy half. If you are going to do it, do it at a size they cannot reinterpret as criticism.

Show up when they win, not only when they lose

Shelly Gable, Gian Gonzaga and Amy Strachman videotaped 79 couples discussing recent good news and recent bad news, coded the responses, and followed the relationships. Responses to the positive event discussions were more closely related to relationship well-being and to breaking up than responses to the negative ones.

Sit with that if your partner's whole identity runs through achievement. The instinct of a loving partner is to be magnificent in a crisis — and the crisis is the moment they are least able to receive you. The promotion, the closed deal, the case that went well: that is the bid, and it usually arrives disguised as something boring told badly at 9pm.

The bids-and-turning-toward language most people know comes from John Gottman's apartment-lab research, published as a coding system by Janice Driver and Gottman in Family Process in 2004. Worth knowing: the famous percentages attached to it — the 86% and 33% you will find on a hundred websites — do not appear in that paper or in any peer-reviewed source I can find. The construct is real and reliably coded. Those specific numbers come from a trade book. I would rather tell you that than repeat them.

What is well evidenced is the loop. Tatum Jolink and colleagues analysed 824 people across four datasets. Feeling responded-to on a given day predicted more affectionate touch that day, and more touch predicted the partner perceiving you as more responsive the next day. It runs in both directions, which means either of you can start it.

The feelings are there. The outward channel is jammed.

The cliché says men cannot identify what they feel. The largest analysis available says otherwise. Jon Mendia and colleagues pooled 120 studies and 88,721 people in 2024 and broke alexithymia into its parts:

• Difficulty identifying feelings: d = −0.24, in men's favour — women scored slightly higher on this one
• Difficulty describing feelings to another person: d = 0.26, men higher
• Externally oriented thinking: d = 0.49, men higher — the only medium effect in the set

He is not blank inside. He is oriented outward, toward the problem and the task, and he is moderately worse at converting what is happening in him into words aimed at you. Those are different faults with different repairs.

Michael Lyvers and colleagues then found the mechanism in 158 adults in relationships: alexithymia correlated with relationship satisfaction at r = −.45, and the effect ran entirely through fear of intimacy. Once fear of intimacy was in the model the direct path stopped being significant. Not being able to name it and being frightened of closeness are the same phenomenon wearing two hats. Some of this is what I have written about elsewhere in why men wall off when they feel challenged and in the strong, silent type.

The move: stop asking “what are you feeling?” It requires the exact translation step that is jammed. Ask something answerable instead — what was the worst ten minutes of that day, what did you think was going to happen, what did you decide not to say. The feeling arrives on its own once the facts are out.

Repair is the skill. Not the absence of rupture.

The best number in this whole field comes from psychotherapy research rather than couples research. Catherine Eubanks, J. Christopher Muran and Jeremy Safran meta-analysed 11 studies and 1,314 patients on what happens when the therapy relationship ruptures and gets repaired: r = .29, d = .62, a moderate and reliable relation between resolving the rupture and a good outcome.

A second meta-analysis in the same paper looked at whether training clinicians in rupture repair improved outcomes. Six studies, 276 trainees, r = .11, not significant. Doing the repair matters. Being taught to do it, on this evidence, did not.

The couples version is older and smaller. Gottman and Robert Levenson had 79 couples discuss a conflict and then discuss something pleasant, and found that the second conversation — the recovery — discriminated who was still married four years later better than the fight did. It is one modest sample and its accuracy figures have been fairly criticised for not being cross-validated, so I would not build a marriage on it. But the shape of the finding matches what I see, and it matches the psychotherapy data, which is better.

With a high-functioning partner, repair fails in a specific way. They apologise like a professional: precisely, once, with a corrective action attached, and then consider the matter closed. You wanted to be understood and got a remediation plan.

The move, and it takes about ninety seconds. Name the specific thing you did. Name the effect it had on the other person, in their words, not yours. Say nothing about your intent — intent is the escape hatch every competent person reaches for. Then ask what would help. Four steps, no defence anywhere in them. Someone who runs a company can learn this in a fortnight once they understand it is a procedure rather than a humiliation.

Three things that are not the problem

The money. A meta-analysis published this year by Stephanie Dashineau and colleagues pooled 47 datasets and 25,171 people and found no significant association between income and marital satisfaction. A 2023 study of 927 adults found income did not even moderate how the usual relationship processes worked. The income is not holding it together and it is not what is breaking it.

The stress coming home. A preregistered meta-analysis of 36 dyadic studies and 1,504 effect sizes found minimal support for work stress crossing over into a partner's family well-being; partners' well-being shared at most 13.7% of variance. Emotional contagion from the job is a smaller force than the folklore suggests.

Which leaves absence. If it is not the money and not the stress travelling home, what is left is the hours themselves and what does not happen during them. Statistics Canada's Survey on Working Conditions, released in January 2026, found 34.6% of managers did unpaid work in their free time several times a month against 17.0% of workers overall, and 50.2% frequently worked to tight deadlines. In an April 2023 supplement, 37.3% of core-aged Canadians in management reported high or very high work stress, against 21.2% nationally.

Why this is worth more effort than it feels like

Sarah Stanton and colleagues followed 1,208 adults across 20 years of the MIDUS study. Change in how responsive people perceived their partner to be predicted how sharply they reacted to daily stressors years later — and that reactivity predicted mortality risk a decade after that. Feeling responded-to is not a soft outcome.

And the traffic runs both ways. A 2025 study in JAMA Network Open of 661 married or partnered physicians found that high partner career support was associated with satisfaction with work-life integration (OR 1.50) and with lower burnout (OR 0.73). The support you give is not charity extended to someone who has everything. It is actually doing something measurable to the person you married.

What I would actually do first

Pick the practical thing, not the emotional one, and make it big enough to be undeniable. Then respond properly the next time they come home with good news — put the phone down, ask two questions, stay in it longer than feels necessary. Then learn the four-step repair and use it the next time you are the one who got it wrong, so they see it done before they are asked to do it.

None of that requires them to become a different person. It requires you to stop reading silence as satisfaction, and it requires them to learn that being seen falling short is survivable. That second part is most of the work, and it is generally not something a couple gets to on their own — which is what couples therapy is for, and what the approach I work in is unusually direct about. If you both also run a business together, the same dynamic has an extra layer that I have written about separately.

The person you are worried about is not withholding. On the numbers, they are the least satisfied person in the house, and the last one who will say so.