What Makes Relationships Last — The Real Psychology
Last reviewed by the Men Women Psychology editorial team.
The evidence
What the research actually shows
John Gottman and Robert Levenson (1992) followed married couples and found they could predict relationship dissolution with striking accuracy from how partners handled conflict. The strongest danger signs were what Gottman later called the Four Horsemen: criticism, contempt, defensiveness, and stonewalling. Of these, contempt — eye-rolling, mockery, an attitude of superiority — was the single best predictor of breakup, because it communicates disgust rather than mere disagreement.
Gottman and Silver's The Seven Principles for Making Marriage Work (1999) distilled the positive side. Stable, happy couples maintained roughly a five-to-one ratio of positive to negative interactions during conflict, made and accepted 'repair attempts' — small gestures that de-escalate tension, like a joke or a softened tone — and consistently turned toward each other's 'bids for connection,' the small everyday invitations for attention and affection. The foundation underneath was friendship: knowing each other deeply and maintaining fondness and admiration.
Two further lines of research round out the picture. Caryl Rusbult's investment model (1980) showed that commitment depends not only on satisfaction but on how much a person has invested and how they appraise their alternatives — which is why people stay through rough patches when they have built a shared life. And Reis and Shaver's (1988) work on intimacy as an interpersonal process found that closeness grows through a cycle of disclosure and responsiveness: feeling understood, validated, and cared for by a partner who is genuinely attentive.
The mechanism
Why this happens
Contempt is corrosive because it attacks a partner's worth rather than a specific behavior, and it accumulates. Gottman's physiological recordings showed that escalating negativity floods partners with stress, which shuts down problem-solving and makes warmth harder to access. A relationship that runs a chronic deficit of positive moments has little buffer when conflict hits, which is why the five-to-one ratio matters more than the absence of conflict.
Repair attempts and bids for connection work because they are constant tests of goodwill. Every day offers dozens of tiny moments — a comment, a question, a sigh — where a partner can turn toward, away, or against the other. Couples who reliably turn toward these bids build a reservoir of trust, so that when a real repair attempt is needed mid-argument, it is more likely to land.
Commitment and responsiveness give the structure its durability. Rusbult's model explains why investment and shared history bind people together beyond moment-to-moment happiness, and Reis and Shaver's intimacy process explains how partners keep feeling close enough to want to stay: through repeated experiences of being understood and cared for. Friendship sits underneath all of it, supplying the affection that makes repair and responsiveness possible in the first place.
In practice
What this looks like in real life
Two couples can argue equally often, yet one thrives and one dissolves. The difference is usually tone and repair: one pair fights, then one of them cracks a gentle joke or says 'I'm sorry, let me try again,' and the temperature drops. The other lets contempt creep in — a sneer, a 'you always' — and each fight leaves a deeper mark.
A partner mentions something small from their day. Turning toward that bid — looking up, asking a follow-up question — seems trivial, but Gottman's research suggests these micro-moments, repeated thousands of times, are where lasting connection is actually built or quietly eroded.
A couple weathers a hard year not because they are perfectly happy in it, but because they have built a shared life and feel deeply understood by each other. Rusbult's investment model and Reis and Shaver's responsiveness work both help explain why commitment and feeling 'gotten' carry a relationship through stretches when satisfaction dips.
Myth vs. evidence
What most people get wrong about this
Many people assume lasting couples are the ones who rarely fight. The research says otherwise: conflict is normal and even healthy, and what distinguishes stable relationships is how conflict is handled — low contempt, working repair attempts, a positive surplus — not its absence. Avoiding all conflict can actually mask problems that fester.
Another common mistake is treating compatibility or initial chemistry as the deciding factor. While they matter early, longevity tracks far more with ongoing behaviors — responsiveness, turning toward bids, managing contempt, and sustained investment — that any committed couple can practice and improve, rather than a fixed match locked in at the start.
Why it matters
What this means for relationships
The practical implications are encouraging because they are largely about behavior, not fate. Watching for contempt and replacing it with specific, non-blaming complaints; protecting the five-to-one ratio with everyday warmth; learning to make and accept repair attempts; and reliably turning toward a partner's bids are all skills that can be strengthened over time by either partner.
Just as important is sustaining the friendship and investment underneath — staying curious about each other, building a shared life, and offering responsive attention when a partner discloses something that matters. None of this is gendered: the same ingredients predict lasting relationships across couples, and both partners contribute to whether they are present.
Where it varies
The nuance
These findings describe patterns across many couples, not guarantees for any one, and the ingredients are remarkably consistent across genders. Janet Hyde's gender similarities hypothesis (2005) is a useful reminder here: men and women respond to contempt, responsiveness, and repair in broadly similar ways, so what makes a relationship last is far more about the dynamic between two people than about their sexes.
Individual differences — attachment style, temperament, life stress, and culture — shape how easily these behaviors come and how they are expressed. The averages hide wide variation, and a relationship's durability ultimately rests on the specific, repeated choices two people make toward each other, not on a formula applied from the outside.
Questions people ask about this
What is the single biggest predictor that a relationship will end?
In Gottman and Levenson's research, contempt stood out as the strongest warning sign — expressions of disgust or superiority like mockery and eye-rolling. It is more damaging than ordinary anger because it attacks a partner's worth. This is a robust average pattern, not a certainty for any individual couple.
Do happy couples just fight less?
Not necessarily. Research suggests lasting couples are not the ones who avoid conflict but the ones who handle it well — low contempt, successful repair attempts, and a strong surplus of positive interactions. Conflict itself is normal; how it is conducted is what tends to matter for the long run.
What is the five-to-one ratio?
Gottman observed that stable couples tended to maintain about five positive interactions for every negative one during conflict. The idea is that a steady reservoir of warmth buffers the inevitable friction. It is a useful guideline drawn from averages rather than an exact threshold that fits every relationship.
What are 'bids for connection'?
They are the small everyday invitations for attention or affection — a comment, a question, a touch. Gottman's work suggests that reliably turning toward these bids, rather than away, builds the trust that holds relationships together. Much of lasting connection is made in these tiny, easily missed moments.
Why do people stay together through unhappy periods?
Rusbult's investment model suggests commitment rests not only on current satisfaction but on accumulated investment and how a person views their alternatives. A shared life, history, and responsiveness can carry a couple through rough stretches. This helps explain endurance without implying anyone should stay in a harmful situation.
Can a struggling relationship be repaired?
Often, yes, because most of the predictors are behaviors rather than fixed traits. Reducing contempt, rebuilding positive interactions, learning repair, and increasing responsiveness can all be practiced. Outcomes vary by couple and circumstance, and serious issues may call for a skilled couples therapist, but change is frequently possible.
Research sources
These references point to the published research and established frameworks behind this page. They are provided for further reading; see our research methodology for how sources are selected.
- Gottman, J. M., & Levenson, R. W. (1992). Marital processes predictive of later dissolution. Journal of Personality and Social Psychology, 63(2), 221–233.
- Gottman, J. M., & Silver, N. (1999). The Seven Principles for Making Marriage Work. Crown.
- Rusbult, C. E. (1980). Commitment and satisfaction in romantic associations: A test of the investment model. Journal of Experimental Social Psychology, 16(2), 172–186.
- Reis, H. T., & Shaver, P. (1988). Intimacy as an interpersonal process. In S. Duck (Ed.), Handbook of Personal Relationships.
- Hyde, J. S. (2005). The gender similarities hypothesis. American Psychologist, 60(6), 581–592.