Actualités

Many people can look back at a separation and recall a stretch when the closeness seemed to drain away. Researchers who followed thousands of people over many years found that, in relationships that did end, satisfaction tended to fall faster in the final year or two. That is a pattern across large groups, though, and it cannot tell any one couple what lies ahead.

What the researchers did

The work was led by Janina Bühler at Johannes Gutenberg University Mainz, with Ulrich Orth at the University of Bern, and appeared in the Journal of Personality and Social Psychology in 2025. A university news release was reported by ScienceDaily in March 2025.

The team used four long-running national surveys, from Germany, Australia, the United Kingdom and the Netherlands. These studies followed people for between 12 and 21 years and regularly asked them to rate how satisfied they were with their current relationship. In total, the analysis covered 11,295 individuals whose relationships ended during the study period.

Instead of lining people up by calendar year, the researchers lined them up by time left before separation. They also compared these people with a control group of similar size in each dataset, made up of people whose relationships did not end. This lets you see whether the shape of the decline is specific to relationships that break down.

What they found

In relationships that ended, satisfaction followed what the authors call a terminal decline with two phases. First came a long stretch, which could last several years, in which satisfaction slipped only a little. Then came a tipping point, after which satisfaction fell much more quickly. The fast-falling phase began roughly 7 to 28 months before the separation.

This fast drop was seen only in the relationships that ended. The comparison groups of continuing relationships did not show it.

The two partners did not necessarily reach the tipping point together. According to the news report, the person who went on to start the separation tended to become dissatisfied earlier, while the person on the receiving end typically reached their tipping point closer to the breakup. The lead author suggested that couples therapy might work better if it begins in the earlier, slower phase, because many couples seek help only after the sharp decline has already started. That is the author's suggestion, not something this study tested.

What this does and does not show

The study looked backwards from separations that had already happened. That makes it good at describing what the lead-up looked like, but it cannot work in reverse. A dip in satisfaction is not a countdown. Most people go through low patches, such as new parenting, illness, money worries or grief, and many couples come out of them. Group averages hide those stories.

The data also came from four wealthy Western countries where separation is legally open to couples. Patterns may differ in places with different laws, cultures or economic pressures, and for relationships in which leaving carries heavy costs. The news report did not set out a list of limits, so these points come from the design itself.

Satisfaction was measured with ratings from each person, not with outside observation. A rating tells you how someone feels, not why. The study does not show what caused the decline, and it does not show that any action would have prevented a separation.

If your own relationship feels like it is drifting, this research is not a reason to panic or to count months. It may simply be a prompt to notice what has changed and to talk about it. This general information does not apply to relationships involving fear, control or violence. In those situations, safety comes first. In Australia, Lifeline is on 13 11 14, and in an emergency call 000.

Questions to ask

• What has changed in how we spend time together, talk or show appreciation over the past year, and is that change something we both notice?
• If one of us feels the closeness slipping, how could we raise it early and gently, rather than waiting until it feels too late?
• Would it help to talk with a suitably qualified relationship counsellor or psychologist, and what would each of us want to get out of that conversation?

Source: Terminal decline of satisfaction in romantic relationships: Evidence from four longitudinal studies DOI: 10.1037/pspp0000551

Illustration of emotional distance within a relationship. AI-generated editorial illustration. This article summarises published research in our own words and is general information, not personal health advice.