What Is Gottman's 'Positive vs. Negative Sentiment Override'?

John Gottman, an American psychologist who spent several decades observing real couples in his University of Washington lab, describes a mechanism that explains why the same sentence, said in the same tone, can land in completely opposite ways depending on which couple hears it.

He calls it “positive sentiment override” and its opposite, “negative sentiment override”: two different default ways of interpreting what the other person does or says, which depend less on the specific thing happening in the moment and more on the general undercurrent of the relationship built up over time.

The concept isn’t originally Gottman’s: it was proposed by psychologist Robert Weiss, of the University of Oregon, in a 1980 chapter on behavioral marital therapy that is the reference usually cited for the concept, and Gottman acknowledges and develops it in The Seven Principles for Making Marriage Work.

The napkin example

Gottman illustrates the concept with a domestic scene with no apparent drama. A couple, Olivia and Nathaniel, are getting dinner ready for guests.

Nathaniel asks where the napkins are; Olivia answers, in a somewhat curt tone, that they’re in the cupboard.

When the underlying friendship is solid

If the underlying friendship between them is solid, the most likely outcome, according to Gottman, is that Nathaniel lets the tone slide and keeps only the useful information: where the napkins are.

He may put Olivia’s curtness down to some passing cause that has nothing to do with him, like not being able to get the cork out of a wine bottle at that very moment.

When the relationship is going through a rough patch

But if the relationship is in a worse place, the same curt answer can be taken as a direct attack and trigger an equally tense reply, something like “then get them yourself,” which in turn feeds a new argument with no real connection to the napkins.

Why the same event gets read in opposite ways

The mechanism works as a background filter, not as a case-by-case judgment.

When “positive sentiment override” prevails, each partner’s favorable thoughts about the other and about the relationship as a whole are so constant that they tend to outweigh passing friction: it takes a noticeably more serious conflict to knock the couple off balance, and the benefit of the doubt is granted almost automatically.

When “negative sentiment override” prevails instead, the opposite happens: any word said in a neutral tone can be taken personally.

Gottman gives another everyday example of this second case: the wife mentions that the microwave shouldn’t be run with nothing inside, a practical remark with no reproach intended, and the husband takes it as an attack on his competence, defensively replying that he already knows how to use it, that’s what he read the manual for.

And so an argument begins that, once again, has no real connection to what set it off.

What the 2002 study found

The idea has a direct empirical test. Hawkins, Carrère and Gottman (2002) recorded conversations of 96 newlywed couples and compared two readings of the same thing: that of trained outside coders and that of each partner about the other’s emotions. When a spouse’s rating departed systematically from the observers’, in a kinder or harsher direction, they interpreted it as sentiment override.

The results were more nuanced than the popular version:

• In wives, the quality of the bond predicted how they rated their husbands’ positive affect and low-intensity negativity, and positive override showed up: with a strong bond, they read small gestures of irritation as less negative.
• For neither husbands nor wives did the bond influence how they rated intense negativity. Clear contempt was perceived as such, whatever the underlying climate.

The authors conclude that the bond acts as a perceptual filter, at least in how the wives in that sample rated their husbands. For us it’s a useful nuance: the benefit of the doubt operates on ambiguity, not on obvious hostility.

A kinship with attributions

The idea fits with a parallel line of research, on attributions in couples. Thomas Bradbury and Frank Fincham (1990) reviewed dozens of studies and found that less satisfied people tend to explain their partner’s negative behavior as stable, global and intentional (“that’s just how they are”), and positive behavior as a one-off. The more satisfied do the opposite. It’s the same filter seen from another theory.

The body-weight analogy

To explain why this background state is so hard to reverse once established, Gottman draws a comparison with the popular “set point” theory of body weight: the idea that the body tends to maintain a particular weight through homeostasis, so that only by changing the baseline metabolism (for example, with regular, sustained exercise) can it be changed for good, beyond one-off diets.

Gottman applies the same logic to a relationship: when the overall “set point” is fixed at a high level of positivity, it takes much more accumulated negativity to damage the relationship than if the starting point were already low.

And if the relationship reaches a predominantly negative state, reversing it takes considerably more effort than avoiding getting there in the first place.

From one state to the other, not overnight

Gottman points out that most relationships start from such a favorable place that neither partner finds it easy to imagine it could ever go off the rails.

But that initial state doesn’t always sustain itself: over time, irritation, anger and unresolved resentment can wear down the underlying friendship until it becomes something increasingly abstract, present in words but less and less in real day-to-day life together.

It’s in that gradual process, not in a single trigger, that Gottman places the shift from positive to negative sentiment override.

He suggests that, once that point is reached, recovering the original bond can feel as hard as rowing backward in the middle of white-water rapids, an image he uses to stress why preventing this decline matters more than trying to reverse it later.

A scene to see it

On an ordinary Tuesday, Laura texts Sergio: “Did you see the electricity bill came?” If the undercurrent is positive, he reads it as information and replies “yes, I’ll pay it tonight.” If the undercurrent has been deteriorating for months, the same message reads as “you forgot again” and the reply comes out defensive. The message is identical; what changes is the filter it’s received through, and that’s what the 2002 study tried to measure.

This mechanism is closely related to the bids for connection Gottman describes in his research: how often a couple turns toward those small everyday gestures, instead of ignoring them or responding with hostility, is precisely one of the factors he associates with keeping the set point positive over time.

You can check how well you recognize them with our bids for connection quiz.

Sources

• Bradbury, T. N. and Fincham, F. D. (1990). Attributions in marriage: Review and critique. Psychological Bulletin, 107(1), 3-33. https://doi.org/10.1037/0033-2909.107.1.3
• Gottman, J. M. and Silver, N. (1999). The Seven Principles for Making Marriage Work. Crown.
• Hawkins, M., Carrère, S. and Gottman, J. M. (2002). Marital sentiment override: Does it influence couples’ perceptions? Journal of Marriage and Family, 64(1), 193-201. https://doi.org/10.1111/j.1741-3737.2002.00193.x
• Weiss, R. L. (1980). Strategic behavioral marital therapy: Toward a model for assessment and intervention. In J. Vincent (Ed.), Advances in Family Intervention, Assessment and Theory (Vol. 1). JAI Press.