A friend says 'calm down' with good intentions, yet the phrase makes the other person feel more alone. Why can an attempt to help make a conversation worse? We need to understand what each person needs and how the support is received.
The conversation continues after the mood changes
Imagine telling your partner that you are frightened about losing your job. They reply with a practical plan before you have finished explaining what happened. The plan might be useful, yet you could feel unheard. Or the same intervention might be exactly what you wanted. This is an editorial illustration, not a reported experiment. It exposes the question at the heart of Beyzanur Arican-Dinc and Shelly Gable's review: when one person tries to change another's feelings, what should count as a successful outcome?
It is tempting to measure only whether the frightened person feels calmer minutes later. The authors argue that this leaves out the person providing help and the relationship between them. A response may alter the immediate emotion and at the same time communicate care, impatience, control or understanding. Those relational meanings influence the next conversation. The review therefore joins two research traditions that have often looked at the same exchange from different sides: affective science asks which regulation strategy changes emotion; relationship science asks how support, shared stress and a continuing bond shape the exchange.
Regulating emotion is more than calming down
Emotion regulation means trying to change an emotion's intensity, duration or expression in pursuit of a goal. A person can reinterpret a difficult event, conceal a reaction or ask another person for help. Earlier affective-science models often contrasted strategies applied before an emotion develops, such as reinterpreting a situation, with strategies applied after it appears, such as suppressing its expression. Reappraisal was once commonly treated as better than suppression. The review explains why that ranking is too simple: effectiveness varies across people and situations, and flexible strategy choice can matter more than allegiance to one supposedly superior technique.
The goal also matters. Sometimes we want less fear or sadness because feeling better is itself the aim: a hedonic goal. At other times we tolerate or even increase discomfort because it serves another purpose: an instrumental goal. Worry may prompt preparation for a deadline; anger may help someone confront an unfair situation. This distinction does not make every unpleasant interaction helpful. It means that the emotional direction of an intervention—toward pleasant or unpleasant feelings—cannot by itself reveal the helper's purpose or the eventual outcome.
When another person enters the process
Interpersonal emotion regulation includes two processes that are easy to confuse. In intrinsic regulation, I seek someone else's company to manage my own emotion: I ask a friend to accompany me to an appointment I dread. In extrinsic regulation, someone takes the initiative to alter my emotion. The difference is who starts with the regulatory goal, not whether the people care about one another. It becomes especially important when the recipient has not asked to feel differently. An attempt to cheer someone up may clash with their wish to remain sad while grieving.
The review presents extrinsic regulation as a sequence: notice the other's emotion accurately, decide what change to seek, choose a strategy and carry it out. A mistake at the first step changes the rest. If a partner mistakes anxiety for anger, their well-meant advice may answer the wrong need. Studies discussed by the authors suggest that responsive validation after a negative experience is often preferred, while encouraging suppression produces mixed results. They also describe research in which help with reframing from a friend produced stronger benefits than reframing alone. These are patterns from different studies, not a universal recipe. The match between a strategy, the emotion, the situation and the recipient matters.
Support and coping are related, but not interchangeable
Relationship science has used a different vocabulary. Social support can come from a partner, friend or family member; it can mean emotional reassurance, information or practical assistance. It need not begin with a crisis. Sharing good news and receiving an engaged response is also a form of relational support. Research distinguishes support that is actually received from the belief that help would be available if needed. Those measures do not always have the same relation to well-being, so counting helpful acts alone can miss how safe or supported a person feels.
Dyadic coping is narrower: two partners face a stressor that affects their shared life. A job loss, illness or financial problem may start with one person but alter the resources and emotions of both. One partner can offer support; they can take over a task to reduce the other's burden; or they can work on the problem together. The review contrasts these positive forms with distant, ambivalent or superficial coping. Meta-analyses it discusses link more positive dyadic coping with better relationship quality and adjustment, but the direction of influence is not always one-way: a stronger relationship may also make joint coping easier. This is why 'my problem' can become 'our problem' without making both partners' needs identical.
The same help can feel different in different relationships
A strategy is not delivered into an empty room. The recipient interprets it through the history of the relationship, their expectations of the helper and their desire for closeness or independence. The review discusses work on attachment and perceived responsiveness: a gesture that one person experiences as reassuring may feel intrusive or doubtful to another. Even holding a spouse's hand has been linked to different responses depending on relationship quality. This is not evidence that touch is always beneficial; it shows why a technique cannot be evaluated apart from the bond in which it is used.
The quantity of support is no substitute for responsiveness. Overprotection may reduce autonomy when assistance exceeds what the other person wants or needs. Sometimes help also carries a cost because receiving it signals inability or debt. The authors connect these findings with a more general question: does the partner feel understood and cared for? A successful change in mood can coexist with damage to that feeling. Conversely, a hard conversation may leave immediate discomfort while building trust. These possibilities call for measuring emotional and relational outcomes separately rather than assuming one stands in for the other.
Unpleasant emotions can be used for different ends
Studies of social support and dyadic coping have concentrated mainly on making people feel better. Interpersonal-regulation research also examines deliberate affect worsening. Relationship research contains related work on inducing guilt or jealousy, but has often studied those emotions under separate labels. Guilt might prompt an apology after a transgression; someone might also provoke it to gain leverage. Jealousy may be elicited to test commitment, but that motive does not establish a good outcome. The review asks researchers to distinguish a deliberate attempt from an unintended effect and to identify whose goal it serves.
Goals can align or conflict. A person may hope to increase a partner's concern about a deadline for the partner's long-term benefit; the partner may see it as pressure. An intervention can be effective in the narrow sense that it produces the intended emotion and still be harmful to autonomy or trust. Equally, easing a partner's guilt after a serious breach may feel kind now while preventing a needed repair. These are explanatory examples, not general empirical conclusions about specific couples. They show why the authors insist on asking who wanted what, over which period and with which consequences for both people.
A framework for two people and a continuing bond
Arican-Dinc and Gable propose dyadic emotion regulation as a framework rather than a newly proven single mechanism. It is bidirectional: the person trying to help can also change emotionally. It recurs over the life of a relationship: a response today affects expectations tomorrow. It is dynamic: people adjust when they see how a partner reacts. And it is goal-driven. The goal may concern oneself, the partner or the relationship itself. These levels can pull in different directions during the same exchange.
Their framework separates the desired direction of emotional change from the strategy chosen to achieve it. One may increase positive emotion by celebrating good news, decrease negative emotion through shared coping, increase negative emotion by drawing attention to a risk, or dampen positive emotion for a perceived instrumental reason. Strategies may work through thought, such as reframing or advice, or through the relationship, such as validation, affection and touch. This map does not tell us that all four directions are equally wise or well supported. It makes the possible pathways visible so researchers can ask which one occurred and assess both the immediate emotion and the relationship's longer course.
What the evidence still needs to explain
The authors identify an imbalance in the literature. We know more about reducing distress than about helping someone enjoy a positive experience. Yet a partner's good news can elicit pride, indifference or envy, and responding well may require a different skill from comforting sadness. The review asks which strategies increase another person's joy or gratitude, and whether people turn to different partners for emotional uplift. It also asks when abundant help becomes overprotection and how a recipient's autonomy should enter the evaluation of a supposedly successful intervention.
Another open question concerns change over time. Partners may develop complementary repertoires: one tends to distract, the other to reinterpret; later each may learn from the other. The review treats this as a direction for research, not an established trajectory for every couple. Much of the evidence it integrates comes from traditions with different definitions, outcomes and study designs. Its contribution is to reveal what each tradition has omitted and to propose a common language for testing the connections. To assess whether help worked, consider whose emotion changed, who achieved what they wanted and how the conversation affected the relationship afterward.
What to keep
- Emotion regulation can be interpersonal.
- Goals can be hedonic or instrumental.
- Relationship effects deserve separate measurement.
The nuance
What this story does not let us claim
What helps depends on both partners’ goals and on whether success means feeling better now or functioning better later.
Go deeper: concepts and optional reading practice
Reversible cards
Test the concept
Topic assessment
Check whether you can explain the idea, not merely recognize it.
Every answer is grounded in the explanation above. Some questions introduce new cases to test whether you can apply and analyze the ideas, not merely recall them.
Source & scope
Dyadic Emotion Regulation. Beyzanur Arican-Dinc & Shelly L. Gable. Annual Review of Psychology, vol. 77 (2026).
This editorial guide is an independent explanation. Consult the original publication for the studies, references and full context.
Open the original review ↗