In Swedish schools, psychiatric wards, and juvenile care facilities, staff are trained in a method called lågaffektivt bemötande, low-affect response, developed and popularised by the Swedish clinical psychologist Bo Hejlskov Elvén. The method’s starting instruction is not about what to say to someone who is escalating. It is about what to do with your own body and voice first, on the theory that the calmest person in the room, not the most persuasive one, usually determines where a tense exchange goes next.
This is a working framework built for specific, often high-stakes care settings, not a peer-reviewed clinical trial with a single outcome measure, and it is worth reading it that way rather than as a settled scientific result.
The idea behind it
Hejlskov Elvén’s method rests on a straightforward claim from affect and stress theory: emotional arousal is contagious. People, and especially people who have difficulty distinguishing their own feelings from someone else’s, tend to pick up and mirror the intensity of the emotion directed at them. Raise your voice at someone already agitated, and there is a good chance you raise theirs further. The practical implication Hejlskov Elvén draws from this is that the person trying to de-escalate a situation needs to actively monitor and lower their own arousal, tone of voice, pace of speech, and body language, rather than focus first on what the other person is doing wrong.
The broader mechanism, that emotional and physiological arousal can pass between people in a conversation, is not specific to Sweden or to this method. Emotional contagion has been studied as a general phenomenon in social psychology for decades, and more recent physiological research has tracked how one person’s stress response can measurably shift a conversation partner’s heart rate and other stress markers during real interactions, a pattern researchers describe as stress contagion.
Where it is actually used, and why
Hejlskov Elvén developed and applied the approach primarily in Swedish special education, psychiatric care, and services for people with autism and intellectual disabilities, contexts where the person on the other side of a conflict may have limited capacity to self-regulate and where physical restraint or confrontation was, for decades, a default response. The method’s ethical starting point is explicit: everyone has the right to say no, and staff carry the responsibility for finding a way to a genuine yes rather than forcing compliance. That framing grew out of Swedish disability and psychiatric care specifically, not a general theory of how any two people should argue.
Hejlskov Elvén states the underlying philosophy plainly on his own site: staff and parents need to take responsibility for a difficult interaction going wrong, because only the person who accepts responsibility is in a position to change their own approach. He also borrows a principle from the American child psychologist Ross W. Greene, that people who are able to behave well generally do, and that a person who is not behaving well is more likely to be lacking a specific skill or capacity in that moment than choosing to be difficult. Applied to staff and parents rather than to the person in distress, the practical upshot is a shift in where the responsibility for de-escalation sits: with the calmer party in the room, not the more distressed one.
The specific moves the method teaches
In practice, the training built around Hejlskov Elvén’s model is fairly concrete rather than purely conceptual. Staff are coached to lower their pitch and volume rather than match a raised voice, to slow their own speech, to increase physical distance rather than close it, and to avoid standing directly face to face with someone who is escalating, since a more angled stance reads as less confrontational. None of these individual moves is unique to Sweden or to this method specifically. What is distinctive is packaging them together as a coherent staff practice, backed by an explicit ethical stance about whose responsibility de-escalation is, and teaching it systematically across schools, psychiatric wards, and care homes rather than leaving it to individual instinct.
What carries over to an ordinary difficult conversation
The specific clinical framework was not built or tested for an argument with a partner, a heated moment in a meeting, or a disagreement with a neighbour. But the underlying mechanism it relies on, that a calmer voice and body can lower the emotional temperature of the person you are talking to, draws on the same general affect-contagion research that applies well beyond care settings. Lowering your own voice rather than matching a raised one, slowing your speech, softening your posture, and creating a bit of physical space are small, testable moves that do not require training in a Swedish clinical method to try.
What does not carry over automatically is the size of the effect, or a guarantee of speed. Hejlskov Elvén’s own model was built around sustained changes in how staff relate to a person over time, not a single instant fix inside one heated exchange, and there is no large body of controlled research validating “low-affect response” as a branded technique the way there is for the underlying stress-contagion mechanism it draws on. The honest version of the claim is that keeping your own arousal down is a reasonable, theory-backed first move, not a proven guarantee that any given conversation will cool down quickly.
What this does not settle
None of this means a calm voice fixes every disagreement, or that the other person’s behaviour is irrelevant. Some conflicts are about a real, substantive disagreement that a calmer tone does not resolve, only makes easier to have. And a method built for a psychologist working with a person in genuine crisis is not automatically the right frame for two adults negotiating who forgot to do the dishes.
What is left, once the specific clinical context is set aside, is a modest and well-supported starting point rather than a trick: arousal moves between people in a room before the content of an argument is even addressed, and the person willing to lower their own first has more influence over where the conversation goes than the one focused only on winning it.