The instinct behind the question is sound, and the reason is not the one the phrasing suggests. There is no look, posture, or silence that has been shown to disarm someone mid-conversation. The research on nonverbal dominance is contested enough that we would not build advice on it.
What has held up is something that happens earlier, requires no confrontation, and is genuinely wordless: knowing how the technique works before it is used on you.
What the meta-analysis found
The idea is old and has a name. Inoculation theory proposes that exposing someone to a weakened version of a persuasive attack, along with an explanation of why it works, builds resistance to the full-strength version later. The parallel with vaccination is deliberate.
The Artful Age
A weekly letter on aging well, family across generations, and the creative life after the kids leave home.
John Banas and Stephen Rains pooled the evidence in a meta-analysis in Communication Monographs in 2010, using 41 cases with 10,660 participants in the main comparison against untreated controls.
The average effect was d = .43. The authors describe that as small in magnitude, and we will not inflate it. Resistance to persuasion is a difficult thing to move at all, which is the context they place it in.
Two of their findings matter more than the headline number.
The first is that it did not matter whether the warning matched the argument. Refuting the same argument someone would later face and refuting a different one produced equivalent resistance. You do not need to anticipate the specific thing that will be said. Understanding the shape of the technique carries across.
That first point is the practically important one, and it is easy to skim past. Most advice about difficult people asks you to memorize warning signs, which only helps if the situation matches the list. This says the useful unit is the move, not the instance. Learn what a false choice is and you are covered against every false choice, including the ones nobody has written an article about.
The second finding is less convenient. Resistance held up at immediate and moderate delays, then decayed after about two weeks.
The version tested at scale
The idea was revived recently at a size that is unusual for psychology. Jon Roozenbeek and Sander van der Linden at Cambridge, with Stephan Lewandowsky at Bristol and Beth Goldberg at Google Jigsaw, published seven experiments in Science Advances in 2022, involving roughly 30,000 people in total.
Six were controlled experiments with 6,464 participants. The seventh ran on YouTube, where 5.4 million users saw short animations and 22,632 of them answered test questions afterward.
The animations were ninety seconds long and taught five specific techniques: scapegoating, deliberate incoherence, false dichotomies, hyper-emotive language, and personal attacks on the speaker rather than the argument. They used examples from Family Guy and Star Wars, which is worth noting because the material was not solemn and still worked.
People who watched were about 5 percent better at spotting those techniques, tested roughly eighteen hours later.
Five percent is a small number and we would rather say so than dress it up. What makes it interesting is the denominator. This ran as ordinary advertising, at around five cents per view, to millions of people who had not asked to be taught anything.
One of the controlled experiments was repeated a year on and the finding held, which sits somewhat awkwardly beside the two-week decay in the earlier meta-analysis. Both results are real and the tension between them is not resolved.
Why this is the wordless version
The thing doing the work here happens before the exchange and inside your own head. There is no confrontation in it, no naming of the other person, and no requirement to win anything.
It also changes what you are looking for. Recognizing a false dichotomy or a shift from the argument to the person is different from deciding what kind of person you are dealing with. The first is a description of a move. The second is a verdict about a human being, and the research does not support handing those out.
That distinction matters for the phrasing of the original question. These studies measured whether people could identify techniques in messages. None of them measured people, sorted anyone into categories, or tested a method for shutting an individual down.
In our reading, that is the useful correction. Someone who uses a false dichotomy on Tuesday is not thereby a manipulator, and treating a technique as a diagnosis tends to make the next conversation worse rather than better.
It is also the more forgiving frame to hold at a kitchen table. Most people reach for these moves occasionally, usually when they are tired, cornered, or care more about the outcome than they are admitting. Noticing the move and not the character leaves room for that.
What this does not cover
The gap between the studies and the situation in the question is real and should be stated.
Both bodies of research are mostly about persuasive messages: advertising, political content, misinformation online. Neither tested a long-running relationship with a specific person who knows exactly which of your sore spots to press. It is plausible the mechanism transfers. It has not been shown.
The samples are also largely Western and, in the YouTube case, self-selected by whoever chose to answer a question after a video. The 5 percent figure belongs to that group.
And the decay finding cuts against the way this kind of advice is usually offered. If resistance fades in a fortnight, then reading one article about manipulation techniques is not a permanent upgrade. It is closer to something that needs refreshing.
If what you are describing is not an argument technique but a pattern where someone controls what you do, who you see, or what you are allowed to believe about your own memory, that is a different matter than persuasion, and a doctor or a domestic abuse helpline is better placed than any article about conversational tactics.
What we keep coming back to is how ordinary the effective version turns out to be. Not a stare, not a silence held a beat too long, not a technique performed at somebody. Just having already read, in advance and at leisure, how the move works.