Body Language & Nonverbal Communication

The conversation under the words

Two people can say the exact same sentence and mean opposite things. "I'm fine." "Sounds great." "No, really, go ahead." The words look identical on the page, yet a listener in the room almost never gets them wrong, because the words were only ever half of it. Underneath every spoken line runs a second channel: the face, the hands, the posture, the distance held between two bodies. That channel is what nonverbal communication describes, and it carries a surprising amount of the meaning we trade in workplace interactions, in interviews, in presentations, and in the relationships that matter most.

This is also where the internet gets one thing badly wrong, so it is worth clearing up first. You have seen the line that "93 percent of communication is nonverbal" and "only 7 percent is the words." It is a misreading of a single set of experiments, and the real finding is narrower and more useful than the slogan. It begins with a psychologist asking a specific question: what happens when a person's words and their face flatly disagree?

Albert Mehrabian published two studies in 1967 that produced the famous numbers. He had people judge how much a speaker liked them from a single spoken word delivered with a tone and an expression that either matched the word or contradicted it. When the channels conflicted, listeners leaned on the face, weighting it at 55 percent, and on the tone, at 38 percent, and mostly discounted the word itself, at 7 percent. That is the whole scope of the result: liking and disliking, judged from one word, in a lab, when the signals fight. It says nothing about how meaning gets carried in an ordinary conversation where words and face agree.

Mehrabian knew his numbers were being abused and said so for decades. On his own site he is plain about it: unless a person is talking about their feelings or attitudes, the equations do not apply. Walk a colleague through a quarterly report and the words carry the report; no facial expression delivers the revenue figure for you. So the honest version is not "body language is 93 percent of everything." It is narrower and more practical: when what you say and what you show disagree, people believe what you show. A manager who says "I'm open to feedback" through a clenched jaw has already given the real answer, and that gap between the spoken and the shown is the part you can actually train.

When the body goes through a screen

Strip the body out and you can watch how much it was doing. A plain text message has no face and no tone, which is why "Fine." and "Sounds great." can land as warm or icy depending entirely on the reader's mood that morning. The channel Mehrabian was studying, the one that resolves a clash between words and feeling, is the first thing a screen deletes. So we improvise replacements.

This is the real reason emoji, punctuation, and response timing carry so much weight in writing. A period at the end of a one-word reply now reads, to many people, as a small cooling-off, because it stepped in to do a job a flat tone used to do out loud. An exclamation point softens a request. A thumbs-up can close a thread warmly or end it curtly, depending on context the sender cannot see. The same flat "no" that sounds reasonable spoken reads as cold on a screen, because the reader has lost the warmth your voice would have carried.

None of this means written communication is worse, only that it asks for different care. When the nonverbal layer is missing, the words do more of the work and the small signals you control matter more. Our guide to remote communication circles this problem repeatedly, because a distributed team runs almost entirely on channels that have had the face and tone stripped out, and the misunderstandings that follow are rarely about the facts. They are about the feeling the facts arrived wrapped in.

A field guide to the common signals

Once you stop chasing the 93 percent myth, the practical question gets simpler: which signals actually carry information, and what do they tend to mean? The honest answer comes with a warning. No single gesture is a word. Body language is closer to weather than to vocabulary; you read patterns, not isolated readings, and the same posture can mean three different things depending on the room temperature and the chair.

Take the most over-interpreted cue of all. Crossed arms get treated as a verdict, as if they prove someone has closed off and rejected you. Usually they prove the person is cold, or the chair has no armrests, or that is simply how they stand. The cue only carries meaning inside a cluster. If crossed arms come with a turned-away body, a wandering gaze, and one-word answers, then yes, you are losing the person. On their own they are noise. This is the single most useful habit in the subject: read the cluster, never the lone gesture.

The table below sorts a handful of common signals by what they tend to communicate. Treat every row as a starting hypothesis to check against the rest of the person, not a translation you can trust on its own.

One distinction in that table is worth dwelling on, because it is the closest thing to a reliable tell. A polite smile uses the mouth; a felt one also pulls in the muscles around the eyes, the ones most people cannot consciously fire on command. That difference is not folk wisdom. It comes out of the precise study of the face that Paul Ekman spent a career building, which raises the next question: how much of the face can really be read?

How to actually change these habits

People who want to improve their own body language usually start from the wrong end. They try to memorize a long list of gestures and then deploy them, which produces exactly the stiff, performed quality they were hoping to avoid. The literature points the other way: the gains come from removing the unhelpful habits you already have, not from bolting on new ones. A few changes, practiced until they stop feeling like a performance, do more than any catalogue of poses.

The first move is to find out what you do. Almost no one has watched themselves talk, which is why the fastest improvement is to record a short conversation or practice run and play it back. It is an unpleasant ten minutes and the most efficient ten available, because you will see in moments what you have been doing for years: the death grip on the lectern, the eyes locked on the floor, the face that forgot it was allowed to move.

From there the useful changes are few and unglamorous. Get your eyes off your notes and onto the people you are talking to, since eye contact does more for how you come across than any other signal. Let your hands out of their hiding place so gestures happen naturally instead of leaking out as fidgeting. Slow down, because nerves speed everyone up and a rushed delivery reads as anxious whatever the words say. Open the posture a little instead of folding inward. None of these is a trick; each one removes a habit that was working against you. The point is to look like yourself on a good day, not like a diagram of a confident person.

The harder truth is that this only changes through repetition, and repetition is uncomfortable before it is automatic. You cannot read your way to better body language any more than you can read your way to a tennis serve. Pick one habit, work it in low-stakes settings until it runs without thinking, then add another. That deliberate practice is the through-line of our guide to persuasive communication, and it separates the people who change from the people who just learn the vocabulary.

The face that gives you away

The most counterintuitive finding in this field is that a handful of expressions mean roughly the same thing almost everywhere, and Paul Ekman is the reason we know it. In the late 1960s he and Wallace Friesen worked with the Fore, a people in New Guinea who had had almost no contact with Western film or photographs. If facial expressions were learned by copying the faces around us, the Fore should not have recognized an outsider's surprise or disgust. They did. A short list of emotions, happiness, sadness, anger, fear, surprise, and disgust among them, showed up on the face in ways people identified across cultures that shared almost nothing else.

Ekman then built a system precise enough to argue with. Working again with Friesen, he developed the Facial Action Coding System, which breaks every visible movement of the face into numbered units: the inner brow raise, the lip-corner pull, the cheek raiser that separates the felt smile from the courteous one. FACS lets a researcher say that a specific muscle fired for a specific fraction of a second, and it still underpins most credible work on the face today, including the software that now claims to read expressions automatically.

Out of that work came the idea most people have heard of: the micro-expression. When someone suppresses a feeling, the real emotion can leak out for as little as a twenty-fifth of a second before the social mask snaps back. In a tense conflict resolution conversation or a hard interview, that flash is genuine information. But this is where the subject gets oversold. Catching a micro-expression tells you a feeling happened; it does not tell you why, and it does not let you detect lies, whatever a crime drama implied. Ekman has been blunt about that limit. Treat a fleeting expression as a question to ask, never a verdict to hand down.

The space between people

How close you stand to another person is communication, and communication you almost never decide on consciously. Edward T. Hall gave this its name, proxemics, the study of how people use the space around their bodies. Hall observed that we each carry a set of invisible bubbles. The intimate zone runs from roughly six to eighteen inches and belongs to family, partners, and the crowded elevator everyone silently tolerates. Just past it sits the personal zone of friends and easy talk. Beyond that lie the social and public zones, for colleagues, strangers, and anyone addressing a room.

What mattered most to Hall was that these distances are not fixed by nature. They are learned, and learned so early that we mistake them for instinct, which means they collide whenever cultures meet. The collision is easy to watch and slightly comic once you know what you are seeing. A person from a closer-talking culture steps in; a person from a more distant one steps back; the first reads the retreat as coldness and closes the gap again, and the two drift across the room in a slow waltz neither one chose. Gestures betray us the same way: the thumbs-up that means "well done" across much of the West is an insult in parts of the Middle East.

Eye contact runs on a similar dial. People tend to hold more of it the farther apart they sit, trading gaze for distance to keep a conversation comfortable, but the meaning of a steady gaze flips depending on where you are. In much of the West it reads as honest and engaged; toward a senior in parts of East Asia and Africa, the same gaze can read as disrespect, and dropping it is the respectful move. This is why memorizing a country-by-country chart is a losing game. The deeper skill, which our guide to active listening leans on too, is to watch the person in front of you and follow their lead. Hall's larger point, that so much of nonverbal behavior is learned rather than universal, runs through all serious work on communicating across cultures.

Reading the room when the room is a webcam

The video call took everything we just covered and made it harder. The signals are still there, but they arrive cropped, flattened, and slightly out of sync, and most people have no idea how they come across on the other end of the lens. The fixes, at least, are concrete and the gains are fast.

Start with the camera, because almost everyone is filming up their own nose. Raise the lens to eye level and the unflattering angle disappears, along with the impression of looking down at whoever you are talking to. Sit back far enough that your hands come into frame when you gesture; a head floating in close-up loses the entire vocabulary of the hands. Light yourself from the front rather than letting a bright window behind you turn you into a silhouette. And talk to the lens, not to the gallery of faces on your screen, because looking at the faces means looking slightly down and away from everyone else's point of view, which reads as evasive even when you are fully engaged.

The single most effective fix, though, is the one that also works off camera: record a real call and watch it back. It is uncomfortable, and it is the fastest correction available, because ten seconds of footage will show you a habit you have repeated for years without knowing. The cues that build trust in a room, steady eye contact, an open posture, a face that moves, all still work through a webcam; they just need the small mechanical adjustments that put them back in view. For teams that live on these calls, our leadership communication guide treats this as a core skill rather than a cosmetic one, because over months, how you show up on screen is most of how you show up at all.

Frequently Asked Questions

CommunicationAbility is reader-supported and independent; we accept no sponsorship or paid placement, and this guide is general professional-development information, not a substitute for professional, counseling, or mental-health advice. Terms apply.

Last reviewed: July 15, 2026