Why Your Brain Decides What a Handshake Means Before You Even Finish Watching It
- 9 hours ago
- 8 min read
Written by Ben Cardall, C.E.O - Investigator
Ben Cardall is a human behaviour and memory expert, bestselling author, and creator of immersive thinking systems that sharpen observation, reasoning, and emotional intelligence. He trains professionals to think faster, remember more, and make better decisions under pressure.
When Trump and Xi shook hands in Beijing, the internet had already decided who won. The problem is, the brain always decides first, and it is almost always wrong. Here is what actually happened, and why it matters for how you read every interaction you will ever witness.
What is body language analysis, and why is most of it wrong?
On 14 May 2026, Donald Trump and Xi Jinping shook hands outside the Great Hall of the People in Beijing. The footage lasted around fifteen seconds. By the time most people had finished watching it, many major outlets and commentators had already published a verdict.
“Xi flipped the power dynamics.” “Trump’s body language signals submission.” “This handshake tells us everything about who holds the power in this relationship.”
Same footage. Fifteen seconds. Dozens of completely different verdicts, all delivered with total confidence.
This is not an accident. It is a feature of how the human brain processes information, and it is the central problem with how body language gets discussed in the media, on YouTube, and in popular culture at large.
Body language analysis is a real scientific field. The study of nonverbal behaviour, proxemics, kinesics, and parakinesics has a genuine and substantial research base. What gets broadcast under that name, however, is something different: a confident interpretation of single behaviours, stripped of context, applied with cultural blinkers on, and packaged to feel like revealed truth.
The researcher Vincent Denault has documented this gap systematically. His peer reviewed work identifies what he calls the pseudoscience of body language in media and security contexts, the systematic mismatch between what the empirical literature actually shows and what gets presented to the public as fact. His conclusion is not that the field is worthless. It is that what is being broadcast under its name routinely is not that field.
You can read Denault’s work in Atenea journal, 2020. It is worth your time if you want to understand the scale of the problem.
Why the brain settles on a story before the evidence is in
To understand why body language content produces so many confident wrong answers, you need to understand what the brain is actually doing when it watches a video of two people interacting.
Daniel Kahneman’s research on cognitive systems, documented in his book Thinking, Fast and Slow, describes two modes of thinking. System 1 is fast, automatic, and pattern-driven. It generates conclusions rapidly and with the feeling of certainty. System 2 is slow, deliberate, and effortful. It checks the conclusions System 1 generates, but only when we make it do so.
Most people watching the Trump and Xi handshake were operating entirely in System 1. They had a prior belief about the power dynamic between the two men. They saw a gesture. System 1 immediately connected the gesture to the prior belief, generated a story, and delivered it with the subjective experience of understanding.
That subjective experience of understanding is not the same as accuracy. It never has been.
The research on human judgment under uncertainty makes this point sharply. Studies consistently show that humans, including trained professionals, perform at roughly chance level when asked to make deception or emotional state judgments from behavioural observation alone. The feeling of confidence in a judgment is not correlated with the accuracy of that judgment in the way most people assume. Smart people do not get away from these flaws, as they have brains too.
Bond and DePaulo’s 2006 meta analysis in the Personality and Social Psychology Review examined the accuracy of deception judgments across hundreds of studies. The results are not flattering for confident body language interpretation.
Lisa Feldman Barrett’s work adds a further layer. Her constructed emotion theory, developed across decades of neuroscience research, demonstrates that emotions are not fixed internal states that leak out through the face and body for trained observers to read. They are constructed by the brain in real time, shaped by prediction, context, and prior experience.
Which means that even if you could perfectly read every physical signal a person produces, and you cannot, you would still be reading the output of a constructed experience, not a window into a fixed underlying truth.
The three mistakes every body language analysis makes, and the Trump and Xi example makes all of them
The coverage of the Beijing handshake is a clean case study in the three errors that appear in almost every piece of popular body language analysis. Naming them is useful because once you can see them, you see them everywhere.
The first error is single behaviour, single meaning. The “power dynamics” narrative selected one feature of the interaction, the double-handed clasp Trump placed over Xi’s hand, and assigned it a fixed meaning. In the scientific literature on nonverbal behaviour, this is called one to one correspondence, which is the assumption that behaviour X reliably maps to meaning Y. The problem is that this does not hold empirically. The same gesture, in different cultural contexts and with different individuals, carries different or even opposite meanings.
The double-handed clasp, for example, has been used by Bill Clinton, by multiple European heads of state, and by diplomats across generations as a signal of personal warmth and engaged interest. It is not inherently submissive. The interpretation of it as submission required selecting one possible meaning and ignoring the rest.
The second error is the absence of baseline. To identify a meaningful deviation in someone’s behaviour, you need to know what their normal looks like first. Trump has a documented, cross-contextual pattern in physical interactions with world leaders, such as extended handshakes, palm down grip, pulling the other person toward him, and prolonged contact as a dominance signal. He did this with Macron in 2017 for twenty-eight seconds. With Shinzo Abe. With Trudeau. With multiple NATO leaders.
With Xi, that pattern was absent. The slower tempo, the warmth gesture, and the absence of a real aggressive pull, that is a genuine deviation from baseline that is worth examining. But the coverage largely skipped the baseline entirely and went straight to the verdict. Without baseline, you have no signal. You have a snapshot you can narrate any way you choose.
The third error is the most important one, for my two pennies: confidence calibrated to narrative rather than evidence. The commentators who looked at fifteen seconds of footage and told you exactly what both men were thinking and feeling were doing something the research on human judgment tells us is almost certainly wrong. Not because they were being inherently dishonest. Because the brain generates plausible explanations with extraordinary fluency, and we mistake that fluency for insight.
The cultural dimension nobody mentioned
There is a fourth error in the coverage of this particular moment that is worth its own section because it illustrates something important about how behavioural analysis goes wrong at a structural level.
The United States and China do not just have different political systems. They operate from fundamentally different cultural orientations toward the relationship between the individual and the group. The psychologist Geert Hofstede spent decades mapping these differences across countries. The United States consistently scores among the highest in the world on individualism. China scores among the lowest, deep in the collectivist range.
In individualist cultures, public emotional expression is not just permitted, it is expected. Warmth is performed visibly. Enthusiasm is communicated physically. A large physical gesture signals confidence and personal investment. In collectivist cultures, the logic runs differently. Public emotional restraint in high-stakes formal settings is not emotional distance. It is social competence. It demonstrates that you understand the weight of the occasion and are not going to subordinate a state-level moment to personal performance.
Xi’s composure, read through a Western individualist lens, looks like withholding. Read through the cultural framework it was actually produced in, it looks like someone performing exactly as his cultural and political context requires. Those are completely different conclusions. Every analyst who produced the “Xi won the handshake” narrative did so by applying a Western cultural operating system to interpret behaviour produced within a Chinese one. They did not acknowledge that an alternative frame existed.
Hofstede’s country comparison data is publicly available at Hofstede Insights. The gap between the US and China on the individualism dimension is significant and directly relevant to how formal public interactions between their leaders should be read.
This is precisely the problem with treating body language as a universal language. It is not one. The claim has been refuted repeatedly in the cross-cultural literature. The consequences of treating it as universal range from embarrassing, such as confidently misreading a diplomatic interaction on YouTube, to genuinely serious, such as misreading cultural signals in professional or investigative contexts where accuracy matters.
What you can actually take from the footage
After all of that, here is what I think the evidence from the Beijing handshake actually supports, stated carefully, with the appropriate qualifications.
First, Trump’s approach to this interaction was meaningfully different from his documented baseline with other world leaders. The absence of the aggressive pull, the slower tempo, and the warmth gesture represent a real deviation that is worth noting. Whether it reflects strategic calculation, a genuine shift in how he is reading the relationship, or real-time adjustment to an interaction partner who has never responded to his usual physical signals, the evidence does not allow a definitive answer. But something was different here. That is an observable fact.
Second, Xi’s composure is consistent with his baseline in formal ceremonial settings and with the cultural framework he operates within. Reading it as a specific dominance response to Trump requires projecting an internal state onto behaviour that is equally well explained by cultural context and personal default register. That projection is not justified by the footage.
Third, and most importantly, the environment communicated more than the handshake did. Xi’s capital. Xi’s choreography. Xi’s military honour guard. The power signal in this interaction was in the architecture of the event, not the geometry of the grip. Reading the handshake in isolation misses where the actual message was.
That is the honest read. It is less satisfying than a clean verdict. It will not get as many YouTube views. That is the point.
The three questions that separate analysis from guessing
The practical thing to take from this is a habit, not a fact. Guessing within the realm of reasoning and thought should, in my humble opinion, be considered swearing. It has the feeling of confidence attached, but by every definition, it is based on nothing more than a confident leap.
Before you make a judgment about what someone’s behaviour means, whether that be in a video, in a meeting, in a negotiation, or in any interaction where you are reading another person, ask yourself three questions.
What is the context, including the cultural context? What is this person’s baseline, what does their normal look like, and am I actually seeing a deviation from it? What are the other possible explanations for what I just observed before I settle on the one that fits my existing story?
If you cannot answer all three, what you have is a first impression. First impressions are useful data. They are not conclusions. The skill, in investigation, in negotiation, and in any professional context where reading people matters, is in knowing which one you are working with and not dressing the first one up as the second. The brain is extraordinarily good at generating stories that feel like understanding. The gap between that feeling and actual accuracy is where most behavioural analysis, popular and professional, quietly lives.
Closing that gap is a discipline. It is learnable. But it starts with being honest about how wide the gap actually is.
Read more from Ben Cardall
Ben Cardall, C.E.O - Investigator
Ben Cardall is a human behaviour and memory expert, bestselling author, and professional investigator specialising in observation, reasoning, and decision-making under pressure. He has worked with elite security teams and personnel from around the world, training high-performing professionals in behavioural intelligence and situational awareness. Through immersive programs, courses, and live experiences, Ben teaches memory mastery, critical thinking, and real-world profiling skills grounded in neuroscience and psychology. His work bridges rigorous science with practical tools for security, investigations, leadership, and personal growth.
Follow Ben and the series at Omniscient Insights on YouTube:
Vincent Denault, Pseudoscience of body language
Daniel Kahneman, Thinking, Fast and Slow
Bond and DePaulo, Accuracy of deception judgments, 2006
Lisa Feldman Barrett, How Emotions Are Made
Geert Hofstede, Culture’s Consequences
Hofstede Insights country comparison tool, USA vs China live data
