There is a claim that turns up often in popular writing about reading people, and it is genuinely appealing. The face can be composed and the hands can be managed, the argument goes, but the feet keep telling the truth. Someone can hold your gaze and nod along while one foot quietly rotates toward the door, and that foot, not the face, reports where the person would rather be.
Tidy, a little conspiratorial, and easy to test against your own memory of awkward parties: the appeal is real. It is also worth slowing down on, because the distance between how confidently the idea is stated and how well it has actually been measured is larger than most retellings admit.
We are writers, not psychologists, and this is a reading of where the claim comes from rather than a verdict on anyone’s living room. What interests us is not whether feet ever reveal anything. It is how a modest observation hardened into a rule.
Where the idea comes from
Joe Navarro, a former FBI counterintelligence agent, put the modern form of this on the map. His book What Every BODY Is Saying (HarperCollins, 2008, written with Marvin Karlins) argues that the lower body is the honest half of a person. His reasoning is partly evolutionary and partly practical. The legs and feet, he suggests, were central to our ancestors’ survival responses, so they react quickly and without much conscious editing. And because we tend to watch faces when we read other people, we also police our own faces when we want to hide something, leaving the feet unsupervised.
Navarro is direct about the basis for this. In his own account of the behaviour, he describes decades of watching people during interviews and says the feet are often more honest than the face, and that a foot turning ninety degrees toward an exit while the person keeps looking at you is, in his experience, a reliable sign of trouble. That is field observation from someone who spent a career doing it. It is not the same thing as a controlled study, and it is useful to keep the two apart before deciding how much weight the claim can carry.
What the leakage hypothesis actually claimed
The academic ancestor of the feet idea is older and more careful than the version that reaches Instagram. In 1969, Paul Ekman and Wallace Friesen published a paper on nonverbal leakage in the journal Psychiatry. They proposed what became known as a leakage hierarchy: some channels of the body are easier to control than others, and because the face is the one we rehearse and the one others watch, it is also the one we manage hardest. The body below the face, on this view, is a comparatively leakier source of feelings a person is trying to conceal.
That is a claim about relative controllability, and it is a reasonable one. A later quantitative summary by Miron Zuckerman, Bella DePaulo and Robert Rosenthal in 1981 did find, across a small set of studies, that the body gave away deception somewhat more readily than the face. So the seed of the popular idea has real roots.
We do rehearse the face.
The trouble begins when a statement about which region is relatively less guarded gets upgraded into a promise that a specific body part reliably reports a specific internal state.
Where the evidence gets thinner
Its most thorough measurement comes from DePaulo and colleagues’ 2003 review in Psychological Bulletin, which pulled together 1,338 estimates of 158 possible cues to deception drawn from 120 independent samples. For anyone hoping for a single honest tell, the finding is sobering. A few cues did distinguish liars from truth-tellers, but many behaviours showed no discernible link, or only a weak one, and the effect sizes overall were small by the usual benchmarks.
On the feet and legs specifically, the picture is not just weak. It is contradictory. As one review of the deception literature lays out, Sporer and Schwandt’s 2007 meta-analysis reported that liars tended to move their feet and legs less, while DePaulo’s 2003 review and a later analysis by Hartwig and Bond in 2011 found no measurable difference at all. Different pooled analyses of the same behaviour reached different answers. That is the opposite of the settled, readable signal the popular claim implies.
Two further wrinkles are worth naming. Charles Bond and colleagues have described a decline effect in this field, where the longer a given cue is studied, the smaller its measured association with lying tends to become. Some researchers have gone further, arguing that the precision of these estimates is low enough that even the modest conclusions deserve caution. None of this is one study being overturned by another. It is a body of work that has slowly grown less confident the more closely it has looked.
The reading that overshoots
There is a further slippage worth catching. Deception research measures lying, and the feet idea as it is usually used is not really about lying at all. It is about engagement and intention: whether someone wants to stay in the conversation. Those are related but not identical, and the engagement version has been studied far less rigorously than the deception one. Its best-tested cousin is already shaky, and the everyday claim rests mostly on expert observation rather than controlled trials.
The cultural assumption underneath all of this is that the body cannot lie, that there is an involuntary channel running beneath the social performance which a trained eye can simply read off. It is a comforting idea, because it promises that other people are legible if you know where to look. Research does not support the strong form of that promise. Behaviour is noisy, context does much of the work, and a foot angled toward the door might mean disengagement, or a full bladder, or nothing, or the plain fact that chairs and rooms push bodies into positions that carry no inner meaning at all.
What is still worth watching
This does not make the observation useless. A person’s orientation, whether they are turning toward you or squaring away, is real information, and in our own experience of interviews and conversations it often tracks whether someone is settling in or getting ready to go. The mistake is treating it as a decoder rather than a hint.
Both extremes miss it. Feet do signal something sometimes, and they are also not the private confession the claim imagines. At best they offer one weak, easily confounded data point among many, worth attention when it lines up with everything else you are noticing and worth ignoring when you have already decided what you want it to mean.
What is worth keeping is smaller than the one that sells books. Watch orientation if you like, and treat a turn toward the exit as a reason to check in, not a reason to conclude. The feet are not lying to you. They are just not saying as much as the confident version insists, and mistaking a hint for a verdict is how people end up flattering themselves that they can read a room.