6 Little Habits That Reveal More Than People Realize
Most of the “body language expert” industry runs on stuff that hasn’t replicated — the eyes-up-and-left lie tell, fidgeting as guilt, crossed arms as defensiveness. Almost none of it survives a meta-analysis. What does survive is much weirder, and usually has nothing to do with whatever folk wisdom claims it means. Six small habits where the underlying research is solid, the correlation is real, and what’s actually being revealed is more interesting than the paperback version.
How fast you walk down a sidewalk predicts how long you’ll live — accurately enough that geriatricians treat it as a vital sign
Self-selected walking speed on flat ground turns out to be a freakishly strong mortality predictor. A 2011 JAMA pooled analysis of nine cohorts by Stephanie Studenski and colleagues tracked 34,485 community-dwelling adults aged 65 and up. Per 0.1 m/s increase in habitual gait speed, all-cause mortality dropped 12%. Predicted 10-year survival at age 75 ranged from 19% to 87% in men, depending on where they sat on the gait-speed distribution.
The predictive power is comparable to combining age, sex, smoking, blood pressure, BMI, and chronic conditions — all of it — in one model. A stopwatch and four meters of hallway outperforms a multi-variable risk panel.
The reason: gait speed is a single integrated readout of cardiopulmonary, neurological, musculoskeletal, and cognitive function at once. The leg moves because all the systems behind it are working. When the leg slows, something upstream is failing, and the body is doing the math whether anyone’s watching or not.
A photo of you from when you were ten years old can predict whether you’ll get divorced — based on whether your smile crinkled around the eyes
The “Duchenne smile” is the one where the orbicularis oculi muscles around the eyes contract along with the mouth, producing crow’s feet and lifted cheeks. It’s notoriously hard to fake voluntarily. Paul Ekman’s group showed in 1990 that it tracks left-frontal EEG activation associated with positive approach emotion.
The twist came almost twenty years later. Matthew Hertenstein and colleagues’ 2009 paper in Motivation and Emotion coded smile intensity in college yearbook photos, and separately in photos from childhood through early adulthood. People in the bottom 10% of smile intensity were roughly five times more likely to be divorced decades later than people in the top 10%. The effect held in both samples.
Smile intensity isn’t a state, in other words — it’s a stable trait that indexes dispositional positivity and emotion regulation. A picture from elementary school is already leaking your divorce odds, because the disposition that produces the smile is the same disposition that maintains a marriage forty years later.
People who use “I” a lot in conversation aren’t narcissistic — they’re depressed, and the narcissism myth is one of psychology’s better-debunked claims
The folk wisdom is that I-heavy speech signals self-importance or narcissism. The data say the opposite. Rude, Gortner, and Pennebaker’s 2004 study in Cognition and Emotion analyzed essays from currently-depressed, formerly-depressed, and never-depressed college students. Currently-depressed students used significantly more first-person singular pronouns. The finding has been replicated many times since.
Then Carey and colleagues’ 2015 meta-analysis explicitly tested the “narcissists say I a lot” claim and found nothing. Narcissists use “I” at roughly normal rates. The pop-psych version is wrong twice over: I-use doesn’t track narcissism, and the people it does track are the ones the popular reading would never suspect.
The depressed person isn’t bragging. Their attentional spotlight is stuck on themselves — what clinical psychology calls self-focused rumination — and the pronouns leak it. The most-revealing thing about a habit can be that everything we say about people who have it is wrong.
A four-second silence in conversation is enough to make people feel they don’t belong — and the people who rush to fill it are leaking their conformability
Silence in conversation is uncomfortable in measurable ways. A 2011 study by Namkje Koudenburg and colleagues in the Journal of Experimental Social Psychology found that a brief silence of just four seconds in a group conversation reliably produced negative emotion, reduced feelings of belonging, lowered self-esteem, and — here’s the part that matters — increased the participants’ motivation to conform to the group’s stated opinion afterward. Participants didn’t consciously notice the silence. Their belonging measures moved anyway.
So an “awkward silence” isn’t subjective squeamishness. It activates the social-pain system on a measurable clock, and the same people who feel it most strongly are the ones whose stated opinions then drift toward the group consensus.
The tell isn’t just discomfort. The talkers who can’t sit with a four-second gap are revealing two things at once: belonging anxiety, and conformability. (Worth flagging: the four-second threshold is Anglophone-specific. Japanese and Finnish conversational norms tolerate much longer silences without distress.)
Vocal fry doesn’t actually correlate with anything about the speaker — it correlates with a measurable bias in the listener, and the bias is sex-specific
Vocal fry is the creaky, low-pitched register at the end of phrases. A 2014 PLOS ONE study by Rindy Anderson and colleagues had a national US sample rate voice clips. Fry voices were rated less competent, less educated, less trustworthy, less attractive, and less hirable. The penalty was substantially larger for female speakers than male ones. Given a hypothetical hiring choice between a normal-voice candidate and a fry-voice candidate, listeners picked the normal voice roughly 80% of the time.
Here’s the thing: vocal fry doesn’t actually correlate with the speaker being less competent. It correlates with how the listener judges them.
So the tell isn’t about the speaker. It’s about whoever’s listening. The habit leaks something specific about the listener’s bias machinery — usually a sex-differentiated bias they wouldn’t admit to if asked directly. The fry isn’t doing the revealing. The reaction to it is.
People who say they can tell when someone is lying are wrong — and the most consistent tell across the lie-detection literature is the believer’s own overconfidence
The classic “lying tells” — avoiding eye contact, fidgeting, touching the face, looking up and to the left, shifting in their seat — don’t survive empirical scrutiny. Aldert Vrij and Pär Anders Granhag’s review of the deception-detection literature, drawing on Bond and DePaulo’s meta-analysis of 206 studies and 24,483 judges, puts average lie/truth discrimination from behavioral cues at 54% — barely above chance.
An earlier DePaulo meta-analysis of 158 candidate cues found that the strongest had small effect sizes around d = 0.1 to 0.3, with massive overlap between liars and truth-tellers. The NLP “eyes-up-and-to-the-left” claim was specifically tested by Wiseman in 2012 and rejected outright.
The real tell isn’t the liar’s body. It’s the listener’s confidence. People who claim they can spot liars — cops, customs officers, “body language experts” — perform at chance in controlled studies but report high subjective certainty about each call. The habit that reveals the most isn’t a tell. It’s believing you have one.
The most-revealing habits aren’t the ones in pop-psych paperbacks. The body-language industry sells the loudest cues and ignores the ones with actual predictive power. You learn more from how fast someone walks than from where they look.
