How a Secret Service agent, CIA officer, mentalist and poker pro read people

There is no trick for spotting a liar at a glance. Four people whose work depends on reading others describe much the same method: learn a person's normal behavior first, then watch for departures from it. All four treat a read as information for a better decision, not as proof.

Evy Poumpouras, a former Secret Service polygraph examiner, is the most cautious. She says the research on body language and deception goes back and forth. What she trusts is narrower: people give off cues, and the cues mean something only against the person's own baseline, which she says you can get in a couple of minutes. Crossed arms are her own resting posture and mean nothing on her. Then she checks whether what a person says, what the body shows and what they do are in harmony. She calls the idea that looking away means lying largely a myth, because eye contact depends on temperament and history. Her rules are never to call people out on what you notice, to ask simple questions, and to weigh behavior over speech. "You don't need the smoking gun from people. They show you."

Jonna Mendez, the CIA's former Chief of Disguise, says liars often try too hard to convince you. She watches for stories that bounce around, nervousness and a point that will not be dropped. Her disguise work shows that an impression is built from a handful of details and a manner, and that both can be manufactured. She tracks her hunches to see later whether they were right.

Oz Pearlman, a mentalist, says there is no foolproof sign. What changes when people lie is their level of detail, their speed and their cadence, measured against their own pattern. A habitual embellisher may give himself away by being brief. He trusts first instincts as a statistical advantage, not a certainty.

Daniel Negreanu, a poker professional, says "Everything everyone does at the poker table conveys information." He makes opponents comfortable, because relaxed people give off more, and he keeps written profiles. He thinks in ranges, holding every possibility open and letting each action remove some. He also separates a good read from a good result.

They agree that over-selling is a sign, since the truth is usually simple, and that charm deserves wariness. They disagree on eye contact. Poumpouras and Pearlman do not accept it as a cue, while Mendez says it is often among the first things to falter. They also differ on how far to trust a first instinct.

The method works face to face with a little time. It fails against trained people, and it fails when you compare someone to a general rule instead of to themselves.

What to do with it: spend the first minutes on easy talk and notice how the person normally stands, speaks and looks at you. Watch for change in either direction. Check that words, body and actions agree, and give most weight to actions. Listen for oaths, surplus detail and selling. Notice whether your question was actually answered. Do not use eye contact as a test. Make people comfortable and keep what you see to yourself. Hold several explanations open. Write down your hunches and check them later. Remember that you are being read too.