Baseline First: Why Behavior Means Nothing Without a Norm

Joe Navarro spent twenty-five years in the FBI's National Security Division working counterintelligence, which means his job was largely to sit across from people who were lying to him professionally. The single instruction he repeats most often in his writing is not about any particular gesture. It is that gesture dictionaries are worthless.

There is no behavior that means deception. There is no posture that means guilt. Crossed arms do not mean defensiveness; they mean the person is cold, or comfortable, or that is what they do with their arms. What carries information is change — a departure from what this specific person does ordinarily, time-locked to a specific stimulus.

This is the most important chapter in Part 2, and it is the one most consistently skipped by the industry that sells this material, because a dictionary can be printed on a card and a baselining protocol cannot.

The reasoning is straightforward once stated. Individual variation in nonverbal behavior is enormous, and it dwarfs the variation produced by emotional states. Some people gesture constantly, blink rapidly, shift in their seats, touch their faces every thirty seconds, and speak in bursts with long pauses. Others sit almost immobile. Both are baselines, not signals. An observer who reads the first person's ordinary manner as anxiety and the second's as coldness has committed a version of the fundamental attribution error — treating stable individual style as evidence of a transient internal state.

The corrective is change detection, and it runs in four steps.

Establish the norm. Spend the opening minutes on neutral, low-stakes conversation, and watch rather than listen: gesture rate, self-touch, posture, foot position, blink rate, speech tempo, pause length. This is what the small talk at the start of an interview is actually for, and the reason a skilled interviewer will not let you skip it.

Introduce a stimulus. A specific question, a document, a name.

Measure the deviation. Not a single sign — a cluster. Multiple channels changing together is the observable; one channel changing is noise. Navarro's rule of thumb is that you want three or more.

And time-lock it. The deviation has to be tied to the stimulus with reasonable precision. A change in manner that arrives forty seconds after a question, or that persists identically through six unrelated questions, is not about the question.

Then, having done all of that correctly, you have established exactly one thing: that this topic produced arousal in this person. Which is where the discipline has to stop, and where nearly everyone in practice keeps going. The next chapter explains why that stopping point is not a technicality but the whole safety margin of the method — an innocent person accused of something serious produces the same physiology as a guilty one, and every miscarriage in this part of the guide lives in the gap between arousal and guilt.