How HR Professionals Read Candidate Body Language in Interviews

Reading candidate body language in interviews comes down to one disciplined method: establish the candidate's behavioural baseline in the first few minutes, then watch for clusters of deviation when specific topics arise. Trained interviewers do not look for a single "tell" — they look for change, in context, repeated across multiple channels.

That distinction matters because most of what passes for interview body language advice is folklore. Crossed arms do not mean deception. Lack of eye contact does not mean dishonesty. What the research and field practice actually support is far more useful — and far more learnable.

Why untrained reading goes wrong before the interview even starts

Alexander Todorov's research at Princeton, including the well-known 2006 study with Janine Willis, showed that people form trait judgements — competence, trustworthiness, likeability — from a face in as little as 100 milliseconds, and that additional exposure time mainly serves to increase confidence in that snap judgement rather than revise it. For an interviewer, that is a problem: by the time the candidate sits down, an impression has already formed, and everything afterwards risks becoming confirmation.

Daniel Kahneman and Amos Tversky's work on judgement under uncertainty explains the mechanism. The anchoring effect means the first impression drags every subsequent assessment towards it; the halo effect means a single positive trait (a confident handshake, a good suit) colours unrelated judgements about competence. Kahneman describes both at length in Thinking, Fast and Slow. An interviewer who has not been trained to notice these biases is not reading the candidate — they are reading their own first 100 milliseconds.

The first job of non-verbal training for HR, then, is not to add skills. It is to subtract noise.

The baseline-deviation method

The approach we teach at BodyLytics — grounded in the work of former FBI counterintelligence agent Joe Navarro, particularly his book What Every BODY Is Saying — has three stages.

1. Establish the baseline during low-stakes conversation

Every candidate arrives with their own normal. Some people fidget constantly; some sit still as statues; some never make sustained eye contact even when describing their proudest achievement. None of that tells you anything by itself. So the first five to ten minutes — the commute, the weather, the walk through their CV chronology — are not throwaway small talk. They are your calibration window. Note their resting posture, gesture rate, speech tempo, blink rate, where their hands live, and how they handle pauses.

2. Watch for deviations from that baseline, tied to specific topics

The signal is never the behaviour itself; it is the change. A candidate who has been animated and forward-leaning for twenty minutes and suddenly stills, withdraws their hands from the table, and angles their torso away when you ask why they left their last role — that shift is worth noting. Navarro frames this through the lens of comfort and discomfort: the limbic system responds to perceived threat with freeze, distancing, and blocking behaviours, and with what he calls pacifying behaviours — neck touching, collar adjusting, leg rubbing — that serve to self-soothe under stress.

3. Read clusters, never single cues — and treat them as questions, not answers

One deviation is weather; a cluster is climate. A trained interviewer waits for two or three discomfort signals occurring together, anchored to the same topic, before doing anything — and the thing they do is not "mark the candidate down". It is to ask a follow-up question. Discomfort tells you where to dig. It does not tell you what you will find. The candidate squirming over their employment gap may be hiding a dismissal — or grieving a parent. The non-verbal data identifies the topic that needs verbal exploration. Nothing more.

What this method explicitly is not: lie detection

This needs saying plainly, because it is where most interview body language content goes badly wrong. Paul Ekman — whose decades of research on facial expression underpin much of modern emotion science — is explicit that there is no single behavioural sign of deception, no "Pinocchio response". His work on micro expressions shows that brief, involuntary facial expressions can leak concealed emotion, but leaked emotion is not the same thing as a lie. A flash of fear when asked about reporting lines might mean the candidate is fabricating — or that their last manager was a bully.

In BodyLytics sessions with HR teams, we consistently see the same pattern: interviewers arrive wanting a deception checklist and leave understanding that the honest, defensible use of this skill is hypothesis generation. You observe, you note, you probe verbally, you verify with references and evidence. The non-verbal channel guides the conversation; it never replaces it.

What trained interviewers actually track

• Congruence between words and body. "I am really excited about this role" delivered with a retreating posture and a half-second delay is a mismatch worth a follow-up. When words and non-verbals conflict on an emotional claim, people tend to weight the non-verbal channel — but this applies specifically to messages about feelings, not communication in general. When words and non-verbals align, the words carry the meaning.
• Topic-linked pacifying behaviours. Navarro's catalogue — neck dimple covering, ventilating a collar, wringing hands — appearing in clusters when a particular subject arises.
• Distancing and blocking. Torso turning away, objects (a notebook, a cup) moved between interviewer and candidate, feet reorienting towards the door.
• Recovery time. Everyone shows stress under a hard question; the more informative observation is how quickly the candidate returns to baseline once they have answered.
• Their own behaviour. Interviewers leak too. A panel member frowning at their notes can suppress a candidate's expressiveness and contaminate the very behaviour everyone is trying to read.

Building this into a fair, structured process

Non-verbal observation belongs inside structured interviewing, not as a substitute for it. The practical implementation we recommend to HR teams across Europe:

• Keep the same low-stakes opening for every candidate, so every baseline is established under comparable conditions.
• Separate observation from interpretation in your notes: write "paused four seconds, touched neck, broke eye contact when asked about the gap" rather than "seemed shifty".
• Probe deviations in the room, while you can. A cluster you never followed up on is wasted data — and an unfair basis for a decision.
• Never score body language directly. Score the verbal answers your observations helped you elicit.
• Debrief biases explicitly: ask the panel what their first-minute impression was, and whether the evidence gathered afterwards genuinely supports it.

Handled this way, behavioural observation makes interviews fairer, not less fair — it pushes interviewers to test their snap judgements against evidence rather than ride them to a decision.

Train the skill properly

Reading people is a trainable, perishable skill, like any other professional competency — and it is far better learnt through guided practice with feedback than from articles, including this one. Our non-verbal communication course for human resources professionals covers baseline-setting, cluster reading, bias management and structured note-taking, with live observation exercises built around real interview scenarios.

If your organisation runs assessment centres, high-volume hiring or senior-level interviews, the Non-Verbal Communication for Human Resources course gives your interviewers a defensible, evidence-grounded method — one that improves the quality of hiring conversations instead of adding pseudoscience to them. Get in touch to discuss a team session in Barcelona or on-site anywhere in Europe.