How to Conduct a Workplace Investigation Interview Without Creating Legal Liability

Picture yourself in a deposition or termination hearing. Opposing counsel leans forward and asks a simple question: "On what basis did you conclude that the respondent was being deceptive during your interview?"

Now imagine your answer is "She broke eye contact when I asked the hard questions" or "He touched his face—that's a tell" or "We were trained to look for microexpressions and clusters of behavior."

Will your workplace investigations training stand up to scrutiny? Maybe not.

That answer doesn't just weaken your case. It exposes you personally. Not the lie detection guru who sold your organization a two-day seminar. Not the vendor whose workbook is still in your drawer. You. Your name is on the report. Your judgment is on the record. Your professional reputation is the one being cross-examined.

This is not a hypothetical. It is the predictable outcome of building workplace investigations on junk science — and it happens more often than most organizations realize.

The Lie Detection Problem Nobody Wants to Talk About

The private sector has a pseudoscience problem.

Walk into almost any HR investigation training, loss prevention certification, or corporate security seminar and you will find some version of behavioral lie detection woven into the curriculum. It goes by different names—behavioral analysis, body language reading, microexpression detection, and baselining—but the underlying claim is the same: that trained observers can detect deception through behavioral cues.

The science says otherwise. Clearly and repeatedly.

Keep so-called truth and lie wizards out of interview and investigation training.

The most comprehensive review of human lie detection ability, conducted by Bond and DePaulo in 2006, examined over 200 studies and more than 24,000 judgments. Their finding was striking in its plainness: people correctly identify truth and deception at a rate of roughly 54 percent — barely better than a coin flip. Trained professionals, including law enforcement officers, did not perform meaningfully better than untrained civilians. The behavioral cues that people rely on most confidently — gaze aversion, fidgeting, nervousness, and speech hesitation — have no reliable relationship to deception in the research literature.

Fifty-four percent. That is your foundation if you build your credibility assessment on behavioral lie detection. That is what you are defending at a deposition, at a termination hearing, at an EEOC proceeding, or in front of a jury.

The "lie wizard" who ran your seminar will not be there with you.

"Soft Skills" Is the Wrong Label—And the Wrong Excuse

Before going further, it is worth addressing a reflex that shows up in some professional circles—the dismissal of interviewing skill as soft, touchy-feely, or beneath the work.

That reflex is expensive.

Workplace violence prevention, misconduct investigation, harassment inquiry, fraud detection — none of these are soft problems. The person sitting across from you during a high-stakes workplace interview is not a soft situation. And the deposition that follows a poorly conducted investigation is definitely not soft.

The investigators and HR professionals who get this right are not the ones with the most confidence going in. They are the ones with the best skills (true confidence) — the ones who know how to gather information cleanly, handle resistance without escalating it, and build a record that survives review. That requires training, not a checklist, and not a seminar built around behavioral myths.

Why Step-by-Step Interview Systems Fail in the Real World

Here is something almost every workplace investigator figures out eventually, usually the hard way: conversations do not follow scripts.

Step-by-step interviewing systems—the kind that walk you through Phase 1, Phase 2, Phase 3, and a prescribed set of moves—are designed for a subject who complies with the choreography. Real people do not. They interrupt. They go sideways. They bring up things you did not expect. They shut down, deflect, get emotional, or circle back to something you thought you had already passed.

When an interviewer is locked into a sequence, or so-called best practices, those moments become problems instead of opportunities. The investigator tries to get back on script rather than following what the conversation is actually offering or how their memory is actually working. Important details are missed. Rapport breaks down. The interview becomes a transaction instead of a genuine cooperative exchange of information.

This is not a flaw in the investigator's execution. It is a flaw in the design of the system.

Step-by-step interrogation models were built around mythical lie detection cues and a confession-driven goal—move the subject from denial to admission through a fixed sequence of pressure and persuasion. That model was not designed for fact-finding. It was designed for closing. And in a workplace investigation, closing prematurely on the wrong conclusion is exactly the outcome you cannot afford.

What works instead is a framework built for adaptability — one that gives the investigator a clear sense of purpose and direction without locking them into a sequence that the subject will inevitably disrupt. Interview planning becomes the foundation: knowing what you are trying to learn, what evidence you have and when/if to use it, and what competing explanations you need to test. The conversation itself stays flexible, responsive, and open.

What Defensible Investigations Actually Require

A workplace investigation is not defended by simply saying, "we did an investigation." It gets defended by showing how it was done — question by question, decision by decision, conclusion by conclusion.

That standard has three practical implications for anyone conducting workplace interviews.

First, credibility assessment needs to be grounded in the statement, not the person's face or gestures. The Strategic Use of Evidence (SUE) technique gives investigators a research-supported method for assessing credibility through what a person says—how their account holds together, where it contradicts known facts, what it omits, and how it responds to evidence when evidence is introduced. Statement-evidence inconsistencies can be documented, explained, and defended. "She seemed nervous and broke eye contact" cannot, at least not in any venue that takes science seriously.

Second, the quality of information gathered is only as good as the quality of the questions asked. Closed yes-or-no questions are necessary, but they yield only closed yes-or-no answers. Investigators who rely on confirmatory questioning — questions designed to verify what they already believe — end up with a record that reflects their hypothesis rather than the facts. Open-ended questions and free narratives consistently produce richer, more detailed, and more accurate accounts. They also give investigators more material to test and follow up on.

Third, rapport is not optional—it is operational. An interviewee who feels judged, cornered, or accused before the conversation begins is unlikely to give a complete account. Rapport is not about being friendly, nice, or engaging in small talk. It is about creating the conditions under which people are willing to talk openly, share details, and correct the record when it is wrong. Research consistently shows that rapport improves both the quantity and quality of information disclosed — and that without it, resistance in high-stakes interviews is significantly harder to overcome.

The Deposition Frame: Think Backwards from Accountability

One of the most clarifying red teaming questions an investigator can ask before every workplace interview is this: If I had to explain every decision I made in this conversation to opposing counsel six months from now, could I?

Could you explain why you asked that question in that sequence? Could you explain what you were testing when you introduced that piece of evidence? Could you explain the basis for your credibility assessment—not as a gut feeling, but as a documented, statement-supported evaluation of the account against the facts?

If the answer is no, the investigation has a problem — one that no final report language will fix.

This is where science-based interviewing gives HR professionals, compliance teams, and corporate investigators a genuine advantage. The methods are documented in peer-reviewed research. The logic is transparent and explainable. The credibility assessments are grounded in the statement and the evidence rather than folklore. When those decisions come under scrutiny—and in consequential cases they will—the investigator has something to stand on.

The lie detection vendor does not share that accountability. The body language trainer has moved on to the next seminar. The pseudoscience stays in your report.

What to Look for in Workplace Investigation Training

Not all investigation training is equal. Some of it actively creates the liability it claims to prevent.

When evaluating training for your team, look for programs that:

• Ground credibility assessment in the statement, the evidence, and documented logic—not behavioral cues or intuition

• Teach active listening as a cognitive discipline rather than a checklist of behaviors

• Build adaptability through interview planning skills rather than rigid step-by-step sequences

• Use research citations from peer-reviewed sources—not marketing materials, appeals to an authority, or anecdotes

• Explicitly address the pseudoscience that still circulates in HR and loss prevention training. Get rid of it!

• Prepare investigators to explain and defend their process, not just their conclusions

Be skeptical of any training that includes behavioral analysis, microexpression detection, neurolinguistic programming (NLP), baselining, or any technique that promises deception detection through observation. The research does not support these methods. More importantly, the courtroom does not forgive them.

The Bottom Line

Workplace investigations carry real stakes—for the people involved, for the organization, and for the professional conducting them. The standard is not just reaching a conclusion. It is reaching a defensible conclusion through a documented, fair, and methodologically sound process.

That standard cannot be met with junk science that still circulates in training curricula, rigid step-by-step scripts, or untested intuitions about what liars look like.

Science-based interviewing was built for exactly this environment — not to make investigations easier, but to improve them. Better questions. Better information. Better decisions. And a process that holds up when it matters most.

Your name is on the report. Make sure what's in it can be defended.

Want to bring science-based interview training to your HR, compliance, law enforcement, or corporate security team? See if we're a fit.

Workplace Investigation References

Bond, C. F., Jr., & DePaulo, B. M. (2006). Accuracy of deception judgments. Personality and Social Psychology Review, 10(3), 214–234. https://doi.org/10.1207/s15327957pspr1003_2

Bond, C. F., Jr., & DePaulo, B. M. (2008). Individual differences in judging deception: Accuracy and bias. Psychological Bulletin, 134(4), 477–492. https://doi.org/10.1037/0033-2909.134.4.477