Rethinking Credibility Assessment: What Science Tells Us About Detecting Deception
IN THIS ISSUE: A research-based approach to credibility assessment that challenges common assumptions about detecting deception in workplace investigations.
The 54% Problem We Do Not Talk About
If you rely on body language, nervous mannerisms, or intuition to detect deception during a workplace investigation, the research offers a cautionary message: those methods generally perform only slightly better than chance.
Research associated with Timothy R. Levine’s Truth-Default Theory reflects a longstanding finding in deception-detection research: people identify deception with approximately 54% accuracy, on average. Professional experience alone does not appear to eliminate this limitation. Organizations whose investigation protocols were developed without considering this research may benefit from reviewing how credibility assessments are conducted.
Key Insights for Workplace Investigations
Approaches That Are Not Reliably Predictive
- Looking for “shifty eyes” or nervous behavior
- Relying on speech hesitations or other verbal patterns in isolation
- Trusting an initial impression about who seems honest
- Treating traditional behavioral cues as reliable indicators of deception
Approaches That May Produce More Useful Information
- Diagnostic questioning: Asking questions designed to elicit information that can help distinguish between competing accounts
- Content analysis: Focusing on what a person says, rather than primarily on how the person appears while saying it
- Progressive interviewing: Moving from an uninterrupted account to focused follow-up questions
- Evidence comparison: Assessing statements against documents, records, witness accounts, and other verifiable information
What Research Suggests About Questioning
Controlled studies indicate that the questions investigators ask can affect the accuracy of credibility assessments. Research on diagnostic questioning has reported accuracy rates above 70% in some experiments, although results vary by study design, circumstances, and the skill with which the questions are asked.
The practical lesson is not that an investigator can identify deception with certainty. Rather, carefully designed questions may help uncover information that can be evaluated against independent evidence.
Five Practical Interviewing Techniques
These techniques require preparation and practice, particularly when an interview involves sensitive allegations or conflicting accounts. Investigators should also consider applicable company policies, contractual obligations, and legal requirements before conducting an interview.
- Start with the story. Begin with an open-ended invitation to provide a complete account before turning to specific details.
- Circle back strategically. Return to important subjects from different angles during the interview. The purpose is not to treat minor inconsistencies as proof of deception, but to test significant details and identify subjects requiring further inquiry.
- Probe the periphery. Ask about surrounding circumstances that may provide context: “What happened immediately before that?” or “Who else may have observed what occurred?”
- Explore the person’s reasoning. Ask questions that clarify how the witness understood the conduct and circumstances at issue. The answers may provide useful context when evaluated alongside the other evidence.
- Address information that may come from other sources. Explain that the investigation may include additional interviews, records, or other evidence. For example: “What do you expect your colleague will say when I speak with them?” Evaluate the response against the information later obtained rather than treating the person’s immediate reaction as proof of deception.
The Truth-Default Tendency
One counterintuitive finding in deception research is that people generally tend to believe what others tell them. This tendency may be adaptive because most ordinary communication is honest.
The practical objective is not to become reflexively suspicious. It is to recognize when the available information warrants additional questions or corroboration.
Circumstances That May Warrant Deeper Inquiry
- Clear contradictions within an account or between an account and other evidence
- Claims that appear implausible in light of the surrounding circumstances
- A possible motive to misrepresent events
- Material gaps that additional questioning may clarify
Applying the Approach: A Simple Example
Instead of: “You seem nervous. Are you lying to me?”
Consider: “Please walk me through exactly what happened after your supervisor left the room. I recognize that some questions may seem repetitive, but the details are important to the investigation.”
The second approach seeks information that can be compared with other evidence. The first treats demeanor as evidence of deception and may cause the interviewee to become defensive without producing useful information.
Why This Matters for Bias Reduction
Traditional approaches to lie detection can reinforce unconscious bias. An investigator may misinterpret anxiety, cultural differences, disability-related behavior, language differences, or an unfamiliar communication style as evidence of dishonesty.
Focusing on the substance of an account, its context, and available corroborating evidence can help reduce reliance on appearance or demeanor. It also creates a clearer basis for explaining credibility findings if they are later reviewed in litigation, by an agency such as the U.S. Equal Employment Opportunity Commission, or by company decision-makers.
Demeanor should not be ignored entirely, but it should be treated cautiously and should not serve as a substitute for examining the evidence.
An Action Plan for Investigators
This week: During your next interview, avoid treating body language as proof of truthfulness or deception. Focus instead on the coherence, plausibility, specificity, and verifiability of the account.
This month: Practice diagnostic questioning in lower-stakes conversations. Use follow-up questions that seek additional detail and information that can be independently checked.
This quarter: Review your investigation protocols. Consider whether credibility assessments rely primarily on demeanor or instead consider communication content, context, corroboration, documentary evidence, and other available information.
Credibility Assessment Is About Information, Not Mind Reading
Effective investigators are not human lie detectors. They are careful interviewers who know how to elicit information and assess it against the available evidence.
The goal is not to identify a person as a liar based on appearance or instinct. The goal is to develop a well-supported understanding of what happened, identify unresolved issues, and document the basis for the investigation’s factual findings.
Further Resources
- Duped: Truth-Default Theory and the Social Science of Lying and Deception by Timothy R. Levine
- Evidence-based interviewing and diagnostic questioning materials
- Training on narrative analysis, question formulation, and corroboration of witness accounts
