Getting to the truth

We are now at a tipping point in the field of lie and truth detection. Peer-reviewed research and high-stake applications are showing that we need two main skillsets to perform better than the average human and the polygraph: 1) recognizing deceptive behavior via multi-channel analysis and 2) using unexpected prompts in interviews to elicit truth.

In this column, I’ll steer you away from the myths and toward the science and hard evidence in recognizing truth, reading emotions and detecting deception during interviews. I’ll avoid snake-oil claims and approaches that only use one or two elements to detect deception, such as “cognitive load,” facial expressions, body language, verbal content, voice and reliance on biometrics, such as polygraphs.

Investigators who depend on just a few of these approaches might yield risky and unreliable results (Porter and ten Brinke, 2010: 69). I encourage you to employ a multi-channel, corroborative approach to help you make reliable decisions (Pearse and Lansley, 2010).

In the “Research Findings” column of the March/April 2019 issue of Fraud Magazine, Vincent Denault, CFE, and Louise Marie Jupe reminded us that that distinguishing “unsubstantiated claims from those that are evidence-based … isn’t always an easy task.” I agree with them.

I’d go a step further and argue that the evidence from some peer-reviewed, deception-detection research is based on dubious experiments involving students completing exercises with low-stake threats.

Those who are interested in discovering high-stakes lies note that many research experiments’ deceptions are trivial (Ekman, 2009). Lies are generally labeled “high stake” when the subject could experience large positive consequences of getting away with the lie (such as money, power, freedom) or large negative consequences of getting caught (death, incarceration, money loss).

And those who are interested in deceptive interactions tend to denounce research outcomes about lies, which result from an artificial, experimental context (Buller and Burgoon, 1996).

Our clients include company owners, bank officials, due-diligence investigators (often engaged by investors/brokers), military organizations, counterintelligence and anti-terrorism agents, and law enforcement officers. The stakes can be life-or-death situations, national security breaches and millions of dollars. We dare not train these professionals to judge lies or truth in their contexts using research findings in which stakes are low — peer reviewed or not. The approaches we use work well when the stakes are high but are less reliable for low-stake contexts (Archer and Lansley).

Truth leakage

People often tell lies using words. However, it’s very difficult to get high-stakes lies past investigators who are properly trained to see and hear liars because all of us leak the truth.

What we know from research (Zuckerman et al., 1981), as Denault and Jupe pointed out in the March/April issue of Fraud Magazine, is that lies will most likely:

• Be linked with emotions.
• Require greater cognitive effort than truthful messages.
• Be associated with arousal.
• Prompt liars to over-control their behaviors.

Power of emotions

Core emotions such as fear, anger, sadness, happiness, disgust, contempt and surprise can happen in us within 400 to 500 milliseconds — less than half a second — of something triggering them. An investigator’s incisive question or the sight of incriminating evidence placed on the desk during an interview could trigger fear in a subject. An interviewer’s oppressive, false accusation could trigger a subject’s anger. The resulting impulse (the first quick physiological change that we can see and hear when a subject experiences an emotion) linked to each of these emotions sets off its own orchestrated, synchronized, simultaneous array of physiological changes across our body in:

• Heart rate.
• Blood pressure.
• Breath rate.
• Voice (e.g., volume, pitch and speed of speech).
• Blood circulation and temperature changes in specific areas of the body.
• Muscular/body tension, freezing and movements.
• Digestion system (de)activation.
• Facial expressions.
• Eye movement, closure, blinking and pupil size.
• Perspiration.

The fact that neuroscientists have identified that impulses occur within about half a second of the triggers — even before the conscious mind is aware that it’s in the grip of an emotion (LeDoux, 1998) — is a gift to fraud examiners who are interested in discerning the truth.

We can see and hear emotional leakage from humans without technological aids even if someone wants to mask or hide their emotions. Passive observation of the face and body has already been shown to be 900 percent more effective in detecting malfeasants than random sampling in U.S. airports (Department of Homeland Security, 2013). This report outlines the improvement in apprehending a range of criminals in airports using behavior detection, which resulted in “30 boarding denials, 79 investigations by law enforcement entities, and 183 arrests.”

First skillset: multi-channel analysis

Though we know that truth leaks through our emotions in interviews, we need to go further. Targeted multi-channel analysis, the first skillset, is the next important step because as Porter and ten Brinke note in Legal and Criminological Psychology, “While the presence of a single behavioural cue may not provide convincing evidence for deception, the co-occurrence of multiple cues from words, body, and face should provide the lie catcher with increased confidence that deception has occurred.” In looking to the future, here are two of Porter and ten Brinke’s suggestions:

• The lie catcher should direct their attention to prominent and ideally consistent changes (across multiple channels) from baseline behavior with a focus on the most reliable specific cues to deception.
• The lie detector should consider the use of strategic interviewing techniques such as manipulating a suspect’s cognitive load by asking unanticipated questions.

We picked up that challenge with our own research. The most useful cues for judging veracity are those signals that leak as a result of cognitive load (such as having to think hard to suppress the truth and simultaneously present a fabricated story) or emotional load (such as the fear of being caught in a lie).

I’ve already offered examples of physiological changes that result from emotions, but I can best present the observable indicators of both emotional and cognitive load as the six communication channels (see Figure 1 below):

• Verbal content (what we say or write).
• Interactional style (the way we say or write our words and phrases).
• Voice (the “music” of the voice including pitch, volume and tone).
• Face (expressions seen in bulges, creases and furrows resulting from movements of one or more of the 43 muscles of the face).
• Body (movements of everything else except what’s covered by the face).
• Psychophysiology (changes within the body that can sometimes show externally — even without biometric apparatus — such as perspiration and breath rate).

Figure 1: Six communication channels (source: “Getting to the Truth – A Practical, Scientific Approach to Behaviour Analysis for Professionals, Behaviour Analysis and Investigative Interviewing Book 1,” by Cliff Lansley, EIA Group Ltd., 2017)

Our research has identified 27 indicators from across these channels (Lansley, 2017), which we’ve labeled Points of Interest or “PIns” — but only when those indicators are inconsistent with the ABC’s:

• The account (i.e., the story they’re trying to convey).
• Their apparent/emerging baseline behavior.
• Context (the influence of the immediate micro setting, e.g. the interview and the wider macro context of culture, politics, events, etc.).

One PIn alone doesn’t mean we’ve found a lie. We ignore one and even two. However, we’re interested the moment we discover a cluster of three PIns. We define a cluster with the 3-2-7 Rule, which is “three or more PIns across two or more communication channels within seven seconds of a probe or question” (Archer and Lansley, 2015). This peer-reviewed article in Corpora shows how interview subjects’ true accounts have limited or no PIns, but subjects’ multiple clusters of PIns exposed false accounts — lies.

Second skillset: power of unexpected probe

The best weapon in a truth-catcher’s armory is surprise. We must pay especially close attention to a subject’s cross-channel behavior immediately after an unpredictable, surprise “stimulus” because we know that within seven seconds “we can reasonably conclude that the behavior is directly associated with the stimulus” (Houston et al. 2012: 30). Stimulus here refers to an interviewer’s unexpected question. The stimuli in fraud examinations could also include tactical use of evidence, such as showing objects or evidence. This can be effective later in an interview where it might contradict what a subject has earlier disclosed in order to analyze their reactions to the object or evidence.

Effective investigative questioning in law enforcement has tended towards a “funnel approach” from the open questions we might ask to get the full account or story from the subject before we explore and verify the subjects’ accounts (Pearse, 2009). (See Figure 2 below.)

Figure 2: Funnel approach of interviewing (source: “Getting to the Truth – A Practical, Scientific Approach to Behaviour Analysis for Professionals, Behaviour Analysis and Investigative Interviewing Book 1,” by Cliff Lansley, EIA Group Ltd., 2017)

Initially, the interviewer draws a full account from the subject. The session progresses, and the interviewer explores the wider context of the account, which might include asking questions that the subject doesn’t anticipate about the time, place and weather. It might also include questions about what happened before and after the episode the subject has volunteered.

Often a perpetrator rehearses the core of their alibi really well, though they might neglect wider contextual issues. A truth teller generally wouldn’t have any problems with such questions.

The interviewer can then move down the funnel to establish if the subject has any cognitive and emotional load that’s out of context by asking what the subject was thinking and feeling around the time of the incident. Many liars also fail to rehearse answers to this type of question. The interviewer can later revisit areas of the account in which they’ve seen or heard clusters of PIns by asking further questions to test the hypothesis that the subject’s deception has caused those PIns.

Sometimes, a casual conversation can be effective because it might minimize the subject’s anxiety level. The interviewer doesn’t ask questions but probes and makes statements that can generate high-quality data. This is often termed “elicitation.” I define elicitation as a process an interviewer uses to draw out information from people, during a communication with a purpose, often without them realizing the elicitor’s purpose for doing so.

Elicitation, conducted by a skilled collector, will appear to be normal social or professional conversation. A person might never realize they were the target of elicitation or that they provided meaningful information (FBI, accessed Feb. 21).

Deception detection takeaways

I counsel fraud examiners to learn these key points:

• Don’t hunt for a lie. The aim is about getting to the truth.
• No “Pinocchio’s Nose” — no single indicator of deception — exists.
• Use the 3-2-7 rule. Only be concerned if you notice three of the 27 (high-stake) research-validated indicators, across two communication channels, within seven seconds of a question.
• Take a hypothesis-driven approach. You might see or hear something, but you don’t know why.
• The hypothesis-driven approach requires good questions within a funnel framework.
• The best weapon for the investigator is surprise. Use elicitation and unexpected questions in a non-oppressive way.

Cliff A. Lansley, Ph.D., is a behavior analyst with the EIA Group in Manchester, U.K. He’s the author of “Getting to the Truth,” a practical approach to behavior analysis for professionals. Lansley is a frequent speaker at ACFE events. Contact him at cliff@eiagroup.co.uk.

Sources

Archer, D. and Lansley, C., “Public appeals, news interviews and crocodile tears: an argument for multi-channel analysis,” Corpora, 10(2), pages 231–258, 2015.

Buller, D. B. and Burgoon, J. K., “Interpersonal deception theory,” Communication theory, 6(3) pp. 203–242, 1996.

Department of Homeland Security, accessed on Feb. 8, 2019, 2013.

EIA Group, “White paper - Observe, target, engage, respond,” accessed Feb. 8, 2017.

Ekman, P., Telling lies: Clues to deceit in the marketplace, politics, and marriage, revised edition, WW Norton & Company, 2009.

Federal Bureau of Investigation (2019), Elicitation Techniques, accessed Feb. 21.

Houston, P., Floyd, M., Carnicero, S. and Tennant, D., 2012. Spy the lie: Former CIA officers teach you how to detect deception. Macmillan.

Lansley, C., “Getting to the Truth – A Practical, Scientific Approach to Behaviour Analysis for Professionals, Behaviour Analysis and Investigative Interviewing Book 1,” EIA Group Ltd., 2017.

LeDoux, J., The emotional brain: The mysterious underpinnings of emotional life, Simon and Schuster, 1998.

Pearse, J. “The investigation of terrorist offences in the United Kingdom: the context and climate for interviewing officers,” R. Bull, T. Valentine & T. Williamson, editors, “Handbook of psychology of investigative interviewing. Current developments and future directions,” pages 69–90, 2009.

Pearse, J. and Lansley, C., “Reading Others – A psychological model for making sense of truth and lies,” Training Journal, 10, 2010.

Porter, S. and ten Brinke, L., “The truth about lies: What works in detecting high-stakes deception?” Legal and criminological Psychology, 15(1), pages 57–75, 2010.

Zuckerman, M., DePaulo, B. M. and Rosenthal, R., “Verbal and Nonverbal Communication of Deception1,” in “Advances in experimental social psychology,” Elsevier, pages 1–59, 1981.