A clue to deception?

Fraud examiners know that a suspect's body language can reveal more than that subject realizes. However, without empirical evidence, interviewers are at risk of assuming guilt or innocence. This column is adapted from a paper written by recent graduates or current students in Carlow University's Master of Science in Fraud and Forensics program for a class taught by Associate Professor Enrique Mu, Ph.D. — ed.

In 1989, a woman jogging in New York's Central Park was brutally beaten and raped. In what became known as the Central Park Jogger case, aggressive investigators coerced five juvenile men into confessing. Police investigators — faced with a lack of forensic evidence — believed that some of the men's nonverbal gestures were indications of guilt. However, the confessions were false, and the investigators' assumptions were wrong. Forensic author and former FBI agent, Joe Navarro, wrote, "Officers mistook nonverbals of stress for deception and pressured the innocent into [making] confessions." (See the book, "What Every BODY is Saying: An Ex-FBI Agent's Guide to Speed-Reading People," by Joe Navarro and Marvin Karlins, page 206.)

This case exemplifies how relying on non-scientific methodologies, including observation and nonverbal indicators as sole measures of determining deception, can be problematic. As poignantly illustrated in the book, "Convicting the Innocent: Where Criminal Prosecutions Go Wrong," by Brandon L. Garrett, false confessions are present in more than 16 percent of wrongful convictions (page 8).

Nonverbal communication is the outward expression of our inner feelings and emotions. (See Fraud Basics.) During the interview process of a fraud examination, body language — also known as nonverbal cues — can be a critical tool in determining deceptive actions. Generally speaking, fraud examiners are able to look at a cluster of signals, not just one, to alert them of possible false statements or omissions.

Commonly accepted nonverbal cues include hand gestures, face touching, self-rubbing, scratching, memory reanimation, tactile gestures affected by culture, types of smiles, cutaneous responses and shifting eyes/non-eye contact (which is the most prevalent cue monitored during an interview). Other responses include fast blinking, yawning and leaning backward and/or forward during questioning.

While all of these tools are extremely useful and have been used successfully in obtaining confessions and criminal convictions, there's still an argument against some of the common assumptions about nonverbal communications as to whether they're actually grounded in science; thus relying on the interview process alone is not enough to secure an evidence-based conviction.

Examples of assumptions about the use of eyes as a detection tool include, "If a person does not make eye contact with you, then he must be lying," or "Our eyes are windows to our soul," both of which are actually examples of common misconceptions in this field. Unfortunately, direct eye contact is the easiest behavior to spot but also one of the easiest to misinterpret. (From a 2011 course, "Communication Analysis," sponsored by the National Drug Intelligence Center, taught by Joelle Fisher.) The evaluation of how nonverbal communication can be used to determine deceptive actions should be further studied to develop empirical research.

NONVERBAL CUES

Researchers have conducted numerous international scientific studies related to the use of nonverbal indicators. One notable researcher on this subject matter is Aldert Vrij, of the University of Portsmouth (U.K.). Several of Vrij's studies affirmed the observation of nonverbal cues as a valid measure of deception. More than 40 studies conducted by Vrij found that the most reliable nonverbal behavioral indicators of deception included the use of illustrators, arm, hand/finger and foot/leg movements. [See Detecting deceit via analysis of verbal and nonverbal behavior, by A. Vrij, K. Edward, K. Roberts, K. and R. Bull in Journal of Nonverbal Behavior, 24(4), pages 239-264 as cited in the article, "Psychopathy and nonverbal indicators of deception in offenders," by J. Klaver, Z. Lee and S.D. Hart, from Law and Human Behavior, Vol. 31(4), August 2007, pages 337-351.]

Research does caution us about relying solely on nonverbal cues to detect deception. This could be for many different reasons: not knowing where to look, not knowing what to look for, or possessing false beliefs or preconceived notions about those behaviors that might be clues to deception — whether valid or not.

One of the most well-documented claims in the literature is that humans are poor at detecting deception. Scientists have conducted numerous studies in an effort to simulate emotions and strategies of a real-life scenario in which a suspect is being interviewed. The fact is, it's difficult to replicate fear, anxiety and guilt in a controlled environment with scripted answers. This raises concerns about the reliability, repeatability and the influence of bias present in these studies.

PROBING FOR THE TRUTH

Probing is a lie-detection strategy designed to monitor a suspect's behaviors during repeated interrogations. It involves an immediate and sometimes unexpected test of a suspect's behavioral responses to repetitive questions during an interview. (See "Repeated interrogations: Verbal and non-verbal cues to deception," by P.A. Granhag and L.A. Stromwall in Applied Cognitive Psychology, Vol. 16, p. 244.) A major misconception people may have is that as an interviewer asks repetitive questions, liars become agitated, and they exaggerate their body language from frustration. However, research suggests the opposite might be true. As the probing strategy continues, liars suppress and control the nonverbal cues that could indicate deceit.

Granhag and Stromwell studied the effects of probing on suspects. In their research study, results showed that truth-tellers averted their gaze more, expressed more smiles and laughter, and carried out more self-manipulations; this behavior was essentially the same throughout the sessions for the truth-tellers. In contrast, as the number of interviews increased for liars, they suppressed these same nonverbal indicators. As expected, the liars appeared tenser, had to think harder and put forth more effort to control their overall emotions, consistent with researchers' predictions (pages 249-252).

Paul Ekman has done extensive research on detection and the role of the observer. He purports that detection of deception requires a mastery of specific skills.

In the article "Detecting Deception" in the July 2001 FBI Law Enforcement Bulletin, authors Joe Navarro and John R. Schafer suggest, "Investigators' abilities to detect deceptive behavior depend largely on their ability to observe, catalogue, and differentiate human behavior."

Fraud examiners also can combine technological tools with interview techniques to assist with the detection of deception.

FUNCTIONAL MAGNETIC RESONANCE IMAGING

Functional Magnetic Resonance Imaging (fMRI) has promising potential to help ferret out liars. The use of this technology is proving important in interviews, according to research by F. Andrew Kozel and his colleagues. Their studies showed that when subjects deceived examiners, certain regions of their brains activated. The researchers were able to prove this assertion by imaging the activity of the prefrontal region and the amygdala of the brain, where detection response was most accurately measured. For the first time, they were able to show that fMRI has the potential to reliably identify brain patterns for deception. (See Detecting Deception Using Functional Magnetic Resonance Imaging, by F. Andrew Kozel, et al, Biological Psychiatry, Volume 58, Issue 8, pages. 605–613, Oct. 15, 2005.)

In another study, researchers noted: "Our results provide clear evidence of functionally dissociable roles of the prefrontal sub regions and amygdala for human deception." Their analysis revealed that the activity of the right anterior prefrontal cortex showed the effects of deception. [See N. Abe, et al., Deceiving others: distinct neural responses of the prefrontal cortex and amygdala in simple fabrication and deception with social interactions, Journal of Cognitive Neuroscience, 19(2), pp. 287-295.]

Studies indicate that subjects who participate in the fMRI must be cooperative for accuracy. However, this technique has come under criticism as potentially invasive and creating a misperception of mind reading. Utilizing a technology that accesses brain processes, which haven't been open to prior scrutiny, poses challenges to privacy conceptions. [See the June 2007 article "Emerging Neurotechnologies for Lie Detection and the Fifth Amendment" by S. Stoller and P. Wolpe in the American Journal of Law & Medicine, Vol. 33 (Issue 2/3), p. 374.] Additionally, researchers indicate that more studies need to be conducted relative to physiological changes to heighten accuracy. (See Detecting Deception Using Functional Magnetic Resonance Imaging, by F. Andrew Kozel, et al, Biological Psychiatry, Volume 58, Issue 8, pages 605–613, Oct. 15, 2005.)

The observation of nonverbal cues can be highly subjective. The use of fMRI can control for this variable through scientific analysis of physiologic and neuronal activity because it can be objectively visualized.

The use of fMRI for lie detection is improving. Work in this field promises standardization of the procedures to utilize this technology to impact lie detection in the future. Lie detectors using fMRI are being refined; the ability of fMRI to image the brain more precisely and directly offers a strong alternative that outstrips existing lie detectors. (See the 2007 article, The Brain and Behavior: Limitations in the Legal Use of Functional Magnetic Resonance Imaging, by Laurence R. Tancredi and Jonathan D. Brodie, American Journal of Law and Medicine, 33, pages 271-294.)

CONCLUSIONS BASED ON FACT, NOT GUT

The practice of observation could be, at best, labeled non-scientific and inductive because it only produces observations — not conclusions. Experienced fraud examiners have made repeated observations relative to the body language nuances of white-collar suspects. These claims can't inform or affect policy development without the necessary empirical scientific validation. An understanding of the nonverbal behavior — in conjunction with the testing of a validated tool such as fMRI — might prove to be more useful to the fraud examiner in determining deception. Practitioners should recognize that most technologies have limitations and stay abreast of the latest research delving into lie detection.

Denise Lawhorn, CFE, is a project manager at the Virginia Department of Forestry.

Colin Pingree, CFE, is an auditor with the U.S. Department of Defense, Office of Inspector General.

Mark Malinowski, CFE, CAMS, is a detective in the Ocean County Prosecutor's Office.

Inger Jensen is an operations analyst at PNC in Pittsburgh, Pennsylvania.

Resource: See the ACFE's 2014 Fraud Examiners Manual: "Nonverbal Behavior/Body Language," 3.314; "Proxemics," 3.327; "Chronemics," 3.28; "Kinetics," 3.328; "Paralinguistics," 3.329; and "Physiology of Deception," 3.348.