Deception: The Quest for DetectionSeptember 2010
Topics: Information Privacy, Social Behavior, Law Enforcement, Criminology, Prevent Terrorism
The Truth About Lies
Our world is filled with deception. When we hit the snooze button with the certainty that we can still arrive at work on time, we are engaging in self-deception. If we fail to shred our unwanted financial documents, criminals may use the information to produce false identification. Many of the advertisements to which we are exposed are misleading or simply false. And when Fido rouses you from your comfy chair by whining at the door but then sits there himself after you've gotten up to let him out, he's once again demonstrated that animals can deceive too. People (and sometimes animals) deceive others to mask their true intentions, avoid the consequences of their actions, and manipulate the beliefs, decisions, and actions of others.
Deception is broadly defined as the manipulation of appearances to convey a false reality and affect the beliefs of others. There are multiple forms of deception, and when one considers the variety of deceptions and their contexts, it becomes difficult to find common features that characterize them all. Although deception is considered immoral by most cultures, some forms like "social lies" and "white lies" are sanctionedor at least toleratedby social customs. Telling jokes, teasing, and fibbing are other examples of untruths that carry little or no social disapproval.
Failure Is Not an Option
Given the prevalence of deception and the evolutionary advantages of being good at both deceiving and recognizing deception, why are humans so bad at it? We tend to underestimate our ability to lie yet overestimate our ability to detect dishonesty. In a recent study, subjects immersed their arm in painfully cold ice water or faked discomfort when their arm was in warm water. Asked which subjects were really experiencing pain, observers were about as likely to pick the fakers. Surprisingly, even police officers, customs officials, polygraphers, and others whose jobs involve detecting deception perform no better than laypersons. The need to detect deception coupled with our inability to reliably do so have driven a century-long quest (since 1895 when Cesare Lombroso measured changes in blood flow during interrogation) to develop technological solutions.
The polygraph (which measures changes in psychophysiological responses including heart rate, blood pressure, respiration rate, and galvanic skin response) is probably the most well-known device for detecting deception. Although it is often referred to as a "lie detector," it really only measures stress and anxiety, not truth or deception. Voice stress analysis, another well-known technique, emerged because the Army wanted a remote alternative to the polygraph. However, subsequent evaluation has demonstrated that it is ineffective for detecting deception.
Several additional devices have been developed that remotely measure some of the same signals as the polygraph. The radar vital signs monitor measures motion processes such as heartbeat, respiration, and eye blinks. Another technique, laser Doppler vibrometry, senses minute vibrations from the body surface, including motions that are associated with respiratory and cardiovascular activity, tremor, muscle activity, and gross body movement.
MITRE is investigating the chemical composition of human sweat to determine whether deceptive individuals emit specific volatile organic compounds. These and other physiological measures may indicate stress or anxiety, but thus far none have directly or uniquely been linked with deception.
Researchers at the U.S. Army Night Vision and Electronic Sensors Directorate (NVESD) are investigating whether a new technology based on thermal imaging may be useful for remotely, and potentially covertly, identifying stress and anxiety without the need for electrodes. MITRE is supporting NVESD by studying the physiological and neurological signatures of deception, emotion, and intent and by providing an operational perspective through interviews with government users of deception detection technologies.
Behavioral approaches have also been used to detect deception, including neurolinguistic programming (NLP), the Facial Action Coding System (FACS), and blob analysis. NLP, a model of communication patterns for psychotherapy, and FACS, which measures facial muscle movements, were repurposed for deception detection. Blob analysis is an automated approach for analyzing nonverbal behavior. The potential contributions of these approaches to detect deception remain uncertain.
More recently, there have been attempts to use noninvasive brain imaging techniques like electroencephalography, functional magnetic resonance imaging, positron emission tomography, magnetoencephalography, and near-infrared spectroscopy to detect deception. These techniques identify brain regions that are most active based on electrical activity measured along the scalp or on cerebral blood flow associated with increased metabolic activity. The hope has been that identifying neural signatures associated with deception would provide a more direct link to mental activity that would avoid the issues that have plagued polygraphy and other psychophysiological approaches.
Using these techniques, laboratory studies comparing subjects who were deceptive and those who were truthful have identified differences in the brain activity patterns of the frontal and limbic regions associated with planning, memory, decision making, executive control, and emotion. However, neither the observed activity patterns nor the activated regions have been uniquely or consistently associated with deception. Despite the commercial promotion of functional magnetic resonance imaging for deception detection, there are no known neural signatures that reliably indicate deception.
Privacy and ethical issues are important aspects of deception detection. Neuroethics and neurolaw are emerging areas that are beginning to address these issues for the brain-based approaches. Detecting and applying deception are important for many of our sponsors. MITRE's role as an independent advisor is critical in evaluating the science underlying the many existing technologies and techniques as well as those to come. Equally critical is our role to work in the public interest by ensuring that our sponsors are fully aware of the associated privacy and ethical concerns.
by Hal Greenwald and Kristin Heckman