Functional Magnetic Resonance Detection of Deception: Great as Fundamental Research, Inadequate as Substantive Evidence - Charles Adelsheim

• Jurisdiction: United States,Federal
• Publication year: 2011
• Citation: Vol. 62 No. 3

Functional Magnetic Resonance Detection of Deception: Great as Fundamental Research, Inadequate as Substantive Evidence

by Charles Adelsheim*

I. Introduction

Essential to the law's pursuit of truth, justice, and the efficient resolution of conflict is assessing the veracity of statements made by individuals both in and out of court. In this judicial context, untruthful statements can be, and no doubt are, made regularly by plaintiffs, defendants, and other witnesses. Humans are generally very skilled at deceiving others, yet they are poor at detecting deception. Because of this disparity, there is a strong demand for reliable scientific techniques to detect deception. The most popular technique is currently the polygraph examination. However, polygraph-based evidence is inadmissible as substantive evidence in nearly all jurisdictions. There are a number of techniques being developed with the hope of filling this unmet demand, one ofwhich is the use offunctional magnetic resonance imaging (fMRI) to detect deception.

While fMRI detection of deception shows promise, and while excellent fundamental research is being conducted, fMRI is not yet ready for deployment in the courtroom. To explain this conclusion, this Article consists of four sections: (1) a discussion of the phenomena of deception and the difficulties attendant to detecting deception; (2) an accessible

* Assistant Director, Center for Science and Innovation Studies, University of California-Davis; Patent Attorney. University of Washington (B.S., with honors, 2004); University of California-Davis School of Law (J.D., 2010).

886 MERCER LAW REVIEW [Vol. 62

primer on MRI, fMRI, and BOLD fMRI technology; (3) a review and analysis of the existent research studies of fMRI detection of deception; and (4) an analysis of why, given the research to date, fMRI detection of deception should not be admitted as substantive evidence in a court of law.

II. The Phenomena of Deception

The ability to deceive one's peers seems to have developed as mankind evolved. There are advantages to being capable of deception, such as the ability to hide and monopolize scarce resources, or in the modern context, the ability to steal and get away with it.¹ Supporting this evolutionary perspective, there appears to be a correlation between the ability to deceive and brain size,² and "[d]eception has been observed in all primate groups."³ Humans certainly excel at deceiving each other. However, there is a marked disparity between our ability to deceive and our ability to detect deceit.⁴ There are abundant examples of this disparity, such as the ability of undercover officers to successfully infiltrate criminal organizations or the ability of adults to lie successfully in paternity cases (especially before the advent of DNA testing).⁵

Because of this disparity between our ability to deceive and our relative inability to detect deceit, mankind has tried to devise "scientific" means to detect deception for at least the last 100 years.⁶ These attempts started with such primitive means as torture and trial by water (for example, witch hunting) and developed towards more scientific techniques. One early example of lie detection technology consisted of a balance table on which a suspect was carefully balanced and then interrogated. The underlying theory was that because lying is more difficult than telling the truth, when a person was lying, more

1. See generally Sean A. Spence & Catherine J. Kaylor-Hughes, Looking For Truth and Finding Lies: The Prospects for a Nascent Neuroimaging of Deception, 14 Neurocase 68, 69 (2008).

2. See Richard W. Byrne & Nadia Corp, Neocortex Size Predicts Deception Rate in Primates, 271 Proc. Royal SocY London B. 1693, 1695 (2004).

3. Jonathan T. Rowell et al., Why Animals Lie: How Dishonesty and BeliefCan Coexist in a Signaling System, 168 Am. Naturalist 180, 181 (2006).

4. See Paul Ekman & Maureen O'Sullivan, Who Can Catch a Liar?, 46 Am. Psychol. 913, 913 (1991); see also Nancy L. Etcoff et al., Lie Detection and Language Comprehension, 405 Nature 139, 139 (2000); Sean A. Spence, The Deceptive Brain, 97 J. Royal Soc'y Med. 6, 6 (2004).

5. See Mark A. Bellis et al., Measuring Paternal Discrepancy and Its Public Health Consequences, 59 J. Epidemial Cmty. Health 749, 751 (2005).

6. See generally Ken Alder, The Lie Detectors: The History of an American

Obsession (2007).

2011] FUNCTIONAL MAGNETIC RESONANCE 887

blood would flow to his head, and the table would consequently tip towards his head. Another early "scientific" means of detecting deception was the rice test employed in China. According to this technique, a suspect would hold grains of rice in his mouth while being interrogated. The underlying theory here was that telling lies would impede saliva production; thus, if the questioned person was unable to wet the rice grains he was lying. Even the best of modern-day lie detection techniques-for example, the polygraph machine-are far from perfect, and the judicial system is aware of this fact.⁷ This is why polygraph examinations are not admissible as substantive evidence in most courtrooms.⁸

The primary method used to determine the veracity of statements in court is to have the fact finder observe the demeanor of the witness. However, studies have shown that even the most highly trained individuals exhibit only slightly better than chance rates of detecting deception based solely on observing demeanor. Even if one were to accept demeanor as a reliable indicator oftruthfulness, the courtroom is a highly formalized atmosphere to which most laypersons are unaccustomed. The courtroom can feel both strange and intimidating; thus, demeanor may be an especially weak indicator of truthfulness in the courtroom. In this light, it is understandable that there is a strong demand for a reliable way to differentiate between truthful and deceptive statements.

The proponents of fMRI detection of deception claim that it offers a reliable way to discriminate between truthful and deceptive statements. There are even two companies currently offering fMRI detection on a commercial basis, No Lie MRI, Inc. and the CEPHOS Corporation.⁹ Certainly this and other neuroscience techniques hold great potential, and support for the proposition that neuroscience will play a role in the courtroom of the future that the upcoming Federal Judicial Center's Reference Manual on Scientific Evidence which will include a chapter on neuroscience.¹⁰ However, thus far no courts have admitted fMRI deception detection as substantive evidence. Courts are justified in taking this position, as will become clear by the close of this Article.

7. See, e.g., United States v. Scheffer, 523 U.S. 303, 309-10 (1998).

8. Id. at 311.

9. For more information on No Lie MRI, Inc., see http://www.noliemri.com, and CEPHOS Corporation, see http://www.cephoscorp.com.

10. For more information regarding the third edition of the Reference Manual on Scientific Evidence, see http://sites.nationalacademies.org/pga/stl/development_manual /index.htm.

888 MERCER LAW REVIEW [Vol. 62

III. Physics and Science of MRI and FMRI

To effectively evaluate the strengths and weaknesses of fMRI lie detection, it is essential to have at least a rudimentary understanding of the science utilized in this technique. Since most lawyers, judges, and other legal practitioners do not have a strong science background, the discussion presented here will not dwell on minutiae and will instead strive to provide a comprehensible overview tailored to actual use of science in a courtroom.¹¹

To understand fMRI, one must first understand "magnetic resonance imaging" (MRI). An explanation of MRI starts with a discussion of the fundamental building blocks of our physical universe. Everything that exists is made ofatoms like carbon, hydrogen, and oxygen. These atoms, in turn, are made of electrons, which orbit a nucleus comprised of protons and neutrons. Essentially, an MRI scanner detects small differences in the density and behavior of certain protons.

This detection is possible because protons possess both angular momentum and a magnetic moment (also known as a dipole moment).¹² A simple yet behaviorally accurate analogy is that a proton can be thought of as a bar magnet spinning about its long axis or as a gyroscope with a bar magnet running along its axis of rotation. When there are an even number of protons in a nucleus, they will align in an anti-parallel manner, effectively negating their magnetic moments. When there are an odd number of protons in a nucleus, however, the nucleus will have a net magnetic moment and angular momentum. These nuclei with an odd number of protons are what an MRI machine can detect. Conveniently, hydrogen has only a single proton and also happens to be the most abundant atom in a human body. This abundance is due to the fact that there are two hydrogen atoms in every molecule of water, and a human body is composed of roughly 70% water.

Generally, the dipole moments of a water-containing body-such as a human brain or knee-are aligned randomly, which is to say that there is no net magnetic moment. It is impossible to extract useful information from randomly orientated magnetic dipoles; this is why an MRI machine creates and manipulates a strong primary magnetic field. In modern MRI machines, this primary field is 1.5 to 4 Tesla, or 30,000 to 80,000 times stronger than the Earth's magnetic field. When a subject

11. For a more complete discussion of the science and technology of fMRI, see Richard B. Buxton, Introduction to Functional Magnetic Resonance Imaging: Principles & Techniques (2002), and Scott A. Huettel et al., Functional Magnetic Resonance Imaging (2004).

12. Huettel et al., supra note 11, at 49.

2011] FUNCTIONAL MAGNETIC RESONANCE 889

or sample is placed into this strong primary magnetic field, the individual dipole moments of the hydrogen protons tend to align with the primary magnetic field, creating a net magnetic moment...