Through a Scanner Darkly: Functional Neuroimaging as Evidence of a Criminal Defendant’s Past Mental States

Teneille Brown & Emily Murphy

Posted in , , , ,

As with phrenology and the polygraph, society is again confronted with a device that the media claims to be capable of reading our minds. Functional magnetic resonance imaging (“fMRI”), along with other types of functional brain imaging technologies, is currently being introduced at various stages of criminal trials as evidence of a criminal defendant’s past mental state. This paper demonstrates that functional brain images should not currently be admitted into courts for this purpose. Within the framework of Federal Rule of Evidence 403 —a rule used to evaluate all types of evidence by weighing the evidence’s probative value against the danger that it may be unfairly prejudicial—we demonstrate that a clear understanding of fMRI methodology exposes brain images as minimally probative when introduced to prove a defendant’s past mental states, and almost certainly substantially more prejudicial than probative on balance. Careful and detailed explanation of the underlying science separates this Article from others that have tended to paint fMRI with the gloss of credibility and certainty for courtroom-relevant applications.

The arguments advanced in this Article are narrowly aimed at functional brain imaging evidence—both images and expert testimony—addressed to issues regarding criminal mens rea. In narrowing our argument in this way, we are explicitly not analyzing the probative value of structural images such as those produced by CT scans or regular MRIs. We are also not critiquing the use of medically-accepted brain scans introduced to prove relevant medical diagnoses (such as brain trauma) when the presence and extent of the medical diagnosis is a key issue of causation, such as in a tort claim. The link between the brain’s structural and functional integrity and its complex mental functions is at the crux of our critique; at present, this link is a matter of scientific research and speculation, rather than diagnostic certainty. Some of our arguments may generalize to other courtroom uses of functional brain imaging; certainly, we expect the scientific background to be useful in many cases that attempt to use functional neuroimaging to make claims about complex mental processes such as lying, experiencing pain, or recalling memories. However, as the fMRI image construction is incredibly complex and its admissibility is specific to the purpose for which it is being offered, we sought to restrict our analysis to the admissibility of fMRI when used by a criminal defendant to argue that he did not have the requisite intent to be held culpable for the charged offense.

Functional brain images are specialized scientific evidence, but they are also just evidence. In order to be admissible, images must be relevant, authenticated, and more probative than unfairly prejudicial; cannot be impermissible hearsay; and, if introduced by an expert, must satisfy the tests for expert scientific testimony in that jurisdiction. This Article focuses on the first three requirements, but does not by exclusion suggest that other rules pertaining to scientific evidence might be incapable of properly evaluating and excluding fMRI images. In particular, the analysis of the minimal probative value of fMRI images could similarly be argued in a Daubert or Frye analysis about the data’s reliability, validity, and general acceptance in the relevant scientific field.

The weak probative value of fMRI evidence with respect to past mental states is best made clear by understanding the process through which an fMRI image is constructed. While a brief summary is provided below, this complex process is covered in greater detail in our full Article.

fMRI is a relatively safe and non-invasive medical technique that indirectly measures the brain’s activity. The emphasis on indirectly cannot be stressed enough. The brain is made up of millions of electrically-active cells called neurons, which use tiny electrical pulses and chemicals to communicate with each other via networks that enable sensation, perception, movement, cognition, emotions, and consciousness. fMRI does not directly measure this neuronal activity, but rather uses a physiological correlate to approximate areas of activity. Because neurons do not have internal reserves of energy, when they send electrical pulses in response to some activity, oxygen-carrying blood must be transported to the neurons to replenish the cell’s metabolic reserves. This is called the “hemodynamic response.” Blood that is carrying oxygen behaves differently in magnetic fields than deoxygenated blood. The difference in the magnetic properties of oxygenated blood allows fMRI to detect changes in blood flow related to activity. This is called the Blood Oxygen Level Dependent (BOLD) response. In simple terms, when a region of the brain is “activated” in response to a perception or to enable a behavior, that region receives more oxygenated blood. Because oxygenated blood behaves differently in a magnetic field, the large magnet in the fMRI device can measure this influx. If the local oxygen use is more than adequately supplied by the influx of blood, then a positive BOLD response will be observed. If the local demand for oxygen exceeds that provided by the regional blood flow, then a negative BOLD response will be observed. The change in the blood oxygenation level in a spatial volume called a voxel (which can be conceptualized as a three-dimensional pixel), does not directly capture the activity of neurons; thus, fMRI does not yet provide detailed physiological information about specific neural mechanisms underlying any specific mental state. Each new finding about the nature of the BOLD response suggests that our understanding of its mechanistic basis and correlation to neuronal activity is just scratching the surface. Understanding this phenomenon at the level of the neuron is critical to understanding if it is capturing little, some, or most of the brain’s actual neuronal activity in the creation of a mental event. In turn, this knowledge is necessary to bridging the gap between a particular cognition or behavior and the neural mechanism underlying it.

Without more confirmed data on what the BOLD response is capturing, inferences about individuals’ past mental states that are drawn from fMRI evidence will be quite useless in a legal setting that demands a strong, or at least probable, causal story. The task-dependency of fMRI may be obscure to the fact finder and, in any event, should severely limit the probative value of the resulting image. This is particularly true given the lack of ecologically valid behavioral tasks we have at our disposal that subjects could conduct in an fMRI scanner. What, exactly, would an assessment of an intentional, knowing, reckless or remorseful mental state be comprised of? Even well-validated behavioral tasks, such as those used to assess executive functions (e.g., judgment, decision making, attentional focus and switching), do not directly map onto Model Penal Code psychological mental state constructs such as purpose, knowledge, recklessness, and negligence. At best, deficiencies in task performance and correlated abnormalities in brain activity are several inferential steps away from a legal lack of intent or lack of capacity—steps so large they are almost certainly fatal to a legally relevant causal story.

In evaluating the weak probative value of fMRI addressed to past mental states, this Article analyzes cross-disciplinary expertise from neuroscience, philosophy, and legal literature. Such concepts as neuroessentialism, ecological validity, the “psycho-legal error,” the “epistemic mismatch” between what the images can say and what they appear to say, the appearance of objectivity and reductionism, and the allure of the colorful image of the brain itself, are explained and evaluated in order to illustrate why the probative value of fMRI evidence is weaker than it might appear at first blush. Further technical details about how fMRI images are constructed expose the many points at which an opportunistic expert could “dial a defect.” Statistical analysis is a critical and often obscure part of all elements of fMRI, from behavioral assessment to filtering functions that attempt to find meaningful “signals” against biological background “noise.” Examples of “dialing a defect” include fudging the statistical threshold, sampling various software packages until you find one that reveals a defect, manipulating base rates, or drawing inferences based on a peculiar reference class of normals. A recent satirical publication showed how researchers were able to demonstrate neural activation in a dead fish’s brain just by throwing enough analytical darts and eventually having something stick—a result obviously attributable to statistical chance. This is possible because of the high level of noise in the blood flow signal on which fMRI relies, and we are also careful to point out this possibility of finding chance activation in a forensic context.

An fMRI image is not a snapshot of an individual’s brain. While thoughtful authentication of the image itself might serve to counteract this “epistemic mismatch,” this is only one hurdle that the evidence needs to overcome. One of these hurdles is a problem of time travel. fMRI can only measure a defendant’s present mental state, which can be linked to his past (and criminally relevant) mental state only by way of inference and supposition. Others have argued that adversarial cross-examination is a powerful tool that can counteract the potential for prejudice and deal with such inferential leaps. But without any professional standards related to the software package, template brains, behavioral tasks or statistical analysis that is appropriate to use in any given setting, such courtroom conversation would waste the court’s time and mislead the jury. While fMRI methodologies are constantly evolving, the purpose of this Article is to provide thoughtful context on how the methodologies would need to improve in order to have the kind of probative value that would warrant their admission as evidence in a court of law. As the methodologies employing fMRI may one day improve such that its probative value is no longer eclipsed by the potential for unfair prejudice, we offer at the end of our Article a non-exhaustive checklist that judges or counsel can use to authenticate the functional brain images and assess the weight these images should be accorded by fact finders.

At this stage, we can only speak to the potential for unfair prejudice, as data is currently being gathered as to just how prejudicial functional brain images are on mock jurors. Even so, a few preliminary findings suggest that the mere presentation of brain images can make a bad argument appear more sensible. We address other forms of unfair prejudice as well, such as fMRI’s potential to mislead jurors, be cumulative of other cheaper psychological evidence, and waste the court’s resources and time.

The familiar story is one of weak circumstantial evidence, unsympathetic defendants, and impressive scientific reports and images. When you mix these three together, you have a prescription for potential injustice that is so compelling it could sway even the most skeptical jurors and convince them that the elements of a weak case are proved beyond a reasonable doubt. If, on the other hand, it is the defendant that catches the court’s sympathies, then the science may swing in the opposite direction and fMRI could be used to make a weak defense appear stronger. This tale has been told before with phrenology, the polygraph, and countless other forensic technologies that have since been discredited. Improper reliance on each of these untested and unreliable technologies has led to unjust outcomes.

Science can appear to be beyond the reach of human distortion. As a result, the more the scientific evidence relies on complex technologies like computers or imaging devices, the greater the risk that it may be endowed with powers to solve hard legal problems. Litigants have used this fact to their advantage, and throughout history have stretched scientific findings to fit them to pre-existing legal conclusions. This may be what is happening with fMRI. The device is not yet capable of capturing past mental states, but because the criminal law may sometimes be desperate to prove the unprovable, we anticipate an increase in proffered evidence and testimony based on this new technology.


Thanks to the MacArthur Funded Law and Neuroscience Project, Jeff Cooper, David Faigman, Jaime King, Michael Saks, Hank Greely, Stephen Morse, Kathryn Abrams, Walter Sinnott-Armstrong, Ryan Calo, R. Duncan Luce, William Uttal, and Sean Mackey for their thoughtful comments on earlier drafts of this paper. Special thanks to Nita Farahany and Adina Roskies for their in-depth feedback and guidance.

Copyright © 2010 Stanford Law Review.

Teneille R. Brown, JD, is an Associate Professor of Law at the University of Utah, S.J. Quinney College of Law. While writing this Article she was a fellow with the Stanford Center for Biomedical Ethics, a fellow at the Stanford Law School Center for Law & the Biosciences, and a research fellow with the MacArthur Foundation Law and Neuroscience Project.  Emily R. Murphy, PhD, is a JD candidate in the Stanford Law School class of 2012.  While writing this article she was a fellow in the Stanford Law School Center for Law and Biosciences and research fellow on the MacArthur Foundation Law and Neuroscience Project.

This Legal Workshop Editorial is based on the following Law Review Article: Teneille Brown & Emily Murphy, Through a Scanner Darkly: Functional Neuroimaging as Evidence of a Criminal Defendant’s Past Mental States, 62 STAN. L. REV. 1119 (2010).

Post a Comment (all fields are required)

You must be logged in to post a comment.