Nadia Eckardt – The University of Washington, School of Law
Nadia is a J.D. candidate at The University of Washington, School of Law. Nadia argues there is an uneven application of neuroscience and neurology expert witness testimony in civil versus criminal contexts, such that the legal system has become more accepting of neuroscience as justification to punish rather than mitigate. She notes that statistics and personal stories can lead a jury to such confirmation bias.
Neuroscience as a courtroom tool began with the infamous case, People v. Weinstein, 156 Misc. 2d 34, 591 N.Y.S.2d 715 (Sup. Ct. 1992). Weinstein argued that a benign cyst pressing on his frontal-temporal region of the brain made him not responsible for the murder of his wife. Since then, neuroscience (“NS”) has presented complex questions to the legal institution by affording the opportunity to observe changes in the brain. This includes the potential for proof of emotional injuries, differing mental capacities, lying, and more. NS is a highly specialized field, requiring expertise in statistics and interpretation of brain scans: it is therefore brought into the courtroom through expert testimony.
Currently, neuroimagery is complex and reflects a shallow understanding of the brain and its functions. Its use in the courtroom to mitigate culpability or increase damage awards has become more popular, and is predicted to increase in use. However, neuroimagery and the use of expert witnesses in explaining their images is in direct tension with the Federal Rules of Evidence § 704(b), which prohibits an expert from speaking directly to a condition of the defendant that constituted an element of the crime. This is in part because neuroimagery relies on correlative data which means any explanation by the expert is necessarily “speaking directly to” culpability in a causal way. Rule 704(b) is perhaps already being overtaken by neuroscience, but it further aggravates the very presentment of neuroimaging and jury heuristics, such as priming, and statistics.
The argument can be made that neuroscience in the civil context is accepted as more causal than in criminal defense. While the premise is largely the same, this neuroimage means that, civil plaintiffs are afforded greater inferences to be drawn than criminal defendants. Defendants who bring neuroscientific evidence to mitigate culpability are constrained by § 704. In comparison, plaintiffs who bring NS to show damages, for instance from a car crash, are not. Considering the heavy burden a criminal sentence lays upon a person, this uneven application of neuroscience in the civil and criminal context should be reconsidered.
The Rules of Evidence and Expert Witnesses
In allowing new neuroscience methods in the court, the trial court must weigh (i) whether the method proffered is generally accepted in the scientific field, and (ii) generally accepted as a diagnostic tool for the particular injury brought by the plaintiff or defendant. The relevance of the technique and its particular application must both apply to the case at hand to pass under Daubert. Daubert v. Merrell Dow Pharm., Inc., 509 U.S. 579, 113 S. Ct. 2786 (1993). Once the original hurdle of admissibility is overcome, NS must be interpreted by the expert witness and presented to the judge and jury. The Rules of Evidence §§ 702-705 apply; of special interest is § 704(b): “… a criminal case, an expert witness must not state an opinion about whether the defendant did or did not have a mental state or condition that constitutes an element of the crime charged or of a defense.” This rule is of particular import because of the emerging role NS is playing in criminal defense: that of reducing culpability due to brain defects. Because “condition” in the language of the statute may be broadly read to include brain abnormality, neuroscience goes to the heart of § 704.
This tension has applied to psychiatrists testifying as well, because “it would be all gobbledygook without the psychiatrist drawing a conclusion as to what he’s saying.” National Mental Health Association, “Myths & Realities,” Hearing Transcript of the National Commission on the Insanity Defense, 30 (1983). Section 704(b) of the Federal Rules of Evidence was enacted relatively recently, and the Senate Report stated that:
The purpose of this amendment is to eliminate the confusing spectacle of competing expert witnesses testifying to directly contradictory conclusions as to the ultimate legal issue to be found by the trier of fact. Under this proposal, expert psychiatric testimony would be limited to presenting and explaining their diagnoses, such as whether the defendant had a severe mental disease or defect and what the characteristics of such a disease or defect, if any, may have been.” S. Rep. No. 225, 98th Cong., 1st Sess. 271, 230 (1983) (emphasis added).
Though the Court has said that “Rule 704(b) forbids only “conclusions as to the ultimate legal issue to be found by the trier of fact,”” this has resulted in a fine line drawn between the aggregate result of a condition and the individual on trial. United States v. Edwards, 819 F.2d 262 (11th Cir. 1987).
A subtle twist to NS testimony in the courtroom is that neuroimaging scans are “the product of a complex set of techniques, subjective decisions…technical choices…, and informed interpretation.” Brigitte Vallabhajosula, Murder in the Courtroom: The Cognitive Neuroscience of Violence 242 (2015). The image shown to the jury silently speaks to the ‘mental state or condition’ because the choices underlying the image itself may either reinforce the expert’s bias, or intentionally highlight a feature that allows for a stronger inference to be made by the jury.
For instance, the court in Weinstein noted that his psychiatrist “would be permitted to testify concerning the quantitative results [of the PET] and to include them in explaining a diagnostic opinion” but would not be allowed to rely upon theories that had also not gained general acceptance to explain the causal link, such as violent tendencies resulting from cysts. But the line between the two is easily blurred: an opinion, even if not directly relying on a theory, can be persuasive when coming from someone who is called an expert.
Issues Using Aggregated Neuroscience Data
First and foremost is the variability of brain structures within the ‘normal’ population which does not compensate for ‘abnormalities.’ Vallabhajosula, at 241. The variety makes it difficult for an expert to speak to the characteristics of a ‘disease or defect’ without resorting to generalities in order to draw any conclusions from the brain scans. Secondly, the understanding of the causal mechanisms between what a brain scan demonstrates and the brain’s output is weakly understood. Id. at 242. Indeed, most of the data links are correlations between aggregate data and outcome measures. Id. at 241. It is “impossible to attribute a particular action to a single, specific, and consistent neurological cause.” Id.
An argument can also be made that juries are vulnerable to heuristics which a competent attorney or expert witness can exploit. One such heuristic (beyond priming) presents an opportunity to inflate neuroscience’s importance: statistics. Statistics are poorly understood by people. Not only is everyone more influenced by small samples sizes, (which lead to more extreme outcomes), but when presented with statistics and a ‘personal’ story, tend to be swayed by the story and take the statistics only as they confirm it (relying on confirmation bias). Erik Johnson, Summarizing Daniel Kahneman: Thinking fast and slow 21 (2011). This, coupled with overestimating the likelihood of rare events, means neuroscience presented in the courtroom not only primes the jury to think there is a causal link, but that the neuroimage shows something that is more likely than the statistics might suggest exists.
While this could be rebutted by noting the jury’s duty is to determine the state of mind of the defendant, irrespective of the methods used to reach it, the fact that neuroscience remains inherently correlative suggests that, on the whole, experts speaking to neuroscience will not be relying on diagnoses, but on opinion. This opinion will hold significant weight due to the impressionability of neuroimaging and statistical heuristics. Because of these reasons, neuroscientific testimony does speak to the state of mind of the defendant in a way that arguably contravenes § 704.
An important policy question to ponder is whether the legal system is, in fact, the correct forum for the intersection between the aggregate and the individual as science continues to improve. Before neuroscience, law was arguably the natural evolution of policing society at large, where nuances in the individual were easily acceded to general criminal policy. With the rise in questions of intent, individual culpability, brain-structure mitigating factors, dependencies, age, gender, etc., whether law will remain the correct mechanism is questionable. That is, as culpability and intent become more nuanced, so too must sentencing, perhaps rendering redundant criminal law.
Dichotomy of Neuroscience in Criminal and Civil Trials
There exists a dichotomy in the use of neuroscience. Contrast, for example, two of the cases: Weinstein and LaMasa v. Bachman, 2008 NY Slip Op 9162, 56 A.D.3d 340, 869 N.Y.S.2d 17 (App. Div.).Weinstein’s expert was not allowed to causally link the cyst to his behavior on the night of the murder, whereas LaMasa’s introduction of PET and qEEG to demonstrate traumatic brain injury (“TBI”) was allowed. The former was a criminal law trial, where the expert could not speak to the defendant’s state of mind, while the latter a civil trial that sought to prove damages from a car accident. Neuroimaging was treated as causal to prove civil damages, but not causal when speaking to a state of mind.
The current dichotomy between the use of causal NS in criminal and civil cases highlights the tension that Rule 704(b) imposes on neuroscience. The legal system is seemingly more accepting of neuroscience as a way to punish (by extracting damages), but less inclined to use it as a mitigating tool (by reducing culpability). Are we erring on the side of false positives? Assuming brain scans are no less correlative for TBI than for impaired cognitive functioning, why are plaintiffs allowed to bring neuroscience, and defendants so limited? See e.g., Paul Rapp et al., Traumatic Brain Injury Detection Using Electrophysiological Methods, 9 Frontiers in Human Neuroscience (2015). The causality of brain scans is therefore encouraged (within Daubert) in civil damage-based claims, but discouraged from the perspective of the defendant’s mens rea. Considering the heavy stakes involved in criminal sentencing, one must question whether § 704(b) has become regressive in the face of neuroscience.
Currently, § 704(b) attempts to prevent experts testifying and presumably disagreeing with one another on the issue of mens rea. In so doing, the jurors must make relatively uninformed decisions based on the evidence presented and their “intuition.” There exists a true risk that Rule 704 is exacerbating folk psychology and encouraging irrational thinking by limiting the expert testimony for criminal defendants.
Neuroscience as it currently stands is more correlative than causal proof, and does not challenge the current system—but it likely will. As neuroscience evolves into causal evidence, it is likely that jurors will continue to be tasked with culpability determinations even when the science is solid, because the legal system is slow to change. This will present the courts with a significant paradigm shift in regard to the role of a jury.
And so, a final question: is the legal system currently shying away from recognizing culpability for acts because it is uncomfortable with the unknown of a new system, or is it because neuroscience does not serve the traditional purposes for which the legal system has existed?