Cognitive neuroscience and the law Brent Garland1,* and Paul W Glimcher2 Advances in cognitive neuroscience now allow us to use that measure a brain feature that correlated, even weakly, physiological techniques to measure and assess mental states with a propensity for violence influence how a court under a growing set of circumstances. The implication of this sentences a convicted felon? Could a more complete growing ability has not been lost on the western legal understanding of the neural mechanism for voluntary community. If biologists can accurately measure mental state, decision-making be used to undermine the notions of then legal conflicts that turn on the true mental states of accountability that are used in criminal convictions? individuals might well be resolvable with techniques ranging Although most neurobiologists agree that these are inter- from electroencephalography to functional magnetic esting questions for the future, neuroscientific evidence is resonance imaging. Therefore, legal practitioners have rapidly entering Western legal systems (in this article we increasingly sought to employ cognitive neuroscientific focus on the US legal system, with which we are most methods and data as evidence to influence legal proceedings. familiar) in ways that would probably surprise and con- This poses a risk, because these scientific methodologies cern many scientists. The result is that the work of have largely been designed and validated for experimental use neuroscientists is being increasingly deployed in various only. Their subsequent use in legal proceedings is an legal contexts, whether the neuroscientists are aware of it application for which they were not intended, and for which or not. In this commentary we argue that the neurobio- those methods are inadequately tested. We propose that logical community must become more aware of how their neurobiologists, who might inadvertently contribute to this work is already beginning to influence the decisions made situation, should be aware of how their papers will be read by by judges, lawyers and legislators. Biologists must the legal community and should play a more active role in become more proactive in their interactions with the legal educating and engaging with that community. community. However, to explain why we (a lawyer and a Addresses neuroscientist) believe this, we have to begin by present- 1,*Currently in private practice. Formerly at: American Association for ing a few key legal concepts with which we believe all the Advancement of Science, 1200 New York Ave NW, Washington, DC, cognitive neurobiologists should become familiar. USA 2 Center for Neural Science, New York University, New York, NY 10003, Use of scientific evidence by the legal USA system Corresponding author: Glimcher, Paul W ([email protected]) First, scientists must understand how and where neu- roscience can be used in legal settings. Many scientists are familiar with the jury trial and with the idea of expert Current Opinion in Neurobiology 2006, 16:130–134 testimony in that setting (see, for example, FED.R.EVID. This commentary comes from a themed issue on 702.). In such cases, if an expert possesses specialized Cognitive neuroscience knowledge that would assist the judge or jury in under- Edited by Paul W Glimcher and Nancy Kanwisher standing the evidence or determining the facts of a case, then parties may seek to have the expert offer an opinion Available online 24th March 2006 into evidence. This formal introduction of expert opinion 0959-4388/$ – see front matter is governed by the rules of evidence. For the purposes of # 2006 Elsevier Ltd. All rights reserved. this article, we talk generally about the law, speaking to DOI 10.1016/j.conb.2006.03.011 general trends and principles, and talking mostly about United States federal law as relates to admissibility of evidence. However, scientists should be aware that in the United States (as in many other countries), the law is Introduction structured such that there are two separate bodies of law During the past two decades, neuroscientific studies have — the federal law and the state law. Although federal law begun to meet the challenge of understanding of cogni- is generally uniform in its nationwide application, the tive function. As the many articles in this themed issue differences between state laws vary more widely. The testify, we now have preliminary biological explanations state and federal courts are separate systems with separate for everything from the control of movements to social jurisdictions that apply separate (and sometimes quite cognition. These physiological insights will challenge, in different) standards to scientific evidence. turn, legal systems that rest on conceptual bases that are often hundreds of years old. A reliable neurobiological The rules of evidence serve as a way of determining what test for willful deception, for example, would indisputa- evidence should be used in considering the case before bly influence our current legal process. Would brain scans the court. The rules governing expert testimony fall Current Opinion in Neurobiology 2006, 16:130–134 www.sciencedirect.com Neuroscience and the law Garland and Glimcher 131 roughly into two approaches in the US — the Frye responses’’. ‘‘In particular...a lateral right-sided region approach (after Frye v. U.S., 54 App. D.C. 46 (1923)) (of the orbitalfrontal cortex) appears particularly respon- and the Daubert approach (Daubert v. Merrell Dow Phar- sive to punishments [2] (in normal subjects)’’. Although maceuticals, 509 U.S. 579 (1993)). In Frye, the Supreme these statements and studies are, of course, directed at a Court of the District of Columbia (a non-federal court) scientific audience and without a doubt reflect our current held that judges, in determining whether to admit scien- understanding of brain function, under current rules of tific evidence, should look to whether the science upon evidence a judge could easily permit these findings to which it is based has ‘‘gained general acceptance in the influence sentencing phase deliberations. Functional particular field in which it belongs’’. Variations of this magnetic resonance imaging (fMRI) data on the hypo- ‘general acceptance’ rule became widely adopted in the or hyperactivity of this punishment-sensitive area in a US by other state and federal courts. Seventy years later, convicted individual could, as a result, be used at a the Supreme Court of the United States set a different sentencing hearing to argue that a convicted criminal standard in Daubert, putting forward a set of four elements was either more or less likely to repeat his or her offence. that judges should consider when evaluating expert scien- tific testimony for admission. That is, whether the under- An even lower standard of evidence is applied in what are lying theory or technique is testable (and whether it has known as ‘pre-formal’ processes. Defense counsel or been tested), whether it has been subjected to peer prosecutors can, for example, use neurobiological test review and publication, its known or potential error rate, results in a dialogue with an individual who has been and its general acceptance in the relevant scientific arrested but not yet charged with any crime — as leverage community. in arguing for a guilty plea, for a reduction in charges, or for a dismissal of the charge. These pre-formal usages The result of these two rulings is that the individual judge typically occur outside of a courtroom setting, and this is in charge of each case, who might have limited (or no) problematic because it makes them difficult to review and scientific training, is placed in the difficult position of places them almost completely beyond the bounds of serving as the ‘gatekeeper’, evaluating the underlying traditional scientific dialogue. science by Daubert or Frye standards, before deciding what to admit into evidence at a particular trial. One As we hope the preceding paragraphs make clear, the can see how this could rapidly become very confusing legal tradition differs from the process with which scien- when scientific evidence conflicts. Which peer reviewed tists are familiar in two ways. First, scientists are accus- article should a judge with no scientific training consider tomed to the validation of fact by peer review, a slow and more seriously, a three page paper published in Nature,a incremental process that seeks to establish ground truth 25 page paper published in the Journal of Neuroscience,ora as a consensus, often during decades of debate. As is clear 50 page paper published in a ‘peer-reviewed’ book pub- from the Frye and Daubert cases, this is simply not the case lished by a vanity press? in the legal world — courts must resolve cases when they are presented with the tools available to them, and cannot Although this alone might raise serious concerns, scien- wait for the development of a scientific consensus. Of tific knowledge can come into play at several other points course this then raises a series of crucial real-world in the legal process at which even the weak and broadly problems. For example, would a neurobiological test interpretable protections of Daubert and Frye might not for willful deception have to perform within a scientist’s apply. This could occur, for example, during a sentencing 95% confidence interval before it is applied in a jury trial? hearing. This is a type of formal hearing that is held The answer is no, the rules of evidence do not require before a judge and that occurs after the guilt of an accused perfection — only relevance, defined as having ‘‘any person has been established. In such hearings, evidence tendency to make the existence of any fact that is of that does not have to meet the Frye or Daubert standards consequence to the determination of the action more for admissibility can be presented to argue for tougher probable or less probable than it would be without the sentencing or for mercy, with evaluations and predictions evidence’’ (FED.R.EVID.