Thursday, October 6, 2016

Guilty or Not Guilty: Policy Considerations for Using Neuroimaging as Evidence in Courts

By Sunidhi Ramesh

This post was written as part of a class assignment from students who took a neuroethics course with Dr. Rommelfanger in Paris of Summer 2016. 

Sunidhi Ramesh, an Atlanta native, is a third year student at Emory University where she is double majoring in Sociology and Neuroscience and Behavioral Biology. She plans to pursue a career in medicine and holds a deep interest in sparking conversation and change around her, particularly in regards to pressing social matters and how education in America is both viewed and handled. In her spare time, Sunidhi is a writer, bridge player, dancer, and violinist.

 In 1893, Dr. Henry Howard Holmes opened his World’s Fair Hotel to the world [1].

But what his guests did not know was that the basement was filled with jars of poison, boxes of bones, and large surgical tables. Chutes from the guest rooms existed only to slide bodies into a pile downstairs. In the few months that the hotel was open for the public, Holmes, dubbed America’s first serial killer, killed an estimated number of 200 guests. Two years later, he was put on trial, found guilty, and sentenced to death [1].

H. H. Holmes, image courtesy WikiCommons
“I was born with the devil in me. I could not help the fact that I was a murderer, no more than the poet can help the inspiration to sing,” Holmes is quoted to have said [1]. But our judicial system does not care much for whether or not a murderer “can help it.” A crime is a crime; a murder is a murder. A guilty crime and a guilty mind are enough to warrant retribution—a punishment.

Holmes’ crimes were premeditated. They were planned, thought out, and acted upon. And that is all that the judge and jury knew for sure. They didn’t know if he was “born with it” or if it was out of his control. They had no way of knowing that, so they hanged him.

But what if we can know? What if we have the technology to determine that a crime is not due to free will but to a series of chemical signals and biological outputs entirely out of conscious control? 

Cue neuroimaging and the ongoing debate on its use in criminal courts.

Functional scanning, including PET, SPECT, and, most recently fMRI, are techniques that are beginning to infiltrate the field of forensics. These technologies assume that “disruptions in the function of discrete parts of the brain can lead to alterations in particular aspects of cognition and behavior” [2]. For example, damage in the prefrontal cortex, a region in the brain critical for reasoning, planning, impulse control, and moral judgment, can imply marred decision-making abilities [2]. Other areas such as the amygdala (involved with sudden emotions, including anger) and the anterior cingulate cortex (involved with empathy and compassion) can be in question in legal settings as well. But ultimately, these scanning techniques simply measure changes in blood flow and correlate these changes with increased or decreased brain activity in specific areas [2].

Recently, functional neuroimaging evidence has been used in criminal cases “in support of the insanity defense, claims of incompetence to stand trial, and pleas for mitigation in sentencing” [2].

In 2007, Peter Braunstein, dressed as a firefighter, set off a smoke bomb and knocked on a woman’s apartment door. After knocking her out with chloroform, Braunstein “sexually assaulted her for the next 13 hours” [2]. During his trial, his defense team used PET scans to argue that “Braunstein had decreased function in his frontal lobes,” the part of the brain that controls “initiation and cessation of behavior, planning, and moral judgment.” The defense contended that Braunstein was “completely unable to plan” and incapable of thinking ahead [2].

Perhaps, if this technology was around in the 1800’s, Holmes’ claims could also have been corroborated by the structures in his brain. If this did happen, America’s first serial killer would have avoided his death. He could have even walked free, with a greatly reduced punishment to follow his crimes.

These are some of the applications of neuroimaging in courts today. It is a matter that needs regulation and guidelines to ensure appropriate and justified use.

According to a study conducted in 2006, “there have been roughly 130 reported opinions involving PET and/or SPECT evidence” [2]. Of those 130 cases, about four-fifths of them have admitted these functional neuroimaging scans as evidence. And although “[neuroimaging] evidence [is most] often admitted along with other neurological and clinical evidence of the party’s mental condition,” the implications of using this evidence alone in future cases still remain [2].

In terms of ethical considerations, the current debate primarily questions matters of privacy, the current factual value of neuroimaging outside of a medical context, and the relationship between brain and thought [3].

That said, current policy on neuroimaging is lacking. Advocates and opponents of using neuroimaging in courts have tried to push the current regulations either way; still, the technology is new enough to where there aren’t enough guidelines in place.

 Here is what we should consider:

Image courtesy of WikiCommons
a) Understanding the meaning and limitations of neuroimaging

Neuroimaging isn’t perfect. A simple fMRI or brain scan generates images that are open to interpretation and human subjectivity. The vast majority of the general public seems to be under the impression that these scans can give direct insight into what a person is thinking or feeling, but this is generally untrue [4].

Thus, first and foremost, if this evidence is to be admissible in courts, judges, juries, and attorneys need to be trained in the value, meaning, and limitations of these scans. They need to understand that these scans yield pictures that are then interpreted as well as the full extent of what this technology can imply. Personality traits, mental illness, sexual preferences, and predisposition to drug addiction are all types of information that can be gathered from neuroimaging [4]. There are, however, important limits to the type and accuracy of this information; these restrictions need to be openly shared and discussed among members of the court as a part of training programs that take place long before actual cases are reviewed.

b) Slippery slope of inferring a state of mind 

Although behavior can be explained by brain evidence, brain evidence cannot directly imply behavior. It is easy to find an anomaly in the brain and point to it directly as the source of any kind of misdemeanor. But no two brains are exactly alike. Where do we draw the line between person-to-person variation and a serious abnormality that caused a crime to be committed?

To address this, a conference must be put into place. This conference, consisting of the most renown ethicists, neurologists, judges, and criminal defense lawyers, must set a goal to determine where this line is drawn. What are the parameters of an “average brain”? How different is “abnormal”? Case studies should be referenced to understand how abnormalities in specific regions could impact behavior.

c) Considering “free will”

What does giving neuroimaging validity in courts imply about our own selves? If we deem all neuroimaging evidence as admissible, are we acknowledging that all crime is committed due to the wiring of the brain? Where do we draw the line between intentionality/free will and the biochemical processes of the brain in regards to committing a crime?

 The implications of this must be considered through further research, and the idea of “free will” as a concept should be brought up during conversations about neuroimaging. Furthermore, these discussions should occur in coordination with philosophers and scientists; ideally, this may begin some level of understanding of the intersection between biological decision-making machinery and “our own” goal-driven decisions.

d) Protecting the privacy of the defendant

It is important to remember that a great deal of neuroscience relies on the assumption that the mind is simply “the processes of the organic brain.” This has huge implications because “it means that when we look at the functioning brain through [these imaging technologies], we are essentially looking into the mind of another person” [5].

 What is critical is that these neuroimaging techniques are used only for what they need to be used for. There is a range of information that can be procured from the scans, but evidence that is not relevant to these cases should not be researched or referred to.

 That being said, criminal court cases, especially those that are tried in higher courts, receive a great deal of media attention. With the nature of technology today, the public eye sees all; how, then, do we keep the defendant’s brain records private? Thus, it is necessary for this information to be concealed and protected. It should be presented solely to the judge, the jury, and the relevant criminal attorneys with security measures taken to prevent it from reaching the general public. Furthermore, the neurologists involved with the testimony should be unrelated to the case, selected by the courts, to prevent any bias from either the plaintiff or the defendant.

e) Reliability and accuracy of results 

With the technology as it is today, one test and one expert cannot be enough to reliably present this information. At least two independent, separate tests must be conducted with more than one neurologist to corroborate the findings. If any inconsistencies are found either between the two scans or between the opinions of the two neurologists, the information should be deemed inadmissible in the courts on the grounds of being unreliable.

f) Lie detection as a basis for neuroimaging policy

Lie detection through the polygraph has been used as to discriminate lying between truth telling since the early 1900’s. Although the technique is far from perfect, many courts use it as a form of supporting evidence. Lie detection poses many similar questions as neuroimaging. How accurate is it? Is there any privacy involved? When can we use it and when can we not [6]?

 In order to address these questions in regards to neuroimaging, we must consider turning to current lie detection policies regarding its use in the courtroom and frame neuroimaging policy around them.

g) Use of neuroimaging when necessary

There is a range of implications that come with the consistent implementation of neuroimaging in court cases [7]. Will criminals end up always relying on this technology to bail them out? Will we get to a point where every criminal is effectively “made innocent” through neuroimaging? Perhaps. To avoid this, stipulations need to be drawn for cases where neuroimaging can be used. This could be in cases where the intentionality of the defendant is uncertain or where no real evidence exists. Neuroimaging should not be open for use in every case, and a strong regulation system should be built to ensure this.

h) Neuroimaging as the “be-all and end-all”

Similar to the current role of lie detection techniques, neuroimaging should be taken with a grain of salt. It should not be used as the only evidence that decides the fate of a case; the judge, once trained in understanding the science behind these techniques, should be given the right to decide whether or not the scans are relevant or admissible in relation to each specific case. Until the technology is refined to near-perfection, neuroimaging should never be the only evidence available on the mental capacity of an individual.

In addition to these policy considerations, more research on the topic is necessary to ensure that this technology is indeed accurate and applicable to this field.

Ideally, neuroimaging as a technology in general would need to be furthered to avoid premature, inappropriate use in criminal courts. Whatever the case, these techniques are picking up speed as relevant tools in courts around the United States. They must be properly regulated, monitored, studied, and discussed to ensure their safe and ethical use as a form of evidence in criminal case proceedings.


 1. “Dr. H. H. Holmes.” The Devil In The White City, by Erik Larson. N.p., 2008. Web. 18 June 2016.

2. Appelbaum, P. S. (2015). Law & psychiatry: Through a glass darkly: Functional neuroimaging evidence enters the courtroom. Psychiatric Services.

3. Agid, Yves, and Ali Benmakhlouf. “Ethical Issues Arising out of Functional Neuroimaging.” National Consultative Ethics Committee for Health and Life Sciences 116 (n.d.): 1-19. Web. 17 June 2016.

4. Feigenson, N. (2006). Brain imaging and courtroom evidence: On the admissibility and persuasiveness of fMRI. International Journal of Law in Context, 2(03), 233-255.

5. Finn, D. P. (2006). Brain imaging and privacy: how recent advances in neuroimaging implicate privacy concerns. bepress Legal Series, 1752.

6. Rusconi, E., & Mitchener-Nissen, T. (2013). Prospects of functional magnetic resonance imaging as lie detector.

7. Baertschi, B. (2011). Neuroimaging in the Courts of Law. Journal of Applied Ethics and Philosophy, 3, 9-16.

Want to cite this post?

Ramesh, S. (2016). Guilty or Not Guilty: Policy Considerations for Using Neuroimaging as Evidence in Courts. The Neuroethics Blog. Retrieved on , from

1 comment:

el roam said…

Thanks for that important and interesting post . Many complications , too many , yet :

Just to put thinks in right proportions :

The law , deals with functions , not definitions !! The judge or the lawmaker, actually, ask themselves, simple question:

Could the perpetrator control his actions at the time , or not ?? in this regard , imaging of the brain , is senseless to some extent at least . The post is missing tensions of times perspective . imaging can actually ,in online terms , describe what we : see , feel , think , in 100 percent accuracy , no doubt !! But , concerning our prospective and retrospective actions , it is almost useless in legal terms . The judge needs to deal with given circumstances in the past , already completed . Imaging , is not helpful almost in this regard . The issue is not one of medical definition , but : functional , legal , circumstantial one . Here for example , article 34 H , from the Israeli penal code ( common law ) , here I quote :

Mental incompetence 34H.

No person shall bear criminal responsibility for an act committed by him, if – at the time the act was committed, because of a disease that adversely affected his spirit or because of a mental impediment – he lacked any real ability – (1) to understand what he did or the wrongful nature of his act; or (2) to abstain from committing the act.

End of quotation :

So we read : " at the time the act was committed …. " So , sometimes a healthy person may be found incompetent in this regard , sometimes a sick man would be found , totally responsible , what counts , is the retrospective observation of the nature of the act while actually committed . The current , categoric , medical definition of the medical state of one person , won't do simply !! but for having certain circumstantial validity , like other evidences indeed ( unless , the perpetrator , was and still is , in total loss state , then a trial , is simply irrelevant then ) .

As such , unless , one could scan the brain of one perpetrator , during his criminal act ( and it can be done generally speaking ) , it can't help much !! the issue of " free will " so called , would continue to be subject for great debate , with or without brains scanned or not .