Subjective experience and the law: should fMRI evidence of high punishment tolerance affect sentencing?

Brooklyn Law School professor Adam J. Kolber's paper "The Experiential Future of the Law" was recently published in the Emory Law Journal. The paper asks whether (and when) the law will take account of fMRIs and other tools that create quantitative indicators of subjective experience -- that is, what happens when someone claims that they can tell you how much pain they've experienced, and compare that to the pain that someone else in the same situation might experience?

Kolber points out that this question is relevant to crime and punishment, whether you're someone who wants to be sure that the person who committed the crime is punished in a way that's commensurate with the pain he caused; or whether you believe that punishments should be calculated to be bad enough to deter criminals.

Subjective experience is key to both questions, but the legal issues are sure to be thorny. If you're being sued for "pain and suffering" after your negligence caused someone's broken nose, should you be allowed to introduce evidence showing that the victim has an unusually high pain threshold and ask the jury to reduce the damages accordingly? Should state prosecutors be able to show that a convicted criminal has a high tolerance for incarceration and ask for a longer sentence to ensure that he suffers as much as a comparable criminal with a lower tolerance and a shorter sentence?

In this Article, I describe some of the ways in which new technologies are shifting the way we measure experiences and will continue to do so more dramatically over the next thirty years. I discuss in general terms how new technologies may improve our assessments of physical pain, pain relief, emotional distress, and a variety of psychiatric disorders. I also discuss more particular applications of such technologies to assess whether: (1) a patient is in a persistent vegetative state, (2) a placebo treatment relieves pain, (3) an alleged victim has been abused as a child, (4) an inmate being executed is in pain, (5) an interrogatee has been tortured, and others.

My central claim is that as new technologies emerge to better reveal people's experiences, the law ought to do more to take these experiences into account. In tort and criminal law, we often ignore or downplay the importance of subjective experience. This is no surprise. During the hundreds of years in which these bodies of law developed, we had very poor methods of making inferences about the experiences of others. As we get better at measuring experiences, however, I make the normative claim that we ought to change fundamental aspects of the law to take better account of people's experiences.


(Image: fMRI one, a Creative Commons Attribution (2.0) image from twitchcraft's photostream)


  1. Why this if it’s already clear for quite some time, that the readings from fMRI could be well explained with biased interpretation or statistical insignificance?

    But then again, polygraphs are legal anyway, so why not taking another mythical machine into the evidence-chain.

  2. It would prefer a culture where game theory calculations on how to get the most from everyone else and become the wealthy king are superseded by true care for the community, where punishment is not constantly calculated for every action.

    America is in love with punishment, we make laws just so we can go after those we don’t like or are afraid of.

  3. Mind you, there’s a lot of evidence that suggests that punishment via the criminal system doesn’t work particularly well as a deterrent. Not just for future criminals or even for ones who’ve already been through the system.

  4. Should we also revoke the medical licenses for doctors and dentists with insanely high pain tolerance then?

    Current punishments for crime don’t generally work for changing the criminals/convicted persons behavior and for preventing further crime. (Apart of death penalty, which is used mostly in the third world countries..)

  5. Lets spend more time and money on making sure people suffer enough, yeah that is just great.

  6. This is a great example of the absurdity of the traditional reductive scientific approach applied to a system which cannot be reduced.

    Let me see if I can follow the chain of reasoning: we put people in jail to deter them (and others) from committing crimes. Therefore, making people suffer is the key to deterring crime. Therefore, if we can quantify and optimize suffering, we can optimize crime deterrence as well!

    Also, Kolber is out of touch with the realities of life on the inside. To many people in the prison system who have been in and out of jail multiple times, prison is the life they know best. Some people even find themselves feeling worse when they’re released than when they were in. What does that mean, we should let those people go to ensure they suffer? Or we should put them in progressively more abusive gulags until they finally have the right brain scans?

    This is not an edge case, there’s a significant portion of prisoners who have learned to be relatively at ease in prison, and some who even thrive, getting degrees and improving their lives. If we look at the justice system as a punishment machine, it has totally failed on those people, when in reality of course they’re amazing successes.

  7. Will never happen, don’t worry.

    Logically, fines levied for minor crimes should be proportional to the assets of the criminal. If a destitute man is fined $20 for jaywalking, a millionaire should be fined $250,000, so that he is given the same level of negative reinforcement of his behavior.

    But since the wealthy can influence the law by buying lawmakers and lawmaking processes, proportional punishment cannot happen without sweeping political and economic changes – that are probably not possible without armed revolt.

    And if there’s an armed revolt, the Teabaggers will win, because they are well-armed and know how to hunt, while their opponents are generally unblooded even if they are armed.

    So it’ll never happen.

  8. Some nations give traffic fines based on the wealth of the lawbreaker. To some extent, this violates the principle that
    “all are equal under the law.” But it also makes some amount of sense to give a punishment that will deter.

    When it comes to criminal sanctions however, as others have noted, the deterrent effect of imprisonment is less certain. Perhaps a well-educated wealthy person is far less likely to commit another crime than a poor uneducated person. That doesn’t indicate the poor should get greater sentences.

    I am in favor of reforming our prison system. It doesn’t make an enormous amount of sense to put someone in a small room for 5 years or 10 years and then generally expect something better to emerge. The main advantage is keeping dangerous people off the streets for an amount of time that seems reasonably connected to their crime. The infliction of pain is neither here nor there.

    1. > Some nations give traffic fines based on the wealth of the lawbreaker. To some extent, this violates the principle that
      “all are equal under the law.”

      How does giving fines proportional to income violate the principle of equality under the law?

      Being fined “10% of your monthly income” seems more equal than being fined a set price, regardless of income.

  9. No, fMRI should not influence the sentencing of individuals. Yes, fMRI should influence the overhaul of the sentencing of all criminals, such that it should serve to demonstrate that, since “cruel” is subjective, some punishments are not “cruel” to some people (psychopaths).

    It’s also (for the US) a violation of the First Amendment, Fourth Amendment, and Fifth Amendment, etcetera as it has been well outlined in case law that an individual cannot be compelled to testify against him/her self, and fMRI could constitute a compelled communication (and it should).

    You can’t determine if punishments are or are not cruel for individuals – thus, there ought not be punishments, but rather disciplines. Everyone can be punished. Not everyone can be disciplined. Those that cannot be disciplined should be incarcerated.

  10. Yes…and I also drew a distinction between them. But if you’re speaking purely on principle (which was how I addressed it), that example can violate the principle. turn_self_off argues how it may not violate the principle in a legalistic sense (though obviously different people will pay different amounts for the same exact action). And I’m not arguing against that scheme of punishment. It makes lots of sense.

    The basic point being that “equal under law” for purposes of punishment depends on how you define the punishment. This article addresses how the punishment may be considered unequal depending on how individual people’s experiences differ.

    As far as principle goes (not substantive nature of the violations, obviously), there is no difference. Fining somebody an amount based on factors external to the action is not different than sentencing somebody for a term based on factors external to the action. That said, as I already said, it doesn’t make a lot of sense because of the substantial difference between the potential deterrent effects.

  11. I suppose we could also see if a defendant posses any real feelings of guilt and if not be allowed to go free. Of course, this does not mean that the perpetrator did not commit the crime, but why would we punish someone who did not have the capability to appreciate his actions.

    1. Revenge. That’s productive.

      And revenge is all this would be. If we start tailoring punishments to their subjective pain for the punished, it legitimizes inequality under the law. Few roads so plainly promise unintended consequences before you even start down them.

  12. Has the notion of justice as rehabilitation really been so very long abandoned that people don’t even bat an eye when someone suggests a method of “improving” it by optimizing suffering inflicted?

    Q: “But will this decrease recidivism or assist reintegration into society after release?”

    A: “Who gives a fuck? The point is making them unhappy.”

  13. “During the hundreds of years in which these bodies of law developed, we had very poor methods of making inferences about the experiences of others.”

    Guess what, Professor, that’s still the case. Get back to us when you understand something about the shortcomings and limited applicability of current fMRI research. (Yes, “the phrenology of the 21st century” is pretty much the phase it’s in right now.) Until then, kindly shut the hell up. Thanks.

  14. Punishment only deters good people from criminal behavior, not the troubled folks the system is made to target. So there’s not much a point in it…

  15. if you want to measure human brain activity, use fMRI. if you want to measure human experience, measure human experience (ie, ask people how they feel). attempting to infer experience from brain activity is a logical error — ‘reverse inference’ — that has been well documented in the past 5-10 years.

    For example, here :

  16. It’s sad that so many purportedly civilised people and countries equate retribution with punishment.

  17. Is it fair that plaintifs can use fMRI to ask for more awards if it can’t be used in the reverse?

    If someone is claiming they are in a great deal of pain, and they bring in these medical records, and an expert can show that this IS in fact a record of their pain, why should they be able to ask for more damages because of this.

    And it happens. And it is accepted.

    That said, I have a background in psychology but haven’t done that many neuro studies. I have friends that have, and I find it very interesting. They have high correlation rates in a lot of studies that aren’t really that disputable. And isn’t this what courts are about? The reasonable assumption?

    In the past, the reasonable assumption was that an eye witness could see something, and it would be evidence. And this still is the case…we over estimate our abilities to recall. Gotta suck if you are a black man in a town where 99.99% of the people are white…someone else of your general skin tone, within +/- 1 ft. of height…maybe the same hairstyle in so far as white people can tell. And then…even though you actually look nothing like the other guy, you are the one arrested.

    Hell, I can say this from being white and having purple hair growing up. I was the one tagged for everything when a wierdo was involved. Guy had a green Mohawk…I have purple skater hair…same difference.

    But at the same time, visual evidence is a lot of times the only thing we have to go on, and the correlational rates of selecting someone from visual is far less than friends are getting from neuro studies.

    And there are reasons people do not like this sort of thing, or are skeptical about it. Lie detectors, in the appropriate hands…are actually pretty accurate. Are they accurate in the hands of a law enforcement official who is already biased? No. I would safely say, unless you have a PhD in behavioral type studies, and are running a standardized test on the individual that has been normed to a large enough population, you aren’t going to get much from it. Then again, what will this give you if you do? Probably not the answers you are looking for because standardized means you can’t ask subjective questions about the event you are looking for. People want to disagree, and its fine that non-experts disagree…but those in the field know what this area of knowledge can actually find.

    I had software years ago that would analyze writing styles…and I could tell quite a bit of personality from it. Sometimes better than some of the better known tests. Its nice that someone thinks that they can get into the mind of a killer or other wackjob, but this could tell pretty easily when applying novel (err…new) inputs to see what it would say. It did a pretty decent job and few false positives. In one case, the false positive ended up being true a few years later when one of the texts we used ended up with the writer admitting to some pretty sick stuff.

    And my work was never really that good…it was highly limited to specific areas. Using things like fMRI and doing analysis of the output one could potentially read ones mind. We know what parts of the brain are used for certain activities. Certain parts are used for retrieval, some are used for creation…we know auditory memories are stored different than visual…we know if you are subvocalizing vs. having other types of though.

    How can that correlate to guilt or otherwise? If you are asking someone to talk about an event, and the major activity is always in the creation portion…and you’ve taken the appropriate baselines to ensure that some how it didn’t drift in this particular interviewee…and they constantly go to that section, you know with high certainty that they are creating the story.

    Are they lying about what actually happened? Are they embellishing? Maybe they are just being dishonest because they are the kind of person that feels the need to lie about everything because they dislike authority. One has a better chance of identifying this with more accuracy with a test like these than one where you interview someone and catch them in a lie. Both would still go to impeach their credibility…and yet, the nonscientific is far more believed.

    It comes down to this…do we want to get more scientific and more accurate with crimes, or do we want to rely on what we’ve always relied on which is HIGHLY inaccurate. I know the reality is, some fucking police force is going to send their detective out for continuing education credits to augment his undergrad education, and they are going to introduce this to the courts…but the fact that others are misusing the technologies does not make the technology bad. It just means people often overstep their abilities and given that, at least in the society I live, experts and the educated are looked at in disdain (i.e., global warming experts having to defend themselves from people with barely a college degree in areas that have nothing to do with climatology)…the technology WILL be misused.

    Again…doesn’t make the technology any less accurate…it just makes people flawed…

    1. @ Anon #32

      Did you read the paper, or even Cory’s excerpt? Kolber isn’t arguing for using psychoanalysis merely to determine guilt. He’s recommending determining sentencing based on a neurological analysis of the convict’s pain/suffering threshold. In other words he wants us to dole out punishments based on a person’s physiological characteristics. If we go that route we might as well replace blindfolded Lady Justice with a statue of of this:

      What part of “equal under the law” is so hard for Kolber and his supporters to grok?

  18. We should just add this to the hate crime laws, admit what we’re doing and open the Ministry of Thought.

    After all, only right thinking and feeling citizens are good for society as a whole (or hole).

  19. The idea of punishment is predicated on the idea that it either helps the victim and/or helps society by reducing the chance of the criminal recidivism.

    The problem is most violent crime is perpetrated by a small percentage of the population that has been shown to be unaffected by punishment of others or even past punishment of themselves. Studies have shown that the lower prefrontal lobe often has structural and/or chemical defect.

    Since such people are an ongoing danger to others, all physical incarceration resources should be reserved for them.

    Instead, over half the people in US prisons are there on non-violent charges like cannabis possession or copyright violation and as a result, repeat violent offenders are released again and again.

    Most violent offenders are the offspring of violent offenders. It could be genetics. It could be environment. It’s likely both. Regardless, it’s should be obvious by now that such people should not be allowed to have or be around children.

    1. @ Anon #35

      Regardless, it’s should be obvious by now that such people should not be allowed to have or be around children.

      I was with you right up until the eugenics suggestion. This “solution” was tried by some of the most horrific regimes in human history. Never again.

  20. Functional MRI is a good way to measure, validate, illustrate and document pain, and has been applied to acute pain states and to fibromyalgia. Other emotions which lend to neuroimaging include depression, anxiety, suicidal thought. fMRI has been adapted commercially to interrogation. Structural neuroimaging, MRS and volumetry have application to cognitive impairment. One good source to follow is the Center for Law and the Biosciences blog at Stanford Law School. I would also recommend “The brain scanner that feels your pain” in the March 3, 2010 issue of The New Scientist

Comments are closed.