[tt] Neuroscience: Law: The Brain on the Stand - Jeffrey Rosen

[Premise Checker]

Neuroscience: Law: The Brain on the Stand - Jeffrey Rosen
http://www.nytimes.com/2007/03/11/magazine/11Neurolaw.t.html

By JEFFREY ROSEN

I. Mr. Weinsteins Cyst When historians of the future try to
identify the moment that neuroscience began to transform the
American legal system, they may point to a little-noticed case from
the early 1990s. The case involved Herbert Weinstein, a 65-year-old
ad executive who was charged with strangling his wife, Barbara, to
death and then, in an effort to make the murder look like a
suicide, throwing her body out the window of their 12th-floor
apartment on East 72nd Street in Manhattan. Before the trial began,
Weinsteins lawyer suggested that his client should not be held
responsible for his actions because of a mental defect namely, an
abnormal cyst nestled in his arachnoid membrane, which surrounds
the brain like a spider web.

The implications of the claim were considerable. American law holds
people criminally responsible unless they act under duress (with a
gun pointed at the head, for example) or if they suffer from a
serious defect in rationality like not being able to tell right
from wrong. But if you suffer from such a serious defect, the law
generally doesnt care why whether its an unhappy childhood or an
arachnoid cyst or both. To suggest that criminals could be excused
because their brains made them do it seems to imply that anyone
whose brain isnt functioning properly could be absolved of
responsibility. But should judges and juries really be in the
business of defining the normal or properly working brain? And
since all behavior is caused by our brains, wouldnt this mean all
behavior could potentially be excused?

The prosecution at first tried to argue that evidence of Weinsteins
arachnoid cyst shouldnt be admitted in court. One of the
governments witnesses, a forensic psychologist named Daniel
Martell, testified that brain-scanning technologies were new and
untested, and their implications werent yet widely accepted by the
scientific community. Ultimately, on Oct. 8, 1992, Judge Richard
Carruthers issued a Solomonic ruling: Weinsteins lawyers could tell
the jury that brain scans had identified an arachnoid cyst, but
they couldnt tell jurors that arachnoid cysts were associated with
violence. Even so, the prosecution team seemed to fear that simply
exhibiting images of Weinsteins brain in court would sway the jury.
Eleven days later, on the morning of jury selection, they agreed to
let Weinstein plead guilty in exchange for a reduced charge of
manslaughter.

After the Weinstein case, Daniel Martell found himself in so much
demand to testify as a expert witness that he started a consulting
business called Forensic Neuroscience. Hired by defense teams and
prosecutors alike, he has testified over the past 15 years in
several hundred criminal and civil cases. In those cases,
neuroscientific evidence has been admitted to show everything from
head trauma to the tendency of violent video games to make children
behave aggressively. But Martell told me that its in death-penalty
litigation that neuroscience evidence is having its most
revolutionary effect. Some sort of organic brain defense has become
de rigueur in any sort of capital defense, he said. Lawyers
routinely order scans of convicted defendants brains and argue that
a neurological impairment prevented them from controlling
themselves. The prosecution counters that the evidence shouldnt be
admitted, but under the relaxed standards for mitigating evidence
during capital sentencing, it usually is. Indeed, a Florida court
has held that the failure to admit neuroscience evidence during
capital sentencing is grounds for a reversal. Martell remains
skeptical about the worth of the brain scans, but he observes that
theyve revolutionized the law.

The extent of that revolution is hotly debated, but the influence
of what some call neurolaw is clearly growing. Neuroscientific
evidence has persuaded jurors to sentence defendants to life
imprisonment rather than to death; courts have also admitted
brain-imaging evidence during criminal trials to support claims
that defendants like John W. Hinckley Jr., who tried to assassinate
President Reagan, are insane. Carter Snead, a law professor at
Notre Dame, drafted a staff working paper on the impact of
neuroscientific evidence in criminal law for President Bushs
Council on Bioethics. The report concludes that neuroimaging
evidence is of mixed reliability but the large number of cases in
which such evidence is presented is striking. That number will no
doubt increase substantially. Proponents of neurolaw say that
neuroscientific evidence will have a large impact not only on
questions of guilt and punishment but also on the detection of lies
and hidden bias, and on the prediction of future criminal behavior.
At the same time, skeptics fear that the use of brain-scanning
technology as a kind of super mind-reading device will threaten our
privacy and mental freedom, leading some to call for the legal
system to respond with a new concept of cognitive liberty.

One of the most enthusiastic proponents of neurolaw is Owen Jones,
a professor of law and biology at Vanderbilt. Jones (who happens to
have been one of my law-school classmates) has joined a group of
prominent neuroscientists and law professors who have applied for a
large MacArthur Foundation grant; they hope to study a wide range
of neurolaw questions, like: Do sexual offenders and violent
teenagers show unusual patterns of brain activity? Is it possible
to capture brain images of chronic neck pain when someone claims to
have suffered whiplash? In the meantime, Jones is turning
Vanderbilt into a kind of Los Alamos for neurolaw. The university
has just opened a $27 million neuroimaging center and has poached
leading neuroscientists from around the world; soon, Jones hopes to
enroll students in the nations first program in law and
neuroscience. Its breathlessly exciting, he says. This is the new
frontier in law and science were peering into the black box to see
how the brain is actually working, that hidden place in the dark
quiet, where we have our private thoughts and private reactions and
the law will inevitably have to decide how to deal with this new
technology.

II. A Visit to Vanderbilt Owen Jones is a disciplined and quietly
intense man, and his enthusiasm for the transformative power of
neuroscience is infectious. With René Marois, a neuroscientist in
the psychology department, Jones has begun a study of how the human
brain reacts when asked to impose various punishments. Informally,
they call the experiment Harm and Punishment and they offered to
make me one of their first subjects.

We met in Joness pristine office, which is decorated with a human
skull and calipers, like those that phrenologists once used to
measure the human head; his father is a dentist, and his
grandfather was an electrical engineer who collected tools. We
walked over to Vanderbilts Institute of Imaging Science, which,
although still surrounded by scaffolding, was as impressive as
Jones had promised. The basement contains one of the few 7-tesla
magnetic-resonance-imaging scanners in the world. For Harm and
Punishment, Jones and Marois use a less powerful 3 tesla, which is
the typical research M.R.I.

We then made our way to the scanner. After removing all metal
objects including a belt and a stray dry-cleaning tag with a staple
I put on earphones and a helmet that was shaped like a birdcage to
hold my head in place. The lab assistant turned off the lights and
left the room; I lay down on the gurney and, clutching a panic
button, was inserted into the magnet. All was dark except for a
screen flashing hypothetical crime scenarios, like this one: John,
who lives at home with his father, decides to kill him for the
insurance money. After convincing his father to help with some
electrical work in the attic, John arranges for him to be
electrocuted. His father survives the electrocution, but he is
hospitalized for three days with injuries caused by the electrical
shock. I was told to press buttons indicating the appropriate level
of punishment, from 0 to 9, as the magnet recorded my brain
activity.

After I spent 45 minutes trying not to move an eyebrow while
assigning punishments to dozens of sordid imaginary criminals,
Marois told me through the intercom to try another experiment:
namely, to think of familiar faces and places in sequence, without
telling him whether I was starting with faces or places. I thought
of my living room, my wife, my parents apartment and my twin sons,
trying all the while to avoid improper thoughts for fear they would
be discovered. Then the experiments were over, and I stumbled out
of the magnet.

The next morning, Owen Jones and I reported to René Maroiss
laboratory for the results. Maroiss graduate students, who had been
up late analyzing my brain, were smiling broadly. Because I had
moved so little in the machine, they explained, my brain activity
was easy to read. Your head movement was incredibly low, and you
were the harshest punisher weve had, Josh Buckholtz, one of the
grad students, said with a happy laugh. You were a researchers
dream come true! Buckholtz tapped the keyboard, and a
high-resolution 3-D image of my brain appeared on the screen in
vivid colors. Tiny dots flickered back and forth, showing my eyes
moving as they read the lurid criminal scenarios. Although I was
only the fifth subject to be put in the scanner, Marois emphasized
that my punishment ratings were higher than average. In one case, I
assigned a 7 where the average punishment was 4. You were focusing
on the intent, and the others focused on the harm, Buckholtz said
reassuringly.

Marois explained that he and Jones wanted to study the interactions
among the emotion-generating regions of the brain, like the
amygdala, and the prefrontal regions responsible for reason. It is
also possible that the prefrontal cortex is critical for
attributing punishment, making the essential decision about what
kind of punishment to assign, he suggested. Marois stressed that in
order to study that possibility, more subjects would have to be put
into the magnet. But if the prefrontal cortex does turn out to be
critical for selecting among punishments, Jones added, it could be
highly relevant for lawyers selecting a jury. For example, he
suggested, lawyers might even select jurors for different cases
based on their different brain-activity patterns. In a complex
insider-trading case, for example, perhaps the defense would like
to have a juror making decisions on maximum deliberation and
minimum emotion; in a government entrapment case, emotional
reactions might be more appropriate.

We then turned to the results of the second experiment, in which I
had been asked to alternate between thinking of faces and places
without disclosing the order. We think we can guess what you were
thinking about, even though you didnt tell us the order you started
with, Marois said proudly. We think you started with places and we
will prove to you that it wasnt just luck. Marois showed me a
picture of my parahippocampus, the area of the brain that responds
strongly to places and the recognition of scenes. Its lighting up
like Christmas on all cylinders, Marois said. It worked
beautifully, even though we havent tried this before here.

He then showed a picture of the fusiform area, which is responsible
for facial recognition. It, too, lighted up every time I thought of
a face. This is a potentially very serious legal implication, Jones
broke in, since the technology allows us to tell what people are
thinking about even if they deny it. He pointed to a series of
practical applications. Because subconscious memories of faces and
places may be more reliable than conscious memories, witness
lineups could be transformed. A child who claimed to have been
victimized by a stranger, moreover, could be shown pictures of the
faces of suspects to see which one lighted up the face-recognition
area in ways suggesting familiarity.

Jones and Marois talked excitedly about the implications of their
experiments for the legal system. If they discovered a significant
gap between peoples hard-wired sense of how severely certain crimes
should be punished and the actual punishments assigned by law,
federal sentencing guidelines might be revised, on the principle
that the law shouldnt diverge too far from deeply shared beliefs.
Experiments might help to develop a deeper understanding of the
criminal brain, or of the typical brain predisposed to criminal
activity.

III. The End of Responsibility? Indeed, as the use of functional
M.R.I. results becomes increasingly common in courtrooms, judges
and juries may be asked to draw new and sometimes troubling lines
between normal and abnormal brains. Ruben Gur, a professor of
psychology at the University of Pennsylvania School of Medicine,
specializes in doing just that. Gur began his expert-witness career
in the mid-1990s when a colleague asked him to help in the trial of
a convicted serial killer in Florida named Bobby Joe Long. Known as
the classified-ad rapist, because he would respond to classified
ads placed by women offering to sell household items, then rape and
kill them, Long was sentenced to death after he committed at least
nine murders in Tampa. Gur was called as a national expert in
positron-emission tomography, or PET scans, in which patients are
injected with a solution containing radioactive markers that
illuminate their brain activity. After examining Longs PET scans,
Gur testified that a motorcycle accident that had left Long in a
coma had also severely damaged his amygdala. It was after emerging
from the coma that Long committed his first rape.

I didnt have the sense that my testimony had a profound impact, Gur
told me recently Long is still filing appeals but he has testified
at more than 20 capital cases since then. He wrote a widely
circulated affidavit arguing that adolescents are not as capable of
controlling their impulses as adults because the development of
neurons in the prefrontal cortex isnt complete until the early 20s.
Based on that affidavit, Gur was asked to contribute to the
preparation of one of the briefs filed by neuroscientists and
others in Roper v. Simmons, the landmark case in which a divided
Supreme Court struck down the death penalty for offenders who
committed crimes when they were under the age of 18.

The leading neurolaw brief in the case, filed by the American
Medical Association and other groups, argued that because
adolescent brains are not fully developed in the prefrontal
regions, adolescents are less able than adults to control their
impulses and should not be held fully accountable for the
immaturity of their neural anatomy. In his majority decision,
Justice Anthony Kennedy declared that as any parent knows and as
the scientific and sociological studies cited in the briefs tend to
confirm, [a] lack of maturity and an underdeveloped sense of
responsibility are found in youth more often than in adults.
Although Kennedy did not cite the neuroscience evidence
specifically, his indirect reference to the scientific studies in
the briefs led some supporters and critics to view the decision as
the Brown v. Board of Education of neurolaw.

One important question raised by the Roper case was the question of
where to draw the line in considering neuroscience evidence as a
legal mitigation or excuse. Should courts be in the business of
deciding when to mitigate someones criminal responsibility because
his brain functions improperly, whether because of age, in-born
defects or trauma? As we learn more about criminals brains, will we
have to redefine our most basic ideas of justice?

Two of the most ardent supporters of the claim that neuroscience
requires the redefinition of guilt and punishment are Joshua D.
Greene, an assistant professor of psychology at Harvard, and
Jonathan D. Cohen, a professor of psychology who directs the
neuroscience program at Princeton. Greene got Cohen interested in
the legal implications of neuroscience, and together they conducted
a series of experiments exploring how peoples brains react to moral
dilemmas involving life and death. In particular, they wanted to
test peoples responses in the f.M.R.I. scanner to variations of the
famous trolley problem, which philosophers have been arguing about
for decades.

The trolley problem goes something like this: Imagine a train
heading toward five people who are going to die if you dont do
anything. If you hit a switch, the train veers onto a side track
and kills another person. Most people confronted with this scenario
say its O.K. to hit the switch. By contrast, imagine that youre
standing on a footbridge that spans the train tracks, and the only
way you can save the five people is to push an obese man standing
next to you off the footbridge so that his body stops the train.
Under these circumstances, most people say its not O.K. to kill one
person to save five.

I wondered why people have such clear intuitions, Greene told me,
and the core idea was to confront people with these two cases in
the scanner and see if we got more of an emotional response in one
case and reasoned response in the other. As it turns out, thats
precisely what happened: Greene and Cohen found that the brain
region associated with deliberate problem solving and self-control,
the dorsolateral prefrontal cortex, was especially active when
subjects confronted the first trolley hypothetical, in which most
of them made a utilitarian judgment about how to save the greatest
number of lives. By contrast, emotional centers in the brain were
more active when subjects confronted the second trolley
hypothetical, in which they tended to recoil at the idea of
personally harming an individual, even under such wrenching
circumstances. This suggests that moral judgment is not a single
thing; its intuitive emotional responses and then cognitive
responses that are duking it out, Greene said.

To a neuroscientist, you are your brain; nothing causes your
behavior other than the operations of your brain, Greene says. If
thats right, it radically changes the way we think about the law.
The official line in the law is all that matters is whether youre
rational, but you can have someone who is totally rational but
whose strings are being pulled by something beyond his control. In
other words, even someone who has the illusion of making a free and
rational choice between soup and salad may be deluding himself,
since the choice of salad over soup is ultimately predestined by
forces hard-wired in his brain. Greene insists that this insight
means that the criminal-justice system should abandon the idea of
retribution the idea that bad people should be punished because
they have freely chosen to act immorally which has been the focus
of American criminal law since the 1970s, when rehabilitation went
out of fashion. Instead, Greene says, the law should focus on
deterring future harms. In some cases, he supposes, this might mean
lighter punishments. If its really true that we dont get any
prevention bang from our punishment buck when we punish that
person, then its not worth punishing that person, he says. (On the
other hand, Carter Snead, the Notre Dame scholar, maintains that
capital defendants who are not considered fully blameworthy under
current rules could be executed more readily under a system that
focused on preventing future harms.)

Others agree with Greene and Cohen that the legal system should be
radically refocused on deterrence rather than on retribution. Since
the celebrated MNaughten case in 1843, involving a paranoid British
assassin, English and American courts have recognized an insanity
defense only for those who are unable to appreciate the difference
between right and wrong. (This is consistent with the idea that
only rational people can be held criminally responsible for their
actions.) According to some neuroscientists, that rule makes no
sense in light of recent brain-imaging studies. You can have a
horrendously damaged brain where someone knows the difference
between right and wrong but nonetheless cant control their
behavior, says Robert Sapolsky, a neurobiologist at Stanford. At
that point, youre dealing with a broken machine, and concepts like
punishment and evil and sin become utterly irrelevant. Does that
mean the person should be dumped back on the street? Absolutely
not. You have a car with the brakes not working, and it shouldnt be
allowed to be near anyone it can hurt.

Even as these debates continue, some skeptics contend that both the
hopes and fears attached to neurolaw are overblown. Theres nothing
new about the neuroscience ideas of responsibility; its just
another material, causal explanation of human behavior, says
Stephen J. Morse, professor of law and psychiatry at the University
of Pennsylvania. How is this different than the Chicago school of
sociology, which tried to explain human behavior in terms of
environment and social structures? How is it different from genetic
explanations or psychological explanations? The only thing
different about neuroscience is that we have prettier pictures and
it appears more scientific.

Morse insists that brains do not commit crimes; people commit
crimes a conclusion he suggests has been ignored by advocates who,
infected and inflamed by stunning advances in our understanding of
the brain . . . all too often make moral and legal claims that the
new neuroscience . . . cannot sustain. He calls this brain
overclaim syndrome and cites as an example the neuroscience briefs
filed in the Supreme Court case Roper v. Simmons to question the
juvenile death penalty. What did the neuroscience add? he asks. If
adolescent brains caused all adolescent behavior, we would expect
the rates of homicide to be the same for 16- and 17-year-olds
everywhere in the world their brains are alike but in fact, the
homicide rates of Danish and Finnish youths are very different than
American youths. Morse agrees that our brains bring about our
behavior Im a thoroughgoing materialist, who believes that all
mental and behavioral activity is the causal product of physical
events in the brain but he disagrees that the law should excuse
certain kinds of criminal conduct as a result. Its a total non
sequitur, he says. So what if theres biological causation?
Causation cant be an excuse for someone who believes that
responsibility is possible. Since all behavior is caused, this
would mean all behavior has to be excused. Morse cites the case of
Charles Whitman, a man who, in 1966, killed his wife and his
mother, then climbed up a tower at the University of Texas and shot
and killed 13 more people before being shot by police officers.
Whitman was discovered after an autopsy to have a tumor that was
putting pressure on his amygdala. Even if his amygdala made him
more angry and volatile, since when are anger and volatility
excusing conditions? Morse asks. Some people are angry because they
had bad mommies and daddies and others because their amygdalas are
mucked up. The question is: When should anger be an excusing
condition?

Still, Morse concedes that there are circumstances under which new
discoveries from neuroscience could challenge the legal system at
its core. Suppose neuroscience could reveal that reason actually
plays no role in determining human behavior, he suggests
tantalizingly. Suppose I could show you that your intentions and
your reasons for your actions are post hoc rationalizations that
somehow your brain generates to explain to you what your brain has
already done without your conscious participation. If neuroscience
could reveal us to be automatons in this respect, Morse is prepared
to agree with Greene and Cohen that criminal law would have to
abandon its current ideas about responsibility and seek other ways
of protecting society.

Some scientists are already pushing in this direction. In a series
of famous experiments in the 1970s and 80s, Benjamin Libet measured
peoples brain activity while telling them to move their fingers
whenever they felt like it. Libet detected brain activity
suggesting a readiness to move the finger half a second before the
actual movement and about 400 milliseconds before people became
aware of their conscious intention to move their finger. Libet
argued that this leaves 100 milliseconds for the conscious self to
veto the brains unconscious decision, or to give way to it
suggesting, in the words of the neuroscientist Vilayanur S.
Ramachandran, that we have not free will but free wont.

Morse is not convinced that the Libet experiments reveal us to be
helpless automatons. But he does think that the study of our
decision-making powers could bear some fruit for the law. Im
interested, he says, in people who suffer from drug addictions,
psychopaths and people who have intermittent explosive disorder
thats people who have no general rationality problem other than
they just go off. In other words, Morse wants to identify the
neural triggers that make people go postal. Suppose we could show
that the higher deliberative centers in the brain seem to be
disabled in these cases, he says. If these are people who cannot
control episodes of gross irrationality, weve learned something
that might be relevant to the legal ascription of responsibility.
That doesnt mean they would be let off the hook, he emphasizes: You
could give people a prison sentence and an opportunity to get
fixed.

IV. Putting the Unconscious on Trial If debates over criminal
responsibility long predate the f.M.R.I., so do debates over the
use of lie-detection technology. Whats new is the prospect that lie
detectors in the courtroom will become much more accurate, and
correspondingly more intrusive. There are, at the moment, two
lie-detection technologies that rely on neuroimaging, although the
value and accuracy of both are sharply contested. The first,
developed by Lawrence Farwell in the 1980s, is known as brain
fingerprinting. Subjects put on an electrode-filled helmet that
measures a brain wave called p300, which, according to Farwell,
changes its frequency when people recognize images, pictures,
sights and smells. After showing a suspect pictures of familiar
places and measuring his p300 activation patterns, government
officials could, at least in theory, show a suspect pictures of
places he may or may not have seen before a Qaeda training camp,
for example, or a crime scene and compare the activation patterns.
(By detecting not only lies but also honest cases of forgetfulness,
the technology could expand our very idea of lie detection.)

The second lie-detection technology uses f.M.R.I. machines to
compare the brain activity of liars and truth tellers. It is based
on a test called Guilty Knowledge, developed by Daniel Langleben at
the University of Pennsylvania in 2001. Langleben gave subjects a
playing card before they entered the magnet and told them to answer
no to a series of questions, including whether they had the card in
question. Langleben and his colleagues found that certain areas of
the brain lighted up when people lied.

Two companies, No Lie MRI and Cephos, are now competing to refine
f.M.R.I. lie-detection technology so that it can be admitted in
court and commercially marketed. I talked to Steven Laken, the
president of Cephos, which plans to begin selling its products this
year. We have two to three people who call every single week, he
told me. Theyre in legal proceedings throughout the world, and
theyre looking to bolster their credibility. Laken said the
technology could have tremendous applications in civil and criminal
cases. On the government side, he said, the technology could
replace highly inaccurate polygraphs in screening for security
clearances, as well as in trying to identify suspected terrorists
native languages and close associates. In lab studies, weve been in
the 80- to 90-percent-accuracy range, Laken says. This is similar
to the accuracy rate for polygraphs, which are not considered
sufficiently reliable to be allowed in most legal cases. Laken says
he hopes to reach the 90-percent- to 95-percent-accuracy range
which should be high enough to satisfy the Supreme Courts standards
for the admission of scientific evidence. Judy Illes, director of
Neuroethics at the Stanford Center for Biomedical Ethics, says, I
would predict that within five years, we will have technology that
is sufficiently reliable at getting at the binary question of
whether someone is lying that it may be utilized in certain legal
settings.

If and when lie-detection f.M.R.I.s are admitted in court, they
will raise vexing questions of self-incrimination and privacy. Hank
Greely, a law professor and head of the Stanford Center for Law and
the Biosciences, notes that prosecution and defense witnesses might
have their credibility questioned if they refused to take a
lie-detection f.M.R.I., as might parties and witnesses in civil
cases. Unless courts found the tests to be shocking invasions of
privacy, like stomach pumps, witnesses could even be compelled to
have their brains scanned. And equally vexing legal questions might
arise as neuroimaging technologies move beyond telling whether or
not someone is lying and begin to identify the actual content of
memories. Michael Gazzaniga, a professor of psychology at the
University of California, Santa Barbara, and author of The Ethical
Brain, notes that within 10 years, neuroscientists may be able to
show that there are neurological differences when people testify
about their own previous acts and when they testify to something
they saw. If you kill someone, you have a procedural memory of
that, whereas if Im standing and watch you kill somebody, thats an
episodic memory that uses a different part of the brain, he told
me. Even if witnesses dont have their brains scanned, neuroscience
may lead judges and jurors to conclude that certain kinds of
memories are more reliable than others because of the area of the
brain in which they are processed. Further into the future, and
closer to science fiction, lies the possibility of memory
downloading. One could even, just barely, imagine a technology that
might be able to read out the witnesss memories, intercepted as
neuronal firings, and translate it directly into voice, text or the
equivalent of a movie, Hank Greely writes.

Greely acknowledges that lie-detection and memory-retrieval
technologies like this could pose a serious challenge to our
freedom of thought, which is now defended largely by the First
Amendment protections for freedom of expression. Freedom of thought
has always been buttressed by the reality that you could only tell
what someone thought based on their behavior, he told me. This
technology holds out the possibility of looking through the skull
and seeing whats really happening, seeing the thoughts themselves.
According to Greely, this may challenge the principle that we
should be held accountable for what we do, not what we think. It
opens up for the first time the possibility of punishing people for
their thoughts rather than their actions, he says. One reason
thought has been free in the harshest dictatorships is that
dictators havent been able to detect it. He adds, Now they may be
able to, putting greater pressure on legal constraints against
government interference with freedom of thought.

In the future, neuroscience could also revolutionize the way jurors
are selected. Steven Laken, the president of Cephos, says that jury
consultants might seek to put prospective jurors in f.M.R.I.s. You
could give videotapes of the lawyers and witnesses to people when
theyre in the magnet and see what parts of their brains light up,
he says. A situation like this would raise vexing questions about
jurors prejudices and what makes for a fair trial. Recent
experiments have suggested that people who believe themselves to be
free of bias may harbor plenty of it all the same.

The experiments, conducted by Elizabeth Phelps, who teaches
psychology at New York University, combine brain scans with a
behavioral test known as the Implicit Association Test, or I.A.T.,
as well as physiological tests of the startle reflex. The I.A.T.
flashes pictures of black and white faces at you and asks you to
associate various adjectives with the faces. Repeated tests have
shown that white subjects take longer to respond when theyre asked
to associate black faces with positive adjectives and white faces
with negative adjectives than vice versa, and this is said to be an
implicit measure of unconscious racism. Phelps and her colleagues
added neurological evidence to this insight by scanning the brains
and testing the startle reflexes of white undergraduates at Yale
before they took the I.A.T. She found that the subjects who showed
the most unconscious bias on the I.A.T. also had the highest
activation in their amygdalas a center of threat perception when
unfamiliar black faces were flashed at them in the scanner. By
contrast, when subjects were shown pictures of familiar black and
white figures like Denzel Washington, Martin Luther King Jr. and
Conan OBrien there was no jump in amygdala activity.

The legal implications of the new experiments involving bias and
neuroscience are hotly disputed. Mahzarin R. Banaji, a psychology
professor at Harvard who helped to pioneer the I.A.T., has argued
that there may be a big gap between the concept of intentional bias
embedded in law and the reality of unconscious racism revealed by
science. When the gap is substantial, she and the U.C.L.A. law
professor Jerry Kang have argued, the law should be changed to
comport with science relaxing, for example, the current focus on
intentional discrimination and trying to root out unconscious bias
in the workplace with structural interventions, which critics say
may be tantamount to racial quotas. One legal scholar has cited
Phelpss work to argue for the elimination of peremptory challenges
to prospective jurors if most whites are unconsciously racist, the
argument goes, then any decision to strike a black juror must be
infected with racism. Much to her displeasure, Phelpss work has
been cited by a journalist to suggest that a white cop who
accidentally shot a black teenager on a Brooklyn rooftop in 2004
must have been responding to a hard-wired fear of unfamiliar black
faces a version of the amygdala made me do it.

Phelps herself says its crazy to link her work to cops who shoot on
the job and insists that it is too early to use her research in the
courtroom. Part of my discomfort is that we havent linked what we
see in the amygdala or any other region of the brain with an
activity outside the magnet that we would call racism, she told me.
We have no evidence whatsoever that activity in the brain is more
predictive of things we care about in the courtroom than the
behaviors themselves that we correlate with brain function. In
other words, just because you have a biased reaction to a
photograph doesnt mean youll act on those biases in the workplace.
Phelps is also concerned that jurors might be unduly influenced by
attention-grabbing pictures of brain scans. Frank Keil, a
psychologist at Yale, has done research suggesting that when you
have a picture of a mechanism, you have a tendency to overestimate
how much you understand the mechanism, she told me. Defense lawyers
confirm this phenomenon. Here was this nice color image we could
enlarge, that the medical expert could point to, Christopher
Plourd, a San Diego criminal defense lawyer, told The Los Angeles
Times in the early 1990s. It documented that this guy had a rotten
spot in his brain. The jury glommed onto that.

Other scholars are even sharper critics of efforts to use
scientific experiments about unconscious bias to transform the law.
I regard that as an extraordinary claim that you could screen
potential jurors or judges for bias; its mind-boggling, I was told
by Philip Tetlock, professor at the Haas School of Business at the
University of California at Berkley. Tetlock has argued that
split-second associations between images of African-Americans and
negative adjectives may reflect simple awareness of the social
reality that some groups are more disadvantaged than others. He has
also written that, according to psychologists, there is virtually
no published research showing a systematic link between racist
attitudes, overt or subconscious, and real-world discrimination. (A
few studies show, Tetlock acknowledges, that openly biased white
people sometimes sit closer to whites than blacks in experiments
that simulate job hiring and promotion.) A light bulb going off in
your brain means nothing unless its correlated with a particular
output, and the brain-scan stuff, heaven help us, we have barely
linked that with anything, agrees Tetlocks co-author, Amy Wax of
the University of Pennsylvania Law School. The claim that homeless
people light up your amygdala more and your frontal cortex less and
we can infer that you will systematically dehumanize homeless
people thats piffle.

V. Are You Responsible for What You Might Do? The attempt to link
unconscious bias to actual acts of discrimination may be dubious.
But are there other ways to look inside the brain and make
predictions about an individuals future behavior? And if so, should
those discoveries be employed to make us safer? Efforts to use
science to predict criminal behavior have a disreputable history.
In the 19th century, the Italian criminologist Cesare Lombroso
championed a theory of biological criminality, which held that
criminals could be identified by physical characteristics, like
large jaws or bushy eyebrows. Nevertheless, neuroscientists are
trying to find the factors in the brain associated with violence.
PET scans of convicted murderers were first studied in the late
1980s by Adrian Raine, a professor of psychology at the University
of Southern California; he found that their prefrontal cortexes,
areas associated with inhibition, had reduced glucose metabolism
and suggested that this might be responsible for their violent
behavior. In a later study, Raine found that subjects who received
a diagnosis of antisocial personality disorder, which correlates
with violent behavior, had 11 percent less gray matter in their
prefrontal cortexes than control groups of healthy subjects and
substance abusers. His current research uses f.M.R.I.s to study
moral decision-making in psychopaths.

Neuroscience, it seems, points two ways: it can absolve individuals
of responsibility for acts theyve committed, but it can also place
individuals in jeopardy for acts they havent committed but might
someday. This opens up a Pandoras box in civilized society that Im
willing to fight against, says Helen S. Mayberg, a professor of
psychiatry, behavioral sciences and neurology at Emory University
School of Medicine, who has testified against the admission of
neuroscience evidence in criminal trials. If you believe at the
time of trial that the picture informs us about what they were like
at the time of the crime, then the picture moves forward. You need
to be prepared for: This spot is a sign of future dangerousness,
when someone is up for parole. They have a scan, the spot is there,
so they dont get out. Its carved in your brain.

Other scholars see little wrong with using brain scans to predict
violent tendencies and sexual predilections as long as the scans
are used within limits. Its not necessarily the case that if
predictions work, you would say take that guy off the street and
throw away the key, says Hank Greely, the Stanford law professor.
You could require counseling, surveillance, G.P.S. transmitters or
warning the neighbors. None of these are necessarily benign, but
they beat the heck out of preventative detention. Greely has little
doubt that predictive technologies will be enlisted in the war on
terror perhaps in radical ways. Even with todays knowledge, I think
we can tell whether someone has a strong emotional reaction to
seeing things, and I can certainly imagine a friend-versus-foe
scanner. If you put everyone who reacts badly to an American flag
in a concentration camp or Guantánamo, that would be bad, but in an
occupation situation, to mark someone down for further
surveillance, that might be appropriate.

Paul Root Wolpe, who teaches social psychiatry and psychiatric
ethics at the University of Pennsylvania School of Medicine, says
he anticipates that neuroscience predictions will move beyond the
courtroom and will be used to make predictions about citizens in
all walks of life.

Will we use brain imaging to track kids in school because weve
discovered that certain brain function or morphology suggests
aptitude? he asks. I work for NASA, and imagine how helpful it
might be for NASA if it could scan your brain to discover whether
you have a good enough spatial sense to be a pilot. Wolpe says that
brain imaging might eventually be used to decide if someone is a
worthy foster or adoptive parent a history of major depression and
cocaine abuse can leave telltale signs on the brain, for example,
and future studies might find parts of the brain that correspond to
nurturing and caring.

The idea of holding people accountable for their predispositions
rather than their actions poses a challenge to one of the central
principles of Anglo-American jurisprudence: namely, that people are
responsible for their behavior, not their proclivities for what
they do, not what they think. Were going to have to make a decision
about the skull as a privacy domain, Wolpe says. Indeed, Wolpe
serves on the board of an organization called the Center for
Cognitive Liberty and Ethics, a group of neuroscientists, legal
scholars and privacy advocates dedicated to protecting and
advancing freedom of thought in the modern world of accelerating
neurotechnologies.

There may be similar cognitive liberty battles over efforts to
repair or enhance broken brains. A remarkable technique called
transcranial magnetic stimulation, for example, has been used to
stimulate or inhibit specific regions of the brain. It can
temporarily alter how we think and feel. Using T.M.S., Ernst Fehr
and Daria Knoch of the University of Zurich temporarily disrupted
each side of the dorsolateral prefrontal cortex in test subjects.
They asked their subjects to participate in an experiment that
economists call the ultimatum game. One person is given $20 and
told to divide it with a partner. If the partner rejects the
proposed amount as too low, neither person gets any money. Subjects
whose prefrontal cortexes were functioning properly tended to
reject offers of $4 or less: they would rather get no money than
accept an offer that struck them as insulting and unfair. But
subjects whose right prefrontal cortexes were suppressed by T.M.S.
tended to accept the $4 offer. Although the offer still struck them
as insulting, they were able to suppress their indignation and to
pursue the selfishly rational conclusion that a low offer is better
than nothing.

Some neuroscientists believe that T.M.S. may be used in the future
to enforce a vision of therapeutic justice, based on the idea that
defective brains can be cured. Maybe somewhere down the line, a
badly damaged brain would be viewed as something that can heal,
like a broken leg that needs to be repaired, the neurobiologist
Robert Sapolsky says, although he acknowledges that defining what
counts as a normal brain is politically and scientifically fraught.
Indeed, efforts to identify normal and abnormal brains have been
responsible for some of the darkest movements in the history of
science and technology, from phrenology to eugenics. How far are we
willing to go to use neurotechnology to change peoples brains we
consider disordered? Wolpe asks. We might find a part of the brain
that seems to be malfunctioning, like a discrete part of the brain
operative in violent or sexually predatory behavior, and then turn
off or inhibit that behavior using transcranial magnetic
stimulation. Even behaviors in the normal range might be fine-tuned
by T.M.S.: jurors, for example, could be made more emotional or
more deliberative with magnetic interventions. Mark George, an
adviser to the Cephos company and also director of the Medical
University of South Carolina Center for Advanced Imaging Research,
has submitted a patent application for a T.M.S. procedure that
supposedly suppresses the area of the brain involved in lying and
makes a person less capable of not telling the truth.

As the new technologies proliferate, even the neurolaw experts
themselves have only begun to think about the questions that lie
ahead. Can the police get a search warrant for someones brain?
Should the Fourth Amendment protect our minds in the same way that
it protects our houses? Can courts order tests of suspects memories
to determine whether they are gang members or police informers, or
would this violate the Fifth Amendments ban on compulsory
self-incrimination? Would punishing people for their thoughts
rather than for their actions violate the Eighth Amendments ban on
cruel and unusual punishment? However astonishing our machines may
become, they cannot tell us how to answer these perplexing
questions. We must instead look to our own powers of reasoning and
intuition, relatively primitive as they may be. As Stephen Morse
puts it, neuroscience itself can never identify the mysterious
point at which people should be excused from responsibility for
their actions because they are not able, in some sense, to control
themselves. That question, he suggests, is moral and ultimately
legal, and it must be answered not in laboratories but in
courtrooms and legislatures. In other words, we must answer it
ourselves.

Jeffrey Rosen, a frequent contributor, is the author most recently
of The Supreme Court: The Personalities and Rivalries That Defined
America.