Summary of Key Points
This article discusses how neuroscientist Keith attempted to use the theory that “brain scans reveal a lack of moral brakes in criminals” to argue for reduced sentences for murderers. However, subsequent studies, including those conducted by his own collaborators, found that this theory lacks solid empirical support. Nevertheless, such unproven neuroscientific or genetic evidence has been widely used in American criminal cases, even in death penalty decisions, highlighting the stark contradiction between the slow pace of scientific correction and the immediate need for legal judgments.
1. A Death Penalty Case Defended on the Basis of “Broken Brain Brakes”
In 2009, murderer Dugan pleaded guilty and was awaiting a sentence (death or life imprisonment). His defense attorney hired neuroscientist Keith to testify. Keith first used the Hare Scale to score Dugan 38.5 out of 40 (a score above 30 indicates psychopathy), and then conducted a brain scan. In normal individuals, the brain regions responsible for emotional responses activate when seeing morally offensive images (such as the burning of crosses by the Ku Klux Klan); however, Dugan’s corresponding brain region did not respond. Keith reasoned, “If his emotional processing ability is that of a 5-year-old, should he be exempt from the death penalty, just like someone with an intellectual disability?” (U.S. law prohibits the death penalty for individuals with intellectual disabilities.) Although the jury was moved by this argument, Dugan was still sentenced to death; the sentence was later commuted to life imprisonment after Illinois abolished the death penalty.
2. Keith’s “Brain Brake” Theory: From Hypothesis to “Court Evidence”
Keith proposed the “marginal system hypothesis,” suggesting that the brain regions responsible for moral emotions and impulse control are dysfunctional in psychopaths. To prove this, he brought a portable MRI machine into the prison and scanned over 3,000 prisoners, claiming to have the largest database of prisoner brain scans in the world. Initially, the academic community accepted his findings; a 2018 review praised his research for its large sample size and rigorous methodology, and the media sensationalized it as suggesting that brain scans could predict criminal behavior. However, there were underlying issues: his 2013 paper on predicting recidivism included too few violent cases for meaningful analysis, and even after adjustments, the predictive accuracy was significantly reduced. Moreover, his methods were not more effective than traditional factors such as criminal records and age.
3. The Collapse of the Theory: Self-Criticism and 20 Years of Lack of Solid Evidence
In 2024, Keith’s former collaborator Deming published a systematic review of 101 studies over 20 years, examining 1,573 instances. The review found that 85.4% of the studies failed to establish a consistent link between psychopathy and the target brain regions. The larger the sample size and the more rigorous the statistical analysis, the more inconclusive the results. Ironically, Keith’s own 2001 study was also included in the review, and all six tests it conducted were ineffective. A paper co-authored by Keith and Konigs also had to be retracted due to data errors. In short, the theory that “a broken brain brake leads to criminal behavior” was refuted by his own research.
4. Unproven Science Already Penetrating the Courts: Over 2,800 Cases Have Used Such Evidence
Despite the lack of evidence, neuroscientific findings have been widely applied in American courts:
- Keith and his wife founded the company Mindset, which provides brain scan and genetic testing services for defense attorneys and has been involved in over 200 death penalty cases;
- Between 2005 and 2015, more than 2,800 judicial opinions discussed neuroscientific evidence, and one-quarter of death penalty decisions relied on such arguments;
- Keith has even given lectures to “almost all federal judges” (although there are only 890 federal judges in total).
Opposition to these practices has persisted, with criticisms such as the fact that many people have similar brain scan results but did not commit crimes, and the fact that some neuroscientists themselves have “psychopathic brains” yet did not commit crimes (as seen in the case of Fallon).
5. The Contradiction Between Slow Scientific Correction and Rapid Legal Judgments
Science follows a process of “hypothesis → verification → revision” (for example, Keith’s theory was overturned after 20 years of research), but the legal system requires immediate decisions. This contradiction leads to situations like the case of Wiles, where genetic evidence was used to argue that “genetics combined with childhood abuse led to violence,” and the prosecution argued that he was always dangerous, resulting in a death sentence. The Supreme Court rejected his appeal, and he remains on death row. Genetic evidence has even become even less reliable: in 2016, the National Institute of Mental Health recommended abandoning such research, yet it is still used in court.
Science can correct its own mistakes, but death penalty decisions are final and irreversible. This highlights the cruel reality of relying on unproven scientific findings in legal proceedings.
This article essentially asks whether we should let unproven science determine human lives when it is not yet mature. The answer is clearly no, yet in practice, such “uncompleted science” has quietly made its way into the courtroom.