By Francis X. Shen
A core failing of the criminal justice system is its inability to individualize criminal sentences and tailor probation and parole to meet the unique profile of each offender.
As legal scholar, and now federal judge Stephanos Bibas has observed, “All too often … sentencing guidelines and statutes act as sledgehammers rather than scalpels.”
As a result, dangerous offenders may be released, while offenders who pose little risk to society are left behind bars. And recidivism is common — the U.S. has an astounding recidivism rate of 80% — in part because the current criminal justice system largely fails to address mental health challenges, which are heavily over-represented in the justice system.
Advances in computational psychiatry, such as the deep phenotyping methods explored in this symposium, offer clinicians newfound abilities to practice precision psychiatry. The idea behind precision psychiatry is both simple and elusive: treat individuals as individuals. Yet advancing such a program in practice is “very ambitious” because no two individual brains — and the experiences those brains have had over a lifetime — are the same.
Deep phenotyping offers the criminal justice system the tools to improve public safety, identify low-risk offenders, and modify decision-making to reduce recidivism. Computational psychiatry can lead to what can be described as precision sentencing.
For instance, all first time drunk driving offenders who do not have a criminal history will typically receive the same offer in a jurisdiction. For some of those first time offenders, the arrest may have been a one-time mistake. But for others, it may have simply been the first time they were caught, and the arrest should serve as an opportunity to take someone dangerous off the streets. Even when observed behavior is the same — for instance, two people commit the same criminal offense — why they behaved as they did may well be different.
The key to precision sentencing is better, more individualized data — and more powerful ways to analyze that data.
Consider the drunk driving example. Wearable devices that can monitor alcohol content in someone’s body are being researched with support by NIH, and already marketed for research use. An alcohol monitor that harnesses smart phone monitoring, actigraphy, and GPS location data would allow for identification of potentially harmful behavioral patterns. For instance, the data could reveal that 3 nights a week the offender is driving to the GPS location of a bar, staying there for 2 hours, experiencing a rising alcohol level, and then driving home. Such a pattern would be deeply troubling — and would be grounds for the state to intervene. Rather than wait for a second drunk driving incident, and the potential loss of life, the probation and parole departments could monitor and intervene in real time — before the harm occurs. Importantly, intervention might not necessarily involve criminal punishment, but rather preventative measures.
Incorporating this type of data would require a culture shift in prosecutors’ plea bargaining and judges’ sentencing. Sentencing and plea bargaining vary county to county, since each locality and judge has great discretion, but within a county — especially within large counties with high volumes of cases — plea bargaining is routine business. There is little precision sentencing and no use of quantitative data of the type I suggest here.
What’s needed is a culture change akin to the data and analytics revolution in sports. Today teams in every major sports league utilize individualized analytics. In the justice system, deep phenotyping offers previously unimaginable opportunities for similar analytics.
The use of analytics in the criminal justice system has been heavily criticized by those who fear that biased data and racist algorithms will lead to a new “Jim Code” and mass e-carceration. Surely the development and implementation of deep phenotyping tools in the criminal justice system must pay attention to these concerns, which we explored in a Petrie-Flom Center event in October 2019.
But, if developed with sensitivity and solutions to these concerns, there is great potential for deep phenotyping methods to facilitate a new era of precision sentencing in criminal justice. Precision sentencing will improve public safety, reduce inequity, and better address the mental health needs of those in the justice system. The technology is a means to reinvigorate a core value in medicine and in law: treating individuals as individuals.
This post is part of our Ethical, Legal, and Social Implications of Deep Phenotyping symposium. All contributions to the symposium are available here.