Decades of research from the fields of criminology and adolescent brain science find that the decisions made in youth — even very unwise decisions — do not crystallize criminality. Instead, as young people age and mature they develop the capacity to make different choices.

Fortunately, more informed policies have begun to replace the punishments that proliferated during the 1990s when fear of “superpredators” and calls for “adult time, adult crime” dominated youth crime policy. The incarceration rate for youth in juvenile and adult systems soared then. Today, youth incarceration in juvenile facilities is now half its level of 20 years ago. Driven by the work of advocates, legislators, researchers and judges, many now readily acknowledge that the juvenile justice system should be used sparingly and only for those who truly need confinement.

Based mostly on the adolescent brain science discoveries that concluded that executive functioning is not fully developed until adulthood, in 2005 the Supreme Court ruled in Roper v. Simmons that the death penalty for juveniles violated the 8th Amendment. The Supreme Court has taken up the issue of severe sanctions for juveniles three more times since then, and as a result there are restrictions in the use of life without the possibility of parole (LWOP) for juveniles.

Changing public sentiment regarding the wisdom of sending young people to adult prisons has led policymakers in many states to revise misguided policies in this area. Many juveniles who would otherwise be languishing in adult prisons are now either in juvenile confinement facilities that are better designed for their needs or have been diverted from confinement altogether. In 2015, the number of juveniles held in adult prisons declined to less than 1,000, an 82 percent drop from the peak year in 1997.

These reforms have not resulted in any adverse public safety outcomes. Crime continues to be at historic lows.

Despite these advances, more than 9,000 people who were convicted as youth still do not stand to benefit from reforms either in the juvenile system or of the restrictions on punishments in the adult system. The Sentencing Project recently released a national study on the prevalence of life sentences nationwide, disaggregated by crime of conviction, race and ethnicity, gender and juvenile status. We obtained data from the states and federal Bureau of Prisons on the number of people serving three categories of life sentences: life with the possibility of parole (LWP), LWOP and virtual or de facto life sentences that amount to terms of 50 years or more. We learned that 1 in 7 prisoners is serving one of these sentences and that nearly 6 percent of the lifer population was under 18 at the time of the crime.

Aside from the roughly 2,300 individuals serving JLWOP there are approximately 7,000 juveniles who are serving parole-eligible life sentences around the country. For them, a statutory mandate or judicial decision has determined that spending the rest of their life in prison is reasonable if parole is not granted sooner. In New York, Georgia and Texas, more than 600 people sentenced for crimes in their youth have parolable life sentences. In California, which leads the nation in the category of life sentences, a notable 2,700 individuals are serving parole-eligible life for a crime committed under age 18.

In addition to the 7,000 juveniles serving life with parole, nationwide 2,000 individuals are serving de facto life sentences of 50 years or more for crimes committed when they were less than 18 years old. Louisiana reports 600 virtual lifers who were juveniles when their crime occurred and Texas reports nearly 450 such individuals.

The crimes committed by these juveniles were typically serious: 82 percent of lifers with the possibility of parole have been convicted of a homicide, and for half of these the crime was a first-degree murder. Among those serving de facto life for crimes committed as juveniles, 56 percent were convicted of a homicide and 94 percent were convicted of violent crime, including 22 percent for aggravated assault.

As with all life sentences, racial disproportionality is evident. African-American youth comprise more than half (53 percent) of the parolable and virtual life sentences, slightly less than their composition among the JLWOP population (63 percent). Overall, youth of color make up 81 percent of those serving life and virtual life sentences.

Some states stand out in the proportion of life and virtual life sentences being served by those who were young at the time of the crime. In Wisconsin, for instance, more than 11 percent of the life-sentenced population was a juvenile at the time of the crime. And while a first opportunity for parole comes after “only” 20 years in Wisconsin, we know from mounting research in parole politics and practices that rates of granting parole have fallen, particularly for those convicted of serious crimes and serving lengthy sentences. In Georgia, which in 2016 reported 600 people serving parole-eligible life sentences for crimes committed in their youth, the first opportunity for parole does not occur for 25 to 30 years. In Tennessee, the first parole hearing occurs only after a minimum of 51 years.

The requirement set forth in Graham v. Florida of a “meaningful opportunity for release based on demonstrated maturity and rehabilitation” has shined a bright light on parole systems’ capacity and willingness to afford a second look, and when this should occur. In June, the Supreme Court ruled in Virginia et al. v. Dennis LeBlanc (582 US ___[2017]) that Virginia’s “geriatric release” mechanism provided a sufficiently meaningful opportunity for release. Geriatric release allows review for those 60 years old and older; this means that LeBLanc and other people in his position will wait more than 40 years before being considered for release.

This and other lengthy terms of imprisonment stand to violate the spirit if not the letter of the recent court rulings. They also conflict with recent recommendations of The American Law Institute, a respected body of legal scholars and law practitioners that proposes a review after 10 years for any juveniles sentenced to terms longer than this.

Revised state laws for sentencing juveniles are being developed in the aftermath of the Supreme Court rulings. It is important to include in these considerations all youth with life and lengthy terms; their potential for reform and maturation is just as likely as for those sentenced to life without parole.

This article originally appeared on Juvenile Justice Information Exchange, the only publication covering juvenile justice and related issues nationally on a consistent, daily basis.