Brain Maturation and the Execution of Juveniles
Some reflections on science and the law.


Illustration ©David Hollenbach

By Ruben C. Gur | By Ruben C. Gur | Should the death penalty be applied to offenders who were “juvenile” but over the age of 16 when they committed their crimes? As I am writing, the U.S. Supreme Court is deliberating this issue.

The attorneys arguing against juvenile execution used evidence from research on brain development to claim that the juvenile brain is insufficiently mature in areas relevant to criminal culpability to warrant the ultimate punishment. This is how I was dragged out of my hiding in the laboratory and into the midst of legal deliberations, culminating in amicus briefs and conferences with glitzy legal teams.

The first hint of the storm came two years ago, with a call from a lawyer, Marc Bookman C’78, from the Defender Association of Philadelphia. He asked if I knew of or could prepare a review of the literature on brain development; he needed one to help a Pennsylvania man who was facing the death penalty for crimes committed as a juvenile. I happened to have reviewed this literature for a manuscript reporting our own data from longitudinal studies performed by Penn’s Brain Behavior Laboratory and the Schizophrenia Center, and so I agreed to augment it and make it more readable for non-experts. He sent it back to me, this time formatted as an affidavit in the case of a Mr. Hector Huertas. He asked whether I would mind reviewing the affidavit and, if I agreed with its contents, to notarize and sign it. Well, I agreed and it apparently worked. The Commonwealth decided not to pursue the death penalty in light of scientific findings that the brain does not mature until early adulthood.  Soon afterward Mr. Bookman called again, this time to help a colleague in Texas who was defending a Mr. Toronto Patterson, a death-row inmate who also committed his crime when he was an adolescent.

The affidavit did not save Mr. Patterson’s life, but while the U.S. Supreme Court refused to hear the appeal, it expressed interest in the scientific evidence from brain research that was presented in the case, and invited such arguments in future cases. The race was on to see what would be the test case determining whether the death penalty will apply to juvenile defendants, and I found myself being asked to sign affidavits from around the country. The “winner” was a Missouri man, Mr. Christopher Simmons (Roper v. Simmons). Below is a summary of the material I have submitted as part of an amicus organized by Mr. Simmons’ defense:

The rate at which the human brain matures has been of considerable interest to neuroscientists, and knowledge of when different brain regions mature in human development may have profound implications for understanding behavioral development. Although the brain and its structure become well differentiated during fetal development, there is overwhelming evidence that much of the maturational process occurs after birth. Indeed, projections from early pioneering work on donated brain tissue have indicated that some brain regions do not reach maturity in humans until adulthood. These projections have been confirmed by more recent neuroimaging studies.

While sophisticated methods for preservation and dissection of postmortem brain tissue had been developed in the first decades of the 20th century, it was not until the 1960s that enough such tissue was available to examine the question of brain maturation in humans. Arguably the largest collection and the most influential work was that of Dr. Paul I. Yakovlev and his colleagues at Harvard University. His work has focused on the  creation of myelin, fatty tissue surrounding nerve fibers. This process, known as myelogenesis, is important for assuring efficient transmission of neuronal signals; myelin surrounds the nerve fibers that carry information across large distances very much in the same way that rubber is used for insulating cables designed to conduct electricity across distance.

Yakovlev examined slices of brain tissue from a wide age range of more than 200 brains, finding that especially late to myelinate were those parts of the brain that inhibit and modulate the more primitive, drive-related activation of the limbic areas. As interpreted by Yakovlev and his colleagues, the anatomic data indicated that the very functions that make us uniquely human are the latest to become fully integrated into the workings of the developing brain.

University of Chicago researcher Peter Huttenlocher uncovered another neurodevelopmental phenomenon apparently taking place during adolescence: pruning.  According to the pruning hypothesis, neurons and their connections that have not been consistently used during childhood “shrivel off” and are eliminated at some point during adolescence, thereby allowing for greater efficiency of the remaining neural systems.

Postmortem tissue studies have contributed important insights into understanding brain maturation, but they have serious limitations, including tissue availability and the inability to trace developmental changes in the same individual.

These difficulties are circumvented by a set of novel techniques—developed in the 1970s and fully implemented by the 1990s—that can be generally referred to as structural imaging. These methods permit visualization and volumetric measurement of brain structure in living people without any risk to the subjects. The method that has become state-of-the-art for these studies is based on magnetic resonance imaging (MRI) procedures. MRI has provided data on the composition of three brain components, or compartments:  gray matter—nerve tissue responsible for information processing; white matter—nerve tissue responsible for information transmission; and cerebrospinal fluid. This division of the brain into these compartments is termed segmentation.

In one of the first studies examining segmented MRI in children and adults, researchers Terry Jernigan and Paula Tallal from the University of California have documented the pruning process. They found that children had higher volumes of gray matter than adults, indicating loss of gray matter during adolescence. In another study Stanford researchers Adolph Pfefferbaum and Calvin Lim demonstrated a clearly different developmental course for gray matter and white matter: The former declined steadily during adolescence while the latter increased in volume until about 20-22 years of age. A subsequent NIH study led by Judith Rappoport pinpointed the greatest delay in myelination for the brain’s fronto-temporal pathways.

In the only study to date that examined segmented MRI volumes from a prospective sample of 28 healthy children aged one month to 10 years, as well as a small adult sample, researchers from Penn and Toyama University in Japan applied segmentation procedures developed by the Penn group. This, actually, was the very study that prompted me to write a review of the literature, which I used for Mr. Bookman’s request. We found that while gray matter volume peaked at about two years of age, the volume of white matter, which indicates brain maturation, continued to increase into adulthood. Furthermore, we found that the frontal lobe showed the greatest maturational lag and its myelination is unlikely to be completed before young adulthood.

Most recently, investigators at UCLA’s brain imaging center analyzed MRI scans of 13 healthy children over a period of eight to 10 years. They concluded that brain areas in the cerebral cortex, responsible for higher-order integration, mature only after lower-order somatosensory and visual cortices are developed.

My review of the data found overwhelming evidence indicating that the main index of maturation, which is the process called myelination, is not complete until some time in the beginning of the third decade of life (probably at around ages 20-22). Other maturational processes, such as the increase and subsequent elimination (“pruning”) in cell number and connectivity, may be completed by late adolescence, perhaps by ages 15-17. (More data are needed to know for sure.) These results have rather profound implications for understanding behavioral development. The cortical regions that are last to mature, particularly those in prefrontal areas, are involved in behavioral facets germane to many aspects of criminal culpability. Perhaps most relevant is the involvement of these brain regions in the control of aggression and other impulses, the process of planning for long-range goals, organization of sequential behavior, the process of abstraction and mental flexibility, and aspects of memory including “working memory.” If the neural substrates of these behaviors have not reached maturity before adulthood, it is unreasonable to expect the behaviors themselves to reflect mature thought processes.

As I stated in my expert opinion for the court, the brain-scan techniques have demonstrated conclusively that the phenomena observed by mental-health professionals in persons under 18, which would render them less morally blameworthy for offenses, have a scientific grounding in neural substrates.  

The evidence now is strong that the brain does not cease to mature until the early 20s in those relevant parts that govern impulsivity, judgment, planning for the future, foresight of consequences, and other characteristics that make people morally culpable. Therefore, from the perspective of neural development, someone under 20 should be considered to have an underdeveloped brain.  Additionally, since brain development in the relevant areas goes in phases that vary in rate and is usually not complete before the early to mid-20s, there is no way to state with any scientific reliability that an individual 17-year-old has a fully matured brain (and should be eligible for the most severe punishment), no matter how many otherwise accurate tests and measures might be applied to him at the time of his trial for capital murder.  This is similar to other physical characteristics such as height. While we know the age at which the average adult reaches his or her maximal height, predictions for individuals are not easy to make. Thus, although 18 is an arbitrary cutoff, given the ongoing development of the brain in most individuals, it must be preferred over 17 to assure that only the most culpable are punished for capital crimes. Indeed, age 21 or 22 would be closer to the “biological” age of maturity.

Dr. Ruben Gur is a professor of psychology in the Department of Psychiatry (with secondary appointments in Radiology and Neurology) and director of the Brain Behavior Laboratory in the School of Medicine.


©2005 The Pennsylvania Gazette
Last modified 01/05/05



FIRST PERSON: Essays

Alumni Voices: The old neighborhood
Elsewhere: “This really makes me want to have a baby”
Expert Opinion: When do brains mature?

Jan | Feb Contents
Gazette Home
Previous issue's column