ANRV372-CP05-20 ARI 19 February 2009 11:37

Adolescent Developmentand Juvenile JusticeLaurence SteinbergDepartment of Psychology, Temple University, Philadelphia, Pennsylvania 19122;email: lds@temple.edu

Annu. Rev. Clin. Psychol. 2009. 5:459–85

The Annual Review of Clinical Psychology is onlineat clinpsy.annualreviews.org

This article’s doi:10.1146/annurev.clinpsy.032408.153603

Copyright c© 2009 by Annual Reviews.All rights reserved

1548-5943/09/0427-0459$20.00

Key Words

adolescence, crime, neuroscience, law, policy

AbstractAlthough justice system policy and practice cannot, and should not, bedictated solely by studies of adolescent development, the ways in whichwe respond to juvenile offending should be informed by the lessons ofdevelopmental science. This review begins with a brief overview of thehistory, rationale, and workings of the American juvenile justice system.Following this, I summarize findings from studies of brain, cognitive,and psychosocial development in adolescence that have implications forthe treatment of juveniles in the justice system. The utility of develop-mental science in this context is illustrated by the application of theseresearch findings to three fundamental issues in contemporary justicepolicy: the criminal culpability of adolescents, adolescents’ competenceto stand trial, and the impact of punitive sanctions on adolescents’ de-velopment and behavior. Taken together, the lessons of developmentalscience offer strong support for the maintenance of a separate juvenilejustice system in which adolescents are judged, tried, and sanctioned indevelopmentally appropriate ways.

459

Ann

u. R

ev. C

lin. P

sych

ol. 2

009.

5:45

9-48

5. D

ownl

oade

d fr

om w

ww

.ann

ualr

evie

ws.

org

by M

assa

chus

etts

Sch

ool o

f Pr

ofes

sion

al P

sych

olog

y (M

SPP)

on

09/0

8/14

. For

per

sona

l use

onl

y.

ANRV372-CP05-20 ARI 19 February 2009 11:37

Contents

INTRODUCTION . . . . . . . . . . . . . . . . . . 460JUVENILE JUSTICE IN AMERICA:

AN OVERVIEW . . . . . . . . . . . . . . . . . . 461The Origins of the Juvenile

Justice System . . . . . . . . . . . . . . . . . . 461Critical Decision Points Along

the Juvenile Justice Pipeline . . . . . 462The Relevance of Developmental

Science to Decision Makingin the Justice System . . . . . . . . . . . . 464

BRAIN, COGNITIVE, ANDPSYCHOSOCIALDEVELOPMENT INADOLESCENCE . . . . . . . . . . . . . . . . . 465Adolescent Brain Development . . . . . 465Adolescent Cognitive

Development . . . . . . . . . . . . . . . . . . . 467Adolescent Psychosocial

Development . . . . . . . . . . . . . . . . . . . 468JUVENILE JUSTICE ISSUES

INFORMED BYDEVELOPMENTALSCIENCE . . . . . . . . . . . . . . . . . . . . . . . . 471Criminal Culpability of Youth . . . . . . 471Competence of Adolescents

to Stand Trial . . . . . . . . . . . . . . . . . . . 473Impact of Punitive Sanctions on

Adolescent Developmentand Behavior . . . . . . . . . . . . . . . . . . . 477

SUMMARY AND CONCLUDINGCOMMENTS. . . . . . . . . . . . . . . . . . . . . 480

INTRODUCTION

Few issues challenge a society’s ideas aboutboth the nature of human development andthe nature of justice as much as serious juvenilecrime. Because we neither expect children tobe criminals nor expect crimes to be commit-ted by children, the unexpected intersectionbetween childhood and criminality createsa dilemma that most people find difficult toresolve. Indeed, the only ways out of thisproblem are either to redefine the offenseas something less serious than a crime or to

redefine the offender as someone who is notreally a child (Zimring 1998).

For most of the twentieth century, Americansociety has most often chosen the firstapproach—redefining the offense—and hastreated most juvenile infractions as matters tobe adjudicated as delinquent acts within a sep-arate juvenile justice system designed, at leastin theory, to recognize the special needs andimmature status of young people and to there-fore emphasize rehabilitation over punishment.Indeed, for much of the past century, statesbelieved that the juvenile justice system was avehicle to protect the public by providing a sys-tem that responds to children who are maturinginto adulthood. States recognized that conductalone—that is, the alleged criminal act—shouldnot be dispositive in deciding when to invokethe heavy hand of the adult criminal justice sys-tem. They recognized that by providing for ac-countability, treatment, and supervision in thejuvenile justice system—and in the communitywhenever possible—they promoted short-termand long-term public safety.

During the last two decades of the twenti-eth century, there was a dramatic shift in theway juvenile crime was viewed by policy mak-ers and the public. Rather than choosing to de-fine offenses committed by youth as delinquent,society increasingly opted to deal with youngoffenders more punitively in the juvenile jus-tice system or to redefine them as adults andtry them in adult criminal court. This trendwas reflected in the growing number of juve-nile offenses adjudicated in adult criminal court,where adolescents are exposed to a far more ad-versarial proceeding than in juvenile court; inthe increasingly punitive response of the crimi-nal justice system to juvenile offenders who arefound guilty; and in what some observers havereferred to as the “criminalization” of the juve-nile justice system itself through increased useof punishment, rather than rehabilitation, as alegitimate juvenile justice goal (Feld 1993).

This transformation of juvenile justice pol-icy and practice raises difficult, but important,questions for psychologists interested in thedevelopment and well-being of young people.

460 Steinberg

Ann

u. R

ev. C

lin. P

sych

ol. 2

009.

5:45

9-48

5. D

ownl

oade

d fr

om w

ww

.ann

ualr

evie

ws.

org

by M

assa

chus

etts

Sch

ool o

f Pr

ofes

sion

al P

sych

olog

y (M

SPP)

on

09/0

8/14

. For

per

sona

l use

onl

y.

ANRV372-CP05-20 ARI 19 February 2009 11:37

These questions are variations of the moregeneral question of whether adolescents arefundamentally different from adults in ways thatwarrant the differential treatment of juvenileswho break the law. In particular:

� Do adolescents have the psychological ca-pabilities necessary to function as compe-tent defendants in adult court?

� Should juveniles accused of crimes beheld to the same standards of blamewor-thiness as adults and punished in the sameways as adult criminals who have commit-ted similar crimes?

� How does exposing juveniles to especiallypunitive sanctions affect their behavior,development, and mental health?

These questions provide this review’s focus.More broadly, the purpose of this review is tointegrate developmental psychological consid-erations into moral, legal, political, and practi-cal analyses of juvenile crime. Because address-ing this issue necessitates at least a rudimentaryunderstanding of the rationale and workings ofthe juvenile justice system, I begin not with adiscussion of the science of adolescent develop-ment, but rather with a short history of juvenilejustice in America and a brief overview of theprocess through which individuals are adjudi-cated within the system.

Following this brief introduction to Ameri-can juvenile justice, I then summarize findingsfrom recent studies of adolescent developmentthat bear on whether adolescents differ fromadults in ways that have implications for jus-tice system policy and practice. Because not allaspects of adolescent development are perti-nent to how young people are, or should be,treated in the justice system, I limit my discus-sion to studies that are especially relevant tothese issues. Readers interested in a broader andmore comprehensive treatment of adolescentdevelopment are encouraged to consult sev-eral recent reviews of this literature (Collins &Steinberg 2006, Smetana et al. 2006) as well as arecently updated handbook on adolescent psy-chology (Lerner & Steinberg 2009). I then lookspecifically at what we know about adolescents’

Competence to standtrial: the ability of adefendant tounderstand the courtproceeding, reasonwith relevant facts, andassist counsel

Criminal culpability:the extent to which anindividual is judged tobe responsible for acrime

competence to stand trial, criminal culpability,and response to various types of sanctions andinterventions.

JUVENILE JUSTICE IN AMERICA:AN OVERVIEW

The Origins of the JuvenileJustice System

Economic recessions in the early nineteenthcentury pushed children out of work inAmerica’s new factory system during the indus-trial revolution. Concerns about poor childrenon the street led to the creation of institutionalcare for children. In New York City, the Societyfor Prevention of Pauperism in 1824 becamethe Society for the Reformation of JuvenileDelinquents, and in 1825 opened the nation’sfirst House of Refuge. Boston followed a yearlater and Philadelphia in 1828. These Houses ofRefuge were designed to maintain class statusand prevent unrest (Krisberg & Austin 1993,Platt 1977).

In 1899, Jane Addams and her Hull Housecolleagues established what is generally ac-cepted as the nation’s first juvenile court.Juvenile court judges, in the early part of thetwentieth century, were authorized to inves-tigate the character and social background ofboth predelinquent and delinquent children.They examined personal motivation as well ascriminal intent, seeking to identify the moralreputation of problematic children (Platt 1977).Ben Lindsey, of Denver, was the juvenile courtjudge whose practice most closely matched therhetoric of the emerging juvenile court:

We should make it our business to study andknow each particular case, because it will gen-erally demand treatment in some little respectdifferent from any other case. . . . (a) Is thechild simply mischievous or criminal in itstendencies? (b) Is the case simply an excep-tional or isolated instance in which a reallygood boy or girl has gone wrong for the firsttime because too weak to resist a strong temp-tation? (c) Is the child a victim of incompetent

www.annualreviews.org • Juvenile Justice 461

Ann

u. R

ev. C

lin. P

sych

ol. 2

009.

5:45

9-48

5. D

ownl

oade

d fr

om w

ww

.ann

ualr

evie

ws.

org

by M

assa

chus

etts

Sch

ool o

f Pr

ofes

sion

al P

sych

olog

y (M

SPP)

on

09/0

8/14

. For

per

sona

l use

onl

y.

ANRV372-CP05-20 ARI 19 February 2009 11:37

Transfer: onemechanism throughwhich juveniles’ casesare referred tocriminal (adult) court

Disposition: in thejuvenile justice system,the outcome of anadjudication;comparable to asentence in criminalcourt

parents? Does the home or parent need cor-rection or assistance? (d) What of environ-ment and association, which, of course, mayembrace substantively all of the points ofstudy? How can the environment be im-proved? Certainly by keeping the child outof the saloon and away from evil examples.(e) Is the child afflicted with what we call “themoving about fever” – that is, is he given toplaying “hookey” from school, or “bumming”and running away, showing an entire lack ofambition or desire to work and settle down toregular habits? [Ben B. Lindsey, “The Boy andthe Court,” Charities 13 ( January 1905):352;cited in Platt (1977)]

Julian Mack, Chicago’s second juvenile courtjudge, similarly described the ideal juvenilecourt proceeding:

The problem for determination by the judge isnot Has this boy or girl committed a specificwrong but What is he, how has he becomewhat he is, and what had best be done in hisinterest and in the interest of the state to savehim from a downward career. It is apparentat once that the ordinary legal evidence in acriminal court is not the sort of evidence to beheard in such a proceeding. (Mack 1909)

It is beyond the scope of this article to dis-cuss the likely causes of the transformation ofthe juvenile justice system away from the re-habilitative ideal espoused by its founders andtoward the more punitive regime that existstoday (but see Scott & Steinberg 2008 for adiscussion). However, it is worth noting thatthe early rhetoric on the rationale and purposeof the juvenile court is significant in two waysthat bear on contemporary debates about jus-tice system policy and practice. First, it is clearthat the founders of the juvenile justice systembegan from the premise that adolescents aredevelopmentally different from adults in waysthat should affect our interpretation and as-sessment of their criminal acts. The questionsraised by Judges Lindsey and Mack are relevantto the most vexing challenges that practition-

ers face today in determining (a) whether anadolescent’s antisocial behavior is due to tran-sient immaturity or contextual disadvantage, asopposed to deep-seated criminal character and(b) how best to construct a response to a juve-nile’s delinquent or criminal acts that will de-crease the likelihood of recidivism. The differ-ence between now and then, however, is thatat the time of the court’s founding, there wasno science available to inform consideration ofeither issue. Owing to the dramatic increase inempirical research on normative and nonnor-mative adolescent development that began inthe late 1970s, there has been a remarkable ex-pansion of the scientific knowledge relevant toeach of these matters.

Critical Decision Points Alongthe Juvenile Justice Pipeline

Juvenile justice is regulated mainly by state law,which makes it difficult to generalize about thesystem in ways that apply universally. Despitewhatever differences exist across jurisdictionsin policies and practices, however, the pointsof decision are essentially similar: referral, in-take, detention, transfer, adjudication, dispo-sition, and release (see Steinberg & Schwartz2000).

Referral. Entrance into the pipeline beginswith a referral to the juvenile justice system or apolice arrest. Depending upon the state, a childmay be too young or too old for the juvenilejustice system. Children who are too young aremost often diverted from the system or sent tothe branch of juvenile court that has jurisdictionover neglected and abused children. Childrenwho are too old are tried as adults. A juvenilemay also be charged with an offense that resultsautomatically in adult prosecution. If the juve-nile is charged as an adult, most states allow forjudges, after a hearing, to decide that the caseshould be transferred to juvenile court if thepublic interest requires it, or if the juvenile canprove that he or she is amenable to treatmentin the juvenile justice system.

462 Steinberg

Ann

u. R

ev. C

lin. P

sych

ol. 2

009.

5:45

9-48

5. D

ownl

oade

d fr

om w

ww

.ann

ualr

evie

ws.

org

by M

assa

chus

etts

Sch

ool o

f Pr

ofes

sion

al P

sych

olog

y (M

SPP)

on

09/0

8/14

. For

per

sona

l use

onl

y.

ANRV372-CP05-20 ARI 19 February 2009 11:37

Intake. If the child enters the juvenile justicesystem after being arrested, referred by a privatepetitioner (such as a school or next-door neigh-bor), or transferred from criminal court, therewill be an intake decision. Should the case pro-ceed, or should the juvenile be diverted? If thelatter, should it be an informal diversion, with-out further involvement by the juvenile court,or should the child be sent to a program, suchas a community panel or teen court (and re-turned to juvenile court if he or she fails to obeya community-ordered disposition)? Some casesare diverted to other systems, such as the mentalhealth system. Some cases are dropped entirelybecause intake officers decide that this particu-lar combination of youth and offense does notbelong in the juvenile justice system. Many fac-tors thus enter into the decision to divert a case:The youth’s age, prior history, the seriousnessof the offense, and the youth’s explanation orattitude will affect the intake decision.

Detention. If the intake officer decides thatthe case should proceed to a hearing, the officermust decide whether the child should be senthome (with or without supervision) or shouldbe detained, either in a maximum-security de-tention center or in a detention alternative.Juveniles and their parents will need to explainto an intake officer how pretrial supervision willoccur, and they will have to convince the offi-cer that the juvenile will appear for trial. If thechild is detained, there will be a court appear-ance within 24–72 hours. Most states call thisfirst court appearance a detention hearing. Herea judge or referee will decide whether to con-tinue the detention status. This is usually thefirst time that the child meets his or her attor-ney. Here the child must be able to discuss withcounsel the circumstances of the arrest and out-of-court issues related to the detention decision(such as school attendance or the presence of aninterested adult in the juvenile’s life).

Transfer. Most persons under the age of 18who are tried as adults are done so because ofstatutory exclusion of their case from the juve-nile justice system. State law may exclude them

because of their age—in New York, for exam-ple, a 16-year-old will be tried as an adult forany offense. Every state excludes some offensesfrom juvenile court jurisdiction if a child is of acertain age (for example, a state can decide that15-year-olds who are charged with armed rob-bery will have their cases begin in adult crim-inal court). Some states permit prosecutors tofile the juvenile’s case directly in the adult sys-tem, where the juvenile may or may not havean opportunity to have the case transferred tojuvenile court. Every state also allows judges totransfer children of a certain age—usually 14,but in some instances, even younger—to crim-inal court if they are charged with an offenseas serious as a felony. States usually must provethat the juvenile is “not amenable to treatment”in the juvenile justice system. At transfer hear-ings, it is important that the juvenile is able, forexample, to discuss with counsel his or her re-cent placement history and its reason for failure.He or she should be able to understand options,such as proposed placements, counseling pro-grams, or plea agreements.

Adjudication. If the child continues to be de-tained within the juvenile justice system, an ad-judicatory hearing (comparable to the trial incriminal court) must be held within 10–30 days.(Although this is the general rule, in some statesjuveniles charged with high-profile crimes suchas murder will have a longer time to wait untiltheir trials.) Demands on juveniles at adjudica-tory hearings are many. They will include theneed to understand the nature of the chargesagainst them and to consult with counsel. Theywill have to weigh the costs and benefits of en-tering an admission (guilty plea). They shouldbe able to help counsel identify potential wit-nesses, know whether an alibi or other defensesare available, and consult with counsel duringcross-examination of state witnesses.

Disposition. If the juvenile admits to the of-fense, or if the juvenile court finds by proofbeyond a reasonable doubt that the child hascommitted the offense, the court will pro-ceed to disposition (sentence). Juveniles are

www.annualreviews.org • Juvenile Justice 463

Ann

u. R

ev. C

lin. P

sych

ol. 2

009.

5:45

9-48

5. D

ownl

oade

d fr

om w

ww

.ann

ualr

evie

ws.

org

by M

assa

chus

etts

Sch

ool o

f Pr

ofes

sion

al P

sych

olog

y (M

SPP)

on

09/0

8/14

. For

per

sona

l use

onl

y.

ANRV372-CP05-20 ARI 19 February 2009 11:37

expected to assist counsel in presenting dispo-sition options to the juvenile court. Assistancemight include suggesting dispositions or help-ing the attorney and experts develop client-specific dispositions. Juvenile dispositionshistorically have been aimed at providing treat-ment, rehabilitation, or supervision in a waythat best serves the needs of the juvenile, al-though in recent years some legislatures alsohave included incapacitation for public safety asa valid rationale. Under any of the models, thejuvenile court will have a range of discretion. Insome states, the juvenile court has wide latitude,from ordering that a child return home un-der supervision (probation) to placing a child inmaximum-security institutions, known as train-ing schools, reform schools, or youth develop-ment centers. In other states, which use a “youthauthority” model, the court will either orderprobation or, if placement is warranted, trans-fer custody of the child to the youth authority,which will then determine the appropriate levelof care.

Release. Most juvenile court dispositions arefor indeterminate periods of time. However,dispositions cannot be for a longer period thanan adult would serve for a similar crime inthe criminal justice system. The court willusually review the juvenile’s case every six tonine months. Sometimes the reviews are for-mal hearings, whereas in other instances theyare informal reviews of reports provided byprobation officers or institutional staff. Manyjuveniles in placement, particularly those withmental health needs or who have been placedin inappropriate placements, end up being re-turned to juvenile court for a new disposition.Most often, those juveniles are placed in de-tention pending a new placement plan. Whenjuveniles are released from institutions, they areplaced on aftercare probation, which is analo-gous to parole. A juvenile who is on probationor on aftercare probation status can have thatstatus revoked, or “violated,” for new offensesor for violating the terms of probation, suchas associating with gang members, truancy, ormissing curfew. A violation of probation may

lead to rearrest, detention, and another hearing,the outcome of which may be a new disposition.

The Relevance of DevelopmentalScience to Decision Makingin the Justice System

Although there are few decision points in thepipeline where the developmental status of thejuvenile is taken into account explicitly, at eachdecision juncture, information about the juve-nile’s stage of development should play an im-portant role in the outcome of the decision. Ajuvenile’s developmental status is relevant withrespect to the adjudication process because ajust and fair hearing requires the competentparticipation of the individual in his or her de-fense. As noted earlier, at both the adjudicationand transfer hearings, certain competencies areexpected to be in place, including those thatpotentially affect the juvenile’s ability to under-stand the charges, assist counsel, and enter pleas(Scott & Grisso 2005). To the extent that thesecompetencies are based on capabilities that de-velop over the course of childhood and adoles-cence, an accurate understanding of how andalong what timetable these capabilities developis crucial to deciding whether an individual pos-sesses the skills necessary to participate in theprocess.

Under the law, characteristics of the of-fender and the circumstances of the offensecan mitigate criminal responsibility and lessenthe punishment that is ordered by the court.A crime that is committed impulsively is pun-ished less severely than one that is premed-itated, as is a crime that is committed un-der coercive pressure from others. Familiaritywith the expected developmental timetables ofphenomena such as self-control, foresight, andsusceptibility to peer pressure is therefore im-portant for making determinations of culpabil-ity. In theory at least, an offender who, by virtueof developmental immaturity, is impulsive,shortsighted, and easily influenced by peersshould be punished less harshly than one who isbetter able to control himself, anticipate the fu-ture consequences of his behavior, and resist the

464 Steinberg

Ann

u. R

ev. C

lin. P

sych

ol. 2

009.

5:45

9-48

5. D

ownl

oade

d fr

om w

ww

.ann

ualr

evie

ws.

org

by M

assa

chus

etts

Sch

ool o

f Pr

ofes

sion

al P

sych

olog

y (M

SPP)

on

09/0

8/14

. For

per

sona

l use

onl

y.

ANRV372-CP05-20 ARI 19 February 2009 11:37

antisocial urgings of his friends (Steinberg &Scott 2003).

Finally, decision makers in the system of-ten must assess the youngster’s potential forchange and risk for future offending when mak-ing transfer or disposition decisions (Mulvey &Leistico 2008). Such determinations of devel-opmental plasticity are especially important attransfer hearings, because a youngster who isor seems hardened and unlikely to profit fromrehabilitation is more likely to be charged asan adult than is one who is or is seen as mal-leable and amenable to intervention. Similarly,a juvenile who is deemed to be at high risk ofrecidivism, either because of a long prior recordof offending or other characteristics associatedwith continued and/or dangerous criminal be-havior (e.g., failure to respond to prior attemptsat rehabilitation, a history of uncontrollable vi-olence, or likelihood of inadequate adult super-vision in the community), will be more likely tobe sent to institutional placement.

In order to make well-informed decisionsabout the treatment of juveniles who have en-tered the juvenile justice pipeline, therefore,policy makers, practitioners, and mental healthprofessionals need to be familiar with the devel-opmental changes that occur during childhoodand adolescence in the capabilities and charac-teristics that are relevant to competence, culpa-bility, and likely response to treatment. Legis-lators need this information in order to createage-related laws and statutes that are develop-mentally appropriate and scientifically reason-able; if, for example, we know that the abilityto understand charges or enter pleas does notgenerally develop until a certain age, it makeslittle sense to draw age boundaries that wouldsubject developmentally incompetent individ-uals to court proceedings that necessitate theirparticipation in order to satisfy ordinary dueprocess requirements. Judges need this infor-mation in order to make wise and fair decisionsin the courtroom; if we know that the capac-ity to regulate one’s own behavior is unlikelyto be present before a certain age, it is impor-tant that this information be taken into accountat the time of sentencing or disposition. Men-

tal health professionals need this informationin order to perform accurate assessments andmake appropriate treatment recommendations;individuals at different stages of developmentmay need very different sorts of interventions.And attorneys need this information in orderto practice law more effectively; prosecutorsmay consider a juvenile’s developmental statusin deciding when it is appropriate to chargean individual as an adult, and defense attorneysneed to know how best to interact with clientswho may not fully understand their situation.Understanding the nature of psychological de-velopment during adolescence, therefore, willlikely improve policymaking, judicial decisionmaking, forensic evaluation, and legal practice.

BRAIN, COGNITIVE, ANDPSYCHOSOCIAL DEVELOPMENTIN ADOLESCENCE

When lawmakers focus on juvenile justice pol-icy, the distinction between adolescence andadulthood, rather than that between childhoodand adolescence, is of primary interest. How-ever, most studies of adolescent developmenthave compared adolescents with children, andonly in recent years has scientific interest fo-cused intensely on the psychological transitionbetween adolescence and adulthood, largely inresponse to new research showing continuedbrain maturation through the end of the ado-lescent period. This work has provided supportfor the uniqueness of adolescence as a stage oflife that is also distinct from adulthood with re-spect to several aspects of brain and psychoso-cial development.

Adolescent Brain Development

Although most of the developmental researchon cognitive and psychosocial functioning dur-ing adolescence involves psychological studies,recent work in developmental neuroscience isbeginning to shed light on the neural under-pinnings of psychological development acrossadolescence and adulthood. In the past sev-eral years, a new perspective on risk taking

www.annualreviews.org • Juvenile Justice 465

Ann

u. R

ev. C

lin. P

sych

ol. 2

009.

5:45

9-48

5. D

ownl

oade

d fr

om w

ww

.ann

ualr

evie

ws.

org

by M

assa

chus

etts

Sch

ool o

f Pr

ofes

sion

al P

sych

olog

y (M

SPP)

on

09/0

8/14

. For

per

sona

l use

onl

y.

ANRV372-CP05-20 ARI 19 February 2009 11:37

Socioemotionalsystem: the brainsystem governing theprocessing of socialand emotionalinformation and theexperience of rewardand punishment

Cognitive controlsystem: the brainsystem governingexecutive function,including deliberativethinking, impulsecontrol, foresight, andthe evaluation of riskand reward

(including antisocial risk taking) during adoles-cence has emerged, one that is informed by ad-vances in developmental neuroscience (Caseyet al. 2008, Steinberg 2008). According to thisview, risky behavior in adolescence is the prod-uct of the interaction between changes in twodistinct neurobiological systems: a socioemo-tional system, which is localized in limbic andparalimbic areas of the brain, including theamygdala, ventral striatum, orbitofrontal cor-tex, medial prefrontal cortex, and superior tem-poral sulcus; and a cognitive control system,which is mainly composed of the lateral pre-frontal and parietal cortices and those parts ofthe anterior cingulate cortex to which they areinterconnected (Steinberg 2007).

According to this dual-systems model, ado-lescent risk taking is hypothesized to be stim-ulated by a rapid and dramatic increase indopaminergic activity within the socioemo-tional system around the time of puberty, whichis presumed to lead to increases in reward seek-ing. However, this increase in reward seekingprecedes the structural maturation of the cogni-tive control system and its connections to areasof the socioemotional system, a maturationalprocess that is gradual, unfolds over the courseof adolescence, and permits more advanced self-regulation and impulse control. The temporalgap between the arousal of the socioemotionalsystem, which is an early adolescent develop-ment, and the full maturation of the cognitivecontrol system, which occurs later, creates a pe-riod of heightened vulnerability to risk takingduring middle adolescence (Steinberg 2008). Asone writer has characterized it, the process maybe akin to “starting the engines without a skilleddriver behind the wheel” (Dahl 2001).

Neurobiological evidence in support of thisdual-systems model is rapidly accumulating. Agrowing literature, derived primarily from ro-dent studies but with implications for humandevelopment, indicates that the remodeling ofthe dopaminergic system within the socioemo-tional network involves an initial postnatal riseand then, starting in preadolescence, a subse-quent reduction of dopamine receptor densityin the striatum and prefrontal cortex; this pat-

tern is more pronounced among males than fe-males (Sisk & Foster 2004, Sisk & Zehr 2005,Teicher et al. 1995). As a result of this remodel-ing, dopaminergic activity in the prefrontal cor-tex increases significantly in early adolescenceand is higher during this period than beforeor after. Because dopamine plays a critical rolein the brain’s reward circuitry, the increase, re-duction, and redistribution of dopamine recep-tor concentration around puberty, especially inprojections from the limbic system to the pre-frontal area, is likely to increase reward-seekingbehavior and, accordingly, sensation seeking.

There is equally compelling neurobiolog-ical evidence for changes in brain structureand function during adolescence and earlyadulthood that facilitate improvements in self-regulation that permit individuals to modulatetheir inclinations to seek rewards, although thisdevelopment is presumed to unfold along a dif-ferent timetable and to be independent of pu-berty (see Paus 2005 for a summary). Becauseof synaptic pruning and the continued myeli-nation of prefrontal brain regions, resulting inimproved connectivity among cortical areas andbetween cortical and subcortical areas, there areimprovements over the course of adolescencein many aspects of executive function, such asresponse inhibition, planning, weighing risksand rewards, and the simultaneous considera-tion of multiple sources of information. Thereis also improved coordination of affect and cog-nition, reflected in improved emotion regula-tion, which is facilitated by the increased con-nectivity between regions associated with thesocioemotional and cognitive control systems.

The development of the cognitive controlsystem, which is manifested chiefly in improvedconnectivity across brain regions, must be dis-tinguished from the well-publicized maturationof the frontal lobes because of synaptic prun-ing. Although both processes result in improvedthinking abilities, they occur at different timesin adolescence and have different implicationsfor cognitive development. Whereas increasesin connectivity take place throughout adoles-cence and well into adulthood, the declinein gray matter density that reflects synaptic

466 Steinberg

Ann

u. R

ev. C

lin. P

sych

ol. 2

009.

5:45

9-48

5. D

ownl

oade

d fr

om w

ww

.ann

ualr

evie

ws.

org

by M

assa

chus

etts

Sch

ool o

f Pr

ofes

sion

al P

sych

olog

y (M

SPP)

on

09/0

8/14

. For

per

sona

l use

onl

y.

ANRV372-CP05-20 ARI 19 February 2009 11:37

pruning takes place in preadolescence and earlyadolescence and is more or less complete by age16. Consequently, performance on tasks that ac-tivate the frontal lobes continues to improvethrough middle adolescence but not beyondage 16 on tasks of moderate difficulty (Conklinet al. 2007, Crone & van der Molen 2004,Hooper et al. 2004, Luna et al. 2001). In con-trast, adult-like performance on more demand-ing cognitive tasks, especially those that requirecoordination between and among multiple cor-tical and subcortical brain regions, is not at-tained until later in development.

The upshot of this developmental neu-roscience is that changes in the socioemo-tional system at puberty may promote reck-less, sensation-seeking behavior in early andmiddle adolescence, while the regions of theprefrontal cortex that govern cognitive controlcontinue to mature over the course of adoles-cence and into young adulthood. This temporalgap between the increase in sensation seekingaround puberty and the later development ofmature self-regulatory competence may com-bine to make adolescence a time of inherentlyimmature judgment. Thus, despite the fact thatin many ways adolescents may appear to be asintelligent as adults (at least as indexed by per-formance on tests of information processingand logical reasoning), their ability to regulatetheir behavior in accord with these advanced in-tellectual abilities is more limited. As the nextsection makes clear, research on adolescent cog-nitive and psychosocial development is consis-tent with this neurobiological profile.

Adolescent Cognitive Development

The application of information about norma-tive adolescent development to policy and prac-tice in the justice system necessitates differ-entiating between cognitive and psychosocialdevelopment, which appear to follow differentdevelopmental trajectories (Steinberg 2008).Briefly, on relatively less-demanding tasks thatare mainly or exclusively cognitive in nature,and where improvement in adolescence is likelydue to synaptic pruning of the frontal lobes,

adolescents evince adult levels of competenceby age 16. In contrast, on more challengingtasks that involve the coordination of affectand cognition, and on many measures of psy-chosocial maturity, performance continues toimprove well into young adulthood, most likelybecause this improvement is mediated by im-proved connectivity across brain regions, a rela-tively later development. As I discuss below, thistemporal disjunction has created a great deal ofconfusion with regard to where we should drawthe legal boundary between adolescence andadulthood, because different developmental lit-eratures suggest different chronological ages.

The most important cognitive capacities in-volved in decision making are understanding(i.e., the ability to comprehend information rel-evant to the decision) and reasoning (i.e., theability to use this information logically to makea choice). These capacities increase throughchildhood into adolescence. Between late child-hood and middle adolescence (roughly betweenthe ages of 11 and 16), individuals show markedimprovements in reasoning (especially deduc-tive reasoning) and in both the efficiency andcapacity of information processing (Hale 1990,Kail 1997, Keating 2004, Overton 1990). Re-search has demonstrated conclusively that, as aresult of gains in these areas, individuals be-come more capable of abstract, multidimen-sional, deliberative, and hypothetical thinkingas they develop from late childhood into mid-dle adolescence (Kuhn 2009). These abilitiesreach an asymptote sometime around 16, andby this age, teens’ capacities for understandingand reasoning in making decisions, at least incontrolled experiments, roughly approximatethose of adults. This comparability betweenmiddle adolescents and adults is not limited tobasic cognitive abilities such as memory or ver-bal fluency or to performance on tasks of log-ical reasoning. Studies of capacity to grant in-formed consent to receive medical treatment orparticipate as a research subject, for example,show little improvement beyond age 16 (Belter& Grisso 1984, Grisso & Vierling 1978,Gustafson & McNamara 1987, Weithorn &Campbell 1982).

www.annualreviews.org • Juvenile Justice 467

Ann

u. R

ev. C

lin. P

sych

ol. 2

009.

5:45

9-48

5. D

ownl

oade

d fr

om w

ww

.ann

ualr

evie

ws.

org

by M

assa

chus

etts

Sch

ool o

f Pr

ofes

sion

al P

sych

olog

y (M

SPP)

on

09/0

8/14

. For

per

sona

l use

onl

y.

ANRV372-CP05-20 ARI 19 February 2009 11:37

The notion that adolescents and adultsdemonstrate comparable capacities for under-standing and reasoning should not be taken tomean that they also demonstrate comparablelevels of maturity of judgment, however. As mycolleagues and I have argued elsewhere, matu-rity of judgment is affected both by cognitivecapabilities as well as psychosocial ones, and al-though the former show adult levels of matu-rity by 16, the latter do not (Steinberg et al.2009). As a result, adolescents may be less ableto deploy their cognitive capacities as effectivelyas adults in exercising judgment in their every-day lives when decisions are influenced by emo-tional and social variables. The development ofthese psychosocial factors is described in thenext section.

Adolescent Psychosocial Development

New perspectives on adolescent “cognition-in-context” emphasize that adolescent thinking ineveryday settings is a function of social andemotional, as well as cognitive, processes, andthat a full account of youthful judgment mustexamine the interaction of all of these influ-ences (Scott et al. 1995, Steinberg & Cauffman1996). Even when adolescent cognitive capaci-ties approximate those of adults, youthful deci-sion making may still differ from that of adultsdue to psychosocial immaturity. Indeed, re-search indicates that psychosocial maturationproceeds more slowly than cognitive develop-ment and that age differences in judgment mayreflect social and emotional differences betweenadolescents and adults that continue well be-yond mid-adolescence. Of particular relevanceto the present discussion are age differencesin susceptibility to peer influence, future ori-entation, reward sensitivity, and the capacityfor self-regulation. Available research indicatesthat adolescents and adults differ significantlywith respect to each of these attributes.

Peer influence. Substantial research evidencesupports the conventional wisdom that teensare more oriented toward peers and responsiveto peer influence than are adults (Steinberg &

Monahan 2007). Resistance to peer influenceincreases between adolescence and adulthoodas individuals begin to form an independentsense of self and develop greater capacity for au-tonomous decision making. Studies of age dif-ferences and age changes in resistance to peerinfluence suggest somewhat different patternsvis-a-vis antisocial versus neutral or proso-cial peer pressure prior to middle adolescence(with resistance to antisocial influence decreas-ing during this time, especially among boys, butresistance to other forms of peer influence in-creasing), but similar patterns after age 14 (withresistance to all forms of peer influence increas-ing). Because the main justice policy and prac-tice questions concern differences between ado-lescents and adults, especially during the latterpart of the adolescent period, it is this increasein resistance to peer influence from age 14 onthat is of particular interest.

Recent studies of the neural underpinningsof resistance to peer influence in adolescence in-dicate that improvements in this capacity maybe linked to the development of greater con-nectivity between cortical and subcortical re-gions, which likely facilitates the better co-ordination of affect and cognition (Grosbraset al. 2007, Paus et al. 2008), although it shouldbe noted that this conclusion is based on stud-ies of individual differences in brain morphol-ogy and function among same-aged adolescentswho differ in their self-reported resistance topeer pressure and not to cross-sectional or lon-gitudinal studies that link age differences inresistance to peer influence to age differencesin brain structure or function. Nevertheless, itis reasonable to speculate that the social andarousal processes that may undermine logicaldecision making during adolescence, when con-nectivity is still maturing, do not have the sameimpact during adulthood. One recent behav-ioral study found, for instance, that adoles-cents, college undergraduates, and adults per-formed similarly on a risk-taking task whenperforming the task alone, but that the pres-ence of same-aged friends doubled risk tak-ing among the adolescents and increased it50% among the undergraduates, but had no

468 Steinberg

Ann

u. R

ev. C

lin. P

sych

ol. 2

009.

5:45

9-48

5. D

ownl

oade

d fr

om w

ww

.ann

ualr

evie

ws.

org

by M

assa

chus

etts

Sch

ool o

f Pr

ofes

sion

al P

sych

olog

y (M

SPP)

on

09/0

8/14

. For

per

sona

l use

onl

y.

ANRV372-CP05-20 ARI 19 February 2009 11:37

impact on the adults (Gardner & Steinberg2005).

Peer influence affects adolescent judgmentboth directly and indirectly. In some contexts,adolescents might make choices in response todirect peer pressure, as when they are coercedto take risks that they might otherwise avoid.More indirectly, adolescents’ desire for peer ap-proval and consequent fear of rejection affectstheir choices even without direct coercion. Theincreased salience of peers in adolescence likelymakes approval seeking especially important ingroup situations. Thus, it is not surprising, per-haps, that adolescents are far more likely thanare adults to commit crimes in groups (Zimring1998). Peers also may provide models for be-havior that adolescents believe will assist themto accomplish their own ends. For example,there is some evidence that during early andmiddle adolescence, teens who engage in cer-tain types of antisocial behavior, such as fight-ing or drinking, may enjoy higher status amongtheir peers as a consequence. Accordingly, someadolescents may engage in antisocial conduct toimpress their friends or to conform to peer ex-pectations; indeed, in one of the most influentialaccounts of so-called adolescence-limited of-fenders (that is, individuals who commit crimesduring adolescence but not before or after), im-itation of higher-status peers is hypothesized tobe a prime motivation for antisocial behavior(Moffitt 1993).

Future orientation. Future orientation, thecapacity and inclination to project events intothe future, may also influence judgment becauseit affects the extent to which individuals con-sider the long-term consequences of their ac-tions in making choices. Over the course of ado-lescence and into young adulthood, individualsbecome more future oriented, with increases intheir consideration of future consequences, intheir concern about the future, and in their abil-ity to plan ahead (Greene 1986, Nurmi 1991,Steinberg et al. 2008b).

There are several plausible explanations forthis age gap in future orientation. In part, ado-lescents’ weaker future orientation may reflect

Adolescence-limitedoffenders: antisocialindividuals whoseoffending begins andends duringadolescence

their more limited life experience (Gardner1993). To a young person, a short-term conse-quence may have far greater salience than onefive years in the future. The latter may seemvery remote simply because five years repre-sents a substantial portion of her life. There isalso evidence linking the differences betweenadolescents and adults in future orientation toage differences in brain structure and function,especially in the prefrontal cortex (Cauffmanet al. 2005).

Reward sensitivity. Research evidence alsosuggests that, relative to adults, adolescentsare more sensitive to rewards and, especially,to immediate rewards, a difference that mayexplain age differences in sensation seekingand risk taking (Galvan et al. 2007, Steinberget al. 2008a). Although it had once been be-lieved that adolescents and adults differ in riskperception, it now appears that age differ-ences in risk taking are more likely mediatedby age differences in reward sensitivity thanby age differences in sensitivity to the poten-tial adverse consequences of a risky decision(Cauffman et al. 2008, Millstein & Halpern-Felsher 2002). Thus, adolescents and adultsmay perceive risks similarly (both in the lab andin the real world) but evaluate rewards differ-ently, especially when the benefits of the riskydecision are weighed against the costs. So, forexample, in deciding whether to speed whiledriving a car, adolescents and adults may es-timate the risks of this behavior (e.g., beingticketed, getting into an accident) similarly, butadolescents may weigh the potential rewards(e.g., the thrill of driving fast, peer approval,getting to one’s destination sooner) more heav-ily than adults, leading to lower risk ratiosfor teens—and a higher likelihood of engagingin the (rewarding) activity. Thus, what distin-guishes adolescents from adults in this regardis not the fact that teens are less knowledgeableabout risks, but rather that they attach greatervalue to the rewards that risk taking provides(Steinberg 2004).

The heightened salience of rewards to ado-lescents, relative to adults, is seen in age

www.annualreviews.org • Juvenile Justice 469

Ann

u. R

ev. C

lin. P

sych

ol. 2

009.

5:45

9-48

5. D

ownl

oade

d fr

om w

ww

.ann

ualr

evie

ws.

org

by M

assa

chus

etts

Sch

ool o

f Pr

ofes

sion

al P

sych

olog

y (M

SPP)

on

09/0

8/14

. For

per

sona

l use

onl

y.

ANRV372-CP05-20 ARI 19 February 2009 11:37

differences in performance on the IowaGambling Task, in which subjects are given fourdecks of cards, face down, and are instructed toturn over cards, one at a time, from any deck.Each card has information about how muchmoney the subject has won or lost by select-ing that card. Two of the decks are “good,” inthat drawing from them will lead to gains overtime, and two of the decks are “bad”; drawingfrom them will produce net losses. Because afew cards in the “bad” decks offer very high re-wards, though, a person who is especially sensi-tive to rewards will be drawn to the “bad” decks,even if he or she keeps losing money as a re-sult. At the beginning of the task, people tendto draw randomly from all four decks, but asthe task progresses, normal adults pick morefrequently from the good decks. Children andyounger adolescents (as well as adults with dam-age to the ventromedial prefrontal cortex) dopoorly on this task (Crone et al. 2005, Crone& van der Molen 2004, Hooper et al. 2004).Performance improves with age, with the mostdramatic improvement taking place duringmiddle adolescence. This likely reflects a de-crease in susceptibility to choosing based onthe prospect of an immediate, attractive reward.Further evidence that adolescents tend to valueimmediate rewards more than adults do is seenin age differences in performance on tests of de-lay discounting, in which individuals are askedto chose between a smaller immediate reward(e.g., receiving $600 tomorrow) and a larger de-layed one (e.g., receiving $1000 in one year)(Steinberg et al. 2008b). Heightened rewardsensitivity, indexed by self-report or task per-formance, is especially pronounced during earlyand middle adolescence, when reward circuitryin the brain is undergoing extensive remodel-ing. There is some evidence from both humanand animal studies that this may be linked topubertal maturation (Dahl 2004).

Self-regulation. In addition to age differencesin susceptibility to peer influence, future orien-tation, and reward sensitivity, adolescents andadults also differ with respect to their ability tocontrol impulsive behavior and choices. Thus,

the widely held stereotype that adolescents aremore reckless than adults is supported by re-search on developmental changes in impulsivityand self-management over the course of ado-lescence (Galvan et al. 2007, Leshem & Glick-sohn 2007). In general, studies show gradualbut steady increases in the capacity for self-direction through adolescence, with gains con-tinuing through the high school years and intoyoung adulthood. Similarly, impulsivity, as ageneral trait, declines linearly between adoles-cence and adulthood (Steinberg et al. 2008a).

An illustration of behavioral research thatsheds light on age differences in impulse con-trol is the study of performance on a task knownas the Tower of London. In this test, the sub-ject is presented with an arrangement of col-ored balls, stacked in a certain order, and sev-eral empty vertical rods onto which the balls canbe moved. The subject is then presented with apicture of a different configuration of balls andasked to turn the original configuration into thenew one by moving one ball at a time, using thefewest number of moves (Berg & Byrd 2002).This task requires thinking ahead, because ex-tra moves must be used to undo a mistake. Inseveral studies, our research group found thatearly and middle adolescents performed simi-larly to adults when the problem presented wasan easy one (i.e., one that could be solved intwo or three moves), but that they did not planahead as much as late adolescents and youngadults on the harder problems; unlike the oldersubjects, the younger individuals spent no moretime before making their first move on the com-plex problems than they did on the simple ones(Steinberg et al. 2008a). These findings are con-sistent with casual observations of teenagers inthe real world, which also suggest that they areless likely than are adults to think ahead beforeacting.

Taken together, these findings from self-report and behavioral studies of psychosocialdevelopment indicate that individuals becomemore resistant to peer influence and orientedto the future, and less drawn to immediate re-wards and impulsive, as they mature from ado-lescence to adulthood. Although the science of

470 Steinberg

Ann

u. R

ev. C

lin. P

sych

ol. 2

009.

5:45

9-48

5. D

ownl

oade

d fr

om w

ww

.ann

ualr

evie

ws.

org

by M

assa

chus

etts

Sch

ool o

f Pr

ofes

sion

al P

sych

olog

y (M

SPP)

on

09/0

8/14

. For

per

sona

l use

onl

y.

ANRV372-CP05-20 ARI 19 February 2009 11:37

adolescent brain development is still in its in-fancy, findings indicate that much of this matu-ration continues well beyond the age by whichindividuals evince adult levels of performanceon tests of cognitive capacity. As I discuss inthe next section, the continued maturation ofcognitive competence through age 16 and thecontinued maturation of psychosocial compe-tence into young adulthood have important im-plications for how we view and respond to thecriminal behavior of juveniles.

JUVENILE JUSTICE ISSUESINFORMED BYDEVELOPMENTAL SCIENCE

Criminal Culpability of Youth

The adult justice system presumes that defen-dants who are found guilty are responsible fortheir own actions, should be held accountable,and should be punished accordingly. Because ofthe relative immaturity of minors, however, itmay not be justified to hold them as account-able as one might hold adults. If, for example,adolescents below a certain age cannot graspthe long-term consequences of their actionsor cannot control their impulses, one cannothold them fully accountable for their actions.In other words, we cannot claim that adoles-cents “ought to know better” if, in fact, the ev-idence indicates that they do not know better,or more accurately, cannot know better, becausethey lack the abilities needed to exercise maturejudgment. It is important to note that culpabil-ity cannot really be researched directly. Becausean individual’s culpability is something that isjudged by someone else, it is largely in the eyeof the beholder. What can be studied, however,are the capabilities and characteristics of indi-viduals that make them potentially blamewor-thy, such as their ability to behave intentionallyor to know right from wrong.

I use the term “culpability” in this review as ashorthand for several interrelated phenomena,including responsibility, accountability, blame-worthiness, and punishability. These notionsare relevant to the adjudication of an individ-

Penalproportionality: theprinciple in Americancriminal law linkingthe severity ofpunishment for acrime to the criminal’sculpability

Mitigation: incriminal law, thelessening of criminalresponsibility

ual’s guilt or innocence, because an individualwho is not responsible for his or her actions bydefinition cannot be guilty, and to the deter-mination of a disposition (in juvenile court) orsentence (in criminal court), in that individualswho are found guilty but less than completelyblameworthy, owing to any number of mitigat-ing circumstances, merit proportionately lesspunishment than do guilty individuals who arefully blameworthy.

The starting point in a discussion of crim-inal culpability is a principle known as penalproportionality. Simply put, penal proportion-ality holds that criminal punishment should bedetermined by two criteria: the harm a personcauses and his blameworthiness in causing thatharm. The law recognizes that different wrong-ful acts cause different levels of harm through acomplex system of offense grading under whichmore serious crimes (rape, for example) arepunished presumptively more severely than lessserious crimes (shoplifting, for example). Be-yond this, though, two people who engage inthe same wrongful conduct may differ in theirblameworthiness. A person may be less culpa-ble than other criminals—or not culpable atall—because he inadvertently (rather than pur-posely) causes the harm, because he is subjectto some endogenous deficiency or incapacitythat impairs his decision making (such as men-tal illness), or because he acts in response to anextraordinary external pressure—a gun to thehead is the classic example. Less-blameworthyoffenders deserve less punishment, and somepersons who cause criminal harm deserve nopunishment at all (Scott & Steinberg 2008).

The concept of mitigation plays an impor-tant role in the law’s calculation of blame andpunishment, although it gets little attention inthe debate about youth crime. Mitigation ap-plies to persons engaging in harmful conductwho are blameworthy enough to meet the min-imum threshold of criminal responsibility butwho deserve less punishment than a typical of-fender would receive. Through mitigation, thecriminal law calculates culpability and punish-ment along a continuum and is not limited tothe options of full responsibility or complete

www.annualreviews.org • Juvenile Justice 471

Ann

u. R

ev. C

lin. P

sych

ol. 2

009.

5:45

9-48

5. D

ownl

oade

d fr

om w

ww

.ann

ualr

evie

ws.

org

by M

assa

chus

etts

Sch

ool o

f Pr

ofes

sion

al P

sych

olog

y (M

SPP)

on

09/0

8/14

. For

per

sona

l use

onl

y.

ANRV372-CP05-20 ARI 19 February 2009 11:37

excuse. Indeed, criminal law incorporates cal-ibrated measures of culpability. For example,the law of homicide operates through a grad-ing scheme under which punishment for killinganother person varies dramatically dependingon the actor’s blameworthiness. Thus, the actorwho kills intentionally is deemed less culpableif he does so without premeditation because hischoice reveals less consideration of the harmfulconsequences of his act, and the actor who neg-ligently causes another’s death is guilty of a lessserious crime than one who intends to kill. Aperson who kills in response to provocation orunder extreme emotional disturbance may beguilty only of manslaughter and not of murder.Under standard homicide doctrine, mitigat-ing circumstances and mental states are trans-lated into lower-grade offenses that warrant lesspunishment.

What makes the conduct of one person lessblameworthy than that of another person whocauses the same harm? Generally, a personwho causes criminal harm is a fully responsiblemoral agent (and deserves full punishment) if, inchoosing to engage in the wrongful conduct, hehas the capacity to make a rational decision anda “fair opportunity” to choose not to engagein the harmful conduct. Under this view, theactor whose thinking is substantially impairedor whose freedom is significantly constrainedis less culpable than is the typical offender anddeserves less punishment—how much less de-pends on the extent of the impairment or co-ercion. Under American criminal law, two verydifferent kinds of persons can show that theircriminal conduct was less culpable than that ofthe offender who deserves full punishment—those who are very different from ordinary per-sons due to impairments that contributed totheir criminal choices and those who are ordi-nary persons whose offenses are responses toextraordinary circumstances or are otherwiseaberrant conduct (Scott & Steinberg 2008).

Although it seems paradoxical, adolescents,in a real sense, belong to both groups. In the firstgroup are individuals with endogenous traitsor conditions that undermine their decision-making capacity, impairing their ability to un-

derstand the nature and consequences of theirwrongful acts or to control their conduct. Inmodern times, this category has been reservedmostly for offenders who suffer from mental ill-ness, mental disability, and other neurologicalimpairments. The criminal law defenses of in-sanity, diminished capacity, extreme emotionaldisturbance, and involuntary act recognize thatpsychological and biological incapacities canundermine decision making in ways that reduceor negate the culpability of criminal choices.

Individuals in the second group are ordinarypersons whose criminal conduct is less culpablebecause it is a response to extraordinary exter-nal circumstances: These cases arise when theactor faces a difficult choice, and his responseof engaging in the criminal conduct is reason-able under the circumstances, as measured bythe likely response of an ordinary law-abidingperson in that situation. Thus, under stan-dard self-defense doctrine, a person who killsa threatening assailant is excused from liabilityif a reasonable person in his place would havefelt that his life was in danger. Similarly, the de-fenses of duress, necessity, and provocation areavailable to actors who can explain their crim-inal conduct in terms of unusual external pres-sures that constrained their ability to choose.

In the preceding section, I described aspectsof psychological development in adolescencethat are relevant to youthful choices to get in-volved in criminal activity and that may distin-guish young offenders from their adult counter-parts. Although youths in mid-adolescence havecognitive capacities for reasoning and under-standing that approximate those of adults, evenat age 18 adolescents are immature in their psy-chosocial and emotional development, and thislikely affects their decisions about involvementin crime in ways that distinguish them fromadults. Teenagers are more susceptible to peerinfluence than are adults and tend to focus moreon rewards and less on risks in making choices.They also tend to focus on short-term ratherthan long-term consequences and are less ca-pable of anticipating future consequences, andthey are more impulsive and volatile in theiremotional responses. When we consider these

472 Steinberg

Ann

u. R

ev. C

lin. P

sych

ol. 2

009.

5:45

9-48

5. D

ownl

oade

d fr

om w

ww

.ann

ualr

evie

ws.

org

by M

assa

chus

etts

Sch

ool o

f Pr

ofes

sion

al P

sych

olog

y (M

SPP)

on

09/0

8/14

. For

per

sona

l use

onl

y.

ANRV372-CP05-20 ARI 19 February 2009 11:37

developmental factors within the conventionalcriminal law framework for assessing blame-worthiness, the unsurprising conclusion is thatadolescent offenders are less culpable than areadults. The mitigating conditions generally rec-ognized in the criminal law—diminished capac-ity and coercive circumstances—are relevant tocriminal acts of adolescents and often character-ize the actions of juvenile offenders. This doesnot excuse adolescents from criminal responsi-bility, but it renders them less blameworthy andless deserving of adult punishment.

Although in general lawmakers have paidminimal attention to the mitigating charac-ter of adolescents’ diminished decision-makingcapacities, some legislatures and courts haverecognized that immature judgment reducesculpability. Most notably, in its considera-tion of the constitutionality of the juveniledeath penalty, the Supreme Court has focusedon this rationale for mitigation. In Roper v.Simmons, the 2005 case that abolished the ju-venile death penalty, the Court adopted the de-velopmental argument for mitigation that fol-lows from the research reviewed above. JusticeKennedy, writing for the majority, describedthree features of adolescence that distinguishyoung offenders from their adult counter-parts in ways that mitigate culpability—featuresthat are familiar to the reader at this point.The first is the diminished decision-makingcapacity of youths, which may contribute to acriminal choice that is “not as morally repre-hensible as that of adults” because of its de-velopmental nature. The Court pointed to thetendency of adolescents to engage in risky be-havior and noted that immaturity and an “un-derdeveloped sense of responsibility” often re-sult in “impetuous and ill-considered decisions”by youths. Second, the Court pointed to the in-creased vulnerability of youths to external coer-cion, including peer pressure. Finally, the Courtemphasized that the unformed nature of adoles-cent identity made it “less supportable to con-clude that even a heinous crime was evidence ofirretrievably depraved character.” Adolescentsare less blameworthy than are adults, the Courtsuggested, because the traits that contribute

Roper v. Simmons:the U.S. SupremeCourt case thatabolished the juveniledeath penalty

to criminal conduct are transient, and becausemost adolescents will outgrow their tendency toget involved in crime as they mature. Althoughthe Court did not elaborate, we have seen thateach of these attributes of adolescence corre-sponds to a conventional source of mitigationin criminal law (Roper v. Simmons 2005).

Does this argument apply to the conductof immature adults? Although most impulsiveyoung risk takers mature into adults with differ-ent values, some adult criminals are impulsive,sensation-seeking risk takers who discount fu-ture consequences and focus on the here andnow. Are these adolescent-like adults also lessculpable than other adult offenders and deserv-ing of reduced punishment? I think not. Unlikethe typical adolescent, the predispositions, val-ues, and preferences that motivate the adult of-fenders are not developmental but charactero-logical, and they are unlikely to change merelywith the passage of time. Adolescent traits thatcontribute to criminal conduct are normative ofadolescence, but they are not typical in adult-hood. In an adult, these traits are often part ofthe personal identity of an individual who doesnot respect the values of the criminal law andwho deserves punishment when he or she vio-lates its prohibitions (Scott & Steinberg 2008).

Competence of Adolescentsto Stand Trial

Before discussing adolescents’ competence tostand trial, it is worth underscoring the distinc-tion between competence and culpability—twovery different constructs that are often con-fused, even by those with expertise in crimi-nal law. Competence to stand trial refers to theability of an individual to function effectivelyas a defendant in a criminal or delinquencyproceeding. In contrast, determinations of cul-pability focus on the defendant’s blameworthi-ness in engaging in the criminal conduct andon whether and to what extent he will be heldresponsible. Although many of the same inca-pacities that excuse or mitigate criminal respon-sibility may also render a defendant incompe-tent, the two issues are analytically distinct and

www.annualreviews.org • Juvenile Justice 473

Ann

u. R

ev. C

lin. P

sych

ol. 2

009.

5:45

9-48

5. D

ownl

oade

d fr

om w

ww

.ann

ualr

evie

ws.

org

by M

assa

chus

etts

Sch

ool o

f Pr

ofes

sion

al P

sych

olog

y (M

SPP)

on

09/0

8/14

. For

per

sona

l use

onl

y.

ANRV372-CP05-20 ARI 19 February 2009 11:37

Dusky v. United

States: the U.S.Supreme Court casethat establishedcriteria forcompetence to standtrial

In re Gault: the U.S.Supreme Court casethat determined thatjuveniles adjudicatedin juvenile court wereentitled to many of thesame proceduralprotections as adultsadjudicated in criminalcourt

Developmentalincompetence: a lackof competence tostand trial due tonormal cognitive orpsychosocialimmaturity, as opposedto mental illness ordisability

separate legal inquiries, and they focus on thedefendant’s mental state at two different pointsin time (the time of the crime and the time ofthe court proceeding).

The reason that competence is required ofdefendants in criminal proceedings is simple:When the state asserts its power against an in-dividual with the goal of taking away his liberty,the accused must be capable of participatingin a meaningful way in the proceeding againsthim. If a defendant is so mentally ill or disabledthat he cannot participate adequately, then thetrial lacks fundamental fairness that is requiredas a part of due process under the FourteenthAmendment to the U.S. Constitution (Scott &Grisso 2005).

In 1960, the Supreme Court announced alegal standard for trial competence in Dusky v.United States that has since been adopted uni-formly by American courts. According to Dusky,when the issue of a defendant’s competence israised in a criminal trial, the court’s determina-tion should focus on “whether the defendanthas sufficient present ability to consult withhis lawyer with a reasonable degree of ratio-nal understanding—and whether he has a ra-tional, as well as factual, understanding of theproceedings against him.” Thus, there are twoparts to the competence requirement: The de-fendant must be able to consult with her attor-ney about planning and making decisions in herdefense, and she must understand the charges,the meaning, and purpose of the proceedingsand the consequences of conviction (Scott &Grisso 2005).

The requirement that criminal defendantsbe competent to stand trial had no place indelinquency proceedings in the traditional ju-venile court. In a system in which the govern-ment’s announced purpose was to rehabilitateand not to punish errant youths, the proce-dural protections accorded adult defendants—including the requirement of adjudicativecompetence—were thought to be unnecessary.This all changed with In re Gault, which led toan extensive restructuring of delinquency pro-ceedings to conform to the requirements ofconstitutional due process. Today, it is generally

accepted that requirements of due process andfundamental fairness are satisfied only if youthsfacing charges in juvenile court are competentto stand trial.

Until the 1990s, the issue of juveniles’ trialcompetence involved a straightforward incor-poration into delinquency proceedings of aprocedural protection that was relevant to a rel-atively small number of mentally impaired adultdefendants, where it was assumed to apply sim-ilarly to a small number of mentally incapaci-tated youths. The regulatory reforms that be-gan in the late 1980s changed the situation byincreasing the punishment stakes facing manyyoung offenders and by eroding the boundarybetween the adult and juvenile systems. The im-portance of this issue was not recognized im-mediately, however. As legislatures across thecountry began to enact laws that dramaticallyaltered the landscape of juvenile crime policy,the procedural issue of whether developmen-tally immature youngsters charged with crimesmight be less able to participate in criminalproceedings than are adult defendants—whatis referred to in this article as developmentalincompetence—was not central to the policydebates.

Given that developmental incompetencelargely escaped the attention of courts and pol-icy makers until recently, it is worth asking di-rectly whether the constitutional prohibitionagainst criminal adjudication of incompetentdefendants must be applied to this form of in-capacity. The answer is surely “yes.” The com-petence requirement is functional at its core,speaking to questions about the impact of cog-nitive deficiencies on trial participation. Func-tionally it makes no difference if the defendantcannot understand the proceeding she faces orassist her attorney, whether due to mental ill-ness or to immaturity (Scott & Grisso 2005).In either case, the fairness of the proceeding isundermined. In short, the same concerns thatsupport the prohibition against trying criminaldefendants who are incompetent due to mentalimpairment apply with equal force when imma-ture youths are subject to criminal proceedings.In the context of the recent changes in juvenile

474 Steinberg

Ann

u. R

ev. C

lin. P

sych

ol. 2

009.

5:45

9-48

5. D

ownl

oade

d fr

om w

ww

.ann

ualr

evie

ws.

org

by M

assa

chus

etts

Sch

ool o

f Pr

ofes

sion

al P

sych

olog

y (M

SPP)

on

09/0

8/14

. For

per

sona

l use

onl

y.

ANRV372-CP05-20 ARI 19 February 2009 11:37

justice policy, it has become important to havea better understanding of how the capacities ofchildren and adolescents to participate in crim-inal proceedings compare with those of adults.In pursuit of this end, I first examine the specificabilities that are required for adjudicative com-petence under the legal standard. I then turn tothe research directly comparing the abilities ofjuveniles and adults.

Three broad types of abilities are implicatedunder the Dusky standard for competence tostand trial: (a) a factual understanding of theproceedings, (b) a rational understanding of theproceedings, and (c) the ability to assist coun-sel (Scott & Grisso 2005). Courts applying thestandard are directed to weigh each factor, butotherwise they exercise substantial discretionin deciding how much competence is enough.Examining each component of competence un-der the Dusky standard and considering how thecapacities of juvenile defendants are likely tocompare with those of adults is instructive.

Factual understanding focuses on the defen-dant’s knowledge and awareness of the chargesand his understanding of available pleas, pos-sible penalties, the general steps in the adjudi-cation process, the roles of various participantsin the pretrial and trial process, and his rightsas a defendant. Intellectual immaturity in ju-veniles may undermine factual understanding,especially given that youths generally have lessexperience and more limited ability to graspconcepts such as rights. Juveniles also may bemore likely than are adults to have extensivedeficits in their basic knowledge of the trial pro-cess, such that more than brief instruction isneeded to attain competence.

The rational understanding requirement ofDusky has been interpreted to mean that defen-dants must comprehend the implications, rele-vance, or significance of what they understandfactually regarding the trial process. Deficitsin rational understanding typically involve dis-torted or erroneous beliefs that nullify factualunderstanding. For example, an immature de-fendant may know that he has a right to re-main silent, yet believe that the judge can takethis “right” away at any time by demanding a

response to questions. (When asked what hethought the “right to remain silent” meant, my12-year-old son said, “It means that you don’thave to say anything until the police ask youa question.”) Intellectual, emotional, and psy-chosocial immaturity may undermine the abil-ity of some adolescents to grasp accurately themeaning and significance of matters that theyseem to understand factually.

Finally, the requirement that the defendantin a criminal proceeding must have the capac-ity to assist counsel encompasses three typesof abilities. The first is the ability to receiveand communicate information adequately to al-low counsel to prepare a defense. This abilitymay be compromised by impairments in atten-tion, memory, and concentration, deficits thatmight undermine the defendant’s ability to re-spond to instructions or to provide importantinformation to his attorney, such as a coher-ent account of the events surrounding the of-fense. As I noted above, these capacities con-tinue to improve through age 16, accordingto studies of cognitive development. Second,the ability to assist counsel requires a ratio-nal perspective regarding the attorney and herrole, free of notions or attitudes that couldimpair the collaborative relationship. For ex-ample, a young defendant may develop a be-lief that all adults involved in the proceed-ing are allied against him, perhaps after seeingdefense attorneys and prosecutors chatting to-gether outside the courtroom. Third, defen-dants must have the capacity to make decisionsabout pleading and the waiver or assertion ofother constitutional rights. These decisions in-volve not only adequate factual and rationalunderstanding, but also the ability to consideralternatives and make a choice in a decision-making process. Immature youths may lackcapacities to process information and exercisereason adequately in making trial decisions, es-pecially when the options are complex and theirconsequences far reaching.

As juveniles’ competence to stand trialbegan to emerge as an important issue inthe mid-1990s, the need for a comprehensivestudy comparing the abilities of adolescents

www.annualreviews.org • Juvenile Justice 475

Ann

u. R

ev. C

lin. P

sych

ol. 2

009.

5:45

9-48

5. D

ownl

oade

d fr

om w

ww

.ann

ualr

evie

ws.

org

by M

assa

chus

etts

Sch

ool o

f Pr

ofes

sion

al P

sych

olog

y (M

SPP)

on

09/0

8/14

. For

per

sona

l use

onl

y.

ANRV372-CP05-20 ARI 19 February 2009 11:37

and adults in this realm became apparent. Be-fore this time, a few small studies had lookedat particular capacities in juveniles that wereimportant at different stages in the justiceprocess. However, no comprehensive researchhad compared the specific capacities of juve-niles and adults that are directly implicated inassessments of adjudicative competence. Inresponse to that need, the MacArthur Founda-tion Research Network on Adolescent Devel-opment and Juvenile Justice sponsored a large-scale study of individuals between the ages of 11and 24—half of whom were in the custody of thejustice system and half of whom had never beendetained—designed to examine empirically therelationship between developmental immatu-rity and the abilities of young defendants toparticipate in their trials (Grisso et al. 2003).The study also probed age differences in psy-chosocial influences on decision making in thecriminal process.

Based on participants’ responses to a struc-tured interview that had been used in previ-ous studies of competence to stand trial amongmentally ill adults, and for which norms hadbeen established to define clinically signifi-cant “impairment,” the researchers found thatcompetence-related abilities improve signifi-cantly between the ages of 11 and 16. On aver-age, youths aged 11 to 13 demonstrated signif-icantly poorer understanding of trial matters,as well as poorer reasoning and recognition ofthe relevance of information for a legal defense,than did 14- and 15-year-olds, who in turn per-formed significantly more poorly than individ-uals aged 16 and older. There were no differ-ences between the 16- and 17-year-olds andthe young adults. The study produced similarresults when adolescents and adults were cate-gorized according to their scores above and be-low the cut-off scores indicating impairment.Nearly one-third of 11- to 13-year-olds andabout one-fifth of 14- and 15-year-olds, butonly 12% of individuals 16 and older, evidencedimpairment at a level comparable to mentallyill adults who had been found incompetent tostand trial with respect to either their ability toreason with facts or understand the trial process.

Individual performance did not differ signifi-cantly by gender, ethnicity, or, in the detainedgroups, as a function of the extent of individuals’prior justice system experience. This last find-ing is important because it indicates that thereare components of immaturity independent of alack of relevant experience that may contributeto elevated rates of incompetence amongjuveniles.

A different structured interview was used toprobe how psychosocial influences affect deci-sion making by assessing participants’ choicesin three hypothetical legal situations involv-ing a police interrogation, consultation with adefense attorney, and the evaluation of a prof-fered plea agreement. Significant age differ-ences were found in responses to police in-terrogation and to the plea agreement. First,youths, including 16- to 17-year-olds, weremuch more likely to recommend waiving con-stitutional rights during an interrogation thanwere adults, with 55% of 11- to 13-year-olds,40% of 14- to 15-year-olds, and 30% of 16-to 17-year-olds choosing to “talk and admit”involvement in an alleged offense (rather than“remaining silent”), but only 15% of the youngadults making this choice. There were also sig-nificant age differences in response to the pleaagreement. This vignette was styled so as not toclearly favor accepting or rejecting the state’soffer, which probably accounted for the factthat young adults were evenly divided in theirresponses. In contrast, 75% of the 11- to 13-year-olds, 65% of the 14- to 15-year-olds, and60% of the 16- to 17-year-olds recommendedaccepting the plea offer. Together, these resultssuggest a much stronger tendency for adoles-cents than for young adults to make choicesin compliance with the perceived desires ofauthority figures (Grisso et al. 2003).

Analysis of participants’ responses to thevignettes also indicated differences betweenthe youngest age group and older subjects inrisk perception and future orientation. Par-ticipants were asked to explain their choices,including their perceptions about positiveand negative consequences of various options;questions probed the subjects’ assessment of the

476 Steinberg

Ann

u. R

ev. C

lin. P

sych

ol. 2

009.

5:45

9-48

5. D

ownl

oade

d fr

om w

ww

.ann

ualr

evie

ws.

org

by M

assa

chus

etts

Sch

ool o

f Pr

ofes

sion

al P

sych

olog

y (M

SPP)

on

09/0

8/14

. For

per

sona

l use

onl

y.

ANRV372-CP05-20 ARI 19 February 2009 11:37

seriousness of risks (the perceived negative con-sequences) and the likelihood of risks material-izing. Analyses indicated age differences for allof these dimensions of “risk perception,” withthe 11- to 13-year-olds less able to see risks than16- to 17-year-olds and young adults. Similarly,in comparison with older adolescents, fewer11- to 13-year-olds mentioned the long-rangeconsequences of their decisions, which suggeststhat future orientation differences exist that areconsistent with those described above.

The study’s findings are consistent withthose of earlier studies that examined variousdimensions of youths’ functioning in the jus-tice system. For example, an important studyof youths’ and adults’ capacities to understandMiranda rights in the early 1980s found that,compared with adults in the criminal justicesystem, 14-year-olds in juvenile detention wereless able to understand the meaning and impor-tance of Miranda warnings (Grisso 1981). Otherstudies using smaller samples also have foundage differences across the adolescent years withregard to knowledge of legal terms and the le-gal process in delinquency and criminal pro-ceedings (e.g., Cooper 1997). Finally, a series ofstudies found significant age differences acrossthe adolescent years in “strategic thinking”about pleas; older adolescents were more likelythan younger subjects to make choices that re-flected calculations of probabilities and costsbased on information provided (e.g., Peterson-Badali & Abramovitch 1993).

In light of what is known about psycholog-ical maturation in early and mid-adolescence,these findings are not surprising. Indeed, giventhe abilities required of defendants in crimi-nal proceedings, it would be puzzling if youthsand adults performed similarly on competence-related measures. This research provides pow-erful and tangible evidence that some youthsfacing criminal charges may function less capa-bly as criminal defendants than do their adultcounterparts. This does not mean, of course,that all youths should be automatically deemedincompetent to stand trial any more than woulda psychiatric diagnosis or low IQ score. Itdoes mean, however, that the risk of incom-

petence is substantially elevated in early andmid-adolescence; it also means that policy mak-ers and practitioners must address developmen-tal incompetence as it affects the treatment ofjuveniles in court (Scott & Grisso 2005).

It is important to emphasize that the pat-tern of age differences in studies of legal deci-sion making more closely resembles that seenin studies of cognitive development (where fewage differences are apparent after 16) than instudies of psychosocial development (where agedifferences are observed in late adolescence andsometimes in young adulthood). This suggeststhat determinations of where to draw a legalboundary between adolescence and adulthoodmust be domain specific. In matters in whichcognitive abilities predominate, and where psy-chosocial factors are of minimal importance(that is, in situations where the influence of ado-lescents’ impulsivity, susceptibility to peer pres-sure, reward sensitivity, and relatively weakerfuture orientation is mitigated), adolescentsolder than 15 should probably be treated likeadults. In situations in which psychosocial fac-tors are substantially more important, drawingthe boundary at an older age is more appro-priate. This is why my colleagues and I haveargued that it is perfectly reasonable to havea lower boundary for adolescents’ autonomousaccess to abortion (a situation in which manda-tory waiting periods limit the impact ofimpulsivity and shortsightedness and whereconsultation with adults likely counters imma-turity of judgment) than for judgments of crim-inal responsibility (because adolescents’ crimesare often impulsive and influenced by peers)(Steinberg et al. 2009).

Impact of Punitive Sanctions onAdolescent Development and Behavior

As noted above, the increasingly punitive ori-entation of the justice system toward juvenileoffenders has resulted in an increase in the num-ber of juveniles tried and sanctioned as adultsand in the use of harsher sanctions in respond-ing to the delinquent behavior of juvenileswho have been retained in the juvenile justice

www.annualreviews.org • Juvenile Justice 477

Ann

u. R

ev. C

lin. P

sych

ol. 2

009.

5:45

9-48

5. D

ownl

oade

d fr

om w

ww

.ann

ualr

evie

ws.

org

by M

assa

chus

etts

Sch

ool o

f Pr

ofes

sion

al P

sych

olog

y (M

SPP)

on

09/0

8/14

. For

per

sona

l use

onl

y.

ANRV372-CP05-20 ARI 19 February 2009 11:37

Life-course-persistent offenders:antisocial individualswhose offendingbegins beforeadolescence andpersists into adulthood

Age-crime curve: incriminology, therelation between ageand crime, showingthat the prevalence ofcriminal activityincreases betweenpreadolescence andlate adolescence, peaksaround age 17, anddeclines thereafter

system. Research on the impact of adult pros-ecution and punishment and on the use ofpunitive sanctions more generally suggests,however, that these policies and practices mayactually increase recidivism and jeopardize thedevelopment and mental health of juveniles(Fagan 2008). Consequently, there is a growingconsensus among social scientists that policiesand practices, such as setting the minimum ageof criminal court jurisdiction below 18 (as aboutone-third of all states currently do), transferringjuveniles to the adult system for a wide range ofcrimes, including nonviolent crimes, relying onincarceration as a primary means of crime con-trol, and exposing juvenile offenders to punitiveprograms such as boot camps, likely do moreharm than good, cost taxpayers much morethan they need spend on crime prevention,and ultimately pose a threat to public safety(Greenwood 2006).

In order to understand why this is the case,it is important to begin with a distinctionbetween adolescence-limited and life-course-persistent offenders (Moffitt 1993). Dozensof longitudinal studies have shown that thevast majority of adolescents who commit an-tisocial acts desist from such activity as theymature into adulthood and that only a smallpercentage—between five and ten percent, ac-cording to most studies—become chronic of-fenders. Thus, nearly all juvenile offenders areadolescent limited. This observation is borneout in inspection of what criminologists referto as the age-crime curve, which shows that theincidence of criminal activity increases betweenpreadolescence and late adolescence, peaks atabout age 17 (slightly younger for nonviolentcrimes and slightly older for violent ones), anddeclines thereafter. These findings, at both theindividual and aggregate level, have emergedfrom many studies that have been conductedin different historical epochs and around theworld (Piquero et al. 2003).

In view of the fact that most juvenile offend-ers mature out of crime (and that most will de-sist whether or not they are caught, arrested,prosecuted, or sanctioned), one must thereforeask how to best hold delinquent youth respon-

sible for their actions and deter future crime(both their own and that of others) without ad-versely affecting their mental health, psycho-logical development, and successful transitioninto adult roles. If the sanctions to which juve-nile offenders are exposed create psychologicaldisturbance, stunt the development of cognitivegrowth and psychosocial maturity, and interferewith the completion of schooling and entranceinto the labor force, these policies are likely toexacerbate rather than ameliorate many of thevery factors that lead juveniles to commit crimesin the first place (mental illness, difficulties inschool or work, and, as reviewed above, psycho-logical immaturity).

It is clear that sanctioning adolescents asadults is counterproductive. One group of re-searchers examining this question compared agroup of 2700 Florida youths transferred tocriminal court, mostly based on prosecutors’discretionary authority under Florida’s direct-file statute, with a matched group of youthsretained in the juvenile system (Bishop &Frazier 2000). In another study, the researcherscompared 15- and 16-year-olds charged withrobbery and burglary in several counties inmetropolitan New York and in demographi-cally similar counties in New Jersey. The le-gal settings differed in that New York juvenilesage 15 and older who are charged with rob-bery and burglary are automatically dealt within the adult system under that state’s legislativewaiver statute, whereas in New Jersey, transferis rarely used, and the juvenile court retains ju-risdiction over almost all youths charged withthese crimes (Fagan 1996).

The New York-New Jersey study found thatyouths convicted of robbery in criminal courtwere rearrested and incarcerated at a higherrate than those who were dealt with in the ju-venile system, but that rates were comparablefor burglary, a less serious crime. The studyalso examined the number of days until rear-rest and found a similar pattern; the youths sen-tenced for robbery in criminal court reoffendedsooner than did their juvenile court coun-terparts. Recidivism was not affected by sen-tence length; longer sentences were not more

478 Steinberg

Ann

u. R

ev. C

lin. P

sych

ol. 2

009.

5:45

9-48

5. D

ownl

oade

d fr

om w

ww

.ann

ualr

evie

ws.

org

by M

assa

chus

etts

Sch

ool o

f Pr

ofes

sion

al P

sych

olog

y (M

SPP)

on

09/0

8/14

. For

per

sona

l use

onl

y.

ANRV372-CP05-20 ARI 19 February 2009 11:37

effective at reducing recidivism than wereshorter sentences. Results of the Florida studyalso support the conclusion that juvenile sanc-tions may reduce recidivism more effectivelythan criminal punishment. This study mea-sured only rearrest rates and found lower ratesfor youths who were retained in juvenile courtthan for youths who were transferred. Thefollow-up period in this study was relativelybrief—less than two years. During this period,transferred youth were more likely to be rear-rested, committed more offenses per year, andreoffended sooner than did juveniles in the ju-venile system. As in the New York-New Jerseystudy, longer sentences did not have a deterrenteffect.

Within the juvenile system, of course, thereis wide variation in the types and severity ofsanctions to which offenders are exposed. Someyouths are incarcerated in prison-like trainingschools, whereas others receive loosely super-vised community probation—neither of whichis effective at changing antisocial behavior. Animportant question therefore is, what can thejuvenile system offer young offenders that willbe effective at reducing recidivism? A detaileddiscussion of the enormous literature evaluat-ing the effects of various sanctions and inter-ventions is beyond the scope of this review, andthis literature has been summarized many times(Greenwood 2006, Lipsey 1999). Here I high-light a few main points.

Until the 1990s, most researchers who studyjuvenile delinquency programs might well haveanswered that the system had little to offerin the way of effective therapeutic interven-tions; the dominant view held by social scien-tists in the 1970s and 1980s was that “nothingworks” to reduce recidivism with young offend-ers. Today the picture is considerably brighter,in large part due to a substantial body of re-search produced over the past 15 years showingthat many juvenile programs, in both commu-nity and institutional settings, have a substantialcrime-reduction effect; for the most promisingprograms, that effect is in the range of 20%to 30%. An increased focus on research-basedprograms and on careful outcome evaluation al-

lows policy makers to assess accurately the im-pact on recidivism rates of particular programsto determine whether the economic costs arejustified. In a real sense, these developmentshave revived rehabilitation as a realistic goal ofjuvenile justice interventions.

In general, successful programs are thosethat attend to the lessons of developmentalpsychology, seeking to provide young offend-ers with supportive social contexts and to as-sist them in acquiring the skills necessary tochange problem behavior and to attain psy-chosocial maturity. In his comprehensive meta-analysis of 400 juvenile programs, Lipsey (1995)found that among the most effective programsin both community and institutional settingswere those that focused on improving social de-velopment skills in the areas of interpersonalrelations, self-control, academic performance,and job skills. Some effective programs focusdirectly on developing skills to avoid antisocialbehavior, often through cognitive behavioraltherapy. Other interventions that have beenshown to have a positive effect on crime re-duction focus on strengthening family support,including Multisystemic Therapy, FunctionalFamily Therapy, and Multidimensional Treat-ment Foster Care, all of which are both effec-tive and cost effective (Greenwood 2006). It isalso clear from these reviews that punitive sanc-tions administered within the juvenile systemhave iatrogenic effects similar to those seen instudies of juveniles tried as adults. Punishment-oriented approaches, such as “Scared Straight”or military-style boot camps, do not deter fu-ture crime and may even inadvertently pro-mote reoffending. Nor do these programs ap-pear to deter other adolescents from offending(Greenwood 2006).

The dearth of evidence supporting the effec-tiveness of tough sanctions in deterring youth-ful criminal activity becomes less puzzling whenwe consider the response of young offend-ers to harsh punishment in light of devel-opmental knowledge about adolescence dis-cussed earlier. Teenagers on the street decidingwhether to hold up a convenience store maysimply be less capable than adults, due to their

www.annualreviews.org • Juvenile Justice 479

Ann

u. R

ev. C

lin. P

sych

ol. 2

009.

5:45

9-48

5. D

ownl

oade

d fr

om w

ww

.ann

ualr

evie

ws.

org

by M

assa

chus

etts

Sch

ool o

f Pr

ofes

sion

al P

sych

olog

y (M

SPP)

on

09/0

8/14

. For

per

sona

l use

onl

y.

ANRV372-CP05-20 ARI 19 February 2009 11:37

psychosocial immaturity, of considering thesanctions they will face. Thus, the develop-mental influences on decision making that mit-igate culpability also may make adolescents lessresponsive to the threat of criminal sanctions(Scott & Steinberg 2008).

In addition, adolescence is a formative pe-riod of development. In mid and late adoles-cence, individuals normally make substantialprogress in acquiring and coordinating skillsthat are essential to filling the conventionalroles of adulthood. First, they begin to developbasic educational and vocational skills to en-able them to function in the workplace as pro-ductive members of society. Second, they alsoacquire the social skills necessary to establishstable intimate relationships and to cooperatein groups. Finally, they must begin to learnto behave responsibly without external super-vision and to set meaningful personal goals forthemselves. For most individuals, the processof completing these developmental tasks ex-tends into early adulthood, but making sub-stantial progress during the formative stage ofadolescence is important. This process of de-velopment toward psychosocial maturity is oneof reciprocal interaction between the individ-ual and her social context. Several environmen-tal conditions are particularly important, suchas the presence of an authoritative parent orguardian, association with prosocial peers, andparticipation in educational, extracurricular, oremployment activities that facilitate the devel-opment of autonomous decision making andcritical thinking. For the youth in the justicesystem, the correctional setting becomes theenvironment for social development and mayaffect whether he acquires the skills necessaryto function successfully in conventional adultroles (Steinberg et al. 2004).

Normative teenagers who get involved incrime do so, in part, because their choices aredriven by developmental influences typical ofadolescence. In theory, they should desist fromcriminal behavior and mature into reasonablyresponsible adults as they attain psychosocialmaturity—and most do, especially as they en-ter into adult work and family responsibilities.

Whether youths successfully make the transi-tion to adulthood, however, depends in parton whether their social context provides op-portunity structures for the completion of thedevelopmental tasks described above. The cor-rectional environment may influence the tra-jectories of normative adolescents in the justicesystem in important ways. Factors such as theavailability (or lack) of good educational, skillbuilding, and rehabilitative programs; the at-titudes and roles of adult supervisors; and theidentity and behavior of other offenders shapethe social context of youths in both the adultand the juvenile systems. These factors may af-fect the inclination of young offenders to de-sist or persist in their criminal activities andmay facilitate or impede their development intoadults who can function adequately in society—in the workplace, in marriage or other intimateunions, and as citizens.

SUMMARY AND CONCLUDINGCOMMENTS

The overarching question I pose in this articleis whether research on adolescent developmentindicates that adolescents and adults differ inways that warrant their differential treatmentwhen they violate the law. More specifically,I ask how this research informs debate aboutthree fundamental questions that continue tochallenge the justice system: (a) Should ado-lescents be held to adult standards of crimi-nal culpability and, accordingly, exposed to thesame punishment as adults; (b) Do adolescentspossess the necessary capabilities to function ascompetent defendants in an adversarial courtproceeding; and (c) How are juvenile offendersaffected by the sorts of punitive sanctions thatbecame increasingly popular during the pastseveral decades?

It is now incontrovertible that psychologi-cal development continues throughout adoles-cence and into young adulthood in ways that arerelevant to all three questions. Although basiccognitive competence matures by the time in-dividuals reach age 16, many of the social andemotional capacities that influence adolescents’

480 Steinberg

Ann

u. R

ev. C

lin. P

sych

ol. 2

009.

5:45

9-48

5. D

ownl

oade

d fr

om w

ww

.ann

ualr

evie

ws.

org

by M

assa

chus

etts

Sch

ool o

f Pr

ofes

sion

al P

sych

olog

y (M

SPP)

on

09/0

8/14

. For

per

sona

l use

onl

y.

ANRV372-CP05-20 ARI 19 February 2009 11:37

judgment and decision making, especially out-side the psychologist’s laboratory, continue tomature into late adolescence and beyond. Com-pared to individuals in their mid to late twenties,adolescents even as old as 18 are more impul-sive, less oriented to the future, and more sus-ceptible to the influence of their peers. In addi-tion, because adolescence is also period duringwhich individuals are still acquiring the psycho-logical capacities they will need to successfullytransition into adult work and family roles, it isimportant that the sanctions to which juvenileoffenders are exposed not adversely affect theirdevelopment. Recent research on the neuralunderpinnings of these developments does notchange the portrait of adolescent immaturitypainted by behavioral research, but it does adddetail and support to the argument that makesthe story more compelling. It is one thing to saythat adolescents don’t control their impulses,stand up to peer pressure, or think through theconsequences of their actions as well as adults;it is quite another to say that don’t because theycan’t.

Because American criminal law clearly pro-vides that diminished judgment mitigates crim-inal responsibility, it is reasonable to arguethat adolescents are inherently less blamewor-thy than their elders in ways should affect deci-sions about criminal punishment; as a class, ado-lescents are inherently less blameworthy thanadults. The picture that emerges from an anal-ysis of the capacities necessary for competenceto stand trial is not the same, however. Here therelevant research indicates that some adoles-cents (generally, those 16 and older) have adult-

like capabilities but that others (generally those15 and younger) may not. Research on the im-pact of punitive sanctions on adolescent devel-opment and behavior, although not explicitlydevelopmental in nature, indicates that tryingadolescents as adults or exposing them to espe-cially harsh sanctions does little to deter offend-ing and may indeed have iatrogenic effects.

Although justice system policy and practicecannot, and should not, be dictated solely bystudies of adolescent development, the ways inwhich we respond to juvenile offending shouldat the very least be informed by the lessonsof developmental science. Taken together, thelessons of developmental science offer strongsupport for the maintenance of a separate ju-venile justice system in which adolescents arejudged, tried, and sanctioned in developmen-tally appropriate ways. Using developmentalscience to inform juvenile justice policy is nota panacea that will solve the problem of youthcrime. Adolescents will always get in trouble,sometimes very serious trouble, and some willcontinue to offend, despite the state’s best ef-forts to respond to their crimes in ways that willdeter future offending. At the same time, the fu-ture prospects of some youths will be harmedby a system that holds them to adult levels ofaccountability for their crimes under our trans-fer rules. No one policy regime will yield goodoutcomes for all young offenders, but lookingto developmental research to guide our decisionmaking provides a solid framework for policiesand practices that will enhance public safety inthe long run by promoting healthy adolescentdevelopment.

SUMMARY POINTS

1. During the past two decades, policies and practices concerning the treatment of juvenileoffenders in the United States became increasingly punitive, as evidenced by the increasein the number of juveniles tried as adults and the expanded use of harsh sanctions withinboth the juvenile and criminal justice systems. This was a break from the traditionalmodel of juvenile justice, which emphasized rehabilitation rather than punishment as itscore purpose, that had prevailed for most of the twentieth century.

www.annualreviews.org • Juvenile Justice 481

Ann

u. R

ev. C

lin. P

sych

ol. 2

009.

5:45

9-48

5. D

ownl

oade

d fr

om w

ww

.ann

ualr

evie

ws.

org

by M

assa

chus

etts

Sch

ool o

f Pr

ofes

sion

al P

sych

olog

y (M

SPP)

on

09/0

8/14

. For

per

sona

l use

onl

y.

ANRV372-CP05-20 ARI 19 February 2009 11:37

2. In order to make well-informed decisions about the treatment of juveniles who haveentered the juvenile justice pipeline, therefore, policymakers, practitioners, and mentalhealth professionals need to be familiar with the developmental changes that occur duringchildhood and adolescence in the capabilities and characteristics that are relevant totheir competence to stand trial, their criminal culpability, and their likely response totreatment.

3. Brain maturation continues well into young adulthood, and although individuals, onaverage, perform at adult levels on tests of basic cognitive ability by the time they are16, most do not attain adult-like levels of social and emotional maturity until very late inadolescence or early in adulthood. Compared to adults, adolescents are more susceptibleto peer influence, less oriented to the future, more sensitive to short-term rewards, andmore impulsive.

4. This research on adolescent brain, cognitive, and psychosocial development supports theview that adolescents are fundamentally different from adults in ways that warrant theirdifferential treatment in the justice system. An analysis of factors that mitigate criminalresponsibility under the law indicates that adolescents are inherently less culpable than areadults and should therefore be punished less severely. In addition, studies of competenceto stand trial indicate that those who are under 16 are more likely to be incompetent thanare adults, raising questions about the appropriateness of trying younger adolescents incriminal court.

5. Studies of the impact of punitive sanctions on adolescent development and behavior,including prosecuting and sanctioning adolescents as adults, indicate that they do notdeter adolescents from breaking the law and may in fact increase recidivism. In contrast,family-based interventions have been shown to be both effective and cost effective.

DISCLOSURE STATEMENT

The author is not aware of any biases that might be perceived as affecting the objectivity of thisreview.

ACKNOWLEDGMENTS

Work on this review was supported by the John D. and Catherine T. MacArthur Foundation. I amgrateful to Elizabeth Cauffman, Thomas Grisso, Elizabeth Scott, and Robert Schwartz for theirpermission to draw on our collaborative work in the preparation of this review.

LITERATURE CITED

Belter R, Grisso T. 1984. Children’s recognition of rights violations in counseling. Prof. Psychol. Res. Pract.15:899–910

Berg W, Byrd D. 2002. The Tower of London spatial problem solving task: enhancing clinical and researchimplementation. J. Exp. Clin. Neuropsychol. 25:586–604

Bishop D, Frazier C. 2000. Consequences of transfer. In The Changing Borders of Juvenile Justice, ed. J Fagan,F Zimring, pp. 227–77. Chicago: Univ. Chicago Press

Casey BJ, Getz S, Galvan A. 2008. The adolescent brain. Dev. Rev. 28:62–77

482 Steinberg

Ann

u. R

ev. C

lin. P

sych

ol. 2

009.

5:45

9-48

5. D

ownl

oade

d fr

om w

ww

.ann

ualr

evie

ws.

org

by M

assa

chus

etts

Sch

ool o

f Pr

ofes

sion

al P

sych

olog

y (M

SPP)

on

09/0

8/14

. For

per

sona

l use

onl

y.

ANRV372-CP05-20 ARI 19 February 2009 11:37

Cauffman E, Shulman E, Claus E, Banich M, Steinberg L, et al. 2008. Responding to reward versus punishment:how adolescents differ from adults in performance on the Iowa Gambling Task. Dept. Psychol. SocialBehav., Univ. Calif., Irvine. Manuscr. under review

Cauffman E, Steinberg L, Piquero A. 2005. Psychological, neuropsychological, and psychophysiological cor-relates of serious antisocial behavior in adolescence: the role of self-control. Criminology 43:133–76

Collins WA, Steinberg L. 2006. Adolescent development in interpersonal context. In Social, Emotional, andPersonality Development. Handbook of Child Psychology, ed. W Damon, R Lerner, N Eisenberg, pp. 1003–67.New York: Wiley

Conklin H, Luciana M, Hooper C, Yarger R. 2007. Working memory performance in typically developingchildren and adolescents: behavioral evidence of protracted frontal lobe development. Dev. Neuropsychol.31:103–28

Cooper D. 1997. Juveniles’ understanding of trial-related information: Are they competent defendants? Behav.Sci. Law 15:167–80

Crone EA, Bunge SA, Latenstein H, Van Der Molen MW. 2005. Characterization of children’s decisionmaking: sensitivity to punishment frequency, not task complexity. Child Neuropsychol. 11:245–63

Crone EA, Van Der Molen MW. 2004. Developmental changes in real life decision making: performance ona gambling task previously shown to depend on the ventromedial prefrontal cortex. Dev. Neuropsychol.25:251–79

Dahl R. 2001. Affect regulation, brain development, and behavioral/emotional health in adolescence.CNS Spectr. 6:1–12

Dahl R. 2004. Adolescent brain development: a period of vulnerabilities and opportunities. Ann. N. Y. Acad.Sci. 1021:1–22

Fagan J. 1996. The comparative impacts of juvenile and criminal court sanctions on adolescent felonyoffenders. Law Policy Rev. 18:77–119

Provides an excellentsummary of research onthe impact of tryingjuveniles as adults onadolescents’ behavior,mental health, andrecidivism.

Fagan J. 2008. Juvenile crime and criminal justice: resolving border disputes. Fut. Child. 18(2):81–118Feld BC. 1993. Criminalizing the American juvenile court. Crime Just. 17:197–280Galvan A, Hare T, Voss H, Glover G, Casey BJ. 2007. Risk-taking and the adolescent brain: Who is at risk?

Dev. Sci. 10:F8–14Gardner M, Steinberg L. 2005. Peer influence on risk-taking, risk preference, and risky decision-making in

adolescence and adulthood: an experimental study. Dev. Psychol. 41:625–35Gardner W. 1993. A life-span rational choice theory of risk taking. In Adolescent Risk Taking, ed. N Bell,

R Bell, pp. 66–83. Newbury Park, CA: SageGreene A. 1986. Future time perspective in adolescence: the present of things future revisited. J. Youth Adolesc.

15:99–113Furnishes acomprehensive analysisof the effectiveness ofvarious approaches topreventing and treatingjuvenile delinquency.

Greenwood P. 2006. Changing Lives: Delinquency Prevention as Crime Control Policy. Chicago: Univ.Chicago Press

Grisso T. 1981. Juveniles’ Waiver of Rights: Legal and Psychological Competence. New York: PlenumGrisso T, Vierling L. 1978. Minors’ consent to treatment: a developmental perspective. Prof. Psychol. 9:412–26

Landmark empiricalstudy that demonstratesthat in comparison toadults, individualsunder 16 are morelikely to be incompetentto stand trial.

Grisso T, Steinberg L, Woolard J, Cauffman E, Scott E, et al. 2003. Juveniles’ competence to standtrial: a comparison of adolescents’ and adults’ capacities as trial defendants. Law Hum. Behav.

27:333–63Grosbras M, Jansen M, Leonard G, McIntosh A, Osswald K, et al. 2007. Neural mechanisms of resistance to

peer influence in early adolescence. J. Neurosci. 27:8040–45Gustafson K, McNamara J. 1987. Confidentiality with minor clients: issues and guidelines for therapists.

Prof. Psychol. Res. Pract. 18:503–8Hale S. 1990. A global developmental trend in cognitive processing speed. Child Dev. 61:653–63Hooper C, Luciana M, Conklin H, Yarger R. 2004. Adolescents’ performance on the Iowa Gam-

bling Task: implications for the development of decision making and ventromedial prefrontal cortex.Dev. Psychol. 40:1148–58

Kail R. 1997. Processing time, imagery, and spatial memory. J. Exp. Child Psychol. 64:67–78Keating D. 2004. Cognitive and brain development. In Handbook of Adolescent Psychology, ed. R Lerner,

L Steinberg, pp. 45–84. New York: Wiley. 2nd ed.

www.annualreviews.org • Juvenile Justice 483

Ann

u. R

ev. C

lin. P

sych

ol. 2

009.

5:45

9-48

5. D

ownl

oade

d fr

om w

ww

.ann

ualr

evie

ws.

org

by M

assa

chus

etts

Sch

ool o

f Pr

ofes

sion

al P

sych

olog

y (M

SPP)

on

09/0

8/14

. For

per

sona

l use

onl

y.

ANRV372-CP05-20 ARI 19 February 2009 11:37

Krisberg B, Austin J. 1993. Reinventing Juvenile Justice. Newbury Park, CA: SageKuhn D. 2009. Adolescent thinking. See Lerner & Steinberg 2009. In pressLerner R, Steinberg L, eds. 2009. Handbook of Adolescent Psychology. New York: Wiley. 3rd ed. In pressLeshem R, Glicksohn J. 2007. The construct of impulsivity revisited. Personal. Individ. Differ. 43:681–91Lipsey M. 1995. What do we learn from 400 research studies on the effectiveness of treatment with juvenile

delinquents? In What Works? Reducing Reoffending, ed. J McGuire, pp. 63–78. New York: WileyLipsey M. 1999. Can rehabilitative programs reduce the recidivism of young offenders? An inquiry into the

effectiveness of practical programs. Va. J. Soc. Policy Law 6:611–41Luna B, Thulborn K, Munoz D, Merriam E, Garver K, et al. 2001. Maturation of widely distributed brain

function subserves cognitive development. Neuroimage 13:786–93Mack J. 1909. The juvenile court. Harv. Law Rev. 23:104–22Millstein S, Halpern-Felsher B. 2002. Perceptions of risk and vulnerability. J. Adolesc. Health 31S:10–27Moffitt T. 1993. Adolescence-limited and life-course persistent antisocial behavior: a developmental taxonomy.

Psychol. Rev. 100:674–701Mulvey E, Leistico A. 2008. Improving professional judgments of risk and amenability in juvenile justice. Fut.

Child. 18(2):35–58Nurmi J. 1991. How do adolescents see their future? A review of the development of future orientation and

planning. Dev. Rev. 11:1–59Overton W. 1990. Competence and procedures: constraints on the development of logical reasoning. In

Reasoning, Necessity, and Logic: Developmental Perspectives, ed. W Overton, pp. 1–32. Hillsdale, NJ: ErlbaumPaus T. 2005. Mapping brain maturation and cognitive development during adolescence. Trends Cogn. Sci.

9:60–68Paus T, Toro R, Leonard G, Lerner J, Lerner R, et al. 2008. Morphological properties of the action-observation

cortical network in adolescents with low and high resistance to peer influence. Soc. Neurosci. In pressPeterson-Badali M, Abramovitch R. 1993. Grade related changes in young people’s reasoning about plea

decisions. Law Hum. Behav. 17:537–52Piquero A, Farrington D, Blumstein A. 2003. The criminal career paradigm: background and recent develop-

ments. Crime Just. 30:359–506Platt A. 1977. The Child Savers: The Invention of Delinquency. Chicago: Univ. Chicago Press. 2nd ed.Roper v. Simmons, 541 U.S. 1040 2005

Provides a legal analysisof how the justicesystem might best takethe developmentalincompetence ofjuveniles into account.Argues that a lowerstandard of competenceshould be used injuvenile than in criminalcourt.

Scott E, Grisso T. 2005. Developmental incompetence, due process, and juvenile justice policy. N. C.

Law Rev. 83:793–846Scott E, Reppucci N, Woolard J. 1995. Evaluating adolescent decision making in legal contexts. Law Hum.

Behav. 19:221–44

Calls for juvenile justicereform based on thescientific study ofadolescentdevelopment. Suppliesuseful summaries ofliteratures onadolescents’ criminalculpability, competenceto stand trial, andresponse tointervention.

Scott E, Steinberg L. 2008. Rethinking Juvenile Justice. Cambridge, MA: Harvard Univ. PressSisk C, Foster D. 2004. The neural basis of puberty and adolescence. Nat. Neurosci. 7:1040–47Sisk C, Zehr J. 2005. Pubertal hormones organize the adolescent brain and behavior. Front. Neuroendocrinol.

26:163–74Smetana J, Campione-Barr N, Metzger A. 2006. Adolescent development in interpersonal and societal con-

texts. Annu. Rev. Psychol. 57:255–84Steinberg L. 2004. Risk-taking in adolescence: What changes, and why? Ann. N. Y. Acad. Sci. 1021:51–58Steinberg L. 2007. Risk-taking in adolescence: new perspectives from brain and behavioral science.

Curr. Dir. Psychol. Sci. 16:55–59

Discusses how braindevelopment inadolescence affects risktaking and recklessbehavior, in which theheightened vulnerabilityof middle adolescence ishighlighted.

Steinberg L. 2008. A social neuroscience perspective on adolescent risk-taking. Dev. Rev. 28:78–106Steinberg L, Albert D, Cauffman E, Banich M, Graham S, Woolard J. 2008a. Age differences in sensa-

tion seeking and impulsivity as indexed by behavior and self-report: evidence for a dual systems model.Dev. Psychol. In press

Steinberg L, Cauffman E. 1996. Maturity of judgment in adolescence: psychosocial factors in adolescentdecisionmaking. Law Hum. Behav. 20:249–72

Steinberg L, Cauffman E, Woolard J, Graham S, Banich M. 2009. Are adolescents less mature than adults?Minors’ access to abortion, the juvenile death penalty, and the alleged APA “flip-flop.” Am. Psychol. Inpress

484 Steinberg

Ann

u. R

ev. C

lin. P

sych

ol. 2

009.

5:45

9-48

5. D

ownl

oade

d fr

om w

ww

.ann

ualr

evie

ws.

org

by M

assa

chus

etts

Sch

ool o

f Pr

ofes

sion

al P

sych

olog

y (M

SPP)

on

09/0

8/14

. For

per

sona

l use

onl

y.

ANRV372-CP05-20 ARI 19 February 2009 11:37

Steinberg L, Chung H, Little M. 2004. Reentry of young offenders from the justice system: a developmentalperspective. Youth Violence Juv. Just. 1:21–38

Steinberg L, Graham S, O’Brien L, Woolard J, Cauffman E, Banich M. 2008b. Age differences in futureorientation and delay discounting. Child Dev. In press

Steinberg L, Monahan K. 2007. Age differences in resistance to peer influence. Dev. Psychol. 43:1531–43Steinberg L, Schwartz R. 2000. Developmental psychology goes to court. In Youth on Trial: A Developmental

Perspective on Juvenile Justice, ed. T Grisso, R Schwartz, pp. 9–31. Chicago: Univ. Chicago Press

Discusses whyadolescents, by virtue ofdevelopmentalimmaturity, areinherently less culpablethan adults. Citedmultiple times by U.S.Supreme Court in itsdecision to abolishjuvenile death penalty.

Steinberg L, Scott E. 2003. Less guilty by reason of adolescence: developmental immaturity, dimin-ished responsibility, and the juvenile death penalty. Am. Psychol. 58:1009–18

Teicher M, Andersen S, Hostetter J. 1995. Evidence for dopamine receptor pruning between adolescence andadulthood in striatum but not nucleus accumbens. Dev. Brain Res. 89:167–72

Weithorn L, Campbell S. 1982. The competency of children and adolescents to make informed treatmentdecisions. Child Dev. 53:1589–98

Zimring F. 1998. American Youth Violence. New York: Oxford Univ. Press

www.annualreviews.org • Juvenile Justice 485

Ann

u. R

ev. C

lin. P

sych

ol. 2

009.

5:45

9-48

5. D

ownl

oade

d fr

om w

ww

.ann

ualr

evie

ws.

org

by M

assa

chus

etts

Sch

ool o

f Pr

ofes

sion

al P

sych

olog

y (M

SPP)

on

09/0

8/14

. For

per

sona

l use

onl

y.

AR372-FM ARI 6 March 2009 17:50

Annual Review ofClinical Psychology

Volume 5, 2009Contents

Construct Validity: Advances in Theory and MethodologyMilton E. Strauss and Gregory T. Smith � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 1

Item Response Theory and Clinical MeasurementSteven P. Reise and Niels G. Waller � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � �27

Methodological Issues in Molecular Genetic Studiesof Mental DisordersCarrie E. Bearden, Anna J. Jasinska, and Nelson B. Freimer � � � � � � � � � � � � � � � � � � � � � � � � � � � � � �49

Statistical Methods for Risk-Outcome Research: Being Sensitiveto Longitudinal StructureDavid A. Cole and Scott E. Maxwell � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � �71

Psychological Treatment of Anxiety: The Evolution of BehaviorTherapy and Cognitive-Behavior TherapyS. Rachman � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � �97

Computer-Aided Psychological Treatments: Evolving IssuesIsaac Marks and Kate Cavanagh � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 121

The Past, Present, and Future of HIV Prevention: IntegratingBehavioral, Biomedical, and Structural Intervention Strategiesfor the Next Generation of HIV PreventionMary Jane Rotheram-Borus, Dallas Swendeman, and Gary Chovnick � � � � � � � � � � � � � � � � � � 143

Evolving Prosocial and Sustainable Neighborhoods and CommunitiesAnthony Biglan and Erika Hinds � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 169

Five-Factor Model of Personality Disorder: A Proposal for DSM-VThomas A. Widiger and Stephanie N. Mullins-Sweatt � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 197

Differentiating the Mood and Anxiety Disorders: A QuadripartiteModelDavid Watson � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 221

When Doors of Perception Close: Bottom-Up Models of DisruptedCognition in SchizophreniaDaniel C. Javitt � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 249

vii

Ann

u. R

ev. C

lin. P

sych

ol. 2

009.

5:45

9-48

5. D

ownl

oade

d fr

om w

ww

.ann

ualr

evie

ws.

org

by M

assa

chus

etts

Sch

ool o

f Pr

ofes

sion

al P

sych

olog

y (M

SPP)

on

09/0

8/14

. For

per

sona

l use

onl

y.

AR372-FM ARI 6 March 2009 17:50

The Treatment of Borderline Personality Disorder: Implicationsof Research on Diagnosis, Etiology, and OutcomeJoel Paris � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 277

Development and Etiology of Disruptive and Delinquent BehaviorRolf Loeber, Jeffrey D. Burke, and Dustin A. Pardini � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 291

Anxiety Disorders During Childhood and Adolescence:Origins and TreatmentRonald M. Rapee, Carolyn A. Schniering, and Jennifer L. Hudson � � � � � � � � � � � � � � � � � � � � � � 311

APOE-4 Genotype and Neurophysiological Vulnerabilityto Alzheimer’s and Cognitive AgingSusan Bookheimer and Alison Burggren � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 343

Depression in Older AdultsAmy Fiske, Julie Loebach Wetherell, and Margaret Gatz � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 363

PedophiliaMichael C. Seto � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 391

Treatment of Smokers with Co-occurring Disorders: Emphasis onIntegration in Mental Health and Addiction Treatment SettingsSharon M. Hall and Judith J. Prochaska � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 409

Environmental Influences on Tobacco Use: Evidence from Societaland Community Influences on Tobacco Use and DependenceK. Michael Cummings, Geoffrey T. Fong, and Ron Borland � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 433

Adolescent Development and Juvenile JusticeLaurence Steinberg � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 459

Indexes

Cumulative Index of Contributing Authors, Volumes 1–5 � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 487

Cumulative Index of Chapter Titles, Volumes 1–5 � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 489

Errata

An online log of corrections to Annual Review of Clinical Psychology articles may befound at http://clinpsy.annualreviews.org

viii Contents

Ann

u. R

ev. C

lin. P

sych

ol. 2

009.

5:45

9-48

5. D

ownl

oade

d fr

om w

ww

.ann

ualr

evie

ws.

org

by M

assa

chus

etts

Sch

ool o

f Pr

ofes

sion

al P

sych

olog

y (M

SPP)

on

09/0

8/14

. For

per

sona

l use

onl

y.

AnnuAl Reviewsit’s about time. Your time. it’s time well spent.

AnnuAl Reviews | Connect with Our expertsTel: 800.523.8635 (us/can) | Tel: 650.493.4400 | Fax: 650.424.0910 | Email: service@annualreviews.org

New From Annual Reviews:Annual Review of Organizational Psychology and Organizational BehaviorVolume 1 • March 2014 • Online & In Print • http://orgpsych.annualreviews.org

Editor: Frederick P. Morgeson, The Eli Broad College of Business, Michigan State UniversityThe Annual Review of Organizational Psychology and Organizational Behavior is devoted to publishing reviews of the industrial and organizational psychology, human resource management, and organizational behavior literature. Topics for review include motivation, selection, teams, training and development, leadership, job performance, strategic HR, cross-cultural issues, work attitudes, entrepreneurship, affect and emotion, organizational change and development, gender and diversity, statistics and research methodologies, and other emerging topics.

Complimentary online access to the first volume will be available until March 2015.TAble oF CoNTeNTs:•An Ounce of Prevention Is Worth a Pound of Cure: Improving

Research Quality Before Data Collection, Herman Aguinis, Robert J. Vandenberg

•Burnout and Work Engagement: The JD-R Approach, Arnold B. Bakker, Evangelia Demerouti, Ana Isabel Sanz-Vergel

•Compassion at Work, Jane E. Dutton, Kristina M. Workman, Ashley E. Hardin

•ConstructivelyManagingConflictinOrganizations, Dean Tjosvold, Alfred S.H. Wong, Nancy Yi Feng Chen

•Coworkers Behaving Badly: The Impact of Coworker Deviant Behavior upon Individual Employees, Sandra L. Robinson, Wei Wang, Christian Kiewitz

•Delineating and Reviewing the Role of Newcomer Capital in Organizational Socialization, Talya N. Bauer, Berrin Erdogan

•Emotional Intelligence in Organizations, Stéphane Côté•Employee Voice and Silence, Elizabeth W. Morrison• Intercultural Competence, Kwok Leung, Soon Ang,

Mei Ling Tan•Learning in the Twenty-First-Century Workplace,

Raymond A. Noe, Alena D.M. Clarke, Howard J. Klein•Pay Dispersion, Jason D. Shaw•Personality and Cognitive Ability as Predictors of Effective

Performance at Work, Neal Schmitt

•Perspectives on Power in Organizations, Cameron Anderson, Sebastien Brion

•Psychological Safety: The History, Renaissance, and Future of an Interpersonal Construct, Amy C. Edmondson, Zhike Lei

•Research on Workplace Creativity: A Review and Redirection, Jing Zhou, Inga J. Hoever

•Talent Management: Conceptual Approaches and Practical Challenges, Peter Cappelli, JR Keller

•The Contemporary Career: A Work–Home Perspective, Jeffrey H. Greenhaus, Ellen Ernst Kossek

•The Fascinating Psychological Microfoundations of Strategy and Competitive Advantage, Robert E. Ployhart, Donald Hale, Jr.

•The Psychology of Entrepreneurship, Michael Frese, Michael M. Gielnik

•The Story of Why We Stay: A Review of Job Embeddedness, Thomas William Lee, Tyler C. Burch, Terence R. Mitchell

•What Was, What Is, and What May Be in OP/OB, Lyman W. Porter, Benjamin Schneider

•Where Global and Virtual Meet: The Value of Examining the Intersection of These Elements in Twenty-First-Century Teams, Cristina B. Gibson, Laura Huang, Bradley L. Kirkman, Debra L. Shapiro

•Work–Family Boundary Dynamics, Tammy D. Allen, Eunae Cho, Laurenz L. Meier

Access this and all other Annual Reviews journals via your institution at www.annualreviews.org.

Ann

u. R

ev. C

lin. P

sych

ol. 2

009.

5:45

9-48

5. D

ownl

oade

d fr

om w

ww

.ann

ualr

evie

ws.

org

by M

assa

chus

etts

Sch

ool o

f Pr

ofes

sion

al P

sych

olog

y (M

SPP)

on

09/0

8/14

. For

per

sona

l use

onl

y.

AnnuAl Reviewsit’s about time. Your time. it’s time well spent.

AnnuAl Reviews | Connect with Our expertsTel: 800.523.8635 (us/can) | Tel: 650.493.4400 | Fax: 650.424.0910 | Email: service@annualreviews.org

New From Annual Reviews:

Annual Review of Statistics and Its ApplicationVolume 1 • Online January 2014 • http://statistics.annualreviews.org

Editor: Stephen E. Fienberg, Carnegie Mellon UniversityAssociate Editors: Nancy Reid, University of Toronto

Stephen M. Stigler, University of ChicagoThe Annual Review of Statistics and Its Application aims to inform statisticians and quantitative methodologists, as well as all scientists and users of statistics about major methodological advances and the computational tools that allow for their implementation. It will include developments in the field of statistics, including theoretical statistical underpinnings of new methodology, as well as developments in specific application domains such as biostatistics and bioinformatics, economics, machine learning, psychology, sociology, and aspects of the physical sciences.

Complimentary online access to the first volume will be available until January 2015. table of contents:•What Is Statistics? Stephen E. Fienberg•A Systematic Statistical Approach to Evaluating Evidence

from Observational Studies, David Madigan, Paul E. Stang, Jesse A. Berlin, Martijn Schuemie, J. Marc Overhage, Marc A. Suchard, Bill Dumouchel, Abraham G. Hartzema, Patrick B. Ryan

•The Role of Statistics in the Discovery of a Higgs Boson, David A. van Dyk

•Brain Imaging Analysis, F. DuBois Bowman•Statistics and Climate, Peter Guttorp•Climate Simulators and Climate Projections,

Jonathan Rougier, Michael Goldstein•Probabilistic Forecasting, Tilmann Gneiting,

Matthias Katzfuss•Bayesian Computational Tools, Christian P. Robert•Bayesian Computation Via Markov Chain Monte Carlo,

Radu V. Craiu, Jeffrey S. Rosenthal•Build, Compute, Critique, Repeat: Data Analysis with Latent

Variable Models, David M. Blei•Structured Regularizers for High-Dimensional Problems:

Statistical and Computational Issues, Martin J. Wainwright

•High-Dimensional Statistics with a View Toward Applications in Biology, Peter Bühlmann, Markus Kalisch, Lukas Meier

•Next-Generation Statistical Genetics: Modeling, Penalization, and Optimization in High-Dimensional Data, Kenneth Lange, Jeanette C. Papp, Janet S. Sinsheimer, Eric M. Sobel

•Breaking Bad: Two Decades of Life-Course Data Analysis in Criminology, Developmental Psychology, and Beyond, Elena A. Erosheva, Ross L. Matsueda, Donatello Telesca

•Event History Analysis, Niels Keiding•StatisticalEvaluationofForensicDNAProfileEvidence,

Christopher D. Steele, David J. Balding•Using League Table Rankings in Public Policy Formation:

Statistical Issues, Harvey Goldstein•Statistical Ecology, Ruth King•Estimating the Number of Species in Microbial Diversity

Studies, John Bunge, Amy Willis, Fiona Walsh•Dynamic Treatment Regimes, Bibhas Chakraborty,

Susan A. Murphy•Statistics and Related Topics in Single-Molecule Biophysics,

Hong Qian, S.C. Kou•Statistics and Quantitative Risk Management for Banking

and Insurance, Paul Embrechts, Marius Hofert

Access this and all other Annual Reviews journals via your institution at www.annualreviews.org.

Ann

u. R

ev. C

lin. P

sych

ol. 2

009.

5:45

9-48

5. D

ownl

oade

d fr

om w

ww

.ann

ualr

evie

ws.

org

by M

assa

chus

etts

Sch

ool o

f Pr

ofes

sion

al P

sych

olog

y (M

SPP)

on

09/0

8/14

. For

per

sona

l use

onl

y.