Juveniles' Competence to Stand Trial (PDF)

-XYHQLOHV &RPSHWHQFHWR6WDQG7ULDO$&RPSDULVRQRI$GROHVFHQWV DQG$GXOWV &DSDFLWLHVDV 7ULDO'HIHQGDQWV $XWKRU V 7KRPDV*ULVVR/DXUHQFH6WHLQEHUJ-HQQLIHU:RRODUG(OL]DEHWK&DXIIPDQ (OL]DEHWK6FRWW6DQGUD*UDKDP)UDQ/H[FHQ1'LFNRQ5HSSXFFLDQG5REHUW6FKZDUW] 5HYLHZHGZRUN V  6RXU...

-XYHQLOHV &RPSHWHQFHWR6WDQG7ULDO$&RPSDULVRQRI$GROHVFHQWV DQG$GXOWV &DSDFLWLHVDV 7ULDO'HIHQGDQWV $XWKRU V 7KRPDV*ULVVR/DXUHQFH6WHLQEHUJ-HQQLIHU:RRODUG(OL]DEHWK&DXIIPDQ (OL]DEHWK6FRWW6DQGUD*UDKDP)UDQ/H[FHQ1'LFNRQ5HSSXFFLDQG5REHUW6FKZDUW] 5HYLHZHGZRUN V  6RXUFH/DZDQG+XPDQ%HKDYLRU9RO1R $XJ SS 3XEOLVKHGE\Springer 6WDEOH85/http://www.jstor.org/stable/1394506. $FFHVVHG Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at. http://www.jstor.org/page/info/about/policies/terms.jsp. JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about JSTOR, please contact [email protected].. Springer is collaborating with JSTOR to digitize, preserve and extend access to Law and Human Behavior. http://www.jstor.org This content downloaded on Tue, 8 Jan 2013 23:21:19 PM All use subject to JSTOR Terms and Conditions Law and HumanBehavior,Vol.27, No. 4, August2003 (? 2003) Juveniles' Competence to Stand Trial:A Comparisonof Adolescents' and Adults' Capacitiesas TrialDefendants Thomas Grisso,1,9Laurence Steinberg,2 Jennifer Woolard,3 Elizabeth Cauffman,4 Elizabeth Scott,5 Sandra Graham,6 Fran Lexcen,' N. Dickon Reppucci,7 and Robert Schwartz8 Abilities associated with adjudicative competence were assessed among 927 adoles- cents in juvenile detentionfacilities and community settings.Adolescents' abilities were compared to those of 466 young adults in jails and in the community. Participants at 4 locations across the United States completed a standardized measure of abili- ties relevant for competence to stand trial (the MacArthur Competence Assessment Tool-Criminal Adjudication) as well as a new procedure for assessing psychosocial influences on legal decisions often required of defendants (MacArthur Judgment Eval- uation). Youths aged 15 and younger performed more poorly than young adults, with a greaterproportion manifesting a level of impairment consistent with that of persons found incompetent to stand trial.Adolescents also tended more often than young adults to make choices (e.g., about plea agreements) that reflected compliance with authority, as well as influences ofpsychosocial immaturity.Implications of these resultsforpolicy and practice are discussed, with an emphasis on the development of legal standards that recognize immaturity as a potential predicate of incompetence to stand trial. KEY WORDS: adolescence;legal competence;delinquency;juvenilejustice. During the 1990s, nationwide legal reforms lowered the age at which youths could be tried in adult criminal court and expanded the range of young offenders subject to adult adjudication and punishment (Snyder & Sickmund, 1999); at the same time, the severity of penalties available to the juvenile court increased (Torbet et al., 1996). 1Universityof MassachusettsMedicalSchool,Worcester,Massachusetts. 2Departmentof Psychology,TempleUniversity,Philadelphia,Pennsylvania. 3Departmentof Psychology,GeorgetownUniversity,Washington,DC. 4WesternPsychiatricInstituteand Clinic,Universityof Pittsburgh,Pittsburgh,Pennsylvania. 5Schoolof Law,Universityof Virginia,Charlottesville,Virginia. 6GraduateSchool of Educationand InformationStudies,Universityof California- Los Angeles, Los Angeles,California. 7Departmentof Psychology,Universityof Virginia,Charlottesville,Virginia. 8JuvenileLaw Center,Philadelphia,Pennsylvania. 9Towhomcorrespondenceshouldbe addressedatDepartmentof Psychiatry, Universityof Massachusetts MedicalSchool,55 Lake AvenueNorth,Worcester,Massachusetts01655. 333 0147-7307/03/0800-0333/1 ? 2003 American Psychology-Law Society/Division 41 of the American Psychology Association This content downloaded on Tue, 8 Jan 2013 23:21:19 PM All use subject to JSTOR Terms and Conditions 334 Grisso,Steinberg,Woolard,Cauffman,Scott,Graham,Lexcen,Reppucci,and Schwartz Theselegaldevelopmentsraisean importantissue thathasreceivedsurprisinglylittle attention:whether youths chargedwith crimes have the developmentalcapacities needed to participateeffectivelyin their trials. It is well establishedthata criminalproceedingmeets the constitutionalrequire- ments of due process only when the defendantis competent to stand trial, which includescapacitiesto assistcounsel and to understandthe natureof the proceeding sufficientlyto participatein it andmakedecisionsaboutrightsaffordedalldefendants (Dusky v. US., 1960;Godinez v. Moran,1993). Although courts and legislaturesin some states have determinedthat youthsadjudicatedin juvenileand criminalcourts mustbe competentto standtrial,the conventionalstandardthathas been appliedfo- cuseson mentalillnessanddisability.In general,therehasbeen little recognitionthat youthsin criminalcourtmay be incompetentbecause of developmentalimmaturity (Bonnie & Grisso,2000;Redding& Frost,2002). Until now, little meaningfuldata have been availableregardingthe capacities of adolescentsrelevantfor adjudicativecompetence.A few studies have examined youths' understandingof the nature of trials and trial procedures(for reviews,see Grisso, 1997, 2000;Mumley,Tillbrook,& Grisso,in press), but typicallythey have includedsmallsamplesizes,measuresthat are seriouslylimitedin content or known reliability,or no comparisonbetween the capacitiesof adolescentsand adults.Sim- ilarly,exploratorystudies have found significantage-related increases in youths' performanceon tasks that paralleldecisionsthat defendantsare expected to make (Abramovitch,Peterson-Badali,& Rohan, 1995;Peterson-Badali& Abramovitch, 1993;Peterson-Badali,Abramovitch,& Duda, 1997), but without comparisonto adults. Informationabout youths'competence to stand trial (CST) is needed for sev- eral reasons.First,states need guidancefor the developmentof meaningfullaws in this area.The doctrineregulatingCSThas focused on adultcriminaldefendantsim- pairedby mentalillness and mentalretardation.Yet basicresearchon cognitiveand psychosocialdevelopmentsuggeststhat some youthswill manifestdeficitsin legally relevantabilitiessimilarto deficitsseen in adultswithmentaldisabilities,but for rea- sons of immaturityratherthan mental disorder(see generallyGrisso & Schwartz, 2000).If therewere empiricalevidencefor this,it wouldsuggestthatthe criminallaw shouldtake immaturityinto considerationwhen evaluatingthe adjudicativecompe- tence of youths in criminalcourt. Second, practitionersneed informationabout youths'capacitiesas trial defen- dants,includingtheir CST.Prosecutorsand defense attorneysmust make case-by- case decisionsaboutwhetherto raisethis issue.Mentalhealthprofessionalswho are asked to performevaluationsof youths'CSTneed guidanceregardingthe potential implicationsof youths'developmentalstatusfor assessingdeficitsin the legally rel- evant abilities.This may require attentionto differentconstructs(immaturity,not only disorder)and a differentlogic (e.g., the "achievement"ratherthan "restora- tion"of competenceamongthose foundincompetent)thanin adultCSTevaluations promptedby putativementalillnesses.Finally,judgesneed guidancein interpreting the lawto protectyoungdefendantswho maybe incompetent,especiallyin theirabil- ities to make decisionsto waive importantrightsin the context of their potentially immatureperspectivesregardingthe implicationsof their choices. This content downloaded on Tue, 8 Jan 2013 23:21:19 PM All use subject to JSTOR Terms and Conditions A Comparisonof Adolescents'and Adults'Capacitiesas ThialDefendants 335 Past analyses of the legal concept of adjudicative competence have outlined the specific functional abilities about which the law is concerned in competence cases (Grisso, 2002), as well as their classification into broader psycholegal constructs (e.g., Understanding, Reasoning, Appreciation; see Grisso & Appelbaum, 1998; Hoge et al., 1997; Otto et al., 1998; Poythress et al., 1999). These abilities focus on the fundamental aspects of adjudicative competence, or what Bonnie (1992,1993) called "competence to proceed": a basic comprehension of the purpose and nature of the trial process (Understanding), the capacity to provide relevant information to counsel and to process information (Reasoning), and the ability to apply information to one's own situation in a manner that is neither distorted nor irrational (Appreciation). Some studies have suggested that preadolescents have less knowledge of tri- als and legal concepts than do older adolescents (e.g., Peterson-Badali et al., 1997) or adults (Grisso, 1981). General developmental research on adolescents' cognitive abilities (see Keating, 1990), however, would not lead us to expect substantial dif- ferences between "average" adolescents and adults in their ability to grasp everyday factual or conceptual information that is provided to them, or their ability to cog- nitively process it to make decisions. In addition, results of basic developmental research on youths' cognitive abilities do not necessarily generalize to abilities to deal specifically with trial-related tasks of comprehension and information process- ing, and to the abilities of the population of juveniles facing adjudication for alleged offenses, many of whom are of below-average intelligence. A few studies have em- ployed brief, specialized "competency screening" measures (e.g., Cooper, 1997) to assess the abilities of youths in this relevant population, but these measures typically focus on "understanding,"failing to assess the wider range of abilities associated with trial competence. Bonnie (1992,1993) suggested that, in addition to defendants' basic understand- ing and reasoning abilities, their "decisional competence" may be significant in cases in which defendants must make important decisions about the waiver of consti- tutional rights. A potentially important difference between adolescents and adults in this regard involves maturity of judgment. Differences between adolescents and adults not only are cognitive, but also involve aspects of psychosocial maturation that include progress toward greater future orientation, better risk perception, and less susceptibility to peer influence. Several authors have hypothesized that these devel- opmental factors could result in differences between adolescents' and adults' decision making about important rights in the adjudicative process (Cauffman & Steinberg, 2000; Scott, 1992; Scott, Reppucci, & Woolard, 1995; Steinberg & Cauffman, 1996). Current law does not include these developmental factors as relevant when considering a defendant's adjudicative competence. For example, when making a decision about waiver of important rights, defendants are free to place a primary value on their immediate gratification at the expense of their future welfare, or to opt to please their friends rather than act in their best interests, as long as they adequately understand and grasp the consequences of their choices. But if adolescents place a relatively higher value on immediate gratification than do adults as a consequence of their developmental immaturity, they may make different legal decisions than they themselves would make in their adult years. Although psychosocial immaturity is not addressed in the formal legal construct of competence to stand trial, it needs to This content downloaded on Tue, 8 Jan 2013 23:21:19 PM All use subject to JSTOR Terms and Conditions 336 Grisso,Steinberg,Woolard,Cauffman,Scott,Graham,Lexcen,Reppucci,and Schwartz be investigatedin this context to provide a comprehensiveaccountof adolescents' capacitiesto participatein the trialprocess. In this study, we used two tools to examine the two types of capacitiesout- lined previously.The MacArthurCompetenceAssessment Tool-CriminalAdjudi- cation (MacCAT-CA)focuses primarilyon the formalfunctionalabilitiestypically associatedwith the legal constructof CST(Bonnie's"competenceto proceed").In a large-scalestudy,the instrumentmanifestedmeaningfuldifferencesamongadultde- fendantswith and withoutfindingsof incompetenceto standtrial(Otto et al., 1998), and a MacCAT-CAmanualsubsequentlywas publishedwithadultnorms(Poythress et al., 1999). Although the instrumentis now widely known amongforensicmental healthprofessionals,the extent of its use is not known,and its recentpublicationhas not yet producedadditionalresearchfindings.There are no reportsof its use with youths. In contrastto the MacCAT-CA,the MacArthurJudgmentEvaluation(MacJEN) was designed for this study as a researchtool to examine the question of immatu- rity of judgment,especiallythe potential relationbetween immaturityand choices that defendantsmake in the courseof adjudication.As describedlater,the MacJEN uses three vignettes and structuredinterviewquestionswith objectivecategoriesof responses.It allows for examinationof differencesin choices acrossages, as well as the relation between choices and three psychosocialfactors (risk appraisal,future orientation,and resistanceto peer influence)with theoreticaldevelopmentalsignif- icance (i.e., are theoreticallyexpected to changefromchildhoodto adulthood).The MacJENmeasuresthese factorsas featuresof respondents'reasonsfor theirchoices. We anticipatedthat adolescentsand adultsmightdiffer on these developmentaldi- mensionsin the reasonsfor theirchoices,and thus in the maturityof theirjudgment. Research to addressthe majorpolicy and practicalquestions about juveniles' adjudicativecompetencerequiresdatafromyouthsrepresentingthe relevantadoles- cent age range,as well as from adultsto whom their performancecan be compared. In this study,we selected 11-17 as the adolescentage range,becausevery few youths are arrestedon delinquenciesbelow 11 and becausejuvenilejustice jurisdictionin most states does not exceed 17. Our comparisongroupincludedyoung adultsaged 18-24 because this representsthe age range most commonlyseen by the criminal courts and permitsa legally relevant,althoughconservative,test of the differences between adolescentsand adultsin abilitiesrelatedto adjudicativecompetence. Two groupsof youths and two groupsof young adultswere studied:those cur- rently detainedwithinthe juvenile detentioncenters or adultjails, and those in the communitywith no current(and little past) juvenile or criminalcourt involvement. Wereasonedthatany age differencesin competence-relevantabilitiescouldbe more reliablyinterpretedif they werefoundin bothjusticesystemandcommunitysamples. In summary,this studywas designedto addressthree basic questions: * Do adolescents differ from young adults in abilities to participatein their trials? * If so, whattypes of youthsmanifestsignificantdifferencesfromyoungadults? * Whatkindsof deficitsin abilitiesare salientfor law,policy,andpracticein this area? This content downloaded on Tue, 8 Jan 2013 23:21:19 PM All use subject to JSTOR Terms and Conditions A Comparisonof Adolescents'and Adults'Capacitiesas TrialDefendants 337 METHOD Greaterdetailregardingthe study'ssample,instruments,andproceduresis avail- able in an archivalreport of the study's method, available at www.mac-adoldev- juvjustice.org. Participantsand Sites The studyused a four-groupdesignof adolescentsandyoungadults,drawnfrom the justice system and the general community.Participantsincluded 927 "youths" aged 11-17 and 466 "youngadults"ages 18-24. (Youths'ages were furthergrouped 11-13, 14-15, and 16-17 during data analysis.)"Detained"participants(detained youths = 453;detainedadults= 233) residedin juveniledetentionfacilitiesor adult jails."Community"participants(communityyouths= 474;communityadults= 233) were individualsresidingin the same or similarcommunitiesas the Detained partici- pants,andwho reportedduringrecruitmentthat they had neverbeen held overnight in a justice system facility and were not currentlychargedwith any offenses. (Re- cruitmentof these participantsis describedunder"Procedure.") These 1,393participantswere among1,429individualsoriginallyenrolledin the study;10 participantswere droppedfrom the analysesbecause of excessivemissing data, 2 because they were 25 years of age, and 24 because they obtainedIQ scores below 60 (for whichone of the study'sdependentmeasureshad not been normed). Tables1 and 2 show the demographiccompositionof the study groups.Males composed66.3%of the Detained sampleand 56.8%of the Communitysample.The ethniccompositionof the samplewas about40%AfricanAmerican,23% Hispanic, 35%non-HispanicWhite,1%Asian, and 1%fromotherethnicidentities;these pro- portionswere similaracrossage andDetained/Communitygroups.Mostparticipants in both the Detained samples(75 and 77%) and Communitysamples(62 and 73%) were classifiedin the two lowest socioeconomicclasses by using the Hollingshead Table1. SampleDemographics Detained Community Youthage groups Youthadults Youthage groups Youthadults 11-13 14-15 16-17 11-17 18-24 11-13 14-15 16-17 11-17 18-24 Participants(n) 74 186 193 453 233 116 159 199 474 233 Male (% of age group) 74 62 62 64 71 52 60 57 57 57 Ethnicity(% of age group) AfricanAmerican 56 32 38 39 43 41 52 33 41 37 Hispanic 21 28 25 26 25 20 20 21 20 24 Non-HispanicWhite 21 35 35 32 32 36 28 44 37 37 Asian and Other 2 5 2 3 0 3 0 2 2 2 SocioeconomicStatus (% of age group) I-II 8 7 11 9 7 15 13 15 14 9 III 12 16 18 16 16 23 26 24 24 18 IV-V 80 77 71 75 77 62 61 61 62 73 This content downloaded on Tue, 8 Jan 2013 23:21:19 PM All use subject to JSTOR Terms and Conditions 338 Grisso,Steinberg,Woolard,Cauffman,Scott,Graham,Lexcen,Reppucci,and Schwartz Table2. Numberof Participantsin Gender/Ethnicityby Age Groups(% of Age Groupin Parentheses) Detained Community 11-13 14-15 16-17 18-24 11-13 14-15 16-17 18-24 Males AfricanAmerican 27 (37) 40 (22) 47 (24) 70 (30) 24 (21) 49 (31) 32 (16) 47 (20) Hispanic 12 (16) 30(16) 25 (13) 47 (20) 15 (13) 16(10) 22(11) 32(14) Non-HispanicWhite 13 (18) 41(22) 46 (24) 48 (21) 20 (17) 30 (19) 55 (28) 50 (21) Asian andother 2(3) 4 (2) 2(1) 0 1 (1) 1 (1) 4(2) 3 (1) Females AfricanAmerican 14 (19) 20 (11) 27 (14) 29 (12) 23 (20) 33 (21) 33 (17) 39 (17) Hispanic 3 (4) 23 (12) 23 (12) 12 (5) 8 (7) 15 (9) 19 (10) 24 (10) Non-HispanicWhite 2 (3) 24 (13) 21 (11) 27 (12) 22 (19) 15 (9) 33 (17) 37 (16) Asian andother 0 4 (2) 1 (1) 0 3 (2) 0 0 1 (1) (1975) system. Current charges for the Detained youths were primarily (about 80%) offenses against persons and offenses against property, in about equal proportions. Charges for the Detained adults were primarily (about 80%) drug-related, against persons, or against property, in about equal proportions. The distribution of charges did not differ between the youth and young adult groups, except that drug-related charges were more frequent in the Detained adult sample (32%) than in the Detained youth sample (10%). To enhance ethnic diversity and minimize bias due to geographic location, the study employed four data collection sites. (A project coordinating team at a fifth site was responsible for cross-site training, consultation, monitoring data collection pro- cedures, and managing the data base.) Data were collected in Los Angeles (n = 404; 29% of total sample), Philadelphia (n = 390; 28%), northern Florida (n = 223; 16%) and northern, central, and western Virginia (n = 376; 27%) at 11 juvenile detention centers, 8 adult jails, and their surrounding communities. Each site contributed an approximately equal number of Detained and Community participants. Each site contributed participants to every age/gender/ethnicity cell, although site contribu- tions to ethnic groups were disproportionate, reflecting the ethnic composition of each site. It was not possible to compare the demographic characteristics of the Detained participants to the characteristics of the total detained population of each detain- ment facility, because total admissions data were unavailable. The age proportions in our Detained youth samples, however, appeared to be representative of juvenile detention centers generally, and the proportions of different ethnic groups in the Detained youth sample were nearly identical to those reported in a national survey of juvenile detention centers (Snyder & Sickmund, 1995). Independent Variables Demographic and Justice System Experience Variables We obtained data regarding age, gender, ethnicity, offense charged (for De- tained participants), and socioeconomic status by self-report. SES was determined using education and occupation according to the Hollingshead system (1975), which This content downloaded on Tue, 8 Jan 2013 23:21:19 PM All use subject to JSTOR Terms and Conditions A Comparisonof Adolescents'and Adults'Capacitiesas TrialDefendants 339 providesfive classificationsrangingfromI (highestclass) to V (lowestclass).In anal- yses in whichthe interactionbetween SES and a second variable(e.g., age) was of interest,SES was treatedas either a five-levelor a three-level(Hollingsheadclasses I-II, III, andIV-V) categoricalvariable.In analysesin whichSES was covaried,SES wastreatedas a continuousvariable,withscoresrangingfrom1 to 5. Forthe Detained samples,experiencein the justice systemwas coded from answersto two questions: whether the participanthad (a) ever before been "foundguilty"of a delinquency or crime and (b) ever before been "lockedup" in a detentioncenter or jail (coding "no"to both questionsas "0,""yes"to only one of the questionsas "1,"and "yes" to both questionsas "2"). Intelligence The WechslerAbbreviatedScale of Intelligence(WASI; PsychologicalCorpora- tion, 1999) producesan estimate of general intellectualability on the basis of two subtests (Vocabularyand MatrixReasoning) that can be administeredin approxi- mately 15 min. The WASI is linked to the WechslerIntelligenceScale for Children (WISC-III)and the WechslerAdult Intelligence Scale (WAIS-III),and has been normedfor individualsaged 6-89 years. MentalHealthProblems The MassachusettsYouth Screening Instrument-SecondVersion (MAYSI-2; Grisso& Barnum,2000;Grisso,Barnum,Fletcher,Cauffman,& Peuschold,2001) is a 52-item,six-scaleyes/noself-reportmentalhealthscreeninginventorythatprovides indexes of degree of disturbanceon six clinicalscales (Alcohol/DrugUse, Angry- Irritable,Depressed-Anxious,SomaticComplaints,SuicideIdeation,ThoughtDis- turbance).Twoitems were modifiedfor use with young adultsin this study;in both, the word "school"was replacedwith the word "work."Alpha coefficientsfor De- tained and Communityyouth and young adultsampleswere comparableto accept- able coefficientspublishedin earlierMAYSI-2reports(Grissoet al., 2001;Grisso& Barnum,2000), with the exception of one scale (ThoughtDisturbance),which did not achieve acceptablealphacoefficientsfor any of the subgroups. Dependent Variables FunctionalAbilitiesRelatedto Competenceto StandTrial The primarydependentvariablewas the MacCAT-CA,designedto assesscrim- inal defendants'abilitiesto participatein theirdefense ("competenceto standtrial"; Otto et al., 1998;Poythresset al., 1999).The instrument'sscoringcriteria,as well as normsbased on large,nationalsamplesof pretrialadultdefendants,are providedin the MacCAT-CAmanual.At the time of the presentstudy,therewereno publications reportinguse of the MacCAT-CAwithadolescents.The standardadministrationand content of the MacCAT-CAwere unalteredfor this study. The 22 MacCAT-CAitems are grouped into three subscales:Understanding, Reasoning,and Appreciation.Understandingassessescomprehensionof courtroom This content downloaded on Tue, 8 Jan 2013 23:21:19 PM All use subject to JSTOR Terms and Conditions 340 Grisso,Steinberg,Woolard,Csalffman,Scott,Graham,Lexcen,Reppucci,and Schwartz proceduresand personnel and the defendant'srights at trial. Reasoning assesses the recognitionof informationrelevantto a legal defense and the abilityto process informationfor legal decision making. The Appreciationsubscale, referringto a person's abilityto recognize the relevance of informationfor one's own situation, assesses whethera defendant'slegal decision makingis influencedby symptomsof mental illness,such as delusionalthinking. Thisstudyemployedmean subscalescores on the three MacCAT-CAscales,as well as a system of classifyingMacCAT-CAsubscalescores into three hierarchical categoriesusingcutoffscoresprovidedin the MacCAT-CAmanualindicating"min- imal or no impairment,""mildimpairment,"or "clinicallysignificantimpairment." The cutoffscorefor "clinicallysignificantimpairment"is set at the score equaling1.5 standarddeviationsbelow the mean of the "presumedcompetent"samples in the original MacCAT-CAnormingstudy (Poythresset al., 1999). Performanceabove 1.0 standarddeviationbelow the mean for those samplesis consideredto represent "minimalor no impairment."Scoresbetween those two cutoffswere labeled "mild impairment." Inter-raterreliabilityfor the MacCAT-CAwas assessed twice using all scorers (25 researchassistants)at allfourof the studysites.ForUnderstanding andReasoning, intra-classcorrelationsfor youth data were marginallyadequate early in the data collection process (.63 and.60, respectively),and better at a later point in the study (.91 and.80;for young adults,.88 and.70). Intra-classcorrelationsfor Appreciation were much more variable,rangingfrom.86 for youths early in the study to.17 for young adults later in the study.The latter result apparentlywas due to highly truncatedscores on Appreciationitems, with almost no young adult participants producing0-creditresponses. Decisions and Judgment in the Adjudicative Process The MacJEN, developed for this study and based on an earlier instrument (Woolard,1998),was designedto providedata regardingage-relateddifferencesin choicesandthe psychosocialfactorsthatmightinfluencethose choices.The MacJEN uses vignettesto pose threelegal decisionscommonin the delinquency/criminal pro- cess: (a) respondingto police interrogation,(b) disclosinginformationduringcon- sultationwith a defense attorney(one half of the participantswere administereda vignette about a publicdefenderand one half a vignette about a privatelyretained attorney);and (c) respondingto a plea agreementfor reducedconsequencesin ex- changefor a guiltyplea and testimonyagainstother defendants. Respondents are given several response choices and asked to recommenda "best choice" and "worstchoice"for the vignette character.Choices for the police interrogationvignetteincludeconfessingto the offense,denyingthe offense, and re- fusingto speak.Choicesfor the attorneyconsultationvignetteincludefull disclosure, partialdisclosure,denyingthe offense,andrefusingto cooperate.Choicesfor the plea agreementvignette include acceptingor rejectingthe offer for reduced chargesin exchange for testimony against other defendants.Participants'"best choice" rec- ommendationsfor the vignette characterwere used to create a variableindicating readinessto complywith authorityfigures. This content downloaded on Tue, 8 Jan 2013 23:21:19 PM All use subject to JSTOR Terms and Conditions A Comparisonof Adolescents'and Adults'Capacitiesas TrialDefendants 341 MacJENresponseswere also scored accordingto criteriadesigned to identify three variablesrepresentingaspectsof psychosocialmaturity:risk appraisal(repre- sented by three indexes),futureorientation,and resistanceto peer influence. To assess risk appraisal,participantswere asked to identify all positive and negative consequences,or risks, of each best and worst choice recommendation, the likelihood of a given set of risks,and the unpleasantnessof those risks.These responsescontributedto three separateindexes conceptualizedas differentaspects of risk appraisal:(a) "riskrecognition,"which summed the total number of risks identifiedacrossthe best andworstchoicesin eachvignetteandthen averagedacross vignettes;(b) "risklikelihood,"whichsummedparticipants'Likert-typeresponsesto questionsthataskedaboutthe likelihoodthatpossiblenegativeconsequenceswould occur;and (c) "riskimpact,"which summedparticipants'Likert-typeresponsesto questionsregardinghow unpleasantthe negativeconsequenceswould be if they did occur. The second psychosocialmaturityvariable,futureorientation,was assessed by coding:all of the risksidentifiedby each participantas reflectingthe short- or long- range nature of their consequences (employing a standardizedclassificationsys- tem). Long-rangeriskswere averagedacrossvignettesto producean index of future orientation. The third psychosocialmaturityvariable,resistanceto peer influence,was as- sessedwithquestionsaskedafterparticipantshadmadetheiroriginalchoices,posing new informationto them about peers' recommendationsthat were contraryto the participants'choices. For each decision-makingvignette,peer resistancewas mea- sured as a dichotomousvariable(retainedoriginalchoice versusswitchedto peers' choice). Procedure Prior to data collection, all site project directorsand researchassistantsmet at one location for several days of trainingby the project coordinatingteam. The projectcoordinatorsthen supervisedresearchteamson site duringpracticeprotocol administrationsat cooperatingfacilities. Researchassistantsvisitedthe participatingjuveniledetentioncentersandadult jails once or twice a week for about 11 months.Theywere assistedby staffto identify new detainees who had arrivedsince the previousvisit, and to determinewhether any detainees had been "screenedout" by staff or participantadvocatesregarding potentialresearchparticipation.Specialprotectionsfor humansubjectsin research were requiredbecause Detained participantswere identifiedas belongingto "vul- nerable"populations.Independentparticipantadvocatesmonitoredthe solicitation of Detained youths,assuringconditionsof voluntaryyouth assent and vetoing spe- cificyouths'participationif it mightpose unnecessarystress.In addition,parentsin some sites were notifiedby mail priorto approachingDetained youths,and youths whose parentsrespondedindicatingthat they objected were not included.All hu- man participantsprocedureswere approvedby the IRB of the universityat which the coordinatingsite was located as well as the IRB of the universityassociatedwith each data collectionsite. This content downloaded on Tue, 8 Jan 2013 23:21:19 PM All use subject to JSTOR Terms and Conditions 342 Grisso,Steinberg,Woolard,Csaffman,Scott,Graham,Lexcen,Reppucci,and Schwartz Any detainees who had not been screened out were approachedby research assistantswith an explanationof the study,the procedure,and a request for assent to participate.As the studyperiodprogressed,Detained youths and adultswere ap- proachedmore selectivelyto augmentcontributionsto age, gender,and/orethnicity subsamplesthat accruedin smallerproportionsin detained settings.Females and very young adolescentsespeciallywere oversampledin proportionto their actual representationin detentionandjail facilities. Communityyouths and young adults were solicited in neighborhoodsserved by the relevantyouth detention or adultjail facilities.Communityyouths were so- licited in schools,youth programs,and Girls'and Boys' Clubs,whereascommunity adultswere solicited in communityclubs,agencies and shelters,and at community colleges, using posters,leaflets,and/ordirectcontactby researchassistants.Human participantsconsiderationsrequired parental consent for Communityyouths. As datacollectionproceeded,the Age x Gender x Ethnicproportionsthatwere accru- ing in the Detained sampleswere examinedperiodicallyto guide a more selective approachto the recruitmentof potential Communityparticipants,aiming at final Communitysamplesthatwere demographicallysimilarto theirrespectiveDetained samples. It was not possible to determineparticipationrates for either the detained or communitysamplesbecauseof the way in whichparticipantswere recruitedin each group.In the case of detained participants,detainees in each facilitywere periodi- cally informedof the studyby facilitystaffpersonsand askedif they were interested in speaking with a research staff member about the project;this was done on a regularbasis because the population of detainees changed daily.Facilitystaff did not approachdetainees who they believed were especiallyvulnerableor otherwise unqualifiedto participatein the study (e.g., due to mental illness, recent trauma, etc.). It was not possibleto monitorhow manydetaineeswere approachedby facility staff and whatproportionrefusedto speak with the researchstaff.Generallyspeak- ing, however, very few detainees with whom we were permittedto speak refused to participate.The recruitmentof the communityparticipantswas done by posting advertisementsabout the study in communitycenters,schools,and recreationcen- ters. Individualswho were interestedin the study were asked to call our research office. All participationwas voluntary.Detained youths and adults received $10 for theirparticipation(or snacksin some facilitiesthat did not allow monetaryawards); communityyouthsand adultsreceived$25.Informedandsignedconsent (assentfor youths)was requiredfor all participants.Confidentialitywas assuredwith the excep- tion of the researchers'obligationto reportto others in cases in whichinformation was obtained that suggestedimminentrisk of harm to self or others, or danger of harmfromothers.Participantsendorsingtwo out of three critical"SuicideIdeation" itemson the MAYSI-2werescreenedfurtherby researchassistantsusinga structured processto obtaindataon seriousnessand recencyof suicidalintent.Responseswere evaluatedin consultationwiththe site projectdirectoror a site clinician,as well as the projectcoordinator,to determinethe need to breachconfidentiality.(Fordetailson frequencywith whichbreachingconfidentialitywas required,see an archivalreport of the study'smethod at www.mac-adoldev-juvjustice.org.) This content downloaded on Tue, 8 Jan 2013 23:21:19 PM All use subject to JSTOR Terms and Conditions A Comparisonof Adolescents'and Adults'Capacitiesas TrialDefendants 343 Table3. Means(SD) for MacCAT-CAUnderstanding,Reasoning,andAppreciationSubscales Age groups(N) Understanding Reasoning Appreciation 11-13 (190) 10.45 (3.31)a 11.30 (2.82)a 9.68 (2.34)a 14-15 (345) 11.27 (2.97)b 12.10 (2.55)b 10.33 (1.79)b 16-17 (392) 12.00 (2.82)c 12.76 (2.34)c 10.65 (1.66)b.c 18-24 (466) 12.13 (2.92)c 12.57 (2.51)b,C 10.77 (1.57)c Note.Superscriptsreferto age groupcomparisonsfor eachsubscaleconsideredseparately.Age groups with differentsuperscriptsdifferedsignificantlyon thatsubscale,at p <.05. When a participant had consented/assented, the research assistant administered the study protocol, obtaining demographic and justice experience data first, followed by the MacCAT-CA, the WASI, the MAYSI-2, and the MacJEN. The protocol typi- cally required between 90 and 180 min to administer. The project coordination team continuously monitored the data collection process and compliance with ethical pro- cedures at the four data collection sites. Research assistants at each site scored their own protocols, and scored copies were sent to the coordinating site for quality checks, data entry, and database management. RESULTS MacCAT-CA: Age Differences in Performance0l Preliminaryanalysesindicatedthat the four age groups did not differ signifi- cantlywith respectto gender or ethnicity,but did differ significantly,albeit slightly, with respectto social class and intelligence.As a consequence,all comparisonsbe- tween age groupscontrolledfor these lattertwo variables. To examine age differencesin MacCAT-CAperformance,a multiple analysis of covariance(MANCOVA) was conductedwith the three MacCAT-CAsubscale scores (Understanding,Reasoning,and Appreciation)as the dependentvariables, age groupas the independentvariable,and the continuousmeasureof SES and IQ as covariates.The analysisindicated a significantmultivariateeffect for age, mul- tivariateF(9, 3263) = 11.32, p <.001,with significantunivariateeffects for age on all three subscales, for Understanding, F(3, 1343) = 16.24, p <.001; for Reason- ing, F(3, 1343) = 19.33, p <.001;for Appreciation, F(3, 1343) = 18.06, p <.001. As shown in Table 3, post hoc contrasts indicated that, on the Understanding subscale, the 11- to 13-year-olds performed significantly worse than the 14- to 15-year-olds, who performed significantly worse than the two older groups; 16- to 17-year-olds and young adults did not differ. On Reasoning, 11- to 13-year-olds performed signif- icantly worse than 14- to 15-year-olds, who performed significantly worse than the 16- to 17-year-olds; 14- to 15-year-olds and 16- to 17-year-olds did not differ from young adults. On the Appreciation subscale, 11- to 13-year-olds performed signifi- cantly worse than 14- to 15-year-olds, who scored lower than young adults, but did not 10Variousstages of the following analyses examined youths' and young adults' performanceon the MacCAT-CAby study site acrossthe demographicvariables.Very few site differenceswere found, no morethanwouldbe expectedby chance. This content downloaded on Tue, 8 Jan 2013 23:21:19 PM All use subject to JSTOR Terms and Conditions 344 Grisso,Steinberg,Woolard,Cauffman,Scott,Graham,Lexcen,Reppucci,and Schwartz differ from 16- to 17-year-olds; as with the other two subscales, 16- to 17-year-olds and young adults did not differ. In general, the magnitude of the difference (i.e., the effect size) between the scores of the 11- to 13-year-olds versus adults was moderate by conventional statistical standards (e.g., ds between.5 and.6) whereas the differences between the scores of the 14- to 15-year-olds and adults were small (e.g., ds between.2 and.3). These patterns of age differences were also seen when chi-square analyses were used to compare age groups (both with and without controlling for IQ and SES) with respect to the proportions of individuals showing various levels of impairment- "no impairment," "mild impairment," or "significant impairment" (Poythress et al., 1999)-in Understanding, X2(6) = 49.60, p <.001, and Reasoning, X2(6) = 37.56, p <.001; see Fig. 1).11 For example, whereas 20% of 11- to 13-year-olds, and 13% of 14- to 15-year-olds, showed significantly impaired Understanding, only 7% of the 16- to 17-year-olds and this same proportion of adults scored in this range. Similarly, proportions of individuals showing significantly impaired Reasoning declined from 16% among 11- to 13-year-olds to 9% among 14- to 15-year-olds, to less than 7% among 16- to 17-year-olds and young adults. It is important to examine the proportions of each age group who show signif- icantly impaired Understanding or Reasoning (or both), because significant impair- ment in either could raise doubts about competence. These results are presented in Fig. 2, illustrating that 30% of 11- to 13-year-olds, and 19% of 14- to 15-year-olds, were significantly impaired on one or both of these subscales; the figures for 16- to 17-year-olds and for young adults were both 12%. To examine whether these patterns of age differences varied as a function of gender, ethnicity, or Detained versus Community status, a 4 (age group) x 2 (gen- der) x 3 (ethnicity) x 2 (Detained/Community status) multiple analysis of covariance (MANCOVA) was conducted with the three MacCAT-CA subscale scores (Under- standing, Reasoning, and Appreciation) as the dependent variables and SES and IQ as covariates. MacCAT-CA performance was unrelated to Detained/Community status, gender, and ethnicity. More importantly, the only interaction to reach signif- icance was the two-way interaction between age and Detained/Community status, multivariate F(9, 3157) = 2.465, p <.01, and the univariate effect was significant only for scores on the Understanding subscale, F(3, 1299) = 6.09, p <.001. Follow- up analyses indicated that whereas Detained and Community young adults differed significantly in Understanding scores, Detained and Community juveniles (of any age) did not. In a separate analysis that treated socioeconomic status as a three-level inde- pendent variable rather than as a covariate, we did not find significant main effects for SES or a significant interaction between SES and age. The absence of significant interactions between age and gender, age and ethnicity, and age and SES indicates 11Wedo not reportproportionsof individualsshowingimpairedappreciation,becauseit is not clearwhat such impairmentmeansin a sampleof individualswithoutseriousmentalillness.This subscaleof the MacCAT-CAwasdevelopedto identifyindividualswhosebeliefsabouttheirtrialwerehighlyirrational or distortedas a resultof seriousmentaldisorder.In this study,the majorityof individualswho showed "impaired"appreciationdid so becausethey couldnot articulatereasonsfor theirresponsesto certain interviewquestions,not becausethey demonstrateddistortedor irrationalthinking. This content downloaded on Tue, 8 Jan 2013 23:21:19 PM All use subject to JSTOR Terms and Conditions  Understanding 100 80 60 (1) 0. 40 20 0 11 to 13 14 to 15 16 to 17 18+ Age Groups Reasoning 100 80 t4-- 60- c) I- QL 40- 2C 0111 to 13 14 to 15 16 to 17 18+ Age Groups Appreciation Cl (,3 Q) D_ 11 to 13 14 to 15 16 to 17 18+ Age Groups Fig. 1. Degree of impairment on MacCAT-CA Un- derstanding, Reasoning, and Appreciation subscales as a function of age. This content downloaded on Tue, 8 Jan 2013 23:21:19 PM All use subject to JSTOR Terms and Conditions 346 Grisso, Steinberg, Woolard, Cauffmunn,Scott, Graham, Lexcen, Reppucci, and Schwartz co 25 10 2 5 11 to 14 to 15 16to 7 18+ Age Groups Fig. 2. Proportionof individualsat different ages who are significantlyimpairedwith respect to either or both MacCAT-CAUnderstandingand Reasoning. that age differences in MacCAT-CA performance were consistent between males and females, across individuals from different ethnic groups, and across individuals from different socioeconomic backgrounds when controlled for IQ. Patterns of age differences in the proportions of individuals showing no impair- ment, mild impairment, or significant impairment on Understanding are shown sepa- rately for the Detained and Community samples in Fig. 3. As the figure illustrates, the proportions of individuals showing mild or significant impairment in Understanding differed between Detained and Community adults but were relatively similar among Detained and Community juveniles. When chi-square analyses were used to compare the Detained and Community groups with respect to the proportions of individuals in each age group who demon- strate significantly impaired Understanding or Reasoning (or both), similar patterns of age differences in the two groups emerged (both with and without controlling for IQ and SES). As Fig. 4 illustrates, although impairment was more common among Detained than Community individuals, in both groups, 11- to 13-year-olds, and to a lesser extent 14- to 15-year-olds, were more likely than adults to show deficits in the capacities related to CTS, for Community participants, X2(3) = 32.37, p <.001; for Detained participants, X2(3) = 17.10, p <.001. In both groups, the performance of 16- to 17-year-olds was comparable to that of young adults. MacCAT-CA: Intelligence, Prior Justice System Experience, Mental Health Problems MacCAT-CA performance may be influenced by factors other than age, includ- ing intelligence (which one would be expected to be positively related to MacCAT performance), mental health problems (expected to be negatively related to MacCAT-CA performance), and prior experience in the justice system (expected to be positively related to MacCAT-CA performance). To examine these relations, we conducted a series of multiple multiple regression regressing scores for each of analyses, regressing regression analyses, This content downloaded on Tue, 8 Jan 2013 23:21:19 PM All use subject to JSTOR Terms and Conditions A Comparison of Adolescents' and Adults' Capacities as Trial Defendants 347 Detained Individuals 100- *-f c a) 0() Q_ 11to 13 14to 15 16 to 17 18+ Age Groups CommunityIndividuals (D Q) cL 1 to 3 14to15 16 to 17 18+ Age Groups Fig.3. Impairment on MacCAT-CA Understand- ing subscale as a function of age and justice system status. the MacCAT-CAsubscaleson age (as a continuousvariable),IQ, each of the six subscalesof the MAYSI-2,and our measure of prior experiencein the justice sys- tem (for justice system participantsonly). As expected, IQ was significantlycorre- latedwithUnderstanding,/ =.415,t(1358) = 15.96,p <.001;Reasoning,P =.421, t(1358) = 16.17, p <.001; and Appreciation, P =.328, t(1358) = 12.10, p <.001. Contraryto expectancy,MacCAT-CAperformancewas unrelatedto our measure of priorexperiencein the justice system and largelyunrelatedto MAYSI-2mental healthproblems. Figure 5 shows the relation between IQ, treated as a three-level categorical variable(IQ = 60-74,75-89, or 90 andabove) andimpairmenton the threeMacCAT- CA subscales.As expected, chi-squareanalyses showed that individualsof lower intelligencewere more likely to have impairedUnderstanding,X2(4)= 211.27,p < This content downloaded on Tue, 8 Jan 2013 23:21:19 PM All use subject to JSTOR Terms and Conditions 348 Grisso, Steinberg, Woolard, Caiiffmnn, Scott, Graham, Lexcen, Reppucci, and Schwartz ' 7 s..... ;.....'. g. 40 f E... 30 C 20 ~;:~....... Une Z tnd. ad Reasoing.a.i c. r i.....:...........c,.(/)finivda Cj i 1o -!l:i 11 to 13 14 to 15 16 to 17 18+ Age Groups Fig. 4. Proportionof individualsat differentages who are significantlyimpairedwithrespectto eitheror both MacCAT- CA Understandingand Reasoningas a functionof age and justicesystemstatus..001, Reasoning, X2(4) = 130.50, p <.001, and Appreciation, x2(4) = 119.55, p <.001. Figure 6 displays the proportions of individuals in each IQ group who were significantly impaired on either the Understanding or the Reasoning scale (or both). Again, individuals of lower intelligence were far more likely to lack these capacities related to participation in trials as defendants. To examine whether the relations between MacCAT-CA performance and intelligence and justice system experience varied with age, we conducted a second series of regressions, in which the main effects of these variables, along with age (as a continuous variable), were entered into the equation followed by the terms repre- senting the interactions between each variable and age. Age remained a significant predictor of MacCAT-CA performance on all three scales, even when intelligence and prior justice system experience were controlled. There were marginally significant interactions between age and IQ for both Understanding, f =.324, t(1361) = 1.696, p =.09, and Appreciation, = -.382, t(1361) = -1.909, p =.06, but no age by ex- perience interaction. To examine the nature of these interactions, follow-up analyses were conducted in which correlations between MacCAT-CA performance and IQ were computed separately for each age group. The interaction between age and IQ in the prediction of MacCAT-CA performance is difficult to interpret; the correlation between IQ and Understanding generally increased with age, but the correlation between IQ and Appreciation generally decreased with age. MacCAT-CA Performance as a Function of Intelligence, Age, and Detained/Community Status It is widely established that, on average, individuals who are detained in the justice system score lower on intelligence tests than demographically comparable This content downloaded on Tue, 8 Jan 2013 23:21:19 PM All use subject to JSTOR Terms and Conditions  Understanding c- Q) C.) a) Q. IQ60-74 IQ75-89 IQ90 + IQ Groups Reasoning 100 80 60 a) 1) 40 20 0 IQ60-74 IQ75-89 IQ90 + IQ Groups Appreciation a 0) Q- 0m IQ60-74 IQ75-89 IQ90 + IQ Groups Fig. 5. Impairmenton MacCAT-CAUnderstanding, Reasoning,andAppreciationsubscalesas a functionof IQ. This content downloaded on Tue, 8 Jan 2013 23:21:19 PM All use subject to JSTOR Terms and Conditions 350 Grisso, Steinberg, Woolard, Cauffman, Scott, Graham, Lexcen, Reppucci, and Schwartz 340 E 20iH:1!i Fig. 6. Proportion of individuals who are significantly im- IQ 60-74 IQ 75-89 1Q90 + (M86.28, 12.95)SD was substanti lower allythanof individuals that fromthe pairedwith respectto either or both MacCAT-CAUnder- standingand Reasoningas a functionof IQ. samples of individuals drawn from the community. As expected, an analysis of variance showed that the average IQ score of detained individuals in this study (M= 86.28, SD = 12.95) was substantially lower than that of individuals from the community, M = 97.46, SD = 15.63; F(3,1385) = 5.64, p <.001, a finding that did not vary as a function of age, F(3, 1385) =.48, ns). In light of this, and in view of the strong relation between IQ and MacCAT-CA performance, we examined the associations between IQ and MacCAT-CA performance separately within each age group, focusing on proportions with scores in the seriously impaired range. Figure 7 shows the proportions of individuals of different levels of intelligence, withineach age group,who were seriouslyimpairedwithrespectto Understandingor Reasoning or both. Younger individuals of lower intelligence were especially likely to be deficientin the necessarycapacitiesassociatedwith trialcompetence.Indeed, among11-to 13-year-olds,morethanone halfwithanIQ between60 and74, andmore thanone thirdwithan IQ between75 and89, were significantlyimpaired.Among 14- to 15-year-olds,approximately40%of those withan IQ between60 and74, andmore than one in four with an IQ between 75 and 89, were comparablyimpaired.These figuresare importantbecause,as Fig.8 illustrates,between one fifthand one quarter of juvenilesaged 15 and youngerin the Detained samplehad IQ scoresbetween 60 and 74, and approximately 40% of Detained juveniles aged 15 and younger had IQ scores between 75 and 89. In other words, approximately two thirds of the Detained juveniles aged 15 and younger had an IQ that was associated with a significant risk of being incompetent to stand trial because of impaired Understanding or Reasoning or both. MacJEN: Age Differences in Choices on Decision-Making Vignettes Tirning now to MacJEN results related to questions about participants' legal decision making, we examined age differences in the participants' recommended best choice, using separate chi-square analyses for each of the three decision-making This content downloaded on Tue, 8 Jan 2013 23:21:19 PM All use subject to JSTOR Terms and Conditions A Comparison of Adolescents' and Adults' Capacities as Trial Defendants 351 80 ag MO 20-: 1 to 13 14 to 15 16 to 17 18+ Age Groups Fig. 7. Proportionof individualsat differentages who are significantlyimpairedwith respectto either or both MacCAT-CAUnderstandingandReasoningas a function of age and IQ. vignettes (police interrogation, attorney consultation, plea agreement; see Fig. 9). Analyses indicated significant age differences for choices regarding police interroga- tion, X2(6) = 158.73, p <.001. The proportion of participants who chose confession as the best choice decreased with age, from about one half of the 11- to 13-year- olds to only one fifth of young adults. No age differences were found for the second vignette regarding consultation with a public defender, X2(9) = 11.59, ns, or a pri- vate attorney, X2(9) = 11.32, ns); over 75% of each age group recommended full disclosure. Significant age differences were found for the plea agreement vignette, X2(3) = 45.58, p <.001, the proportion accepting the plea agreement decreasing from 74% among 11- to 13-year-olds to 50% of young adults. Because Anayss the fourdffre n e differed age groups o hiesignificantly eadnoieitroa with respect to social class tio, 2() 158.73 and intelligence, - rprioofpriciat p examined 01h hoecneso separate chi-square analyses age differences within three 1 o1-er as cho~~~~~~~icederase it ag,fo bu n afo h olds g yugaut.Nf0iftif f rneswr fudfo h scn vignette regarding consultation ,~, 6,:;i::.'... ;'*':'i~i':iiii??ii ?;'~4:'-!;i......i,~?-~i? with apublic --ns, pri-~~~~~'lli X2(9)aor defender,11.59, vateI torey,X()-1.2 a s;oe5 fahaegoprcmeddfl disclosure. (13to1 Significant agedifferences were found for the vignette,~~iB$ plea agreement X23)=4558 60rpotinacepig