Cool and Hot Executive Function Impairments in Violent Offenders with Antisocial Personality ...

12 Pages·2013·0.4 MB·English

Cool and Hot Executive Function Impairments in Violent
Offenders with Antisocial Personality Disorder with and
without Psychopathy
Stephane A. De Brito1*, Essi Viding2, Veena Kumari3, Nigel Blackwood4, Sheilagh Hodgins4,5
1SchoolofPsychology,UniversityofBirmingham,Birmingham,UnitedKingdom,2DivisionofPsychologyandLanguageSciences,UniversityCollegeLondon,London,
United Kingdom, 3Department of Psychology, Institute of Psychiatry, King’s College London, London, United Kingdom, 4Department of Forensic and
Neurodevelopmental Science, Institute of Psychiatry, King’s College London, London, United Kingdom, 5De´partement de Psychiatrie, Universite´ de Montre´al,
Montre´al,Que´bec,Canada
Abstract
Background:Impairments in executive function characterize offenders with antisocial personality disorder (ASPD) and
offenderswithpsychopathy.However,theextenttowhichthoseimpairmentsareassociatedwithASPD,psychopathy,or
both isunknown.
Methods:The present study examined 17 violent offenders with ASPD and psychopathy (ASPD+P), 28 violent offenders
withASPDwithoutpsychopathy(ASPD2P),and21healthynon-offendersontasksassessingcool(verbalworkingmemory
andalterationofmotorresponsestospatiallocations)andhot(reversallearning,decision-makingunderrisk,andstimulus-
reinforcement-based decision-making) executivefunction.
Results:Incomparison to healthy non-offenders, violent offenders with ASPD+P and those with ASPD2P showed similar
impairments in verbal working memory and adaptive decision-making. They failed to learn from punishment cues, to
changetheirbehaviourinthefaceofchangingcontingencies,andmadepoorerqualitydecisionsdespitelongerperiodsof
deliberation.Intriguingly,thetwogroupsofoffendersdidnotdiffersignificantlyfromthenon-offendersintermsoftheir
alteration of motor responses to spatial locations and their levels of risk-taking, indicated by betting, and impulsivity,
measuredasdelayaversion.Theperformanceofthetwogroupsofoffendersonthemeasuresofcoolandhotexecutive
function did notdiffer, indicatingshareddeficits.
Conclusions:These documented impairments may help to explain the persistence of antisocial behaviours despite the
known risksofthe negativeconsequences ofsuch behaviours.
Citation:DeBritoSA,VidingE,KumariV,BlackwoodN,HodginsS(2013)CoolandHotExecutiveFunctionImpairmentsinViolentOffenderswithAntisocial
PersonalityDisorderwithandwithoutPsychopathy.PLoSONE8(6):e65566.doi:10.1371/journal.pone.0065566
Editor:CarlesSoriano-Mas,BellvitgeBiomedicalResearchInstitute-IDIBELL,Spain
ReceivedMarch17,2013;AcceptedApril26,2013;PublishedJune20,2013
Copyright:(cid:2)2013DeBritoetal.Thisisanopen-accessarticledistributedunderthetermsoftheCreativeCommonsAttributionLicense,whichpermits
unrestricteduse,distribution,andreproductioninanymedium,providedtheoriginalauthorandsourcearecredited.
Funding:ThisresearchwasfundedbyresearchgrantsfromtheDepartmentofHealth(theNationalForensicMentalHealthR&Dprogramme),theMinistryof
Justice(aDSPDprogrammegrant),theNIHRBiomedicalResearchCentre,SouthLondon,MaudsleyNHSFoundationTrust,andInstituteofPsychiatry(King’s
CollegeLondon).SDBwassupportedbyanMRCPh.D.studentshipandaresearchfellowshipfromtheSwissNationalScienceFoundation(FNSPA00P1_139586),
EVissupportedbyESRC(RES-062-23-2202).Thefundershadnoroleinstudydesign,datacollectionandanalysis,decisiontopublish,orpreparationofthe
manuscript.
CompetingInterests:Theauthorshavedeclaredthatnocompetinginterestsexist.
*E-mail:[email protected]
Introduction ThereisaccumulatingevidencethatmenwithASPDrepresent
a heterogeneous population with respect to personality traits,
Most violent crimes are committed by a small group of males
aggressive behaviour, offending patterns, [4], and engagement
who display persistent antisocial and aggressive behaviour from
with, and response to, cognitive-behavioural offender rehabilita-
childhood onwards [1,2]. This life-long pattern of behaviour is
tionprograms[12,13].Whileallwithinthispopulationpresentan
indexedbyDSM-IVdiagnosesofConductDisorder(CD)priorto
earlyonsetofantisocialbehaviourthatremainsstableoverthelife-
age 15 and Antisocial Personality Disorder (ASPD) in adulthood span,asub-groupadditionallypresentpsychopathy(ASPD+P),as
[3]. Life-long patterns of risk taking and impulsivity are central
defined by the Psychopathy Checklist-Revised (PCL-R; [14,15]).
features of ASPD [4]. Illegal behaviours persist despite repeated Psychopathy is a syndrome characterized by a constellation of
criminal sanctions. Neuropsychological deficits in executive affective,interpersonal,andbehaviouralfeatures[14],includinga
function (EF) reflecting the higher order cognitive control of lack of empathy, callousness, shallow affect and a failure to take
thought, action, and emotion [5] have been hypothesized to be responsibility for one’s actions, and a pathological interpersonal
central to the onset and persistence of severe antisocial and style involving grandiosity, glibness, superficial charm, and the
aggressive behaviour[6–11]. manipulationofothers[16].Muchresearchhasdemonstratedthat
PLOSONE | www.plosone.org 1 June2013 | Volume 8 | Issue 6 | e65566
NeuropsychologyofAntisocialPersonalityDisorder
in comparison to offenders without psychopathy, those with comes from studies showing that they make more commission
ASPD+P begin offending at a younger age [17], more often errors (i.e., responses to stimuli paired with negative reinforce-
engage in instrumental aggression [18], and acquire more ment) than non-psychopathic offenders on passive avoidance
convictions orcharges forviolent offences [19,20]. learning tasks assessing stimulus-reinforcement-based decision-
Consistent with the differences in personality traits and making.
aggressive behaviour that distinguish adult men with ASPD+P Only two studies have been published that examined violent
and those with ASPD and not psychopathy (ASPD2P), recent offenderswithclearlydelineatedASPD2P [39,40].Themajority
evidence suggests that the two groups show distinct emotional of investigations of cool EF suggest that men with ASPD2P
impairments [20], affective processing [21], brain response to performlikehealthymenonattentionalset-shiftingtasks,planning
emotional stimuli when engaged in goal-directed behaviour [22], andmeasuresofverbalorspatialworkingmemory,alldependent
and structural brain anomalies [23]. Those with ASPD2P are on DLPFC functioning ([39,41,45,56]; but see [57,58]). By
hypothesized to present a low threshold for engaging in reactive contrast, there is consistent evidence that men with ASPD2P
aggressive behaviour towards others due to a hyper-sensitivity to arecharacterizedbyhotEFdeficitsasindicatedbytheirimpaired
threat, both real and perceived, as evidenced by hyperactivity in performance on tasks such response reversal and affective
the amygdala [6]. By contrast, individuals with ASPD+P show decision-making indexing the functional integrity of the VMPFC
hypo-activity in the amygdala in response to threat (e.g., [24]). [43,44,46,57].
Importantly, these distinct phenotypes emerge early in childhood To date, only one study [40] has compared violent offenders
[25]. The lack of responsiveness in the amgydala, especially to with ASPD+P and violent offenders with ASPD2P to healthy
fearfulfaces,amongadultswithASPD+P(e.g.,[26])andchildren participants on tasks assessing cool EF (the Stockings of Cam-
showing the antecedents of this condition (e.g., [27,28]) may bridge planning task, attentional set-shifting on the intra-dimen-
underlie their impairment in stimulus-reinforcement learning sional/extradimensional [ID/ED] set-shifting task, behavioural
central to passive avoidance paradigms [29]. This impairment inhibition on a Go/No-Go task) and hot EF (response reversal
has been hypothesized to be a core deficit of ASPD+P that components of the ID/ED task). Based on their scores on the
emerges in childhood and limits learning not to engage in Psychopathy Checklist: Screening Version (PCL: SV; [59]) the
instrumental antisocial and aggressive behaviour and learning to violentoffenderswithASPDweredividedintothreegroups(‘low’
engageinprosocialbehaviour[29].Evidencefromdevelopmental = PCL:SV#15,‘medium’ = PCL:SV =16–19;and‘high’ =
studies examining children and adolescents is indeed consistent PCL: SV .19). Results indicated that, regardless of psychopathy
withthenotionthatdifferentEFimpairmentsareassociatedwith scores, offenders with ASPD, as compared to the healthy
distinct forms of antisocial behaviour and patterns of aggressive participants, exhibited subtle impairments in cool EF (planning,
behaviour [30,31]. attentional set-shifting, response inhibition), but no hot EF
Cool EF refers to top-down processes subsumed primarily by impairment (i.e., reversal learning). In correlational analyses
thedorsolateralprefrontalcortex(DLPFC)andventrolateralPFC psychopathy scores were not related to performance on any of
that are distinctly cognitive in nature and usually elicited by thetasks.Takentogether,theresultsofthisinvestigationsuggested
abstract, decontextualized problems. Working memory, response that violent offenders with ASPD+P and those with ASPD2P
inhibition, planning, sustained attention, and attentional set- exhibit similar cool EF impairments as measured by attentional
shiftingareconsideredtobecoolEF[32,33].Bycontrast,hotEF set-shifting, and similar performance on one index of hot EF
referstocognitiveprocessesthathaveanaffective,motivational,or (reversal learning). However, in view of the results Dolan [40]
incentive/reward component; these processes are generally concluded that ‘‘further studies using a range of DLPFC and
consideredtobesubsumedbyventromedialpathwaysconnecting VMPFC tasks’’ (p.8)were needed.
mesolimbicrewardcircuitry,includingtheamygdalaandstriatum, KnowledgeofcoolandhotEFthatareimpairedorpreservedin
to the ventromedial prefrontal cortex (VMPFC) [34]. Appraising each type of violent offender could be used to improve the
the motivational significance of a stimulus in affective decision- effectiveness of rehabilitation programs aimed at reducing
making paradigms and reappraising it in response reversal recidivism[60].Whilecognitive-behaviouralprogramshavebeen
paradigms areconsidered hot EF[32,35]. shown to reduce criminal recidivism [61,62], offenders with
TheextenttowhichcoolandhotEFdifferinthetwotypesof ASPD+P fail to benefit [12,13]. Further, such knowledge will
violentoffendersisdifficulttodeterminefromtheextantliterature contribute to unravelling the etiology of persistent violent
asfewstudieshavedirectlycomparedEFofviolentoffenderswith behaviour that isaprerequisite forpreventing it.
ASPD2PandASPD+P.Mostpreviousstudiescomparedoffend- The present study employed a broad range of neuropsycho-
ers with ASPD+P to offenders without psychopathy([36,37], but logical tests to assess both cool and hot EF among violent
see[38]),whileotherscomparedindividualswithASPDfromthe offenders with ASPD+P, violent offenders with ASPD2P, and
community who had not been assessed for psychopathy (but see healthy non-offenders. Tests of EF were selected because they
[39,40]fortwostudiesonviolentoffenders)toeitherhealthynon- have been validated in studies of subjects with lesions in specific
offenders [41–44] or patients without ASPD with substance use brain regions, all but one (Cambridge Gamble Task [CGT])
disorders [45,46]. have been used in previous studies of ASPD+P or ASPD2P,
AlargenumberofstudiessuggestthatoffenderswithASPD+P, they index processes that have been shown to play an important
in comparison to offenders without psychopathy, do not present role in the display of aggressive behaviour (e.g., impairment in
impairments in cool EF such as attentional set-shifting, planning, working memory), or to be related to core features of either
and verbal working memory indexing the functional integrity of ASPD+P or ASPD2P (e.g., insensitivity to punishment, impul-
theDLPFC[47–53].Thesestudies,however,showthatoffenders sivity and risk-taking) (Detailed justifications available online in
with ASPD+P present impairments in cool EF such as response Text S1).
inhibition(e.g.,[50])andinhotEFtaskssuchasresponsereversal, BothviolentoffenderswithASPD+PandASPD2Pexhibitlife-
behaviouralextinction,andaffectivedecision-makingindexingthe longantisocialbehaviour,buttheyarecharacterizedbydifferences
functional integrity of the VMPFC (e.g., [51,52,54,55]). Further inpersonalitytraits,aggressivebehaviour,emotionprocessing,and
evidence of hot EF impairments among offenders with ASPD+P inresponsetointerventionsaimedatreducingantisocial/criminal
PLOSONE | www.plosone.org 2 June2013 | Volume 8 | Issue 6 | e65566
NeuropsychologyofAntisocialPersonalityDisorder
behaviour.Consequently,wereasonedthattheywouldshowboth Spatial Alternation Task [47]. This task assesses the
similarities and differences in neurocognitive performance. First, alteration of motor responses to spatial locations on the basis of
wehypothesizedthatboththeviolentoffenderswithASPD+Pand reinforcementinformation.Tworedcarsappearedoneitherside
those with ASPD2P would show similarly poor performance on ofthecomputerscreenoneachtrial.Theparticipanthadtolearn
twotestsassessingcoolEF(theDigitSpan–BackwardandSpatial that the side on which the £20 note was located was being
Alternation Task), and on several tests of hot EF, (CGT, more alternated after each correct response. The dependent variable
reversalerrorsontheProbabilisticResponseReversalTask,more was the number of errors committed before achieving 12
commission on the Passive Avoidance Learning Task) as consecutive correctresponses.
compared to healthy non-offenders. Second, we hypothesized Probabilistic Response Reversal Task [54]. In the
thatthetwogroupsofviolentoffenderswouldshowoneimportant acquisition phase, the task assesses the ability to learn stimulus-
differenceinperformanceonthesetestsofhotEF.Consistentwith response associations and, in the reversal phase, the task assesses
muchpreviousevidenceandtheorizingaboutpsychopathy[6,37], the ability to alter stimulus-response associations as a function of
we hypothesized that the violent offenders with ASPD+P would contingency change. The reinforcement contingencies were
makemorecommissionerrorsonthePassiveAvoidanceTaskthan probabilistic: the ‘correct’ pair was not always rewarded and the
theviolent offenders withASPD2P. ‘incorrect’pairwasnotalwayspunished.Thereweretwotestpairs
that changed contingency (reversing pairs) and four ‘dummy’ or
Materials and Methods non-reversing pairs. The two reversing pairs had the following
probabilisticcontingencies:100–0;80–20.Thedependentvariable
Participants
wasthenumberoferrorscommittedbeforereachingthelearning
Violent male offenders with ASPD and healthy male non- criterionofeightconsecutivecorrectresponses.Iftheparticipants
offenderswithEnglish as a first languagewere recruitedfrom the did not meet the learning criterion, total errors made were
communityforastudyoftheneurobiologicalcorrelatesofpersistent analysed.
aggression.Diagnosticinterviewsindicatedthatnonehadalife-time CGT[68]. Oneachtrial,theparticipantwasgiven100points
historyofseverementalillnessorasubstanceusedisorderinthepast andpresentedwitharowof10boxesacrossthetopofthescreen,
month,andshowedthatallobtainedascoreof70orhigheronthe someofwhichwereredandsomeofwhichwereblue.Theratioof
WechslerAdultIntelligenceScale(WAIS-III;[63]). red:blue boxes varied from 1:9 to9:1 ina pseudo-random order.
Violent offenders with ASPD. Violent offenders were The participant was instructed that the computer had hidden a
recruited from theNational Probation Service. Probation officers token in one of the boxes, and that they must guess whether the
identified potential participants with convictions for violent tokenhadbeenhiddeninoftheredoroneoftheblueboxes.On
offences (murder, rape, attempted murder, and grievous bodily each trial, the participant first selected the colour (decision stage)
harm) confirmed by official criminal records. Offenders with a and then betted a proportion of his total points on his colour
diagnosis of ASPD who obtained a total PCL-R score $25 were decision(gamblingstage).Eachbetwaspresentedfor2.5seconds
assigned to the ASPD+P group (n=17), and those with a score andofferedindescending(95%,75%,50%,25%,5%ofcurrent
$25were assignedtotheASPD2Pgroup (n=28). points) or ascending (5%, 25%, 50%, 75%, 95% of the current
Healthynon-offenders. Non-offenders(n=21)wererecruit- points)sequences.Afterthebetwasplaced,thehiddentokenwas
edbymeansofadvertisementsinlocalnewspapersandnoticesin revealed and the bet was added to or subtracted from the total
the community. Those retained for the study had no criminal
score. The five principal dependent measures were: (1) Deliber-
record,nomentaldisorderotherthanpastsubstancemisuse,anda
ation Time defined as the mean latency in milliseconds from
PCL-R score of24or less.
presentationofthecolouredboxestotheparticipant’sresponse;(2)
QualityofDecision-Makingdefinedastheproportionoftrialson
Classification measures whichtheparticipantchosetogambleonthemorelikelyoutcome,
StructuredclinicalinterviewforDSM-IV. Allparticipants i.e. the colour of the greatest number of boxes; (3) Risk Taking
completedtheStructuralClinicalInterviewforDSM-IV,IandII, definedasthepercentageofthecurrentpointsthattheparticipant
(SCID; [64]) administered by trained forensic psychiatrists to bet. To maintain the independence of betting behaviour and
provide life-time andcurrent DSM-IV diagnoses. choice behaviour, analyses were limited to the trials where the
Psychopathychecklist–revised. ThePCL-R[16]consists participantsselectedthecolourofthemajorityofboxes,i.e.trials
of20itemsthatarescoredbyatrainedrateronthebasisofafile on which they had more chance of winning than losing; (4) Risk
review and a semi-structured interview. Each of the 20 items is Adjustmentwasdefinedasthedegreetowhichaparticipantvaries
scored on a three-point scale (0–2), with the total score ranging their risk-taking in response to the ratios of red to blue boxes
from0to40.Consistentwithavalidationstudy[65],ascoreof25 within each trial; and (5) Delay Aversion was defined as the
or higher identified the syndrome of psychopathy among these difference between risk-taking scores in the descending and the
European offenders. Forensic psychiatrists and psychologists ascendingconditions.Highbetsinbothascendinganddescending
trained to use the PCL-R administered interviews and extracted conditions reflect genuine risk-taking behaviour, whereas betting
information from files in order to rate the scale. Interviews were early in both the ascending and descending conditions reflects
videotaped and a random 25% sample was rerated by a second impulsivity(theparticipantdoesnotwaitforthebettoincreasein
trained psychologist. Intra-class correlation coefficient values for the ascending condition or to decrease in the descending
PCL-R total scores were acceptable (0.81). Scores for the four condition).
facets andtotal scoreswere calculated [66]. Passive Avoidance Learning Task [36]. The goal was to
learn to respond to stimuli that lead to reward and to avoid
Neuropsychological measures (Detailed descriptions responding to stimuli that lead to punishment. The participant
available in Text S1) was presented with 10 blocks of eight trials of distinct number
DigitSpan–Backward[63]. TheDigitSpan–Backwardis identity. Each number was presented once during a block. Four
averbalsubtestoftheWAIS-Rusedtomeasureofverbalworking numberswereassociatedwithpunishment(theS–)andfourwith
memory[67].Therawscorewasusedasthedependentvariable. reward (the S +). Participants were randomly assigned to one of
PLOSONE | www.plosone.org 3 June2013 | Volume 8 | Issue 6 | e65566
NeuropsychologyofAntisocialPersonalityDisorder
twoversionsofthetask:thenumbersthatweretheS+andtheS Results
– for one task were the S – and the S + in the other task,
respectively. Reinforcement values were plus or minus 1, 700, Final sample
1400, and 2000 points for the four different S +/S –. The ThecharacteristicsofthethreegroupsarereportedinTable1.
dependent variables were the number of passive avoidance The three groups were similar with respect to age, IQ, and
(commission) errors (i.e., when participants approached a S2) ethnicity. As intended, there were significant differences between
and the number of omission errors (i.e., when the participants all three groups on total and 4-facet PCL-R scores. The non-
did not approach a S+). offenders additionally differed from the offenders by presenting
Ethics statement. The study was approved by the Joint significantly fewer conduct disorder symptoms prior to age 15,
South London and Maudsley and the Institute of Psychiatry lower scores for proactive and reactive aggression, and less
substance misuse. Consistent with previous studies, the ASPD+P
NHS Research and Ethics Committee (reference 06/Q0706/87).
offenders, as compared to the ASPD2P offenders, presented
significantly more symptoms of conduct disorder prior to age 15,
Procedure and apparatus
weresignificantlyyoungeratfirstconvictionforaviolentoffence,
Atthefirstinterview,thestudywasfullyexplainedbothverbally
obtainedhigherscoresforproactiveaggressionandsimilarscores
andinwritingtopotentialparticipants.Afteralloftheirquestions
forreactiveaggression,andtherewasatrendsuggestingthatthey
were answered, participants signed consent forms. All potential
hadaccumulatedmoreconvictionsforviolentcrimesandtheyhad
participants who declined to participate or otherwise did not
higher scores for proactive aggression. The proportions of
participate were not disadvantaged in any other way by not
ASPD+P and ASPD2P offenders with substance use disorders
participatinginthestudy.Participantsincludedinthestudywere
were similar.
reimbursed at minimum hourly wage for each hour of testing
completed. Participants were strongly encouraged to desist from
Digit Span – Backward
using substances two weeks prior to participation and during the
In line with the a priori hypothesis, there was a statistically
period of testing.
significantgroupdifferenceinscoresontheDigitSpan-Backward
Afteralldiagnosticinterviewswerecompleted,anappointment
task, F’ (2, 37.15) =3.57, p=.038, g 2 =.11. Post-hoc tests
was scheduled for neuropsychological testing. Participants were p
indicated that the non-offenders (M=7.33, SD =3.76) repeated
reminded not to use drugs/alcohol prior to testing and that on
moredigitsthantheASPD+P(M=4.69,SD =2.21;p=.009)and
arrivalatthelaboratorysalivaandurinesampleswouldbetaken.
the ASPD2P group, but only at a trend level (M=5.75, SD
Each participant was tested individually in a quiet interview
=2.62; p=.07). Scores for the two ASPD groups did not differ
room. The computer administered tasks were presented on a
(p=.26).
Dell Inspiron 510m Laptop computer with a 15-in. (38.1 cm)
colour monitor with participants seated about 0.5 m from the
Spatial Alternation Task
computer.
No participant failed the task. Contrary to the a priori
hypothesis, scores for the three groups on the Spatial Alteration
Data analytic strategy Task did not differ, F(2, 62) =.36, p=.70, g 2 =.01 (ASPD+P:
p
Data for quality of decision-making on the CGT were highly M=3.12, SD =2.52; ASPD2P: M=3.96, SD =7.13; non-
negatively skewed, with many participants selecting the likely offenders: M=4.52, SD =3.16). Five participants (one ASPD+P,
outcome on the large majority of the trials. Normality could not one ASPD2P and two non-offenders) were identified as outliers
be achieved using an arcsin transformation, thus data for quality withrespecttotheirgroups,butremovingthemfromtheanalyses
of decision-making were analysed using non-parametric Kruskal did not alter the pattern of results, F(2, 57) =1.38, p=.26, g 2
p
–Wallis tests in each condition (ascending versus descending), =.05.
collapsing across box ratio, and in each ratio (9:1, 8:2, 7:3, 6:4)
collapsing across conditions. Significant between-group effects Probabilistic Response Reversal Task
werefollowed-upusingpair-wisepost-hoctests.Deliberationtime
Allparticipantsreachedthelearningcriterionfortheacquisition
data were positively skewed, with many participants responding
and reversal phases of the pair with the 100–0 contingency.
quickly. The distribution of these values was successfully However,threenon-offendersfailedtoreachthecriterionforthe
normalised using a logarithmic (log-10) transformation [69]. acquisition of the pair with the 80–20 contingency. In line with
Data presented in the tables and figures are untransformed. Budhani etal. [54], their data were excluded from the analyses
Continuousvariablesthatconformedtoparametricassumptions sinceitwasuncleariftheseparticipantshadlearnedthestimulus-
wereanalysedusingStudent’st-test,univariateanalysisofvariance responseassociationsothatresponsereversalcouldbeexamined.
(ANOVA) or repeated measures ANOVA. The Welch t’ and F’ Inaddition,datafromoneASPD+PandthreeASPD2Poffenders
tests and the Greenhouse-Geisser correction were applied where andthreenon-offenderswereexcludedfromtheanalysesastheir
assumptions about homogeneity of variance and sphericity were scores were more than 2.5 standard deviations above their
violated, respectively[69].Significantbetween-groupeffects were respective groupmeans.
followed-up using pair-wise comparisons with Fisher’s LSD A 3 (group: ASPD+P, ASPD2P, non-offenders) 62 (contin-
procedure, which is the most powerful technique for post-hoc gency:100–0versus80–20)62(phase:acquisitionversusreversal)
tests involving three groups [69,70]. Effect sizes are reported as mixed model ANOVA revealed that there was a statistically
partial eta-squared (g 2; small $.01, medium $.06, large $.14) significantmaineffectofgroup,F(2,52) =3.94,p=.03,g 2 =.13.
p p
[69]. Categorical variables were analysed using Chi square tests. Post-hoc tests indicated that men with ASPD2P (M=15.7, SD
Results were considered statistically significant at p,.05, two- =6.7; p=.01) and those with ASPD+P, albeit at a trend level
tailed.Notalltheparticipantscompletedallthetasks,sodegreesof (M=14.1, SD =8.6; p=.07), committed more errors than the
freedom vary slightlyacrossanalyses. non-offenders (M=9.6, SD =3.5). Scores of ASPD+P and
ASPD2P offenders did not differ (p=.45). There was a highly
significant main effect of phase, F(1, 52) =51.64, p,.001, g 2
p
PLOSONE | www.plosone.org 4 June2013 | Volume 8 | Issue 6 | e65566
NeuropsychologyofAntisocialPersonalityDisorder
Table1. ComparisonsofSociodemographic, Clinical,and BehaviouralCharacteristics ofNon-offenders,Violent Offenders with
ASPD2P,and ViolentOffenders with ASPD+P.
Measure Non-offenders(n=21) ASPD–P(n=28) ASPD+P(n=17) Groupdifferences
Ageinyears 35.0(8.2) 35.8(8.4) 40.0(9.0) F(63) =1.87
FullScaleIQ 95.1(11.0) 91.9(10.2) 88.9(9.9) F(63) =1.72
%Caucasian 61.9 67.9 41.2 x2(2) =3.22
%withPDotherthanASPD
ClusterA 0 10.7 17.6 x2(2) =3.69
ClusterB 0 14.3 23.5 x2(2) =5.1#
ClusterC 0 7.1 11.8 x2(2) =2.38
PCL–Rtotal 3.8(2.8)a 16.7(4.1)b 28.3(2.1)c F(63) =260.73***
PCL–RInterpersonalfacet 0.4(1.0)a 1.7(1.4)b 3.9(1.3)c F(63) =36.14***
PCL–RAffectivefacet 0.6(0.9)a 2.9(1.8)b 6.2(2.9)c F(63) =63.37***
PCL–RLifestyle 1.9(1.5)a 5.1(2.1)b 6.9(1.8)c F(63) =34.91***
PCL–RAntisocial 0.3(0.6)a 5.8(2.1)b 8.6(1.4)c F(63) =134.60***
CDsymptomsCounts 0.7(1.2)a 4.4(2.8)b 7.6(3.4)a F(63) =32.60***
Ageatfirstviolentconvictions n/a 23.4(8.1)a 16.8(3.3)b t’(43) =23.75**
Numberofviolentconvictions n/a 4.7(3.4) 6.9(5.2) t(43) =1.70#
RPAQAggressiontotal 7.3(3.4)a 17.4(9.1)b 22.3(11.3)b F(62) =15.90***
Proactiveaggression 2.3(3.3)a 8.4(5.4)b 12.5(7.1)c F(62) =17.66***
Reactiveaggression 4.9(3.1)a 9.0(5.8)b 11.5(6.5)b F(62) =7.49**
%Alcohol
Abuse 11.8 25.0 26.7 x2(2) =1.39
Dependence 5.9a 39.3b 26.7b x2(2) =6.04*
%Cannabis
Abuse 5.9 14.3 13.3 x2(2) =.78
Dependence 11.8 32.0 25.0 x2(2) =2.27
%Cocaine
Abuse 0 0 6.7 x2(2) =3.05
Dependence 0 20.0 25.0 x2(2) =4.48
%Stimulants
Abuse 0 3.6 0 x2(2) =1.16
Dependence 0 4.2 0 x2(2) =1.23
%Sedatives
Abuse 0 3.6 0 x2(2) =1.16
Dependence 0 4.0 0 x2(2) =1.18
%Opioid
Abuse 0 3.6 13.3 x2(2) =3.21
Dependence 0 12.0 8.3 x2(2) =2.14
%Hallucinogenics
Abuse 0 7.1 0 x2(2) =2.37
Dependence 0 0 0 n/a
Note.Unlessotherwisestated,meansarepresentedwithstandarddeviationsinparenthesesforeachgroup.Meanswithdifferentsuperscriptswithineachrowindicate
asignificantdifference.PD = PersonalityDisorder;ASPD–P = AntisocialPersonalityDisorderwithoutPsychopathy;ASPD+P = AntisocialPersonalityDisorderwith
Psychopathy;n/a = NotApplicable;PCL–R = PsychopathyChecklist–Revised(Hare,2003);RPAQ = ReactiveProactiveAggressionQuestionnaire(Raineetal.,2006).
OneoffenderwithASPD–PdidnotcompletetheRPAQAggressionQuestionnaire.
#p,.10.**p,.01.***p,.001.
doi:10.1371/journal.pone.0065566.t001
=.50, indicating that participants committed more errors during committed more errors on the stimulus pair with a 80–20
thereversalphase(M=10.6,SD =6.9)thantheacquisitionphase contingency(M=10.3,SD =6.8)thanonthestimuluspairwitha
(M=2.9, SD =2.7) (Figure 1). In addition, there was a highly 100–0 contingency (M=1.8, SD =0.2). There was also a
significant main effect of contingency, F(1, 52) =49.23, p,.001, significant phase by contingency interaction, F(1, 52) =10.84,
g 2 =.49. As can been seen from Figure 1, participants p=.002, g 2 =.17. Importantly, there were also a significant
p p
PLOSONE | www.plosone.org 5 June2013 | Volume 8 | Issue 6 | e65566
NeuropsychologyofAntisocialPersonalityDisorder
group by phase interaction, F(2, 52) =58.38, p=.02, g 2 =.15, Similarly, follow-up post-hoc-tests for the descending condition
p
and a significantgroup byphase bycontingency interaction, F(1, revealed significant differences between the non-offenders and
52) =49.23, p=.049, g 2 =.11. While the three groups boththeASPD+P(p=.01)andASPD2Poffenders(p=.01),while
p
committed a similar number of errors for the two contingency thetwoASPDgroupsdidnotdifferfromeachother(p=.72).The
pairsintheacquisitionphaseandthereversalphaseofthe100–0 quality of decision-making did not differ across groups for the
contingencypair,incomparisontothenon-offenders(M=4.3,SD ratios9:1,x2(2) =3.61,p=.17,g 2 =.05,and8:2,x2(2) =1.14,
p
=3.5),offenderswithASPD2P(M=10.5,SD =6.7;p=.005)and p=.57, g 2 =.02, but there were statistically significant group
p
those with ASPD+P, albeit at a trend level (M=8.1, SD =7.8; differencesinthequalityofdecision-makingforthelessfavourable
p=.10),committedmoreerrorsonthereversalphaseoftheofthe ratios7:3,x2(2) =7.33,p=.03.g 2 =.11,and6:4,x2(2) =8.98,
p
80–20 contingency pair (Figure 1). The comparison between p=.01, g 2 =.14. Post-hoc tests for the ratio 7:3 revealed that,
p
ASPD+PandASPD2P wasnot significant (p=.26). while the there was no difference between the two ASPD groups
(p=.97), there was a significant difference between the non-
CGT offendersandboththeASPD2Poffenders(p=.01)andASPD+P
Deliberation time. A 3 (group: ASPD+P, ASPD2P, non- offenders(p=.03).Asimilarpatternwasobservedfortheratio6:4:
offenders)64 (ratio: 9:1, 8:2, 7:3, 6:4) mixed model ANOVA on while the there was no difference between the two ASPD groups
deliberation time indicated that participants took more time to (p=.87), there was a significant difference between the non-
makedecisionsontrialswithlessfavourableratiosasindicatedby offendersandboththeASPD2Poffenders(p=.01)andASPD+P
a significant main effect of ratio, F(3, 189) =6.38, p,.001, g 2 offenders (p=.02).
p
=.09(Figure2).Therewasastatisticallysignificantmaineffectof Risk-taking, risk adjustment, and delay aversion. A 3
group, F(2, 63) =5.69, p=.005, g 2 =.15. Both offenders with (group: ASPD+P, ASPD2P, non-offenders) 64 (ratio: 9:1, 8:2,
p
ASPD+P (M=3629.6, SD =1285.5; p=.001) and those with 7:3,6:4)62(condition:ascendingversusdescending)ANOVAon
ASPD2P (M=3060.5, SD =1103.2; p=.04) took more time to risk-takingidentifiedasignificantmaineffectofcondition,F(1,61)
makedecisionsthanthenon-offenders(M=2571.7,SD =991.0). =90.48,p,.001,g 2 =.60,andofratio, F(1.52,92.41) =83.13,
p
Inaddition,therewasasignificantgroupbyratiointeraction,F(6, p,.001, g 2 =.58. (The degree of freedom for the repeated
p
189) =3.53, p=.002, g 2 =.10. There were differences in ANOVAis61insteadof63becausetherisk-takingmeasurecould
p
deliberation time between some of the ratios among men with notbecalculatedforoneASPD+PandoneASPD2P,astheybet
ASPD+P (9:1 vs. 7:3, p=.009; 9.1 vs. 6:4, p=004) and among on the colour in the minority (i.e. the less likely outcome) – see
thosewithASPD2P(6:4vs.7:3,p=.008;6:4vs.8:2,p=.001;6:4 description of how Risk-taking is calculated in Text S1.)
vs. 9:1, p=.048), but this was not observed for the non- offender Participants bet more on the descending condition and less as
group (allps..69). the ratio of boxes became less favourable (Figure 2). The main
Qualityofdecision-making. Therewerestatisticallysignif- effect of group and the interactions terms were not statistically
icant group differences in the quality of decision-making in both significant (all Fs ,2.23). A one-way ANOVA on the risk
the ascending, x2 (2) =6.74, p=.034, g 2 =.10, and the adjustment measure indicated that there was a trend for a group
p
descending, conditions, x2 (2) =9.42, p=.009, g 2 =.14. difference,F(2,63) =2.97,p=.058,g 2 =.60,suggestingthatthe
p p
Follow-up post-hoc tests for the ascending condition indicated two ASPD groups adjusted their betting less than the non-
significant differences between the non-offenders and both the offenders(Figure2).Finally,therewasnomaineffectofgroupon
ASPD2P(p=.02)andASPD+Poffenders(p=.04).Nodifferences the delay aversion measure, F(2, 61) =.12, p=.99, g 2 =.00,
p
weredetectedinscoresofthetwoASPDgroups(p=1)(Figure2). indicating nogroup difference inimpulsivity.
Figure1.PerformanceofthethreegroupsontheProbabilisticResponseReversalTaskasindicatedbythenumberoferrorsto
criterionmadeintheacquisitionandreversalphasesofthepair100–0(left)andofthepair80–20(right).Maximumerrors =40.Error
barsindicatestandarderrorofthemean.ASPD–P = AntisocialPersonalityDisorderwithoutPsychopathy;ASPD+P = AntisocialPersonalityDisorder
withPsychopathy.
doi:10.1371/journal.pone.0065566.g001
PLOSONE | www.plosone.org 6 June2013 | Volume 8 | Issue 6 | e65566
NeuropsychologyofAntisocialPersonalityDisorder
Figure2.PerformanceofthethreegroupsontheCGTasindicatedbythedeliberationtimebyratio(topleft),qualityofdecision-
makingby ratio (top right), qualityof decision-making by condition(bottomleft), risk-takingby ratio (bottomright). Errorbars
indicatestandarderrorofthemean.ASPD–P = AntisocialPersonalityDisorderwithoutPsychopathy;ASPD+P = AntisocialPersonalityDisorderwith
Psychopathy.
doi:10.1371/journal.pone.0065566.g002
Passive Avoidance Learning Task was not statistically significant, F(10.49, 325.25) =1.69, p=.08,
Following R.J.R. Blair etal [36], each initial presentation of a gp2 =.05.
stimulus was treated as a learning trial, so results from the first
block were omitted from the analysis. A 3 (group: ASPD+P, Discussion
ASPD2P, non-offenders) 64 (punishment values: 1, 700, 1400,
Thepresentstudyisthefirsttouseacomprehensivebatteryof
2000)69(blocks)modelANOVAwasperformedonthenumber
neuropsychologicalteststoassesscoolandhotEF,comparingtwo
of commission errors. There was a statistically significant main
groups of violent offenders and one group of healthy non-
effect of block, F(5.91, 366.63) =13.67, p,.001, g 2 =.18,
p offenders. A summary of the results is presented in Table 2. As
indicating a decrease in the number of commission errors as the
hypothesized, both theviolent offenders with ASPD+P and those
task progressed (Figure 3). The main effect of group fell short of
with ASPD2P showed similarly poor performance on the Digit
statistical significance, F(2, 62) =2.92, p=.06, g 2 =.09,
p Span–BackwardtestassessingcoolEF,andonseveraltestsofhot
suggesting that the two ASPD groups tended to make more
EFascomparedtohealthynon-offenders.However,infirmingour
commission errors than the non-offenders. There was no
second hypothesis, the ASPD+P offenders did not make more
statistically significant main effect of punishment or interaction
commission errors than the ASPD2P offenders on the Passive
effects (all Fs ,1.1).
Avoidance Learning Task. In fact, the performance of the two
A 3 (group: ASPD+P, ASPD2P, non-offenders) 64 (reward
groups of violent offendersdid not differ on anyof thetasks.
values: 1, 700, 1400, 2000) mixed model ANOVA conducted on
Importantly, violent offenders with a life-long history of
the omission errors revealed a main effect of reward, F(2.47,
antisocialandaggressivebehaviourascomparedtonon-offenders
153.26) =2.94, p=.045, g 2 =.05. As illustrated in Figure 3,
p matchedforage,IQandethnicity,showeddeficitsinanarrayof
participantsmadefeweromissionerrorsforsmallerrewardvalues.
both cool and hot EF. Both offenders with ASPD+P and those
There was also a main effect of block, F(5.25, 325.23) =15.83, withASPD2PshowedimpairedperformanceontheDigitSpan–
p,.001, gp2 =.20. As can be seen from Figure 3, participants Backward,ameasureofworkingverbalmemoryindexingcoolEF.
mademoreerrorsasthetaskprogressed.Whilethemaineffectof Impairedverbalworkingmemorylimitsreflectionduringproblem
groupwasnotsignificant,F(2,62) =1.17,p=.32,gp2 =.04,there solving, particularly in situations requiring adaptive social
was a significant group by reward interaction, F(4.94, 153.26) responses[7,71,72].Whileapreviousstudyreportedthatoffenders
=2.65,p=.03,gp2 =.08.WhiletheperformanceoftheASPD2P with ASPD+P performed similarly to offenders without psychop-
and the non-offender groups was not influenced by the level of athy on this task [53], the results of the present study show that,
reward, the ASPD+P offenders committed fewer errors at the when compared to healthy non-offenders, both offenders with
lowestlevelofreward(+1)ascomparedtolevels+700and+2000 ASPD+P and those with ASPD2P display deficits in verbal
(ps =.018and.008,respectively).Thegroupbyblockinteraction workingmemoryasdopersistentlyaggressivechildrenandadults
[11,72]. Interestingly, two previous studies did not find a verbal
PLOSONE | www.plosone.org 7 June2013 | Volume 8 | Issue 6 | e65566
NeuropsychologyofAntisocialPersonalityDisorder
Figure 3. Performance of the three groups on the Passive Avoidance Learning Task as indicated by the number of passive
avoidanceerrorsbyblock(topleft),numberofpassiveavoidanceerrorsbypunishmentlevels(topright),numberofomission
errorsbyblock(bottomleft),andnumberofomissionerrorsbyrewardlevels(bottomright).Errorbarsindicatestandarderrorofthe
mean.ASPD–P = AntisocialPersonalityDisorderwithoutPsychopathy;ASPD+P = AntisocialPersonalityDisorderwithPsychopathy.
doi:10.1371/journal.pone.0065566.g003
Table2.SummaryofTaskPerformanceofNon-offenders,ViolentOffenderswithASPD2P,andViolentOffenderswithASPD+P.
Neuropsychologicalmeasure GroupandInteractionEffects Post-hoc{
DigitSpan–Backward Group Non-offenders.ASPD+P,ASPD2P#
SpatialAlternationTask – –
PassiveAvoidanceLearningTask
Commissionerrors Group# Non-offenders.ASPD+P#,ASPD2P#
Omissionerrors – –
ProbabilisticResponseReversal Group Non-offenders.ASPD+P#,ASPD2P
Acquisitionerrors – –
Reversalerrors80–20pair GroupxPhase Non-offenders.ASPD+P#,ASPD2P
CambridgeGambleTask
Deliberationtime Group Non-offenders.ASPD+P,ASPD2P
Qualityofdecision-making GroupqQ Non-offenders.ASPD+P,ASPD2P
Groupratio7:3 Non-offenders.ASPD+P,ASPD2P
Groupratio6:4 Non-offenders.ASPD+P,ASPD2P
Risk-taking – –
Riskadjustment Group# Non-offenders.ASPD+P#,ASPD2P#
Delayaversion – –
Note.Betterperformance.worseperformance.
{TheperformanceoftheASPD+PandASPD2Pdidnotdifferonanyofthetasks.
#Trendforgroupdifference.
–Nostatisticallysignificantgroupdifference.
qAscendingcondition.
QDescendingcondition.
doi:10.1371/journal.pone.0065566.t002
PLOSONE | www.plosone.org 8 June2013 | Volume 8 | Issue 6 | e65566
NeuropsychologyofAntisocialPersonalityDisorder
working memory deficit among men with ASPD2P with no scores,areknowntobeimpairedonsomeoftheseprocesses,this
historyofcriminalityorsubstancemisuse[43,73].Takentogether, mightexplainresultsofpreviousstudiesofrisk-takingontheIowa
the results of the present study and the extant literature may be Gambling Task. Much evidence indicates that men with ASPD,
interpreted to suggest that impaired verbal working memory is regardless of psychopathy, show impulsive behaviour in the form
associated with a life-longpattern ofaggressive behaviour. of impaired response inhibition (e.g., [50,57]). Results of the
While both groups of violent offenders showed poorer verbal presentstudysuggestthattheymaynotdisplayimpulsivitydefined
working memory than the non-offenders, on another test of cool as delay aversion. A previous study examined delay aversion
EF, the Spatial Alternation Task, they showed no impairment. amongoffenderswithandwithoutpsychopathy[78].Lowanxious
Thisfindingisconsistentwithresultsoftheonlypreviousstudyto psychopaths, in comparison to low anxious non-psychopaths,
assessoffenderswiththistaskinwhichoffenderswithandwithout delayed gratification less often in the condition that involved
psychopathy performed similarly [47]. Our results extend the rewards and punishments, but not in the condition that involved
previousfindingsbyshowingthat,whiletheroleoftheDLPFCin rewards only. Clearly, additional research examining different
thealterationofmotorresponsestospatiallocationsonthebasisof forms of impulsivity in relation to ASPD and psychopathy is
reinforcement information is not impaired in offenders with warranted.
ASPD+PandthosewithASPD2P,theydoexhibitimpairmentsin On another test of hot EF, the Passive Avoidance Learning
verbal working memory, another cool EF subsumed by the Task,theviolentoffenderswereimpairedascomparedtothenon-
DLPFC. offenders. There was a trend (p=.06) indicating that the two
Thetwogroupsofviolentoffendersalsoshowedimpairmentsin offender groups made more commission errors than the non-
hotEFascomparedtothenon-offenders.Inthereversalphaseof offenders, but, contrary to our second hypothesis, no evidence of
the Probabilistic Response Reversal Task, offenders with anincreasednumberofcommissionerrorsintheASPD+Pgroup
ASPD2P committed significantly more errors than the non- comparedtotheASPD2Pgroup.Ashypothesized,therewasno
offendersintheconditionwherethestimulus-responseassociation evidence of a group difference in omission errors. These results
was less clear (i.e., 80–20 contingency pair). The offenders with showthatthisfailuretolearnfrompunishmentcuescharacterizes
ASPD+Pshoweda trendinthesamedirection.Aprevious study not only violent offenders with ASPD+P, but also those with
of offenders using the same paradigm reported that those with ASPD2P. The results of the present study suggest that both
psychopathy, as compared to those without psychopathy, com- violent offenders with ASPD+P and those with ASPD2P have
mittedmoreerrorsonthereversalphaseofthe100–0and80–20 difficulty in stimulus-punishment associations.
pairs [54]. However, the finding that the ASPD2P offenders Violent offenders with ASPD, both those with and without
showed impairments on the response reversal task is consistent additional diagnoses of psychopathy, showed impairments in
with a previous study [74] showing that offenders with moderate verbal working memory, and in adaptive affective decision-
PCL-Rscores(between21–29;insufficienttowarrantadiagnosis making. They failed to learn from punishment cues, to change
ofpsychopathyintheU.S.)wereimpairedinresponsereversalin their behaviour in the face of changing contingencies, and made
comparison to offenders without psychopathy (PCL-R scores poorer quality decisions despite longer periods of deliberation
range: 0–20). beforesuchdecisions.Thecombinationoftheseimpairmentsmay
OntheCGT,thetwogroupsofoffenders,ascomparedtothe gosomewaytowardsexplainingwhyviolentoffenderswithASPD
non-offenders, also displayed poorer quality of decision-making with and without psychopathy are characterized by irresponsibil-
despite increased deliberation times and a strong trend for less ity,recklessness, persistent aggressivebehaviour, andengagement
modulationoftheirbettingastheprobabilityoflossincreased,but in multiple other types of antisocial behaviours despite knowing
similar levels of impulsivity and risk-taking. The two groups of that such behaviour will likely lead to negative consequences for
violentoffenders,likethenon-offenders,deliberatedlongerbefore themselves and/orothers [14].
makingadecisionastheboxratiobecamelessfavourable,thereby Thesefindingsneedtobereplicated.Theabsenceofstatistically
showinganunderstandingofthetrial-by-trialprobabilitiesandof significantdifferencesinperformanceonanyoftheneuropsycho-
the increased risk of losing points. This pattern of results – long logicaltasksbetweentheviolentoffenderswithASPD+Pandthose
delayandpoordecisionmaking-resemblesthatobservedamong withASPD2PsuggestsshareddeficitsincoolandhotEF,atleast
patients with lesions in the VMPFC (e.g., [68,75]; but see [76]). basedonthetasksusedandtheprocessestheyindex.Theseresults
Thus, although they were aware of the increased risk of loss, the areconsistent,however,withourstructuralbrainimagingfindings
offenders failed to adjust their behaviour to the increasing risk of on an overlapping sample showing differences between violent
losing points, just as they persist in engaging in antisocial offenders with ASPD+P and those with ASPD2P in gray matter
behaviour despite knowing that it will likely lead to negative volume of the superior/medial prefrontal cortex and temporal
consequences. poles, but no differences in gray matter volume the amygdala,
Perhaps one of the most novel aspects of this study are the VMPFC or DLPFC [23]. Additionally, reduced fractional
results of the delay aversion and risk taking measures, which anisotropy in the right uncinate fasciculus (the primary white
indicate that the two groups of ASPD offenders were no more matter tract connecting the VMPFC and the anterior temporal
impulsive or risk-taking (at least not at the two most favourable lobes) has been demonstrated in both violent men with ASPD+P
ratios) than the non-offenders. These results are likely due to the [79,80]andthosewithASPD2P[81].Thus,theresultsfromthe
fact that the CGT is a decision-making task in which outcome present study are consistent with this emerging evidence from
probabilities andtheassociatedrisksareexplicit.Bycontrast,the brain imaging studies and might reflect that fact that, while the
few studies that have examined affective decision-making of men offenderswithASPD+Pscoredtwiceashighastheoffenderswith
with ASPD+P or ASPD2P and shown increased risk-taking ASPD2P on the PCL-R facet 1 and facet 2 indexing the core
behaviour used the Iowa Gambling Task in which outcome interpersonal and affective features of the syndrome of psychop-
probabilities are unknown. This latter task relies on the integrity athy,thesetwogroupsofviolentoffenderssharemanycharacter-
and coordination of several processes, including stimulus-rein- istics, most importantly a childhood onset of conduct problems
forcement learning, reversal learning, set-shifting, and working that persist into adulthood and violent behaviour. The present
memory [77]. Since men with ASPD, regardless of psychopathy results suggest that both offenders with ASPD+P and those with
PLOSONE | www.plosone.org 9 June2013 | Volume 8 | Issue 6 | e65566
NeuropsychologyofAntisocialPersonalityDisorder
ASPD2P present similar EF impairments despite differences in The findings from this study provide novel evidence that, in
thetypesofaggressivebehaviourinwhichtheyengage,personality comparison to healthy non-offenders, violent offenders with
traits, emotion processing, and structural and functional brain ASPD+P and violent offenders with ASPD2P present impair-
anomalies. ments in both cool and hot EF such as verbal working memory,
Several methodological limitations should be considered in response reversal, affective decision-making under risk, and
interpretingtheresultsofthepresentstudy.One,theremayhave stimulus-reinforcement-based decision-making. The performance
beenalackofstatisticalpowertodetectgroupdifferencesresulting of the two groups of offenders on these tasks did not differ
from the relatively small number of violent offenders with suggesting shared deficits in EF, at least based on the tasks used
ASPD+P. The number of participants, however, was similar to and the processes they index. The combination of these
manyofthepreviousneuropsychologicalstudiesinthefield(e.g., impairments may help to explain why violent offenders with
[51,54,82]).Two,theuseofthevalidatedPCL-Rcut-offscorefor ASPD, both those with and without psychopathy, persist in
Europeanoffenderstoidentifythesyndromeofpsychopathymay engaging in antisocial behaviours despite knowing the risks of
have lessened the likelihood of observing cognitive impairments. negativeconsequencestothemselvesand/orothers[14].Crucial-
Howeverthisisunlikelyasthepatternofresultsgenerallyshowed ly,giventhedifferencesintheirresponsestocognitive-behavioural
that impairments previously reported as characterizing offenders rehabilitationprogramsaimedatreducingviolenceandrecidivism
withpsychopathyalsocharacterizedthosewithASPD2P.Three, [12,13],additionalresearchisneededtofurtherunderstandingof
as is evident from the review of the literature and the present the neurobiological and psychological similarities and differences
results,findingsaboutpsychopathydependtoalargeextentonthe in these two types of offenders. Functional magnetic resonance
comparisongroupusedineachstudy.Therefore,allanalyseswere imaging,whichhasneverbeenusedtodirectlycomparethesetwo
re-run excludingsix offenderswithASPD2Pwhosetotal PCL-R groups of violent offenders, would be highly informative in this
scores were between 24 and 20. Again, no significant group regard since it can detect subtle alterations in neural processing
differences between offenders with ASPD+P and ASPD2P were that maynot be observablewith behavioural indices.
found. Four, the violent offenders with ASPD, like almost all
peoplewithASPD[83],hadahistoryofsubstancemisuse.While Supporting Information
objective tests assured that the participants were not tested when
TextS1 Supporting text.
intoxicated,itispossiblethatpastsubstancemisuseledtosomeof
(DOCX)
thedeficitsinperformancethatwereobserved.Finally,theuseof
digits as central stimuli in two of the tasks (i.e., the Digit Span –
Acknowledgments
Backward and the Passive Avoidance Learning Task) may not
havebeenidealfortestingindividualswithlowlevelsofeducation. Wethanktheparticipantsfortheirtime,theNationalProbationServicefor
Thisstudyalsohasseveralstrengths.One,itisthefirststudyto theircollaborationandthefollowingpersonswhorecruitedandassessed
participants: Ms. Sam Prior, Ms. Clare Goodwin, Mr. William Wain-
directlycontrastvariousaspectsofhotEFinviolentoffenderswith
wright,Ms.RebeccaBrewer,MsSarahGregory,Mr.RubenAzevedo,Mr.
ASPD+PandviolentoffenderswithASPD2Pandtocomparetest
Francis Vergunst, Ms. Lucy Butler, Ms. Leila Niknejad, Dr. Anna
performancetothatofhealthynon-offenders.Two,thisisthefirst
Plodowski,Dr.PhilipBaker,Dr.TimothyRogers,Dr.PreethiChabbra,
study to include offenders who were convicted of several violent Dr. Stephen Attard, Dr. Seema Sukhwal, Dr. Nathan Kolla, Dr. Paul
crimes, diagnosed by forensic psychiatrists using standardized, Wallang,andDr.ClareConway.
validated interview protocols, and examined using a comprehen-
sivebatteryofneuropsychologicalteststhatassessedbothcooland Author Contributions
hot EF. Three, this study was the first to examine affective
Conceivedanddesignedtheexperiments:SDBEVVKNBSH.Performed
decision-making under risk among men with ASPD+P and men
the experiments: SDB. Analyzed the data: SDB. Contributed reagents/
withASPD2P.Finally,thethreegroupsdidnotdifferintermsof materials/analysistools:SDBEVNBVKSH.Wrotethepaper:SDBEV
age, IQ,and ethnicity. VKNBSH.
References
1. KratzerL,HodginsS(1999)Atypologyofoffenders:AtestofMoffitt’stheory 10. RaineA,MoffittTE,CaspiA,LoeberR,Stouthamer-LoeberM,etal.(2005)
amongmalesandfemalesfromchildhoodtoage30.CriminalBehaviourand Neurocognitiveimpairmentsinboysonthelife-coursepersistentantisocialpath.
MentalHealth9:57–73. JournalofAbnormalPsychology114:38–49.
2. MoffittTE,CaspiA,HarringtonH,MilneBJ(2002)Malesonthelife-course- 11. Se´guinJR,SylversP,LilienfeldS(2007)Theneuropsychologyofviolence.In:
persistentandadolescence-limitedantisocialpathways:follow-upatage26years. FlanneryDJ,VazsonyiAT,WaldmanID,editors.TheCambridgehandbookof
DevelopmentandPsychopathology14:179–207. violentbehaviorandaggression.NewYork:CambridgeUniversityPress.187–
3. AmericanPsychiatricAssociation (2000) Diagnostic and statistical manual of 214.
mentaldisorders(4thed.,Textrevision).Washington,DC:AmericanPsychiatric 12. OgloffJRP,WongS,GreenwoodA(1990)Treatingcriminalpsychopathsina
Association. therapeuticcommunityprogram.BehavioralSciencesandtheLaw8:181–190.
4. DeBritoSA,HodginsS(2009)Antisocialpersonalitydisorder.In:McMurran 13. Olver ME, Wong S (2011) Predictors of sex offender treatment dropout:
M,HowardR,editors.Personality,personalitydisorder,andriskofviolence:An Psychopathy,sexoffenderrisk,andresponsivityimplications.Psychology,Crime
evidencebasedapproachUnitedKingdom:Wiley.133–154. andLaw17:457–471.
5. Zelazo PD, Mu¨ller U (2002) Executive function in typical and atypical 14. HareRD(2003)ManualfortheRevisedPsychopathyChecklist.Toronto,ON,
development In: Goswami U, editor. Handbook of childhood cognitive Canada:Multi-HealthSystems.
development.Malden,MA:Blackwell.445–469. 15. HareRD,HartSD,HarpurTJ(1991)PsychopathyandtheDSM-IVcriteriafor
6. BlairRJR,MitchellDGV,BlairKS(2005)Thepsychopath:Emotionandthe antisocialpersonalitydisorder.JournalofAbnormalPsychology100:391–398.
brain.Oxford:Blackwell. 16. HareRD(1991)TheHarePsychopathyChecklist-Revised.Toronto,Ontario:
7. Giancola PR (1995) Evidence for dorsolateral and orbital prefrontal cortical Multi-HealthSystems.
involvementintheexpressionofaggressivebehavior.AggressiveBehavior21: 17. WongS(1985)Criminalandinstitutionalbehavioursofpsychopaths.Ottawa,
431–450. Ontario:MinistryoftheSolicitorGeneralofCanada.
8. MoffittTE(1993)Theneuropsychologyofconductdisorder.Developmentand 18. PorterS,PorterS(2007)Psychopathyandviolentcrime.In:Herve´H,YuilleJC,
Psychopathology5:135–151. editors. The Psychopath: Theory, Research, and Practice. Mahwah, NJ:
9. OgilvieJM,StewartAL,ChanRCK,ShumDHK(2011)Neuropsychological LawrenceErlbaumAssociatesPublishers.287–300.
measures of executive function and antisocial behavior: A meta-analysis. 19. HareRD,McPhersonLM(1984)Violentandaggressivebehaviorbycriminal
Criminology49:1063–1107. psychopaths.InternationalJournalofLawandPsychiatry7:35–50.
PLOSONE | www.plosone.org 10 June2013 | Volume 8 | Issue 6 | e65566