Neuropsychiatry of Frontal Lobe Dysfunction in Violent and Criminal Behaviour: a Critical Review

J Neurol Neurosurg Psychiatry: first published as 10.1136/jnnp.71.6.720 on 1 December 2001. Downloaded from 720 J Neurol Neurosurg Psychiatry 2001;71:720–726

ADVANCES IN NEUROPSYCHIATRY

Neuropsychiatry of frontal lobe dysfunction in violent and criminal behaviour: a critical review

M C Brower, B H Price

Abstract impulsive, trivially motivated, or habitual Objectives—To establish the link between aggression.5–7 But whereas clinical observation frontal lobe dysfunction and violent and and current theories of prefrontal network criminal behaviour, based on a review of function suggest that frontal lobe disorders relevant literature. may contribute to violent and criminal behav- Methods—Articles relating evidence of iour, the strength of this hypothesised associ- frontal lobe dysfunction with violence or ation has yet to be established. crime were collected through a MEDLINE This paper evaluates the evidence for a search using the keyword “frontal lobe” causal relationship between abnormal frontal combined with the terms “aggression,” lobe function and violent crime, based on a “violence,” “crime,” “antisocial personal- review of current research literature. We ity disorder,” “psychopathy,” “impulse located articles using a MEDLINE search from control disorders”, and “episodic dyscon- 1966 through 2000, combining the keyword trol.” Reference lists were then searched “frontal lobe” with the terms “aggression,” for additional articles. “violence,” “crime,” “antisocial personality Results—High rates of neuropsychiatric disorder,” “psychopathy,” “impulse control abnormalities reported in persons with disorders”, and “episodic dyscontrol.” We then violent and criminal behaviour suggest an conducted a hand search of relevant reference association between aggressive dyscontrol lists. Articles were selected for review if they and brain injury, especially involving the contained clinical, laboratory, or neuropsycho- frontal lobes. The studies reviewed sup- logical test data relating frontal lobe function to port an association between frontal lobe aggression, crime, or violence. In this review, dysfunction and increased aggressive and we adopt the definition of aggression as any antisocial behaviour. Focal orbitofrontal threatening or physically assaultive behaviour injury is specifically associated with in- directed at persons or the environment. creased aggression. Deficits in frontal “Violence” refers to actions that inflict physical executive function may increase the likeli- harm in violation of social norms. We divide

hood of future aggression, but no study our findings under the following headings: (1) http://jnnp.bmj.com/ Secure Care Program, has reliably demonstrated a characteristic studies relating clinical focal frontal lobe disor- Department of pattern of frontal network dysfunction ders to violent behaviour; (2) studies reporting Psychiatry, Taunton neuropsychological measures of frontal lobe State Hospital, predictive of violent crime. Taunton, MA, USA, Conclusions—Clinically significant focal function in aggressive and antisocial subjects; and Harvard Medical frontal lobe dysfunction is associated with (3) studies of clinical neurological findings in School, Boston, MA, aggressive dyscontrol, but the increased violent and criminal populations; and (4) neu- USA risk of violence seems less than is widely roimaging studies of aggressive and violent

M C Brower presumed. Evidence is strongest for an subjects. We conclude by assessing the magni- on October 2, 2021 by guest. Protected copyright. tude and speciﬁcity of the hypothesised link Departments of association between focal prefrontal dam- Neurology, McLean age and an impulsive subtype of aggres- between frontal lobe dysfunction and violence, Hospital, Belmont, sive behaviour. and discuss implications for future research. MA, Massachusetts (J Neurol Neurosurg Psychiatry 2001;71:720–726) General Hospital, Boston, MA, and Keywords: frontal lobe dysfunction; aggression; vio- Harvard Medical lence Focal frontal lobe disorders and violent School, Boston, MA, behaviour USA Case studies as far back as 1835 have reported BHPrice Reports describing high rates of neuropsychiat- the onset of antisocial personality traits after frontal lobe injury.8 Such cases typically involve Correspondence to: ric abnormalities among death row inmates, Dr B H Price, Department of forensic psychiatric inpatients, and other per- damage to the orbitofrontal cortex, which Neurology, Mclean Hospital, sons with histories of violence have led to clinical observation has associated with “poor 115 Mill Street, Belmont, impulse control, explosive aggressive out- MA 02478, USA assertions that evidence of brain-behavioural [email protected] impairment may mitigate or excuse criminal bursts, inappropriate verbal lewdness, jocular- conduct.1–4 Frontal lobe dysfunction in particu- ity, and lack of interpersonal sensitivity.”9 Such Received 18 July 2000 and in gross dysregulation of aVect and behaviour ﬁnal form lar, has been invoked to explain the actions of 7 June 2001 some persons charged with, or convicted of, may occur while cognitive, motor, and sensory Accepted 2 July 2001 violent crimes, who apparently fail to inhibit functioning remain relatively intact.10 Blumer

www.jnnp.com J Neurol Neurosurg Psychiatry: first published as 10.1136/jnnp.71.6.720 on 1 December 2001. Downloaded from Neuropsychiatry of frontal lobe dysfunction in violent and criminal behaviour 721

and Benson dubbed this orbitofrontal syn- aggression, substance misuse, stability of em- drome “pseudopsychopathy,”8 based on simi- ployment, socioeconomic status, the presence larities to psychopathy—a personality type of psychiatric symptoms or disorders other that, as defined by reliable and valid checklist than depression, or criminal charges or other criteria, is strongly associated with violence and legal involvement. Without such data, it criminality.11–13 remains unclear how much of the increases in Case reports have described a similar aggressive behaviour found can be specifically syndrome of “acquired sociopathy” in persons attributed to focal frontal lobe injury. who had ventromedial prefrontal injury in adulthood.14–16 Although showing minimal im- Neuropsychological studies of aggressive pairments on standard neuropsychological and antisocial subjects tests of intelligence and executive functions, A previous comprehensive review of neuro- these subjects display marked deficits in real psychological studies by Kandel and Freed life tasks demanding judgment, awareness of (1989) found that “evidence for the association socially appropriate conduct, and the capacity between specifically violent criminal behaviour to assess future consequences.17 Persons with and frontal lobe dysfunction is weak at best.”27 frontal network damage acquired before the A subsequent review by Pennington and age of 8 have also been reported to have adult OzonoV concluded that comorbid attention histories of recurrent impulsive, aggressive, and deficit hyperactivity disorder (ADHD) most antisocial behaviour, associated with primary likely accounted for deficits in frontal executive deficits in tests of executive function, poor function linked with adolescent conduct disor- abstract conceptual thinking, inability to envi- der, but considered that ADHD might worsen sion another person’s subjective experience, aggression in such cases.28 Table 1 summarises and immature moral reasoning.18–20 One report, results of relevant neuropsychological studies however, has described two cases of improve- reported since 1989.29–36 ment in impulsive and antisocial behaviour One study reported that errors on a single after frontal traumatic brain injury in adult- subtest in a battery of executive function meas- hood.21 ures correlated significantly with a diagnosis of Large systematic studies on cohorts of war antisocial personality disorder in a male veterans with head injury have also tended to community sample.29 A small study of subjects find an association between frontal lobe lesions addicted to cocaine, all of whom met diagnos- and aggressive or antisocial behaviour, al- tic criteria for antisocial personality disorder, though the prevalence of actual violent crime found that high violence subjects as a group seems small. Among German researchers who scored significantly better than low violence described personality changes in first world subjects on a widely accepted measure of fron- war and second world war veterans with frontal tal executive functioning.30 By contrast, two lobe injuries, Kleist found a consistent relation studies using a laboratory based procedure between orbitofrontal lesions and subsequent designed to elicit aggressive behaviour have antisocial behaviour.8 Five patients (3%) in a correlated decreased performance on executive sample of 144 British second world war veter- function tests with increased aggression in ans with penetrating head injury committed community samples of male subjects without “crimes and misdemeanors,” though all five neurological, psychiatric, or substance misuse had damage limited to the frontal lobes.22 Less histories.31 32 than 5% of all subjects with frontal lobe injury A prospective study found that low scores on http://jnnp.bmj.com/ in a similar study of Finnish second world war executive function tests significantly predicted veterans had a history of criminal conviction, self reported aggression in 10 to 12 year old and only one had committed a violent boys with paternal histories of substance oVence.23 misuse, but the results did not control for The Vietnam Head Injury Study (VHIS) ADHD.33 The same lead authors conducted a found that subjects with lesions limited to the subsequent case-control study of aggression in frontal lobes tended to show more aggressive conduct disordered adolescent females, con- and violent behaviours compared with patients trolling for ADHD: low executive function on October 2, 2021 by guest. Protected copyright. with non-frontal head injury and controls scores retained a significant independent corre- without head injury.24 About 14% of subjects lation with physically aggressive antisocial with frontal lobe injury engaged in fights or behaviour.34 Ina1yearprospective study of damaged property, compared with about 4% of forensic psychiatric inpatients who had com- controls without head injury. The study also mitted a violent crime, low scores on three tests found a significant association between in- of frontal executive function significantly pre- creased aggression and focal mediofrontal and dicted frequency of aggression, accounting for orbitofrontal injury identified on brain CT. 57% of the variance.35 Although studies of psy- Reports have also found higher rates of antiso- chopathic subjects have not demonstrated cial behaviour (including stealing, physical frontal executive dysfunction,27 one report assault, and sexual comments or advances) in found that, compared with non-psychopathic patients with frontotemporal dementia, even criminals, psychopathic criminals showed sig- when compared with equally cognitively im- nificant deficits on tests specifically selected to paired patients with Alzheimer’s disease.25 26 assess orbitofrontal and ventromedial function- All of these studies were retrospective, and ing.36 most did not adequately control for known Overall, these neuropsychological studies violence risk factors. The VHIS study, for tend to support a significant association example, did not report on prior history of between prefrontal executive dysfunction

www.jnnp.com J Neurol Neurosurg Psychiatry: first published as 10.1136/jnnp.71.6.720 on 1 December 2001. Downloaded from 722 Brower, Price

measured by neuropsychological testing and and assaultive behaviour.45 A study of 333 pris- increased antisocial and aggressive behaviour. oners referred for evaluation after being In populations with prior risk of antisocial charged with a violent crime specifically related behaviour or aggression, the presence of frontal EEG findings to “habitual physical executive function deficits may have value in aggression or explosive rages.”46 After exclusion assessing the future likelihood of aggression. of subjects with clinical evidence of structural Studies of psychopathic subjects, however, brain damage, 56.9% of habitually aggressive suggest that standard tests of executive func- subjects had EEG abnormalities (62.2% fron- tion may miss orbital or ventromedial prefrontal), compared with 11.8% of other subjects tal dysfunction relevant to aggression. Matura- who had committed a single, isolated aggres- tional delay in the development of the sive act. prefrontal cortex,37 or deficient education or Another report of neurological findings in 31 socialisation related to psychosocial deprivation, or both29 34 38 may also account for deficits subjects, who were referred by attorneys in found in neuropsychological test performance connection with claims of mitigation related to linked to aggression. murder charges, found that 64.5% showed “some physical evidence of frontal dysfunction.”47 Signs elicited included snout, suck and grasp reflexes, paratonia, abnormal smooth Clinical neurological findings in violent pursuit eye movements, diminished word and criminal populations fluency, and reciprocal hand movement errors. A previous review concluded that antisocial Examination disclosed three or more signs in subjects had more EEG abnormalities, pre- 39 32.3%, two signs in 9.7%, and one sign in dominantly anterior. Subsequent studies have 22.6%. A retrospective chart review found that continued to find abnormal frontal EEG activ- a frontal lobe lesion was the best predictor of ity, as well as diminished frontal event related involvement in a violent episode among potentials, correlating with antisocial personal- inpatients on a neuropsychiatric unit, account- ity disorder or histories of aggression.40–42 In a ing for 11% of the variance, ahead of number of study of adult male drug misusers, for example, inpatient days, seizure disorder, history of subjects rated as “high aggressive” on a self 48 report scale showed statistically significant alcohol misuse, and aVective psychosis. A frontal EEG slowing relative to the “low prospective study of frontal lobe function and aggressive” group.43 A case series describing violence in psychiatric inpatients with mood recurrent, severe aggression in mentally re- and psychotic disorders found no significant tarded subjects attributed their behaviour to diVerence in frontal or other neurological find- frontal lobe seizures, based on phenomenologi- ings between violent and non-violent patients.49 cal similarities to frontal ictal automatisms; Persistently violent patients, however, had 60% of subjects had abnormal EEGs, with significantly more frontal lobe impairment 20% having focal frontal or frontotemporal than transiently violent patients and their findings.44 Case reports have also linked orbit- behaviour seemed less responsive to environ- ofrontal EEG spiking to violent hallucinations mental factors. A related study by the same

Table 1 Neuropsychological studies of frontal lobe function in aggressive and antisocial subjects

Reference Subjects Neuropsychological test measure Results http://jnnp.bmj.com/

Deckel et al29 89 men, age 21–25, recruited from + WCST, COWT, PMT + PMT VII Maze added significantly to community + EEG prediction of ASPD + Decrease in left versus right frontal EEG activation in ASPD Rosse et al30 14 male crack cocaine addicted inpatients WCST + Low violence group made significantly with ASPD more perseverative errors than high violence group Giancola and Laboratory aggression in community SOP, CAT + Increased aggression correlated with Zeichner31 sample, 72 white men aged 18–32 decreased FL test performance on October 2, 2021 by guest. Protected copyright. Lau et al32 Laboratory aggression and alcohol SOP, CAT + Increased intoxication correlated with intoxication in community sample, 114 increased aggression male social drinkers + Lowest 25% on FL tests significantly more aggressive Giancola et al33 Self reported aggression in 198 males aged PMT, vigilance task, forbidden toy task, + Significantly lower EF scores in boys at risk 10–12 at risk for substance misuse, and motor restraint task, WISC-R block design for substance abuse controls; 2 year prospective follow up + Low EF scores significantly predicted increased aggression Giancola et al34 Case-control study of aggressive antisocial PMT, vigilance task, motor restraint task, + CD subjects had significantly lower EF behaviour in 249 conduct disordered SCWT, WISC-R/WAIS-R block design, scores females, age 14–18, and controls picture arrangement and object assembly + Low EF scores significantly correlated with aggression, even when controlling for ADHD Foster et al35 One year prospective study of aggression in + SCWT, JLOT, SDMT, TONI, WCST, + Scores on SCWT, JLOT, and EPT only 23 male forensic psychiatric inpatients EPT significantly predicted frequency, but not + Overt aggression scale severity of aggression LaPierre et al36 30 psychopathic male criminals compared + Visual go/no-go, PMT qualitative error + Psychopathic subjects significantly with 30 non-psychopathic male criminals score, WCST, mental rotation task impaired on visual go/no-go, PMT errors and olfactory identification + Modular smell identification test + WCST and mental rotation no diVerent from controls

WCST=Wisconsin card sorting test; COWT=controlled oral word ﬂuency test; PMT=Porteus maze test; WISC-R=Wechsler intelligence scale for children-revised; WAIS-R=Wechsler adult intelligence scale-revised; FL=frontal lobe; EF=executive functions; SOP=self ordered pointing task; CAT=conditional associative learning task; SCWT=Stroop colour-word task; JLOT=judgment of line orientation test; SDMT=symbol digit modalities test; TONI=test of non-verbal intelligence; EPT=emotional perception test; ASPD=antisocial personality disorder; CD=conduct disorder.

Table 2 Neuroimaging studies in violent and aggressive subjects

Reference Subjects Methods Results

Raine et al51 21 community volunteers with ASPD, Prefrontal volume as measured using MRI + ASPD significant 11.0% reduction in compared with 26 substance dependent brain scans prefrontal gray matter compared with subjects, 21 psychiatric controls, and 34 controls; 13.9% reduction compared with healthy controls substance dependent group Woermann et al52 24 aggressive patients with TLE, compared Voxel by voxel analysis of grey matter density + Aggressive TLE patients had decreased left with 24 non-aggressive TLE patients and 35 using MRI brain scans frontal grey matter compared with controls non-aggressive TLE and controls Kuruoglu et al53 40 alcohol-dependent subjects (15 with Resting SPECT + ASPD subjects had significant anterior ASPD), compared with 10 age and sex frontal hypoperfusion compared with other matched controls alcohol dependent subjects and controls Amen et al54 40 aggressive psychiatric patients compared Resting SPECT + Aggressive subjects showed decreased with 40 non-aggressive psychiatric controls activity in prefrontal cortex, increased activity in left subcortical structures Hirono et al55 10 aggressive dementia patients compared Resting SPECT + Aggressive subjects had significant left with 10 non-aggressive dementia patients anterior temporal and bilateral dorsofrontal hypoperfusion + No diVerences in orbitofrontal regions Volkow and Four forensic psychiatric patients with Resting PET (15O-water and 18FDG) + Significantly decreased left temporal CBF Tancredi56 repetitive violence compared with four and metabolism in four patients normal controls + Significant frontal decreases in two subjects with “no remorse” Volkow et al57 Eight psychiatric patients with repetitive Resting 18FDG PET + Seven of eight violent patients, one control violence compared with eight normal controls subject, showed multiple areas with significantly decreased metabolism + Violent patients showed significantly greater reduction in bilateral prefrontal and medial temporal regions Goyer et al58 17 subjects with DSM-III personality + Activated 18FDG PET + Increased MAS correlated with decreased disorder (PD), 43 controls + Self reported “impulsive aggression” on orbitofrontal metabolism in PD subjects modified aggression scale (MAS) + No diVerences in CPT performance + CPT to assess prefrontal function Raine et al59 Attorney referrals of 41 persons charged with 18FDG PET with frontal activation by CPT + “Murderers” showed significant bilateral murder or manslaughter, matched controls metabolic decreases in prefrontal cortex, and left subcortical structures + No diVerences in CPT performance Pietrini et al62 15 young healthy volunteers selected for 15O-water PET superimposed on averaged + Evoked aggressive imagery correlated with visual imagery abilities brain MR scans significant decreases in ventromedial frontal CBF

PET=Positron emission tomography; SPECT=single photon emission computed tomography; 15O water=15Oxygen water; 18FDG=18Flouorodeoxyglucose; CBF=cerebral blood flow; CPT=continuous performance task; TLE=temporal lobe epilepsy. authors reported that frontal executive dys- traits; and reduced overall prefrontal grey mat- function was significantly associated with a his- ter volume. An MRI study of interictal aggres- tory of community violence, but did not sion in temporal lobe epilepsy found that, com- predict inpatient assaults.50 pared with both normal controls and non- These studies indicate that clinical signs of aggressive patients with temporal lobe epilepsy, frontal lobe dysfunction are prevalent in popu- patients with temporal lobe epilepsy with lations of persons prone to violent and antiso- recurrent episodic aggression had statistically cial behaviour. Most of the subjects, however, significant frontal grey matter reductions; the were either referred by attorneys, or had known area of maximum diVerence involved the left http://jnnp.bmj.com/ or suspected neuropsychiatric disorders, and so anterior frontolateral cortex.52 do not represent violent criminals in general. Three studies have used single photon emis- The mere presence of EEG abnormalities or sion computed tomography (SPECT) brain frontal neurological signs also does not explain scanning to evaluate antisocial and aggressive whether, or how, such findings contributed to behaviour.53–55 A comparison of alcohol de- behaviour at the time of an alleged crime. Two pendent subjects with healthy controls found parallel prospective studies do not support a that alcoholic subjects with an antisocial retrospective report that frontal lobe findings personality disorder had significantly greater on October 2, 2021 by guest. Protected copyright. predict violent behaviour in inpatient settings. frontal hypoperfusion than other alcoholic Clinical evidence of frontal lobe dysfunction, subjects.53 As a group, adolescent and adult however, does seem to be associated with psychiatric patients who had physically at- recurrent or persistent aggression.46–50 tacked another person or destroyed property showed significantly decreased prefrontal activity compared with matched, non-aggressive Neuroimaging in aggressive and violent psychiatric patient controls.54 Compared with subjects patients with non-aggressive dementia, patients Morphometric and functional neuroimaging with aggressive dementia with the same degree studies of aggressive and violent subjects have of cognitive and psychiatric impairments had consistently found frontal lobe abnormalities significant left anterior temporal and bilateral (table 2).51–62 A well designed MRI brain volu- superior frontal hypoperfusion, but no signifi- metric study compared an antisocial personal- cant diVerences in orbitofrontal regions.55 ity disorder group with substance dependent, Two positron emission tomography (PET) psychiatric and normal control groups.51 The brain scan studies, which compared forensic subjects with an antisocial personality disorder psychiatric patients with normal controls, showed significant diVerences on three meas- documented decreased frontal cortical blood ures: more violent crimes, more psychopathic flow or metabolism associated with “repetitive”

www.jnnp.com J Neurol Neurosurg Psychiatry: first published as 10.1136/jnnp.71.6.720 on 1 December 2001. Downloaded from 724 Brower, Price

and “purposeless” violent behaviour.56 57 An- aVective aggression, as opposed to premedi- other PET study of “impulsive aggression” tated or predatory behaviour, supports this found that, compared with non-psychiatric interpretation. controls, patients with personality disorders (chiefly antisocial, borderline, and narcissistic) Discussion showed decreased anterior medial and left The studies surveyed in this review indicate anterior orbitofrontal metabolism, which cor- that clinically significant frontal lobe dysfunc- related with increased scores on a self reported tion is associated with aggressive dyscontrol. aggression scale.58 Frontal cortex metabolism Subjects with both traumatic and neurodegen- did not distinguish patients with antisocial per- erative disorders primarily involving the pre- sonality disorder from controls. frontal cortex display increased rates of aggres- Another study examined 41 persons charged sive and antisocial behaviour compared with with murder or manslaughter, who were subjects who have no, or non-frontal brain injury. Studies employing neuropsychological referred for PET in connection with psychiatric testing, neurological examination, EEG, and evaluations for criminal responsibility, compe- neuroimaging have also tended to find evi- tence to stand trial, or claims of mitigation.59 dence for increased rates of prefrontal network Compared with controls (matched for age, sex, dysfunction among aggressive and antisocial and diagnosis of schizophrenia, if present), subjects. Prefrontal network dysfunction seems “murderers” as a group showed statistically to be most specifically associated with a recur- significant bilateral prefrontal metabolic de- rent, impulsive subtype of aggression that may creases during a frontal lobe activation task. A contribute to some violent behaviour. Two follow up report on the same subjects found prospective studies suggest that in populations that only those subjects blindly rated as lacking at risk for antisocial or aggressive behaviour, histories of psychosocial deprivation had sig- performance on neuropsychological tests of nificantly lower overall prefrontal metabolic executive function may have value in assessing rates.60 A further study separated these same future likelihood of aggression.33 35 No study, subjects into “predatory” versus “aVective” however, shows that disorders of prefrontal murderers, based on a forensic typology distin- cortex predict violent crime. guishing controlled, purposeful aggression to Methodological problems in this literature achieve a desired goal from impulsive, emo- include a lack of prospective data, small subject tionally charged aggression.61 AVective mur- numbers and lack of adequate controls for derers had significantly lower prefrontal meta- known violence risk factors. Study samples bolic activity compared with controls, whereas often draw from groups (prisoners, attorney frontal metabolism in predatory murderers referrals, or those with severe neurological or resembled controls. In a PET study of healthy psychiatric illness) that do not mirror the gen- volunteer subjects who were instructed to eral population or even the larger criminal imagine a scenario involving their own aggres- population. Reports describing persons sive behaviour, visual evocation of unrestrained charged with violent crimes tend to cite gross aggression correlated with significant focal measures of brain function with low specificity reductions in ventromedial frontal blood flow, and questionable clinical significance, while compared with an emotionally neutral sce- failing suYciently to relate the clinical data to 62 the specific aggressive behaviours in question.

nario. http://jnnp.bmj.com/ The cumulative evidence from these neu- Standard neuropsychological tests of executive roimaging studies points to a strong association function typically employed in studies of between increased aggression and reduced antisocial subjects also may not detect orbito- prefrontal cortical size or activity. Although frontal or ventromedial dysfunction relevant to most studies cite bilateral prefrontal abnor- aggressive behaviour. Although the bulk of malities, others specifically cite left anterior research on violent and criminal behaviour points to multiple, probably interacting, causal frontal or orbitofrontal findings, as well as non- factors, few studies attributing violent crime to frontal brain regions. These inconsistencies on October 2, 2021 by guest. Protected copyright. frontal lobe dysfunction adequately address may reflect variation related to experimental concurrent psychosocial variables such as conditions, limitations of imaging technology, emotional stress, drug and alcohol misuse, or subject selection. Most of the subjects in physical and sexual abuse, family breakdown, these studies had known or suspected psychiat- and poverty.4 ric disorders potentially contributing to altera- Studies of subjects with acquired frontal lobe tions in prefrontal function. Studies using PET injury support the expected association of have documented focal decreases in frontal increased aggression with focal orbitofrontal, cortical activity associated with various neu- or ventromedial frontal injury, or both. The ropsychiatric disorders, as well as transient neuropsychological literature, however, tends mental states, such as induced sadness, and to find increased aggressive behaviour associ- episodes of mood disorder.63–65 The reported ated with deficits in executive function, which reductions in prefrontal size or activity may, correlate with dorsolateral prefrontal dysfunc- therefore, represent a predisposition to aVec- tion.31 32 One hypothesis to account for discrep- tive states relevant to aggressive behaviour, ant localisation data is that orbitofrontal and without necessarily signifying an incapacity to dorsolateral prefrontal dysfunction contribute avoid actual violent acts.66 The trend in to aggressive dyscontrol in diVerent ways. Dor- neuroimaging findings, which associates pre- solateral dysfunction may predominate in per- frontal abnormalities with “purposeless” or sons with comorbid features of fetal or birth

related brain injury, developmental learning We acknowledge Drs Shervert Frazier, Paul S Appelbaum, and Kenneth L Appelbaum, for their encouragement and support. disorders, attention deficit hyperactivity disor- We also thank Dr James Ellison for his helpful review of the der, substance misuse, and antisocial personal- manuscript. ity disorder.28 67 Elliot characterised this group 1 Lewis DO, Pincus JH, Feldman M, et al. Psychiatric, neuro- as having episodic aggressive dyscontrol rooted logical, and psychoeducational characteristics of 15 death in “developmental deviance” manifested by row inmates in the United States. Am J Psychiatry 1986;143:838–45. “attention deficit disorder and minimal brain 2 Lewis DO, Pincus JH, Bard B, et al. Neuropsychiatric, psy- dysfunction,” and associated with neurological choeducational, and family characteristics of 14 juveniles 68 condemned to death in the United States. Am J Psychiatry soft signs and executive function deficits. 1988;145:584–9. 3 Martell DA. Estimating the prevalence of organic brain dys- Resulting educational and social failure likely function in maximum-security forensic psychiatric pa- contribute to aggressive and antisocial life tients. J Forensic Sci 1992;37:878–93. 4 Filley CM, Price BH, Nell V, et al. Toward an understanding adaptation, as well as to associated poor neuro- of violence: neurobehavioral aspects of unwarranted physi- psychological test performance. Executive cal aggression. Neuropsychiatry Neuropsychol Behav Neurol 2001;14:1–4. function deficits, therefore, may increase the 5 Bear D, Freeman R, Greenberg M. Alterations in personal- risk of violence via direct eVects on impulse ity associated with neurologic illnesses. Psychiatry 1985;1: 1–13. control or through associated psychosocial 6 Pincus JH. Aggression, criminality, and the frontal lobes. In: eVects, or both, either interactively or inde- Miller BL, Cummings JL, eds, The human frontal lobes: functions and disorders. New York: The Guildford Press, 1999. pendently. 7 Brower MC, Price BH. Epilepsy and violence: when is the Persons who have clinically evident neu- brain to blame? Epilepsy Behavior 2000;1:145–9. 8 Blumer D, Benson DF. Personality changes with frontal and ropsychiatric disorders involving focal injury to temporal lobe lesions. In: Benson DF, Blumer D, eds. Psy- chiatric aspects of neurological disease. New York: Grune and structural-functional components of the fron- Stratton, 1975. tal network, particularly the orbital and ventro- 9DuVy JD, Campbell JJ III. The regional prefrontal syndromes: a theoretical and clinical overview. J Neuropsy- medial prefrontal cortex, comprise a diVerent chiatry Clin Neurosci 1994;6:379–87. group. Retrospective data strongly support a 10 Mesulam, MM. Frontal cortex and behavior. Ann Neurol 1986;19:320–5. link between the disinhibited type of frontal 11 Hare RD, Harpur TJ, Hakstian AR, et al. The revised network syndrome and aggressive dyscontrol. psychopathy checklist: reliability and factor structure. Psy- chol Assess 1990;2:338–41. Case descriptions suggest that focal orbitofron- 12 Bodholdt RH, Richards HR, Gacono CB. Assessing tal injury specifically impairs capacities for psychopathy in adults: the psychopathy checklist-revised and screening version. In: Gacono CB, ed. The clinical and social judgment, risk avoidance, and empathy forensic assessment of psychopathy: a practicioner’s guide. that inhibit inappropriate or reflexive aggres- London: Lawrence Erlbaum Associates, 2000. 13 Hare RD, McPherson LM. Violent and aggressive behavior sion. The actual frequency of violent behav- by criminal psychopaths. Special issue: empirical ap- iour, however, seems relatively low. Based on proaches to law and psychiatry. Int J Law Psychiatry 1984; 7:35–50. results reviewed here,24 48 a reasonable conjec- 14 Tranel D. Acquired sociopathy: the development of sociopathic behavior following focal brain damage. In: ture for the increased risk of violence associ- Fowles DC, P Sutker P, SH Goodman SH, eds. Progress in ated with clinically significant focal frontal lobe experimental personality and psychopathology research. Vol 17. New York: Springer, 1994. injury might be 10% over the base rate for a 15 Meyers CA, Berman SA, Scheibel RS, et al. Case report: given population. Confirmation of this esti- acquired antisocial personality disorder with unilateral left orbital frontal lobe damage. J Psychiatr Neurosci 1992;17: mate must await prospective studies. 121–5. In addition to using prospective design, 16 Blair RJR, Cipolotti L. Impaired social response reversal: a case of acquired sociopathy. Brain 2000;123:1122–41. future studies testing the relation between 17 Bechara A, Tranel D, Damasio H. Characterization of the frontal lobe dysfunction and aggression should decision-making deficit of patients with ventromedial prefrontal cortex lesions. Brain 2000;123:2189–202.

incorporate controls for known risk factors 18 Price BH, DaVner KR, Stowe RM, et al. The comportmen- http://jnnp.bmj.com/ contributing to violent behaviour. Clinical tal learning disabilities of early frontal lobe damage. Brain 1990;113:1383–93. description of the nature and extent of frontal 19 Eslinger PJ, Grattan LM, Damasio H, et al. Developmental lobe impairments, coupled with attention to consequences of childhood frontal lobe damage. Arch Neu- rol 1992;49:764–9. the type (premeditated versus impulsive), 20 Anderson SW, Bechara A, Damasio H, et al. Impairment of social and moral behavior related to early damage in frequency, and severity of aggressive behaviour, human prefrontal cortex. Nat Neurosci 1999;2:1032–7. should help to clarify the brain-behaviour rela- 21 Labbatte LA, Warden D, Murray GB. Salutary change after frontal brain trauma. Ann Clin Psychiatry 1997;9:27–30. tions involved. Accurate measurement of the 22 Lishman WA. Brain damage in relation to psychiatric

increased risk of violence in subjects with pre- disability after head injury. Br J Psychiatry 1968;114:373– on October 2, 2021 by guest. Protected copyright. 410. frontal dysfunction also requires comparison 23 Virkkunen M, Nuutila A, Huusko S. EVect of brain injury with rates of aggression in appropriate controls. on social adaptability. Acta Psychiatr Scand 1976;53:168– 72. The neuropsychiatric evaluation of violent 24 Grafman J, Schwab K, Warden D, et al. Frontal lobe patients should include clinical assessment for injuries, violence and aggression: a report of the Vietnam head injury study. Neurology 1996;46:1231–8. frontal lobe impairment and neuropsychologi- 25 Stip E. Compulsive disorder and acquired antisocial behavior in frontal lobe dementia. J Neuropsychiatry Clin Neurosci cal evaluation of executive functions, particu- 1995;7:116. larly in cases involving recurrent, impulsive 26 Miller BL, Darby A, Benson DF, et al. Aggressive, socially disruptive and antisocial behavior associated with fronto- aggression. Further progress in the study of temporal dementia. Br J Psychiatry 1997;170:150–5. aggression and frontal lobe dysfunction will 27 Kandel E, Freed D. Frontal-lobe dysfunction and antisocial behavior: a review. J Clin Psychol 1989;45:404–413. require a forensically informed, interdiscipli- 28 Pennington BF, OzonoV S. Executive functions and nary approach that integrates neuropsychiat- developmental psychopathology. J Child Psychol Psychiatry 1996;37:51–87. ric, neuropsychological, and psychophysiologi- 29 Deckel AW, Hesselbrock V, Bauer L. Antisocial personality cal methods for the study of brain localisation, disorder, childhood delinquency, and frontal brain functioning: EEG and neuropsychological ﬁndings. J Clin social cognition, and emotional processing. Psychol 1996;52:639–50. Better understanding of brain injury and 30 Rosse RB, Miller MW, Deutsch SI. Violent antisocial behavior and Wisconsin card sorting test performance in aggression can then inform medical, public cocaine addicts [letter]. Am J Psychol 1993;150:170. 31 Giancola PR, Zeichner A. Neuropsychological performance health, and social policy interventions to on tests of frontal-lobe functioning and aggressive behavior 69 prevent violence. in men. J Abnorm Psychol 1994;103:832–5.

www.jnnp.com J Neurol Neurosurg Psychiatry: first published as 10.1136/jnnp.71.6.720 on 1 December 2001. Downloaded from 726 Brower, Price

32 Lau MA, Pihl RO, Peterson JB. Provocation, acute alcohol 51 Raine A, Lencz T, Bihrle S, et al. Reduced prefrontal gray intoxication, cognitive performance, and aggression. J matter volume and reduced autonomic activity in antisocial Abnorm Psychol 1995;104:150–5. personality disorder. Arch Gen Psychiatry 2000;57:119–27. 33 Giancola PR, Moss HB, Martin CS, et al. Executive cogni- 52 Woerman FG, van Elst LT, Koepp MJ, et al. Reduction of tive functioning predicts reactive aggression in boys at high frontal neocortical grey matter associated with aVective risk for substance abuse: a prospective study. Alcohol Clin aggression in patients with temporal lobe epilepsy: an Exp Res 1996;20:740–4. objective voxel by voxel analysis of automatically seg- 34 Giancola PR, Mezzich AC, Tarter RE. Executive cognitive mented MRI. J Neurol Neurosurg Psychiatry 2000;68:162– functioning, temperament, and antisocial behavior in 9. conduct-disordered adolescent females. J Abnorm Psychol 53 Kuroglu AC, Arikan Z, Vural G, et al. Single photon 1998;107:629–41. emission computerized tomography in chronic alcoholism. 35 Foster HG, Hillbrand M, Silverstein M. Neuropsychologi- Br J Psychiatry 1996;169:348–54. cal deficit and aggressive behavior: a prospective study. Prog 54 Amen DG, Stubblefield M, Carmichael B, . Brain 1993; :939–46. et al Neuropsychopharmacol Biol Psychiatry 17 SPECT findings and aggressiveness. Ann Clin Psychiatry 36 LaPierre D, Braun CMJ, Hodgins S. Ventral frontal deficits 1996: ;129–37. in psychopathy: neuropsychological test findings. 8 Neuropsy- 55 Hirono N, Mega M, Dinov I, . Left frontotemporal chologia 1995;131:139–51. et al 37 Pontius AA, Ruttiger KF. Frontal lobe system maturational hypoperfusion is associated with aggression in patients with lag in juvenile delinquents shown in the narratives test. dementia. Arch Neurol 2000;57:861–6. Adolescence 1976;11:509–18. 56 Volkow ND, Tancredi L. Neural substrates of violent 38 Lueger RJ, Gill KJ. Frontal-lobe cognitive dysfunction in behavior: a preliminary study with positron emission tom- conduct disorder adolescents. J Clin Psychol 1990;46:696– ography. B J Psychiatry 1987;151:668–73. 706. 57 Volkow ND, Tancredi LR, Grant C, et al. Brain glucose 39 Mednik SA, Volavka J, Gabrielli WF, et al. EEG as a predic- metabolism in violent psychiatric patients: a preliminary tor of antisocial behavior. Criminology 1982;19:219–31. study. Psychiatry Res 1995;61:243–53. 40 Bauer LO, O’Connor S, Hesselbrock VM. Frontal P300 58 Goyer PF, Andreason PJ, Semple WE, et al. Positron- decrements in antisocial personality disorder. Alcohol: Clin emission tomography and personality disorders. Neuropsy- Exp Res 1994;18:1300–5. chopharmacology 1994;10:21–8. 41 O’Connor S, Bauer L, Tasman A, et al. Reduced P3 ampli- 59 Raine A, Buchsbaum M, LaCasse L. Brain abnormalities in tudes are associated with both a family history of murderers indicated by positron emission tomography. Biol alcoholism and antisocial personality disorder. Prog Neu- Psychiatry 1997;42:495–508. ropsychopharmacol Biol Psychiatry 1994;18:1307–21. 60 Raine A, Stoddard J, Bihrle S, et al. Prefrontal glucose defi- 42 Finn PR, Ramsey SE, Earleywine M. Frontal EEG response cits in murderers lacking psychosocial deprivation. Neu- to threat, aggressive traits and a family history of ropsychiatry Neuropsychol Behav Neurol 1998;11:1–7. alcoholism: a preliminary study. J Stud Alcohol 2000;61:38– 61 Raine A, Meloy JR, Bihrle S, et al. Reduced prefrontal and 45. increased subcortical brain functioning assessed using 43 Fishbein DH, Herning RI, Pickworth WB, et al. EEG and positron emission tomography in predatory and aVective brainstem auditory evoked response potentials in adult murderers. Behav Sci Law 1998;16:319–32. male drug abusers with self-reported histories of aggressive 62 Pietrini P, Guazzelli M, Basso G, et al. Neural correlates of behavior. Biol Psychiatry 1989;26:595–611. imaginal aggressive behavior assessed by positron emission 44 Gedye A. Episodic rage and aggression attributed to frontal tomography in healthy subjects. Am J Psychiatry 2000;157: lobe seizures. Journal of Mental Deficiency Research 1772–81. 1989;33:369–79. 63 Volkow ND, Fowler JS. Neuropsychiatric disorders: investi- 45 Fornazzari L, Farcnik K, Smith I, et al. Violent visual hallucinations and aggression in frontal lobe dysfunction: gation of schizophrenia and substance abuse. Semin Nucl 1992; :254–67. clinical manifestations of deep orbitofrontal foci. J Med 22 Neuropsychiatry Clin Neursci 1992;4:42–4. 64 Mayberg H. Limbic-cortical dysregulation: a proposed 46 Williams D. Neural factors related to habitual aggression: model of depression. In: Salloway S, Malloy P, Cummings consideration of diVerences between those habitual aggres- JL, eds The neuropsychiatry of limbic and subcortical disorders. sives and others who have committed crimes of violence. Washington, DC: American Psychiatric Press, 1997. Brain 1969;92:503–20. 65 Blumberg HP, Stern E, Ricketts S, et al. Rostral and orbital 47 Blake PY, Pincus JH, Buckner C. Neurologic abnormalities prefrontal cortex dysfunction in the manic state of bipolar in murderers. Neurology 1995;45:1641–7. disorder. Am J Psychiatry 1999;156:1986–8. 48 Heinrichs RW.Frontal cerebral lesions and violent incidents 66 Mayberg H. Medical-legal inferences from functional in chronic neuropsychiatric patients. Biol Psychiatry 1989; neuroimaging evidence. Semin Clin Neuropsychiatry 1996; 25:174–8. 1:195–201. 49 Krakowski M, Czobor P. Violence in psychiatric patients: 67 Kandel E. Biology, violence and antisocial personality. J the role of psychosis, frontal lobe impairment, and ward Forensic Sci 1992;37:912–18. turmoil. Compr Psychiatry 1997;38:230–6. 68 Elliot FA. Neurology of aggression and episodic dyscontrol. 50 Krakowski M, Czobor P, Carpenter MD, et al. Community Semin Neurol 1990;10:303–12. violence and inpatient assaults: neurobiological deficits. J 69 Filley CM, Kelly JP, Price BH. Violence and the brain: an Neuropsychiatry Clin Neurosci 1997;9:549–55. urgent need for research. Scientist 2001;15:39. http://jnnp.bmj.com/ on October 2, 2021 by guest. Protected copyright.