of punishment information (perhaps as a function of noradrenergic disturbance). I will argue

0010-0277/$ - see front matter Published by Elsevier B.V.

* Tel.: +1 301 496 5198; fax: +1 301 594 9959.E-mail address: blairj@intra.nimh.nih.gov.

www.elsevier.com/locate/COGNIT

Cognition 101 (2006) 414442that the ability of the amygdala to form the stimuluspunishment associations necessary forsuccessful socialization is disrupted and that because of this, individuals with psychopathydo not learn to avoid actions that will harm others. It is noted that this model follows the neu-ropsychological approach to the study of developmental disorders, an approach that has beenrecently criticized. I will argue that these criticisms are less applicable to psychopathy. Indeed,animal work on the development of the neural systems necessary for emotion, does not sup-port a constructivist approach with respect to aect. Importantly, such work indicates thatwhile environmental eects can alter the responsiveness of the basic neural architecture medi-ating emotion, environmental eects do not construct this architecture. However, caveats tothe neuropsychological approach with reference to this disorder are noted.Published by Elsevier B.V.

Keywords: Psychopathy; Reward and punishment; Development; AmygdalaThe emergence of psychopathy: Implicationsfor the neuropsychological approach

to developmental disorders

R.J.R. Blair *

Mood and Anxiety Disorders Program, National Institute of Mental Health, USA

National Institutes of Health, Department of Health and Human Services, USA

Abstract

In this paper, I am going to examine the disorder of psychopathy and consider how geneticanomalies could give rise to the relatively specic neuro-cognitive impairments seen in individ-uals with this disorder. I will argue that genetic anomalies in psychopathy reduce the saliencedoi:10.1016/j.cognition.2006.04.005

R.J.R. Blair / Cognition 101 (2006) 414442 4151. Introduction

The intellectual roots of psychopathy can be traced to Pritchard (1835); see(Pichot, 1978). Pritchard developed the concept of moral insanity to accountfor socially damaging or irresponsible behavior that was not associated with knownforms of mental disorder. He attributed morally objectionable behavior to be a con-sequence of a diseased moral faculty. Psychiatric and legal classications have fol-lowed this tradition; Antisocial Personality Disorder is a classication within theAmerican Psychiatric Associations DSM-IV.

Unfortunately, the psychiatric diagnoses of Conduct Disorder (CD) and Antiso-cial Personality Disorder (APD: American Psychiatric Association, 1994) are seri-ously awed. These diagnoses of CD and APD focus on the presence of antisocialbehavior and tend to identify a highly heterogeneous sample. Indeed, DSM-IVacknowledges this heterogeneity by specifying two forms of CD: childhood- andadolescent- onset types. In childhood-onset type, onset of at least one criterion char-acteristic of CD must have occurred prior to 10 years of age. In adolescent-onsettype there should be not be any criteria characteristic of CD prior to 10 years ofage. Because of their lack of precision, the diagnostic rate of CD can reach 16%of boys in mainstream education (American Psychiatric Association, 1994) whilethe diagnostic rate of APD can reach over 80% in adult forensic institutions (Hart& Hare, 1996). Unsurprisingly, therefore, diagnoses of CD and APD are relativelyuninformative regarding an individuals prognosis.

In sharp contrast, the classication of psychopathy, introduced by Hare (1980,1991), is highly informative. The classication is a very useful predictor of a patientsfuture behavior (Hare, 1991) and is at the basis of many recidivism risk assessments.The classication of psychopathy involves aective-interpersonal (e.g., such as lackof empathy and guilt) and behavioral components (e.g., criminal activity and poorbehavioral controls); (Frick, OBrien, Wootton, & McBurnett, 1994; Hare, 1980,1991; Harpur, Hare, & Hakstian, 1989). It is a developmental disorder (Harpur &Hare, 1994). In childhood and adolescence, psychopathic tendencies are identiedprincipally by either the use of the Antisocial Process Screening Device (Frick &Hare, 2001) or by the Psychopathy Checklist: Youth Version (Forth, Kosson, &Hare, 2003; Kosson, Cyterski, Steuerwald, Neumann, & Walker-Matthews, 2002).In adulthood, psychopathy is identied though use of the Psychopathy Checklist Revised (Hare, 1991).

A remarkable feature of the behavioral prole of individuals with psychopathy istheir excessive displays of instrumental aggression (Cornell et al., 1996; Williamson,Hare, & Wong, 1987). The term instrumental aggression (also referred to as proac-tive aggression) is used to distinguish aggression that is purposeful and goal directed.The aggression is used instrumentally to achieve a specic desired goal such asobtaining the victims possessions or to increase status within a group hierarchy(Berkowitz, 1993). Bullying is an example of instrumental aggression and, unsurpris-ingly, individuals who engage in bullying behaviors, frequently engage in other formsof instrumental antisocial behavior in other contexts (Roland & Idsoe, 2001). Incontrast, reactive aggression (also referred to as aective aggression) occurs when

416 R.J.R. Blair / Cognition 101 (2006) 414442a frustrating or threatening event triggers the aggressive act and frequently alsoinduces anger (Barratt, Stanford, Dowdy, Liebman, & Kent, 1999; Barratt, Stan-ford, Kent, & Felthous, 1997; Berkowitz, 1993; Crick & Dodge, 1996; Linnoilaet al., 1983). Importantly, the aggression is initiated without regard for any potentialgoal. Elevated levels of reactive aggression are found in many disorders; e.g., psy-chopathy, depression, anxiety, bipolar disorder and post traumatic stress disorder(see, for a review, Blair, 2003a). However, only individuals with psychopathy showelevated levels of instrumental aggression. Instrumental aggression is, of course, aform of instrumental behavior. Whether an individual displays a particular instru-mental behavior is a product of their developmental learning experiences and deci-sion-making regarding the current context. The suggestion that will be made hereis that the impairment seen in individuals with psychopathy disrupts their capacityfor normal socialization such that antisocial behavior is more likely to be displayed(Blair, 1995; Lykken, 1957; Trasler, 1973).

The disorder of psychopathy should be of interest to developmental cognitiveneuroscientists for at least two main reasons. First, psychopathy allows us to under-stand the cognitive mechanisms necessary for the development of morality, an issuethat is of clear intrinsic interest. Second, and more importantly, psychopathy repre-sents a developmental disorder which aects a relatively specic set of neuro-cogni-tive systems. The pathology seen in individuals with psychopathy does not lead toglobal decits in the development of a variety of neuro-cognitive systems. Instead,the eects of the pathology inuence a relatively specic series of systems. In thispaper, I will consider why this is. However, rst I will describe the functional impair-ment seen in individuals with psychopathy.

1.1. What is the functional impairment shown by individuals with psychopathy?

Psychopathy is associated with a core set of empirically demonstrated behavioraldiculties. There are clear diculties with emotional processing (Blair, 1995;Blair, Jones, Clark, & Smith, 1997; Cleckley, 1976; Eysenck, 1964; Fowles, 1988;Gray,1987; Lykken, 1995; Mealey, 1995; Patrick, 1994; Pichot, 1978; Trasler, 1978, 1973)and some indications of diculty with some forms of attentional processing (Jutai &Hare, 1983; Raine & Venables, 1988). I will concentrate on the diculties withemotional processing as these have the clearest developmental implications.

There is a considerable body of literature revealing that individuals with psychop-athy are impaired in the processing of fear-related stimuli. Thus, individuals withpsychopathy show impairment in: (1) aversive conditioning (Flor, Birbaumer, Her-mann, Ziegler, & Patrick, 2002; Lykken, 1957); (2) generating autonomic responsesto anticipated threat (Hare, 1982; Oglo & Wong, 1990); (3) the augmentation of thestartle reex to visual threat primes (Herpertz et al., 2001; Levenston, Patrick, Brad-ley, & Lang, 2000); (4) passive avoidance learning (Lykken, 1957; Newman & Kos-son, 1986); and (5) response reversal (Mitchell, Colledge, Leonard, & Blair, 2002;Newman, Patterson, & Kosson, 1987). In addition, individuals with psychopathyshow clear diculties in empathic responding. Thus, individuals with psychopathyshow impaired processing of the sadness and fear of others (Aniskiewicz, 1979; Blair,

R.J.R. Blair / Cognition 101 (2006) 414442 4171999; Blair et al., 1997; House & Milligan, 1976). In addition, individuals with psy-chopathy show impaired recognition of/autonomic responding to sad and fearfulfacial and vocal expressions (Aniskiewicz, 1979; Blair et al., 2002; Blair, Colledge,& Mitchell, 2001; Blair, Colledge, Murray, & Mitchell, 2001; Blair, Monson, & Fred-erickson, 2001; Blair et al., 1997).

The fear-related data have led to repeated suggestions that individuals with psy-chopathy show impairment in the neuro-physiological systems modulating fearbehavior (Cleckley, 1976; Eysenck, 1964; Fowles, 1988; Gray, 1987; Lykken, 1995;Mealey, 1995; Patrick, 1994; Pichot, 1978; Trasler, 1978, 1973). Developmentally,these positions all assume that moral socialization is achieved through the use ofpunishment (Eysenck & Gudjonsson, 1989; Trasler, 1978). In essence, they assumethat the healthy individual is frightened by punishment and associates this fear withthe action that resulted in the punishment thus making the individual less likely toengage in the action in the future. They suggest that individuals with psychopathy,because they are less aversively aroused by punishment, make weaker associationsand thus are more likely to engage in the punished action in the future than healthyindividuals. However, several researchers have questioned the assumption that con-ditioned fear responses play a crucial role in moral socialization (Blackburn, 1988;Blair & Morton, 1995). Thus, the developmental literature indicates that moralsocialization is not achieved through the formation of conditioned fear responsesbut rather through the induction and fostering of empathy (Homan, 1984). Studieshave shown, for example, that moral socialization is better achieved through the useof induction (reasoning that draws childrens attention to the eects of their misde-meanors on others and increases empathy) than through harsh authoritarian orpower assertive parenting practices which rely on the use of punishment (Baumrind,1971; Baumrind, 1983; Homan & Saltzstein, 1967). Indeed, several researchers havesuggested that while empathy may facilitate moral socialization, fear may actuallyobstruct it (Brody & Shaer, 1982; Homan, 1994).

On a related note, if healthy individuals learn to avoid antisocial behavior becauseof fear of punishment, it must be assumed that the healthy child judge all rules/trans-gressions in a similar way. In other words, if we learn to avoid talking in class andhitting other individuals because we are punished when we commit these actions,there is no reason for us to distinguish between these two transgressions. However,healthy developing children make a distinction between moral (victim-based) andconventional (social order based) transgressions from the age of 36 months (Sme-tana, 1981, 1985, 1993). In other words, children do not judge all transgressions inan identical fashion. Instead, they dierentiate between those transgressions thatresult in harm to another from those that simply cause social disorder.

The importance of empathy for moral socialization was one of the reasons for thedevelopment of the Violence Inhibition Mechanism model of psychopathy and mor-al development (Blair, 1995; Blair et al., 1997). At its simplest, I argued that the Vio-lence Inhibition Mechanism (VIM) was a system that when activated by distress cues(the sad and fearful expressions of others) results in increased autonomic activity,attention and activation of the brainstem threat response system (usually resultingin freezing); (Blair, 1995); see Fig. 1. I considered the VIM to be activated whenever

418 R.J.R. Blair / Cognition 101 (2006) 414442distress cues are displayed rather than being reliant upon contextual informationabout ongoing violence for activation. In line with the model, the display of distresscues has been found to result in the inhibition of not only aggression (Perry & Perry,1974) but also non-violent disputes over property ownership (Camras, 1977) and sex-ual activity (Chaplin, Rice, & Harris, 1995).

The main focus of the model was to describe the cognitive prerequisites for moraldevelopment. According to the model, moral socialization occurs through thepairing of the activation of the mechanism by distress cues with representations ofthe act that caused the distress cues (Blair, 1995). Through association theserepresentations of moral transgressions become triggers for the mechanism. Theappropriately developing child thus initially nds the pain of others aversive andthen, through socialization, the thoughts of acts that cause pain to others aversive.I proposed that individuals with psychopathy have had disruption to this systemsuch that representations of acts that cause harm to others do not become triggersfor the VIM (Blair, 1995).

One early index of appropriate moral socialization, and thus the developmentalintegrity of the VIM, is the demonstration by the child of the moral/conventional

Fig. 1. The Violence Inhibition Mechanism model. Key to Fig. 1: arrows indicate information ow. Thebroken arrow represents that transgression stimuli will only come to activate the VIM after learning hasoccurred; pairings of the transgression representations with distress cues. Boxes indicate putativemechanisms.

R.J.R. Blair / Cognition 101 (2006) 414442 419distinction mentioned above. Children with psychopathic tendencies and adults withpsychopathy show impairment in distinguishing moral and conventional transgres-sions (Blair, 1995, 1997; Blair, Jones, Clark, & Smith, 1995; Blair, Sellers, et al.,1995; Blair, Colledge et al., 2001; Blair, Colledge, Murray et al., 2001; Blair, Monsonet al., 2001); see, for related work with children with Behavior Disorder and CD(Arsenio & Fleiss, 1996; Nucci & Herman, 1982). In addition, and in line with theVIM position, psychopathic adults show reduced comprehension of situations likelyto induce guilt although they show appropriate comprehension of happiness, sadnessand even complex social emotions such as embarrassment (Blair, Jones et al., 1995;Blair, Sellers, et al., 1995). Moreover, and a direct prediction of the model, childrenand adults with psychopathy show pronounced impairment in processing sad andfearful facial and vocal expressions (Aniskiewicz, 1979; Blair et al., 2002; Blair,1999; Blair, Colledge et al., 2001; Blair, Colledge, Murray et al., 2001; Blair, Monsonet al., 2001; Blair et al., 1997; House & Milligan, 1976).

However, while the original VIM model could provide an account of the emer-gence of instrumental antisocial behavior in individuals with psychopathy andwhile it did generate a variety of predictions that have been empirically con-rmed, it faced a serious diculty; it could not account for the data associatedwith the fear hypotheses. Moreover, it could not account for data on the interac-tion of temperament and socialization practice on the development of moraldevelopment/ conscience. Kochanska has stressed the role of fearfulness as theimportant temperamental factor (Kochanska, 1993, 1997). Indeed, she and othershave found fearful children to show higher levels of moral development/con-science using a variety of measures (Asendorpf & Nunner-Winkler, 1992;Kochanska, 1997; Kochanska, De Vet, Goldman, Murray, & Putman, 1994;Rothbart, Ahadi, & Hershey, 1994). In addition, Kochanska has stressed that dif-ferent socialization practices may promote moral development in children withdierent temperaments (Kochanska, 1993, 1997). In line with this, she found thatfor fearful children, maternal gentle discipline promoted moral/conscience devel-opment. In contrast, for fearless children, alternative socialization practices,presumably capitalizing on mother-child positive orientation (secure attachment,maternal responsiveness), promoted the development of conscience (Kochanska,1997). Because of these diculties for the VIM model, I recently expanded itat both the cognitive and neural levels (Blair, 2003a, 2003b, 2003c, 2004; Blair& Charney, 2003); see below.

1.1.1. SummaryIndividuals with psychopathy are profoundly impaired in the processing of fear-

related stimuli and in the processing of the fearful and sad expressions of others. TheViolence Inhibition Mechanism (VIM) model provided an account of the impair-ment in the processing of sad and fearful expressions in individuals with psychopathyand the implications of this impairment for moral development in these individuals.However, the VIM model cannot be considered a complete model of psychopathy;notably, it could not account for the impairments in the processing of fear-relatedstimuli in individuals with psychopathy.

420 R.J.R. Blair / Cognition 101 (2006) 414442In the following sections of this paper, I will consider a series of issues. First, I willconsider whether there is a genetic basis to this disorder. Second, I will consider theneural and neuro-chemical systems on which I believe this genetic contribution maybe operating (the systems which mediate the basic response to threat). Two optionswill be considered: (a) the genetic anomalies disrupt the functioning of the amygdala;and (b) the genetic anomalies disrupt the functioning of specic neurotransmitter(s)which are involved in specic aspects of amygdala functioning, in particular codingpunishment information. Third, I will consider other neural systems that may be dys-functional in psychopathy, in particular, the extent of orbital frontal cortex dysfunc-tion in psychopathy. Within these second and third issues, I will develop a neuro-cognitive account of psychopathy, the Integrated Emotion Systems model. Fourth,I will consider the dissimilarities between psychopathy and another disorder linkedto impairment in social cognition, autism. Fifth, I will consider psychopathy withrespect to the neuropsychological approach to developmental disorders. I will pointout the considerable benet the study of psychopathy has gained from the neuro-psychological approach. However, I will also point out the limits of the neuro-psy-chological approach; specically, the pathology associated with psychopathy is notequivalent to that seen subsequent to an amygdala lesion.

1.2. What is the ultimate cause of psychopathy?

Growing evidence suggests a genetic contribution to psychopathy. Early twin,adoption, and family studies indicated that antisocial behavior was heritable (Rhee& Waldman, 2002). However, such studies are dicult to interpret. Much antisocialbehavior is goal directed: the individual mugs the victim for their wallet, the individ-ual steals the bag to obtain its contents, the individual engages in an elaborate stingoperation to gain another persons money. It is extremely unlikely that there is adirect genetic contribution to these behaviors or at least it is as likely as there is adirect genetic contribution to the use of a light switch to enable the navigation ofa dark room. An individual might learn to use a light switch and under particularconditions, alternatively he/she might learn to mug people for their wallets. Howev-er, where genetics are likely to play a role is in determining the probability that theindividual will learn an antisocial strategy to gain money (mugging other people) asopposed to a strategy sanctioned by society (using an ATM machine at the end ofthe workday). Many individuals have argued that the emotional dysfunction shownby individuals with psychopathy makes them more likely to learn antisocial strate-gies to reach goals (Blair, 1995; Eysenck, 1964; Lykken, 1995; Trasler, 1973). Theargument developed here is that there may be a genetic contribution to the emotionaldysfunction and that it is this which results in an apparent genetic contribution toantisocial behavior. Recent data suggests that this is indeed the case.

Blonigen, Carlson, Krueger, and Patrick (2003) collected data from 353 adultmale twins using the self-report Psychopathic Personality Inventory [PPI] (Lilienfeld& Andrews, 1996). The PPI includes 163 items and forms a global index of psychop-athy with eight sub-scales. Most of these subscales showed moderate heritability(h2 = 0.290.56) and negligible shared environmental inuence (Blonigen et al.,

R.J.R. Blair / Cognition 101 (2006) 414442 4212003). Moreover, in a considerably larger study, examining almost 3500 twin pairswithin the Twins Early Development Study (TEDS), the callous and unemotionalcomponent of psychopathic tendencies was indexed at age 7 (Viding, Blair, Mott,& Plomin, 2005). This study revealed a signicant group heritability of h2g :67 andno shared environmental inuence on the callous-unemotional component; i.e.,genetic factors account for two thirds of the dierence between the callous-unemo-tional probands and the population.

As yet, the roles of environmental inuences on the emergence of psychopathyremain unclear. The genetics studies above suggested relatively little inuence. How-ever, there is a considerable literature indicating a relationship between socioeco-nomic status (SES) and antisocial behavior (Raine, 1993). With respect topsychopathy, it would be surprising if social variables did not impact on the proba-bility of antisocial behavior; SES, for example, is likely to constrain the possibility ofalternative behavioral choices to antisocial behavior as well as increase the salienceof the money contained in a potential victims wallet. Indeed, in line with this, a rela-tionship between SES and the antisocial behavior component of psychopathy hasbeen reported (Hare, 2003). Thus, genetics would determine the level of emotionaldysfunction while the environment would inuence how this genetically determinedemotion dysfunction was expressed.

1.2.1. SummaryIt is likely that there is a genetic contribution to psychopathy. I argue that this

contributes to the emotional dysfunction that is the basis of the disorder. The emo-tional dysfunction, in turn, puts the individuals for learning antisocial behaviors tosolve goals. Social environmental variables are likely to have an inuence, mostimportantly constraining the possibility of alternative behavioral choices to antiso-cial behavior as well as potentially increasing the salience of the rewards the mightaccrue through antisocial behavior. In the following section, I will consider the neu-ral and neuro-chemical architecture upon which this genetic contribution may beoperating.

1.3. The basic threat response

Three basic responses to threat are freezing, escape behaviors and reactive aggres-sion (a rage attack at the threat); (Blanchard, Blanchard, & Takahashi, 1977). Whichbehavioral response is initiated is determined by the threats intensity and proximity.At low levels of intensity, from a distant threat, the animal will freeze. At higher lev-els, from a closer threat, the animal will attempt to escape the environment. At high-er levels still, when the threat is very close and escape is impossible, the animal willdisplay reactive aggression (Blanchard et al., 1977).

These responses to threat are mediated by a neural circuit that is common tomany mammalian species (Gregg & Siegel, 2001; Panksepp, 1998). It runs frommedial amygdaloidal areas downward, largely via the stria terminalis to the medialhypothalamus, and from there to the dorsal half of the periaqueductal gray (PAG).With respect to the basic response to threat, these systems function as a unitary

422 R.J.R. Blair / Cognition 101 (2006) 414442entity. Stimulation of any of these systems may induce the basic threat responses.Which one is elicited is determined by the level of stimulation.

An important neuro-chemical response to threat involves the noradrenergic sys-tem (Charney, 2003; Francis & Meaney, 1999). When specic neurons in the centralnucleus of the amygdala are activated by threat, they activate, in turn, the locus coe-ruleus, leading to an increase in noradrenaline release. Rogers and others haveargued that noradrenaline inuences the salience of aversive cues (Rogers, Lancas-ter, Wakeley, & Bhagwager, 2004); greater noradrenaline levels should allow fasterlearning about information associated with the aversive cues.

With regards to psychopathy, I suggest that the genetic anomalies potentiallypresent in individuals with psychopathy (Blonigen et al., 2003; Viding et al., 2005),aect particularly a component of this circuit; the amygdala. I argue that this leadsto an individual who is less responsive to aversive stimuli. I will consider here twopossibilities regarding how this might occur. First, the genetic anomalies disruptthe functioning of the amygdala (Blair, 2001, 2002; Blair, Morris, Frith, Perrett, &Dolan, 1999; Patrick, 1994). Second, the genetic anomalies disrupt the functioningof specic neurotransmitter(s) which are involved in specic aspects of amygdalafunctioning, in particular coding punishment information (Blair, Leonard, Morton,& Blair, 2006a; Blair et al., 2006b).

1.3.1. SummaryThe mammalian basic response to threat is mediated by a circuit that runs from

the amygdala to the hypothalamus and from there to the peri-aqueductal gray. Animportant neuro-chemical response to threat involves the noradrenergic system. Inthe following sections we will consider whether the genetic contribution to psychop-athy disrupts the functioning of the amygdala or, perhaps, disrupts the functioningof specic neurotransmitter(s) which are involved in specic aspects of amygdalafunctioning.

1.4. Psychopathy and amygdala dysfunction

The suggestion that psychopathy might be due to early amygdala dysfunction hasgenerated considerable research. In line with the suggestion, individuals with psy-chopathy show reduced amygdaloid volume relative to comparison individuals(Tiihonen et al., 2000) and reduced amygdala activation during emotional memory(Kiehl et al., 2001) and aversive conditioning tasks (Veit et al., 2002). Moreover,the positions origins in the neuropsychological approach to developmental disorders(Baron-Cohen, 1998; Frith, 2002; Leslie, 1992) has allowed the generation of a vari-ety of predictions regarding the functional impairment. Thus, human and animalneuropsychological work has informed us that the eects of amygdala lesions includeimpairment in: (1) aversive conditioning (Bechara, Damasio, Damasio, & Lee, 1999;Bechara et al., 1995; LaBar, Gatenby, Gore, LeDoux, & Phelps, 1998); (2) theaugmentation of the startle reex to visual threat primes (Angrilli et al., 1996;Funayama, Grillon, Davis, & Phelps, 2001); (3) passive avoidance learning(Ambrogi Lorenzini, Baldi, Bucherelli, Sacchetti, & Tassoni, 1999); and (4)

R.J.R. Blair / Cognition 101 (2006) 414442 423expression recognition that is particularly marked for fearful expressions (Adolphs,2002; Blair, 2003a). If psychopathy is associated with amygdala dysfunction, the neu-ropsychological approach would predict that individuals with psychopathy areimpaired in the above tasks. Considerable data shows that they are (Blair, Colledgeet al., 2001; Blair, Colledge, Murray et al., 2001; Blair, Monson et al., 2001; Floret al., 2002; Herpertz et al., 2001; Levenston et al., 2000; Lykken, 1957; Newman &Kosson, 1986).

Not only has the amygdala dysfunction position generated a host of predictions ithas also allowed the reconciliation of fear (Lykken, 1995; Patrick, 1994) and empa-thy (Blair, 1995) dysfunction positions. Many of the ndings associated with bothpositions are direct results of amygdala dysfunction (see above). Moreover, itbecomes easier to understand ndings indicating the importance of fearfulnessas an important temperamental factor in socialization (Kochanska, 1993, 1997).The temperamental variable fearfulness can be understood as an index of theintegrity of the amygdala (Blair, 2003a). While fear conditioning is not consideredto be important in socialization (Brody & Shaer, 1982; Homan, 1994), I argue thatthe amygdalas response to the fear and sadness of victims, during empathy induc-tion, is crucial for socialization. Murray and colleagues argue that the amygdala iscrucially involved in the formation of stimulusreward and stimuluspunishmentassociations (Baxter & Murray, 2002). I argue that the associations required forappropriate socialization are associations between moral transgressions and victimsdistress. Empathy induction and other positive parenting techniques should aid theformation of associations between the moral transgression and the victims distress(empathy induction focuses the childs attention on the aversive stimulus that is thevictims distress). A considerable literature shows that these parenting techniques doreduce antisocial behavior in children (Brody & Shaer, 1982). Individuals with psy-chopathy are less likely to form these associations and so should be more likely toengage in actions that will harm others. Indeed, interestingly, while positive parent-ing techniques are associated with reduced antisocial behavior levels in healthy chil-dren, they have no impact on the level of antisocial behavior expressed by childrenwho present with the emotional dysfunction associated with psychopathy (Wootton,Frick, Shelton, & Silverthorn, 1997).

Despite the success of the amygdala dysfunction position, two caveats should beconsidered here. First, very few patients with amygdala lesions acquired in adult-hood present with instrumental aggression [though this may not be the case if thelesion occurs early in life (Fine, Lumsden, & Blair, 2001)]. Of course, this shouldbe expected. My argument is that the amygdala is required for the formation of mor-al transgressionvictims distress associations. However, typically, goal directedbehavior does not require the accessing of these associations. Instead, these moraltransgressionvictims distress associations resulted in the individual failing to learnantisocial behaviors to achieve their goals. The behaviors the individual does use toachieve their goals will be a function of their learning history. While the individualwith psychopathy might learn that a solution to a lack of money was to mug another(rewarded by the money but not punished by the victims distress), the individualwith an acquired amygdala lesion would have learnt, before their lesion, that this

424 R.J.R. Blair / Cognition 101 (2006) 414442was not appropriate (such actions were punished by the victims distress). When con-sidering how to rectify a lack of money, the individual with an acquired amygdalalesion would activate previously successful motor programs for achieving this goal(a visit to an ATM).

The second caveat is slightly more problematic. As noted above, the amygdala iscrucially involved in the formation of stimulusreward and stimuluspunishmentassociations; animals with amygdala lesions show impairment in both reward andpunishment related behavior (Baxter & Murray, 2002). Yet, the impairment in indi-viduals with psychopathy is far more marked for processing dependent on stimuluspunishment associations than for stimulusreward associations (Levenston et al.,2000; Patrick, 1994; Blair et al., 2006a, 2006b). Thus, whereas individuals with psy-chopathy do not show augmentation of the startle reex following a negative visualprime relative to comparison individuals, they do show a comparable reduction instartle reex following a positive visual prime relative to comparison individuals(Levenston et al., 2000; Patrick, 1994). Moreover, in an aective priming study, Pesc-hardt et al. found that whereas individuals with psychopathy showed no congruencefacilitation for negative target words, they did show congruence facilitation for posi-tive target words (Blair et al., 2006a). In other words, while a positive prime primedthe output response associated with positive valence in individuals with psychopathy(albeit less so than in comparison individuals), a negative prime did not prime theoutput response associated with negative valence. Finally, in a decision makingstudy, individuals with psychopathy showed particular diculty, relative to controls,when choosing between stimuli associated with dierent levels of punishment. Theirimpairment in choosing between stimuli associated with dierent levels of rewardwas far less marked (Blair et al., 2006a).

In addition to the above, the amygdala has also been considered to play a role incertain aspects of social cognition, at least in humans and possibly primates; in par-ticular, aect-related judgments based on facial stimuli (Adolphs, 2003; Baron-Co-hen et al., 2000). Thus, in one paradigm, participants were shown pictures ofindividuals in natural poses and asked to judge the trustworthiness of these individ-uals. While healthy individuals typically judge some individuals faces less trustwor-thy than others, patients with amygdala lesions present with atypical judgmentpatterns (Adolphs, Tranel, & Damasio, 1998). Moreover, recent neuro-imaging workhas indicated that healthy individuals show greater amygdala activation to facesjudged to be untrustworthy relative to faces judged to be trustworthy (Winston,Strange, ODoherty, & Dolan, 2002). In a second paradigm, participants must judgethe complex social emotion being displayed by an individual based on informationonly from the eye region (Baron-Cohen, Wheelwright, & Jolie, 1997). Individualswith amygdala lesions show impairment on this task (Adolphs, Baron-Cohen, &Tranel, 2002; Stone, Baron-Cohen, Calder, Keane, & Young, 2003) while neuro-im-aging work has indicated the involvement of the amygdala whilst performing thistask (Baron-Cohen et al., 1999). However, despite this apparent role of the amygdalain these two aspects of social cognition, individuals with psychopathy are notimpaired either in making trustworthiness judgments (Richell et al., 2005) or injudging complex social emotions from the eyes (Richell et al., 2003).

R.J.R. Blair / Cognition 101 (2006) 414442 4251.4.1. Summary

Individuals with psychopathy are impaired on many tasks that require thefunctional integrity of the amygdala. Both the impairments in fear-related andempathic responding found in individuals with psychopathy can be attributedto amygdala dysfunction. In short, predictions developed from human and animalneuropsychological work with respect to amygdala dysfunction in individuals withpsychopathy have been useful with respect to the impairment in emotional learningand basic emotional responding seen in individuals with psychopathy. However,individuals with psychopathy are not equivalent to patients with amygdala lesions;functions that appear to require the amygdala such as the formation of stimulusreward associations and certain aspects of social cognition are instead only mildly,or not at all, impaired in individuals with psychopathy. This suggests that the geneticanomalies present in individuals with psychopathy do not lead to the development ofthe disorder through a global disruption of the functioning of the amygdala. Instead,they may have a more selective eect, disrupting the functioning of specicneurotransmitter(s) which are involved in specic aspects of amygdala functioning.This will be considered in the following section.

1.5. Psychopathy, neurotransmitter dysfunction and developing the neuro-cognitive

model

Polymorphisms of particular genes can alter the functioning of specic neuro-transmitter systems (Lichter et al., 1993; Shih, Chen, & Ridd, 1999; Vandenberghet al., 1992). Currently, it remains unclear which neurotransmitter systems mightbe dysfunctional in individuals with psychopathy. However, one possibility is thatthe noradrenergic response to stress/threat stimuli, described above, is disturbed inindividuals with psychopathy (Blair, 2003a; Blair et al., 2006b). Interestingly, therehave been recent suggestions that noradrenaline is involved in mediating the impactof aversive cues in human choice (Rogers et al., 2004). Moreover, recent pharmaco-logical data imply that noradrenergic manipulations selectively impact on the pro-cessing of sad expressions (Harmer, Perrett, Cowen, & Goodwin, 2001; Sustrik,Coupland, & Blair, in preparation). Further support for this suggestion comes fromstudies linking NA abnormalities to antisocial behavior/Conduct Disorder (Raine,1993; Rogeness, Cepeda, Macedo, Fischer, & Harris, 1990; Rogeness, Javors, Mass,& Macedo, 1990). In this regard it is interesting to note that NA function appears tobe increased in a range of anxiety disorders (Charney, Heninger, & Breier, 1984); i.e.,it is increased in populations that present with a heightened responsiveness to aver-sive cues, the opposite of the emotional impairment seen in psychopathy. I thereforeargue that the genetic anomalies considered to be present in individuals with psy-chopathy disrupt the functioning of the noradrenergic system such that the impactof aversive stimuli is muted.

The integrated emotion systems (IES) model represents a neuro-cognitive modelof the interactions of the systems involved in emotional processing (Blair, 2004). Ini-tial components of this model, those most related to the earlier VIM model, aredepicted in Fig. 2. Three systems are depicted. One allows sensory representations;

426 R.J.R. Blair / Cognition 101 (2006) 414442e.g., the representation of objects currently in the environment, including moraltransgressions. At the neural level, this system is primarily implemented by temporalcortex. The second contains valence representations; i.e., representations that canbe activated by unconditioned stimuli (e.g., loud noises but also distress cues). At theneural level, this system is implemented by the amygdala. The third will be discussedbelow.

The valence representations can represent both appetitive and aversive stimulifollowing claims that the amygdala is crucially involved in the formation of stimu-lusreward and stimuluspunishment associations (Baxter & Murray, 2002). Thestrength of the connections between representations in the dierent systems increasethrough Hebbian learning (Hebb, 1949); if two representations in dierent systemsare simultaneously active the connection between these two representations willstrengthen. In short, activation of a valence representation by an observed distresscue will increase the connection between this valence representation and currentlyactive sensory representations (perhaps a representation of the moral transgressionthat has caused this distress cue). This computation was also the basis of the VIMmodel (see above and Fig. 1). The connections between the representations in the dif-ferent systems are reciprocal, reecting the interconnections of the amygdala with

Fig. 2. The basics of the Integrated Emotion Systems model. Key to Fig. 2: the ovals indicaterepresentational systems. The lled in circles indicate representations within these systems. The linesindicate connections between the systems while the arrows denote the direction of information ow.CS = Conditioned stimulus; US = Unconditioned stimulus; S1 and S2 are competing sensory represen-tations; R1 and R2 are alternative motor responses.

R.J.R. Blair / Cognition 101 (2006) 414442 427cortical regions (Amaral, Price, Pitkanen, & Carmichael, 1992). These reciprocalconnections allow the valence representations to bias the processing of emotionalsensory representations (an emotional sensory representation will activate thevalence representations, leading to greater activation of the emotional sensory repre-sentation). In short, activation of the valence representations can increase attentionto the sensory representation which activated them.

If we assume that the genetic anomalies considered to be present in individualswith psychopathy disrupt the functioning of the noradrenergic system such that neg-ative valence representations are less activated by aversive stimuli, we can predictmany of the behavioral phenomena associated with psychopathy. Aversive condi-tioning will be impaired, and passive avoidance errors will be more common, becausesensory representations will be more weakly associated with negative valence (Floret al., 2002; Lykken, 1957; Newman & Kosson, 1986). Reduced activation of nega-tive valence representations by sad and, particularly, fearful expressions will result inreduced autonomic responses to these expressions and may impair the naming ofthese expressions (Aniskiewicz, 1979; Blair, 1999; Blair et al., 1997; House & Milli-gan, 1976; Stevens, Charman, & Blair, 2001).

The third system depicted in Fig. 2 mediates the production of motor responses(and, at the neural level, is implemented by basal ganglia and motor cortex). Thevalence representations are not involved in the formation of stimulusresponse asso-ciations (the linking of sensory representations to motor responses). At the neurallevel, it is known that the amygdala is not necessary for the formation of stimu-lusresponse associations (Baxter & Murray, 2002). Instrumental learning tasks reli-ant on stimulusresponse associations, such as object discrimination, are notimpaired by amygdala lesions while instrumental learning tasks that are reliant onstimulusreinforcement associations (within the IES, the linking of sensory represen-tations with valence representations), such as passive avoidance are (Baxter & Mur-ray, 2002). Interestingly, and in line with the suggestion of disruption of the valencerepresentations, individuals with psychopathy are not impaired on object discrimina-tion tasks (Mitchell et al., 2002) but are impaired on passive avoidance learning(Newman & Kosson, 1986).

1.5.1. Summary

In this section, I have suggested that the genetic contribution to psychopathy mayinuence the functioning of the noradrenergic system such that the formation ofstimulusreinforcement, particularly stimuluspunishment, associations is impaired.I have also begun to develop the neuro-cognitive Integrated Emotion Systems model.

1.6. Psychopathy and orbital frontal cortex

Orbital frontal cortex is considerably involved in the regulation and mediation ofemotional behavior. Patients with orbital frontal cortex lesions may present withemotional dys-regulation and antisocial behavior (Anderson, Bechara, Damasio,Tranel, & Damasio, 1999; Blair & Cipolotti, 2000; Blumer & Benson, 1975; Dama-sio, 1994; Pennington & Bennetto, 1993). This has led to considerable claims that

428 R.J.R. Blair / Cognition 101 (2006) 414442psychopathy might be due to early orbital frontal cortex damage and that patientswith orbital frontal cortex damage and psychopathy present similarly (Damasio,1994). Such claims have received questionable empirical support however.

Orbital frontal cortex does play a role in the regulation of the brainstem systemsinvolved that mediate the basic responses to threat (including reactive aggression);(Gregg & Siegel, 2001; Panksepp, 1998). Damage to orbital frontal cortex should thuslead to a dys-regulation of brainstem systems involved in mediating the basic respons-es to threat and potentially increase the risk of threat/ frustration-based reactiveaggression. Patients with lesions of orbital frontal cortex do indeed display a height-ened risk of reactive aggression (Anderson et al., 1999; Blair & Cipolotti, 2000; Pen-nington & Bennetto, 1993). However, patients with lesions of orbital frontal cortex donot display a heightened risk of goal-directed, instrumental aggression, even if theirlesions occur in early childhood (Anderson et al., 1999; Pennington & Bennetto,1993). Yet, individuals with psychopathy are particularly notable for their use ofinstrumental aggression (Cornell et al., 1996; Williamson et al., 1987). Moreover,empirically, while patients with orbital frontal cortex lesions display generallyimpaired expression recognition (Blair & Cipolotti, 2000; Hornak, Rolls, & Wade,1996), individuals with psychopathy do not show such general impairment but rathera relatively more selective impairment for fearful, and to a lesser extent, sad expres-sions (Blair, Colledge et al., 2001; Blair, Colledge, Murray et al., 2001; Blair, Monsonet al., 2001). Additionally, ndings such as the impairment in aversive conditioningshown by individuals with psychopathy (Flor et al., 2002; Lykken, 1957) are notfound in patients with orbital frontal cortex lesions (Bechara et al., 1999).

This is not to suggest that individuals with psychopathy are without impairmentfor functions in which orbital frontal cortex is involved. There is a considerable neu-ropsychological and neuro-imaging literature demonstrating that orbital frontal cor-tex is crucially involved in response reversal and extinction (Cools, Clark, Owen, &Robbins, 2002; Rahman, Sahakian, Hodges, Rogers, & Robbins, 1999; Rolls, Hor-nak, Wade, & McGrath, 1994). Response reversal and extinction involve changing aresponse to a stimulus as a function of a change in contingency; i.e., learning to with-hold a response that is now punished though previously it had been rewarded (Rolls,1997). The dierence between response reversal and extinction is that in responsereversal tasks, the participant learns to respond to a second stimulus rather thanthe rst while in extinction tasks the participant learns to withhold responding alltogether.

In Fig. 3, the depiction of the IES model is extended to include two systems inu-enced by input from the valence representations and therefore likely to be disruptedin individuals with psychopathy. The rst of these, the response selection system,represents expected level of reinforcement associated with the stimulus (and, poten-tially, the form of response to this stimulus), information that is provided by thevalence representations. If two or more stimuli are present in the environment, eachof which is associated with a dierent response, it is important that the system canrapidly select the stimulus that will elicit the greatest level of reward. The responseselection system allows this to occur (the computational details of this process aredescribed elsewhere, Blair, 2004). At the neural level, this system is implemented

R.J.R. Blair / Cognition 101 (2006) 414442 429by medial regions of orbital frontal cortex. The second, the response gating system, isparticularly activated when expected reinforcement information, provided by theresponse selection system, diers from actually received reinforcement. In Fig. 3, Ipropose that this system alters stimulusresponse associations (the connectionsbetween the sensory representations and the motor responses) to alter the probabilitythat that response to that stimulus will be made in the future (usually to reduce theprobability that that response will be made in the future). This system is crucial forresponse reversal. At the neural level, this system is implemented by ventrolateralfrontal cortex.

Adults with psychopathy are notably impaired on response reversal andextinction tasks (Mitchell et al., 2002; Newman et al., 1987). However, childrenwith psychopathic tendencies are only mildly impaired on these tasks. InNewmans card playing task (Newman et al., 1987), the participant has todecide whether to play a card. Initially, the participants choice to play is always

Fig. 3. An expansion of the Integrated Emotion Systems model. Key to Fig. 3: the ovals indicaterepresentational systems. The lled in circles indicate representations within these systems. The linesindicate connections between the systems while the arrows denote the direction of information ow. Thelines from the response gating system to the connections between the sensory representation and motorresponse systems, that terminating in small circles represent the gating by the response gating system ofresponses activated by these connections. CS = Conditioned stimulus; US = Unconditioned stimulus; S1and S2 are competing sensory representations; R1 and R2 are alternative motor responses. Thecomputational details of response selection are described elsewhere (Blair, 2004).

430 R.J.R. Blair / Cognition 101 (2006) 414442reinforcing; if the participant plays the card he/she will win points or money.However, as the participant progresses through the pack of cards, their proba-bility of reward decreases. The participant should terminate his/her respondingbefore he/she receives greater levels of punishment than reward. Children withpsychopathic tendencies, like adults with psychopathy, are impaired on this task(Fisher & Blair, 1998; Newman et al., 1987; OBrien & Frick, 1996). In theID-ED task (Dias, Robbins, & Roberts, 1996), the participant learns thatresponding to one of two stimuli gains reward while responding to the otheris punished. This contingency is then reversed; i.e., responding to the rst stim-ulus is no longer rewarded but punished while responding to the second is nowrewarded. While adults with psychopathy are impaired on this task, childrenwith psychopathic tendencies are not (Blair, Colledge et al., 2001; Blair,Colledge, Murray et al., 2001; Blair, Monson et al., 2001; Mitchell et al.,2002). A major dierence between these two tasks is in the salience of thecontingency change. In the card-playing task, the probability of reinforcementdecreases by 10% over every ten trials. In the ID-ED task, the probability ofreinforcement changes from 100% to 0% once the initial learning criterion hasbeen achieved. This indicates that while both children with psychopathic tenden-cies and adult individuals with psychopathy are impaired in the detection ofcontingency change, this impairment is markedly greater in the adult individualswith psychopathy. Moreover, this suggests that if we reduce the salience of thecontingency change, we should see impairment in the children with psychopathictendencies and that the degree of impairment will be a function of the salienceof the contingency change. This was tested using a probabilistic response rever-sal paradigm. Participants were presented with pairs of stimuli. The probabilityof reward was dierent across pairs (i.e., for pair 1, stimulus A was rewarded100% of the time, for pair 2, stimulus C was rewarded 90% of the time etc).Following a set number of trials the contingency was reversed (i.e., for pair1, stimulus B was rewarded 100% of the time, for pair 2, stimulus D wasrewarded 90% of the time). While the children with psychopathic tendenciesshowed no diculty reversing their responses for salient contingency changes[pair 1], they did show signicant diculty as the salience of the contingencychange decreased [pair 2] (Budhani & Blair, 2005).

This dierence between children and adults with psychopathy is of interest.Response reversal and extinction occur following the successful identication of vio-lations between expected reward and received punishment (or absence of reward).Failure to perform response reversal and extinction could be due to impairment insystems necessary for: (1) detecting the violation; (2) changing the stimulusrewardassociations; or (3) representing the punishment. We have argued that there isreduced punishment sensitivity in individuals with psychopathy. Children with psy-chopathic tendencies seem to be at least as insensitive to punishing stimuli as adultswith psychopathy. They are comparably impaired on tasks such as passive avoidancelearning (Newman & Kosson, 1986; Newman, Widom, & Nathan, 1985), the pro-cessing of fearful expressions (Blair, Colledge et al., 2001; Blair, Colledge, Murrayet al., 2001; Blair, Monson et al., 2001) and aversive conditioning (Lykken, 1957;

R.J.R. Blair / Cognition 101 (2006) 414442 431Raine, Venables, & Williams, 1996). However, the impairment of children with psy-chopathic tendencies for response reversal is far less marked. This suggests thatadults with psychopathy are impaired in systems necessary for detecting the viola-tion/changing the stimulusreward associations; i.e., orbital/ventrolateral frontalcortex dysfunction.

Given the evidence of selective amygdala dysfunction discussed above, thereare several possibilities regarding the origins of the orbital frontal cortexpathology found in adults with psychopathy. First, the orbital frontal cortexpathology could be developmentally independent of the amygdala pathology.For example, the genetic anomalies associated with psychopathy might aectthe development of the amygdala and orbital frontal cortex independently ofone another. Second, there are considerable interconnections between the amyg-dala and orbital frontal cortex (Amaral et al., 1992; Carmichael & Price, 1995).It is possible that a lack of aerent input from the amygdala to orbital frontalcortex could disrupt the development of orbital frontal cortex to an increasinglygreater degree as development progresses. Thirdly, individuals with psychopathypresent with higher levels of drug abuse, dependence, and poly-drug use thancomparison individuals (Hemphill, Hart, & Hare, 1994; Smith & Newman,1990). Alcohol and drug dependent individuals are impaired on measuresassessing the functioning or orbital frontal cortex (Bechara et al., 2001; Grant,Contoreggi, & London, 2000; Rogers & Robbins, 2001). It is thus also possiblethat the apparent developmental eect seen here is simply a consequence of thelifestyle chosen by psychopathic individuals.

1.6.1. Summary

Damasio and others have suggested that psychopathy might be due to orbitalfrontal cortex damage. However, recent data suggest that the decits expected fromorbital frontal cortex damage do not mirror those seen in individuals with psychop-athy. This is not to suggest that there is no orbital frontal cortex dysfunction in psy-chopathy. Children and adults with psychopathic tendencies show impairment inresponse reversal, a function known to recruit orbital frontal cortex. Interestingly,while the children and adults with psychopathic tendencies perform similarly withrespect to their lack of guilt for their actions and on measures of, or based on, stim-ulusreinforcement learning, the impairment in response reversal is far more markedin adults with psychopathy than children with psychopathic tendencies. This sug-gests a developmental progression to the orbital frontal cortex pathology found inpsychopathy. The reasons for this progression remain unclear.

1.7. Psychopathy and autism

Psychopathy is not the only disorder linked to impairments in social cognition.Autism is a severe developmental disorder where there is marked neuro-cognitiveimpairment. It is described by the American Psychiatric Associations Diagnosticand Statistical Manual (DSM-IV, American Psychiatric Association, 1994) as thepresence of markedly abnormal or impaired development in social interaction and

432 R.J.R. Blair / Cognition 101 (2006) 414442communication and a markedly restricted repertoire of activities and interests (p.66). Like psychopathy, autism has also been lined to impairments in amygdala func-tioning (Baron-Cohen et al., 2000).

Individuals with autism may show structural amygdala abnormalities. However,these have been reported as an increase in grey matter volume in the amygdala/peri-amygdaloid cortex (Abell et al., 1999) rather than the decrease seen in individualswith psychopathy (Tiihonen et al., 2000). Functionally, the situation is complex.

As discussed above, the amygdala is involved in the formation of stimuluspun-ishment and stimulusreward associations as well as some aect laden componentsof social cognition. My colleagues and I argue that the sensitivity of individuals withpsychopathy to punishment, and therefore the ability to form stimuluspunishmentassociations, is disproportionately impaired (Blair et al., 2006a). However, there isno reason to believe that this is the case in individuals with autism. Indeed, individ-uals with autism are at increased risk of presenting with anxiety (Gillott, Furniss, &Walter, 2001; Rumsey, Rapoport, & Sceery, 1985), rather than the decreased riskseen in individuals with psychopathy (Frick, Lilienfeld, Ellis, Loney, & Silverthorn,1999; Patrick, 1994; Verona, Patrick, & Joiner, 2001). In contrast, there may beimpairment in aspects of social cognition that involve the amygdale in individualswith autism. Thus, patients with autism, unlike individuals with psychopathy, areimpaired on both the Eyes task and the face trustworthiness judgment tasks dis-cussed earlier (Adolphs, Sears, & Piven, 2001; Baron-Cohen, Wheelwright, Hill, Ras-te, & Plumb, 2001). Of course, these impairments may not reect amygdaladysfunction per se. It is possible that they are developmental consequences of impov-erished face representations due to dysfunction in regions such as fusiform gyrus andsuperior temporal sulcus (Schultz et al., 2003).

1.7.1. Summary

Autism, like psychopathy, is associated with impairment in social cognition and,at the anatomical level, amygdala dysfunction. However, while psychopathy is asso-ciated with impairment in the formation of stimulusreinforcement associations, afunction that animal and human work, has clearly indicated amygdala involvementin, there is little reason to suspect such impairment in autism. In contrast, severalaspects of social cognition that have been related to amygdala functioning, thoughwhich may reect the functioning of nearby neural systems, are impaired in autismbut are not impaired in individuals with psychopathy.

1.8. Psychopathy and the neuropsychological approach to developmental disorders

The model that I have developed here follows a tradition of pre-supposing theexistence of relatively independent neuro-cognitive systems that may be selectivelyimpaired in developmental disorders (Baron-Cohen, 1998; Frith, 2002; Leslie,1992). With respect to the predictions of the position, I have taken care taken to con-sider neuropsychological ndings with adult lesions cases and, sometimes, to usetasks that have been previously used with these patients (Blair et al., 2002; Blair,Colledge et al., 2001; Blair, Colledge, Murray et al., 2001; Blair, Monson et al.,

R.J.R. Blair / Cognition 101 (2006) 414442 4332001; Mitchell et al., 2002). Indeed, it was this neuropsychological approach whichallowed the reconciliation of the fear-relevant and empathy-relevant data describedabove; most could be accounted for by assuming dysfunction in systems involved inemotional processing, in particular the amygdala. However, this approachhas been recently criticized and sometimes considered to be not developmental(Karmilo-Smith, Brown, Grice, & Paterson, 2003; Paterson, Brown, Gsodl,Johnson, & Karmilo-Smith, 1999). In short, Karmilo-Smith and colleagues arguethat the existence of separable neuro-cognitive systems in the adult does not implythat these separable systems are present in the young child. They suggest that theseparable systems seen in the adult may be constructed through development andmay not be discrete systems at birth.

I do not propose to debate these issues fully here. The systems at which theseattacks are made (systems for number and language processing) are usually corticallyrepresented as opposed to the interaction of cortical and sub-cortical systems consid-ered here. Whether or not this dichotomy is important, it is clear that such criticismsmust be tempered when considering psychopathy. Work with animals allows consid-erable specication of the emotional systems that are dysfunctional in individualswith psychopathy as well as their development. Thus, let us consider the basic neuralarchitecture for emotional processing; i.e., the amygdala, hypothalamus and peri-acqueductal gray. This neural architecture is present from birth, primarily geneticallydetermined, and highly similar to that found in other mammalian species. In short, itis not constructed from environmental experience. Of course, this does not mean thatthis architecture functions equivalently in the child and adult or that environmentaleects do not eect its functioning. Interestingly, animals during infancy do not dem-onstrate HPA axis responses to stress. However, infant animals exposed to stressorsdemonstrate increases in the expression of immediate early genes (e.g., c-fos andnerve growth factor inducible gene) in the paraventricular nucleus of the hypothal-amus (Charney, 2003). In short, early postnatal adverse experiences increase theresponsiveness of these stress response systems, eects which last throughout the life-time of the individual (Heim, Owens, Plotsky, & Nemero, 1997; Liu et al., 1997;Plotsky &Meaney, 1993). Thus, the environment has an eect, by partially determin-ing the responsiveness of these basic threat systems. However, it is important toagain note here that this neural architecture is not constructed out of environmentalexperience [in contrast to processes proposed to underlie the development of somecortical neuro-cognitive systems (Karmilo-Smith et al., 2003; Paterson et al.,1999)], it is present from birth.

But more generally, how appropriate is a neuropsychological approach when con-sidering psychopathy? In many respects, the empirical literature on psychopathydemonstrates the clear advantages of the neuropsychological approach. However,at the same time, it identies the need for caution. Following the neuropsychologicalapproach, and assuming amygdala dysfunction, has allowed the generation of anarray of predictions, many of which have been conrmed. The amygdala is involvedin the formation of stimuluspunishment associations (Baxter & Murray, 2002) andindividuals with psychopathy are impaired in a variety of tasks dependent on the for-mation of stimuluspunishment associations. However, the amygdala is involved in

434 R.J.R. Blair / Cognition 101 (2006) 414442the formation of stimulusreward associations and certain aect-relevant aspects ofsocial cognition. Individuals with psychopathy show far less impairment in theseaspects of amygdala functioning. Thus, the impairment in psychopathy is not equiv-alent in impact to the eects of an amygdala lesion to a healthy adult.

However, it should be noted that even results where data obtained from individ-uals with psychopathy diers from that of patients with amygdala lesions are infor-mative. Thus, the neuro-psychological and neuro-imaging literatures indicate a roleof the amygdala in specic aect-relevant aspects of social cognition. Faces judged tobe untrustworthy activate the amygdala (Winston et al., 2002) and patients withamygdala lesions are poor at judging a faces trustworthiness (Adolphs et al.,1998). The neuropsychological approach might therefore predict problems intrustworthiness judgments in individuals with psychopathy but this prediction isnot supported (Richell et al., 2003). This suggests either that the ability to maketrustworthiness judgments involves neuro-cognitive systems independent of thoseinvolved in stimuluspunishment association formation (taking a rened neuro-psychological approach) or that individuals with psychopathy are makingtrustworthiness judgments using a neuro-cognitive architecture that is very dierentfrom that used by healthy individuals. The predictions of these contrasting positionsare clear. The rst suggests that individuals with psychopathy will show appropriateamygdala responses to untrustworthy faces even if their amygdala responses duringaversive conditioning are reduced (Veit et al., 2002). The second suggests that theneural responses of individuals with psychopathy to untrustworthy faces will be verydierent from those of comparison individuals. The data to resolve these contrastingpredictions should soon be available.

1.8.1. Summary

The study of psychopathy has gained considerably from the neuro-psychologicalapproach. Predictions generated from following this approach have been systemati-cally conrmed allowing a relatively tight specication of the impairment at the basisof this disorder. However, the neuropsychological approach cannot be followed slav-ishly. Psychopathy is not the equivalent to an amygdala lesion. The decits in psy-chopathy are only a subset of those functions mediated by the amygdala.

1.9. Conclusions

The main goal of this paper was to consider how genetic anomalies could giverise to the relatively specic neuro-cognitive impairments seen in individuals withpsychopathy. With respect to this goal, the suggestion is that genetic anomaliesreduce the salience of punishment information (perhaps as a function of noradren-ergic disturbance). This impairs various aspects of amygdala function, most impor-tantly the ability to form stimuluspunishment associations. In order to achievesuccessful socialization, the child needs to form associations between representa-tions of moral transgressions (acts which harm others) and the aversive punish-ment caused by the victims distress. This allows the child to learn to avoidactions that will harm others. As individuals with psychopathy nd the distress

of psychopathy but may contribute to the full behavioral prole of the disorder.

R.J.R. Blair / Cognition 101 (2006) 414442 435Certainly, by analogy, some individuals with autism show executive dysfunctionin addition to their diculties with Theory of Mind. While the executive dysfunc-tion does not appear to be related to their social diculties, it does appear to berelated to other components of the behavioral prole such as their repetitivebehavior (Turner, 1997). Individuals with psychopathy may show certain attention-al abnormalities (Jutai & Hare, 1983; Raine & Venables, 1988). The role of thesediculties in the development of the emotional component of psychopathy remainsunclear. However, it is possible that they reect the more general inuences ofgenetically mediated neurotransmitter anomalies. Alternatively, of course, thegenetic anomalies may inuence the functioning of one particular neurotransmitterreceptor. Certain neurotransmitter receptors appear to be far more neural locationspecic than others. If this possibility is the case, the relatively selective nature ofthe decits seen in individuals with psychopathy may reect, for example, theimpact of genetic anomalies on the density of particular neurotransmitter receptorsin the amygdale crucial for punishment coding. At present, it is impossible to dis-tinguish between these speculations. However, future work will allow the situationto be resolved.

The model that has been developed here follows a tradition of pre-supposing theexistence of relatively independent neuro-cognitive systems that may be selectivelyimpaired in developmental disorders. This neuropsychological approach has beenrecently criticized and sometimes considered to be not developmental. However, Iwould argue, at least with respect to psychopathy, and particularly given animal dataon the development of the neural systems mediating emotion, that these criticismsare unfounded. I suggest that the neuropsychological approach will continue to bean invaluable tool with respect to the understanding, and ultimately the cure, of thispernicious disorder.

References

Abell, F., Krams, M., Ashburner, J., Passingham, R., Friston, K., Frackowiak, R., et al. (1999). Theneuroanatomy of autism: a voxel-based whole brain analysis of structural scans. Neuroreport, 10,16471651.

Adolphs, R. (2002). Neural systems for recognizing emotion. Current Opinion in Neurobiology, 12,169177.

Adolphs, R. (2003). Is the human amygdala specialized for processing social information?. Annals of theNew York Academy of Sciences 985, 326340.of the victim signicantly less aversive, they are less likely to learn to avoid actionsthat will harm others.

Why should the genetic eects operate so selectively, aecting primarily the sal-ience of punishment information? It is likely that if a neurotransmitter such as nor-adrenaline is implicated in psychopathy, it may serve more functions than simplythe coding of the salience of punishment information. This may mean that othersystems are aected in individuals with psychopathy. Dysfunction in these othersystems may have no causal role in the development of the core emotional aspects

436 R.J.R. Blair / Cognition 101 (2006) 414442Adolphs, R., Baron-Cohen, S., & Tranel, D. (2002). Impaired recognition of social emotions followingamygdala damage. Journal of Cognitive Neuroscience, 14, 12641274.

Adolphs, R., Sears, L., & Piven, J. (2001). Abnormal processing of social information from faces inautism. Journal of Cognitive Neuroscience, 13, 232240.

Adolphs, R., Tranel, D., & Damasio, A. R. (1998). The human amygdala in social judgment. Nature, 393,470474.

Amaral, D. G., Price, J. L., Pitkanen, A., & Carmichael, S. T. (1992). Anatomical organization of theprimate amygdaloid complex. In J. P. Aggleton (Ed.), The Amygdala: neurobiological aspects ofemotion, memory, and mental dysfunction (pp. 166). New York: Wiley.

Ambrogi Lorenzini, C. G., Baldi, E., Bucherelli, C., Sacchetti, B., & Tassoni, G. (1999). Neuraltopography and chronology of memory consolidation: a review of functional inactivation ndings.Neurobiology of Learning and Memory, 71, 118.

American Psychiatric Association. (1994). Diagnostic and Statistical Manual of Mental Disorders (4th ed.)Washington, DC: American Psychiatric Association.

Anderson, S. W., Bechara, A., Damasio, H., Tranel, D., & Damasio, A. R. (1999). Impairment of socialand moral behaviour related to early damage in human prefrontal cortex. Nature Neuroscience, 2,10321037.

Angrilli, A., Mauri, A., Palomba, D., Flor, H., Birhaumer, N., Sartori, G., et al. (1996). Startle reex andemotion modulation impairment after a right amygdala lesion. Brain, 119, 19912000.

Aniskiewicz, A. S. (1979). Autonomic components of vicarious conditioning and psychopathy. Journal ofClinical Psychology, 35, 6067.

Arsenio, W. F., & Fleiss, K. (1996). Typical and behaviourally disruptive childrens understanding of theemotion consequences of socio-moral events. British Journal of Developmental Psychology, 14,173186.

Asendorpf, J. B., & Nunner-Winkler, G. (1992). Childrens moral motive strength and temperamentalinhibition reduce their immoral behaviour in real moral conicts. Child Development, 63,12231235.

Baron-Cohen, S. (1998). Modularity in developmental cognitive neuropsychology: evidence from autismand Gilles de la Tourette syndrome. In J. A. Burack, R. M. Hodapp, & E. Zigler (Eds.), Handbook ofmental retardation and development (pp. 335). Cambridge, England: Cambridge University Press.

Baron-Cohen, S., Ring, H. A., Bullmore, E. T., Wheelwright, S., Ashwin, C., & Williams, S. C. (2000). Theamygdala theory of autism. Neuroscience Biobehavior Review, 24, 355364.

Baron-Cohen, S., Ring, H. A., Wheelwright, S., Bullmore, E. T., Brammer, M. J., Simmons, A., et al.(1999). Social intelligence in the normal and autistic brain: an fMRI study. European Journal ofNeuroscience, 11, 18911898.

Baron-Cohen, S., Wheelwright, S., Hill, J., Raste, Y., & Plumb, I. (2001). The Reading the Mind in theEyes Test revised version: a study with normal adults, and adults with Asperger syndrome or high-functioning autism. Journal of Child Psychology and Psychiatry, 42, 241251.

Baron-Cohen, S., Wheelwright, S., & Jolie, T. (1997). Is there a language of the eyes? Evidence fromnormal adults, and adults with autism or Asperger syndrome. Visual Cognition, 4, 311331.

Barratt, E. S., Stanford, M. S., Dowdy, L., Liebman, M. J., & Kent, T. A. (1999). Impulsive andpremeditated aggression: a factor analysis of self-reported acts. Psychiatry Research, 86, 163173.

Barratt, E. S., Stanford, M. S., Kent, T. A., & Felthous, A. (1997). Neuropsychological and cognitivepsychophysiological substrates of impulsive aggression. Biological Psychiatry, 41, 10451061.

Baumrind, D. (1971). Current patterns of parental authority. Developmental Psychology Monographs, 4,1103.

Baumrind, D. (1983). Rejoinder to Lewiss interpretation of parental rm control eects: are authoritativefamilies really harmonious?. Psychological Bulletin 94, 132142.

Baxter, M. G., & Murray, E. A. (2002). The amygdala and reward. Natural Reviews. Neuroscience, 3,563573.

Bechara, A., Damasio, H., Damasio, A. R., & Lee, G. P. (1999). Dierent contributions of the humanamygdala and ventromedial prefrontal cortex to decision-making. Journal of Neuroscience, 19,54735481.

R.J.R. Blair / Cognition 101 (2006) 414442 437Bechara, A., Dolan, S., Denburg, N., Hindes, A., Anderson, S. W., & Nathan, P. E. (2001). Decision-making decits, linked to a dysfunctional ventromedial prefrontal cortex, revealed in alcohol andstimulant abusers. Neuropsychologia, 39, 376389.

Bechara, A., Tranel, D., Damasio, H., Adolphs, R., Rockland, C., & Damasio, A. R. (1995). Doubledissociation of conditioning and declarative knowledge relative to the amygdala and hippocampus inhumans. Science, 269, 11151118.

Berkowitz, L. (1993). Aggression: its causes, consequences, and control. Philadelphia: Temple UniversityPress.

Blackburn, R. (1988). On moral judgements and personality disorders: the myth of psychopathicpersonality revisited. British Journal of Psychiatry, 153, 505512.

Blair, K. S., Leonard, A., Morton, J., & Blair, R. J. R. (2006a). Impaired decision making on the basis ofboth reward and punishment information in individuals with psychopathy. Personality and IndividualDierences, 41, 155165.

Blair, K. S., et al. (2006b, in press).Blair, R. J. (2003a). Facial expressions, their communicatory functions and neuro-cognitive substrates.

Philosophical Transactions of the Royal Society of London. Series B, Biological Sciences, 358, 561572.Blair, R. J., Colledge, E., & Mitchell, D. G. (2001). Somatic markers and response reversal: is there

orbitofrontal cortex dysfunction in boys with psychopathic tendencies?. Journal of Abnormal ChildPsychology 29, 499511.

Blair, R. J., Mitchell, D. G., Richell, R. A., Kelly, S., Leonard, A., Newman, C., et al. (2002). Turning adeaf ear to fear: impaired recognition of vocal aect in psychopathic individuals. Journal of AbnormalPsychology, 111, 682686.

Blair, R. J. R. (1995). A cognitive developmental approach to morality: investigating the psychopath.Cognition, 57, 129.

Blair, R. J. R. (1997). Moral reasoning in the child with psychopathic tendencies. Personality andIndividual Dierences, 22, 731739.

Blair, R. J. R. (1999). Responsiveness to distress cues in the child with psychopathic tendencies.Personality and Individual Dierences, 27, 135145.

Blair, R. J. R. (2001). Neuro-cognitive models of aggression, the antisocial personality disorders andpsychopathy. Journal of Neurology, Neurosurgery and Psychiatry, 71, 727731.

Blair, R. J. R. (2002). A neuro-cognitive model of the psychopathic individual. In T. Robbins & M. Ron(Eds.), Disorders of brain and mind II. Cambridge, UK: Cambridge University Press.

Blair, R. J. R. (2003b). Neurobiological basis of psychopathy. British Journal of Psychiatry, 182, 57.Blair, R. J. R. (2003c). A neurocognitive model of the psychopathic individual. In M. A. Ron & T. W.

Robbins (Eds.), Disorders of brain and mind 2 (pp. 400420). Cambridge: Cambridge University Press.Blair, R. J. R. (2004). The roles of orbital frontal cortex in the modulation of antisocial behavior. Brain

and Cognition, 55, 198208.Blair, R. J. R., & Charney, D. S. (2003). Emotion regulation: an aective neuroscience approach. In M. P.

Mattson (Ed.), Neurobiology of aggression: Understanding and preventing violence (pp. 2132). Totowa,New Jersey: Humana Press.

Blair, R. J. R., & Cipolotti, L. (2000). Impaired social response reversal: a case of acquired sociopathy.Brain, 123, 11221141.

Blair, R. J. R., Colledge, E., Murray, L., & Mitchell, D. G. (2001). A selective impairment in theprocessing of sad and fearful expressions in children with psychopathic tendencies. Journal ofAbnormal Child Psychology, 29, 491498.

Blair, R. J. R., Jones, L., Clark, F., & Smith, M. (1995). Is the psychopath morally insane?. Personalityand Individual Dierences 19, 741752.

Blair, R. J. R., Jones, L., Clark, F., & Smith, M. (1997). The psychopathic individual: a lack ofresponsiveness to distress cues?. Psychophysiology 34, 192198.

Blair, R. J. R., Monson, J., & Frederickson, N. (2001). Moral reasoning and conduct problems in childrenwith emotional and behavioural diculties. Personality and Individual Dierences, 31, 799811.

Blair, R. J. R., Morris, J. S., Frith, C. D., Perrett, D. I., & Dolan, R. (1999). Dissociable neural responsesto facial expressions of sadness and anger. Brain, 122, 883893.

438 R.J.R. Blair / Cognition 101 (2006) 414442Blair, R. J. R., & Morton, J. (1995). Putting cognition into sociopathy. Brain and Behavioral Science, 18,548.

Blair, R. J. R., Sellars, C., Strickland, I., Clark, F., Williams, A. O., Smith, M., et al. (1995). Emotionattributions in the psychopath. Personality and Individual Dierences, 19, 431437.

Blanchard, R. J., Blanchard, D. C., & Takahashi, L. K. (1977). Attack and defensive behaviour in thealbino rat. Animal Behavior, 25, 197224.

Blonigen, D. M., Carlson, R. F., Krueger, R. F., & Patrick, C. J. (2003). A twin study of self-reportedpsychopathic personality traits. Personality and Individual Dierences, 35, 179197.

Blumer, D., & Benson, D. F. (1975). Personality changes with frontal and temporal lobe lesions. In D. F.Benson & D. Blumer (Eds.), Psychiatric aspects of neurological disease (pp. 151170). New York:Grune & Stratton.

Brody, G. H., & Shaer, D. R. (1982). Contributions of parents and peers to childrens moralsocialisation. Developmental Review, 2, 3175.

Budhani, S., & Blair, R. J. (2005). Response reversal and children with psychopathic tendencies: success isa function of salience of contingency change. Journal of Child Psychology Psychiatry, 46, 972981.

Camras, L. A. (1977). Facial expressions used by children in a conict situation. Child Development, 48,14311435.

Carmichael, S. T., & Price, J. L. (1995). Sensory and premotor connections of the orbital and medialprefrontal cortex of macaque monkeys. Journal of Comparative Neurology, 363, 642664.

Chaplin, T. C., Rice, M. E., & Harris, G. T. (1995). Salient victim suering and the sexual responses ofchild molesters. Journal of Consulting and Clinical Psychology, 63, 249255.

Charney, D. S. (2003). Neuroanatomical circuits modulating fear and anxiety behaviors. Acta PsychiatricaScandinavica Supplement, 417, 3850.

Charney, D. S., Heninger, G. R., & Breier, A. (1984). Noradrenergic function in panic anxiety. Eects ofyohimbine in healthy subjects and patients with agoraphobia and panic disorder. Archives of GeneralPsychiatry, 41, 751763.

Cleckley, H. M. (1976). The mask of sanity. St. Louis, MO: Mosby.Cools, R., Clark, L., Owen, A. M., & Robbins, T. W. (2002). Dening the neural mechanisms of

probabilistic reversal learning using event-related functional magnetic resonance imaging. Journal ofNeuroscience, 22, 45634567.

Cornell, D. G., Warren, J., Hawk, G., Staord, E., Oram, G., & Pine, D. (1996). Psychopathy ininstrumental and reactive violent oenders. Journal of consulting and clinical psychology, 64, 783790.

Crick, N. R., & Dodge, K. A. (1996). Social information-processing mechanisms on reactive and proactiveaggression. Child Development, 67, 9931002.

Damasio, A. R. (1994). Descartes error: emotion, rationality and the human brain. New York: Putnam(Grosset Books).

Dias, R., Robbins, T. W., & Roberts, A. C. (1996). Dissociation in prefrontal cortex of aective andattentional shifts. Nature, 380, 6972.

Eysenck, H. J. (1964). Crime and personality. London: Routledge & Kegan Paul.Eysenck, H. J., & Gudjonsson, G. H. (1989). The causes and cures of criminality. London: Plenum Press.Fine, C., Lumsden, J., & Blair, R. J. (2001). Dissociation between theory of mind and executive functions

in a patient with early left amygdala damage. Brain, 124, 287298.Fisher, L., & Blair, R. J. R. (1998). Cognitive impairment and its relationship to psychopathic tendencies

in children with emotional and behavioural diculties. Journal of Abnormal Child Psychology, 26,511519.

Flor, H., Birbaumer, N., Hermann, C., Ziegler, S., & Patrick, C. J. (2002). Aversive Pavlovianconditioning in psychopaths: peripheral and central correlates. Psychophysiology, 39, 505518.

Forth, A. E., Kosson, D. S., & Hare, R. D. (2003). The Psychopathy Checklist: Youth Version. Toronto,Ontario, Canada: Multi-Health Systems.

Fowles, D. C. (1988). Psychophysiology and psychopathy: a motivational approach. Psychophysiology, 25,373391.

Francis, D. D., & Meaney, M. J. (1999). Maternal care and the development of stress responses. CurrentOpinion in Neurobiology, 9, 128134.

R.J.R. Blair / Cognition 101 (2006) 414442 439Frick, P. J., & Hare, R. D. (2001). The antisocial process screening device. Toronto: Multi-Health Systems.Frick, P. J., Lilienfeld, S. O., Ellis, M., Loney, B., & Silverthorn, P. (1999). The association between

anxiety and psychopathy dimensions in children. Journal of Abnormal Child Psychology, 27, 383392.Frick, P. J., OBrien, B. S., Wootton, J. M., & McBurnett, K. (1994). Psychopathy and conduct problems

in children. Journal of Abnormal Psychology, 103, 700707.Frith, U. (2002). Autism: explaining the enigma. Oxford: Blackwell.Funayama, E. S., Grillon, C., Davis, M., & Phelps, E. A. (2001). A double dissociation in the aective

modulation of startle in humans: eects of unilateral temporal lobectomy. Journal of CognitiveNeuroscience, 13, 721729.

Gillott, A., Furniss, F., & Walter, A. (2001). Anxiety in high-functioning children with autism. Autism, 5,277286.

Grant, S., Contoreggi, C., & London, E. D. (2000). Drug abusers show impaired performance in alaboratory test of decision-making. Neuropsychologia, 38, 11801187.

Gray, J. A. (1987). The psychology of fear and stress. Cambridge: University of Cambridge Press.Gregg, T. R., & Siegel, A. (2001). Brain structures and neurotransmitters regulating aggression in cats:

implications for human aggression. Progess in Neuropsychopharmacology and Biological Psychiatry, 25,91140.

Hare, R. D. (1980). A research scale for the assessment of psychopathy in criminal populations.Personality and Individual Dierences, 1, 111119.

Hare, R. D. (1982). Psychopathy and physiological activity during anticipation of an aversive stimulus in adistraction paradigm. Psychophysiology, 19, 266271.

Hare, R. D. (1991). The hare psychopathy checklist-revised. Toronto, Ontario: Multi-Health Systems.Hare, R. D. (2003). Hare psychopathy checklist-revised (PCL-R) (2nd ed.). Toronto: MHS.Harmer, C. J., Perrett, D. I., Cowen, P. J., & Goodwin, G. M. (2001). Administration of the beta-

adrenoceptor blocker propranolol impairs the processing of facial expressions of sadness. Psycho-pharmacology (Berl), 154, 383389.

Harpur, T. J., & Hare, R. D. (1994). Assessment of psychopathy as a function of age. Journal of AbnormalPsychology, 103, 604609.

Harpur, T. J., Hare, R. D., & Hakstian, A. R. (1989). Two-factor conceptualization of psychopathy:construct validity and assessment implications. Psychological Assessment: A Journal of Consulting andClinical Psychology, 1, 617.

Hart, S. D., & Hare, R. D. (1996). Psychopathy and antisocial personality disorder. Current Opinion inPsychiatry, 9, 129132.

Hebb, D. O. (1949). The organization of behavior. New York: Wiley.Heim, C., Owens, M. J., Plotsky, P. M., & Nemero, C. B. (1997). Persistent changes in corticotropin-

releasing factor systems due to early life stress: relationship to the pathophysiology of major depressionand post-traumatic stress disorder. Psychopharmacology Bulletin, 33, 185192.

Hemphill, J. F., Hart, S. D., & Hare, R. D. (1994). Psychopathy and substance use. Journal of PersonalityDisorders, 8, 169180.

Herpertz, S. C., Werth, U., Lukas, G., Qunaibi, M., Schuerkens, A., Kunert, H. J., et al. (2001). Emotionin criminal oenders with psychopathy and borderline personality disorder. Archives of GeneralPsychiatry, 58, 737745.

Homan,M. L. (1984). Empathy, its limitations, and its role in a comprehensive moral theory. In J. Gewirtz&W.Kurtines (Eds.),Morality, moral development, andmoral behavior (pp. 283302). NewYork:Wiley.

Homan, M. L. (1994). Discipline and internalisation. Developmental Psychology, 30, 2628.Homan, M. L., & Saltzstein, H. D. (1967). Parent discipline and the childs moral development. Journal

of Personality and Social Psychology, 5, 4557.Hornak, J., Rolls, E. T., & Wade, D. (1996). Face and voice expression identication in patients with

emotional and behavioural changes following ventral frontal damage. Neuropsychologia, 34, 247261.House, T. H., & Milligan, W. L. (1976). Autonomic responses to modeled distress in prison psychopaths.

Journal of Personality and Social Psychology, 34, 556560.Jutai, J. W., & Hare, R. D. (1983). Psychopathy and selective attention during performance of a complex

perceptual-motor task. Psychophysiology, 20, 146151.

440 R.J.R. Blair / Cognition 101 (2006) 414442Karmilo-Smith, A., Brown, J. H., Grice, S., & Paterson, S. (2003). Dethroning the myth: cognitivedissociations and innate modularity in Williams syndrome. Developmental Neuropsychology, 23,227242.

Kiehl, K. A., Smith, A. M., Hare, R. D., Mendrek, A., Forster, B. B., Brink, J., et al. (2001). Limbicabnormalities in aective processing by criminal psychopaths as revealed by functional magneticresonance imaging. Biological Psychiatry, 50, 677684.

Kochanska, G. (1993). Toward a synthesis of parental socialization and child temperament in earlydevelopment of conscience. Child Development, 64, 325347.

Kochanska, G. (1997). Multiple pathways to conscience for children with dierent temperaments: fromtoddlerhood to age 5. Developmental Psychology, 33, 228240.

Kochanska, G., De Vet, K., Goldman, M., Murray, K., & Putman, P. (1994). Maternal reports ofconscience development and temperament in young children. Child Development, 65, 852868.

Kosson, D. S., Cyterski, T. D., Steuerwald, B. L., Neumann, C. S., & Walker-Matthews, S. (2002). Thereliability and validity of the psychopathy checklist: youth version (PCL:YV) in nonincarceratedadolescent males. Psychological Assessement, 14, 97109.

LaBar,K. S., Gatenby, J. C., Gore, J. C., LeDoux, J. E., & Phelps, E. A. (1998). Human amygdala activationduring conditioned fear acquisition and extinction: a mixed-trial fMRI study. Neuron, 20, 937945.

Leslie, A. M. (1992). Pretence, autism, and the theory-of-mind-module. Current Directions in PsychologicalScience, 1, 1821.

Levenston, G. K., Patrick, C. J., Bradley, M. M., & Lang, P. J. (2000). The psychopath as observer:emotion and attention in picture processing. Journal of Abnormal Psychology, 109, 373386.

Lichter, J. B., Barr, C. L., Kennedy, J. L., Van Tol, H. H., Kidd, K. K., & Livak, K. J. (1993). Ahypervariable segment in the human dopamine receptor D4 (DRD4) gene. Human Molecular Genetics,2, 767773.

Lilienfeld, S. O., & Andrews, B. P. (1996). Development and preliminary validation of a self-reportmeasure of psychopathic personality traits in noncriminal populations. Journal of PersonalityAssessement, 66, 488524.

Linnoila, M., Virkkunen, M., Scheinin, M., Nuutila, A., Rimon, R., & Goodwin, F. K. (1983). Lowcerebrospinal uid 5-hydroxy indoleacetic acid concentration dierentiates impulsive from nonimpul-sive violent behavior. Life Sciences, 33, 26092614.

Liu, D., Diorio, J., Tannenbaum, B., Caldji, C., Francis, D.,